(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Adjustment To College A Study Of 10 000 Veteran And Nonveteran Students In Sixteen American Colleges"

378.37 F85 cop 1 



Keep Your Card in This Pocket 

Books will be issued only on presentation of proper 
library cards. 

Unless labeled otherwise, books may be retained 
for two weeks. Borrowers finding books marked, de- 
laced or mutilated are expected to report same at 
library desk; otherwise the last borrower will be held 
responsible for all imperfections discovered. 

The card holder is responsible for all books drawn 
on this card. 

Penalty for over-due books 2c a day plus cost of 
notices. 

Lost cards and change of residence must be re- 
ported promptly. 




Public Library 

Kansas City, Mo. 



TENSION ENVELOPE CORP. 



3 1148 00699 9940 





A STUDY OF 10,000 VETERAN 
AND NONVETERAN STUDENTS 

IN SIXTEEN 
AMERICAN COLLEGES 



NORMAN JFREDERIKSEN 

AND 

W. B. SCHRADER 



EDUCATIONAL TESTING SERVICE 

PRINCETON, NEW JERSEY 
195 1 



This study was made possible "by funds 
granted by Carnegie Corporation of Nev York. 
That Corporation is not, however, the author, 
owner, publisher, or proprietor of this 
publication, and is not to be understood as 
approving by virtue of its grant any of the 
statements made or vievs expressed therein. 



TABLE OF CONTENTS 

Page 

Preface ............................. vii 

List of Tables ......................... ix 

List of Figures ......................... xiv 

CHAPTER I. THE FINDINGS AND THEIR IMPLICATIONS: A SUMMARY . . 1 

II. THE PLAN OF THE STUDY ............... 63 

III. THE ACADEMIC ADJUSTMENT OF VETERAN AND NOFVETERAN 
STUDENTS ..................... 96 

IV. SOME CHARACTERISTICS OF VETERAN STUDENTS ..... 185 

V. AGE AND GENERAL BACKGROUND OF VETERAN AND NON- 
VETERAN STUDENTS ................. 215 

VI. SOME FACTORS RELATED TO MOTIVATION ........ 236 

"VII. THE WORRIES OF COLLEGE STUDENTS .......... 258 

VIII. HOW COLLEGE STUDENTS SPEND THEIR TIME ....... 285 

IX. THE GI BILL .................... 310. 

X. THE STUDENT VIEWS HIS COLLEGE ........... 328 

APPENDIX A. DERAILED RESULTS OF QUESTIONNAIRE ANALYSIS .... 355 

APPENDIX B. COMPUTATIONAL PROCEDURES 

1. ANALYSIS -OF COVARIANCE PROCEDURE ......... k?>6 

2. DEO^IRMINATION OF ADJUSTED AVERAGE GRADES ..... Mi- 5 

3. TESTING SIGNIFICANCE OF RESPONSES BY THE F-TEST . . 
APPENDIX C* QUESTIONNAIRE ADMINISTRATION AND CODING 

1. DIRECTIONS FOR GROUP ADMINISTRATION OF THE STUDENT 
OPINION QUESTIONNAIRE ............... 



2. THE STUDENT OPINION QUESTIONNAIRE ......... 

3. EXCERPTS FROM CODING KEY AND CODING MANUAL .... 

BIBLIOGRAPHY .......................... 

INDEX ............................. 



Preface ' 



In the years immediately following World War II, when there was a 
good deal of speculation and little dependable information a"bout the ad- 
justment of veteran students to college, the idea of making a thorough 
investigation of the problem vas suggested by the Carnegie Corporation. 
After considerable thought and discussion of the possible values of such 
a study and the methods of obtaining and analyzing data, the investiga- 
tion described here was undertaken by the College Entrance Examination 
Board with the .support of the Carnegie Corporation. Following the merger 
which led to the formation of the Educational Testing Service, the study 
was carried to completion by ETS. 

It is believed that the findings reported will be of value in a 
number of ways and to various groups. Findings on the value of tests 
and high school record for predicting academic success in college should 
be of interest to guidance personnel, educational psychologists, and col- 
lege administrators, particularly admissions officers. The results of 
the analysis of the questionnaire items may have some significance for 
psychologists interested in personality as well as to guidance officers 
and educational psychologists. A considerable amount of information on 
background characteristics and attitudes of college students should be 
of interest to college officials . Finally, the findings with respect to 
college success of low income students and veterans enabled to attend 
college through the educational benefits of the GI Bill may have some 
significance with respect to scholarship programs. 

A great many people contributed to the study in a variety of ways. 
It is unfortunate that acknowledgment by name cannot be made to all those 
people at the various colleges and universities who permitted the study 
to be made and who supplied the data; these people cannot be named be- 
cause of the decision not to reveal the identities of the participating 
institutions. 

Acknowledgment is due to Dr. Charles Bollard, President of the Car- 
negie Corporation, and to Dr. 0. C. Carmichael, President of the Carnegie 
Foundation for the Advancement of Teaching, for their helpful suggestions 
in planning the general objectives and outlines of the study. The members 
of an Advisory Committee, consisting of Professor Philip Rulon, Mr. Lyle 
Spencer, Dr. Kenneth Yaughn, Professor Frederick Stephan, and Mr. Felix 
Moore, also participated in the Initial phases of the planning and made 
many useful suggestions regarding hypotheses to be tested and procedures 
for obtaining data. 

A great many members of the staff of the Educational Testing Service 
contributed to the study. Mr. Henry Chauncey, President of the Educa- 
tional Testing Service, contributed many useful suggestions and criticized 



vii 



portions of the manuscript. The technical assistance of Dr. Ledyard R 
Tucker,, Professor Harold Gulliksen, Dr. Frederic Lord, and particularly 
Professor S. S. Wilks has "been invaluable. Dr. John Clausen, who had 
major responsibility for the development of the Student Opinion Quest ion- 
pglre, contributed greatly through his broad experience in survey and 
questionnaire studies. Dr. Eobert Myers developed the questionnaire cod- 
ing manual and gave general supervision to the coding operation. Miss 
Henrietta Gallagher vas the direct supervisor of the coders. Mrs, Judith 
Aronson was in charge of computing and Mr, Harry Garrison supervised the 
punching and tabulating operations. Mr. Donald Peterson assisted in making 
arrangements with the participating colleges for obtaining data. Mrs. Mary 
McCabe supervised the transcription and preparation of much of the data. 
Mrs. Margaret Kostritsky aided in preparing the bibliography. 

The entire manuscript was read by Dr. William . Turnbull, who made 
many helpful suggestions. Portions of the manuscript were also read by 
Professor A. B. Crawford, Dr. Douglas Schultz, and Professor A. P. Horst, 
who also contributed many useful suggestions. Miss Evelyn Wicoff contrib- 
uted greatly through painstaking editorial work on the manuscript. The 
index was prepared by Mrs. Eleanor Apter. Finally, acknowledgment is 
gratefully made to Mrs. Sally Matlack, who spent many hours of careful 
work in typing the manuscript. 

Norman Frederiksen 
W. B. Schrader 



TABLES 

Page 

1. Percentage of Agreement in Initial Coding of Questionnaire 

Item ^5 for Several Samples . 82 



2. The Twenty-Five Groups Included in the Study , Arranged Ac- 
cording to University ................... 



3. Inter correlations of Scholastic Aptitude Test Scores , Date 
Tests Were Taken (for Veterans Only), Adjusted School Bank, 
Predicted Grade, and First -Year College Average Grades. 
Adams University, College of Arts and Science, Freshmen, 

o ....... 99 



k. Comparison of Average Grades Earned by Veteran and Non- 
veteran Male Students. Adams University, College of Arts 
and Science, Freshmen, 1946-1947 . . . .......... 101 

5- Comparison of Average Grades Earned by Veteran and Non- 

veteran Male Students. Stewart University, College of Arts 

and Science, Freshmen, 1946-1947 . . . ......... 10J 

6. Intercorrelations of Test Scores, High School Average, Year 
of High School Graduation (for Veterans Only), College 
Credit Hours, and First-Year College Grades. Douglas Uni- 
versity, College of Arts and Science, Male Freshmen, 1946- 
1947 ........... .......... ...... 106 

7- Intercorrelations of Test Scores, High School Average, Col- 
lege Credit Hours, and First -Year College Grades. Douglas 
University, College of Arts and Science, Female Freshmen, 

.......... ........... 107 



8. Comparison of Average Grades Earned by Veteran and Nonveteran 
Male Students. Douglas -University, College of Arts and Sci- 

ence, Freshmen, 1946-1947 ........ . . . . ..... 109 

9. Comparison of Average Grades Earned by Male Nonveterans and 
Female Nonveterans Douglas University, College of Arts 
and Science, Freshmen, 1946-1947 . . . . ......... 

10. Comparison of Average Grades Earned by Male Veterans and Fe- 
male Nonveterans . Douglas University, College of Arts and 
Science, Freshmen, 19^6-19^7 ....... ........ 

11, Comparison of Average Grades Earned by Veteran and Nonveteran 
Male Students . Harris University, College of Arts and Sci- 

ence, Freshmen, 1946-1947 ............ ..... 113 



TABLES 

Page 

32, Comparison of Average Grades Earned "by Yeteran and Nonveteran 
Male Students, Miller University, College of Arts and Sci- 
ence, Iresnsten, 1946-1947 . . . 115 

13. Comparison of Average Grades Earned "by Teteran and Uonveteran 
Male Students* Evans University, College of Arts and Sci- 
ence, Freshmen, 1946-1947 .-.,.... 118 

14. Comparison, of Average Grades Earned "by Veteran and Nonveteran 
Male Students, Taylor University, College of Arts and Sci- 
ence. (Students vho entered as freshmen in fall, 1945, or 

fall, 1946.) . . . . . . . . . 120 

15. Intercorrelations of American Council Psychological Examina- 
tion Scores, High School Average, and First-Year College 
Average Grades, Western State University, College of Arts 

and Science, Male Freshmen, 1946-1947 . . 122 

16. Intercorrelations of American Council Psychological Examina- 
tion Scores, High School Average, and First-Tear College 
Average Grades. Westerii State University, College of Arts 

and Science, Female Freshmen, 1946-194-7 , . 12J 

17. Comparison of Average Grades Earned "by Veteran and Nonveteran 
Male Students. Western State University, College of Arts and 
Science, Freshmen, 1946-1947 125 

18. Comparison of Average Grades Earned "by Male Nonveteran and 
Female Students. Western State University, College of Arts 

and Science, Freshmen, 1946-1947 ..>.. 126 

19. Comparison of Average Grades Earned "by Male Yeteran and Fe- 
male Students. Western State University, College of Arts 

and Science, Freshmen, 1946-1947 . , . . 128 

20 o Comparison of Average Grades Earned "by Yeteran and Nonveteran 
Male Students. Central State University, College of Arts 
and Science, Freshmen, 1946-1947 . . 150 

21. Comparison of Average Grades Earned by Yeteran and Nonveteran 
Male Students. Central State University, College of Arts 
and Science. (Students entering as freshmen during the aca- 
demic year 1945*1946.) .,...... 151 

22. Comparison of Average Grades Earned "by Yeteran and Nonveteran 
Male Students. Llttletown State University, College of Arts 

and Science, Freshmen, 1946-1947 . ...,.. 155 



TABLES 



23. Comparison of Ayerage Grades Earned "by Yeteran and Nonveteran 
Male Student s. Eastern City University, College of Arts and 
Science, Freshmen, 1946-1947 . . . . 135 

2k. Comparison of Average Grades Earned "by Veteran and Nonveteran 
Male Students. Midwest City University, College of Arts and 
Science, Freshmen, 1946-1947 . , 9 . . 137 

25. Intercorrelatlons of Test Scores, High School Average, Col- 
lege Credit Hours, and First -Year College Grades. Midwest 
Tt yhnological University, College of Engineering, Freshmen, 

1946-1947 ............ . 14 o 

26. Comparison of Average Grades Earned by Yeteran and Nonveteran 
Male Students. Midwest Technological University, College of 
Engineering, Freshmen, 1946-1947 .............. 142 

27. Intercorrelatlons of American Council Psychological Examina- 
tion Total Score, High School Rank, Year of High School Grad- 
uation, and First-Year College Grades. Middle State Uni- 
versity, College of Engineering, Freshmen, 1946-1947 . , * 144 

28. Comparison of Average Grades Earned by Yeteran and Nonveteran 
Male Students. Middle State University, College of Engineer- 
ing, Freshmen, 1946-1947 .. 9 ........ 146 

29. Comparison of Drawing Grades Earned by Yeteran and Nonveteran 
Male Students. Middle State University, College of Engineer- 
ing, Freshmen, 1946-1947 147 

30. Comparison of Mathematics Grades Earned by Yeteran and Non- 
veteran Male Students. Middle State University, College of 
Engineering, Freshmen, 1946-1947 , , . . 148 

31 Comparison of Average Grades Earned by Yeteran and Nonveteran 
Male Students. Midwest City University, College of Engineer- 
ing, Freshmen, 1946-1947 ......... 150 

32. Comparison of Average Grades Earned by Yeteran and Nonveteran 
Male Students, Southern Technological University, College of 
Engineering- (Students who entered as freshmen in fall, 1945; 

or fall, 19460 152 

33, Intercorrelations of Test Scores, High School Average, Col- 
lege Credit Hours, and First-Year College Grades. Midwest 
Technological University, College of Agriculture, Freshmen, 
1946-1947 ............... ^ .......... 15^ 



TABLES 

Page 

J14-. Comparison of Average Grades Earned by Veteran and Nonveteran 
Male Students. Midwest Technological University,, College of 
Agriculture , Freshmen, 19^6-19^7 .............. 156 

35. Comparison of Average Grades Earned by Veteran and Nonveteran 
.Male Students. Midwest State University, College of Business, 
Freshmen, 19^6-19^7 158 

36 Comparison of Average Grades Earned by Veteran and Nonveteran 
Male Students. Littletovn State University, College of Busi- 
ness,, Freshmen, 19^6-19^7 159 

37, Intercorrelations of Scholastic Aptitude Test Scores, Ad- 
justed School Rank, Predicted Grade, and First and Second - 
Year College Average Grades, Adams University, College of 
Arts and Science. (Veterans who completed one year before war 
service and one year after war service; nonveterans who 

entered in the fall of 19^5 and completed two years,) . . , . 162 

38, Comparison of Average Fourth -Semester Grades Earned by Inter- 
rupted Veteran and Uninterrupted Nonveteran Male Students. 
Adams University, College of Arts and Science. (Veterans who 
completed one year before war service, one year after war 
service; nonveterans who entered in the fall of 19^5 an & com- 
pleted two years . ) * ...*..., , 1,6k 

39- Intercorrelations of First and Second -Year College Average 
Grades. Stewart University, College of Arts and Science. 
(Veterans who completed one year before war service and one 
year after war service; nonveterans who entered in the fall 
of 19^5 an & completed two years.) , 166 

40. Comparison of Average Fourth-Semester Grades Earned by Inter- 
rupted Veteran and Uninterrupted Nonveteran Male Students. 
Stewart University, College of Arts and Science. (Veterans 
who completed one year before war service, one year after war 
service; nonveterans who entered in the fall of 19^5 and com- 
pleted two years . ) . . . . 167 

41. Comparison of Average Fourth-Semester Grades Earned by Inter- 
rupted Veteran and Uninterrupted Nonveteran Male Students. 
Eastern City University, College of Arts and Science. 
(Veterans who completed one year before war service; one year 
after war service; nonveterans who entered in the fall of I9k^ 

and completed two years .).,...... 169 



TABLES 

Page 

42. Comparison of Average Eighth-Quarter Grades Earned by Inter- 
rupted Veteran and Uninterrupted Non-veteran Students. Mid- 
vest Technological University, College of Engineering. 
(Veterans who completed two to six quarters before "war ser- 
vice and two to six quarters after war service for a total 
of at least eight quarters; nonveterans who entered in the 

fall of 1939 and completed eight quarters.) ......... 171 

43. Comparison of Average Eighth-Quarter Grades Earned by Inter- 
rupted Veteran and Uninterrupted Nonveteran Male Students, 
Midwest Technological University, College of Agriculture. 
(Veterans who completed two to six quarters before war ser- 
vice and two to six quarters after war service for a total 
of at least eight quarters; nonveterans who entered in the 

fall of 1939 and completed eight quarters.) ......... 173 

44. Correlations of American Council Psychological Examination 
Scores and Measures of High School Standing with First -Year 
College Average Grade .................... 175 

45. Median Hours Per Week Spent Attending Classes,, Laboratories, 
and Other Regularly Scheduled Course Conferences: Item 

22(a) ............................ 288 

46. Median Hours Per Week Spent Studying in Boom, in Library, or 
Elsewhere: Item 22 (b) ................... 292 

47. Comparison of Average Grades Earned by Veteran Male Students 
(A) Who Probably Would Have Attended College and (B) Who 
Probably Would Not Have Attended College, Without Veterans 8 
Benefits. Miller University, College of Arts and Science, 
Preshmen, 1946-1947 ..................... 



48. Outcome of Significance Tests When Veterans Are Classified On 
the Basis of Whether or Not They Would Have Attended College 
Without Veterans' Benefits As Well As On Questionnaire Re- 
sponse, for Ten Responses Which Are Significant for Veterans 
at Both Central State and Miller and Not Significant for Non- 
veterans in Either College Group. (Miller University, 425 
freshman male veterans enrolled in liberal arts.) ...... 

49. Relation Between College Plans As Reported On Items 8(0) and 
8(m) and Various Measures of Aptitude and Achievement. 

I. Middle State University, 352 Freshmen in Engineering. 

II. Central State University, 466 Preshmen in Liberal Arts. . 323 



FIGURES 

Page 

1. Per Cent of Veterans Excelling the Average Nonveteran with 
Respect to Adjusted Average Grade in Each of Twenty-Five 

College Groups ...................... . k 

2. Median Per Cent Making Selected Eesponses to Questionnaire 

Items Among Veteran and Nonveteran Students ...... 17 

3. Median Per Cent of Veteran and Nonveteran Male Students 

Bothered "by Various Problems .... ............ 22 



Per Cent Making Selected Eesponses to Questionnaire Items 

Among Older Male Veterans and Younger Male Veterans .... kk 

Per Cent Making Selected Responses to Questionnaire Items 
Among Older Male Veterans, Younger Male Veterans , Male Non- 
veterans, and Female Nonveterans . . . ........ <, . . ^5 



6. Hypothetical Relationships Between Non-Aptitude Determiners 
of College-Going and (l) Per Cent of High School Graduates 
Who Entered College and (2) Estimated Motivation for College 

Work Among Students Who Actually Attend .......... Ij.9 

7. Median Per Cent Making Selected Responses to Questionnaire 
Items Among Veteran Male, Nonveteran Male, and Female Stu- 

dents . . . o ....... ............... 51 

8. Median Per Cent of Nonveteran Male and Female Students 

Bothered by Various Problems .... ..... 8 ...... 55 

9. Relation Between Two Independent Codings of Question ^5 
(Comments Regarding College) for 166 Questionnaires of Evano 
University Students .............. ..... 8l 

10. Determination of the Proportion of Veterans Excelling the 
.Average Nonveteran, Assuming Normal Distribution and Equal 
Standard Deviations ...... ...... , ...... 88 

11. Illustration of a Sign Test for Testing the Hypothesis 
That More Male Veterans (M?) Than Male Nonveterans (MN) 
Possess the Characteristics Associated with Superior Ad- 

justed Average Grade (AAG) ............. , 9-5 

12. Trends in Grades of Interrupted Veterans and Uninterrupted 
Nonveterans . , . t ..... . 



13. Trends in Grades of Veteran and Nonveteran Male Students . . 179 



FIGURES 

Page 

Ik. Distribution of Per Cent of Veterans Excelling the Average 
Nonveteran vith Eespect to Adjusted Average Grade for the 
Ten Groups in Which High School Standing Was Used in Combina- 
tion with a Test or Tests and for the Ten Groups in Which 
Tests Only Were Used ............... . . . . . 2,8k 

15. Branch of Service: Item 8(a) ....... ........ 188 

16, Length of Service: Item 8(b) . . ............. 190 

iy. Highest Bating, Bank or Grade Held While in Service: Item 

8(c) ..................... ....... 192 



18. Service Outside the United States: Item 8(f ) 



19. Influence of Service Experience on Eagerness to Attend Col- 

lege: Item 8(p) .......... , ......... . . 199 

20. Influence of Service Experience on Scholastic Ability: Item 

8(qj ..... . ...... . .......... . . . . . 201 

21. Time of Decision to Attend College: Item 8(j) ....... 205 

22. Year of Last Full-Time Attendance in Secondary School: Item 

6(b) ........ ....... .......... . . . 207 

23. Marital Status: Item ?k .................. 210 

2k. Year of Birth, Male Veteran Students: Item 32 ....... 217 

25. Year of Birth, Male Nonveteran Students: Item 32 ..... 218 

6. Length of Full-Time Employment Before Entering War Service 

or College: Item 9(b) ........ ........... 221 

27. Size of Home Community While in Secondary School: Item 31 - 

28. Head of Family's Income During Student's Secondary School 
Career: Item 43 ......... * ...... . ..... 



29. Father's Education: Item kk . . . * ............ 228 

30. Evaluation of Preparation for College: Item 26(a) ..... 232 

31. Chief Eeasons for Attending College : Item 10 ....... 237 

32. Vocational Objective: Item 11 . . . . . ..... ..... 



52. Hours Per Week Spent in Social Activities: Item 22 (e) . . . 



FIGURES 

Page 

33. Certainty of Vocational Choice: Item 12 . ..... . , 242 

3!*. Importance of College Graduation for Vocational Objectives: 
Item 13 ............... . ....... .0. 

35. Importance of College Grades for Vocational Opportunities: 
Item Ik ............... o . . . ...... . 

36. Length of Time Planned for Earning Degree: Item 21 ...... 

37. Difficulty of College Work: Item 20 ...... ...... 251 

38. Evaluation of Effort Exerted in Doing College Work: Item 28 253 

39. Tendency to Worry: Item 39 ........... * . . . . 259 

ifO, Worry and Anxiety About Concentration: Item 40(g) ..... 262 

kl, Worry and Anxiety About Getting Accustomed to College Study; 

Item 40(f ) ................ ......... 26k 

42. Worry aixd Anxiety About Deciding What Course of Study to 

Follows Item 4o(k) ...... t ..... ......., 265 

43* Worry and Anxiety About Making Ends Meet; Item 40(a) . . 268 
kk. Worry and Anxiety About Peelings of Inferiority: Item 40(j) 270 
45.. Worry and Anxiety About Nervousness: Item 40(d) ...... 272 

k6. Worry aiad Anxiety About Getting to Know People Socially: 

Item kO(h) ......... . ...... , ......... , ^k 



if 7. Extent to Which Worries and Anxieties Have Interfered vith 

College Work: Item k . . ........ ., ..... 28l 

k&. Hours Per Week Spent Attending Classes: Item 22 (a) , , 287 

49. Hours Per Week Spent Studying: Item 22 (b) * ....... . 290 

50. Hours Per Week Spent in Athletics: Item 22 (c) ..... 295 

51. Hours Per Week Spent in Organized Extracurricular Activi- 

ties; Item 22(d) ............... ....... P 



XVI 



FIGURES 

Page 

53. Hours Per Week Spent Attending Public Lectures, Concerts, and 

Other Cultural Activities: Item 22 (f) ...... . . . . 299 

5k. Hours Per Week Spent in Bull Sessions^ Item 22 (g) ..... 301 

55. Hours Per Week Spent in Paid Employments Item 22 (h) . . , 303 

56, Hours Per Week Spent in Voluntary Reading and Study: Item 



305 



57. Attendance at Evening Lectures During Previous Four Weeks: 

Item 25 . ..... * o ..... . . . o ...... 307 

58. Importance of Financial Aid Provided by Veterans' Benefits 
in Decision of Veteran Students to Come to College: Item 

8(0) , . . . . ....... . . . ........... . . 311 

59. Per Cent Making Various Selected Responses to Questionnaire 
Items Among Veterans Who Probably Would Have Attended Without 
Veterans' Benefits (Subgroup A) and Those Who fxobably Would 

Not Have Attended Without These Benefits (Subgroup B) ... 315 

60. Per Cent Making Various Selected Responses to Questionnaire 
Items Among: Veterans Who I^robably Would Have Attended With- 
out Veterans' Benefits (MV:A), Veterans Who Probably Would 
Not Have Attended Without These Benefits (MV:B), Male Non- 
veterans (MN) and Women Students (F) . . . . ........ 317 

61. Likelihood of Student 1 a Attending College If He Had Not 

Entered Service: Item 8(m) . ......... ..... 321 

62. Veterans* Educational Benefits Applied For: Item 8(n) ... 3^5 

63. Proportion of Courses Beally of Interest to Student: Item 

37 ........ ....... .............. 329 

6k . Evaluation of College Studies: Item 38 .......... 332 

65. Satisfaction With Education: Item 36 .......... 

66. Satisfaction With Present School or Division: Item 16 . . . 

67. Attitude Toward Faculty: Item 17 ............. 338 

68. Adequacy of Place for Study: Item 2? .... . ...... 

69. Most Frequent Suggestions for Improving the College: Item 
45-2 (Part 2) .... ........ . .......... 



70. Satisfaction with Present University: Item 15 ....... 351 

xvSi 



Chapter I 
TEE PIKDBTGS AISD THEIR IMPLICATIONS s A SUMMARY 

Introduction 

The influx of about one million veterans into American colleges and 
universities at the close of World War II was a remarkable phenomenon in 
American higher education. Besides bringing college enrollments to a 
peak veil beyond any previous level, the veterans were obviously distinc- 
tive in at least three other wayss First, they brought a background of 
experiences which often had no counterpart in the backgrounds of civilian 
students; those who had not been in combat had at least undergone the ex- 
perience of service in the armed forces in time of war. Second, they were 
enough older than their nonveteran fellow students to change the general 
appearance of the student body. Third, because of the educational pro- 
visions of the GI Bill of Rights, their decision to attend college, and 
their choice of college, was undoubtedly less affected by the economic 
status of their family than would usually be the case. Prom a psycho- 
logical viewpoint, a need was evident for getting behind these more con- 
spicuous characteristics of veterans and to describe veteran-nonveteran 
differences in terms of more meaningful psychological and educational 
variables achievement, aptitudes, worries, attitudes. 

Some characteristics of the veteran group had decisive administra- 
tive implications, so that speculation, judgment, study, and interpreta- 
tion were brought heavily to bear on these issues. Veterans' educational 
plans were important in college planning for staff and facilities Their 
marital status and family responsibilities affected college housing plans 
Their emotional stability or instability bore on the question of psychi- 
atric and psychological services,, Their ability to form wise and realis- 
tic vocational and educational plans affected counselling needs. Their 
formal and informal educational experiences in the service created a need 
for aids in evaluating these experiences; the American Council oil Educa- 
tion provided the Guide to the Evaluation of Educational Experiences in 
the Alined Serviceo and the General Educational Development tests to aid 
in this matter. Their ability to do college work without having completed 
the usual prerequisites had important implications for admissions pro- 
cedures. Their "ruotineao" in academic pursuits was made the basis for 
refresher and other course provisions to ease their transition into the 
stream of college life. Their desire for acceleration influenced college 
calendars and student programs, and led to procedures for admitting stu- 
dents at other than the usual time- Their reasons for coming to college 
and attitudes toward the conventional academic curriculum bore on the 
touchy subject of curriculum adjustment. All of these subjects have been 
dealt with in the extensive published literature on this fascinating groui 
and no doubt college files contain many unpublished reports on these sub- 
jects. 



ADJUSTMENT TO COLLEGE 



Need was evident for a systematic comparison of veterans and non- 
veterans with respect to such things as background, attitudes and motives, 
worries, and participation in various aspects of college life. Such in- 
formation, valuable for its own sake, should also aid in understanding the 
dynamics of academic success and failure and in accounting for veteran- 
nonveteran differences In academic success * 

A particular need was recognized for specific study of the students 
who would not have attended college without the aid provided by the GI 
Bill an inquiry which would consider not only academic success but also 
various personal characteristics of these students. 

The study to be reported here was designed to meet these varied 
needs. This chapter Is primarily a summary of findings; only enough atten- 
tion to method will be given to make evident the basis for the results re- 
ported,, In summarizing the findings, the following sequence wlU be fol- 
lowed, First, an effort will be made to determine whether veterans did 
earn better grades in college, relative to ability, than did nonveterans. 
Second, veterans and nonveterans will be compared with regard to background 
and attitudes, as reported by them on a questionnaire. TOiIrd, the value 
of the questionnaire items in identifying promising students will be con- 
sidered. Fourth, the special study of veterans brought Into college by 
the GI Bill will be summarized. Fifth, the Information obtained from the 
questionnaire will be used in an attempt to account for differences In 
academic performance between veterans and nonveterans. Sixth, some com- 
parisons of men and women students with respect to academic success, back- 
ground, and attitudes will be made. Seventh, the findings of this study 
regarding the effectiveness of conventional predictors of academic success 
will be summarized* Eighth, the possibilities of using data from the stu- 
dent questionnaire for obtaining a description of a college will be Illus- 
trated. 



Did Veterans Succeed Better in College Than Nonveterans? 

The academic success of veterans in college was early recognized as 
an important subject for Investigation. Evidence was needed promptly to 
aid In making short-term adjustments In admissions 7 placement, counselling, 
and curriculum. More Intensive study was also clearly needed to evaluate 
the possible implications of the veterans" success for long-range formula- 
tions of educational and public policy* It was clear that the perform- 
ance of veterans had an important bearing on two major Issues In higher 
education: who should go to college? and at what age should the typical 
student enter college? It was also clear that the performance of veterans 
in college was relevant to the problem of predicting college success, a 
matter which has received considerable attention from psychologists, 
especially during the past thirty years. 

A major purpose of the present study was to provide a reasonably 
clear answer to the questions how did veterans and nonveterans differ 
with respect to academic success? On examination of this question, It 



THE FINDINGS AND THEIR IMPLICATIONS 



"became evident that a number of specific steps must be taken in order to 
minimize influences which might obscure the differences and thus present 
a misleading estimate of their importance. In particular^ the following 
procedures were carried out* 

1. Kecognizing the diversity of American colleges 
and universities, studies were made in cooperation 
with l6 colleges and universities so chosen as to in- 
clude private colleges, state universities, and munic- 
ipal universities! coeducational and men's colleges ; 
large universities and relatively small colleges; col- 
leges with great financial resources and less wealthy 
institutions $ and colleges located in large cities as 
well as colleges in small towns. 

2. The crucial comparisons in this study were between 
male veterans and male nonveterans who were enrolled in 
the same division (e.g., liberal arts or engineering), 
in the same class (e.g., freshman or sophomore), of the 
same college or university. 

3. Differences between veterans and nonveterans in 
aptitude for college, as measured by conventional pre- 
dictors of college success, were controlled by the use 
of analysis of covariance. 

k. Those veterans who had received a substantial amount 
of college training in the V-12 program or in the Army 
Specialized Training Program were excluded from the com- 
parisons . 

It is not necessary at this point to describe in detail the analysis 
of covariance procedures used in evaluating veteran-nonveteran differ- 
ences in academic achievement; the laethods are described fully in Chapter 
II. Separate comparisons were made for veteran and nonveteran students 
in each of twenty-five separate groups, each of which was homogeneous 
with respect to institution, division within the institution, and aca- 
demic status of the students; and each comparison involved the use, as 
the criterion, of a measure of achievement -relative-to-ability called the 
Adjusted Average Grade (MG) . In effect, then, in each college group 
veteran students were compared with nonveterans of the same ability. 
The estimates of ability which were employed were based, in most instances, 
on a test or tests of scholastic aptitude and achievement, or on a measure 
of high school success used in combination with a test or tests of apti- 
tude and achievement. 

What, then, were the findings with respect to veteran-nonveteran 
differences in Adjusted Average Grade? In Figure 1 twenty-five separate 
answers are provided, one for each group studied It will be seen that 
the names used for the sixteen colleges and universities are fictitious; 
the code names are used throughout the report in order to preserve the 



ADJUSTMENT TO COLLEGE 



Group 


Tear of 
Entrance 


Per Gent of "Veterans Excelling the 
Average Nonveteran 
Ifi 2p 3p \0 5p 6p 7p 8,0 Sp 100 



GEOUPS INCLUDING VETERANS WHO ENTERED COLLEGE AS FRESHMEN AFTER WAR SERVICE 



Midwest Tech*, Engr. 
Midwest Tectu, Agri. 
Southern Teclu, Engr. 
Central State, Arts 
Middle State, Engr,, 
Midwest State, Bus. 
Midwest City, Arts 
Evans, Arts 
Little town State, Bus 
Western State, Arts 
Miller, Arts 
Stewart, Arts 
Central State, Arts 
Harris, Arts 
Adams, Arts 
Douglas, Arts 
Little town State, Arts 
Midwest City, Engr. 
Eastern City, Arts 
Turner, Arts 



-iolj.6 //////////////.{/ /////////ffJ/WJ A 



/JMJ///./////////////////J///A 




GROUPS rWCHlDING YETERAHS IfflD BEPORBDED TO COLLEaE AFTER WAR SERVICE AMD 
U01WE1ERMS WHO EKTEKED COIUSGE OT THE TEAE SPECIFIED 



Eastern City, Arts 
Stewart, Arts 
Midwest Tech., Agri. 
Adams, Arts 
Midwest Tech., Engr. 




Level of Significance;; 
level J I Hot significant 
level L i Ambiguoua 



FIGURE 1. PER CENT OF VETERANS EXCELLING TBE AVERAGE WONVETEBAN WITH 
KESPECT TO ADJUSTED AVERAGE GRADE JOT EACH QF TWENTT-FIVE COLLEGE GROUPS. 



THi FINDINGS AND THESR IMPLICATIONS 



anonymity of the cooperating institutions. Let us consider first the 
results for the twenty groups containing freshman student s 

For ease of interpretation,, the results are presented in terms of 
the per cent of veterans excelling the average nonveteran in Adjusted 
Average Grade, that is, freshman average grade adjusted in such a manner 
that any ability differences between veterans and nonveterans have been 
cancelledo If veterans and nonveterans were exactly alike, the two AAG 
distributions and the mean AAGs would coincide, and 50 per cent of the 
veterans would be found to excel the average nonveteran* If veteran stu- 
dents are slightly superior, the two distributions will not coincide but 
will overlap, and that proportion of the veterans' distribution which falls 
to the right of the mean AAG for nonveterans will exceed 50 per cent (see 
Figure 10, p 11-26). 

Referring to Figure 1, we may note the two most extreme cases; 76 
per cent of the veterans excelled the average nonveteran in Adjusted 
Average Grade in the engineering school of Midwest Tech (a midwestern 
land -grant college), while at the opposite extreme only 39 per cent of 
the veterans excelled the average nonveteran at Turner (a private coedu- 
cational university) . In the latter case the veterans were actually 
inferior to the nonveterans. 

Veteran superiority in grades relative to ability is then not a uni- 
versal tendency. What about the over -all results? It will be observed 
that in 16 of the 20 comparisons involving freshman students the veteran 
subgroup was superior; one would expect by chance to find 16 out of 20 
differences in one direction less than 5 times in a hundred trials. This 
result in itself, being significant at the yfa level, may be considered 
as moderately convincing evidence of superiority of veteran students in 
academic work when ability differences are kept constant . In the remain- 
ing four comparisons the nonveterans were superior (although for liberal 
arts students at Littletown State the superiority of the nonveterans is 
too small to be apparent in Figure l) Q 

Are the differences between veterans and nonveterana statistically 
significant in each of the 20 college groups now being discussed? Evi- 
dence on this question is shown by the shading of the bars in Figure 1. 
The black bars indicate a highly significant difference (the 1$ level of 
confidence), the diagonally shaded bars indicate significance at the 556 
level, and the white bars indicate results which are not significant at 
the 5$ level. Bars outlined with broken lines indicate that the results 
of the significance test were ambiguous (because of technical considera- 
tions related to the fact that errors of estimate are greater for one 
subgroup than for the other), 

In three of the comparisons, those for engineering students at Mid- 
west Tech and Middle State and for liberal arts students at Central State, 
who entered lu 19^6, the veterans are significantly superior (at the 1$ 
level) to ixonveterans . In three additional comparisons the differences 
are significant at the 5$ level,, The remaining ten comparisons which, 



ADJUSTMENT TO COLLEGE 



showed veterans to "be superior were either ambiguous or not significant 
(at the 5$ level) In none of the four cases where nonveterans were 
superior was the difference significant. 

On the whole, therefore, it must "be concluded from the studies of 
freshman students that there jL a tendency for veterans to achieve higher 
grades in relation to ability than do nonveteran students,, The actual 
magnitude of the difference is small, however. In the most extreme case, 
the advantage of the veterans would on the average amount to no more than 
the difference between a C and a C+o In other institutions the difference 
was much smaller, and as we have seen the direction of the difference is 
even reversed in some colleges. 

Now let us consider the remaining five comparisons those involving 
"interrupted " veterans. Substantially all the veteran students in these 
five comparisons were those who had completed at least two quarters of 
freshman work as ordinary civilians, who then entered military service, 
and who returned to college after discharge and completed at least an 
additional two quarters of academic work. The nonveterans with whom the 
interrupted veterans were compared were ordinary civilian students who 
completed without interruption the same amount of academic work as the 
veterans o In three of the college groups these nonveterans were post- 
war students who entered college as freshmen in 19^5; but at Midwest 
Tech they were prewar students who had entered in 1939. In all five of 
these groups, college grades earned early in a student's academic career 
were taken as the measure of his ability; those earned later in his col- 
lege career were used as the measure of his academic success. For the 
veterans, of course, a considerable amount of time elapsed between their 
early and later academic work. 

In all five groups involving interrupted veterans, veteran students 
were superior to nonveterans of equal ability. In four of the five com- 
parisons, the difference in Adjusted Average Grade was significant at the 
1% level o The proportion of veterans who excelled the average nonveteran 
ranged from 72 per cent at Eastern City and Stewart to 57 per cent at the 
Midwest Tech engineering school,. The evidence from these studies of 
Interrupted students thus strongly supports the hypothesis that veterano 
lo excel nonveterans of equal ability with respect to achievement in col- 
lege. 

One possible interpretation of ouch results for interrupted veterano 
Ls that the students let down noticeably in effort during the term just 
>rior to induction,, Usually the student knew that his induction was immi- 
Lent, and it was in some cases a question as to whether or not he would 
>e able to complete the term's work. According to this hypothesis, the 
Lifference in grades before and after war service is due to a let-down 
.n effort before war service rather than improvement after war service 
,vidence available for two college groups suggests, however, that this 
ypothesis does not hold. The trends in average grade earned during the 
wo semesters of the freshman year were alhnost the same for the veterans 
nd their nonveteran controls. Furthermore, in all five groups definite 



THE FINDINGS AND THEIR IMPLICATIONS 



steps were taken to ensure that the measure of ability was not "biased by a 
possible letdown in effort during the last term in school. (A more exten- 
sive discussion of the difficulties in interpreting these results is pre- 
sented on pages 172 - 



Relation to Other Studies 

Although the method employed in the present study differed in certain 
respects from that employed in any previous study of veteran -nonveteran 
differences in academic success, it is desirable to view the present results 
in the light of the numerous previous comparisons of veterans and nonveterar 
(At the cost of interrupting the presentation of other results of this 
study, a fairly detailed review of other studies comparing the academic 
success of veterans and nonveterans is introduced at this point, since over- 
all evaluation of the veterans* academic success is a major objective.) 

The studies to be reviewed may be classified according to the specific 
question which each kind of study attempted to answer. The first group of 
studies took quite literally the question: do veterans do better than 
nonveterans in college work? All veterans in a particular administrative 
jurisdiction were compared with all nonveterans in that group with respect 
to college grades. The second group of studies modified the question to 
read somewhat as follows: other things except age being roughly equal, 
do veterans do better than nonveterans in college work? In the present 
investigation, the studies of the 20 groups which included entering fresh- 
men fall in this category. A third group of studies asked, in effect, "the 
questions do returning veterans do better in college after the war than 
they did before their wartime service? In the present investigation, the 
five studies of interrupted veterans belong to this category. A fourth 
group of studies, which emphasized the role of age in relationship to 
veteran-nonveteran differences in academic success, will be discussed in 
a later section on possible explanations of the veterans 1 superiority. 

The initial approach to the question of how well veterans were suc- 
ceeding in college took the form of inquiries directed to college presi- 
dents and other administrative officers or to college faculty members. 
President Walters of the University of Cincinnati queried a large number 
of university presidents as to how well veterans were getting along and 
got almost unanimously favorable opinions of the veteran students. The 
reports indicated that, on the average, they were doing as well as or 
better than their nonveteran fellow-students . A number of examples of 
replies to this questionnaire were reported in an address made by Presi- 
dent Walters to a conference on veterans' education sponsored by the 
American Council on Education in July, I$k6 (100). In a later survey 
based on the opinions of the presidents of 98 large universities, the 
results again were favorable to the veterans; they were making grades 
higher than the prewar average, were adjusting well, and were serious- 
minded (99). Dean Bender (6) reported that although it was difficult to 
generalize about all veterans at Harvard, the veterans had "at least done 
no worse than nonveterans" and that the percentage who made Dean's List 



ADJUSTMENT TO COLLEGE 



records vas higher than for any prewar group . Mathewson (62), on the "basis 
of an informal survey,, reported that veterans are on the whole "high-grade 
students" with favorable attitudes toward college work. In another early 
study, Young (105) surveyed the faculty of Shrivenham American University, 
and found that they rated their soldier-students superior in interest in 
academic work and in general intellectual power to their former students 
in American colleges . 

On the basis of a questionnaire distributed in May, 19^6 > Kamm and 
Wrenn (55) reported that among 122 coeducational universities and liberal 
arts colleges who replied, two thirds reported that veterans were excell- 
ing nonveterans in scholastic achievement The remaining one third re- 
ported no difference between the two groups. Fine (32) learned during a 
tour of colleges in the East and Middle West that veterans had impressed 
teachers by their maturity and eagerness and that they were earning better- 
than-average grades at many colleges,, 

These survey reports undoubtedly helped to reassure those who had 
feared that the "veterans 9 bulge" represented a threat to academic stand- 
ards. On the whole, the generally favorable tone of these reports has 
been borne out by later, more technical studies . 

A number of over -all comparisons of grades earned by veterans and 
nonveterans enrolled in a particular university or in a particular uni- 
versity division were made, especially during the academic year 19^5- 
1946. In part, use of these broader groups was necessitated by the 
limited number of students available in certain finer classifications 
either in the veteran or in the nonveteran group . In spite of the compli- 
cation in interpretation introduced by the relatively heterogeneous groups 
of veterans and nonveterans included, these studies served a valuable pur- 
pose in indicating the order of magnitude of the difference between 
veterans and nonveterans. On the whole, it became evident that the dif- 
ference in academic achievement was not very large Most, but not all, 
Df the average grades reported favored the veterans 

A rather comprehensive study was reported by Tlbbetts and Hunter 
'93)o This study was based on the records of 857 male veterana and 846 
aale nonveterans at the University of Michigan during the fall term, 
19^5-19^6 Among the six divisions studied, however, only one had more 
ihan 50 veterans and more than 50 nonveterans In this large division 
;he veterans averaged 250 while the nonveterans averaged 2, ,51, For 
ill six divisions, the average grade was for veterans 2,56, for non- 
-eterans, 2,55,, Junior and senior veterans excelled the nonveterans at 
.he same academic level; nonveterans excelled in the freshman and oopho- 
Lore groups . In none of the four years did the difference exceed one 
eventh of a letter grade Tibbetts and Hunter noted that freshman 
eterans were apparently not inferior to the freshman rionveterano in the 
ests of ability and achievement given at entrance, 

Another extensive study was made by Thompson and Fleoher (90) at 
hio State University u Grades earned during the winter quarter of the 
cademic year 19^5-19^6 in agriculture, arts, commerce and education col- 



THE FINDINGS AND THEIR IMPLICATIONS 



leges were included It was found that 1,399 freshman male veterans 
earned an average grade of 2 33 as compared with 2d8 for 1,072 male non- 
veterans o In the other three -undergraduate classes, 579 male veterans 
earned an average of 2 75 as compared with 2,56 for 6ll male nonveterans. 
Here, again, the maximum difference is only about one fifth of a letter 
grade o 

Orr (66) reported average grades for veterans and nonveterans at Okla- 
homa Agricultural and Mechanical College in each of seven divisions during 
the academic years 1945-1946 and 1946-1947. He found veterans superior in 
all except engineering during 1945-1946 and superior in all except the 
Graduate School during 1946 -1947 . The number of students included in each 
comparison was not reported. The over-all average in 1945-1946 for male 
veterans was 2 .76; for male nonveterans, it was 2 39 In 1946-1947, the 
corresponding figures weres male veterans, 2. 53 1 male nonveterans, 2 42 
In the absence of a breakdown of these figures by college class, the dif- 
ferences must be interpreted with some cant ion 

Atkinson (5) studied the average grades earned by groups of veterans 
and nonveterans during the second semester of the 19^5 -1946 academic year 
and the first semester of the 1946-1947 academic year in the University of 
California at Los Angeles. As in Orr's study, the pooling of results from 
the four undergraduate classes makes interpretation somewhat difficult,, 
Atkinson found that in 1945-1946, the veterans excelled the nonveterans 
in all of the five divisions for which adequate data were available. In 
1946-1947, the nonveterans excelled in the group of science majors in 
Letters and Science; the veterans were superior in the other four groups. 
The largest of the ten differences amounted to about one fourth of a 
letter grade. 

Other less extensive studies of this general type may be summarized 
brief lys Eiemer (73) reported a study of University of Wisconsin second- 
semester grades in 19^5-1946 which showed veterans excelling nonveterans 
in each of the four class years and which showed over-all averages of 1.66 
for 4,201 male veterans and 1*57 for 1,296 male nonveterans; Welborn (103) 
reported that 109 veterans excelled 92 nonveterans in grades earned at 
Indiana State Teachers College, Terre Haute, during the Winter Quarter, 
1945-1946; Epler (30) reported that at Yanport Extension Center of the 
Oregon State System of Higher Education in 1946-1947 the average grade 
of 100 veterans was 2*58 while for 64 nonveterans the average was 2 ,47; 
Weintraub and Salley (102) found that the male veterans admitted to Hunter 
College normally a woman's college earned a first -semester average of 
242 as compared with the 2*36 average earned by the freshman women, 
although the women had noticeably higher high school averages; Davidson 
(26) reported that only 13 per cent of l62 veterans failed at the Uni- 
versity of Colorado while 18 per cent of 135 nonveterans failed; Taylor 
(88) found that veterans excelled nonveterans in English grades at the 
University of Southern California; Kvaraceus and Baker (58) reported that 
veterans excelled nonveterans in the final examination of an educational 
measurements course at Boston University; Deignan (27) found that 104 
veterans enrolled in Clark College of Clark University earned a higher 



10 ADJUSTMENT TO COLLEGE 



first-semester average in 19^6-19^7 than did 5^- male nonveterans; and 
Tepping (89) found that 373 nonveterans at the University of Colorado Ex- 
tension Center in Denver earned an average grade 0.26 of a letter grade 
higher than did the 6lO veteran students. Except for Tepping 's study, 
these studies are consistently favorable to the veteran group . 

An additional study which may be included in this group was carried 
out by Stewart and Davis (82). This study compared veterans of World War I 
with nonveterans at the University of Colorado during the period 1919-1926, 
The veterans were compared with nonveterans in the same division who had 
completed the same number of quarters during the period of the study. The 
general average for 251 male veterans was 77.9 while that for 263 male 
nonveterans was 78-7* I& the three divisions which included 50 r more 
of each group , veterans excelled slightly in the College of Engineering 
while nonveterans excelled slightly in Law and in Arts and Sciences, None 
of the differences was statistically significant. 

A second group of studies of veterans and nonveterans involved com- 
parisons between veterans and nonveterans who had been matched in various 
respects . 

A controlled comparison of veterans and nonveterans of World War II 
was reported by Love and Hutchison (6l) in November, 19^6. In this study, 
each of 1(A freshman veterans in the College of Education at Ohio State 
University was paired with a nonveteran on the basis of the Ohio State 
Psychological Examination, and, as far as possible, on the .basis of aca- 
demic program. It was not possible, however, to limit this study to male 
students. It turned out that the veterans earned an average grade of 2.^5 
as compared with 2.31 for nonveterans. The difference was not statistically 
significant. 

Gowan (40, 4l) carried out a doctoral dissertation at Iowa State 
College comparing the performance of veteran and nonveteran freshmen who 
entered college in the fall of 1945. This study included Ik6 veterans 
and 365 male nonveterans. Gowan found that in each of three divisions- 
engineering, science, and agriculture --the veterans excelled the nonveterane 
in academic grades during each of the three quarters. For the total group 
studied, the difference in grades was statistically significant at the 1% 
level during each of the three quarters. In a special analysis of first- 
quarter grades, application of analysis of covariance, taking ability into 
account, further enhanced the advantage of the veteran group. For the 
group of students as a whole, the nonveterans were superior to the veterans 
both in their mean high school average grade and in their mean American 
Council Psychological Examination score. The difference in the aptitude 
test score, however, was only about two points of raw score for the total 
group, and in the case of science students, the veteran group was slightly 
superior in this measure. Gowan' s results are relatively clear-cut; how- 
ever, only in the case of engineering students did the number of veterans 
and nonveterans exceed fifty in each group. 



THE FINDINGS AND THEBR IMPLICATIONS if 



Clark (l8) investigated the performance of veterans and nonveterans 
-in liberal arts, commerce , journalism, and speech education at Northwestern 
University, Included in his final population after balancing of ability 
between the two groups, were 562 veterans and 272 nonveterans who entered 
in 19^-60 A check on the equating of the groups indicated that veterans 
and nonveterans had been satisfactorily matched with respect both to high 
school standing in graduating class and in scholastic -aptitude score. 
Clark found that during the first quarter of residence, the veterans earned 
a grade point average of 3. 79 as compared with 3 .48 for the nonveterans. 
He noted that this difference in the average grade between the two groups 
was statistically significant. He also found that only 39 per cent of the 
nonveterans exceeded the median of veterans in grade point average, while 
64 per cent of the veterans exceeded the median of the nonveterans. 

A carefully controlled study of the academic achievement of veteran 
and nonveteran freshmen at the State University of Iowa was carred out by 
Garmezy and Grose (35) <, In this study, which was based upon freshmen who 
entered in September 19^6, the variables of sex, marital status, race, and 
college aptitude were held constant. Only single, white, male students 
were included In the comparison. All students in this study were enrolled 
in the college of liberal arts. It was found that it would be impossible 
to match the veterans and nonveterans with respect to age. The average 
grade earned by 2^5 veterans turned out to be 2 19 while the correspond- 
ing figure for the matching nonveterans was 2.09. This difference was 
not quite great enough to be statistically significant at the 5$ level. 
The findings of Garmezy and Grose with respect to age will be discussed in 
a later sect ion < 

In a study based on 170 veterans and 250 nonveterans enrolled in the 
same mathematics course at Princeton University, Brederiksen (33) found 
by use of analysis of covariance technique that there was no significant 
difference between veterans and nonveterans in grades when account was 
taken of ability differences. 

On the whole, the controlled studies, in which an effort was made to 
eliminate various obscuring influences which might affect veteran-non- 
veteran differences, showed an advantage for veterans over the nonveterans 
in academic grades. The relatively small size of the differences obtained, 
however, underlines the importance of careful control in the study of this 
problem. 

Perhaps the most dramatic of the studies of veteran students were 
those in which the academic performance of returning veterans was compared 
with performance of the same students during their pre- service educational 
career As it happened, however, a number of these studies did not 
directly take into account the possibility of an upward trend in students 1 
grades during their college career. As a result, the differences in 
average grade obtained in these studies cannot be taken at full face value. 



12 ADJUSTMENT TO COLLEGE 



In the fall of 19^6, Love and Hutchison (6l) reported comparisons 
"based on 102 veterans in the College of Education and 117 In the College 
of Agriculture at Ohio State University, They found that the veterans in 
the College of Education earned an average of 2^6 after the war as com- 
pared vith 2.03 before their war service . For the students in the Col- 
lege of Agriculture,, the postwar average was 2.86 as compared with a pre- 
war average of 2.25- They noted that only about ten per cent of the 219 
veterans did less well after the war than they did before. In another 
early study, Welborn (103) reported results for 107 veterans at Indiana 
State Teachers College In Terre Haute He found that the gain on the 
average amounted to about three fifths of a letter grade In his group 
about one fifth had earned lower grades after the war than before. Early 
in 19^7, President Day of Cornell University reported that returning 
veterans increased their average from a value of 71-5 or the last term 
before their service to 78 after their return to college (77), A study 
made "by Justman (54), of Brooklyn College students in the summer of 1946, 
Indicated that 66 per cent of a sample of 900 returned veterans had 
earned better grades during the first semester after their return than 
in the last semester before leaving the university for service. About 
73 per cent had achieved a better total record as veterans than they had 
In their prewar college worko In a study based on 400 veterans, again at 
the Ohio State University College of Education, Pultz (71) reported a 
gain in median grade amounting to about three fifths of a letter grade 

Hansen and Paterson (43) studied the prewar and postwar average 
grades of 265 veterans enrolled in the Junior division of the College of 
Science,, Literature, and the Arts of the University of Minnesota. All 
men In this study had been in the junior college before the war for at 
least two quarters and had completed two quarters In the Junior college 
after the war. The increase in grade point average for the 265 veterans 
amounted to .72 grade points. 

Deignan (27) compared a group of 60 veteran upper classmen, who had 
completed at least a semester's work in Clark College of Clark University 
before entering the service and at least one semester's work at Clark 
since their discharge, with various groups of nonveterans whose education 
had not been Interrupted by the war* He found that the 60 veterans showed 
a reliably higher average after their war service, the critical ratio 
being 5.00, Their gain in average grade amounted to 4.60 points; the 
largest gain shown by any of the other three groups was 2.71. One of the 
three comparison groups showed a slight decrease In average grade dixring 
the comparable period Deignan also found a statistically reliable dif- 
ference in grades earned during the early part of their college career 
In favor of a group of 30 men who entered college with the Interrupted 
veterans but who were allowed to finish college before interruption, as 
compared with his interrupted veterans. 

In a carefully controlled study of this question, Thompson and 
Pressey (91) used the records of 108 veterans who had completed at least 
four quarters at Ohio State University before entering service, who had 
completed at least three quarters after discharge from the service, and 



THE FINDINGS AND THIIR IMPLICATIONS 13 



had been graduated. The matching group of nonveterans had been graduated 
during the period 19^1-19^6. Matching was on the basis of age at" initial 
entry into the university, percentile on the Ohio State Psychological 
Examination, college, program "within the college, and cumulative average 
at the end of the first three quarters of college worko The median grade 
for the last three quarters was 2, 91 for the veterans as compared to 2.73 
for the nonveteranSo Thompson and Pressey note that for many of the 
veterans, the three quarters included in the measure of postwar perform- 
ance the difficult first quarter after return from the service. 

On the whole, the evidence regarding the improvement in academic 
grades following war service is distinctly favorable to the hypothesis 
that war service led in some way to improved performance of the student 
who returnedo It is possible, of course, that the students who actually 
returned to college after their service represented a selected group; 
that is, they were more highly motivated or more serious, perhaps, than 
the veterans who failed to resume their interrupted college careers. The 
difficulty in interpretation arising from the fact that average grades 
may tend to show some upward movement even if no interruption had occurred 
has already been noted. Only in the studies by Deignan and by Thompson 
and Pressey was specific attention given to this particular difficulty. 
Particularly on the basis of the latter study it would appear that the 
net gain for postwar as compared to prewar grades for these students 
would be relatively small. 

The studies reviewed thus far show a rather surprising degree of 
consistency in their general tendency in spite of the variation in pro- 
cedure involved Whether the comparison of veterans and nonveterans is 
based upon Judgments of administrators and faculty members, on compari- 
sons of mean grades of veterans and nonveterans in a particular adminis- 
trative unit, on controlled comparisons of the two groups, or on prewar 
and postwar performance of the same veterans, the evidence suggests that 
veteran status is associated with better-thanaverage academic perform- 
ance. 

However, when the magnitude of the difference which occurred typi- 
cally between veterans and nonveterans is the main consideration, the 
need for careful control is evident,, The assumption that various con- 
flicting determiners will cancel each other out is rather risky when 
differences of the size being investigated here are in question. In par- 
ticular, the need for eliminating the possible effects of variations in 
typical grades from one division of a university to another and during 
the four years of the academic program is evident,- Differences in ability, 
as measured by the usual admission data, are also sufficiently important 
to make their control necessary,, 

In the twenty entering freshman groups of the present study, when 
university, university division, sex, and class rank are controlled, but 
no account is taken of ability differences, the veterans excel the non- 
veterans in average grade In ten comparisons and are excelled by the 
nonveterans in the other ten comparisons. Part of the difference between 



14 ADJUSTMENT TO COLLEGE 



this finding and the typical finding of the earlier studies involving 
direct comparisons of grades earned by veterans and nonveterans may result 
from the exclusion from the present study of veterans who had had a sub- 
stantial amount of college training during their military service; it 
appears likely that more rigorous control of such matters as class rank 
and the division in which the students were enrolled would have reduced 
considerably the veterans 9 advantage in actual grades earned. The possi- 
bility that findings favorable to the veterans were more likely to be 
submitted for publication than those in which the findings were incon- 
clusive cannot be entirely ruled out in evaluating the published reports. 

When the evidence from other studies and the results of the present 
study are viewed in relation to each other, it appears that the veterans 
did do better college work relative to their ability than did the non- 
veterans. In the present study, as well as in most of the other con- 
trolled studies of this question, the veterans tended to have a slight 
advantage. It should be emphasized, however, that the advantage of the 
veterans, after allowance for ability differences, amounted to about 
one tenth of a letter grade, on the average, in 18 groups in the present 
study which included entering freshmen, (in two groups, letter grades 
were not available.) The largest difference found among these groups 
amounted to less than one third of a letter grade. 

Evidence from the present study regarding the interrupted veterans 
suggested that the gain shown by these students was slightly greater. In 
three of these college groups where letter grades were used, the typical 
advantage of postwar as compared to prewar averages amounted to perhaps 
one fourth of a letter grade. It should be kept in mind that In this 
estimate account has been taken of the gain shown by a control group 
whose college work was not interrupted by war service . 



Some Differences between Veterans and Nonveterans 
in Background and Attitudes 

From the outset, this study was designed to go beyond veteran-non- 
veteran differences in academic success and to provide a broader picture 
of these differences. Accordingly, a questionnaire was drafted, pre- 
tested, revised, and administered in the spring of the academic year 
19^.6-1947. A detailed description of the questionnaire and of its 
preparation, administration, coding, and analysis will be found in 
Chapter II. A copy of the questionnaire Itself is included In Appendix 
C. The present discussion is concerned mainly with difference between 
veterans and nonveterans in matters covered by the questionnaire. 

Veteran Characteristics 

The differences between veterans and nonveterans may be Interpreted 
more adequately If some attention Is given to the service careers, educa- 
tional history, and marital status of the veteran group Included In this 
study. 



THE FINDINGS AND THEIR IMPLICATIONS 15 



The great majority of the veterans in the typical selected college 
group had completed "between one and three years of service; on the average, 
only one fifth had completed three years or more of service. Almost 75 
per cent of these veterans had been overseas; most of those "who had over- 
seas service reported a year or more of such duty. About 80 per cent held 
enlisted ratings above corporal or seaman first class; less than 10 per 
cent held a commissioned rank. Interestingly enough, a higher proportion 
reported service in the Navy than in the Army; this result is plausible in 
view of the large proportion of young men in the Navy and the demobiliza- 
tion policies of the Army and Navy. Relatively few of the veterans in 
this study had served in the Marines, the Coast Guard, or Field Services. 

In a survey of 1,630 veterans in New York state colleges, Miller and 
Allen (6k) found that 6l per cent had served in the Army, and 33 per cent 
had served in the Navy. In their study, 56 per cent had served between 2k 
and 39 months. The typical rank was that of sergeant. When it is con- 
sidered that only 39 P*" cent of the students in Miller and Allen's study 
were freshmen, these results would seem to be reasonably similar to those 
of the present study. 

Almost 75 P er cent of the veterans reported that military service had 
increased their eagerness to attend college; less than 5 P er cent reported 
a decreased eagerness. Miller and Allen (6k) report an even higher figure 
on this point; 82 per cent reported that their service had increased their 
desire to attend college. These findings are in line with the observation 
of Kelly (57) that war seems to increase the demand for higher education. 

Over one third reported that service had increased their scholastic 
ability; slightly less than one fourth reported a decreased ability to do 
college work. Almost as many veterans thought they were doing lees well 
in college as a result of service experience as thought they were doing 
better. 

These evaluations of the effects of service may be compared with 
those obtained by Cottrell and Stouffer (21) from a cross-section study 
made in November, 19^5, based on high school graduates less than 25 years 
of age. Only about kO per cent of these soldiers thought helpful effects 
of Army experience outweighed the harmful effects; about 55 per cent con- 
sidered harmful effects predominant. Although no rigorous conclusions 
may b drawn from this comparison, there is some suggestion that the 
veterans in the present study may have taken a more favorable view of 
their service experiences than did the typical veteran. 

About one half of the veterans in the typical group had last at- 
tended school in 19^3 or 19^4; less than 20 per cent reported high school 
attendance during 19^5 or 19^6. Only a small proportion of veterans in 
this study had participated for any length of time in a college training 
program during their service, since such students were deliberately ex- 
cluded from the study wherever possible. Less than 15 per cent had taken 
on or more United States Armed Forces Institute (USAFI) courses during 
their service. 



16 ADJUSTMENT TO COLLEGI 



In this study, slightly more than 10 per cent of the veterans were 
married . This proportion is lower than that usually reported for veterans : 
a result which is presumably due, at least in part, to the limitation of 
the present summary to freshmen who entered in the fall of 1946. It may 
be noted here that Clark (18) reported that eight per cent of the fresh- 
man veterans in his Northwestern University study were married. 

The veterans who were included in this summary typically had enough 
service, enough responsibility during their service, and a long enough 
interruption of their educational careers to give a fair picture of the 
relationship between war service and various personal qualities. (A more 
detailed description of these veterans is given in Chapter TV. ) 

yeteran~Nonveteran Differences in Questionnaire Responses 

The statements in this section are based on results for male freshman 
veterans and nonveterans in twelve selected college groups. In general, 
in evaluating veteran-nonveteran differences consideration has been given 
to two lines of evidence. The difference between veterans and nonveterans 
in the median proportion of each group choosing each response provides 
one basis of comparison. Along with this, the consistency with which the 
differences between veterans and nonveterans were in the same direction 
in each of the twelve groups aids in identifying general tendencies. A 
degree of consistency defined as eleven or more differences in the same 
direction will arise by chance less than one time in 100. Figure 2 shows 
the median proportion of veterans and of nonveterans giving each of a 
number of questionnaire responses on which veterans and nonveterans dif- 
fered consistently from each other. In addition to the responses shown 
in Figure 2, four responses pertaining to worry (worry about finances, 
about inability to concentrate, about getting to know people socially, 
and about feelings of inferiority) also met the required standard of 
consistency. These responses, along with other responses pertaining to 
worry, are shown in Figure 3* 

Differences In Background Characteristics., When only freshmen are con- 
sidered, veterans are inevitably older, on the average, than nonveterane. 
Our findings indicate that the typical nonveteran entered college at IS, 
the usual college entrance age, and that the typical veteran entered at 
21 a difference of three years. Veterans varied considerably in age at 
entrance, as would be expected, since length of military service varied 
considerably. The nonveterans, on the other hand, showed very little 
variation in age 'at entrance 

With respect to other background characteristics it was found that 
veterans tend to have had more full-time work experience (other than 
military service) than nonveterans, and they were more likely to come 
from communities of between 2,500 and 100,000 than were the nonveterane 
Their fathers in general have had less formal education than is true for 
the nonveterans, a result which agrees with that found by Clark (18) for 
parents of Northwestern University veterans and nonveterans. According 
to the questionnaire findings, the family income at the time of high 



THE FINDINGS AND THESR IMPLICATIONS 



17 



EESPONSE 



Six months or more of full- 
time work between high school 
and service or college: 
Item 9(b), Category C 

Preparation for "better -pay ing 
job chief reason for attending 
college : 
Item 10, Category A 

College life, contacts, social 
pressure, or aid in career- 
choice chief reason for coming: 
Item 10, Category D 

Likely, but not certain, to 
enter occupation already chosen: 
Item 12, Category B 



32 
6 




MEDIAN PER CENT 

50 75 






Planning to take degree in less 
than the usual amount of time: 
Item 21 f Category A 

Relatively many hours per week, 
as compared to own college group, 
spent attending classes: 
Item 22 (a), Category C 

Two hours or more per week spent 

in organized extracurricular 

activities: 

Item 22 (d), Categories B, C 



ko 
B 




E 




Legend 1 . 

IBJMale veteran 

I IMkle nonveteran 



ITGURE 2 (PART l). MEDIAN BER CENT MAKING SELECTED BESPONSES TO QUESTIOMAIR1 
I03EMS AMONG VETERAN AND NONVETERAN STUDENTS. (MEDIAN VALUES, BASED ON THE 
TWELVE BASIC GROUPS.) 



18 ADJUSTMENT TO COLLEGE 



o m 



Generally behind schedule in g 

study assignments: 

Item 29, Category C 10 



Living in apartment or house 
(self -rented or owned): 
Item 30, Category E 



Father's income under $2,000 
in student's high school years: 
Item 43, Category D 



Father graduated from college: 
Item kk, Category C 



Hesitant to answer all the 

questions in the questionnaire 

frankly: 

Item 46 , Category B 



E 





Legend 

(( Male veteran 
I J Male nonveteran 



FIGURE 2 (PART 2). MEDIAE HER CENT MAJ8JNG SELECTED RESPONSES TO 

ITEMS AMONG "VETERAN AND NONVETERAN STUDENTS. (MEDIAN VALUES, BASED ON THE 

TWELVE BASIC GROUPS,) 



THE FINDINGS AND THEIR IMPLICATIONS 19 



school attendance was definitely lower than for nonveterans; this finding 
must be discounted to some extent, however, because of the general in- 
crease in income levels between the time the veterans had been in high 
school and that when the nonveterans were in high school. Veterans were 
more likely to be providing their own housing than were nonveterans; they 
were less likely to be living in dormitories than were nonveterans. When 
students were asked to evaluate their preparation for doing college work, 
the nonveterans were found more often than veterans to feel that they 
were very well prepared (The detailed findings on general background 
characteristics are presented in Chapter V.) 

Certain of these findings suggest that veterans within a particular 
college group came from families with less educational background " and 
lower income than did nonveterans in the same college. 

Mot i va t i onal Fac t or s . In view of the popular notion that veteran stu- 
dents are characterized by a greater seriousness of purpose and greater 
maturity, it will be of interest to examine the questionnaire evidence 
which bears on motivational factors. It is recognized that a question- 
naire is a rather crude tool to use for the investigation of* such sub- 
jective characteristics; but since we are only interested in comparing 
groups, where great accuracy of measurement is unnecessary, the tech- 
nique may be of some value . 

Some results which have to do with motivational factors may be sum- 
marized as follows : Differences between veterans and nonveterans with 
respect to vocational plans were generally slight although nonveterans 
were more inclined toward professions requiring graduate study. Veterans 
expressed certainty of being able to carry out their vocational plans 
somewhat more often than nonveterans, which is in agreement with the 
notion that veterans are more mature. Gowan (^0) also found somewhat 
greater certainty of vocational plans for veterans than for nonveterans . 
Nonveterans were slightly more likely to consider graduation from college 
essential to their vocational plans than were veterans. Veterans also 
assigned slightly less importance to college grades in relation to the 
kind of opportunities available to them after college; in view of the 
somewhat more extended credentials which a veteran could present to a 
prospective employer than the typical nonveteran, it is perhaps signifi- 
cant that the veterans nevertheless considered college grades almost as 
important as did the nonveterans . 

Nearly one half (k6 per cent) of veterans and about one third of 
nonveterans in the median group gave preparation for a better-paying job 
as their first reason for coming to college; this is a finding of some 
importance. The further result that almost one third of veterans and 
somewhat more than one third of nonveterans gave professional training 
as their chief reason for attending emphasizes the prevalence of the 
view that college is a means of getting ahead in the world. That veter- 
ans were likely to overemphasize career preparation was noted by the 
Educational Policies Commission (29) in 19^, by President Stoke (78), 
then at the University of New Hampshire, in 19^5, and by a number of 



20 ADJUSTMENT "TO COkLBGE 



other writers, including Bolte .(8), Byers (15.),, and .Humphreys (51). These 
predictions appear to be borne out by the findings of this ;study. (it 
may be noted that Katz and Allport ,(56) also found a heavy stress on col- 
lege as the path to success in their study at -Syracuse in 1926 . ) The long- 
range implications of this emphasis on economic gains are also of consider- 
able significance, Bowles (11), Thompson and .Pressey (91)7 Jordan (52), 
and Atkinson (3) i^ve stressed the importance .of a satisfying transition 
from college to an effective life-career as a major feature of the veterans f 
educational career, and Atkinson (k) has recently reported on the placement 
of veterans graduating in 191*7. President Henry of Wayne -University (kB) 
has pointed out that it is not always as obvious to the .student as to the 
educator that successful completion of college does not guarantee .profes- 
sional or economic success in line with the student ? s expectations. The 
recent rather pessimistic evaluation of the economic prospects for col- 
lege graduates given t?y .Harris (k6) makes the heavy stress placed by the 
students in our study on college as a means to ; personal economic advance - 
ment worthy of considerable thought. One might hope of course that these 
attitudes changed during the last three years of colleges the present study 
offers no evidence on this point* 

It has already been noted that veterans' were less likely than non- 
veterans to give ^preparation for entering a profession as their chief 
reason for attending college ; this finding may be attributed to the greater 
age of the veteran group on entering college. Nonveterans were also more 
likely to give reasons classed as "other" in this study social contacts, 
family pressure, postponing vocational choice, or coming because it was 
the "thing to do." 

The biggest veteran-nonveteran difference was found in plans for 
acceleration of the college program; about kO per cent of veterans and 
only about 10 per cent of nonveterans planned to graduate in less than 
the usual amount of time, Nonveterans were somewhat less successful in 
keeping up-to-date in their assignments than were veterans; the differ- 
ence was slight but consistent. 

Insofar as the questions permitted, the veterans did give evidence 
of greater seriousness of purpose than the nonveterans . They expre-ssed 
a definite desire to graduate in less than the usual amount of time, even 
though this presumably meant reduced vacation time. Veterans did not 
show as great an advantage with respect to certainty of vocational choice 
as would be expected from their reputed seriousness of purpose; it is 
possible, of course, that they were more aware of obstacles In the way 
of obtaining their objectives than nonveterans. Veterans were no more 
likely to report that they usually exerted strong effort on their courses 
than were nonveterans; it is conceivable, however, that they had a dif- 
ferent idea of what "strong effort" meant from the idea held by non- 
veterans ,> They were somewhat less likely to report that they were behind 
schedule on theias study assignments than were nonveterans. Even when 
allowance is made for complications In interpretation, however, the dif- 
ferences In motivation are generally small and plausible rather than 
large and spectacular. 



THE FINDINGS AND THEIR IMPUCATSONS 21 



Worries_ > What about the worries of veteran and nonveteran college students? 
Some early reports Indicated a certain apprehensiveness about the emotional 
problems of veterans, but experience with the veteran student indicated 
that, in the opinion of college faculties, the problems did not materialize 
except in isolated cases. Responses of students to various questionnaire 
items relating to worry and anxiety tend to corroborate this finding. For 
example, when a general question vas asked about tendencies to feel anxious 
or upset, the responses showed no tendency for veterans to worry more than 
nonveteransj if anything, they worried less . 

Figure 3 summarizes the results of a series of questions about 
specific sources of worry. The items are ordered with respect to their 
importance as sources of worry for nonveterans . It will be seen that by 
and large the most important sources of worry are related to academic 
problems concentration, getting accustomed to college study, and deciding 
what course of study to follow Ignoring financial worries for the moment, 
we find that emotional problems feelings of inferiority and nervousness- 
are the next most common causes of worry . Lower in the list comes worries 
about social relationships getting to know people socially and relations 
with members of the opposite sex| and far down in the list axe worries 
about .health, illness in the family, and, last of all, housing. 

In only one instance does a striking difference appear between the 
black and the white bars in Figure 3* veterans and nonveterans differed 
markedly only in what the questionnaire termed "making ends meet f inan^ 
cially " Veterans apparently worried considerably more about money than 
did nonveterans in spite of their allowances through the GI Bill. 

Veterans worried somewhat more than did nonveterans about inability 
to concentrate, which agrees with reports by Cottrell and Stouffer (21) 
and Crespi and Shapleigh (24) that veterans reported that war experi- 
ences had made them more restless,, They worried slightly more about 
illness or death in their family, which may reflect their greater age. 

Nonveterans worried a bit more than veterans about "deciding what 
course of study to follow," "feelings of inferiority," and "getting to 
know people socially ?t The first of these results is consistent with 
the finding that veterans are more certain of their vocational goals, 
and does suggest somewhat greater "maturity" on the part of the veterans. 
The tendency for nonveterans to worry more about feelings of inferiority 
may merely be a function of their presence on the campus with the veterans; 
an 18-year-old freshman, Just out of high school, might be expected to 
feel inferior in the presence of a group of combat veterans. The finding 
of Cottrell and Stouffer (21) that 5! per cent of veterans -under 25 
thought that Army service had increased their self-confidence, and the 
report by Crespi and Shapleigh (24) that 80 per cent of 199 Princeton 
veterans thought that service experiences had made them more independent 
are relevant here. Although this result tends to justify the observation 
expressed by Steele (8l) and Little (60) that veterans* preferences of 
various kinds may have had, adverse effects on the young nonveterans com- 
peting with them, the difference is so slight, however, as to support the 
view of Strom (84) that nonveterans as well as veterans benefit from going 
to school together. 



22 



ADJUSTMENT TO COLLEGE 



Source of Worry 



Median Per Cent Bothered Some or Very Much 
25 50 75 100 



Concentration 

Getting accustomed to study 

Deciding what course of study to 
follow 

Making ends meet 

Feelings of inferiority 

Nervousness 

Getting to know people socially 

Making up a deficiency 

Kelations with opposite sex 

Health problems 

Illness in family 

Strained personal relations 

Housing 




Legend 
Male Veteran 
Male Nonveteran 



FIGURE 3 . MEDIAN PER CENT OF VETERAN AND NONVETERAN MALE STUDENTS BOTHERED 
BY VARIOUS PROBLEMS. (MEDIAN VALUES, BASED ON THE TWELVE BASIC GROUPS*) 



THE FINDINGS AND THEIR IMPLICATIONS 23 



That veterans are concerned about financial problems agrees with a 
number of reports that the veterans -were finding it difficult to live on 
the allotments provided. In the fall of 19^6 , Cronbach (25) noted a study 
by Spurr which showed that average expenses of veterans were running above 
the amounts provided by the GI Bill; School and Society (79) also presented 
a summary of Spurr f s study. President Walters (99) told the American 
Association of Colleges early in 19^-7 that relatively few of the single men 
were able to live on their GI subsidy and that the financial situation for 
the married veterans was considerably worse. In the fall of 19^7> Little 
(60) reported that veterans at the University of Wisconsin were spending 
some $^0-65 P er month in. excess of their subsistence allowance, Aaronson 
(1) reported a questionnaire survey of students who failed to return to the 
University of Minnesota at various quarters of the academic year 19^6-19^7* 
Of those returning questionnaires, kl per cent of those who failed to return 
in the fall of 19^7 reported inadequate subsistence payments as a major 
reason for their withdrawal. The corresponding figure for the fall of 
was 31 per cent. Early in 19*1-8, Strom (85) reported on the basis of a 
nationwide survey that both married and single veterans had average ex* 
penses well above the GI subsistence allowance. It is apparent, then, that 
the GI Bill, however beneficial, did not typically free the veterans from 
economic worries. 

The general impression whioh one gains from the information about 
worry is that veterans and nonveterans are pretty much alike with regard 
to both amount of anxiety and things worried about. There is one notable 
exception, and that is worry about making ends meet. In addition, however, 
the tendency for veterans to worry more about inability to concentrate, 
and about illness or death in their family, and the tendency for non- 
veterans to worry more about feelings of inferiority, about choosing a 
course of study, and about getting to know people socially are sufficiently 
consistent in the twelve college groups to be statistically significant. 

Expenditure of Time. Another area of inquiry had to do with how veteran 
and nonveteran students spent their time how many hours per week were 
devoted to attending classes, studying, athletics, extracurricular activi- 
ties, social affairs, and so forth. It vae thought that such information 
might throw some light on the motives and interests of veterans and non~ 
veterans and thus might help to account for the difference In achievement 
relative to ability which, was found between veterans and nonveterans. 

The findings may be summarized briefly* Honveterans reported spend- 
ing about two hours more per week attending classes, laboratories, and 
other course conferences; presumably this difference (which was verified 
for two Institutions on the basis of information from transcripts) is due 
to the exemption of veterans from the usual physical education or mili- 
tary science requirements. On the other hand, veterans reported a sligjhtly 
greater amount of time spent In studying; the difference on the average 
amounted to sllgjhtly more than one hour per week, a finding which agrees 
witlx that of G-owan (ko) at Iowa State. It Is probable, of course, that a 
more elaborate plan for Investigating study hours would have yielded more 
accurate reeulte; and a more Intensive Inquiry Into study methods and 



24 ADJUSTMENT TO COLLEGE 



degree of application during study might have "been particularly significant 
in accounting for veteran-nonveteran differences in achievement. One com- 
parison of our results with those of Crawford's careful investigation (23) 
at Yale University, done nearly 25 years ago, may be mentioned here, Craw- 
ford found that Yale freshmen in the Class of 1929 reported a mean of 23-3 
hours per "week spent in study. The colleges in the present study most 
nearly comparable to Yale are Adams, where veterans had a median of 23.5 
hours per week and nonveterans had the same median, and Stewart, where 
veterans had a median of 22.7 hours per week and nonveterans had a median 
of 21.6. In many respects, this agreement in results may be considered re- 
markable . 

Early reports that veterans were participating very little in extra- 
curricular phases of college life were not confirmed by this study. Inso- 
far as athletic activities and physical recreation are concerned, veterans 
were apparently devoting only slightly less time than nonveterans to such 
activities. Veterans did show a consistent tendency to take less part than 
nonveterans in other organized extracurricular activities; 70 P er cent of 
veterans spent one hour or less per week as compared to about 60 per cent 
of nonveterans. About 15 per cent of veterans and about 20 per cent of 
nonveterans were devoting four hours or more per week to such activities. 

With respect to leisure activities, veterans showed a slight tendency 
to spend more time than nonveterans on social activities (but not enough 
more to indicate any appreciable number of free riders), and a slight 
tendency to spend more time on voluntary reading or study. Nonveterane, 
on the other hand, were more likely to spend more time on paid employment 
and in attending public lectures, concerts, and other cultural activities 
than were veterans. Time spent on bull sessions was about the same for 
both groups. 

The evidence reviewed above suggests that there Is some tendency 
for veterans to take a slightly more serious attitude than nonveterans 
toward their academic work. Apparently veterans had slightly more free 
time than nonveterans at their disposal, since the course load tended to 
be light, and they were less likely to have a part-time Job. They were 
somewhat more likely than nonveterans to spend their free time In studying 
and somewhat less likely to spend It on organized student activities, but 
the differences are typically small. 

Attitudes Toward the College Environment. In still another series of items, 
students were asked to Indicate their attitudes toward college courses, 
instructors, study facilities, and toward their university and the kind of 
education they were getting In Its more general aspects. The attitude of 
students toward various aspects of the college environment la a matter of 
considerable concern In evaluating the adjustment of veterans to college. 
Differences between veterans and nonveterans were typically so small that 
they can be treated together in the discussion that follows. 

Taken as a whole the findings Indicate that the typical student Is 
fairly well satisfied with his college. A substantial minority, comprising 



THE FINDINGS AND THEIR IMPLICATIONS 2 5 



perhaps one fifth of the students, appeared to "be somewhat dissatisfied 
and roughly the same proportion were rather enthusiastic about their college 
program. A majority thought that most of their instructors ere good 
teachers. Only about one student in ten thought that his study facilities 
were "quite unsatisfactory." Nearly all students were in the school or 
division of their choice . A majority of the students definitely preferred 
their own college to any other, although there was considerable variation 
in the proportion of students who preferred their own institution to any 
other. Thus, In spite of the competition for admission in the fall of 
191*6, most students actually in college the following spring were reason- 
ably well satisfied. This finding tends to provide indirect support for 
the view expressed by Russell (76), on the basis of a survey of officials, 
that substantially all qualified students were being accommodated by the 
colleges. Evidently the overcrowding did not result In a large number of 
students enrolled in one college or division but wishing to be In another. 
Of course, if many veterans refrained from entering college because they 
could not secure admission to the "name" college of their choice, as sug- 
gested by Clausen (19), the present figures would present a somewhat over- 
optimistic picture of the success of the colleges in meeting the needs of 
the total veteran group 

Students were also given an opportunity to suggest changes in the col- 
lege which would help them to get what they were after in a college educa- 
tion. About six out of every seven students made one or more suggestions. 
The most commonly given suggestions were for better instructors or courses, 
fewer (or different) required courses, changes In general requirements, 
especially with regard to grades and examinations, and more courses, more 
teachers, or more classrooms,, (This last type of suggestion undoubtedly 
reflected the crowded condition on college campuses at the time.) 

Howard (50), on the basis of questionnaire replies and letters of 
4,000 former University of Illinois students in the services in the late 
summer of 19W-, Justice (53), on the basis of questionnaires returned by 
49 veterans enrolled in 10 colleges, and Morris (65), on the basis of 
interviews with students in three New York City universities In the fall 
of 19^6 found about the same major suggestions as those reported in the 
present study. Better teachers, better instruction, and modification of 
course requirements were prominent In their findings, 'Since none of these 
studies considered the attitudes of nonveterans, however, the extent to 
which these responses were typical of all college students was not evi- 
dent. Vinocour (98) on the basis of Informal contacts with veterans 
argued that veterans were greatly dissatisfied! his conclusions were 
seriously questioned, however, by Ansley (2), Bush (13), and Coulton and 
Justman (22). The findings of the present study tend to agree with the 
more moderate views expressed by Vinocour 's critics 

Although Strom ? s (87) questionnaire study of November, 19^-7, is not 
strictly comparable to the present one, it may be noted that he found that 

inadequate courses and instructors, financial difficulties, and large 
classes were the most prevalent sources of irritation to veterans* He 
also found y on the basis of replies to a specific question, that about 



26 ADJUSTMENT TO COLLEGE 



one third of veterans in literal arts colleges, and about one fifth, of 
veterans in other types of colleges (except for teachers colleges) Judged 
that a majority of their courses did not pertain directly to their voca- 
tion. 

In the typical college group in this study 9 it was found that about 
one student in four expressed a need for fever or different required 
courses. This finding appears to be pertinent to the timely question of 
general and special education. General Education in a. Free Society (kj) , 
A Desig* 1 foy. General Education ($3)V ^ e Heport ctf the. President's Commls~ 
"sion on Higher Educat ion (97 ) / and more recently President ConanV s Educa- 
tion in a_ Divided World (20) have f ocussed attention on the need for a 
proper balance between specialized education directly connected with career 
plans and general education leading to individual development as a person 
and a citizen. Many writers called attention to the needs of veterans (and 
other students) for a realistic curriculum that would be clearly relevant 
to their plans and goals; among these vere Peatman (68), Little (60), the 
Committee on College Training of the Connecticut Beemployment Commission, 
cited in Hollis and Flynt 0*9), Feder (31), Bitchie (7*0, Kogers (75), and 
Hansom (72), The results of the present study suggest that for veterans 
and nonveterans alike, a substantial number were taking courses whose use- 
fulness was not apparent to them. Whether this resulted from an excessively 
narrow view of the kind of education they wanted or from instruction which 
did not make clear the importance of what they were expected to learn, or 
from a combination of these and other causes is not clear. The presence 
in our colleges of students who find some phase of their course program 
undesirable is undoubtedly familiar to the colleges; the fact that a sub- 
stantial proportion of students made specific comments in this matter 
and that overachievers were about as likely as under achievers to write 
these comments axe deserving of the thoughtful consideration of college 
administrators, curriculum coimittees, and faculty members. Besearch with 
respect to educational method, as urged by Gilmer (37) > and with respect 
to essential content, as urged by Brett (12), might well contribute to a 
narrowing of the gap between some students' conception of what they think 
they need and what they are getting as college freshmen. 

A few differences between veterans and nonveterans may be noted. 
The veterans tended to give slightly less favorable Judgments of their 
teachers than did the nonveterans _, and more of the nonveterans commented 
on the need for better guidance and placement services. Otherwise the 
two groups could scarcely have been more similar in the frequency with 
which they made various comments. This result is in general agreement 
with the findings of Gowan (hO) at Iowa State College, although the non- 
veterans in Gowan 's study were slightly less critical than the veterans. 

Attitude Toward the Questionnaire . One questionnaire item remains to be 
discussed. The last item in the questionnaire asked, "How did you feel 
about answering the questions contained in this questionnaire?" More 
than ninety per cent of the students, in a typical college group, checked 
the response, "Felt I could answer all frankly." A bit more than five per 
cent were "hesitant to answer all frankly," while less than one per cent 



THE FINDINGS AND THEIR IMPLICATIONS 27 



"felt it foolish to answer some frankly." (It was noticed that a number 
of the students mentioned specifically that they vere hesitant about answer- 
ing the item about father's income,,) Although the percentages were almost 
the same for veterans and nonveterans, nonveterans tended to "be more hesi- 
tant about answering all questions frankly. Apparently the students gener- 
ally accepted the questionnaire in good faith and did not try to distort 
their views. However, this result cannot be accepted as proof of the 
honesty of the responses; if a student wished to misrepresent himself he 
could do so on this item just as easily as on any other. Nevertheless 
the answers to the questionnaire items hang together generally in a way 
which in itself is evidence that students tended to answer the questions 
sincerely. 

Summary . The survey of veteran-nonveteran differences with respect to a 
variety of characteristics as assessed by mesns of the questionnaire has 
revealed that the similarities far outweigh the differences. Veterans 
are older, of course; this is by far the most clear-cut difference and the 
only one where there is very little overlap between the two distributions. 
There are suggestions that veterans have less family resources behind 
them: they worry more about finances, are more likely to have been employed 
full-time, and have less -well -educated fathers. There are also some indications 
that veterans are more "mature" than nonveterans: they are a bit more cer- 
tain of their vocational objectives, they worry less about deciding on a 
course of study, and they are less concerned about feelings of inferiority 
and about social adjustment. Some evidence that they have a greater "seri- 
ousness of purpose" is provided by the questionnaire responses: they study 
a little more than nonveterans and spend less time on organized extracur- 
ricular activities . On the whole, they attached less importance to college 
grades and to college graduation than the nonveterans. Motivation is 
slightly different ; veterans attend college in order to get a better- 
paying job somewhat more frequently than nonveterans, and less often to 
prepare for a profession. A somewhat lighter course load may give the 
veterans a slight advantage; but the veterans far more often than the non- 
veterans plan to accelerate their college program,, 

If these characteristics possessed by veterans more often than non- 
veterans are those which are associated with a tendency to "overachieve" 
in college, then the reasons for the veteran superiority in grades rela- 
tive to ability will be clearer. In the following section we will inquire 
into the problem of what characteristics are related to Adjusted Average 
Grade . 



Who Are the Over achievers? 



As in the extensive pioneer study carried out by Crawford (23) in 
1926, the plan of the present study called for relating information about 
students obtained from a questionnaire to their success in college. In 
this study, fir at -year average grades, adjusted to allow for ability dif- 
ferences, were generally used as the criterion measure. For veterans and 
nonveterans separately, in each of sixteen college groups, the mean Ad- 



28 ADJUSTMENT TO COLLEGE 



justed Average Grade earned by students giving each questionnaire response 
was computed. The use of Adjusted Average Grades is especially convenient 
because this procedure snakes it unnecessary to consider whether the group 
selecting a particular questionnaire response is superior or inferior in 
scholastic ability as measured by tests and high school average,, 

Two different methods were used in evaluating the significance of dif- 
ferences in mean AAG. One was an adaptation of the F-testj this test was 
applied to the veterans or nonveterans in each separate college group in 
order to determine whether the mean AAG of those choosing a particular 
response is significantly different from the mean AAG of students who chose 
other responses to an item,, The other approach to the study of significance 
was based on the consistency in the direction of the differences for veterans 
and nonveterans separately in twelve selected college groups. In this sum- 
mary, primary consideration has been given to the second approach^; the re- 
sults of the F-test are mainly useful in identifying characteristics associ- 
ated with overachievement in a particular college group,, 1 (See Chapter II 
and Appendix B2 for a more complete description of the methods employed in 
this part of the study, ) The specific evidence on which the following 
statements are based will be found in Chapters TV through X and in Appen- 
dix A, 



Background Factors 

With respect to background factors, the following groups of veterans 
earned higher grades relative to ability than did veteran students in 
generals 

1. Veterans who had last attended high school six years or 
more before entering college 

2. Veterans who were married . 

3. Veterans who had had three years or more of active duty, 
IK Veterans who had not served outside the United States, 

either during or after hostilities,, 

Although data on last year of high school attendance and on marital 
status were collected for nonveterans as well as veterans, the data did 
not permit any useful analysis for the nonveteran group, since virtually 
all the nonveterans entered college directly from high school and ware 
single . 

In an earlier study at the University of Pennsylvania, Prederiksen, (3*0 
found that freshman veterans who had been out of school longer earned higher 

^Relating the findings of the present study to previous work in this broad 
field is beyond the scope of this report , Summaries of this literature 
have been prepared by Harris (W-, Vj) and Borow (9, 10), 



THE FINDINGS AND THEIR IMPLICATIONS 29 



grades than those whose schooling had "been interrupted for a shorter time. 
The finding that married veterans did "better academically than unmarried 
veterans is in line with the results of several other studies. Early in 
19^7, Riemer (73) reported that married veteran men were making higher 
grades than single veteran men at all four undergraduate class levels in 
the University of Wisconsin. His study was based on. grades earned during 
the second semester of the academic year 19^5-19^6. He also reported, 
although on the has is of decidedly fewer cases, that married veterans with 
children, living in the trailer camp, earned somewhat higher average grades 
than married veterans without children, also living in the trailer camp. 
This latter finding was not consistent at all four class levels, probably 
because of the small number of cases available for study. Thompson and 
Pressey (91) found that at Ohio State University the average grade for kkk 
married veterans was 2.69 as compared with 2.48 for 1,58^ single veterans. 
Veterans having children excelled the married veterans by only .03 in the 
same study, with respect to average grade. Only in the study by Thompson 
and Pressey was specific attention given to the question of aptitude dif- 
ferences. Their finding that the married veterans had a median aptitude 
test percentile rank of 51 -^ as compared with 50.2 for the single veterans 
suggests that the difference was not primarily a matter of aptitude differ- 
ences. Epler (30) and Orr (66) also reported an advantage for married 
veterans as compared to single veterans in college grades. The possibility 
that the lumping of married students and single students enrolled in dif- 
ferent class years and also in different university divisions into a single 
over -all comparison may have obscured the basic differences between the 
two groups makes rigorous interpretation of x hese studies difficult. 

The finding that the "stateside" veteran tended to overachieve would 
seem to contradict the rather plausible hypothesis that the superior veters 
students would be those who had the broadest experiences, travel, and 
combat . 

With respect to age, the following groups were found to excel their 
fellow students in Adjusted Average Grades 

5. Veterans who were 23 years of age, or older, when they 
entered college as freshmen. 

6. Kfonveterans who were 17 years of age, or younger, when 
they entered college as freshmen. 

The finding that for veterans greater age and for nonveterans lesser 
age is associated with high MG seems paradoxical at first. The findings 
are reasonable, however, when viewed in the light of the different selec- 
tive factors that were presumably operating. For nonveterans, the rela- 
tionship is undoubtedly a reflection of the usual finding that scholastic 
ability and age are negatively correlated in a secondary school population. 
This negative correlation results from a tendency to accelerate the best 
pupils and retard the poorest; the youngest nonveterans are continuing to 
show in college the same characteristics which caused their arrival in 



30 ADJUSTMENT TO COLLEGE 



o 

college at an earlier age. Among veterans, selective factors of a dif- 
ferent sort are presumably operating. The veteran who enters college at 
23 would normally be 27 when he graduates j such a student would probably 
not enter college at all unless -he were very strongly motivated to do aca- 
demic work or had unusual Incentives to go to college. Thus a process of 
self -select Ion may be presumed to account for the relationship between age 
and AAG for veterans 

Three additional groups which tended to overachieve weres 

TO Nonveterans who came from a city of over 100,000 population. 

8. Students who came from a family whose Income was under $2,000 
a year. 

9. Students (in seven out of eight possible comparisons In three 
universities having relatively many private school graduates) 
who had attended a public school. 

What kind of student earns high grades relative to ability? It Is 
difficult to write a single generalization which covers all the diverse 
characteristics mentioned above ; but it is hard to escape the Impression 
that the over achieving student Is the one who has had the most to overcome 
In the way of economic and social barriers to college . More will be said 
about this in a later section. 



Some Factors Belated to Motivation 

Although knowledge of a student's motivation obtained from a question- 
naire has obvious limitations, the following groups were identified as 
overachievers from their questionnaire responses. 

1. Students who went to college because they felt that a college 
degree was necessary in order to enter a chosen profession. 

2. Students who planned to enter a profession requiring college 
graduation or graduate study. 

3. Students who were almost certain that they would do the kind 
of work that they planned to do. 

k. Students who were majoring in a school or division which 
represented the field of their first choice. 

So far as the above generalizations are concerned, they agree In 
Indicating that the student who earns high grades relative to his ability 
tends to be one who plans to enter a profession which requires college or 
graduate training and who is reasonably certain of his vocational objective, 
In view of the Importance of college grades for admission to graduate pro- 
grams, the tendency to overachieve may represent a realistic attitude 
toward the problem on the part of these students,, 

2 
The tendency for younger students to do well In college is brought out 

clearly in a recent review of the literature In Educational Accelera- 
tion by Pressey (70) - 



THE FINDINGS AND THEIR IMPLICATIONS 31 



Worries 

Amount and kind of worry also appears to be related to Adjusted Average 
Grade. The following groups of students tended to achieve higher grades 
relative to ability than did their fellow students. 

1. Students who said that they seldom or never felt worried or 
anxious and upset. 

2. Students who were bothered little or not at all about being 
unable to concentrate. 

3. Students who were bothered little or not at all about getting 
accustomed to college study. 

4. Students who were bothered little or not at all about trying 
to make up a deficiency in preparation for some course. 

5. Students who were bothered little or not at all about trying 
to decide what course of study to follow. 

6. Students who were bothered little or not at all about making 
ends meet financially. 

7. Students who were bothered little or not at all about feelings 
of infer iority, inability to compete with others or to live up 
to their own standards. 

8. Students who were bothered some, or "bothered very much, about 
getting to know people socially. 

By and large, the overachieving student is one who is relatively free 
from worry. This generalization holds both for generalized worry without 
regard to nature of the thing worried about, and for a number of specific 
sources of worry. In view of the fact that the most common source of worr^ 
was the student's own scholastic adjustment, one should not assume that 
freedom from worry is a causal factor; if anything, it may be the other 
way around. The only exception is that worry about social relationships 
is associated with high rather than low grades relative to ability; this 
finding is in line with the possible tendency for overachievers to be 
less socially inclined than under achie ver s . 

The fact that freedom from financial worry is associated with effi- 
cient use of scholastic abilities may have some significance as far as 
scholarships and other financial aid are concerned. It should be noted, 
however, that the relationship is not very marked, possibly because for 
some students it may involve adverse economic circumstances, for other 
students merely poor planning of expenditures. In some instances, since 
large scholarships are awarded to students of high promise, this selec- 
tion may lead to relative freedom from financial worry as well as academic 
success . 

Expenditure of Time 

With respect to variations in how students spent their time, the fol- 
lowing groups of students were found to earn high grades relative to their 
scholastic ability s 



32 ADJUSTMENT TO COLLEGE 



1. Students who spent a greater amount of time than the average 
student attending classes, laboratories, and other regularly 
scheduled course conferences 

2. Students who spent a greater amount of time studying- than the 
average student. 

3. Students who spent a moderate amount of time in bull sessions. 
k. Students who spent less time than the average student in social 

activities and recreation dates, parties, movies, etc. 
5. Students who attended evening lectures given "by visiting lec- 
turers or local faculty members, but not required "by any spe- 
cific course, more frequently than the average student. 

The evidence from the study of how students spent their time thus 
shows that the academic overachiever was likely to take more courses, 
generally studied more than the average student y attended more evening 
lectures, and spent less time on such "frivolous" activities as going to 
parties and movies. He did engage in bull sessions, but in moderation. 
He seems to have been a pretty serious sort of fellow with definite aca- 
demic interests. 

Some Ambiguous Finding^ 

The questionnaire was filled out by the students in the spring of 
19^7, near the end of the academic year,, By this time the students knew 
their first -semester grades or, in colleges with a quarter system, their 
grades for the first two quarters. In addition, they knew their grades 
on midterm examinations and quizzes- for the final term. This knowledge 
must have had some effect on the way students answered certain of the ques- 
tionnaire items, and, since grades correlate rather highly with Adjusted 
Average Grade, it had an effect on certain of the relationships with which 
we are concerned in this section. It should be stressed that although 
Adjusted Average Grades have a correlation of zero with the predictors 
used in computing them, they correlate highly with the measure of success,' 5 
For certain kinds of items, such as those concerned with year of birth or 
amount of time spent in attending classes and labs, the knowledge of grades 
would presumably have no effect on the answers given . But on other typec 
of items the knowledge of grades is possibly a very important factor. For 
example, when a student is asked how well he was prepared, by virtue of 
previous education and experience, for getting the most out of his college 
course, the failing student may be tempted to think that it wao poor 
preparation that caused his trouble while the high ranking student may 
generously assign some of the credit for his achievement to his (presumably) 

Tor example, if the multiple correlation of the predictors with grades is 
.65, the Adjusted Average Grades will correlate .76 with the original 
grades. Even where the predictors correlate .71 vith grades, the Ad- 
justed Average Grades correlate ,JO with the original grades,, In general, 
the sum of the squares of the two coefficients will equal 1.00, which is 
reasonable, since Adjusted Average Grade includes all the variance in 
grades which is not accounted for "by the predictors. 



THE FINDINGS AND THEIR IMPLICATIONS 33 



good preparation., This tendency to rationalize may "be especially serious 
with the failing student, who may embrace any plausible suggestion as the 
reason for his poor performance. 

There is obviously no way to tell with certainty for which items this 
kind of rationalization was an important factor and for which items it 
was not. One can only look at the item and the results and exercise his 
best judgment. The following Items are related to Adjusted Average Grade 
but, in the judgment of the writers, may at least In part be reactions to 
success or failure rather than characteristics which have contributed to 
producing good or poor work In any case, the following groups tended to 
have better-than-average grades relative to ability. 

1, Students who believed that they were very well prepared, by 
virtue of previous education and experience, for getting the 
most out of their courses. 

2, Students who believed that they must have a college degree in 
order to do the kind of work they were planning to do. 

3, Students who believed that college grades would be very impor- 
tant in relation to the kind of opportunities available after 
college . 

k Students who found It less difficult to keep up in their work 

than they had expected. 
5. Students who claimed that they usually exerted strong effort 

to do good work in their courses. 
6 Students who claimed that they usually had their assignments 

done before they were due 

7. Students who believed that worry had not interfered at all 
with their college work. 

8. Students who claimed to be really interested in a majority of 
their courses. 

9, Students who reported that they were enjoying their studies as 
much as or more than they had anticipated. 

10 , Students who claimed that they seldom or never felt that the 
things they were studying in college were not worth the time 
spent on them. 

11. Students who felt that on the whole they were very well satis- 
fled with the kind of education they were getting. 

12,, Students who believed that most or all of the faculty members 
who had taught their courses were good teachers . 

13. Veterans who felt that their experience while in the service 
made them more eager to go to college. 

Ik, Veterans who believed that their military service experience 
had increased their ability to do good scholastic work In col- 
lege . 

15. Veterans who believed that they were doing better in their 

college work than they would have done if they had gone on with 
their schooling instead of going into the service. 

Although it would perhaps be useful to a counsellor to know that these 
attitudes were associated with earning good grades relative to ability, 

the Interpretation of these character Is tics is made ambiguous by the possi- 



34 ADJUSTMENT TO COLLEGE 



bility that their relationship to Adjusted Average Grade might be much less 
close if the student did not know how well he was succeeding in college. 

The GI Bills Who Goes to College? 

In 1938; Learned and Wood (59) in The Student and Els Knowledge, re- 
ported that among 4,000 students accepted by Pennsylvania colleges, there 
were nearly 1,000 who made test scores as seniors in high school lower than 
the average high scnool senior who did not go on to college. Moreover, 
among the seniors who did not enter college, there were some 3,000 who 
scored above the average of those who did enter. That many high school 
graduates of high ability do not enter college has "been demonstrated by a 
number of other studies as well. 

The Report of the Presidents Commission on Higher Education (97) , 
the Harvard Report (^7), and Education in a Divided World (20) have re- 
cently called attention to the relevance of such findings to the general 
question of equality of educational opportunity. Science, the Endless 
Frontier (l4) emphasized their relevance to the problem of securing an 
adequate supply of professional and technical personnel. 

What would happen if a substantial number of students who otherwise 
could not have attended college were given financial aid which made such 
attendance possible? How would such students compare with the usual stu- 
dent "body of the college? Certain findings of the present study have a 
bearing on these questions. These findings have to do with the character- 
istics of students who were enabled to attend college by the educational 
provisions of the GI Bill. 

The provisions of the GI Bill undoubtedly changed the educational 
plans of veterans in a variety of ways. Some were enabled to go to col- 
lege who otherwise would not have attended; others attended a college or 
iniversity outside their home community rather than one at home, or shifted 
from a less expensive to a more expensive college. In the questionnaire 
employed in this study, several items were included in order to investigate 
;he influence of the GI Bill on college attendance. 

The veteran students were asked, "Do you think you would have come 
io college after completing your military service if the financial aid 
xrovided by veterans' benefits had not been available to you?" The four 
esponsee used in the analysis were (A) yes, I am quite sure I would have 
;ome anyway; (B) I probably would have come, but I r m not sure; (C) I might 
tave come, but I probably would not have come; and (D) no, I am quite sure 
! would not have come to college. 

In a typical college group about 20 per cent of the veterans were 
.pparently influenced appreciably by the GI Bill (or other veterans 1 ben- 
'its) in their decision to enter college. About ten per cent of the 
eterans definitely would not have come and another ten per cent probably 
ould not have done so without such financial assistance. The proportion 



THE FINDINGS AND THEIR IMPLICATIONS 35 



In these categories varied widely, however; in some colleges only one in a 
hundred was influenced by the veterans' benefits, while in others more than 
a third of the veterans definitely or probably would not have come to col- 
lege without the GI Bill. The over -all value obtained in this study agrees 
well with the findings of a survey conducted by Strom (86), reported in 
April, 19^8. In his study the proportion of veterans who reported that 
they would not have attended without federal aid was approximately 20 per 
cent in state universities, private and men's schools, municipal universi- 
ties and liberal arts colleges. About 5 per cent, in addition, reported 
that they did not know whether or not they would have attended without 
federal aid. Strom reported that the percentage who would not have at- 
tended among students enrolled in junior colleges and teachers colleges 
was about 30 P er cento It should be noted, however, that Strom's study 
sampled all veterans, not freshmen only. 

In interpreting these proportions, it should be kept in mind that 
the GI Bill, however generous and unprecedented its provisions, did not 
eliminate but only lowered financial barriers to higher education. As 
President Stoddard of the University of Illinois (83) recently pointed 
out, in another connection, a large part of the cost of higher education 
to a student is the income foregone by not holding a full-time job. A 
comparison of GI benefits with the earnings which bright and healthy young 
men could have earned in industry indicates that most if not all veterans 
made some financial sacrifice while attending college. 

What kind of student was brought to college by the GI Bill? The evi- 
dence favors the conclusion that they were quite similar to their veteran 
student classmates, who would have gone to college anyway, in academic 
performance relative to ability. The slight difference that was found 
showed that those veterans who definitely would not have attended college 
without the GI Bill were superior in Adjusted Average Grade. So far as 
measures of ability are concerned, differences tended to be slight and 
did not consistently favor any one category, although there was some indi- 
cation that those who would definitely have gone to college even without 
the GI Bill were slightly higher in ability measures than the remaining 
veterans. It was further found, using data from the university which con- 
tained the largest number of veterans who would not have gone without GI 
aid, that the relationship between ability measures and freshman average 
grade Is essentially the same for both veteran groups those who certainly 
or probably would have gone and those who certainly or probably would not 
have gone without the GI Bill. The same regression equation was found to 
be appropriate for use with either group. 

Using data from two colleges, an investigation was also made of the 
differences between the two kinds of veterans the "would have gones" and 
the "would not have gones" with respect to other characteristics assessed 
by the questionnaire. Differences which were statistically significant 
(at at least the 5$ level) at both institutions are reported here. It 
was found that students who were Influenced by the GI Bill to attend col- 
lege were older, had been out of school longer, had had a longer period of 
military service, and had served overseas longer. Their fathers had had 



36 ADJUSFMBNT TO COLLEGE 



le-ss formal education- and probably were less -well-off financially. These 
veterans were more likely to "be married and, were less likely to "be planning 
to enter- a profession than were the other veterans. There is, to "be sure, 
considerable overlapping "between the two groups with respect to most of 
these; char act eristics. When veterans who would not have gone were compared 
with mal nonveterans, thie differences were ordinarily greater than those 
de.sG,ribd above and in the same direction^. The veteran-nonveteran differ- 
ences 1 , tfcus appear to be enhanced when oaly the- "would not have gone" 
veterans are considered., 

When veterans were asked, ".. .do you think you actually would have 
goose to college if you hadn*t entered military service?" their responses 
were distributed in about the same manner as for the item relating to the 
(JI Bill. The relation of these items to each other and to MG suggests 
that those students who would not have gone without the GI Bill tend to 
be the same students as those who would not have gone if they hadn't 
entered military service . (The group of students who would have foregone 
their prewar college plans if there had been no GI Bill when they were 
discharged were evidently balanced by a group who had not planned to go 
before the war but who would have attended after military service even 
without the GI Bill. Both of these atypical groups were presumably small.) 
It is interesting, however, that for this phrasing of the question, a sig- 
nificant association exists between the "would not have gone" response and 
superior MG. 

No distinction has so far been made between disabled veterans drawing 
benefits under Public Law 16 and the veterans drawing benefits under other 
laws, principally Public Law ^46 (the GI Bill). The veterans were asked, 
in the questionnaire., to state whether or not they were drawing benefito 
and, if they were, under what law or laws. Fewer than five per cent of 
the veterans drew benefits under PL 16. Wo significant difference was 
found between the disabled veterans and the other veterans in Adjusted 
Average Grade. 

Fewer than five per cent of the veterans were drawing no veterans 
benefits. Since men vho served in the merchant marine and field services 
were considered to be veterans in this study ; it ia clear that a very 
small proportion indeed of those eligible were not drawing benefits. It 
appears likely that at least a few of the eligible veterans were Having 
their educational benefits to use for later professional training. Those 
not drawing benefits appeared to be slightly superior In mean Adjusted 
Average Grade, although their superiority cannot be said to be significant 
in a statistical sense,, 

When it was first suggested that a study be made of some of the 
effects of lowering economic barriers to higher education through the 
provisions of the GI Bill, considerable thought was given to procedures 
to be used. In these discussions it waa agreed that it would be desir- 
able to distinguish three kinds of veteran students: (l) those who were 
economically able to go to college and who would have gone without subsidy; 
(2) those who could not have gone to college, because of economic factors, 



THE- FINDINGS AND THEIR IMPLICATIONS* 37 



without- the educational benefits of the: GI Bill; and. ('3) the "'free rtdejns." 
The free risers were the veterans who, took advantage; of the GI Bill to at- 
tend college for a JQBT paarely for the- good times and amusement which could 
be associated with such aiL experience:-., Ho satisfactory method was found fear 
identifying; the free riders, however;- and as has been seen,, no use was made, 
of such a group in the analysis of thee problem. There is, nn< reason to 
believe- that the number <oEf free ridera? among the veterans"- was any greater 
than among the nonvetersns . At any rate, it has been f onn<t that whan a 
system of federal scholarships is instituted by the femoral government for- 
administration by the colleges (whicfc is in effect, what happened with the 
GI Bill),, a substantial group of students who could, not otherwise have at- 
tended was matriculated. They proved to* be just about aa able as 1 the stu- 
dents who could have paid their owtuvay, and if anything they earned better 
grades relative to their ability than did the students with means to attend 
college and all. this with possibles free riders included. Ability to pay 
for a college education is obviously not perfectly correlated with ability 
to achieve the acadeaic goals of college. The results indicate that th 
veterans who needed financial a Distance to attend college could and did 
make proper use of the opportunities afforded them by the GI Bill. 



Why Did the Veterans Excel? 

Ident if ying Relevant Characteristics 

We have looked into the various ways in which veterans were found to 
differ from noaveterans, and we have found certain characteristics which 
are associated with the tenctency to earn high grades relative to ability, 
It is now appropriate to consider whether or not any of these findings 
can be combined in such a way as to account for veteran-nonveteran dif- 
ferences in grades relative to ability. 

One particular kind of item was thought to be especially relevant 
in accounting for veteran-nonveteran differences in grades relative to 
ability. In this kind of item, there is a clear relationship between item 
responses and Adjusted Average Grade; and at the same time there is a 
marked difference between veterans and nonveterans in their pattern of 
response. Suppose, for example , that on some particular item veterans 
are considerably more likely than nonveterans to choose the response 
associated with superior Adjusted Average Grade. We can then deduce that 
if veterans and nonveterans were alike in the quality identified by this 
item, the advantage of the veterans in Adjusted Average Grade would be 
lessened. Similarly, if veterans were much less likely to choose the 
response associated with inferior Adjusted Average Grade, the item would 
again tend to account for veteran superiority in the over-all results. 
It is of course possible that an item would function in the opposite 
way; that is, the results might indicate that veteran superiority would 
be increased if it were controlled. It need hardly be added that such 
findings cannot be interpreted mechanically; a positive result for an 
item merely indicates that it may throw light on the question of veteran- 



38 ADJUSTMENT TO COLLEGE 



non-veteran differences In Adjusted Average Grade. In practice, a definite 
procedure was worked out for identifying the more promising items in ac- 
cordance with these concept ions ; the steps In this procedure are described 
In detail at the end of Chapter II. As part of the process of Identifying 
items for further scrutiny, a simple statistical test was made in order to 
minimize the role of chance variations in selecting items. This test was 
based on the consistency with which the veterans showed a greater tendency 
than the nonveterans to select the responses associated with high Adjusted 
Average Grades. (In some Items,, the responses associated with low Adjusted 
Average Grades were made the basis of the comparison. ) 

Now we are ready to look at the results. For how many items were sig- 
nificant results found? One item was found to be significant at the 1$ 
level, and three additional Items were found to be significant at the 5$ 
level. Since the proportion of significant items is only slightly greater 
than the number which, in view of the number of items tested, would be ex- 
pected by chance, the significance of these four items must be considered 
as doubtful. Nevertheless it may be worth-while to examine the four Items 
which were picked out by the significance test. 

Some Characteristics of Possible Significance 

Amount of Time Devoted to Class Attendance* The Item which was found to 
be significant at the 1$ level Is Item 22 (a), which asks for number of 
hours per week spent In attending classes, laboratories, and other regularly 
scheduled course conferences. This item provides a measure of course load 
in which laboratory courses would be given greater weight than would usually 
be provided by "credit" hours * The finding is that students who spend rela* 
tively many hours attending class meetings tend to be above average in AAG, 
and that nonveterans possess this characteristic more frequently than 
veterans. Eleven subgroups were found in which both veterans and non- 
veterans were above average in mean AAG; ten of these subgroups contained 
relatively more nonveterans than veterans, and in the eleventh group the 
.percentages were the same,, This amount of consistency In direction of dif- 
ferences, or signs, would be expected to occur by chance lees than once in 
a hundred times,, 

Our only highly significant item, then, leads us to the expectation 
that nonveterans would excel the veterans in grades relative to ability, 
which of course Is contrary to our actual findings. The reason may be 
attributed to extraneous factors which influence the course load of veteran 
students. The general tendency for heavy course load to be associated with 
high AAG may be attributed to selective factors; only the more able and 
highly motivated students elect to take an unusually heavy cours'e load. 
The course load of the veteran student tends to be lighter because, at 
most universities, he is excused from the usual military science or 
physical education requirements rather than because he is less able or 
less strongly motivated. Probably we should therefore discard our highly 
significant finding as being an artifact produced by these modified uni- 
versity regulations and thus Irrelevant to the Issue. 



THE FINDINGS AND THEIR IMPIJCAT1ONS 39 



Keeping Up-to-date in Assignments. One of the items significant at the 5$ 
level is Item 29, which asks, "In general, how well do you keep up-to-date 
in your study assignments?" Fourteen subgroups were found in which "both 
veterans and nonveterans were below average in AAG. In all twelve colleges 
included, the subgroup containing students who said they were usually behind 
in their assignments was below average, and at two colleges students who 
just kept up-to-date were below average. Out of these Ik subgroups of 
below average student s ; the percentage of veterans was smaller than the per- 
centage of nonveterans in twelve groups; this sign test result is significant 
at the 5$ level. We find, then, that the characteristic associated with 
underachievement not keeping ahead in completing assignments --is possessed 
more frequently by nonveterans than by veterans. This finding might, then, 
help to account for the observed fact that veterans do tend to earn higher 
grades relative to ability than nonveterans. It is unlikely, however, that 
this is anything more than a symptom of some more important underlying 
determiner . 

Worry _about Concentration, Thirteen subgroups were found in which both 
veterans and nonveterans were above average. The characteristic was "being 
bothered little, or not at all, about being unable to concentrate" (but at 
one college students who said they were bothered "some" were also above 
average). In eleven of these 13 subgroups the proportion of nonveterans 
was greater than the proportion of veterans , The finding is significant 
at the 5$ level. Here we have an item which would lead to the expectation 
that nonve-terans , rather than veterans, would excel in Adjusted Average 
Grade, since relatively more nonveterans than veterans are free from worry 
about concentration (the characteristic associated with high AAG). We mus' 
conclude that veterans were superior in AAG in spite of greater worry abou 
concentration^, and that the observed superiority of veterans would have be< 
greater if they were not handicapped by this difficulty. 

Worry about Getting to Know People Socially Here we find eleven superior 
subgroups The characteristic associated with superiority is "tendency to 
worry about social relationships." In ten of the eleven subgroups there 
are relatively more nonveterans than veterans, and the sign test shows 
significance at the 5$ level. Again we find results which would lead to 
the expectation that nonveterans would excel in grades relative to ability, 
insofar as this evidence is concerned. However, it is doubtful that much ' 
stress should be placed on this finding since the relationship between 
worry about social relationship and overachievement is presumably rather 
indirect. 

As was previously stated, the number of items found to be significant 
is only slightly greater than the number which would be expected "by chance 
alone. Therefore the significance of the four items must be discounted. 
One of the items apparently involves an artifact which further detracts 
from its significance. The remaining three results, which at best are 
significant at only the 5$ level, do not consistently favor either veterans 
or nonveterans. We are left with no reasonable hypothesis from our statis- 
tical approach to the problem. 



40 ADJUSTMENT TO COLLEGE 



On "Giving the Veterans a Break" 

It has been suggested that the superiority of veteran students Is 
simply due to a tendency for Instructors to give the veterans a "break in 
assigning grades According to this hypothesis, the teacher was slightly 
more lenient with veterans. If the grade was on the borderline "between a 
B and a B+, there was a greater tendency to call It B+ for a veteran than 
for a nonveteran, or a penalty for lateness in turning in a term paper was 
more likely to be remitted for a veteran than a nonveteran student. Such 
an hypothesis would lead one to expect a slight rather than a big differ- 
ence, and would lead to the expectancy that the difference would he more 
noticeable for older veterans (who to the instructor are more obviously 
veterans) than to younger veterans. The facts are in reasonably good 
agreement with such expectations but may be due to other causes. 

Unfortunately no satisfactory way to test the hypothesis has been 
found. The matter has been discussed with a number of college teachers, 
and their opinions vary. Although there are no safe grounds on which 
either to accept or reject the hypothesis, there are two reasons for 
believing that faculty bias played a very minor role in the results of 
this study: (l) The widespread use of objective examinations in under* 
class courses makes the hypothesis less tenable, since subjective opinion 
is Involved only to a minor degree, (2) The large classes so often found 
In underclass teaching makes it unlikely that instructors would know very 
many of the students by name. Knowing the students by name would appear 
to be an essential condition for the hypothesis to hold. While no direct 
evidence is available, the authors are inclined to believe that relatively 
little importance can be attached to any tendency for teachers to give the 
veterans a break In assigning grades. The fact that there IB no tendency 
for veterans in colleges with small classes (where students would be more 
1-fkely to be known by name) to excel more than in colleges with large 
classes provides Indirect evidence against the hypothesis. 

The Vexing Question of Age_ Differences 

The role of age differences in accounting for veteran-nonveteran dif- 
ferences in academic success Is worthy of serious consideration. In deal- 
ing with this variable a difficult technical problem presents Itself. If 
the comparison is not limited to a single class year (for example, fresh- 
man), the obtained relationship between age and average grade may merely 
reflect differences In the typical grades earned by freshmen, sophomores, 
etc. On the other hand,. If the comparison Is limited to a single class 
year, only the youngest members of the veterans' group will overlap In 
age with the oldest members of the nonveteran group . In a sens, then, 
age and veteran status become the same thing. The older students are the 
veterans, and the younger students are the nonveterans, 

Several studies of age, in which the matter of class rank was not 
controlled, have been reported,, Shaffer (80) by matching on age and 
allowing everything else except sex to vary, was able to ohow that male 



THE FINDINGS AND THEIR IMPLICATIONS 41 



nonveterans excelled male Teterans at every age level. This was true al- 
though, veterans excelled nonveterans on the whole "by a margin of .01 grade 
points. Owens and Owens (67) found a correlation of 37 between age and 
grade point average for 194 male veterans at Vinona State Teachers College. 
They also found that age contributed to the prediction of success when com- 
bined vith American Council Psychological Examination Scores. 

Pultz (71) reports figures which show a clear-cut upward trend in the 
grades earned by veterans in the Ohio State University College of Education 
as successively older age groups were considered. Thus, 97 veterans in the 
17-19 year age group earned a median average grade of 2.16 as compared with 
a median grade of 2*87 for 22 veterans who were 32 years of age or older. 

Although the facts reported "by these studies must, of course, be ac- 
cepted, the possibility that the relationships found are strongly influenced 
by the failure to control the factor of class year makes it imperative to 
withhold Judgment regarding any intrinsic relationship between age and aca- 
demic success. 

Fortunately, Garmezy and Croee (35) and Pier son (69) have reported evi- 
dence on the relationship between age and academic success for veterans 
during a single class year. The correlation found by Garmezy and Grose 
was .00. This was based on the results for 564 veterans at the State Uni- 
versity of Iowa during the academic year 1946-1947. Upon more detailed 
examination of their data, they found that the tendency, if any, was for 
youth rather than age to be associated with higher grades. In Pier son* s 
study of students at Michigan State College, holding class year constant 
by considering only students completing the sophomore year during the 
spring or summer quarters of 19*1-7, veterans in the oldest and youngest 
age groups earned the" highest average grades. Those veterans entering at 
ages up to 18 had an average grade of 1,47, those entering at 25 or older 
averaged 1*37;> and two intermediate groups averaged 1.30. The groups 
varied in size from 53 to 140 students. These results, taken in conjunc- 
tion with the findings of the present study, indicate that the correlation 
of age with grades within the veterans group is not sufficiently high to 
warrant the conclusion that it in and of itself is a major determiner of 
veteran-nonveteran differences in academic performance. In this connec- 
tion it may be well to note the observation made in 1944 by Williamson (104) 
that increased age and work experience do not necessarily bring maturity 
and seriousness of outlook,, Although he was discussing the effects of war- 
work on students rather than the effects of military service, it would ap- 
pear that his point is equally pertinent here. 

Thompson and Pressey (91) succeeded in carrying out a controlled 
experiment in which veterans were compared with nonveterans who had 
entered college at the same older age during the period 1941-1946. Their 
study was based on 187 pairs of students, matched on the basis of per- 
cent ile on general ability test, college, program within college, and 
cumulative average during first three quarters in college. They found 
that the median average grade of the veterans was 2.63 as compared with 
2.55 for the nonveterans, during their first three quarters in the uni- 



42 ADJUSTMENT TO COLLEGE 



versity. They suggest that the superiority of the veteran may have re- 
sulted from "broader experience during the time he was out of school or 
from having more time to study as a result of federal support . It should 
"be added that the difference in average grade of only .08 in favor of the 
veterans may underestimate the influence of veteran status, since it is 
more exceptional for a nonveteran to return at an older age than is true 
for a veteran^ thus, the older nonveterans may "be a very highly motivated 
group of students. 

In the present study^ no controlled comparison was possible because 
there vas virtually no overlap in age between the veteran and nonveteran 
groups. However, it was Judged important to study the relationship between 
age and Adjusted Average Grade for veterans and nonveterans separately. 
And, as has been noted in the section on overachievers, it was found that 
the youngest nonveterans and the oldest veterans made the best records 
relative to ability. The finding for nonveterans is of course what we 
would expect from earlier studies of this question,, The finding for veter- 
ans, though in line with the many observations that veterans excelled by 
reason of greater maturity, is worthy of further examination,, 

Although a rigorous analysis of the contribution of age (and the 
determiners associated with it) did not seem possible in terms of the 
conceptions of the present study, one further step in the analysis was 
undertaken. This step was aimed at determining whether or not the re- 
moval of the oldest group of veterans would introduce a substantial change 
in veteran-nonveteran differences,, Accordingly, veterans born in 192if or 
earlier were removed from each of the twelve veteran groups. The youngest 
of the veterans thus excluded were within a few months of their 22nd birth- 
day when they entered college. Put in another way, this older group of 
students were entering college at or beyond the customary age for complet- 
ing college. Although this procedure did not by any means eliminate age 
differences between veterans and nonveterans, it undoubtedly reduced con- 
siderably the effect of age difference on the results. 

The outcome of this step was rather interesting. In the twelve "basic 
college groups, the results had favored the veterans In ten groups, favored 
the nonveterans in one group, and one group was tied. After the older 
veterans had been removed, there were three groups (Douglas, Little town 
State, and Midwest City) where the nonveterans slightly excelled the veter- 
ans and one group (Western State) showed a tie. In only three groups 
(Central State, Midwest Tech, and Middle State) did the veterans have an 
advantage amounting to more than five points in Adjusted Average Grade, 
and the median difference was reduced from six to only three points. It 
would appear, then, that when the older veterans were removed, the differ- 
ence between veterans and nonveterans, which was not very great to begin 
with, is reduced to a point where it can no longer be considered signifi- 
cant. 

Three points must be stressed, however, in relation to this statement. 
First, the above conclusion is concerned with the extent to which being a 
veteran, in and of itself, contributed to success in college; there is no 



THE FINDINGS AND THEIR IMPLICATIONS 



intention to minimize the remarkable achievement of the veteran in return- 
ing to college after a lapse of several years in his academic career and 
outdoing his nonveteran fellow students. Second, even when the advantage 
of the veteran is deliberately reduced "by eliminating a subgroup of rela- 
tively outstanding performers, the advantage still rests with the veteran 
group. Third, the removal of the oldest veterans after it was found that 
they were the highest of the three age groups among the veterans may be 
questioned on the grounds that the hypothesis was constructed after the 
results were in; in other words, that this procedure may have tended to 
capitalize on chance fluctuations in the results. In defense of this 
procedure, however, it may be 'observed that there was ample justification 
on the basis of previous reports for thinking that the oldest veterans 
were contributing disproportionately to the reported superiority of veter- 
ans. Moreover, the relatively large number of separate groups involved in 
the study tends to reduce the danger involved in applying a hypothesis 
to the same data which gave rise to it 

It appears that age may at least be regarded as providing some clues 
aa to why veterans did better. To take advantage of this finding, it was 
thought desirable to find out in what respects the older group of veterans 
did better than the younger group. 

Some Characteristics of Older Veterans 

Accordingly a supplementary study was made for veterans at one large 
college in order to compare the older veterans with the younger veterans 
with respect to some kO selected questionnaire items. (Central State was 
chosen because it showed a relatively large difference between veterans 
and nonveterans and included a large number of older veterans . ) The main 
purpose was to find out to what extent this group of veterans, who were 
most different from nonveterans in age, possessed characteristics which 
differentiated them from the younger veterans. Responses significant at 
the 1$ level are shown in Figures k and 5. Veterans at Central State 
who were born in 1924 or earlier were compared with those born later 
than 1924* There were 147 veterans in the older group and 317 in the 
younger group*, It was found that veterans at Central State who were 
born in 1924 or earlier did differ from veterans born later than 1924 in 
several important respects. Differences which are significant at at least 
the 5$ level of confidence are summarized below. 

Background CharacteristicSo So far as aspects of military service are 
concerned, more of the older veterans were commissioned officers, their 
military service was longer (the medians were about 37 months and 2k months 
for the older and younger groups respectively), and more had overseas 
service, as compared with younger veterans. Considerably more of the 
older veterans had served overseas for 18 months or more. These differ- 
ences undoubtedly result from the fact that the older veterans had gotten 
into the war in its earlier stages. 

The older veterans had graduated from high school at an earlier date, 
which of course is to be expected. The typical older veteran graduated 
in 1941, and the typical younger veteran in 1944. The other findings sug- 



44 



ADJUSTMENT -TO 'COLLEGE 



"RESPONSE 



SUB- 
GROUP 



PER GMT 
50 



100 



3? our "years or more 

between high -school 

and college: 

Item 6(b), Categories A, B 

Three years or .more 

on active duty: 

Item 8(b), Category D 

Held .rank of commissioned 
officer (major, lieuten- 
,ant commander "or lower):: 
.item 8(c), Category D 

Had no overseas service 
(either on "sea .duty -or In 
land -areas outside U. S.): 
Item:8(f.), Category A 

Eighteen months or more 
on .overseas duty:: 
Item 8(g), Category E 

First decided while in 
high .school to go to 
college : 
Item 8(j), Category A 

Probably would have come 
if -hadn't entered service: 
Item'8(mj, Categories A, B 

Probably would have come 

even without veterans ' 

benefits: 

Item>8(o'), Categories A, B 

Married, now or previously; 
Item 34, Category C 



A 
B 








Legend 

Subgroup A = older veteran students, born 1924 or earlier (N = 
Subgroup B = younger veteran students, born 1925 or later (N = 317) 



. PER CEUT MAKING SELECTED RESPONSES TO QUESTIOMAIRE ITEMS AMONG 
OLDER MAI VETERANS AKD YOUNGER MALE VETERANS. (BASED ON CENTRAL STATE IBBSH- 
MEK WHO ENTERED IN THE FALL OF 



'THE XfNntNCS .AND "THBIR 



RESPONSE 



Six months or more of 
full-time work between 
high school and = service 
or college : 
Item 9 (t>.), Category C 



Planning on profession 
which requires graduate 
training : 
Item 11, Category A 



SUB- 
.GROUP 



7Z5 





Planning to .take degree 
in less .than the usual 
amount df time: 
Item 21, Category A 



Living in apartment or 
house (rented or owned 
by respondent;) : 
Item 30, Category E 



Father T s income under 
$2,000 in student's high 
school years: 
Item lj-3, Category D 





Legend 

older veteran students, born 1924 or earlier (N 

^^^ M?;B = younger \veteran students, born 1925 or later (N = 517 '') 
L I MN* = male nonveterari students (N = 166) 
F = women students (IT = 272 ; ) 



FIGURE 5. PER CENT MAKING ; SELECTED KESPONSES TO 'QTBESTIONNA1RE ITEMS AMONG 
OLDER MALE YEMRANS, TOIMGER MALE OTTE3RANS, MALE WOmmmHStBj AMD KEMAII^ 
VETERANS. (BASED ON CE1W!RAL .STATE 'FRESHMEN WHO ENIDEREB IN THE FAIJL OF 



46 ADJUSTMENT TO COLLEGE 



gest that the older veterans tended to come to college in spite of more 
adverse economic factors. Only about JO per cent of the older group had 
decided to go to college "before graduation from high school, while 70 per 
cent of the younger veterans had made the decision at that time. Almost 
a fourth of the older men were married, while less than 10 per cent of the 
younger veterans had a wife at the time the questionnaire was filled out. 
Related to this result on marital status is the finding that more of the 
younger veterans lived in fraternity houses, and more of the older ones 
were renting or owned their house or apartment. Almost three fourths of 
the older group had been employed on a full-time basis for six months or 
longer, while only about 15 per cent of the younger group had worked full 
time for six months. About a fourth of the older veterans, as compared 
with about 15 per cent of the younger men, reported that while they were 
in secondary school their father's annual income was less than $2,000; 
this finding may be affected, however, by a general increase in income 
during the period under consideration. 

Factors Related to Motivation. The older veterans more often gave as 
their chief reason for attending college "to increase general knowledge/ 1 
a result which is in line with the observation of President Oliver C. 
Carmichael (l?), then Chancellor of Vanderbilt University, in 19^5* based 
on the veterans who came back first, that the veterans tended to be 
interested in fundamental courses. The older veterans also were less 
likely than the younger veterans to select the reason, "A college degree 
is necessary in order to enter the profession I have chosen." In answer 
to a question about vocational objectives, the older veterans less often 
named a profession requiring graduate training and more often gave an 
occupation which probably requires a college degree but not necessarily 
any graduate work. Presumably the older men did not wish to embark on a 
training program which would further delay their economic independence . 
The older veterans were less inclined to consider college grades 'Very 
important," and they more often planned to accelerate their college pro- 
gram. 

Worries, The older veterans apparently did not differ greatly from the 
younger veterans in amount of worry. However, three sources of worry 
were found which concerned the older men more than the younger. The 
older group was bothered more about nervousness, getting to know people 
socially, and housing. So far as housing is concerned, the problem may 
have been related to the fact that a larger proportion of the older veter- 
ans were married. Why more of the older veterans should be worried about 
nervousness is not clear, unless it has something to do with their 
greater feeling of urgency. Greater concern about social relationships 
might possibly result from the fact that the older men tended to seek 
friends off the campus. But they were not worried more about making ends 
meet financially,, 

Expenditure of Time.. The older veterans were found to spend a signifi- 
cantly greater amount of time than nonveterans in three kinds of activity. 
One was studying! the median number of study hours reported by the older 
men was about 19 1/2 as compared with 17 for the younger veterans. The 



THE FINDINGS AND THEIR IMPLICATIONS 47 



older men also spent more hours per week reading and studying materials re- 
lated to their courses but not required. Finally, the older veterans were 
more likely to "be engaged in paid employment. These results suggest that 
the older veterans were more interested or more strongly motivated to do 
academic work, and fit the hypothesis that their financial need was greater. 

The GI Bill, The results clearly indicate that the educational aspect of 
the GI Bill was a more important factor in getting the older veterans into 
college than was true for the younger veterans . Only about 65 per cent of 
the former group would definitely or probably have come to college even 
without the benefits provided , while about 85 per cent of the younger veter- 
ans would have come anyway. Similar results were found for the question, 
".. .do you think you actually would have gone to college if you hadn't 
entered military service?" almost 90 per cent of the younger veterans would 
have come anyway, as compared with about two thirds of the older veterans. 
The educational benefits of the GI Bill seem to be the deciding factor, 
and since the consideration is a financial one, it again suggests that the 
economic factors were more adverse for the older veterans. This finding 
agrees reasonably well with that of Strom (86), who found that only 50 P er 
cent of veterans 24 or more years old, with no preservice college training, 
would have returned without the GI Bill, and an additional 6 per cent were 
uncertain. 

Attitudes Toward the University. No significant difference was found in 
ratings of faculty members as teachers or judgments about the degree of 
satisfaction with the kind of education the men were getting. 

Self ^Select ion and Ye t er an Sue cess 

The results of the questionnaire in relation to veteran-nonveteran 
differences in academic success may be used more properly to formulate a 
hypothesis than to draw conclusions. The hypothesis to be proposed is 
that veteran-nonveteran differences reflect a process of se If ~sele ct ion . 
Let us consider how such a process might work. 

First, we may safely assume that determiners other than scholastic 
aptitude and high school record ezert a considerable influence on whether 
or not a student will or will not attend college. Evidence that many 
college -age students of very high ability do not attend college has been 
provided by many writers; much of this evidence has been summarized re- 
cently in Science , the Endless Frontier (l^). Studies by Toops (9^-), 
Bittner (7), GoetschT?^), Warner, Havighurst and Loeb (101), and earlier 
studies growing out of Counts' pioneering investigation of school-leaving 
have indicated what some of these influences are. From these studies, it 
may be judged that the following items included in the present study would 
be relevant to likelihood of college attendance: father's income, father's 
education, and age. With respect to age, Toops reported that studies in 
the 1930 's in two large Ohio colleges indicated that very few students 
entered college more than two years after they completed high school. It 
is also plausible that the student's marital status and his report regard- 
ing the likelihood that he would have attended college without the GI Bill 
fall in this same general field. We might, then, imagine a measure which 



48 ADJUSTMENT TO COLLEGE 



would be a composite of all of these non-aptitude determiners of college - 
going. For students of a given aptitude level, we might construct a 
curve something like that shown in the upper part of Figure 6 to represent 
the relationship between this over-all composite and the probability of 
college attendance., 

Let us consider, then, the students who actually attend college. It 
would seem that a student drawn from the relatively unfavorable end of the 
curve would not elect to go to college unless he had a special incentive 
to do so, had an unusually keen interest in and liking for academic pur- 
suits, or for some similar reason was strongly motivated to endure tempo- 
rary financial insecurity in order to achieve a college education. A stu- 
dent from the "favorable" end of the distribution would not need special 
personal incentives to attend college. Considering only the students who 
actually do go to college, it would therefore seem that students from the 
more favorable end of the scale will display less drive than those from 
the less favorable end. The relationship which exists might be something 
like that shown in the lower portion of Figure 6. It may be added that 
the actual form of this curve is not essential to the argtiment as long as 
a tendency for greater drive to be associated with greater adversity of 
non-aptitude determiners is present. 

This hypothesis appears to fit especially well the findings of the 
present study with respect to the superior performance of older veterans 
and married veterans. The findings with respect to father's income, both 
for veterans and nonveterans (although the interpretation for veterans is 
complicated by shifts in income levels during the period covered by the 
question) tend also to fit the hypothesis, at least as far as low incomes 
are concerned. The findings with respect to the importance attached to 
the GI Bill by various veterans in regard to their college-going are 
probably less dramatic than this hypothesis would lead one to suspect, 
although the trend is in the expected direction. The finding that veter- 
ans who would probably not have attended college if they hadn't entered 
military service were overachievers fits this hypothesis . (It is con- 
ceivable, of course 1 , that the economic benefits of the GI Bill tended to 
obscure rather than enhance the relationship between economic self -selec- 
tion and drive.) The absence of a relationship between father f o education 
and Adjusted Average Grade is contrary to the hypothesis being considered; 
this finding suggests that father f s education does not belong in the com- 
posite of non-ability determiners of college -go ing insofar as the present 
hypothesis is concerned. 

Why did the veterans excel? According to this hypothesis, the su- 
periority of the veteran student was not du primarily to any psychological 
characteristics associated with greater age or with experiences connected 
with military service. His superiority, we suggest, was due to a procoeo 
of self -selection growing out of a complex of circumstance a which included 
the educational benefits of the GI Bill and the delaying of college matricu- 
lation on the part of veterans. Those veterans who decided to go to college 
included a larger proportion of strongly motivated and academically-minded 
men than would otherwise have gone to college^ those with less drive and 



THE FINDINGS AND THEIR IMPLICATIONS 



100 -I 



Unfavorable Favorable 

NON-APTITUDE DETERMINERS OF COLLEGE -GOING 



Unfavorable Favorable 

NON-APTITUDE DETERMINERS OF COLLEGE-GOING 



FIGURE 6. HYPOTHETICAL RELATIONSHIPS BETWEEN NON-APTITUDE DETERMINERS 
OF COLLEGE -GOING AND (l) PER CENT OF HIGH SCHOOL GRADUATES WHO ENTERED 
COLLEGE AND (2) ESTIMATED MOTIVATION FOR COLLEGE WORK AMONG STUDENTS 
WHO ACTUALLY ATTEND, 



50 ADJUSTMENT TO COLLEGE 



Interest tended not to go to college because of economic and social condl- 
|tlons associated with greater age which functioned as deterrents to col- 
lege attendance . 

In conclusion, we should like to repeat that this hypothesis is of- 
fered, not as an explanation of veteran-nonveteran differences, but as a 
means of bringing together a number of findings into a more coherent pic- 
ture. The reader may speculate on the scope of a study which would be 
needed to make a rigorous test of this hypothesis as an explanation of why 
veterans excelled. More fruitfully, perhaps, he might consider the impli- 
cations of this hypothesis for future research on the long-range questions: 
who goes to college and why? and. who succeeds In college and why? 

Some Comparisons of Men and Women Students 

Strictly speaking, women students could have been excluded from the 
present study, since Insufficient female veterans were enrolled In any of 
the college groups to Justify a separate analysis. However, it was thought 
desirable to make some study of sex differences In order to obtain a more 
complete picture of the groups In which the veterans were enrolled, par- 
ticularly since findings in this area might have considerable educational 
significance. Accordingly, attention was given to sex differences In 
grades relative to ability In two college groups and to the questionnaire 
responses of women In nine college groups . In this summary the discussion 
will be limited to differences between women and nonveteran men* (The 
median values for veteran men on selected responses are shown in Figure 7 
for comparison.) 

Sex Differences In Grades .Relative to Ability 

At a university known in this study as Douglas University (a private 
coeducational university located in a southern city), data were obtained 
for 119 male nonveterans and 93 female nonveterans* When ability as meas- 
ured by the ACIE and high school average grade was controlled, no signifi- 
cant difference In freshman average grade was found. The women's mean Ad- 
Justed Average Grade was higher by only .06 of a letter grade unit. At the 
other university, Western State (a coeducational state university located 
in a western city), data were available for 222 male nonveterans and 1*82 
women students. Ability was again measured "by ACEE scores and high school 
average grade . In this case no difference In AAG was found within two 
decimal places of a letter grade unit. It must be concluded that so far 
as these two cong&arlsons are concerned there IB no significant tendency 
for either men or women students to excel in college achievement when the 
factor of ability 10 kept constant. 

In both institutions it was found that the male students were slightly- 
superior so far as ACEE mean scores are concerned, while the women were 
superior with respect to high school average grades. 



THE FINDINGS AND THEIR IMPLICATIONS 



5! 



BESPONSE 



MEDIAN PER CENT 



25 



iqo 



Preparation for better - 
paying job chief reason 
for attending college: 
Item 10, Category A 



College degree essential to 
chosen profession chief reason 
for attending college: 
Item 10, Category B 





To increase general 
knowledge chief reason 
for attending college : 
Item 10, Category C 




777//7/7A 



Planning on profession 
vhich requires graduate 
training : 
Item 11, Category A 




Planning on profession 
which probably requires 
college degree only: 
Item 11, Category B 




26 



Legend 
Male veteran 
Male nonveteran 



FIGURE 7 (PART l) . MEDIAN PER CENT MAKING SELECTED RESPONSES TO QUESTION- 
NAIRE ITEMS AMONG VETERAN MALE, NONVETERAN MALE, AUD FEMALE STUDENTS. 
(MEDIAE VALUES, BASED ON THE NINE GROUPS FOE WHICH THE RESPONSES OF WOMEN 
WERE TABULATED.) 



52 



ADJUSTMENT TO COLLSGE 



RESPONSE 



Eight hours or more per 
veek spent in athletics 
and physical recreation: 
Item 22(c), Category C 



MEDIAN PER CENT 




Two hours or more per 
veek spent in organized 
extracurricular activities : 
Item 22 (d), Categories B, C 



Ten hours or more per 
week spent in social 
activities : 
Item 22(e), Category C 



30 
39 
62 



7 777/y7/y//////( 




Tvo hours or more per 
week spent attending 
public lectures, concerts, 
other cultural activities: 
Item 22 (f), Category! 

Two hours or more per 
week spent in voluntary 
reading and study: 
Item 24, Category C 



28 
30 

ko 



35 



//////A 



Legend 

HUBI Male veteran 
I J Male nonveteran 
Y/A Female 



FIGURE) 7 (PART 2). MEDIAN PER CENT MAKING SELECTED [RESPONSES TO QDESTIONHAIRE 
ITEMS AMONG "VETERAN MALE, NONVET0RAN MALE, AM) IEMALE STUDENTS. (MEDIAN VALUES 
BASED ON THE NINE GROUPS FCR WHICH THE RESPONSES OF WOMEN WERE TABULATED.) 



THE FINDINGS AND THEIR SUPPLICATIONS 53 



Sex Differences in Questionnaire Responses 

As in the discussion of veteran-nonveteran differences in response to 
the various questionnaire items , two lines of evidence will be considered. 
The size of the difference between women and nonveteran men offers one 
kind of evidence . Another kind is provided by the consistency with which 
the difference between women and men goes in the same direction in each of 
the nine college groups. The latter kind of evidence is especially perti- 
nent In evaluating the generality of the trends. Figure 7 shows the median 
proportion of women , of nonveteran men, and of veteran men for responses 
which show the same direction of difference between women and nonveteran 
men in all nine of the groups In addition, three responses shown in Figure 
8 (worry about finances, about illness or death in family, and about nervous- 
ness) are consistent in all nine groups. Such consistency would be expected 
to arise by chance less thin once in 100 times. 

Background Characteristics . Women and men on the average showed rather 
similar background characteristics. There was a slight tendency for a 
greater proportion of women to be in the "born before 1928" category in 
age, perhaps because the draft was taking some of the older nonveteran 
men. Women were somewhat less likely to have held a full-time job. Dif- 
ferences in family income tended to be small on the average, but women 
tended to report higher Incomes. However, women were somewhat more likely 
than men to omit this question. Relatively more women reported that their 
fathers had completed college. Women were more likely than men to come 
from cities of 100,000 or more population, and were more likely to be living 
at home or in college dormitories than men. On the whole, the findings 
show considerable similarity between men and women students with respect to 
background characteristics. 

Motivational Factors . As shown In Figure 7, relatively few women had plans 
for a career involving graduate study; they were, however, somewhat more 
likely than the men to be planning for work which required a college degree 
but no graduate work. About half of the women named occupations which did 
not require college graduation or named broad vocational fields. In all 
but one of the nine groups, women were more likely than men to be unde- 
cided about their vocational plans; Katz and Allport's (56) study noted a 
similar tendency. Their chief reason for attending college tended to place 
less stress on career plans and more stress on desire to increase general 
knowledge and on social reasons than was true for the men. Women considered 
college graduation less important for their post -college plans, felt that 
college grades were somewhat less important in relation to their later op- 
portunities, and were somewhat less likely to want to accelerate their col- 
lege training. With regard to keeping up-to-date in their assignments, the 
women less often reported that they kept ahead or fell behind; they more 
often reported that they completed assignments just on time . The general 
tenor of these findings suggests that the motivations of women toward their 
college work placed less weight on future occupational goals and perhaps 
more stress upon their immediate college program than was true for the 
men. 



54 ADJUSTMENT TO COLLEGE 



Worries . Figure 8 summarizes the findings for male nonveterans and women 
students. The "bars which represent the median percentages of male non- 
veterans who were "bothered "some" or "very much" in Figure 3 are not iden- 
tical with those in Figure 8 because three of the groups in the earlier 
figure included institutions in which there were no women students . The 
order of importance for male nonveterans of the various sources of worry 
is about the same, however; first in importance are those related to aca- 
demic problems (concentration, getting accustomed to college study, and 
deciding what course of study to follow); then come financial worries, 
worry about personality problems (inferiority and nervousness), worry about 
social relationships, health problems, and housing. 

The tendencies for women to worry more about illness or death of 
loved ones and about nervousness, and to worry less about finances are 
consistent in the nine college groups. Women also expressed somewhat 
more concern about choosing their course of study and about strained 
personal relations. On the other hand, they worried less than the men 
about getting to know people socially and about housing problems. These 
tendencies are in the direction which would be expected in a culture which 
favored greater emotional expressiveness in women than in men and which 
tended to shelter women from practical problems. The small size of the 
differences, however, indicates that among the college women in this 
study, the effects of such tendencies are relatively small. 

Figure 8 shows an over -all tendency for women to report worries more 
frequently than men The greater tendency for women to worry, or at least 
to report that they worry, was also found in the responses to a general 
question, "Do you sometimes feel worried and anxious or upset?" Women re- 
ported somewhat more often than the male students that they worried fre- 
quently. In spite of this, women tended to claim slightly more often than 
men, that their worries had not interfered with their college work. 

The somewhat greater tendency for women to report worry should not, 
of course, be taken at face value as indicating that they are more sus- 
ceptible to worry and anxiety, particularly since the difference is slight. 
Indeed, the difference is less than would probably be expected on the 
basis of popular stereotypes. 

gxpenditur e of Time . The comparison of male nonveterans and women students 
with respect to amount of time spent in various types of activity showed 
no significant difference in time spent in studying and in bull sessions. 
Women spent considerably more time than men in extracurricular activities 
other than athletics, attending lectures and concerts, and in social activi- 
tiesdates, parties, movies, etc. and less time in athletics, physical 
recreations and in voluntary course reading, as is apparent from Figure 7 
(Part 2). In general they spent less time than men in attending classes 
(presumably because they tend to take fewer laboratory courses), and were 
less likely to have a part-time job. 

Attitudes Toward the College Environment With regard to a number of ques- 
tionnaire items which relate to attitudes toward the college environment, 
differences tended to be slight. Women students were slightly more often 



THE FINDINGS AND THSIR IMPLICATIONS 



55 



Source of Worry 



Median Per Cent Bothered Some or Very Much 
25 5p 75 IPO 



Concentration 

Getting accustomed to study 

Deciding what course of study to 
follow 

Making ends meet 

Peelings of inferiority 

Nervousness 

Getting to know people socially 

Relations with opposite sex 

Making up a deficiency 

Health problems 

Strained personal relations 

Illness in family 

Housing 



7/7 '/////// / ///t 



te 



/////////// / 



7T7I 



17 /// / /^ 



23 



/ / / ../ / / / / i 



/ / / / / 7771 



/ / / //I 



////.//I 

Ca 



ZJ 



Legend 

Male Nonveteran 
Female 






FIGURE 8. MEDIAN PER CENT OF NONVETERAN MALE AND FEMALE STUDENTS BOTBEKED BS 
VAEIOUS PROBLEMS. (MEDIAN VALUES, BASED ON THE NINE GROUPS FOB WHICH THE 
RESPONSES OF WOMEN WERE TABULATED.) 



56 ADJUSTMENT TO COLLEGE 



critical about having a satisfactory place to study; and they vere somevhat 
less likely to criticize the quality of instruction provided. So far as 
interest in courses, enjoyment of their studies, satisfaction with the 
kind of education they were getting, preference for the school or division 
in which they vere enrolled, and ratings of their instructors are concerned, 
essentially no difference was found between the men and women. Women were 
no more inclined than the men to make comments, in a free -answer situation, 
regarding changes they would like to see made in the program or organiza- 
tion of education at their college. 

Summary. In their responses to the questionnaire, women differed most 
strikingly from the nonveteran men in their vocational plans and in their 
reasons for attending college. Less than ten per cent of the women, as 
compared with about 35 per cent of the men, were planning to enter a pro- 
fession requiring graduate study. On the other hand, about one fourth of 
the women were planning a career which required college graduation hut not 
advanced training, as compared with about 15 per cent of the nonveteran men. 
Acquiring general knowledge was much more likely to be the chief reason for 
attending college for women than was true for the men. Understandably, 
preparation for professional work or for a better-paying job was relatively 
less important for women than men. These differences undoubtedly reflect, 
at least in part, the prevailing expectations of the parents of these stu- 
dents and the prevailing conditions with regard to "marriage y_s. a career* 1 
in our culture. 

Women differed substantially from men in the way they spent their time. 
Organized extra-curricular activities other than athletics, social activi- 
ties and recreation, and attendance at public lectures and concerts were 
relatively more popular with women than with men; while athletics and 
voluntary reading were more popular with men. Except for the last differ- 
ence mentioned, these differences probably would agree reasonably well with 
the stereotype of men students and coeds held by college students. 

The tendency for women to report more worry about nervousness may re- 
flect greater social acceptability of this type of worry in a woman than 
in a man; the tendencies for women to report more worry about illness and 
death in their family and less about making ends meet financially may re- 
flect somewhat closer emotional and economic ties to the family for the 
women. 

It must be added that any general summary may overlook important dif- 
ferences in the questionnaire responses of men and women in a particular 
college or a particular curriculum; the fact that even the strongest group 
tendencies showed many individual exceptions shpuld also be recognized in 
interpreting these results. 



THi FINDINGS AND THEIR IMPLICATIONS 57 



Effectiveness of Conventional Predictors of Academic Success 

In the process of evaluating the differences between veterans and non- 
veterans -with respect to grades relative to ability, a considerable amount 
of information "was obtained relating to the value of tests and high school 
record for predicting college grades. The number of validity studies con- 
ducted as an incidental part of the study was unusually large, and it seems 
appropriate to summarize here the results of this aspect of the study. k 

Test Scores and High School Record as Predictors 

The American Council Psychological Examination (ACPE) was the test of 
ability most commonly employed. It was used as a predictor for twelve 
separate college subgroups! since veterans and nonveterans were treated 
separately and two of these subgroups included women students, 26 separate 
correlations between total score on the ACEE and freshman average grade 
were obtained. There was considerable variation among these coefficients, 
presumably because of sampling error due to the small size of certain of 
the subgroups; the coefficients ranged from .28 to ,6l. The median of the 
2k male subgroups was .^7. It may be observed parenthetically that Thur- 
stone and Thur stone (92), in 1932, remarked that the correlations between 
American Council Psychological Examination scores and grades averaged 
around .50, which is in good agreement with the present findings. 

The median validity coefficient for the twelve male veteran subgroups 
was somewhat higher than for the male nonveteran subgroups; the two medians 
were A8 and .^3 for veterans and nonveterans respectively. While the dif- 
ference is not very great, the finding is consistent with the hypothesis 
that veterans tend more than nonveterans to achieve the grades they are 
capable of earning. In other words, veterans may be more uniformly moti- 
vated to work at maximum capacity; differences among veterans in grades 
earned are to a lesser extent a function of such nonintellectual variables 
as interest and motivation than is the case for nonveterans. 

Another predictor which was available in a large number of colleges was 
some sort of measure based on high school record average grade, rank in 
class, or rank adjusted in some manner on the basis of differences between 
various types of secondary schools Such a measure was employed in eleven 
college groups, and 22 validity coefficients for male veterans and nonveter- 
ans were therefore computed. The validity coefficients varied from .33 to 
.68, and the median value was .5?- T& e high school record thus is found to 
furnish a somewhat more accurate prediction, for these groups, than the ACPE* 
In six college groups both ACPE and high school record were employed as pre- 
dictors, thus affording twelve direct comparisons of the two predictors 
among male groups. In eleven of the twelve comparisons the validity co- 
efficient was higher for high school standing. Such consistency would be 
expected to occur by chance fewer than once in a hundred times. 

4 
Wo attempt is made to review the extensive literature in this field. 

Recent reviews in this general area have been published by Cain, Michaelis, 
and Eurich (l6), Garrett (36), and Travers (95). 



58 ADJUSTMENT TO COLLEGE 



Comparing veterans and nonveterans with respect to the predictive 
value of high school record, we find that the median of the validity co- 
efficients is .53 for veterans and .6l for nonveterans. It will be re- 
called that validities were higher for veterans when the ACPE was con- 
sidered. The reverse finding for high school record seems reasonable, In 
the light of the greater time elapsing between high school graduation and 
college entrance for veteran students. The findings are also consistent 
with the previously mentioned hypothesis that veterans tend somewhat more 
than nonveterans to work at maximum capacity because of more uniform moti- 
vation. High school grades presumably reflect motivational and other non- 
intellectual factors as well as ability to do academic work; to the extent 
that these have changed more for some veterans than for others, the corre- 
lation would be lowered. To the extent that high school grades reflect 
knowledge and skills directly useful in college work, their predictive 
effectiveness from this viewpoint would be lowered also, since it is 
plausible that the educational effects of service were not uniform for 
all veterans. 

Although it is of some theoretical interest that better prediction 
of freshman grades was obtained from the high school record than from the 
ACPE, the more critical question is how well the two function together as 
a team. These two predictors were used as the predictive team in six col- 
lege groups. Considering the twelve multiple correlations obtained from 
the veteran and nonveteran subgroups, a range of .53 to j6 was obtained, 
and the median value was ,64. The median multiple correlation coefficients 
for veterans and nonveterans respectively were .60 and .68. The use of 
the two predictors in combination thus furnishes a better basis for pre- 
dicting freshman grades than either ACHE scores or high school record used 
alone. The magnitude of the correlations Is great enough to indicate that 
the combined measures provide a really useful prediction of how well a 
particular student is likely to succeed In his freshman year of college. 

Prediction of Sophomore or Junior Grades 

In the studies of interrupted veterans the measure of ability used 
was grades earned during the freshman year* It was decided to employ 
freshman grades rather than tests and high school record on the assumption 
that freshman grades would provide a better prediction; this assumption 
vas tested at one institution. At Adams, a prediction based on a combina- 
tion of College Board tests and adjusted high school rank correlated (for 
veterans) .51 with fourth-semester average grade, while the correlation of 
first -semester grade with the same criterion was .59* For nonveterans, 
the difference was smaller; the analogous correlations were ,6k and .66. 

In three interrupted groups, first -term grades were used as the pre- 
dictive measure and fourth-term grades were used as the measure of suc- 
cess. In two other groups, average grades for the first two quarters 
were the predictor and eighth-quarter grades were the criterion measure. 
The median prediction coefficient in the five nonveteran groups was .62; 
In the five veteran groups It was .57, The difference in coefficients 
is plausible in view of the interruption in the educational careers of 
the veterans. 



THE FINDINGS AND THEIR IMPLICATIONS 59 



The data of this study made it possible to compare the effectiveness 
with which later college grades were predicted from earlier college grades 
with the effectiveness with which first-year college grades were predicted 
by a combination of high school record and test scores. Comparisons were 
based on nonveteran students in the same college and division. In the 
three groups where fourth-semester grades were predicted from first -semester 
grades, the validity coefficients for nonveterans were ,66, .68, and .60; 
the corresponding figures for prediction of freshman grades from preadmis- 
sion data were 065, .66, and 5^. I& the two groups where eighth-quarter 
grades were predicted from average grades in the first two quarters, the 
validities were only .4 5 and ,62 as compared with validity coefficients of 
,76 and .70 for the prediction of freshman grades. Thus, first-year col- 
lege grades can be predicted about as adequately from data available at 
entrance as later college grades for a single term can be predicted from 
Initial college average. 

The Effect on Validity of Time of Taking Aptitude Tests 

At two institutions (Adams and Stewart) there was considerable varia- 
tion among veterans with respect to time of taking the aptitude test. The 
test employed at these two institutions was the Scholastic Aptitude Test 
of the College Entrance Examiruatlon Board. Many of the veterans had been 
admitted to college, after taking the tests, at the time of graduation 
from high school, although they did not matriculate until after their war 
service. Others applied, were tested, and were admitted after war ser- 
vice. This situation made it possible to study the time of testing as a 
variable in relation to the predictor and criterion measures. Such a study 
is Important because at most institutions the testing occurred at the time 
of entrance, which was soon after high school graduation for the nonveterans, 
but a varying number of years after graduation for the veteran students. 
The problem is also of interest to college admission officers, who may 
feel that they should discount the results of tests taken a year or two 
prior to application to college. 

At Adams, the correlations involving the variable date of 'testing 
were uniformly low, varying from -.08 to ,21, and at Stewart they ranged 
from -.(A- to ,10. Except for the possibility of a slight Increase in 
verbal ability scores, the evidence Indicates that, within the time limits 
and particular tests used, date of testing is a matter of little signifi- 
cance , 

The results indicate that, at least so far as the College Board Scho- 
lastic Aptitude Test is concerned and for the time period here studied, 
the time of taking the test has little effect on the predictive value of 
the test and little relationship to the predictive measures eigployed. 



60 ADJUSTMENT TO COLLEGE 



Describing a College from Its Students 1 Questionnaires 

A questionnaire such as the one employed in this study may be a use- 
ful device when employed by a college administration for the purpose of 
inquiring into the characteristics of the students at its college and 
their attitudes toward the educational program. Such an instrument is 
particularly useful when reference data are available which permit students 
in the particular college group to be compared with students at other insti- 
tutions. Some of the outstanding characteristics of two college groups as 
revealed in the questionnaire responses will be described here, merely to 
illustrate the potential value of a questionnaire like the Student Opinion 
Questionnaire which was employed in this study for getting a picture of 
an institution or one of its divisions. The results will be based on the 
findings for freshman nonveteran male students only. The "average college" 
referred to in the following discussion is the median of the twelve basic 
college groups employed in this study. 

Adams University is a private college for men located in an eastern 
city. Midwest Tech is a coeducational land-grant college located in a mid- 
western, city. The code names are of course used in order to preserve the 
anonymity of the colleges. (Adams had the largest number of nonveteran stu- 
dents among the twelve basic groups; Midwest Tech engineering students were 
selected to represent the three engineering college groups.) 

Freshmen at Adams University 

Background Characteristics . Freshmen at Adams were of about the same age 
as those in the average university, and they came from small towns and 
large cities in about the same proportions as students at the average col- 
lege. Considerably fewer had had full-time work experience; 93 P*" cent 
had never had a full- time job, while in the average group the percentage 
was about 75. The fathers of Adams freshmen were better-off financially; 
60 per cent reported a family income greater than $6,000, as compared with 
about 25 per cent in the average college. The fathers were also better- 
educated; 60 per cent were college graduates, as compared with about 25 
per cent in the average college About two-thirds of the freahmen had at- 
tended private schools, while at most colleges almost all had attended 
public high schools. More than half of the freshmen considered themselves 
very well prepared for college, as compared with about JO per cent in the 
average group. 

factors Related to Motivation, Adams freshmen gave "general knowledge" 
as their reason for attending college considerably more often than usual, 
and less often said they went because a college degree was necessary in 
order to enter a chosen profession. They planned to enter a profession 
requiring college graduation or graduate study no more often than the 
freshmen in the average college, and they resembled the typical freshmen 
with regard to certainty of vocational choice. There were no striking 
differences with respect to judgments about the difficulty of college work, 
the inrportance of college grades or graduation from college, or tendency to 
keep up-to-date in assignments. 



THE FINDINGS AND THEIR IMPLICATIONS 61 



Worries. Adams freshmen did not in general differ very much from freshmen 
in the average university with regard to tendencies to worry. They were 
near the median group in amount of worry about such things as deciding 
what course of study to follow, inferiority, nervousness, getting to know 
people socially, and making up a deficiency in preparation for some course. 
It is particularly interesting to note that they worried about making ends 
meet about as much as the freshman student in the average college. Adams 
freshmen worried slightly less than usual about being unable to concentrate 
and getting accustomed to college study, and they worried somewhat more 
about relations with girls . More of them felt that worries had not Inter- 
fered with college work than in the average freshman group. 

Expenditure of Time . Adams freshmen spent considerably more time studying 
than freshmen usually do; almost half reported 25 hours or more a week as 
compared with about one fourth in the average group. They also spent con- 
siderably more time in bull sessions (almost half devoted six or more hours 
per week to this activity) and attending evening lectures given by visiting 
lecturers or local faculty members. They spent less time than freshmen in 
the average college group in social activities (perhaps because Adams is 
not coeducational) and in reading or studying material related to courses 
but not assigned. They did not differ appreciably in time spent attending 
classes, in athletics, extracurricular activities, or paid employment. 

Attitudes Toward the College. An unusually high proportion of Adams fresh* 
men expressed satisfaction with the kind of education they were getting, 
and more than usual felt that most or all of their teachers were good 
teachers. On the other hand, the proportion who felt they were really 
interested in most or all of their courses is somewhat lower than in the 
average group. Adams students resembled those in the average institution 
with respect to feelings about the worth-whileness of college studies, 
their evaluation of their study facilities, and amount of enjoyment of 
their studies, 

Freshmen In Engineering at Midwest Technological University 

Background Character istics a With regard to background characteristics, 
Midwest Tech engineering freshmen differed from those in the average uni- 
versity group most notably with respect to size of home community; kO per 
cent came from farms or towns of less than 2,500 people, while in a typical 
group the proportion was less than 15 per cento These freshmen tended 
more than usual to consider themselves poorly prepared for getting the 
most out of their courses* They resembled the average freshman group in 
age, amount of full-time employment, and father s t income and education. 
Practically all had attended a public high school. 

factors Belated to Mot iyat ion , , Freshman engineers at Midwest Tech appear 
in general to differ somewhat from those In the average group with respect 
to certain motivational factors. They more often went to college in order 
to prepare for a better-paying job and less often to get necessary training 
for entering a profession. They less often planned to enter a profession 
which requires graduate study, and fewer were certain of their vocational 



62 ADJUSTMENT TO COLLEGE 



choice. Fewer considered college graduation absolutely necessary in order 
to do the kind of work planned, and they tended more to consider college 
grades jfejyjf important rather than very important. They did not differ 
markedly from freshmen in the average group with respect to judgments 
about the difficulty of college work or amount of effort exerted in course 
work, but they tended more to keep ahead in completing assignments. 

Worries. Midwest Tech freshmen tended to resemble freshman students in the 
arerage college group with respect to amount and kind of worry. They 
worried somewhat more about feelings of inferiority and about making up a 
deficiency in preparation for some course. With respect to other sources 
of worryinability to concentrate, getting accustomed to college study, 
deciding what course to follow, soaking ends meet, nervousness, getting to 
know people socially, relations with girls, health problems, and housing 
they resembled the freshmen in the average group. They did not tend to 
feel more or less than usual that worry had interfered with their college 
work. 

Expenditure of . Tiae. With respect to expenditure of time, only one out- 
standing characteristic of the Midwest Tech engineering freshmen may be 
noted-~they spent a much greater amount of time in attending classes, 
laboratories, and other regularly scheduled course meetings than students 
in the average college group. This finding is undoubtedly a function of 
the fact that students in an engineering college have a great deal of 
laboratory work; but time spent in classes and labs by Midwest Tech non- 
reteran, freshen was greater than for the two other engineering schools 
included in the twelve basic groups. The median member of hours is about 
51, as compared with 28 and 21 in the other two engineering schools. With 
respect to other activities studying, athletics, extracurricular activities, 
social activities, attending lectures and concerts, bull sessions, paid 
employment, and voluntary reading and study the Midwest Tech freebusen were 
quite similar to freshmen in the average college group, A slight tendency 
wafl noted on a number of these activities for a somewhat greater proportion 
of the students than usual to fall in the intermediate or moderate category, 

Attitudes Toward the College. Midwest Tech engineering freshmen expressed 
a greater amount of dissatisfaction with the kind of education they were 
getting, more often preferred some other school or division at the university 
than the one they were enrolled in, and less often expressed the opinion 
that most or all of their teachers were good teachers than freshmeu in the 
average college. However, they expressed about th usual interest in their 
courses, enjoyment of their studies, and evaluation of the worth-whilenese 
of college study, and were satisfied with tbe study facilities to about the 
usual extent. 

In interpreting summaries such as the foregoing, knowledge of the char- 
acteristics of the particular college would obviously be essential. Many 
of the f indijoge might be of considerable importance when viewed in the light 
of some local procedure or custom. The foregoing brief descriptions are in- 
tended merely as illustrations of the potential value of a study of student 
characteristics and attitudes by use of a suitable cpiestionnaire. 



63 



Chapter II 
PLAN OF THE STUDY 

Purposes of the Study 

Since about 19^5, American colleges have been crowded with students, a 
large proportion of whom are veterans of the recent war. In part, the influx 
was an outcome of a new feature in the educational scene the educational pro- 
visions of the Federal law commonly called the GI Bill of Rights, Many of 
these veteran students would not have attended college without the aid thus 
provided. At the beginning, considerable concern was, expressed regarding the 
possible effects of combat and of other features of wartime service upon the 
adjustment of veterans to the life of colleges typically designed for a less 
widely experienced student body. 

The experience of university faculties quickly demonstrated that the more 
pessimistic views were unfounded. University teachers and deans reported that 
the veterans were alert and industrious students, that their influence on the 
undergraduate body as a whole was wholesome, and that the incidence of events 
that could be traced to battle shock was much smaller than had been anticipated. 
Numerous articles were published in newspapers and popular magazines in which 
rather glowing accounts of the scholastic success of veteran students were pre- 
sented. Veteran students were almost universally reported to be superior to 
nonveterans in academic achievement. 

The Carnegie Foundation for the Advancement of Teaching became interested 
in the problem of the academic success of the veteran student in college, and 
the College Entrance Examination Board was requested to make a study investi- 
gating the relationship between veteran-nonveteran status and academic success 
in college. The College Board agreed to conduct such an investigation, with 
the support of the Carnegie Corporation. The primary objectives of the study 
were to answer the following questions: 

1. Do veteran students make better grades in college, in 
relation to their ability, than nonveteran students? 

2. What light does information about background, attitudes, 
and other qualities throw upon veteran-nonveteran differ- 
ences? 

3. How do veterans who could not have attended college 
without the financial assistance provided through 
the GI Bill compare with veteran students who were 
financially able to attend college? 

The third of the three objectives, that of comparing veterans who were 
enabled to attend college by the educational provisions of the GI Bill with 
those who were financially able to attend college, has evident implications 
for any sort of plan for subsidizing higher education through scholarships 
or other types of financial aid. 



ADJUSTMENT TO COLLEGE 



The analysis of the data needed to study the three points specified 
would furnish information relating to numerous other questions of interest 
to educators. How well can scholastic success in college "be predicted from 
aptitude test scores and measures of hj^h school achievement? What improve- 
ment in prediction can "be effected "by combining various predictors of college 
success? Are there differences between veterans and nonveterans with regard 
to the relative effectiveness of predictors for forecasting college achieve- 
ment? Of what significance is the differential in time of taking aptitude 
tests in relation to time of entering college? Are veterans who enter college 
as freshmen handicapped during the first term, in comparison with nonveterans, 
so that a warm-up or refresher period might "be desirable? In the case of the 
veteran whose education was interrupted by the war, is a period of readjustment 
necessary, or does he come "back fired with enthusiasm which leads to temporary 
overachie vement ? Questions of this sort were taken into account in designing 
the study. 

The best opportunity for making such a study presented itself in the fall 
of 19^6 } which was the time when the maximum number of veterans (particularly 
veterans who had actually experienced combat overseas) were enrolling in col- 
lege. It was felt that the academic year 19^6-19^7 v &s the optimal time to 
study the question of the effect of war service on college achievement Stu- 
dents enrolled at that time would possess to a marked degree the characteristics 
which make veterans different from nonveterans and would be sufficiently 
numerous to yield statistically stable results. 



The General Plan 



The general plan of the study, as it finally evolved, may b briefly out- 
lined as follows : 

Cho ice of Co liege s . It was desired to obtain data for both veteran and non- 
veteran students from a number of institutions which were of varied types, 
size, and location, Criteria bo be used in selecting colleges included number 
of cases, availability of suitable predictor scores, and availability of suit- 
able criterion data. Insofar as possible, it was also desired to follow the 
principle of diversity with duplication in the choice of colleges: that is, 
to select matched pairs representing a variety of types. The selection of 
institutions was made primarily on the basis of questionnaires which were 
mailed to thirty-six colleges and universities which were considered likely 
to meet the various requirements of the study, 

Collection of Data. During the spring of 194?, substantially all of the par- 
ticipating colleges were visited by a member of the College Board staff, at 
which time a paid supervisor was selected to have charge of data collection,, 
In conference with the supervisor and other college officials, decisions were 
made regarding the groups of students to be studied, and regarding the cri- 
terion and predictor variables to be used. These data were obtained usually 
during the summer, after completion of the spring term,, 



PLAN OF THE STUDY 55 



Preparation and Administration of a Questionnaire. In order to obtain Infor- 
mation on personal characteristics which might account for any observed dif- 
ference betveen veterans and nonveterans in college achievement, a question- 
naire was prepared. This questionnaire contained items dealing with biograph- 
ical history, attitudes toward college, vocational alms, disposition of time, 
worries, and other areas thought to have a possible relationship to college 
success,, The questionnaire was administered by the supervisor at each college 
to students in the selected populations at his institution* 

Coding of Questionnaires, A staff of carefully selected coders was trained to 
code the questionnaire items. Coding was necessary in order to permit the use 
of tabulating machine equipment in the analysis. Many of the items were pre- 
coded and therefore presented no particular problems; others, however, required 
careful judgment in order to assign each response to one of a number of cate- 
gories that were chosen on the basis of study of samples of questionnaires. 

Analysis of Academic Data In order to bring veteran-nonveteran differences 
into sharper focus, each separate analysis of academic data was based upon a 
carefully defined group of students. Each of these defined groups was limited 
to students enrolled in a specific division of a particular university who had 
entered that division at a specified time (or at specified times). Twenty-five 
such groups were selected for study in the sixteen colleges , These groups were 
in turn subdivided into male veterans, male nonveterans, and (in nine of the 
groups) female nonveterans,, In all, fifty- two such subgroups were included in 
the analysis of the academic data. For each of these, intercorrelations, means, 
and standard deviations of predictor and criterion measures were computed. 

From the outset, it was considered essential that, in any comparisons of 
the relative achievement of veteran and nonveteran students, allowance be made 
for any possible differences in ability. More specifically, the comparisons 
should depend upon how far each of the two groups exceeded or fell short of 
the level of achievement expected of it on the basis of scores on suitable pre- 
dictors,, At this point, however, a basic problem arises : are the conventional 
predictors of academic success equally appropriate for both groups? 

By means of analysis of oovariance procedures, it is possible to make a 
rigorous check on the appropriateness of the predictor before proceeding to 
the actual comparisons of achievement relative to ability. Thus, in the pro- 
cedure followed in the study, the first steps provided a basis for evaluating 
the comparability of the predicted grades; if the results of these steps met 
specific requirements, comparisons of veteran and nonveteran grades (after 
allowance for ability differences) could be made with reasonable confidence. 

Analys is of Quest ionnair e Data . An Adjusted Average Grade (AAG) was calculated 
for each male student in sixteen of the twenty-five groups studied, for use in 
analysis of questionnaire responses . A student's AAG is a measure of the extent 
to which his grade was higher (or lower) than would be expected on the basis of 
his predictor scores. The use of this index made it possible to determine 
readily whether the students who gave any particular response to a question 
were, on the average, performing above or below their expected level in aca- 
demic work, This type of information Is particularly informative when veterans 
who chose a particular response to a question are compared with nonveterans who 
chose the same response,, In addition to mean AAG's for each item, the number 



66 ADJUSTMENT TO COLLEGE 



of students choosing each response was also tabulated and studied. The 
analysis of questionnaires had, therefore, two purposes: to throw light upon 
the "ororachievement" of veteran and nonveteran students, and to provide 
descriptive information regarding baolcground, experiences, and attitudes of 
the students. 

The tabulation of mean MG for each response to a question for veterans 
and nonveterans separately made it possible to determine, for each college 
group, whether the difference between veterans and nonveterans choosing a par- 
ticular response was greater than, equal to, or less than the difference in 
that college group as a whole. By counting the number of times that the differ- 
ence in MG was more favorable to the veterans for each response and for each 
college group, it was possible to determine whether a particular question helped 
to account for the difference in achievement between veterans and nonveterans. 

In order to study the problem of the relation of economic factors to col- 
lege achievement, the student's own statements as to the effect of the GI Bill 
upon his decision to attend college was made the basis of further investigation. 
The relation to MG was studied for the sixteen college groups, and a more 
comprehensive study, including cross -tabulations with other questionnaire 
material, was carried out in one college group* 

The above paragraphs give a very brief outline of the general procedures 
used in the study. Various aspects of this plan are discussed in somewhat 
greater detail below. 

Selection of Colleges 

It was desired that the study of veterans' achievement in college be 
based on data obtained from a variety of types of institutions, including 
coeducational and men's colleges, private colleges and colleges supported by 
state or municipal funds, large and small institutions, colleges with various 
curricular emphases (such as liberal arts, engineering, and agriculture), 
both heavily endowed colleges and colleges with lees endowment, and colleges 
representing various geographical regions of the United States. It was also 
desired that pairs of colleges with roughly the same characteristics be chosen. 
Further restrictions in the choice of colleges included adequate numbers of 
students, availability of suitable data, and willingness of the college to 
participate in the study. 

The first step in selecting colleges was to study Good's Guide to Col^ 
legea, Universities, and Professional Schools in the United State a (American 
Counc il on Education, 19^5) . A tentative list of colleges which seemed to 
meet the criteria for inclusion in the study was prepared. The number of 
colleges in this list was reduced to 3&" on the basis of conferences with 
people who had wide acquaintance with colleges throughout the country. 

Letters were sent to the presidents of the 36 colleges by Dr, 0. C. 
Caxmichael of the Carnegie Foundation for the Advancement of Teaching; each 
letter briefly described the objectives of the gtudy. A few days later a 
letter and a brief questionnaire were sent from the College Entrance Examina- 



PLAN OF THE STUDY 67 



tion Board to the same persons. Anonymity of the colleges in the published 
report was promised to the institutions who wished to participate in the 

study . 

The questionnaire sent to the college presidents was designed to provide 
information which would be useful in judging whether or not the institution 
was an appropriate one to include in the study from the standpoint of adequacy 
of data. It furnished a convenient and uniform method for the college to re- 
port: (l) the number of veteran and nonveteran students entering each college 
or division of the university in the fall of 19^5 and in the fall of 19^6; (2) 
information regarding the program of intelligence or scholastic aptitude testing, 
including names of tests used, whether tests were normally taken near the comple- 
tion of secondary school or at the time of entrance to college, and the proportion 
of students having scores; (3) whether or not some over-all measure of achievement 
(such as average grade or grade-point average) was routinely available for these 
students; (k) what specific courses were taken in common by all or by a large 
proportion of the freshman students; and (5) whether or not some measure of 
scholastic success in high school was available for a large proportion of the 
entering freshmen. A question was included to ascertain whether or not the 
institution wished to participate in the study. 

Twenty-six of the colleges indicated that they wished to participate in 
the veterans study and returned questionnaires. Seven institutions could not 
or did not wish to participate, and from the remaining three colleges there 
was no reply or a noncommittal reply, with no reply to follow-up letters. The 
usual reason given for not wishing to participate was the pressure of work in 
the registrar's office, which is understandable in view of the heavy enrollments 
during this period of time. 

Nine of the twenty-six colleges which expressed willingness to participate 
in the study were not included because of an inadequate number of cases in one 
or more of the groups to be studied or because of the lack of certain crucial 
information. Data were obtained from seventeen colleges. One institution was 
dropped from the study after the data had been collected, since needed data 
were available for too few students; thus the statistical analysis, is based 
on groups from sixteen colleges and universities. 

The participating colleges will be referred to in this report by code names. 
The private colleges were given pseudonyms which are common American surnames 
assigned at random with no attempt to make the name carry any implication as 
to any characteristic of the institution. The publicly supported institutions, 
including both state and municipal colleges, were given geographical names 
which describe in a general way the location of the institution. The following 
is a list, by geographical section, giving the code name and a brief description 
of each institution: 

Eastern C ity Univer a ity : A coeducational publicly- supported college of 
arts and science located in an astern city. 

Adams Uni ver s ity : A private university for men located in an eastern 
city. 



68 ADJUSTMENT TO COLLEGE 



Stewart University. A private university for men located in a small 
eastern city. 

Miller University; A private coeducational university located in an 
eastern city. 

Midwest State University: A large coeducational state university lo- 
cated in a midvestern city* 

Midwe st C ity Univer s ity ; A coeducational, publicly-supported university 
located in a midvestern city. 

Littletovm State University: A coeducational state university located 
in a small middle we stern city. 

Harris University: A private coeducational university located in a 
small midvestern city. 

Evans University: A private coeducational university located in a 
midvestern city* 

Central State University; A large coeducational state university 
located in a small midvestern city. 

Turner University: A private coeducational university located in a 
midvestern city. 

Midwest Technological University s A coeducational land -grant college 
emphasizing agriculture, engineering, and other applied arts, located in a 
small midvestern city 

Middle State University; A large coeducational state university lo- 
cated in a midvestern city- 

Douglas : Phi ver -a 1 ty : A private coeducational university located in a 
southern city* 

Southern Technological University^ A coeducational land -grant college,, 
emphasizing agriculture, engineering, and other applied arts, located in a 
southern city,, 

Western State Univer s ity? A coeducational state university located in 
a ve stern city* 



Collection of Data 

A member of the College Board staff visited each of the participating 
institutions, with the exception of Western State University, for the purpose 
of making detailed arrangements for the collection of data and for the admin- 



PLAN OF THE STUDY 69 



istration of the questionnaire. Western State was not visited "because of 
considerations of time and distance, and the arrangements there were made 
entirely "by correspondence. Prior to the visits to the remaining institutions, 
a letter "was written requesting that preliminary arrangements be made for 
selecting a supervisor to have charge of the data collection and question- 
naire administration. 

The contribution of the supervisors to the execution of the study vas 
substantial. In a number of schools, a member of the college personnel staff 
or of the Psychology Department acted as supervisor; in others, graduate stu- 
dents carried this responsibility,^ The proposed study was discussed in some 
detail with the supervisor and other interested persons at each institution 
visited, detailed plans were drawn up for the data-collection and quest ionnaire- 
administration, and the plans were summarized on a check-list previously pre- 
pared for that purpose. This check-list was designed to permit the recording 
of (l) definitions of the groups for whom data were to be obtained, (2) a list 
of the criterion data to be obtained for each group, (3) a list of the predictor 
data to be obtained for each group, (4) detailed plans for administration of 
questionnaires, and (5) arrangements for special administrations of achievement 
tests- 

The original plan of the study called for the administration of achieve- 
ment tests in mathematics, physics, or chemistry in institutions where the 
curriculum for the group to be studied included such courses for all students. 
The intention was to administer such examinations during the final examination 
period for the purpose of supplementing regular course grades by objective 
examinations which could be used as additional criteria of academic success. 
Most colleges, however, were unable to cooperate in this phase of the study. 
Indeed, several institutions stipulated that their participation in the study 
was contingent on the agreement that no achievement tests be given. Arrange- 
ments for administering achievement tests were made at only one Institution, 
However, when the tests were actually administered and scored, it was found 
that the difficulty of the tests was not appropriate for the students. No 
specially administered achievement tests were therefore used in the study. 

Groups for Whom Data Were Obtained. At the time when the data were collected, 
it was conceived that three main types of study could be executed: (1) com- 
parisons of male veteran students who entered in the fall of 19^6 with male 
nonveterans and female nonveterans who entered at the same time ; (2) similar 
studies for students who entered in the fan of 19^5; and (3) comparisons of 
male veterans whose college careers were interrupted by war service and who 
returned to college after the war with the best available control group, No 
effort was made to include female veterans, since the preliminary survey 
showed that this group was too small to warrant study. 



"W. Donald Peterson and the senior author of this report were responsi- 
ble for making these arrangements with the cooperating colleges. 

2 It is regretted that individual acknowledgment in this report is pre- 
cluded by the decision that complete anonymity of the colleges should be 
maintained 



70 - ADJUSTMENT TO COLLEGE 



Collection of data on adequate numbers of students who entered in the 
fall of 19^-6 proved to be relatively straightforward; similar efforts to 
obtain groups of adequate size among students entering in 19^5 proved to 
"be unexpectedly difficult. Interestingly enough, it was the nonveteran 
group which was frequently inadequate in size; the operation of selective 
service after the cessation of hostilities, together with the rather 
rigorous requirements for inclusion in the group, appeared to "be the main 
sources of the difficulty* As a result, only one of the twenty-five groups 
actually studied was limited to students entering during the academic year 
19^5-19W. In a few other instances, students entering in the fall of 19^.5 
were pooled with students entering in the fall of 19^-6 for analysis. 

In the collection of data for the study of veterans whose schooling 
was interrupted, the "basic plan was to limit the group to veterans who had 
completed one year prior to interruption and one year after interruption. 
For three groups, this plan was followed. In two additional groups of 
interrupted veterans, however, it was found desirable to increase the flexi- 
bility of the defining pattern in order to obtain larger groups - 

Logically, it would seem desirable to use male nonveterans who had 
experienced a similar interruption for comparison with the interrupted 
veterans o In practice, however, it proved to be impossible to locate more 
than thirty interrupted nonveterans in any one group Consequently, non- 
veterans whose schooling was not interrupted, but who were otherwise as 
similar as possible to the interrupted veterans, were selected for this 
purpose. Thus, for three of the groups, male nonveterans who entered in 
the fall of 19^5 and who completed two full academic years consecutively 
formed the comparison group. For the two interrupted groups including 
veterans having a flexible pattern of interruption, the uninterrupted non- 
veteran comparison group was composed of students entering in the fall of 
1939, Care was taken, of course, to ensure that there was no reason to 
doubt the comparability of grading standards before determining that the 
control group was suitable, 

Criterion Data. The criterion data included average grades, as determined 
by the college in which the student was enrolled, or by calculation from 
data supplied by the college. In addition, grades in specific courses, 
such as English or mathematics, were obtained in oases where the course 
had been taken by all or practically all the students * In deciding whether 
or not a particular variable was to be obtained, it was considered desirable 
to include the variable if there seemed to be a reasonable chance that it 
would be available for practically all students and if it appeared to have 
particular significance. 

Other types of criterion data which were considered and included in 
certain instances were such variables as academic standing (e.g, . "godd 
standing, probation, dropped") and number of visits to the health depart- 
ment for psychiatric problems . Because of the very small proportion of 
students in these extreme categories, however, and for other practical 
reasons, these data were not used in the statistical analysis* Attendance 
at classes was considered but not included in any instance because of varia- 



PLAN OF THE STUDY 71 



bility in the accuracy with which absences were reported, and because of 
differences which were likely to occur between veterans and nonveterans in 
the keeping of attendance records due to certain Veterans Administration 
requirements. 

Predictor Variables. The colleges had originally been selected in part on 
the basis of the availability of adequate predictor data. Scores on some 
intelligence or scholastic aptitude test were available in all colleges; 
the most usual test of this sort was the American Council Psychological 
Examination (ACFE) In most cases , some measure of high school achievement 
was also obtained, such as high school average grade or rank in high school 
class. In addition to the intelligence test score and high school grade, 
other data which were thought to be potentially good predictors of college 
achievement were included, achievement tests used by the college for admis- 
sions or for sectioning being the most common example . In some colleges 
composite scores or predicted freshman grades were available, and these were 
also included. Still other variables of theoretical interest were obtained, 
although they were not expected to be good predictors of achievement. Examples 
of such variables are date of high school graduation and date of taking apti- 
tude and achievement tests; there was considerable variability on these 
factors for veteran students in certain institutions. 

As was true for criterion data, data for any predictor variable was 
included if there seemed to be a reasonably good likelihood that it would 
prove useful. Data on a number of variables were later excluded, mainly be- 
cause more detailed examination showed missing data on an excessive proportion 
of students or because other available variables were judged to be measuring 
much the same thing. 

A variety of methods of recording data were used hand-copying to rosters, 
photostats, microfilm, and punched cards depending upon such considerations 
as cost and the facilities available at the particular institution,. 

Preparation and Administration of the Questionnaire 

The primary purpose of the questionnaire was to discover what factors 
are related to any observed tendency for veteran students to overachieve in 
comparison with nonveterans. The questionnaire which was developed accord- 
ingly contained items relating to as many hypotheses as could be developed 
on an a priori basis for explaining veteran superiority in academic achieve- 
ment. 

Development of the Questionnaire. The first step in the questionnaire develop- 
ment was to jot down ideas about possible reasons for veteran-nonveteran dif- 
ferences in achievement or about factors thought to be generally related to 
academic achievement. Many ideas were contributed by members of the Advisory 
Committee; others were developed in conference with members of the Veterans 
Administration Office of Coordination and Planning. Some of the ideas are 
shown in the following list: 



72 ADJUSTMENT TO COLLEGE 



Type of military service 

Branch, of service 

Number of USAFI courses taken 

Attitudes toward military service 

Presence of physical handicaps 

financial status 

Housing conditions 

Type of preparatory school 

Age 

Marital status 

Number of children 

Study conditions 

Vocational aims (nature and definiteness) 

Attitudes toward teachers 

Extra-curricular activities 

Social maturity 

Peeling of "urgency" 

Tendencies toward neurotic ism 

Reasons (or rationalizations) for going to college 

Satisfaction with college attended 

A number of interviews with veteran students were conducted "by those 
who developed the questionnaire with the view of getting further insights 
and hypotheses, Further leads for questionnaire items were obtained from 
the responses of ninety-nine university presidents to a questionnaire sent 
out in January of 19^7 by President Raymond Walters of the University of 
Cincinnati. President Walters had questioned these college presidents con- 
cerning the problems of the veteran in college, and he generously made the 
completed questionnaires available to the staff for further study. 

Early in March of 19*1.7 two experimental versions of the questionnaire 
were developed and tried out on approximately 400 freshman students at a 
large eastern coeducational university not used in the major part of the study. 
The questionnaires were administered to freshman students in English sections,, 
The two forms differed with respect to method of getting at certain biographi- 
cal information and in the number of free -answer items. The two methods of 
getting biographical information which were tried were (l) a tabular method, 
in which students were instructed to enter, for each year from 1939 to 19W>, 
the number of months spent in certain activities; and (2) a series of multiple- 
choice items covering the same ground. Primarily because of greater ease of 
coding, the latter method was selected for the final form of the questionnaire. 
Comparison of free -answer responses with analogous multiple-choice responses 
provided information as to the adequacy of the latter items and suggested ap- 
propriate revisions. Free-answer comments were also used to suggest addi- 
tional items for inclusion in the questionnaire. 

Another experimental variable was introduced in the trial questionnaire 
administration for the purpose of studying the effect of requiring a signature. 
Half of the Form A questionnaires were administered with a face sheet which 

-T)r. John Clausen was primarily responsible for the development and pretest- 
ing of the questionnaire. 



PLAN OF THE STUDY 73 



required a signature and half with a face sheet which did not require a 
signature. After coding and tabulating the responses, the frequencies with 
which the multiple -choice categories had been checked for signed and unsigned 
questionnaires "were compared. The comparison shoved that requiring a signa- 
ture had little effect on the distribution of responses-. 

No attempt will be mad.e to summarize here the detailed results of the 
questionnaire tryout. However, the results for certain items may be of 
interest. One item on which one might expect the signature to affect stu- 
dents 1 responses was, "How would you rate, as teachers, the faculty members 
who have taught you this past term?" 

The results, for 129 signed and 1^3 unsigned questionnaires, were as 
follows s 

Answer Signed Unsigned 

All are good teachers C&-$ 05$ 

Most are good teachers . 39$ If 3 $ 

Some are good, some rather poor 53$ ^-8$ 

Most are rather poor teachers Qk<f> Qk<jb 

All are rather poor teachers 

Another item where the effect of requiring a signature would seem to 
be extremely important was, "If you could be admitted to (and could get 
housing at) any other university you might choose, do you think you would 
still want to attend the institution at which you are now studying?" The 
results for 329 signed and 142 unsigned questionnaires were as follows: 

Answer Signed Unsigned 

Yes, I'm quite sure I would still 

want to attend the university I 

am now at tending . 68$ 69$ 

I might want to go elsewhere, 

but I f m not sure, 2^$ 23$ 

No, I would definitely attend 
some other university. 

On a few items, which tend to involve the self-esteem of the student 
rather than his evaluation of the institution, slight differences did ap- 
pear. For example, one item was, "How often, during the past four weeks, 
have you gone to evening lectures given by visiting lecturers or local 
faculty members but not required by any specific course?" The results 
from 127 signed and 1^3 unsigned questionnaires were as follows: 

Answer Signed Unsigned 

Not at all 5^$ 65$ 

Once 26$ 18$ 

Twice 13$ 13$ 

Three or more times 07$ 



74 ADJUSTMENT TO COLLEGE 



The questionnaires hich asked for a signature on the face sheet carried 
these instructions; 

"Please print your name in the space below Soon 
after the questionnaires have been collected, a number 
will "be assigned to identify your quest ionnaire, and 
this cover sheet will be torn off. After certain other 
data have been obtained from the Registrar, you will be 
known by number only. No one working with the question- 
naire will know the name of the person who filled it out." 

In the actual study, it was of course necessary to obtain the names of 
the respondents, in order to collate the questionnaire responses with academic 
data. The instructions used were similar to those above (see Appendix Cl and 
C2). In the final form of the questionnaire, however, a separate identity 
sheet was used which was inserted under the front cover of the questionnaire 
booklet. Identity sheets and booklets bore corresponding serial numbers; this 
permitted students to record their names and turn the identity sheets in to the 
administrator separately. The conditions were thus, from the student's point 
of view, slightly more favorable from the standpoint of anonymity than in the 
case of the signed pretest questionnaires. It is therefore judged that the effect 
of requiring the students to identify themselves in the regular administration of 
the questionnaire had small influence on the nature of their responses. 

In the final form of the questionnaire, a few new items were added, a 
few items were eliminated, and a number were revised. The final version of 
the questionnaire which was used in this study is included in Appendix C2 
of this report; it was called the Student Opinion Questionnaire. 

The Student Opinion Questionnaire contains a variety of types of items, 
most "of which were to be answered by all students. One section, however, was 
to be answered by veterans only and another by married students only. In order 
to indicate briefly the general nature of the questionnaire, the various types 
of items will be indicated. 

A number of items deal with facts of personal history and status,, In- 
cluded in this category are such items as kind of secondary school attended, 
date of last full-time attendance in secondary school, length of any full- 
time employment, father's education, type of living quarters, and, for veterans 
only, length of service outside the United States, highest rank or rating, and 
amount of college training ^received while in the service. Items in this cate- 
gory were for the most part objective and factual and would be little influenced 
by the particular time when the questionnaire was administered. 

Another group of items is related to attitudes of students toward college 
and college grades; these items deal with such questions as importance of col- 
lege grades, satisfaction with present institution, enjoyment of studies, 
interest in present courses, and reasons for coming to college. Still another 
category includes items dealing with attitudes toward self: evaluations of 
one's own effort and work habits and judgments of the extent to which worries 
have interfered with college work. Responses to items of these types are com- 
plex Judgments which are highly subjective; since the judgments were made 



PLAN OF THE STUDY 



75 



after the student had some knowledge of his success in college, the inter- 
pretation of the responses as rationalizations cannot "be overlooked. 

One page of the questionnaire is devoted to worries and anxieties. The 
student was to indicate whether he was "bothered very much, some, or little 
by each of a list of common problems, including making ends meet, health, 
concentration, nervousness, relations with members of the opposite sex, etc. 
He was also given an opportunity to list other problems in an open-end ques- 
tion. Another important item was concerned with the student f s disposition 
of time, he was asked to indicate the number of hours spent in a typical week 
in attending classes, studying, athletics, bull sessions, paid employment, 
etc. 

In addition, an item was included which was intended to furnish directly 
a means of classifying veteran students with respect to the importance of the 
educational benefits of the GI Bill in determining college attendance. Several 
other items, intended as check items on this point, were also included* 

Directions for administering the questionnaire were prepared and dis- 
tributed along with the questionnaire to the participating institutions; a 
copy of these directions is included as Appendix Cl. 

Administration of the Questionnaire. ThB method of questionnaire administra- 
tion was selected on the occasion of the visit to the participating colleges. 
Whenever possible, the questionnaire was filled out in groups, using the direc- 
tions for group administration. The most common method was to give the ques- 
tionnaire to freshman English sections or to students in some other course 
which contained a majority of the students desired. This sometimes necessi- 
tated administering considerably more questionnaires than were to be used in 
the study, but was nevertheless the most satisfactory method of getting the 
data,. In other instances, the questionnaire was administered at a special 
assembly of students. Students who belonged in the groups to be studied but 
who did not get the questionnaire in the group administration were reached 
either by mailing the booklet with an appropriate letter and instructions 
or, at some institutions, by calling the students in to the university test- 
ing bureau or personnel office. 

At some universities it did not prove to be possible to employ the group 
method of administration. The most common method then resorted to was that 
of mailing the questionnaires-, With the assistance of the supervisors at 
these institutions, letters were prepared which made use of appeals which 
were thought to be particularly appropriate to the type of student involved. 
Follow-up letters were also sent when necessary in order to improve the 
proportion of returns. In other cases, it wap possible to reach the students 
through their dormitory counselors or through the dean's office. The methods 
used at the various participating colleges are summarized below. 

Eastern City University. 

Freshmen: Administered in English sections. 
Sophomores: Distributed by mail. 



76 ADJUSTMENT TO COLLEGE 



Adams University. 

Freshmen: Distributed by freshmen counselors . 
Sophomores: Distributed by mail. 

Stewart University 

Freshmen: Administered in English sections and 

distributed by mail. 
Sophomores: Distributed by mail* 

Miller University. 

Administered in English sections 

Midwest State University- 
Administered in Business Organization and Economics 
classes. 

Midwest City University. 

Engineering: Administered in classes. 
Liberal Arts: Distributed in classes. 

Littletown State University. 

Administered in English sections. 

Harris University. 

Administered individually by testing bureau. 

Evans University. 

Administered at an assembly of freshmen and 
sophomores. 

Central State University. 

Administered in English sections. 

Turner University. 

Distributed by mail. 

Midwest Technological University. 

Freshmen: Administered in English sections. 
Sophomores: Administered individually by testing 
bureau. 

Middle State University. 

Freshmen: Administered in drawing sections. 
Sophomores: Distributed by mail. 

Douglas University. 

Administered at an assembly. 

Southern Technological University, 
Administered at an -assembly. 

Western State University. 

Administered in English sections. 



PLAN OF THE STUDY 77 



The question of possible bias introduced by incompleteness of question- 
naire returns will be discussed in connection with the results of the ques- 
tionnaire analysis. 



Coding the Questionnaires 

A total of about 2^,000 completed questionnaires were contributed by the 
sixteen institutions participating in the veterans study, an average of about 
1500 for each college. More than half of these questionnaires ere rejected 
before coding, however. The most common reasons for rejecting questionnaires 
were as follows: 

1. The respondent was not a member of one of the defined 
subgroups selected for study. 

2 . The respondent was a member of a subgroup which was 
found to be too small to warrant statistical analysis. 

3- The respondent lacked essential data on predictor 
and criterion variables. 

About 11,000 questionnaires remained after the preliminary editing and were 
coded. 

The purpose of coding is of course to make possible the quantitative 
analysis of data which consist of verbal responses to questions. Many of 
the questionnaire items were preceded, i.e., the multiple-choice responses 
in the printed booklet bore numbers which were used to represent the answers 
selected by the respondents. Such items presented few difficulties. Other 
items, however, were of the free-answer type. The response consisted, in 
some cases, of filling In a number (to represent, for example, the number of 
hours per week spent In studying) . In other cases the response consisted of 
a statement in the respondent's own words reflecting his attitude toward a 
particular question. In coding the responses, each answer must be given a 
number representing one of a number of categories Into which all the responses 
can be classified. The first problem concerning questionnaire coding had 
to do with determining what these categories should be. 

The first steps in determining the categories to be used in coding the 
open-end questions consisted in examining responses made by a sample of stu- 
dents, classifying their responses into trial categories, and trying out 
these categories on a second sample of questionnaires. Jar this purpose, a 



Mr. Robert C. fleers contributed extensively to this .phase of the study In 
designing the coding plan, and in general supervision of the whole program 
of coding. Miss Henrietta Gallagher was the Immediate supervisor of the 
coders . 



78 ADJUSTMENT TO COLLEGE 



sample was drawn of about 100 questionnaires each from Douglas, Harris, 
Central State, Midwest Tech, and Southern Tech universities The experience 
gained earlier from the pretest questionnaires also proved to be useful in 
planning the questionnaire -coding. After this preliminary work, a first 
draft of an outline of procedures and a coding key vere prepared. Following 
discussions of this preliminary outline by the staff and tryout of suggestions 
for revision on additional questionnaires, a complete coding manual was 
developed; those portions of the coding manual which pertain to the major 
open-end questions are reproduced in Appendix C3- 

The coding of Item 45 of the questionnaire deserves special comment. 
This item asks, "Briefly, what are the main changes you would like to see 
made in the program or organization of education at this college, in order to 
help you get what you are after in a college education?" Eight lines were 
provided for the student T s answer. The purpose of including this item was 
to give the student an opportunity to make concrete suggestions or complaints 
about his college in a relatively unstructured situation. The item was sug- 
gested by some of the responses to President Walters' questionnaire, on which 
a number of college officials had commented on the veterans contributing a 
"more mature and purposeful tone" to undergraduate life, showing "broader 
social concepts" than the nonveterans, providing a "more mature outlook," 
and their "willingness to speak their mind and complain when they feel that 
something is wrong with any portion of the University administration." One 
purpose of the item was to compare veterans and nonveterans with regard to 
the number and nature of the responses given; the item is also interesting 
as a means of discovering the criticisms, complaints, and suggestions made 
by undergraduate students about their colleges. 

In the preliminary examination of samples of questionnaires, it was 
found that comments given under Item 36 ("On the whole, how well satisfied 
are you with the kind of education you are getting?") and Item 38 ("Bo you 
ever feel that the things you are studying in college are not really worth 
the time spent on them?") were essentially suggestions of the same nature 
as those made in answer to Item 45- It was therefore decided that for coding 
purposes all comments appended under Items 36 an<3- 38 would be examined in 
conjunction with answers to Item V? in determining the categories of remarks 
made. 

After examination and experimental coding of several hundred question- 
naires, a code comprised of forty-six categories and subcategories was set 
up for Item 4-5. This code was used in the coding of questionnaires for three 
ins titut ions -~Adams, Midwest Tech, and Stewart. A check on the reliability of 
coding which was made at this time indicated that coders were unable to dis- 
criminate among the categories with sufficient accuracy . JOT this reason, and 
because the coding proved to be too time-consuming, an abbreviated code was de- 
vised containing only twenty categories. A plan was also developed for consoli- 
dating the coding already completed for the first three Institutions Into the 
new code The consolidated code was obtained in large part on the basis of plots 
made from the forty -six- category coding, in which the specific disagreements 
of a first and second independent coding could be noted. Q3ie result was that 
the "collapsing" of the codes for the first three institutions removed most 
of the disagreements,, As will be seen later, subsequent studies of coder 



PLAN OF THE STUDY 79 



reliability showed satisfactory reliability,, An excerpt from the revised 
coding manual (Appendix C3) shows the new categories for coding Item kj and 
the method of consolidation used for the first three colleges coded. 

A second problem involved in the coding of Item 4$ responses had to do 
with the maximum number of responses to be coded for any one questionnaire. 
Tabulations for several hundred questionnaires showed that about 90 per cent 
of the respondents gave three or fewer codable comments. It was therefore 
decided to code three comment s, if that many were made, and to reject any 
comments beyond that number a This decision of course resulted in the problem 
of which three comments to code, if more than three were made. A rigorous 
system was therefore set up designed to ensure that any two persons inde- 
pendently coding the same questionnaire would accept exactly the same comments 
for coding, and also to make sure that comments to Item k^ would not be over- 
emphasized to the exclusion of comments to Items 36 and 38. 

Somewhat similar but less complex problems were involved in the coding 
of other free-answer questionnaire items. The coding of these items need 
not be discussed here in detail, since the section of the Coding Manual and 
Coding Key which are reproduced in the Appendix will make clear the solutions 
which were accepted,, 

Coders were selected with considerable care. The maximum number working 
at any one time was nine and the average five All were women college gradu- 
ates. Accuracy rather than speed was emphasized; a good deal of attention 
was devoted to giving the coders an appreciation of the general objectives 
of the study and to giving them as a group a common basis of understanding 
with regard to the various coding categories,, Morale was maintained at a 
high level throughout the three months required for the coding operation, 
despite the generally tedious nature of the work. 

The coding of all questionnaire items was checked throughout the coding 
period. The coding of every item was checked by a second coder, and a spot- 
check of every fifth questionnaire was made by a supervisor. In addition, 
occasional studies were made of coding reliability for free-answer items by 
having a sample of questionnaires independently coded by two coders; then 
scatter plots were made, using the questionnaire categories assigned by the 
first coder and the second coder as the variables. 

In connection with the coding of Item ^5, a rather elaborate procedure 
was developed to ensure consistency of coding, which routinely involved inde- 
pendent coding, by two coders, of the same responses. The first coder wrote 
her selected code numbers on the left-hand margin of the front cover of the 
questionnaire booklet; the second coder recorded her code numbers in the ap- 
propriate code boxes. Then a clerk checked the two sets of code numbers by 
folding the booklet in such a way as to bring the two sets of numbers into 
juxtaposition. Disagreements were then examined by two different coders who 
in collaboration tried to agree on how the doubtful responses should be coded. 
In the occasional instances where no agreement could be reached, the final 
decision was made by a supervisor. Studies of the consistency of coding this 
item were made by comparing the two independent judgments of the first two 
coders. Agreement was found to be high even at this point before discussion 
of the disagreements. 



80 ADJUSTMENT TO COLLiCE 



Figure 9 illustrates a typical plot made for the purpose of studying 
the reliability of coding Item if 5. In this example, based on 166 question- 
naires from Ivans University, the tvo independent judgments made by the 
coders agreed in 91 per cent of the oases. (The agreement is slightly en- 
hanced by the fact that the matching was based on sets of three codes rather 
than individual paired codes The entries in the diagonal represent the 
oases where there was agreement; those off the diagonal indicate the number 
and nature of the disagreements. (The interpretation of the numerical codes 
is given in Appendix C3-) It may be noted that the agreement to be expected 
by chance, in this Figure, is about 15 per cent. The code "T" was used to 
indicate no response. It will be noted that about half of the disagreements 
were in the Y categories; in other words, much of the disagreement had to do 
with whether or not a particular , comment could be given a specific code under 
the rules laid down in the Coding Manual. A total of forty- five of the 498 
responses were in disagreement The disagreements in this nine per cent of 
the cases were, in accordance with the standard procedure outlined above, 
resolved by conference of coders not involved in the original coding. 

Plots similar to that shown in Figure 9 were made for a number of separate 
samples of questionnaires. The over -all percentage of agreement in coding 
Item ^5 ( a "t this stage prior to study of disagreements) was found, on the 
basis of tabulations of 1567 questionnaires, to be 89.6 per cent. The varia- 
tion from sample to sample is indicated in Table 1 below, (in showing numbers 
of cases, MV means male veteran, MR means male nonveteran, and ITT means female 
nonveteranTJ" 

Since these data for the coding of Item 4 5 are based on the stage prior 
to the discussion of the responses not agreed upon, it is Judged that the 
final reliability of coding this item was quite satisfactory. It of course 
should not be assumed that the disagreements were entirely the fault of the 
coders; many of the responses not agreed upon were sufficiently ambiguous 
that it was sometimes quite arbitrary as to whether they should be coded at 
all, or which coding category should be used. 

The accuracy of coding other open-end questionnaire items was studied 
in a similar manner. Item 11 ("What kind of work are you planning to do 
after you finish your studies? Describe the job as specifically as you can.") 
was the one other item where a high degree of subjective judgment was required 
of the coders and where a large proportion of the students gave responses 
For three samples, comprising a total of 308 questionnaires, the first and 
second coders, working independently, agreed in 88.0 per cent of the cases. 
The percentage of agreement for the three samples was 92.0 per cent, 89.3 
per cent, and 83.0 per cent. 

It should be remarked that the accuracy of coding undoubtedly increased 
as the coding operation continued. Part of the training of coders inevitably 
occurred on the job. Some of the refinements as to procedures were instituted 
after coding had progressed through several institutions. Coding was probably 
not uniformly accurate for the questionnaires from a particular institution, 
since time was required for the coders to familiarize themselves with the 
unique types of responses which were likely to arise among students at a 



PLAN OF THE STUDY 



81 



1 


OJOonrHtr\onrHt*-onOHO HOOOOO LAVO rH O ON 
\O OJ H on rH OJ H H H -= O on 
H H 


00 


vols 


P 






S H 


O 






P5 O -P 


H O 


rH rH rH rH rH H -=t 


o 






on 


_jHJ- 


fe -P "^1 


O 


H 


H 


OJ -H 








X-v H 


N"\ 






pr"1 r-J (-Q 


H 






W CQ 


O 






H * 3 


JH O 


H H H 

ON 


on 

ON 


03 -P 
O 0) 


@ 






O o -p" 


OJ 






5 ^ H 


pq 






R tJ ft 








ffi d) 


O 






^jj jH QJ 


JM 





Q 


r^ oh J r{ 




-d- 


-d" 


S ,-] _|j 


EH 






CM tO 


CO 






03 M 








CQ 03 8 


O 


OJ H rH 


-=h 


K OJ 3 


t 






H rH g 


-zf- 


H H on 


LTN 


g g P 


vo 


H 


H 


O fl 








O U3 -H 


on 









VO 
OJ 







LT\ O <D 

-Sf- rH 


\Q 






(D 








fe a -p 


H 







O <U H 


\o 






H rH 03 


s 


H ON H 


H 
H 


1-i 

D 


OJ 


rH 


H 




LT\ 






CO 








[TI P| QJ 


LTN 




O 


O H nd 

625 o 













H 


H 


O Q 


LTN 














H EH <U 


rH 


rH 


rH 


O JM 


O 




O 


{JH 


-d- 






P-j tH D 










OJ 


H OJ 


on 

rH 


H eg -P 


on 


iH 




P ^ 


rH 


H VD 


t*~ 


p3 rH O 


on 






^ rH 


o 




o 


fi CD 


on 






H }3 rH ' 


OJ 
OJ 


rH H ON H 


OJ 
OJ 


<JJ O -P 


H 


H H 


CU 


B JS S 


OJ 


rH 


H 


PR -P O 










o 


ON H OJ H H rH 


LT\ 


p| CO f 


OJ 


OJ 


on 


H CO QJ tjQ 










H 


H H H H OJ CM 


"3 


i B 1 <u 


O 


OJ f- H on 


on 


m ^ "S -P 


rH 


H 




O O -H bp 


rH 


03 OJ H H 


OJ 


H H *=! 


O 


ir\ 


V> 










<d co OJ o 




HOHOHOJOHOJQHOrHOJOHOJ on^t O -"" *^* " - 

oHHOJOjojonon on^- ^j- u-\ LT\ ur\vo VOMDVDVJOD H pq pq 


^ 


PH O/ *H ,3 




hH O !>s O JH O 


S 










O\ 




s i Pi 








^ ^ N_, 




O 


H30QQ OKEQOaS IS OHIODO 


s 



82 



ADJUSTMENT TO COLLEGE 



Table 1 

Percentage of Agreement in Initial Coding of 
Questionnaire Item 45 ** Several Samples 



University 


Number of Cases 


Per Cent of 
Agreement 


MV 


MN 


FN 


Total 


Midwest City 


11 


18 


29 


58 


94.8 


Southern Tech 


88 


31 





119 


93-8 


Midwest City 


58 


54 


25 


137 


92.9 


Central State 


8? 


24 


ko 


151 


91-6 


Evans 


88 


33 


^ 


166 


91.0 


Miller 


281 








281 


89.8 


Turner 


19 


40 


94 


153 


8?.l 


Southern Tech 


22 


42 





64 


87.0 


Miller 





206 





206 


86.9 


Eastern City 


23 


56 


154 


233 


85.7 



PLAN OF THE STUDY 



particular college. It was felt, however, that the limitations in relia- 
bility are those which characterize free-answer questionnaire coding gener- 
ally; and that in general, the coding procedure was sufficiently sound to 
justify reasonable confidence in the results for these items. 

The large majority of the questionnaire items were pre-coded, so that 
the question of coding reliability did not arise. In the few free-answer 
items not discussed above (e.g., highest rank attained during service, prob- 
lems not included on the check-list) it was judged that the coding was so 
nearly objective or the responses were so few that formal reliability studies 
were not warranted. It appears, then, that reliability of coding is not a 
matter of particular concern except on Items 11 and ^5. 



Analysis of Academic Data 



The primary purpose of the analysis of academic data was to determine 
whether or not veteran students made higher grades, in relation to their 
ability, than did nonveteran students. In order to obtain as precise a 
comparison as possible, it was considered essential to analyze data sepa- 
rately for each university, and, in institutions of complex organization, 
separately for each college or division. Analyses were based on groups 
homogeneous with respect to institution attended, the division within that 
university in which enrolled, and time of entrance in college. Each such 
group ordinarily included both veteran and nonveteran male students; in 
some cases female nonveteran students were also included. A total of twenty- 
five separate groups were defined by this process. The groups within each 
institution are shown in Table 2. 

Colleges where differentiation into various divisions is delayed until 
after the freshman or sophomore year are classified as colleges of arts and 
science in this study. In all, seventeen groups were from colleges of arts 
and science (so defined), four were from engineering colleges, two from 
colleges of business, one from an agricultural college, and six groups in- 
volved interrupted veterans from arts and science and technological colleges. 

Preparation of the Academic Data. In defining the specific students from each 
college to be included in the statistical analyses and in making the numerous 
detailed decisions required before analysis could begin, primary consideratior 
was given to two objectives: 

a. to bring into sharp focus any basic difference in relative 
achievement between veteran and nonveteran students; and 

b. to provide data which would be as comparable as possible 
for the twenty-five groups chosen for analysis. 

For practical reasons, the detailed planning of the study and the preparation 
of the academic data for analysis developed concurrently. In order that all 
analyses be based on the same students within a particular college group, the 



84 



ADJUSTMENT TO COLLEGE 



Table 2 
THE TWENTT-FIVE GROUPS INCLUDED IN THE STUDY, , ARRANGED ACCORDING TO UNIVERSITY 



University 


Division 


Time of 
Entrance 


Number of Cases in Each Sample 


Academic 
W* MN M 


Questionnaire 
W* MN FN 


Adams 


Arts & Science 


Fall 1946 


531 694 


^65 612 


n 


11 ! 


Fall 1945** 


134 111 


135 97 


Central State 


tt tt 


Fall 191*6 


466 166 


k66 166 272 


it r> 


If TJ 


Fall or 


135 59 


63 






Winter 1945 






Douglas 


tt n 


Fall 1946 


77 119 93 


77 119 93 


Eastern City 


tt tt 


ti 


53 147 --- 


53 3A-7 285 


M at 


tt tt 


Fall 1945** 


70 99 


70 99 


Evens 


It tt 


Fall 1946 


283 94 


283 9^ 159 


Harris 


tt tt 


n tt 


105 lW - 


105 146 213 


Littletown State 


tt tl 


it it 


103 107 


103 107 134 


H It 


Business 


it tt 


llj-2 65 


142 65 


Middle State 


Engineering 


tt tt 


352 98 


352 98 


Midvest City 


Arts & Science 


it tt 


83 72 


83 72 


u t 


Engineering 


it tt 


167 171 


167 171 


Midvest State 


Business 


n tt 


232 58 


232 58 


Midwest Techno- 


Agriculture 


n tt 


UK) 102 


140 100 


logical 










TT IT 


tt 


Fall 1939** 


57 106 


53 


It ft 


Engineering 


Fall 1946 


271 128 


267 122 


n TI 


n 


Fall 1939** 


3AQ 215 


118 


Miller 


Arts & Science 


Fall 1946 


te5 193 


425 193 216 


Southern Techno- 


Engineering 


Fall 1945 or 


120 50 


120 50 


logical 




Fall 1946 






Stevart 


Arts & Science 


Fall 1946 


187 3^ 


150 323 


ti 


rt it 


Fall 1945** 


55 70 


32 40 


Turner 


tt ft 


Fall 1945 or 


100 101 


98 94 121 






Fall 1946 






Western State 


n tt 


Fall 1946 


lf-33 222 W2 


272 148 37^ 



Hxe abbreviations MV^ MN and FN stand for male veteran, male nonveteran., and 
female nonveteran respectively. 

X- 

The date given is for nonveterans. The veterans with vhom they are compared 
entered college at various times and had their college careers interrupted by 
war service. 



PLAN OF THE STUDY 85 



inclusion of any variable on which data were not complete required the ex- 
clusion from the analysis of students who lacked data on that variable. 
Thus in the final determination of the members of each of the twenty-five 
subgroups it was necessary to balance the information to be obtained from 
a particular variable against the proportion of cases to be dropped and 
the size and representativeness of the sample remaining if the variable 
were included. Fortunately, two facts aided greatly in making the deci- 
sions: first, that many of the scores most appropriate for the study were 
available for substantially all cases; and second, that it was necessary or 
desirable to exclude students lacking data on certain variables (e.g., first- 
year grades) as a means of defining the sample. 

Generalized Description of the Samples. The students included in the various 
studies were male veteran students, male nonveteran student s, or female non- 
veteran students who belonged to one of the twenty-five groups chosen for 
study. These groups were defined in terms of the college or university, the 
college class (e.g., freshman in the fall of 19^6) and the university division 
(e.g., College of Arts and Science ) In general, each student included had 
earned at least a specified amount of college credit at specified periods 
of time. He had complete data on all variables chosen for statistical analy- 
sis in his group o He had not attended, as a civilian student, any other col- 
lege or university; and in most groups, he had not received substantial 
credit (10 quarter hours) in specific courses for college work he had taken 
while in the armed services. In universities having several divisions, he 
had been enrolled in the designated division during the entire defined period, 
which was typically one academic year. Except for the editing on credit for 
armed service college training programs, the procedures were uniform for 
veterans and nonveterans. Typically, the veterans and nonveterans were not 
separated until the editing was completed. 

Direct comparisons of grades of veterans and nonveterans (excepting 
only the "interrupted" groups) are thus limited to grades earned at the 
same time, in the same university, and in the same division within the uni- 
versity by veteran and nonveteran students. Students who have done part of 
their academic work at another college are excluded and veteran students who 
had extensive college training in basic academic subjects during their mili- 
tary career are usually excluded from the comparisons. 

One additional detail of procedure should be noted here : the determina- 
tion of veteran status* For making this classification, two main sources of 
data were possible: first, the student's own definition of his status in 
response to Item 7 of the questionnaire; and second, an indication by his 
college regarding his veteran status The exclusive use of the first of 
these would have led to the elimination from the study of all students who 
did not complete questionnaires. On the other hand, exclusive use of the 
second source in accordance with a uniform definition would have involved 
excessive practical difficulties. Students having data from both sources 
were classified as follows: the data provided by the college were used to 
determine veteran status for the large freshman groups at Adams, Stewart, 
and Western State Universities and for the two interrupted veteran groups 
and their controls at Midwest Technological University; in all other groups, 
the questionnaire response was used as the chief basis of classification,, 



86 ADJUSTMENT TO COLLEGE 



Comparison of the two sources in a number of institutions showed a 
high degree of agreement between the two methods of classification; accord- 
ingly, the veteran status of a small proportion of students who lacked ques- 
tionnaires vas determined from data provided by the colleges in several of 
the groups. 

The first step in the actual analysis was to compute the intercorrela- 
tiona of predictor and criterion variables 5 JOT each group, separate tables 
of inter correlations were computed for male veterans, male nonveterans, and, 
if they were included, female nonveterans- For certain institutions (Midwest 
Tech, Middle State, Western State, Douglas, and Mams) a larger number of vari- 
ables was included tten for the remaining institutions The more complete 
analyses were in, general made in instances where they were justified by the 
jsdae of the sample or by the availability of variables thought to be of par- 
ticular interest. The purpose was not only to study the relationships of 
the predictors to college achievement, but also to provide information for 
use in selecting variables to be employed in the analysis of covariance* 

The analysis of covariance method employed in the study is one developed 
by S. S Vilka. The method permits one to test successively three hypotheses 
regarding the regressions of a criterion, on a predictor for two (or more) 
groups. Hypothesis A is the hypothesis that the errors of estimate about 
the regression lines (or planes) are the same for both (or all) groups; 
hypothesis B is the hypothesis that the slopes of the regression lines (or 
planes) are the saxnej and hypothesis C is the hypothesis that the intercepts 
of the regression lines (or planea) on the criterion axis are the same. The 
test of Hypothesis B is legitimately applied if Hypothesis A is not disproved, 
and, similarly, the test of Hypothesis C is legitimately applied if the 
liypothesis that the regression slopes are alike is not disproved* If the 
intercepts of the regression lines (or planes) on the criterion axis do 
prove to be significantly different, the interpretation of course is that 
the members of one group show a higfter perf oamance on the criterion than do 
fflBJribers of the aecond group -who are of similar ability as measured by pre- 
dictor scores. The method has been generalized and computational procedures 
have been evolved for regression planes based on more than one predictor as 
well as for the single predictor situation*" 



extensive tabulations and analyses required by this study were executed 
by the Department of Statistical Analysis, of which Dr. Ledyard B Tucker is 
Bead. Particular acknowledgments are also due to Mr, Hairy Garrison, who 
W&s in charge of the IBM work, to Miss Henrietta Gallagher, who coordinated 
the analysis, and to Mrs, Judith Aronson, Head of the Confuting Section*, 



contribution of Dr, Lsdyard R Tucker, who developed an effective, sys- 
tematic procedure for coasting the necessary constants aided subsl^ntially 
in making the use of this procedure feasible. 



PLAN OF THE STUDY 87 



- In this study, the regression lines (or planes) to be compared are the 
regressions of freshman average grades on the predictor or predictors chosen 
for a particular group for male veteran and male nonveteran students in that 
group. Disproof of Bypothesis C is evidence that male veterans and nonveter- 
ans of similar ability differ in their college achievement as measured by 
grades <> 

The conceptions involved in the analysis of covariance procedure suggest 
a convenient and meaningful method for evaluating the amount of the difference 
between two groups * The procedure yields an estimate of the percentage of 
veterans excelling the average nonveteran, after allowing for differences in 
ability . 

If it is found (in the test of Hypothesis B) that the regression lines 
or' planes may be assumed to be parallel, then they may actually be made par- 
allel by calculating a common slope . (The common slopes were already avail- 
able from the test of Hypothesis BO Each of the parallel regression lines 
or planes will intersect the criterion axis at some point j and the difference 
between the two points (the intercepts), measured on the criterion axis, is 
a measure of the extent to which one group excels the other. Since the units 
used for the criterion vary from one institution to another, it is desirable 
to find a measure which is more nearly comparable from group to group * Such 
a measure would be provided by dividing the obtained difference in intercepts 
by a suitable standard deviation . It was decided that the square root of the 
pooled error of estimate, based on the common slopes used in computing these 
intercepts, would yield the most appropriate denominator. This of course 
assumes that the pooled error of estimate is appropriate for both sugroupsi 
the tests of Hypotheses A and B provide a check on this assumption. It is 
apparent that the resulting measure is a standard score whose unit is a 
standard error of estimate. From such units the proportion of veterans who 
excel the average nonveteran may be estimated by use of a table of normal 
curve areas . 

This concept is illustrated in Figure 10, The left-hand distribution is 
for nonveterans and the right-hand distribution is for veterans; the means 
of these distributions are separated by an amount equal to the difference 
between the regression lines (or planes) expressed in standard error of esti- 
mate units. The proportion of cases falling in the diagonally shaded area 
may be obtained from a table of the normal curve j this value is an appro- 
priate estimate of the proportion of veterans excelling the average nonveteran 
when ability is assumed to be equivalent In the comparisons of male veterans 
with male nonveterans, the per cent of veterans excelling the average non- 
veteran is always reported. Percentages of less than 50 then indicate superi- 
ority of the nonveteran subgroup., Percentages greater than 50 indicate su- 
periority of the veteran subgroup. 

The criterion which was uniformly used in the analyses of covariance 
was freshman average grade, point-hour ratio, or some similar index based 
on course grades obtained during the freshman year. Grades in specific 
courses were also used in a limited number of analysis of covariance studies. 
The predictors varied from school to school, but typically two measures, com- 
bined through use of multiple correlation techniques, were employed s a 



88 



ADJUSTMENT TO COLLEGE 



Diff 




Veterans who score 
above average 
nonveteran in AAG 

Veterans who score 
below average 
nonveteran in AAG 



College Grades 
(Adjusted to allow for Ability Differences) 



FIGURE 10. DETERMINATION OF THE PROPORTION OF VETERANS EXCELLING 
THE AVERAGE NOWVETERAN^ ASSUMING NORMAL DISTRIBUTION AND EQUAL 
STANDARD DEVIATIONS . 



PLAN OF THE STUDY 89 



measure of high school achievement, such as average grade or rank in class, 
and a test or composite of test scores. The most commonly used test "was 
the American Council Psychological Examination, Scores on College Entrance 
Examination Board tests were used for two colleges, and in a number of 
instances a composite score based on various scholastic aptitude and 
achievement tests or tests of tool skills "was employed. 

A slightly different approach -was used in the case of interrupted 
veterans. Here it was assumed that the veteran's freshman grades, earned 
"before war service, furnished the best possible predictor of grades earned 
after discharge from military service. In the typical study of interrupted 
veterans, therefore, the regressions of second-semester sophomore grades on 
first-term freshman grades for interrupted veterans and uninterrupted non- 
veterans were compared. The semester or quarter average grades falling be* 
tween the first term of the freshman year^and the Jast term of the sophomore 
year vere not used in the analysis. The term just prior to induction, it was 
thought, might have suffered because of knowledge of the impending induction 
into the armed forces; and the term occurring immediately after the return 
from service might not have been typical, either because of need to readjust 
to academic life or because of a temporary enthusiasm leading to overachieve- 
ment. 

A total of fifty-two tables of inter correlations, involving varying 
numbers of variables, and thirty-two analyses of covariance were computed 
in this phase of the study. The results of the analysis of academic data 
are reported in Chapter III. 

Analysis of Questionnaire Data 

The purposes of the questionnaire analysis may be stated, as follows: 

1. To provide a summary of the opinions, attitudes, and 
biographical background of veteran and nonveteran stu- 
dents, as they are reflected in the Student Opinion 
Questionnaire 

2. To see if the grades of students, equated in ability, 
are related to characteristics measured by the question- 
naire items. 

3- To see if relatively more veterans than nonveterans 
possess the characteristics which are associated with 
superior Adjusted Average Grade. 



90 ADJUSTMENT TO COLLEGE 



The first purpose, to provide a summary of the opinions of students as 
they are reflected in the questionnaire, may "be achieved merely by reporting 
the frequencies with which the various response categories of the question- 
naire items are chosen. The attainment of the other purposes cannot so easily 
be achieved. 

It would of course be possible to report, in addition to frequencies, 
the mean aptitude test score and the mean freshman grade of those students 
who chose each particular response category to each item. One could then 
draw certain conclusions about the relation of each item, to college apti- 
tude and to college grades. It was felt, however, that more meaningful re- 
sults could be obtained if the items were analyzed in the light of a measure 
based on both aptitude and college achievement. A measure called Adjusted 
Average Grade was therefore employed in the analysis. The Adjusted Average 
Grade (AAG) is a measure, based on the standard error of estimate, of the 
extent to which a student "overachieves" or "underachieves"; it indicates 
the extent to which his grade falls above or below the regression <Line for 
his group, including both veteran and nonveteran students. Noting ^tnat the 
standard deviation of these scores is the standard error of estimate and 
that the mean deviation is zero, it is a relatively straightforward matter 
to obtain deviation scores having any desired mean and standard deviation. 
In this study, 130 was chosen for the mean and kO for the standard deviation, 
rhe computational procedures are outlined in Appendix B2. 

MG*s were not computed for all groups used in the analysis of academic 
lata; it was judged that the labor of computing AAG and tabulating the re- 
sults was justified only for groups which were of reasonable size and which 
rere particularly appropriate from the standpoint of the objectives of this 
study. AAG's were not computed for female students in any college nor for 
:hose groups which because of small size, lack of a control group, or other 
jonsiderations were least useful. In all, sixteen groups were judged to be 
mitable for the computations of AAG's. 

For every item, then, a table was prepared showing, for each of the 
;wenty-five groups, the per cent of male veterans, male nonveterans, and 
"emale nonveterans (if any) who chose each category of that item. In addi- 
iion, for the sixteen selected groups, the mean AAG is shown separately for 
ihose male veterans and male nonveterans in each group who chose each cate- 
;ory of the item. These detailed tables are included in Appendix A. 

The second purpose of the questionnaire analysis was to determine 
whether or not a statistically significant degree of relationship exists 
between an item and AAG. A method of attacking this problem was desired 
whicTi would be feasible in the light of the large number of subgroups and 
of questionnaire items involved in this study. The solution found was an 
adaptation of the F-test, For each item, the mean AAG of students who chose 
a particular response was compared with the mean AAG of the students who 
chose other responses to the item; this procedure was carried out separately 
for veteran and nonveteran students in each subgroup for whidh AAG's were 
computed. Tables were devised to facilitate the application of the F-test. 
(The procedure followed is described in greater detail in Appendix BJ.) 



PLAN OF THE STUDY 9! 



Each test thus resulted in a determination of whether an association 
(stronger than would "be expected by chance) existed "between the item and 
AAG for a particular subgroup of veterans or nonveterans. 

Achievement of the third purpose required a method by which the back- 
ground, attitudes, and other personal qualities of veterans and nonveterans 
could be tied in with the results from analyses of academic data. The aim 
is to try to identify those characteristics which may help to account for 
any difference found between veterans and nonveterans in college achieve- 
ment relative to ability. 

In view of the large number of questionnaire items and college groups 
involved, it was obviously necessary to choose a method of analysis which 
would be simple and yet which would give a straightforward answer to the 
question of whether either of the groups veterans or nonveterans tended 
more often than the other to possess the characteristics which are associ- 
ated with superior Adjusted Average Grade. 

The method chosen is based on the fact that studying several college 
groups constitutes several replications of an experiment. If it can be 
assumed that there is equal probability of obtaining positive or negative 
findings for any one group, then a sign test may be used in evaluating the 
statistical significance of finding any particular number of the results 
to be positive. 1 An extensive discussion of the sign test has been pre- 
sented by Dixon and Mood (28) . 

We are considering the problem of whether veteran students possess 
more often than nonveterans a characteristic which is associated with a 
tendency to earn high grades relative to ability. If they do, we may per- 
haps assume that that characteristic helps to explain a tendency for the 
veterans to earn higher grades, relative to ability, than nonveteran stu- 
dents. The actual procedure used in the study of a particular question- 
naire item was as follows. 

First, all instances were identified where for a particular college 
group the mean Adjusted Average Grade of both veterans and nonveterans who 
chose a particular response category was above average for that group. 
For example, it might be found that at college A veterans choosing the 
first response to Item 36 were above the average of all veterans at that 
college in mean AAG, and a similar result was obtained for colleges B, C, 
D, E, F, G, H, and J. Also for nonveterans at college A let us suppose 
that the mean AAG of those giving the same response was above the average 
of all nonveterans at College A in mean AAG, and a similar result was 
found for colleges B, C, D, E, F, G, H, and J (the same colleges). The 
next step is to examine the percentages of veterans and nonveterans giving 
the first response to Item 36 in these nine colleges in order to determine 
in how many instances the percentage is higher for veterans. Suppose that 
relatively more nonveterans than veterans at College A were found to give 

^Acknowledgment is made to Professor S. S. Wilks, who suggested the use of 
a sign test in this portion of the analysis. 



92 ADJUSTMENT TO COLLEGE 



Response 1 to Item 36 > but in all other colleges the percentage was higher 
for veterans. The final step is to determine what is the probability of 
getting by chance the obtained number of instances where relatively more 
veterans than nonveterans chose Response 1. Referring to the tables pro- 
vided by Dixon and Mood,, we find that the probability of getting eight out 
of nine differences with the same sign is five in a hundred) so we may say 
that our finding is significant at the 5$ level. The possibility that 
veterans tend to be superior in AAG because they possess more than nonveterans 
the characteristic described by the first response to Item 36 may then be 
considered. 

It will be apparent that it is also possible to consider category mean 
AAG's which are below average rather than above. For an item with only two 
categories, the interpretation of the results of a test based on the below- 
average categories would in general be the same as one based on above -average 
categories. But on an item with more than two categories, especially if 
the categories are not ordered (do not form a continuum) or if the relation 
of the item to AAG is curvilinear, the interpretation would not necessarily 
be the same. Tests could of course be made for both the above- and below- 
average subgroups. However, it was felt that in order to avoid undue 
capitalization on chance it would be preferable to make only one test 
per item. The following procedure was therefore employed. The number of 
instances where both subgroups were above average in mean AAG was deter- 
mined; then the number of instances where both subgroups were below average 
was determined. The larger number of subgroups was chosen for use in the 
subsequent portion of the test. 

A graphic illustration may make the method clearer. In Figure 11 re- 
sults are plotted for a hypothetical three -category item. Each point is 
plotted to show the percentage of veterans and the percentage of nonveterans 
at a particular college who gave one of the responses to the item. If the 
percentages for veterans and nonveterans are the same, the point will 
obviously fall on the diagonal. The solid circles represent colleges and 
categories for which both veterans and nonveterans were above average in 
AAG. The open circles represent groups and categories for which both 
veterans and nonveterans were below average in AAG. The crosses represent 
the remaining group-categories , for which only veterans or only nonveterans 
were above average in AAG, or where one or both groups were equal to the 
average . 

We find that there are eleven open circles and only nine solid 
circles. The open circles, representing below-average mean AAG's, are 
therefore chosen for use in the test. Of these open circles we find that 
ten are below the diagonal, one falls on the diagonal, and none are 
above } i.e., out of eleven instances of category means which are below 
average, in ten cases veterans possess the characteristic relatively less 
often than nonveterans . Getting 10 1/2 plus signs out of eleven would 
occur by chance less than once in a hundred; so we may say that whatever 
characteristic is assessed by this item may help to account for veteran 
superiority in grades relative to ability. 



PLAN OF THE STUDY 



J.UU- 

QO- 






















$ 




y\j 
80- 


















\ 


$ 


& 




70 H 






9 


jffi 




-f 






/ 


6.-^ 






60 
















/ 










ern_. 






& 


. 




/ 


/ 












P u 
40- 









4- 


X 


3 






o 








"3(X- 








y 


i 
















20- 






'/ 






4- 




O 














/ 




o 

O <J 


-f 
















1O- 




/ 










9 


Both ] 

1Rrv+-"h 1 


W and 
vnr 5=1 nfl 


MPT abc 

MAT "ho" 


Dve av 


srage in AAG 




X 


+ 






O 




4. 


MV an 
below 


i MN n 
avera/ 


Dt "botl 

a;e In J- 


i above 

IAG 


3 or both 




' r If 
















1 









10 20 30 40 50 60 ?0 80 

Per Cent of Nbnveterans 



90 



100 



FIGURE 11. IILUSTRATIOIT OF A SIGN TEST FOR TESTING THE 
HYPOTHESIS THAT MORE MALE VETERANS (MV) THAN MAUS NONVETERANS 
(MHT) POSSESS THE CHARACTEEISTICS ASSOCIATED WITH SUPERIOR 
ADJUSTED AVERAGE GRADE (AAG). 



94 ADJUSTMENT TO COLLEGE 



Essentially, then, the procedure is quite simple,, Por each item,, the 
answer is sought to one of the following questions s (l) Are veterans more 
likely than nonveterans to choose responses associated with better -than- 
average AAG*s? or (2) Are veterans more likely than nonveterans to choose 
responses associated -with poorer-than-average AAG's? Which question was 
asked depended only on which question provided the larger number of con- 
sistent responses. In either case, it was only necessary to count the 
number of times the veterans were higher on the critical responses* The 
sign test then permitted an easy interpretation of this result in terms of 
the usual conceptions of statistical significance 

It y be added that the convenience of the sign test and the relatively 
single assumptions required for its use led to its extensive application in 
this study to other questions than whether or not questionnaire items con- 
tributed to an understanding of veteran-nonveteran differences in Adjusted 
Average Grades* For example, if veterans irho chose a particular question- 
naire response earned better~than~average AAG's in eleven out of twelve 
veteran groups, use of the sign test permitted the conclusion that,, for 
veterans, the response was associated with Adjusted Average Grade, and that 
ths association was significant at the 1$ level . 

The fact that the sign test was ordinarily based directly on a rather 
small number of observations, usually about twelve, is less serious when it 
is recalled that each of the observations which entered into the count was 
based on a substantial number of cases. 

In applying this technique to determining whether each questionnaire 
item contributed to- an understanding of veteran-nonveteran differences, 
twelve of the twenty-five groups were selected. These groups, selected for 
inclusion in this phase of the questionnaire analysis, were the basic twelve 
groups emphasized in the analysis and interpretation of other questionnaire 
results. All were limited to freshman students who entered in the fall of 
L946. In all twelve groups, first-year college average grade is the cri- 
terion. Each of the twelve groups represents a different university. Hone 
3f the twelve groups has fewer than 75 members in either the male veteran 
JT the mle nonveteran group? most of the groups are much larger than this 
ainiaum. Of these twelve "sign-test" colleges, stx are private colleges, 
ind six are public? nine are arts and science groups, while three are enuri- 
leering groups, 

Presentation of Results of the Questionnaire Analysis 

Because of the relatively large number of results obtained for each 
questionnaire item, and because of their generally systematic pattern, a 
standard method of presenting these results was developed. Variations in 
tha method of presentation were of course necessary In items which did not 
fit the general pattern, e.g., questions which TOre answered only by veterans. 

The results of the questionnaire analysis are presented in two forms : 
E series of graphs, interspersed through the chapters IV through X, and a 
series of tables, presented in Appendix A The graphs are intended to 



PLAN OF THE STUDY 95 



portray the more general trends, while the tables contain the detailed re- 
sults of the analysis. A separate table is presented for each questionnaire 
item. The tabular presentations of the results are intended for those 
readers vho are interested in making a detailed study of a particular item 
or vho are interested in tracing through the tables the detailed results 
for a particular college or group of colleges. A full explanation of these 
tables vill be found in Appendix A. 

The graphs presented in chapters IV through X are based only on the re- 
sults for twelve college groups, selected so as to meet the following re- 
quirements: first, they include only freshman students who entered in 
the fall of 19^6,' second, in each group there are at least 75 veterans 
and at least 75 nonveterans vho filled out the questionnaire; and third, 
not more than one group is included from any one university. Nine groups 
of students vho entered arts and science colleges in the fall of 19^6 and 
three groups vho entered engineering colleges in the fall of 19^6 are in- 
cluded. The nine arts college groups are from Central State, Evans, Western 
State, Miller, Stevart, Harris, Adams, Douglas, and Littletovn State uni- 
versities. The three engineering groups are from Midwest Tech, Middle 
State, and Midwest City universities* Six private and six publicly- 
supported universities are included. 

The top portion of each graph presents the per cent of students in 
each of the tvelve basic groups vho chose each category of the questionnaire 
item. A class -interval of 5 vas used in constructing these distributions. 
Male veteran and male nonveteran results are shown separately, in the case 
of items anevered by both. The arrowhead at the left of each cluster of 
points represents the per cent selecting the response in the median sub- 
group. The top portion of the figure may be used for three main compari- 
sons: First, the popularity of any category relative to the other cate- 
gories ; second, the relative frequency vith vhich any category vas chosen 
by veterans, as compared vith nonveterans; and third, the amount of varia- 
bility , among the various college groups in the per cent of veterans or non- 
veterans selecting any category. 

The bottom portion of each figure gives a general indication of the 
relationship betveen various responses to the item and the Adjusted Average 
Grade. JOT each response, the median value of the tvelve mean AAG f s earned 
by veteran subgroups vas computed, and vas plotted as a solid circle. 
Similarly, the median value of the tvelve mean AAG's earned by nonveterans 
vho chose this response vas computed; this vas plotted as an open circle. 
At the right of each graph are also ehovn, for purposes of comparison, the 
median of the tvelve mean AAG's earned by all veterans and the correspond- 
ing median" value for all nonveterans. These "total group" medians are, of 
course, uniform tor all items. In general, the median values for veterans 
should be compared vith the over-all median for the veteran subgroups, vhile 
the median values for nonveterans should be compared vith the over -all non- 
veteran median in determining vhether a particular response is associated 
vith superior or inferior Adjusted Average Grades. 



96 



Chapter III 
TEE ACADEMIC ADJUSTMENT OF VETERAN AHD NONVETERAN STUDENTS 



A primary objective of this study was to determine whether or not 
veteran students earn higher grades in college than nonveteran students of 
equal ability. To this end, a series of twenty -five separate, but related, 
comparisons of veteran and nonveteran students was carried out- -one compari- 
son for each of the twenty-five groups listed in Chapter II In addition, 
a number of supplementary analyses were conducted to aid in the interpreta- 
tion of the basic findings The groups were studied separately so that each 
comparison would be based on veteran and nonveteran students who were as 
similar as possible with respect to such factors as college program, previous 
college training as a civilian, and educational environment while in college,, 
(Age was not controlled directly in any of the studies, nor was educational 
experience during service in the armed forces except where this experience 
led to substantial credit in specific academic courses . ) 

In order to take account of ability differences in comparisons of 
veterans and nonveterans, it was necessary to define ability in terms of 
specific measures, such as scores on a test of scholastic aptitude-, In 
every comparison the suitability of the measure of ability employed was 
evaluated through the use of an analysis of covariance method. This pro- 
cedure minimized the likelihood that a predictor chosen for equalizing 
ability might introduce some bias into the comparison through its closer 
relationship with grades in one group than in the other* The procedure pro- 
vided not only an estimate of the amount of difference between veterans and 
nonveterans, but also an estimate of the probability that a greater differ- 
ence than the one obtained might have arisen by chance. 

In selecting measures of ability for the purpose of the statistical 
analysis,, the guiding principle was to seek comparability among the various 
analyses without attempting to force all the studies into an identical 
design. By permitting some flexibility, it was possible to take advantage 
of the more extensive data at some colleges for the light they might throw 
on certain special problems. At the same time, the predictors chosen for 
the final comparisons were considered to be sufficiently similar to permit 
general conclusions to be drawn from the series of separate analyses,. 

The choice of predictors within a fairly definite framework was also 
thought to be desirable in order to avoid the proliferation of studies 
which would have resulted if varying combinations of predictors were used, 
and to reduce the danger of capitalizing on chance variations resulting 
from numerous comparisons involving the same group. The type of predictor 
to be used was therefore designated in advance . 

For the entering freshman groups, the usual team of predictors was 
some measure of high school standing used in combination with some test of 
scholastic aptitude, (in colleges where suitable data on high school stand- 
ing were not available, test scores provided the only predictive measure-,) 
For the groups which included interrupted veterans, grades earned during 
the freshman year (usually first-semester grades) were used as the predictor 
of later success . 



ACADEMIC ADJUSTMENT OF STUDENTS 97 



In order to carry out the statistical analyses, it was necessary to 
select not only measures of ability (the predictors), "but also a measure of 
college success (the criterion) The criterion chosen for the twenty groups 
of beginning students was ordinarily the freshman average grade. For the 
interrupted veterans and their nonveteran control groups, fourth-semester 
average grade was typically used as the criterion. In supplementary analy- 
ses, grades in specific courses were also employed * The average grades were 
generally obtained directly from the college records, although in a few cases 
they were coimputed from data appearing on transcripts with slight modifica- 
tions in the system us.ed by the college in ccaaputing averages 

In discussing the results of comparisons in the various groups studied, 
the following order will be followed s (a) students of arts and science in 
private universities) 1 (b) students of arts and science in state and municipal 
universities; (c) students of engineering in state and municipal universities; 
(d) students of agriculture and of business in state and municipal universi- 
ties; and (e) interrupted veterans and their uninterrupted nonveteran control 
groups o ithin each of these "main divisions, those groups upon which the more 
extensive analyses were done will generally be considered first. 

Grades of Teteran and Uonveteran Students in 
Arts and Science Colleges of Private Universities 

Adams University In addition to the basic comparison of achievement relative 
to ability, certain other pertinent problems were studied at Adams for veterans 
and nonveteranss the interrelations among various predictors (including Col- 
lege Board test scores, date of taking tests for admission, and secondary 
school standing) and the validities of these predictors in relation to term 
grades 

Adams University is a large private university for men All candidates 
for admission are required to present scores on College Entrance Examination 
Board tests, along with evidence regarding their secondary school achievement. 
For the 1<&6 group, about two -thirds of the entering students had attended 
private schools . Although specific courses are not required of freshmen, the 
selection of courses is limited, and all freshmen carry a uniform number of 
courses 

The group of students included in this study may be defined as follows s 
students who entered as beginning freshmen in the fall of 19^6 -and who com- 
pleted a full year fl s work during the academic year, 19^6-19^-7 -Che number of 
veterans in the group is 531 ; of nonveterans, 69^-0 A few items of informa- 
tion obtained from group members who completed questionnaires may be relevant 
here s The veteran group in this analysis is relatively young (about kO per 
cent were under twenty years of age at time of entrance), and the tour of 
active duty for the majority of them was comparatively short (only about ^5 
per cent had served two years of more) Only 3 per cent were married a About 
95 P^ cent of them reported that they would have attended college without 



98 ADJUSTMENT TO COLLEGE 



the aid given by the GI Bill (although presumably some of these students 
would have needed financial aid in the form of a scholarship or loan) . For 
veterans and nonveterans alike, 60 per cent of the group reported that their 
fathers vere college graduates. Eighty per cent of veterans and about 95 
per cent of nonveterans vere living in the college dormitories. 

The predictors used for Adams were the Verbal and Mathematical scores 
of the College Entrance Examination Board Scholastic Aptitude Test (SAT), 
the Adjusted School Bank, and a Predicted Grade, computed by the university, 
which is based on adjusted rank in school and College Board test scores. 
The adjustment of the rank in school is based on past records at Adams of 
students from each particular secondary school. The criterion measures are 
First -Semester and Second -Semester College Average Grade. Grades are "per- 
centage" grades based on a 100-point scale. The date of taking College Board 
tests "was included as still another variable. The inter correlations of these 
variables are shown in Table 3. 

"Date of Tests" vas included in order to investigate the relationship 
of amount of time elapsing between testing and college entrance to measures 
of achievement in college. Many of the veterans had taken the College Board 
tests prior to war service, while practically all nonveterans took the tests 
in the spring just prior to entrance at Adams University. The mean date of 
testing for veterans was found to be between 19^4 and 19^5, with a standard 
deviation of 1.12 years. It is apparent that there is sufficient variation 
among the veterans to make the results of the analysis meaningful. 

The correlations involving year of testing are all low; they range from 
-.08 to .21. The correlations .with term averages are .03 and .06". The high- 
est correlation ( .21) is with the Verbal score of the Scholastic Aptitude 
Test. This positive relationship might be accounted for in terms of higher 
standards for admission in the more recent years; however, such an hypothesis 
is not borne out by the correlations with the SAT -Mathematical score and the 
Adjusted School Bank, which are -.02 and -.08 respectively. The slightly 
higher correlation with the SAT-Verbal score may merely reflect growth in 
vocabulary and other verbal abilities with age. By and large, it appears 
that date of taking the aptitude tests is a matter of little importance within 
the limits of age and time found in this study. This finding is of consider- 
able significance in the interpretation of data in this investigation, since 
in most groups the veteran students were tested after war service, while non- 
veterans were tested soon after graduation from high school . 

The correlation of First -Semester Average Grade with Second -Semester 
Average Grade is .81 for each of the subgroups. This value may be considered 
as an indication of the reliability of semester grades, although it is pre- 
sumably an underestimate. The reliability of average grade for the freshman 
year may be estimated at .90 or higher, using the Spearman-Brown formula. 

The veteran subgroup is seen to have obtained lower mean scores than 
the nonveteran subgroup on all the variables on which a comparison is pos- 
sible. It is particularly interesting to note that the veterans were pre- 
dicted to earn a mean average grade of about 72 and they actually earned a 
First-Semester Average Grade of 75^, while the nonveterans were predicted 
to have a mean average grade of 7^.8 and actually earned a First -Semester 
Average of 75-7; a somewhat smaller difference. 



ACADEMIC ADJUSTMENT OF STUDENTS 



Table 3 

IMERCORREIATIONS OF SCHOLASTIC APTITUDE TEST SCORES, DATE TESTS WERE TAKER 

(FOR VETERANS OKLY), ADJUSTED SCHOOL BAM, PREDICTED GKADE, 

AMD KDRST-YEAB COLLEGE AVERAGE GRADES 

Adams University, College of Arts and Science, Freshmen, 1946-1947 





Male Nonveterans (1^694) 


4J 03 


Ist-Sesu Average 
2nd-Sem. Average 
1st -Year Average 


9 


Q 

to 




SAT -Verbal 


X. .17 -31 .65 


.37 .37 -39 


565 


86 




SAT-Mathemtical 


.23 N. .33 -44 


.32 .25 -30 


594 


80 




Date of Tests 


.21 -.oe NV 








H 

oo 

Lf\ 

i 


Adjusted School Bank 
Predicted Grade 


.29 .37 -.08 X. .82 
.62 .46 .16 .76 X. 


-57 .57 .60 
.61 .62 .65 


76,2 
74.8 


7.0 
6.0 


1st -Semester Average 


.41 .37 .03 -53 -6l 


\v .81 


75-7 


6.8 


1 


2nd-Semester Average 


.37 .26 .06 .50 .58 


.81 \v 


75-8 


6.6 


d 

i 


First -Tear Average 


-41 .33 -55 -63 


N^ 


75-8 


6.4 


Mean 


536 576 4.42 72.2 72.0 


75.4 74.8 75-1 






Standard Deviation 


93 84 1.12 7.7 6.9 


6.5 6.3 6*1 







100 ADJUSTMENT TO COLLEGI 



It will be noted in Table 3 that the best single predictor, for both 
Teterans and nonveterans, ia the Predicted Grade, and that the Adjusted 
School Rank alone ia nearly as good, especially for nonveterans , The Pre- 
dicted Grade as tried as a variable for use in controlling ability in the 
analysis of coyariance. It as found, however, when this Tariable was used, 
that the slopes of the regression lines were significantly different (that 
for nonveterans being steeper), so that the interpretation of the results 
of later steps would be doubtful. It was also found that the Adjusted School 
Bank was responsible for the different slopes,, The greater slope of the re* 
gression of grades on Adjusted School Bank for nonveterans means, of course, 
that for a given Increment in school rank there WHS on the average a greater 
increase in average grade for nonveterans than for veterans. One can only 
speculate as to the reason for this finding! but perhaps it is related to 
the nonintellectual factors which Influenced high school achievement and which 
have somehow been modified In veterans during the years- of military service 
Since the College Board test scores, used alone*, did not have significantly 
different slopes for veterans and nonveterans, they were chosen as the pre- 
dictors to be held constant* It must be added that this choice resulted In 
some loss of predictive effectiveness. 

Eesults pertaining to the analysis of covariance are shown in detail in 
Table 4. In the first section of the table (l) are shown the intercorrela- 
tionSj means, and standard deviations for the variables selected- The cri- 
terion is First -Tear College Average Grade* The correlations In this instance 
are seen to be very similar for veterans and nonveterans * The nonveterans 
obtained higher average scores for both the Verbal and Mathematical parts of 
the Scholastic Aptitude Test, and their freshman average grade was also 
slightly higher. Variability of test scores was slightly greater for veterans, 
as shown by larger standard deviations, but the nonveterans were slightly 
more variable with respect to average grades a Part II of Table k shows that 
the multiple correlations, based on the two SAT scores,, are JW and k6 for 
veterans and nonveterans respectively.. For both subgroups combined, the 
multiple correlation is Ajo In Interpreting these correlations, it is im- 
portant to remember that the tests were used in selection of the students. 

Part IJI of the table presents the results of the three significance 
tests in the analysis of covariance. Differences In errors of estimate and 
slopes are not significant. The Intercepts of the regression planes, also, 
are not significantly different| when ability (as measured by SAT scores) 
is made equivalent, the grades earned by veterans and nonveterans are too 
sljailar to warrant the conclusion that either group is superior. 

Part IV of the table highlights the findings? The veterans have a 
slight advantage over nonveterans of equivalent ability; the advantage, ex- 
pressed in Adams University grade units, is 36. (la terms of the mean 
grades shown in Part I of the table, the nonveterans have an advantage of 
?) The advantage of the veterans, expressed in standard error of estimate 
units, is only o06 Perhaps the most meaningful measure of the veteran stu- 
dents 1 advantage is the percentage of veterans found to exceed the average 
nonveteran in grades adjusted for ability differences; in the case of Adams, 
this turns out to be only 52. If there were no difference, this per cent 
would of course be 50. The difference, aa noted above, is too small to b 
statistically significant. 



ACADEMIC ADJUSTMENT OF STUDENTS 



101 



Table k 

COMPAEISON OF AVERAGE GRADES EARNED BY "VETERAN AND HOHVETERAN MALE STUDENTS 
Adams University, College of Arts and Science, Freshmen, 1946-19^7 

I. Correlations, Means, and Standard Deviations: 







Correlation with: 










Sub- 










Variable 


group 


SAT-Verbal 


SAT-Math. 


First -Year 
Avg. Grade 


Mean 


3D 


IT 


1, Scholastic Aptitude 


W 




.23 


.ia 


536* 


93 


531 


Test-Verbal 


m 




17 


39 


565 


66 


69^ 


2. Scholastic Aptitude 


m 


.23 




33 


576 


84 


531 


Test -Mathematical 


m 


-17 




.30 


59^ 


80 


694 


3, First -Tear College 


m 


-U 


.33 




75-1 


6.1 


531 


Average Grade 


m 


-39 


-30 




75,8 


6.4 


694 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Male Veterans 
Male Nonveterans 
Combined Group 



Multiple E 



A6 



III Analysis of Covariance Results s 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 


2 008 


1 


Between *10 and .20 


B. Equality of slopes 


0-733 


2 


Between ,50 and 70 


C. Equality of intercepts 


1.213 


1 


Between .20 and ,30 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 
0,36 
06 

52 
Hot significant 



ADJUSTMENT TO COLLEGE 



Stewart University. For the Stewart University students entering as fresh- 
men in 1946, the analysis design was much like that for Mams. At Stewart, 
Intercbrrelatlons among College Board Scholastic Aptitude Test scores, Ad- 
Justed School Rank,, date of taking admissions tests, and First-Year College 
Average Grades were obtained. In comparing the relative achievement of 
veteran and nonveteran male students, allowance was made for three predictors: 
Scholastic Aptitude Test-Verbal, Scholastic Aptitude Test -Mathematical, and 
Adjusted School Rank, 

Stewart is a private institution for men, similar in many respects to 
Adams .University, Students are required to present College Board scores, 
along with their secondary school records, for admission. As at Adams, the 
majority (about three -fourths) of the 19^6 entering students had attended 
private schools. As freshmen, Stewart students select their program within 
a restricted framework. 

The group of entering freshmen may be defined as follows: students who 
entered as beginning freshmen in the fall of 19^6 and who completed a full 
year's work during the academic year, 1946-19^7. There were 187 veteran and 
3^-8 nonveteran students. Since all students carry a fixed number of courses, 
it was not necessary to consider variations in academic load in defining the 
sample. Further information about this group, obtained from their question- 
naires, may be summarized as follows: The veterans were young (about 55 per 
cent were not yet twenty when they entered college); the great majority had 
had relatively little active duty (slightly under 30 per cent had served two 
years or more); and none was jsarried. Virtually all Indicated that they 
probably would have come to college without G-I Bill assistance (although pre- 
sumably some would have needed financial aid in some other form) , Sixty per 
cent of veterans and about 65 per cent of nonveterans reported that their 
fathers were college graduates. As at Adams, substa.ntia.lly all students 
(about 95 per cent of both veterans and nonveterans) were living in college 
dormitories. 

The statistical results involved in the comparison of achievement of 
veteran and nonveteran students are presented in Table 5. The predictors 
again included .the College Board SAT scores, both Verbal and Mathematical. 
Their predictive value was very nearly the same as at Adams. Adjusted School 
Bank was also used at Stewart, and its predictive value was high (r = .53 
for veterans and .62 for nonvetera.ns) . At Stewart, the adjustment of the 
secondary school ra.nk is made on the basis of the grades at Stewart of former 
students from each particular secondary school. 

The multiple correlations (based on all three prediqtors) were higher 
than for Adams because of the inclusion of the measure of high school achieve 
ment; the JR was .60 for veterans, .66 for nonveterans, and .65 for both sub- 
groups combined. As at Adams, the nonveteran students on the average had 
higher scores on all the predictors and also a higher First-Year Average 
G-ra.de. The actual difference in mean grades is .31 on the grading system 
^ised at Stewart. Stewart uses a seven-step grading system; in this analysis 
the highest grade was given the value 1, the next 6, and so on, 1 being the 
Lowest . 



ACADEMIC ADJUSTMENT OF STUDENTS 



103 



Table 5 

COMPARISON OP AT7ERAGE GRADES EABHED BY VETERAN AHD HOHVETERAN MAXE STODEOTS 
Stewart University, College of Arts and Science, Freshmen, 1946-1947 

I. Correlations, Means, and Standard Deviations-; 



Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


H 


SAT 
Verbal 


SAT 
Math. 


Adj. 
School 
Bank 


Ist-Yr. 
Average 
Gradfe 


1. Scholastic* Aptitude 
Test -Verbal 


MV 

m 




.21 
.09 


.29 
-31 


.ko 

.40 


536 
570 


86 
82 


187 
348 


2. Scholastic Aptitude 
Test-Mathematical 


m 
m 


.21 
.09 




32 
37 


.28 
.25 


570 
590 


77 
7^ 


187 
348 


3. Adjusted School 
Rank 


MV 

m 


.29 
31 


32 
-3? 




-53 
.62 


4.12 
4.70 


.16 
.78 


187 
348 


4. First -Tear College 
Average Grade 


m 
m 


.40 

.4o 


.28 
.25 


:i 




4.76 
5.07 


77 
.83 


187 
348 



II. Multiple Correlations (Variables 1, 2, and 3 vs. Variable 4): 



Sample 



Male Veterans 
Male Honveterans 
Combined Group 



Multiple R 



.60 
.66 
.65 



III* Analysis of Covar lance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Sxeedom 


Probability that a value 
greater than obtained 
would arise by chance 


A, Equality of errors of estimate 


0.014 


1 


Between .90 and .95 


B. Equality of slopes 


3-375 


3 


Between -30 and .50 


C. Equality of Intercepts 


2.143 


1 


Between .10 and .20 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 
0.09 
0.14 

56 
Not significant 



104 ADJUSTMENT TO COLLEGE 



The analysis of covariance results shov that the hypotheses of equal 
errors of estimate and equal slopes of regression planes are not disproved. 
The difference "between veteran and no-nveteran students in intercepts of 
their regression planes is veil within the range of chance expectancy. The 
advantage is, then, in favor of the veteran student at Stewart; lout the ad- 
vantage is only .09 grade units or .Ik standard error of estimate units, 
when ability measures are taken into account. Fifty-six per cent of the 
veterans exceed the average nonveteran. The advantage of the veteran is 
again found not to be significant, when account is taken of differences in 
ability. 

In addition to the variables directly involved in the analysis of co- 
variance, attention was given to the year during which College Board tests 
were taken. The results for this variable were similar to those found at 
Adams. While practically all the nonveteran students were tested in 19*6, 
the mean date of testing for veterans was midway between 19^ and 19^5, 
with a standard deviation of one year. The correlations of year of testing 
with other variables ranged from .10 (with Adjusted School Bank) to -.04 
(with SAT-M and First-Year Average Grade) . The correlation with SAT-V was 
.07 at Stewart. The time of testing was again found to be a factor of 
negligible importance within the conditions of this study. 

Douglas University. Several points not investigated in the first two analy- 
ses were included in the study at Douglas. In particular, data for a group 
of female nonveterans were analyzed. Douglas is the first of many groups 
for which American Council Psychological Examination scores were available; 
in this analysis, both part and total scores were studied. Year of High 
School Graduation and number of hours of credit were other variables of 
special interest included in the Douglas study. 

Douglas University is a coeducational private university located in a 
southern city, The typical method of admission at Douglas depends primarily 
upon the student's secondary school record. In the 19^6 group, about h^ per 
cent had attended private schools . Certain required courses are set up for 
freshmen in arts and science, required subjects being English, mathematics, 
and social science. A foreign language and a natural science course are also 
required during the freshman or sophomore year. 

The group studied at Douglas met the following requirements: all 
entered as beginning freshmen in the fall of 19U6, completed three full 
quarters during the academic year 19^6-19^7, and returned a questionnaire. 
(Very few students in this group failed to complete a questionnaire.) The 
resulting group included 77 male veterans, 119 male nonveterans, and 93 
female nonveterans. From the questionnaires, the following characteristics 
were noted: The veterans were relatively young (about half were under age 
twenty at time of entrance); only about kQ per cent had had two or more 
y^ars of active duty; and slightly under 10 per cent were married. Some- 
what more than 90 per cent reported that they would probably have attended 
college without the aid of the GI Bill. When veterans were compared with 
the two nonveteran subgroups with respect to father's education, it turned 
out that slightly more than 20 per cent of th.9 veterans' fathers had been 
graduated from college ; as compared with slightly more than UO per cent for 



ACADEMIC ADJUSTMENT OF STUDENTS 105 



the male nonveterans The percentage for -female nonveterans was slightly 
greater than for male nonveterans. The great majority of Douglas students 
(about 80 per cent of the veterans and 75 per cent of the nonveterans) live 
either at home or in one of the college dormitories, A much larger propor- 
tion of male veterans (60 per cent) than of male nonveterans (slightly over 
30 per cent) reported that they lived at home or with near relatives. 

The predictors used at Douglas included raw scores on the ACE Psycho- 
logical Examination (Quantitative, Linguistic, and total score), an English 
Placement Test and a Mathematics Test administered to freshmen at Douglas, 
and High School Average Grade. The high achool grade, which was expressed 
in letters, was converted into numerical form, as follows: A * 8, A- 7, 
B* = 6, B = 5, B- 4, etc. The Year of High School Graduation was also in- 
cluded, for veteran students only. Criterion measures included the First - 
Year College Average Grade, score on an English Achievement Test given near 
the end of the freshman year, and grades in the two freshman English courses. 
The grades were based on a four-category scale, the units of which had 
numerical values of 3 to 0* The number of First-Year College Credit hours 
for which students were registered was included aa another variable. 

The intercorrelations of the variables are shown in Tables 6 and 7. 
Comparison of the mean scores reveals that the women students were superior 
to both the other subgroups in measures of English aptitude and achievement, 
which is consistent with the sex difference usually found* Osn mathematical 
tests, they were definitely poorer than the male nonveterans and similar to 
the male veterans. The male veteran subgroup tended to be poorest on both 
predictor and criterion measures. 

For the male subgroups, the best predictor of the First-Year Average 
Grade was High School Average Grade, while for the women students, the 
measures of verbal ability (ACPE total and L-scores and the English Place- 
ment Test) were the beet predictors. 

Year of High School Graduation was included as a variable for male 
veterans only, since practically all the nonveterans graduated from high 
school in 19W>. The average veteran graduated from high school in 19Mf 
(the last digit of the year of graduation was used as the variable), and 
there was considerable variability as indicated by the standard deviation 
of 1.7. Year of High School Graduation has essentially aero correlations 
with all the predictors except High School Average, where the correlation 
is -.18. This correlation indicates a very slight tendency for students who 
graduated most recently to have lower High School Averages. Such, a rela- 
tionship might easily have been the result of a tendency for admissions 
officers, to admit the older veterans only if they had exceptionally good 
high school records, or 6f a self -selection process in the older veterans . 
Correlations with the measures of college achievement were also negative; 
the most recent high school graduates tended to make poorer grades. The 
highest correlation (~*28) is with the First-Year Average Grade, but the 
relationship also holds for the three measures of achievement in English* 



106 



ADJUSTMENT TO COLLEGE 










CO 



* o 








ACADEMIC ADJUSTMENT OF STUDENTS 



107 



pa 



EH 



C 

* 

i-i 
vo 



H 

g 

J? 

-H 

to 

^ 

I 

iS 



a -s 



VTOT ' 



i-aaov 



tr\o 




oj 



O\COC3O CUOO O 

ro en 



C3O 

vif 



OJ cnt*-vooo no 
in tr\ir\ ir\ co 



CO rf r-lj* ONVD 

no tr\ ir\i 



oo ir\ 
\r\^t- 



VD t CVJ 

tr\ir\vo 



vo 



82 

CO O 
O 



-H o 

" 



o 



CQ i-l 

38 



s^ 

Of ^ EH 03 
1 1 1 H 



OJ 



no 




108 ADJUSTMENT TO COLLEGE 



Number of houra credit correlates positively with, all other measures for 
all three subgroups, indicating a tendency for the better students to take 
the heavier loads,, Hie correlation is especially high ith the Mathematics 
Test in all three groups The heaviest course load is taken by the male non- 
veterans and the lowest "by the male veterans j the biggest mean xlif f erenoe 
amounts to two credit hours. 

Three analyses of covar lance were computed. The results of the first, 
comparing veteran and nonveteran male students, are shown in Table 8. The 
correlations, means, and standard deviations for the criterion and the selected 
predictors are taken from the more complete table of intercorrelations . The 
American Council Psychological Examination total score and Sigh School Average 
Grade were chosen as predictors. The multiple correlations were found to be 
.59 sad >7^ for veterans and nonveterans. respectively, and .71 $r both sub- 
groups combined. The higher validities for nonyeterans is undoubtedly due in 
part to their greater variability. 

Differences in errors of estimate and slopes of the regression planes 
were found to be no greater than would be expected by chance, and the differ- 
ence in intercepts of the regression planes was also not significant. When 
account is taken of differences between the subgroups in. ability, the mean 
difference in, freshman average grade is only .02 (in fayor of the veterans) 
as compared with, a difference of ,26 (in favor of the nonveterans) when no 
adjustment is madeo In standard error of estimate units, the difference is 
04j 52 per cent of the veterans excel the average nonveteran* 

The analysis of covariance results for the male-female nonveteran compari- 
son, shown in Table 9? also shows negative results. The hypotheses of equality 
of errors of estimate, slopes, and intercepts are not disproved. JPemale non<- 
veterans are found to excel the male nonveterans by .06 grade units, when 
ability is considered, a difference which is not significant* Only 46 per 
cent of the male nonveterans excelled the average female nonveteran* 

The third analysis, the results of which are snown in Table 10, involves 
the comparison of male veterans and female nonveteran students* Again the 
three hypotheses are not disproved,, The female students were found to earn 
better grades, in relation to ability, than thie male veterans, the difference 
being ,09 grade units* Forty-three per cent of the male veterans excelled 
the average female student, a difference which, is not significant, 

Harris Ifodversitva The analysis of data for students in Harris University 
was limited to. that necessary for a comparison of relative achievement of 
veteran and nonveteran students after allowing for differences In high 
school standing and a composite score based on five entrance tests <> 

Harris is a coeducational private midwestern college of arts and science 
which is bound by ties of tradition to one of the major religious denomina- 
tions. Harris students are drawn from the upper half of their high jechool 
graduating classes Sreshmen choose their own program, within a series of 
general requirements, only English being specifically required. 



ACADEMIC ADJUSTMENT OF STUDENTS 



109 



Table 8 

COMPARISON OF AVERAGE GRADES EARNED BY "VETERAN" AHD NONVE'i'KHAH MALE STUDENTS 
Douglas University, College of Arta and Science, Freshmen, 1946-194? 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


3D 


H 


ACPE 


H. 8. 
Average 
Grade 


Ist-Yr. 
Average 
Grade 


1. ACPE Total 
(raw score) 


MV 

m 




.13 
.32 


35 
55 


106 
117 


19 
20 


77 
119 


2. High School 
Average Grade 


MV" 
MN 


.13 
.32 




'? 

.65 


3.26 
4.0? 


2.04 
2.30 


77 
119 


3. First-Year College 
Average Grade 


MV 

m 


35 
.55 


-52 
.65 




1.14 
i.4o 


.64 
.75 


77 
119 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Male Veterans 
Male Nbnveterans 
Combined Group 



Multiple R 



.59 
.74 
71 



III. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


0.0506 
2.048 
0.0587 


1 
2 
1 


Between .80 and .90 
Between *30 and .50 
Between .80 and .90 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 
0.02 
0.04 

52 
Not significant 



110 



ADJUSTMENT TO COLLiGI 



Table 9 
COMPARISON OF AWRASE GRADES EAEHED BI MAUE NOUVETERANS AHD FEMALE EOH7ETERAHS 



of Arts and Science _ } 



^ .1946 " 



I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation vith: 


Mean 


3D 


K 


ACEE 


H. S. 
Average 
Grade 


ist-lr. 
Average 
Grade 


1. ACFE Total 
(raw score) 


m 

JN 




-32 

,20 


-55 
.51 


117 
111 


20 
20 


119 
93 


2. High School 
Average Grade 


m 

M 


.32 
.20 




.65 

M 


k. 07 
5.3^ 


2.30 
1.97 


119 
93 


3. First-Tear College 
Average Grade 


m 
m 


55 
,51 


65 

M 




lAo 
1.57 


.75 
,6l 


119 
93 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3); 



Sample 



Male Nonveterans 
Female Nonveterans 
Combined Group 



Multiple B 



.63 
-70 



III. Analysis of Covar lance Ee suits: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercept a 


0.275 
3*246 


1 
2 
1 


Between .50 and .70 
Between ,10 and .20 
Between ,30 and *50 



IV. Difference between Subgroups with Ability Held Constant : 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of male nonveterans excelling the average female nonveteran 

Level of significance of difference (from IIIC above) 



Female Nonveteran 

0.06 

0.11 

46 

Not significant 



ACADEMIC ADJUSTMENT OF STUDENTS 



111 



Table 10 

COMPABISQJiT OP AVERAGE GRADES EARUED BT MAIE VETERANS AKD FMAE HOHVETERAHS 
Douglas University, College of Arts and Science, Freshman^ 

I. Correlations, Means, and Standard Deflations: 







Correlation with: 








Variable 


Sub- 




Ha 
D. 


Ist-Yr. 


Mean 


SD 


1ST 




group 


ACEE 


Average 


Average 














Grade 


Grade 








1. ACEE3 Total 


m 




.13 


-35 


106 


19 


77 


(raw score) 


FN 




.20 




111 


20 


93 


2. High School 


m 


.13 




.52 


3.26 


2.04 


77 


Average Grade 


m 


.20 






5-34 


1.97 


93 


3. First-Year College 


m 


,35 


-52 




i.i4 


.64 


77 


Average Grade 


m 


51 






1-57 


.61 


93 



II. Multiple Correlations (Variables 1 and 2 TS. Variable 3): 



Sample 



Male Veterans 
Female Honveterans 
Combined Group 



Multiple R 



59 
.63 
.66 



III. Analysis of CoYariance Ee suits; 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 


0.479 


1 


Between -30 and .50 


B. Equality of slopes 


1.3^7 


2 


Between ,50 and .70 


C. Equality of intercepts 


0.990 


1 


Between .30 and .50 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of male veterans excelling the average feaale nonveteran 

Level of significance of difference (from IIIC above) 



Female Honveteran 

0.09 

0.17 

43 

Not significant 



112 ADJUSTMENT TO COLLEGE 



For study at Harris University, the following group of students was 
selected: students who entered as beginning freshmen in the fall of 19^6, 
who completed substantially a full year's work during the academic year 
19lj.6-19lf7, and who returned a questionnaire. Students who received ten or 
more hours of credit in specific subjects for training received in the armed 
services were judged not to be "beginning students/' and students who carried 
ten or fewer credit hours in any term were judged not to have completed a 
full year's work; both of these groups were excluded. It may be added that 
virtually all students completed questionnaires , so that this was not an im- 
portant cause of rejection of students from the analysis. The group selected 
for study included 105 veteran and 1**6 nonveteran students. 

From the questionnaires, the following points may be added to the 
description, of the group: Only about 30 per cent of the group of veterans 
were under twenty years of age upon entrance to college j slightly under 60 pel- 
cent served two or more years of active duty) and just under 10 per cent of 
the veterans were married . Almost 90 per cent would probably have attended 
college without the GI Bill. With respect to father's education, the picture 
is somewhat similar to that at Douglas: slightly over 20 per cent of the 
veterans and about 35 per cent of the male nonveterans reported that their 
fathers had been graduated from college . A little more than half of the male 
veteran and male nonveteran students were living in fraternity houses,- Harris 
is the only college included in the study where the fraternity house was the 
predominant living arrangement for the groups studied. 

The predictors used at Harris included High School Rank and a Composite 
Test score. (High School Eank was converted to a standard score scale having 
a mean of 13 and a standard deviation of 1*.) The composite score, which is 
computed routinely at Harris, is based on the American Council Psychological 
Examination (l9i*2 edition) and four Form T Cooperative tests- the Coopera- 
tive English Test (consisting of Mechanics of Expression, Effectiveness of 
Expression and Reading Comprehension) and the three Cooperative General 
Achievement Tests in social studies, natural sciences and mathematics. The 
composite score was obtained by adding to the ACPE total score twice the sum 
of the scaled scores on. the four achievement tests, after both the ACPE total 
and the sum of the scaled scores had been converted to standard scores based 
on local norms. The criterion as usual is the First -Year College Average 
Grade. A six-step grading system is used at Harris; in, this analysis, the 
numerical equivalents of the steps were 3, 2, 1, 0, -1, and -2. 

The results of the analysis are shown in Table 11. The nonveterans at 
Harris proved to be superior on. the average to veterans on both predictor 
measures and also with respect to freshman average grade. Validity coeffi- 
cients are satisfactorily high. The greater validity of High School Rank for 
nonveterans may be due in part to the greater variability of nonveterans on 
this measure. The multiple correlations, based on both the Composite Test 
score and High School Rank, were .61 and .68 for veterans and nonveterans 
respectively, and for both subgroups combined it was .65. 

The analysis of covariance results show that the hypotheses of equality 
of errors of estimate angi equality of slopes of regression planes are not 
disproved, and that the difference between the intercepts of the generalized 



ACADEMIC ADJUSTMENT OF STUDENTS 



113 



Table 11 



COMPARISON OF AVERAGE GRADES EARNED BY VETERAN AND NONVETERAN MALE STUDENTS 
Harris University, College of Arts and Science, Freshaien, 1946-1947 

I. Correlations, Means, and Standard Deviations: 







Correlation with: 










Sub- 












Variable 


group 


Composite 
Test 


H. S, Rank 


First -Year 
Avg . Grade 


Mean 


SD 


N 


1. Composite Test 


MV 




.46 


-55 


164 


28 


10*5 




MN 




.54 


52 


175 


32 


146 


2. High School Rank 


MV 


.46 




49 


15-0 


3*0 


105 


(converted score) 


MN 


.54 




-65 


17.2 


3.4 


146 


3. First-Year College 


MV 


-55 


.49 




1.20 


72 


IP 1 ? 


Average Grade 


MN 


.52 


.65 




1.43 


.71 


146 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Male Veterans 
Male Nonveterans 
Combined Group 



Multiple E 



.61 
.68 
.65 



III. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A, Equality of errors of estimate 


0.901 


1 


Between .30 and .50 


B. Equality of slopes 


4.024 


2 


Between .10 and .20 


C. Equality of intercepts 


0.676 


1 


Between .30 and .50 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 
0.06 
0.11 

^ 

Not significant 



114 ADJUSTMENT TO COLLEGE 



regression planea Is not signlf leant The veteran subgroup Is slightly 
superior* Although tha nonveterana achieve a raw mean average grade which 
la .23 higher than for veterans, the difference becomes 06 In favor of the 
veterans when the Influence of the ability measures Is taken Into account. 
In error of estimate units, the advantage of veterans Is only .U. Fifty- 
four per cent of the veterans exceed the average nonveteran with respect to 
fresianan average grade at Darrls* The advantage of veterans is clearly not 
significant* 

Miller. IMyerfllty* As at I/arris, the analysis, of the acadefialc data at Killer 
vas liaitecl to that required for comparing the relative achievement of male 
veteran and nonveteran students, The predictors employed were the American 
Council Psychological Examination total score and High .School Rank* The cri- 
terion was First-Year College Average Grade* 

Miller is a large coeducational private university located in an. eastern 
city. Students. In the College of Arts and Science are typically drawn from 
the upper three-fifths, of graduating classes of accredited high schools. The 
only course required of all freshmen is English; as usualj however, the formu- 
lation, of the student ' s program is determined in part by broader requirements , 

The group selected for study may be def ined as follows: students who 
entered as beginning freshmen in the fall of IJ&fi, vho completed a full year 
of work (eleven or more hours each semester) during the academic year 19^ 
and (since questionnaire returns were satisfactoyy) vho returned a questionnaire. 
Because one of the variables included in the analysis was score on the 19^6 
edition of the American Council Psychological Examination, a few students who 
had taken the 19^5 edition were excluded. Further information about the group 
was obtained from the questionnaires, as follows: Veterans at Miller tended 
to be somewhat older than those in other arts and science groups entering in 
19^6 (only about 15 per cent were under twenty years of age at time of entrance); 
they also had served a longer period of active duty (about 75 per cent had served 
two years or mpre); and roughly 15 per cent were married* Sixty-five per cent 
of the group Judged that they would probably have attended college without the 
GI Bill. With respect to father' 1 s education, slightly more than 10 per cent 
of male veterans reported that their fathers were college graduates ; for the 
male nonveterans the percentage was 20. The majority of students in this 
group lived at home or with near relatives; 70 per cent of the male veterans 
and slightly over 80 per cent of the male nonveterana were doing 00 * It may 
also be noted that about 10 per cent of the veterans in this group rented or 
owned their own house or apartment; among the groups previously described, 
only Douglas with slightly under 10 per cent has more than a HTnall propor- 
tion In this category. 

Results of the analysis of data for Miller students are shown in Table 12 
In interpreting means and standard deviations, the following information may 
be useful: The grading system at Miller is based on a five-step sjcale; the 
numerical equivalents of the categories range from, 3 through to -Is For use 
In statistical analysis^ Ugh School Rank was converted to a standard acale 
having a jnean of 13 and a atandard deviation of \. 



ACADEMIC ADJUSTMENT OF STUDENTS 



115 



12 

COMPARISON OP AVERAGE GRADES EARNED BT VETERAN AND N03TOTERAN MALE STUDENTS 
Milled University, College of Arts and Science, 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


3D 


N 


ACEE 


H.S. 
Bank 


lst*Yr. 
Average 
Grade 


1. ACEE (1946) Total 
(raw score) 


m 

m 




.26 

.32 


.41 
39 


111 
112 


20 
20 


425 
193 


2. High School Rank 
(converted score) 


m 
m 


.26 

32 




.43 
-58 


14,8 
15.8 


3.3 
3.4 


425 
193 


3- First-Tear College 
Average Grade 


m 

MN 


.41 
39 


.43 
*58 




1.49 
1.50 


.56 

.64 


425 

193 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 


Multiple E 


Male Veterans 


53 


Male Nonveterans 


.62 


Combined Group 


-56 



III, Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes" 
C. Equality of intercepts 


1.136 
7-027 
2.732 


1 
2 
1 


Between ,20 and .30 
Between 02 and .05 

Between .05 and .10 , 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 

0.07 

0.14 
56 
Not significant 



116 ADJUSTMENT TO COLLECT 



Veteran and, nonveteran students at Killer Tftaiversity have almost ex- 
actly the same mean First-Tear Average Grade. Their average raw scores on 
the ACPE are almost identical, bit with respect to High School Bank the non- 
yeterans are slightly superior* 

Veteran students proved to be somewhat less predictable than nonyeterans 
in terms of the multiple correlations, "where are -53 and ,62 respectively,, 
The difference is due to the greater predictive value of high school standing 
for nonveterans, which ia characteristic in some degree of almost all the 
groups studied where a measure of high school standing was available. Errors 
of estimate were not significantly different. Toe slopes, however, were 
significantly different at the 5$ level. It was judged that a difference in 
slopes significant at this level did not preclude continuing the analysis of 
covariance, particularly in view of the relatively large size of the group at 
Miller The difference in intercepts between the groups turned out not to be 
significant. 

Male veterans tend to excel nonyeteran students of equivalent ability, 
the difference in grade unite, however, being only .07* In standard error of 
estimate units, the difference is .14. Fifty-six per cant of the veteran sub- 
group exceed the average nonyeteran; the difference is not significant. The 
difference in slopes of the regression planes y however, casts some doubt on 
the accuracy of the evaluation of the difference in intercepts. 

Efrans Uhiyeraitv^ For students entering Evans University as freshmen, the 
comparison of achievement between veterans and nonveterana was carried out 
using only a single predictor total score of the American Council Psychologi- 
cal Examination as a basis for equating aptitude The criterion, as usual, 
was First -Year College Average Grade. 

Evans is a coeducational, private, church-connected college of arts and 
science located in a midwestern city. Admission is based typically upon the 
high school record of the student. At Evans, English is a required subject; 
the remaining requirements in the freshman program allow some choice of sub- 



The group chosen for analysis at Evans was, limited to students entering 
as beginning freshmen in the fall of 19^6 who completed two semesters of 
eleven or more hours each, during the academic year 19^6-19^7, and who completed 
a questionnaire* Veteran students who received ntoe or more credit hours in 
specific subjects for college training during military service were considered 
not to be "beginning students" and were excluded* The veteran subgroup com- 
prised 283 students, and the numtxer of ndnveterans was 9^. Questionnaire 
responses provided the following additional information about the Evans group 
chosen for studys Lite the Miller group, the Evans veteran group was rela- 
tively old (only about 20 per cent were under twenty years of age); almost 
70 per cent of the veterans had completed two years or more of active dutyj 
and close to 20 per cent were married As at Miller, slightly under 70 per 
cent indicated that they would probably have attended college without the 
aid provided by the GI Billo Ten per cent of the jmale veterans and just 
under 20 per cent of the nonyeterans indicated that their fathers were col- 



ACADEMIC ADJUSTMENT OF STUDENTS 117 



lege graduates . Finally, these Evans students generally lived at borne or with 
near relatives , slightly under 60 per cent of veterans and about 70 per cent 
of male nonveterans used this arrangement. About 25 per cent of veterans 
lived in a rooming or boarding house, and approximately 10 per cent owned or 
rented their own house or apartment. 

The results of the analysis are shown in Table 13. At Evans University , 
only one predictor was used, the total score on the American Council Psycho- 
logical Examination. A measure of high school standing was not used because 
data were not available for a sufficiently large proportion of the group The 
ACFE proved to have a fairly good predictive value at Evans. The nonveterans 
were slightly superior in ACPE score, while the veterans earned First-Year Average 
Grades which were slightly higher than those of nonveterans . The superiority 
of the veteran group was found to amount to .15 in grade units (.28 in stand- 
ard error of estimate units), when account is taken of the difference in ability. 
The percentage of veterans exceeding the average nonveteran was found to be 6l; 
this advantage of the veterans is significant at the 2$ level. 

Turner Univers ity . At least two special features characterize the analysis of 
data for freshmen at Turner University: first, the average grades are based 
on achievement test scores used by the college rather than on the course 
grades; and second, students entering in the fall of 19^5 have been combined 
with those entering in the fall of 19^6 in making up the group. The inclusion 
of both years was necessary in order to obtain a group of sufficient size. One 
hundred veteran and 101 nonveteran students were included. 

Turner University is a private coeducational university located in a large 
midwestern city. In addition to the student's high school record , a special 
battery of tests is used in the admissions procedure. The customary depart- 
mental ^divisions are minimized in the freshman program of Turner students. 

Students who entered as beginning freshmei) in the fall of 19^5 or in the 
fall of 19^6, and who had data on three comprehensive examinations taken dur- 
ing the first year --in eluding one in. physical or biological science and one in 
another typical freshman course- -constituted the group' for analysis. Further 
information, derived from questionnaire responses includes the following: The 
veteran group had longer service (almost 90 per cent served two years or more) 
than, any of the groups previously discussed. Twenty per cent of the veteran 
group were married. Somewhat more than 60 per cent would have attended col- 
lege without the aid provided by the GI Bill of Eights, Almost 20 per cent 
of the veterans and slightly over 30 per cent of the male nonveterans reported 
that their fathers had completed college. Just under 80 per cent of male non- 
veterans lived at home or in. college dormitories, as compared Vith approxi- 
mately 50 per cent for the male veterans . Bpughly 20 per cent of male veterans 
reported owning or renting their own. house or apartoopent. (Age of veterans is 
not reported for Turner, since the method of determining age at entrance is 
not adapted to groups whose members did not enter at a fixed time . ) 

High school rank was not available for a sufficiently high proportion of 
the students to justify its use as a predictor. The prediqtor which was 
chosen proved to be a rather good one, however. It is a Composite Test score 



118 



ADJUSTMENT TO COLLEGE 



Table 13 

COMPARISON OP AVERAGE GRADES EARliED BT VE1ERAI AM) UONVSTERAI MALE STUDEHTS 
Evans University, College of Arts and Science, Freshmen,, 

I. Correlations, Means, and Standard Deviations: 



Variable 


group 


Correlation with: 


Mean 


3D 


ft 


ACPE 


First-Tear 
Avg- Grade 


ACPE Total 
(converted score) 


m 
m 




.51 
-45 


12.0 
13,1 


3,3 

3-1 


283 
94 


First-Tear College 
Average Grade 


m 
m 


51 




2.10 
2.05 


.61 

.60 


283 



Validity coefficient for combined group: 



II. Analysis of Covariance Besults: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedoni 


Probability that a value 
greater than obtained 
would arise, by chance 


A. Equality of errors of estimate 
B Equality of slopes 
C Equality of intercepts 


0,0285 
0.148 
5-504 


1 
1 
1 


Betveen ,80 and .90 
Between ,70 and .80 
Between .01 and ,02 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade unita 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonyetereua 

Level of significance of difference (from IIC above) 



Male Veteran 
0.15 



9$ 



ACADEMIC ADJUSTMENT OF STUDENTS 119 



"based on the American Council Psychological Examination total score, "a test 
of reading comprehension, and a test of writing skills, the last two tests 
having been prepared at Turner. As shown in Table Ik, the validity of this 
Composite Test score was .61 for the group as a vhole, and for veterans and 
nonveterans respectively the validity coefficients were 6l and .60. The 
nonveteran subgroup was superior on both the Composite Test score and the 
criterion, First-Year College Average Grade. Standard deviations of "both 
measures were larger for nonveterans . 

It was found through the analysis of covariance that the difference in 
standard errors of estimate was significant at the jf> level: the grades of 
veterans were somewhat more predictable than those of nonveterans This find- 
ing casts some doubt on the legitimacy of testing the other two hypotheses 
concerning the regression lines. The remaining two tests were carried out; 
their results must be interpreted with caution. It was found that the slopes 
were not significantly different and that the difference in intercepts was 
not significant. 

Turner is the first institution so far encountered where the nonveteran 
students are superior to veteran students . Their advantage in average grade 
is .26 in. grade point units when differences in ability are not considered; 
taking into account the difference in Composite Test score, the grade differ- 
ence is reduced to.l6. Only 39 per cent of the veterans at Turner exceed 
the average nonveteran. It should be kept in mind that the results based on 
adjusted grades must be discounted somewhat in view of the rejection of Hypothe- 
sis A; it is believed, however, that the tendency for nonveterans to excel 
the veterans is a valid finding for the present data. 

Summary . Summarizing the results fox the private institutions, it is found 
that the superior subgroup in six instances is the male veteran and in only 
one instance the male nonveteran subgroup. The difference favoring veterans 
is significant at the 2$ level for one group; none of the other differences 
in intercepts is significant. 

The validity coefficients obtained for tests of ability are in general 
very similar for veterans and nonveterans; but for measures based on high 
school standing, the predictive value is consistently greater for the non- 
veteran subgroups . This finding appears to be reasonable in the light of 
the greater time elapsing between high school graduation and college entrance 
for veterans than for nonveterans, permitting greater opportunity for change 
in motivation, interests, efficiency of work habits, etc. Evidence from 
Adams and Stewart, where there was considerable variation in the time of 
taking tests, indicates that the time of testing has little effect on the 
predictive value of the tests and little relationship to the predictive 
measures . 



120 



ADJUSTMENT TO COLLEGE 



Table Ik 
COMPARISON OP AVERAGE GRADES EARNED BY VETERAN AHD NONVETERAN MALE STUDENTS 

Turner. University^ College of Arts and Science 
(Students who entered as freshmen in fall", 194 5 / or fall, 1946) 

I. Correlations , Means, and Standard Deviations: 



"Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


N 


Composite 
Test 


First-Year 
Avg. Grade 


Composite Test 
(converted score) 


MV 
MN 




.61 
.60 


66,3 
68.1 


7-3 
8.0 


100 
101 


First-Year College 
Average Grade 


MV 
MN 


.61 

,60 




2.4? 
2,73 


.66 

.81 


100 

101 



Validity coefficient for combined group: 



.61 



II. Analysis of Covariance Results; 



Hypo the sis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate. 
B. Equality of slopes 
C. Equality of intercepts 


4.540 
0.241 
3.753 


1 

1 
1 


Between .02 and .05 
Between .50 and ,70 
Between .05 and ,,10 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Nonveteran 
0.16 
0.28 

39 
Not significant 



ACADEMIC ADJUSTMENT OF STUDENTS 121 



Grades of Teteran and Nonveteran Students In Arts and Science 
Colleges of State and Municipal Universities 

Western State University a At Western State a fairly complete analysis of the 
available data was carried out. As at Douglas, a female subgroup was included- 
(At Western State , however , 16 female veterans were included in the 482 members 
of this subgroup o) Predictors studied included part and total scores on the 
American Council Psychological S&amination, and High School Average Gradej cri- 
terion measures were the average grades earned during each quarter and during 
the entire college freshman year* In the comparison of veterans with the non- 
veteran males and with the female group, ^First-Year College Average Grade was 
the criterion measure, as usualj the predictors were High School Average Grade 
and ACEE total score 

Western State University is a large coeducational institution. The 
typical within-state student is admitted primarily on the basis of his high 
school record, the basic standard being an average grade of W C" or higher. 
As a freshman in the college of arts and science, he is required to take 
English j otherwise, his program is determined by broader requirements. 

T&B group included in the analysis may be defined as follows; students 
who entered in the ;f aH of 19^6 as beginning freshmen and who completed ten 
or more quarter hours of work during each of the three quarters. Students who 
were not enrolled in freshman English during any of the three quarters were ex- 
cluded in order to increase the homogeneity of the group. The group included 
^33 veteran and 222 male nonveteran students. Tha following additional features 
of the group were determined from the questionnaires s About 25 per cent of the 
male veteran group were under twenty years of age upon entrance; almost 65 per 
cent had two years or more of active duty; and about 15 per cent were married. 
Slightly over 80 per cent of the veterans would probably have attended college 
without the GI Bill. Among the male veterans, less than 20 per cent reported 
that their fathers had completed college^ among male nonveterans the percentage 
was slightly higher; and among females, it was about 25 per cent. With respect 
to housing, about 50 per cent of the male veterans lived at home or with near 
relatives, and a little over 10 per cent said they owned or rented their own 
house or apartment . Slightly less than 70 per cent of the male nonveterans and 
60 per cent of the females were living at home or with relatives. 

Inter correlations were computed for each of the three subgroups, based on 
a larger number of variables than were used in the analysis of covariancej the 
results are shown in Tables 15 and 16. Average grades were reported on a five- 
step scale, the numerical values being k to 0* The three sets of inter correla- 
tions show that the validity coefficients for the Q, L, and total scores on the 
ACEE axe rather similar for the three subgroups, although the male nonveterans 
appear to be slightly more predictable on the basis of the total score- The 
validity of high school standing as usual is higher for nonveterans than for 
veterans | but the highest coefficient (.62) is obtained for the women students. 
Contrary to the findings at Douglas University, where the verbal tests gave 
the best prediction of fresnman grade for women, the best predictor of women f s 
grades at Western State is the High School Average* 



122 



ADJUSTMENT TO COLLEGE 



Table 15 

OP AMERICA!? COUNCIL PSYCHOLOGICAL EXAMINATION SCORES, HIGH 
SCHOOL AVERAGE, AM) KtRST-IEAB COLLEGE AVERAGE GRADES 

Western State University, College of Arts and Science, Male Freshmen, 1946-1947 





Male Nonveterans (N222) 


5-Q 

5-L 
3-T 
i School Average 


-Quarter Average 
Quarter Average 
-Quarter Average 
-Year Average 








W 


p 'd d -P 
w S h to 
H (M m H 


^ 


CQ 




ACH5 -Quantitative (rav score) 


\.44 .74 .38 


.32 .24 .30 .32 


^5 


10 




ACRE -Linguistic (rav score) 


.45\>89 .39 


.47 .37 .42 .48 


69 


15 


/H 


ACPE-Total (rav score) 


,77 .90 >v .47 


.48 .39 ,45 .51 


114 


21 


II 


High School Average 


.29 .32 .36 N^ 


.57 -W .51 -59 


2.93 


51 


1st -Quarter College Average 


.25 .42 ,41 .46 


\, .66 ,62 .85 


a.iA 


.60 


0) 

p 


2nd-Quarter College Average 


.27 .40 .40 M 


.61N. .63 .88 


2. 41 


.65 


Q) 


3rd-Quarter College Average 


.27 .38 .39 *45 


-57 .65 \-87 


2.43 


.66 




l^irst-Year College Average 


.31 .46 n 46 .53 


,82 .88 .85 \v 


2.43 


,56 


Mean 


43 71 114 2.69 


2.38 2.38 2.42 2.39 






Standard Deviation 


10 15 22 .50 


.68 .72 .71 .62 







ACADEMIC ADJUSTMENT OF STUDENTS 



123 



Table l6 

IHTERCOREEIATIONS OF AMERICAN COUNCIL PSYCHOLOGICAL EXAMINATION SCORES, HIGH 
SCHOOL AVERAGE, AND FIRST -YEAR COLLEGE AVERAGE GRADES 

Western State University, College of Arts and Science, Female Freshmen, 





Female Students (N=482) 





fl) nj fl> 

S> W w 








1 


CD CD CO 
h fc fc (U 
4> 1) <U M 
fe P fe ro 
< aj < ^i 








1 1 1 


0) o> <u <J 
S, 9 P S 

| 

H <53 CO H 


a 


CO 


ACEE -Quantitative (raw score) 


,51 .80 .33 


.36 .23 .29 .33 


39 


10 


ACPE -Linguistic (raw score) 


90 . 44 


A2 ,38 .3k A5 


67 


15 


ACTS -Total (raw score) 


.1*6 


A3 .36 .36 A3 


106 


22 


High School Average 




.56 .53 A9 .62 


3.08 


A9 


1st -Quarter College Average 




,60 .5^ .82 


2.W 


59 


2nd-Quarter College Average 




.58 .84 


2. 47 


.58 


3rd-Quarter College Average 




.82 


2.47 


.60 


First -Year College Average 






2.46 


.50 



Includes 16 female veterans. 



124 ADJUSTMENT TO COLLEGE 



A rough estimate of the reliability of quarter average grades is fur* 
nished "by the inter correlations of the quarter averages. These values range 
from .5^ to .66. Assuming that the reliability of the quarter average grades 
is .60, use of the Spearman-Brown formula would suggest that the reliability 
of the freshman average grade variable is about C 82. These values are probably 
under estiioates as compared with the hypothetical split -half reliability. 

With regard to mean aptitude scores, it is found that the ACFE means are 
almost identical for veteran and nonveteran male students, but lower for female 
students, especially on the Quantitative score. The female students on the 
other hand have the best high school standing, the male veterans being poorest 
The same rank order prevails with regard to all the mean average grades, althougl 
the differences are slight'. 

The analysis of covariance results for the male veteran-male nonveteran 
comparison is shown in Table 17. The ACFE total score and High School Average 
Grade were selected as the predictors; the correlations, means, and standard 
deviations are taken from Table 15. The multiple correlations are .60 and 
.65 for the two subgroups and ,6l for both combined. 

Turning to the analysis of covariance results under III, we see that the 
standard errors of estimate are significantly different at the 2$ levelo 
Strictly speaking the slopes are therefore not comparable The test becomes 
fairly sensitive, however, with a group as large as that employed in this 
malysis; so it was considered permissible to proceed to the test of Hypothe- 
sis B that the slopes are not significantly different. The intercepts of 
;he regression planes are significantly different at the 5$ level. The raw 
Lifference in mean freshman grade is .04 in favor of the nonveteran; when al- 
.owance is made for the difference in ability, the difference is .09 in favor 
if the male veteran student. This is equivalent to .18 in standard error of 
stimate units, from which we find that 57 per cent of the veterans excel the 
verage nonveteran It may be concluded that the grades of veteran students 
re less predictable than those of male nonveterans; the difference in errors 
f estimate is significant at the 2$ level. The grades of the veterans are 
ore variable than those of nonveterans, perhaps because of greater hetero- 
eneity with regard to courses chosen by veteran students. The findings with 
egard to superiority in grades relative to ability are ambiguous. 

Two additional analyses of covariance, comparing male nonveterans with 
smale students and male veterans with female students, are possible from 
aese data. Table 18 shows the results for the first of these comparisons. 

In spite of the identical multiple correlations (.65), the hypothesis of 
jual errors of estimate is disproved at the 5$ level of significance; the 
*ror in prediction is greater for the male students. The inequality is ap- 
irently due to the difference in variability of freshman average grades (the 
;andard deviations are .56 and .50); the test becomes quite sensitive with 
irge F's. The slopes of the regression lines are not proved to be different, 
>r is the difference in intercepts greater than would be expected by chance. 
ten allowance is made for differences in ability, the male nonveterans and 
imen achieve grades which are almost exactly the same. This is consist- 



ACADEMIC ADJUSTMENT OF STUDENTS 



125 



Table 1? 

COMPARISON OF AVERAGE GRADES EARNED BY VETERAN AM) NONVETERAN MAJJE STUDENTS 
Western State University,, College of Arts and Science, Freshmen, 1946-194? 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


K 


ACPE 


E. S. 
Aver&ge 
Grade 


Ist-Yr. 
Average 
Grade 


1. ACEE Total 
(raw score) 


MV 

m 




.36 

A? 


.46 
-51 


114 
114 


22 

21 


^33 
222 


2. High School 
Average Grade 


m 
m 


.36 
A? 




53 
.59 


2.69 
2.93 


.50 

.51 


^33 
222 


3. First -Year College 
Average Grade 


m 
m 


.46 
.51 


53 
59 




2.39 
2.43 


.62 
.56 


^33 
222 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 


Multiple E 


Male Veterans 


.60 


Male Nonveterans 


.65 


Combined Group 


.61 



III. Analysis of Covariance Results : 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
"would arise by chance 


A. Equality of errors of estimate 


6.0537 


1 


Between .01 and .02 


B. Equality of slopes 


0.3921 


2 


Between .80 and .90 


C. Equality of intercepts 


4.5398 


1 


Between .02 and .05 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 

0.09 
0.18 
57 
Ambiguous 



126 



ADJUSTMENT TO COLLEGE 



Table 18 

COMPARISON OP AVERAGE GRADES EARNED BY MAIE NONVETERAN AND JEMAIE STUDENTS 
Western State University, College of Arts and Science, Freshmen, 1946-1947 

I. Correlations, Means, and Standard Deviations: 



Sub- 


Correlation with: 


Mean 


SD 


N 




H. S. 


Ist-Yr. 




ACPE 


Average 
Grade 


Average 
Grade 








1. ACPE Total MN 
(raw score) F 




.47 
.46 


.51 


114 
106 


21 
22 


222 
482 


2. High School MN 
Average Grade F 


.47 
.46 




-59 
.62 


2.93 
3.08 


-51 
.49 


222 
482 


3. First -Year College MN 
Average Grade F 


-51 

.45 


.59 

.62 




2.43 
2.46 


.56 
.50 


222 
482 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Male Nonveterans 
Female Students 
Combined Group 



Multiple R 



65 
65 
.65 



III. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise bv chance 


A. Equality of errors of estimate 


4.1558 


1 


Between .02 and .05 


B. Equality of slppes 


3-2419 


2 


Between .10 and .20 


C. Equality of intercepts 


0.0648 


1 


Between -70 and .80 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of male nonveterans excelling the average female student 

Level of significance of difference (from IIIC above) 



Male Nonveteran 

.00 

.01 

50 

Not significant 



ACADEMIC ADJUSTMENT OF STUDENTS 127 



ent with the results at Douglas University, "where no significant difference 
was found between male and female nonveteran students. The results for 
Western State must "be interpreted with caution, however, in view of the 
significant difference in errors of estimate. 

Turning to the analysis of covariance of male veteran and female stu- 
dents (the two largest subgroups), the results of which are shown in Table 
19, we find that the multiple correlations are respectively .60 and .65, and 
that the hypothesis of equal errors of estimate is disproved at the 1$ level 
Probably all we are justified in concluding is that male veteran and female 
students at Western State are not equally predictable, and that we cannot 
draw any conclusion as to which subgroup achieves higher grades in relation to 
ability. Grades of female students are distinctly more predictable, in terms 
of standard errors of estimate, than those of iiale veteran students. 

Central State University. At Central State, two separate groups were studied. 
The first of these included freshmen entering in 19^6; the second, freshmen 
entering in 19^5 The latter group was unique in the study; in no other insti- 
tution were veteran and nonveteran male students who entered in 19^5 compared. 
For the 19^6 group, the comparison of achievement of veterans and nonveterans 
was made after allowing for differences in High School Average Grade and in a 
composite entrance test score For the 19^5 group, only the Composite Test 
was used in the analysis of covariance . 

Central State is a large midwestern state university which admits stu- 
dents living in the state upon completion of an appropriate high school course 
of study Although it is coeducational, the data for the female nonveterans 
were not included in the academic phase of this study. As usual, freshmen in 
arts and science found that their program permitted a number of alternative 
ways of meeting various requirements. 

For the larger group at Central State, the following definition may be 
stated: male students who entered as freshmen in the fall of 19^6, vho com- 
pleted two semesters of academic work during the academic year 19^6-19^7> 
and who returned a questionnaire. The group contained k66 male veterans and 
166 male nonveterans. The following additional information regarding this 
group was determined from the questionnaires: About 25 per cent of the 
veteran group were under twenty at time of entrance; approximately 65 per 
cent were in the service two years or longer; and slightly over 10 per cent 
were married. Almost 80 per cent of the veterans indicated that they would 
probably have attended college without federal aid. About 15 per cent of 
the veterans' fathers had been graduated from college as compared with about 
25 per cent for the male nonveterans. These figures are strikingly similar to 
those for the corresponding groups at Western State. However, the two insti- 
tutions differ considerably on method of housing students B A variety of hous- 
ing arrangements prevail at Central State, with slightly over 30 per cent of 
the veterans and a little over 40 per cent of nonveterans living in college 
dormitories, while only about 10 per cent of veterans and no nonveterans had 
done so at Western State. Somewhat over 10 per cent of the veterans at 
Central State supplied their own housing through rental or ownership of a 
house or apartment, and just over 20 per cent of veterans and of nonveterans 
lived in rooming or boarding houses. Only about 10 per cent of each group 
lived with parents or other near relatives. 



128 



ADJUSTMENT TO COLLEGE 



Table 19 

COMPARISON OP AVERAGE GRADES EARNED BI MAO1 VETERAN AMD FEMALE STUDENTS 
Western State University^ College of Arts and Science r Freshmen, 1946-1947 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


3D 


ff 


ACPE 


H. S. 
Average 
Grade 


Ist-Yr. 
Average 
Grade 


1. ACPi Total 
(ray score) 


m 

F 




.36 

-46 


,46 
*45 


114 
106 


22 
22 


433 
482 


2, High School 
Average Grade 


MV 
F 


.36 
.46 




-53 
.62 


2.69 
3-08 


.50 
.49 


433 
482 


3. First -Year College 
Average Grade 


m 

i 


,46 
.45 


.53 
.62 




2-39 
2.46 


.62 
.50 


433 
482 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Male Veterans 
Female Students 
Combined Group 



Multiple R 



.60 
.65 
,62 



III. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
vould arise by chance 


A. Equality of errors of estimate 


31.0457 

8O7QA 


1 


Less than .01 




61 coc 








J-PO 







IV Difference beteen Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of male veterans excelling the average female student 

Level of significance of difference (from III above) 



Male Veteran 

0.08 

0.19 

58 

Ambiguous 



ACADEMIC ADJUSTMENT OF STUDENTS 



The following information may be useful in interpreting the results pre- 
sented in Table 20 The Composite Test score was reported as a percentile 
rank based on six tests: Correctness and Effectiveness of Expression, Bead- 
ing Materials in Social Studies, Beading Haterials in Natural Science, Inter- 
pretation of Literary Materials, General Mathematical Ability, and General 
Vocabulary. The percentile scores were concerted to a standard scale with a 
mean of 13 and a standard deviation of 4. The criterion was the First-Tear 
College Average Grade., The grading system used five categories having the 
numerical values k- to 0* 

The results for students entering in the fall of 19^6 are presented in 
Table 2Q It will be noted that the Composite Test has a somewhat higher 
validity for veterans than for nonveterans, and that the validity of High 
School Average is the same for both subgroups j in most institutions it has 
been found that high school standing is more predictive of college grades 
for nonveterans than for veterans * Honveterans have higher means for both 
predictors, but the means are practically the same for freshman average grade. 
The hypotheses of equal errors of estimate and equal slopes are not disproved, 
and the difference in intercepts is significant at the 1$ level. The differ- 
ence in intercepts amounts to .19 grade units* In standard error of estimate 
units, the advantage of veterans is 39* Sixty-five per cent of the male 
veterans excel the average nonveteran, the most extreme difference so far 
encountered o The difference in freshman grades, when, aptitude is considered, 
is highly significant . 

The group of students which entered in 19^5 Bay be defined as follows 2 
siale students who entered in the fall or winter semester of 19*4-5, and who 

completed at least three semesters before the summer of 19*4-7* An eight -week 
summer term was counted as one semester for three of the 135 veterans. Ques- 
tionnaire information obtained from. 63 of the veterans indicated that they 
rere older, as would be expected, than the 19^6 group, and that about 80 per 
sent had completed two years or more of active duty* Roughly 35 per cent 
were married at the tJme of the questionnaire administration (the spring of 
19^7) . Sixty-five per cent indicated that they probably would have attended 
college without the veterans 1 aid. About 15 per cent had fathers who had 
completed college, a figure which agrees closely with that for the veterans 
antering Central State in 19^-6 . About 1*0 per cent of these veterans lived 
In dormitories, and about 20 per cent had their own house or apartment, 

The analysis plan for the group entering in 19^5 differed from that for 
bhe 19^6 group in only two respects : first College Average 

was used as the criterion, end School was not tocluded in 
she analysis of covariance, Basic results are shown in 21. OJbe number 
>f cases is rather small for the 19^5 group; there were only 59 in the male 
lonveteran subgroup . The results must therefore be interpreted with caution. 

The validity coefficients for the Composite Test score were A3 and .1*9 
for veterans and nonveterans respectively Those for School Average 
turned out to be *33 and 065 for the two subgroups. Since the use of both 
)redictore yielded regression planes which differed reliably in elopes (at 
:he 20 level of significance), it as decided to use only the Composite feet 



130 



ADJUSTMENT TO COLLEGE 



Table 20 

COMPARISON OF AVERAGE GRADES EARKED BY VETERAN AM) NOTOETERAN MALB STUDEliTS 
Central State University , College of Arte and 3oie_noe f 

I. Correlations, Means, and Standard Deflations: 







Correlation with: 








Variable 


Sub- 
group 






Mean 


3D 


H 


Composite 


H. S. Avg. 


First -Tear 






Test 


Grade 


Ayg. Grade 








1. Composite Test 


W 




.W 


.58 


12.5 


3.9 


W6 


(converted score) 


m 




.43 


.51 


13.5 


3.6 


166 


2. High Sohool 


m 


. 




.61 


2.W 


.62 


466 


Average Grade 


m 


A3 




.61 


2.7l* 


.6k 


166 


3, First-Year College 


.MV 


.58 


.61 




2.20 


.66 


466 


Average Grade 


m 


51 


.61 




2.19 


.66 


166 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3)j 



Sample 



Male Veterans 
Male Kbnveterans 
Combined Group 



Multiple R 



.69 
6? 
.68 



III. Analysis of Covarlance Results: 



Hypothesis 


Chl- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A, Equality of errors of estimate 


0.016 


1 


Between .80 and .90 


B. Equality of slopes 


0.815 


2 


Between .50 and .70 


C. Equality of intercepts 


17 . 753 


1 


Less than .01 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed In standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 
0.19 
0.39 
65 



ACADEMIC ADJUSTMENT OF STUDENTS 



131 



21 

COMPARISON OF AVERAGE GRADES EABHED BI VETEBAST AM) HOMVE1EKRAN MALE STODE1TS 

Central State University, College of Arts and Science 
( Students entering as freshmen during the academic year 1945-1946 ) 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


JS 


Composite 


1st. 3 Sem. 






Test 


Avg. Grade 








Composite Test 
(converted score) 


m 
m 




.43 
-49 


13.5 
14.5 


3.8 
4.2 


135 
59 


First Three Semesters 
College Average Grade 


m 
m 


A3 
.49 




2.36 
2.38 


33 
.65 


135 

59 



Validity coefficient for combined group: 



II. Analysis of Covariance Eesults: 



Hypothesis 


Chi- 
square 


Degrees 
of 
freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A, Equality of errors of estimate 
B. Equality of slopes 
C. Equality of Intercepts 


2.792 
0.795 
0.313 


1 
1 
1 


Between .05 and .10 
Between 30 and ,50 
Between .50 and .70 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 

0.04 

0.09 

54 
Not significant 



132 ADJUSTMENT TO COLLEGE 



as the predictor, From Table 21, It will be seen tliat the difference in 
errora of estimate falls between the 5$ and 10$ levels of significance. The 
differences In slopes and intercepts of the regression lines, are not signifi- 
cant. The superior subgroup is the male veteran* The difference in inter- 
cepts, however, is only .04 grade units* In standard error of estimate units, 
the advantage of the veterans is .09, which means that $k per cent of the 
veterans exceed the average nonveteran in average grade* 

Littletown State University. Tiro groups, both entering as freshmen in the 
fall of 19^6, were studied at Littletown State. Of these, the group entering 
the College of Arts and Science will be discussed in this sectionj the other 
group, which entered the College of Business, will be discussed in a later 
section. For the arts and science studenta, the analysis was limited to a 
comparison of the achievement of veteran and nonveteran students after allow- 
ing for differences in American Council Psychological Examination total scores. 

Littletown State University is a coeducational state university located 
in a small midwestern city. Students are drawn mainly from the high schools 
of the state, and are admitted on the basis of high school graduation* Stu- 
dents with poor averages in high school are admitted on a probationary basis, 
ffreshmen in arts and science are required to take an English coursej the re- 
mainder of their program is determined by broader requirements. 

The sample under consideration may be defined as follows s students who 
entered as beginning freshman in the fall of 19^6, who completed eleven or 
more hours of work in the College of Arts and Science in each of the two semes- 
ters of the academic year, 19^-6-19^7^ a&d (since the proportion of question- 
naires returned was quite high) who completed a questionnaire. The number of 
students was 103 and 107 for veterans and nonveterans respectively. From the 
questionnaires, further information about the group was obtained, as follows: 
About 30 per cent of the veterans were under twenty years of age at time of 
entrance | slightly fewer than 50 per cent served for two years or more in the 
armed services ^ and 3 per cent were married. Somewhat over 80 per cent of the 
veterans indicated that they probably would have attended college even without 
GI money. About 20 per cent of the veterans indicated that their fathers had 
completed college,' the corresponding figure for nonveterans was slightly below 
30 per cent. Forty per cent of the veterans and about 65 per cent of the non- 
veterans were living in college dormitories at the time of the study. 

The predictor selected for use at Littletown State was again the American 
Council Psychological lamination total score. Kaw scores were uaed. Although 
data on the quarter of the claas in which a student stood were available, it 
was decided to omit this variable so that students for whom it was not avail- 
able could be included in the study * The criterion, First -Tear College Average 
Grade, was actually a point -hour ratio, basad on a scale of five steps having 
the numerical values of k to 0. 

The results are shown in Table 22. Although the validity coefficient of 
the ACPE was somewhat higher for veterans than for nonveterans, the analysis 
of coyariance shows that differences in errora of estimate and slope are no 
greater than would be expected to arise frequently by chance. Toe nonveteran 



ACADEMIC ADJUSTMENT OF STUDENTS 



133 



Table 22 

COMPARISON OB 1 AVERAGE GRADES EARIED IT VETERAN AND -NOMJOTIHAN MALE STUDENTS 
Littletown State University, College of Arts and Science, ffreBhmen, 1946-1947 

I. Correlations, Means , and Standard Deviations; 





Sub- 


Correlation with: 








"TTe-yt-t ftVil > 








M a 


cm 


"KT 




group 


ACPE 


Fir at -Tear 




OiJ 


JM 








Avg. Grade 








ACPE Total 


m 




4p, 


in 


23 


103 


(raw score) 


m 




**l4 


115 


20 


107 


First -Tear College 


m 


49 




2.29 


.68 


103 


Average Grade 


m 


^4 




2.35 


.69 


107 



Validity coefficient for combined groups 



.42 



II. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
[Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C Equality of intercepts 


0*8l4 
0.464 
0.00193 


1 

1 
1 


Between ,30 and .50 
Between .30 and .50 
Between .95 and .98 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from 1IC above) 



Male Nonveteran 

0.00 

0.01 

50 

Not significant 



134 ADJUSTMENT TO COLLEGE 



subgroup obtained higher scores, on the average, for "both the predictor and 
the criterion. The difference in intercepts of the regression lines is 
smaller than usually would arise "by chance; the probability that a value of 
Chi-square greater than that obtained would occur by chance is between ,95 and 
.98 When allowance is made for ability differences, then, the average grades 
of veterans and nonveterans at Littletown State are almost identical and, of 
course, the very slight difference in favor of nonveterans is not significant. 

Eastern City University. For students entering Eastern City in the fall of 
19^6, the analysis was limited to a comparison of male veteran and nonveteran 
students, after allowing for differences in aptitude reflected in High School 
Average Grade and Composite Score on entrance tests. The criterion measure 
was First-Year College Average Grade. 

Eastern City is a municipal college of arts and science located in a large 
city* Although it is coeducational, only male students were included in the 
academic analysis. (Questionnaire results for female students were tabulated 
and will be reported in later chapters.) Students are admitted to Eastern City 
on a competitive basis, using primarily their high school records and their 
scores on entrance examinations administered by the college. Although the 
freshman program in arts and science is not prescribed, a relatively large 
number of required courses, in English, mathematics, and social sciences, 
must be completed before graduation. 

The groups selected for statistical analysis may be described as follows: 
male students who enrolled as beginning freshmen in the fall of 19^6, who com- 
pleted two semesters of academic work, and who returned a questionnaire,, The 
limitation of the group to students entering in the fall greatly reduced the 
dumber of veterans available for study, since many veterans had been admitted 
luring the previous spring. It was considered desirable, however, to exclude 
these veterans in order to obtain as comparable conditions as possible between 
veteran and nonveteran student s The following added information about the 
group was obtained from the questionnaire analysis: Like the Adams, Stewart, 
a,nd Douglas veterans, the veterans at Eastern City were relatively young 
[about kO per cent were under twenty at time of entrance); about kO per cent 
lad completed two years or more in the service; and none were married. Ap- 
proximately 65 per cent of the veterans would probably have attended college 
fithout federal aid Only about 5 per cent of veterans were sons of college 
graduates; for the nonveterans, the corresponding figure was about 15 per cent. 
Practically all of the students, both veteran and nonveteran, lived at home 
:>r with near relatives. 

For the statistical analysis, High School Average and Composite Test Score 
rere combined, using equal weights. (The standard deviations were roughly 
jqual.) The Composite Test Score was based on performance on eight Coopera- 
tive tests: Mechanics of Expression, Effectiveness of Expression, Vocabulary, 
Speed of Comprehension, Level of Comprehension, Mathematics, Natural Science, 
md Social Studies. 

The results of the analysis are presented in Table 23. The validities 
ire somewhat lower than usually found by Eastern City, possibly because the 
subgroups have been made more homogeneous by separating the sexes. The pre- 



ACADEMIC ADJUSTMENT OF STUDENTS 



135 



Table 23 

CQMPABISOH OF AVEBAOE GRADES EARHED BY VETERAN AHD HONVETERAJf MALE 
Eastern City University, College of Arts and Science, Freshmen, 

I, Correlations, Means, and Standard Deviations: 



STODEKTS 



Variable 


group 


Correlation with: 


Mean 


SD 


V 


Comp. Test 


First -Tear 






+ H-S. Avg. 


Avg. Grade 








Composite Test Score 
plus High School Average 


m 




: 


165.5 
167.5 


k.6 
6.5 


lP 7 


First -Tear College 


m 


AS 




0.26 


53 


53 


Average Grade 


m 


5^ 




0.37 


-59 


147 



Validity coefficient for combined group: 



-53 



II. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C> Equality of intercepts 


0.128 
0.018*1. 

O.Olj-lll. 


1 
1 
1 


Betveen -70 and .80 
Between .80 and ,90 
Between .80 and .90 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Nonveteran 
0.02 



Not significant 



136 ADJUSTMENT TO COLLEGE 



dictor score for veterans is lover than for nonveterans, and the veterans 
also earned a lower mean freshman average grade. Che errors of estimate, 
slopes, and intercepts are all somewhat more similar than would be expected 
by chance . When the effect of ability is eliminated , the nonveterans are 
found to have a very slight advantage^ 02 grade units. Forty-eight per cent 
of the veterans exceed the average nonveteranj the amount of overlapping of 
the distributions is very great,, and, of course, the difference is not signifi- 
cant. 

Midwest City Ufaiversityo Two groups, both entering as freshmen in I$k6? were v 
studied at Midwest City* The present section is concerned only with those 
entering the College of Arts and Science! those entering the College of Engi- 
neering will be discussed in the following section,, For Midwest City students, 
the analysis was limited to a comparison of the relative achievement of male 
veteran and nonveteran students after eliminating the effect of differences in 
performance on the American Council Psychological Examination 

Midwest City is a large coeducational municipally-supported institution. 
Students are admitted on the basis primarily of high school record, being se- 
lected from students graduated in the upper two-thirds of their graduating 
classes, Freshmen in arts and science are required to take English and a sur- 
vey course in western civilization and to fulfill additional broader require- 
ments , 

The group studied may be defined as follows i students who entered as be- 
ginning freshmen in the fall of 19^6> who completed seven or more semester 
hours of academic work during each semester, and who returned a questionnaire* 
From the questionnaires completed by this group, the following information may 
be noted s Among the veterans, about 35 per cent were under twenty years old; 
about 50 per cent served two years or more in the armed services; and 10 per 
cent were married. Slightly over 80 per cent indicated that they probably would 
have attended college without their federal scholarship* Slightly less than 
20 per cent of veterans and about 30 per cent of nonveterans reported that their 
fathers had completed college. These results (except for the higher proportion 
of married students) are quite similar to those for Little town State. Substan- 
tially all students In the group studied were living at home or with near rela- 
tives, the percentages being roughly 85 for veterans and just over 90 or non- 
veterans c 

Total raw scores on the ACEE were used as the only predictor for this 
group, data on high school standing were available for only a relatively small 
proportion of the students. The criterion, the First-Tear College Average 
Grade, is based on a scale of six points corresponding to the values 5 "to 0. 
Hie results of the analysis are shown in Table 2k* 

An unusually large difference between the validity coefficients for 
veterans and nonveterans was found for this group at Midwest City; the corre- 
lations are 57 and 29 for veterans and nonveterans respectively,, The means 
and standard deviations of ACEE scores are identical for the two subgroups, 
but the veterans achieve an average grade which is .22' higher than for non- 
veterans. None of the three hypotheses are disproved o The veteran subgroup 



ACADEMIC ADJUSTMENT OF STUDENTS 



137 



Table 2k 

COMPABISOQST OF AVERAGE GRADES EARNED BT VETERAN AM) NOKYEEEBAN MALE STODEMTS 
Midwest City University, College of Arts and Science, Freshmen, 19^6-1947 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


N 


ACEE 


First -Year 
Avg . Grade 


ACEE Total 
(raw score) 


m 

m 




57 
29 


J20 
120 


23 
23 


83 
72 


Pirst-Year College 
Average Grade 


m 

m 


-57 
.29 




3.^0 
3.18 


77 
.80 


83 

72 



Validity coefficient for combined group: 



II. Analysis of Covariance Results: 



Btypotaeais 


CM- 
square 


Degrees 
of 
Freedom 


Probability tnat a value 
greater than obtained 
vould arise by chance 


A* Equality of errors of estimate 


2-872 


1 


Betveen .05 and .10 


B, Equality of slope i 


3-323 


1 


Between .05 and ,10 


C, Equality of intercepts 


3.622 


1 


Between ,05 and .10 



III. Difference between Subgroups with Ability Held Constants 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 

0.22 

0.31 

62 

Wot significant 



138 ADJUSTMENT TO COLLEGE 



is superior by the same amount , .22 grade units, when ability is considered, 
since the two subgroups have the same means and standard deviations on the 
ACEE. Although 62 per cent of the veterans exceed the average nonveteran, 
the difference is not significant because of the small number of cases. 

Summary. The results for the six groups from five state and municipal col- 
leges of arts and science are in general similar to those found for the pri- 
vate institutions. For the private colleges, the veterans showed higher 
achievement in six out of seven cases, but the difference was significant (at 
the 2$ level) in only one case. For the public arts colleges, the veterans 
vere superior in four out of six cases. The superiority of the veterans was 
significant at the 1$ level for one institution (Central State ) At another 
(Western State) the results of the significance test were ambiguous because of 
difference in errors of estimate, which was significant at the 2$ level. 

The tendency for high school standing to have greater validity for non- 
veterans than for veterans is confirmed in one case, but at Central State the 
coefficient was the same for both subgroups. 

Grades of Veteran and Nonveteran Students in Engineering 
Colleges of State and Municipal Universities 

Midwes t Te chnologi cal Uhiver s i ty . At Midwest Technological University a 
relatively extensive analysis was made. Inter correlations were determined 
for the following variables: part and total scores on the American Council 
Psychological Examination, English and mathematics placement test scores, 
High School Average Grade, average grades for each quarter of the freshman 
college year, number of college credits per quarter, and course grades in 
English and mathematics. The comparison of veteran and nonveteran engineer- 
ing students took into account differences between the two subgroups in High 
School Average Grade and in total score on the ACPE. 

Midwest Technological University is a m land-grant college in a state which 
also supports a state university. Students who are residents of the state are 
admitted upon completion of an appropriate high school program. Students in 
the different branches of engineering take a common freshman year, which in- 
cludes chemistry, drawing, English, college algebra, trigonometry and analytic 
geometry. 

The group selected for study may be defined as follows: students who 
entered as beginning freshmen in the fall of 19^6, who attended classes on 
the main campus, and who completed three quarters of engineering college 
work, with ten or more hours of academic work each quarter. 'Veterans who 
received ten or more credits in specific courses for training received during 
their service were considered not to be beginning students. The limitation of 
the study to those students studying on the main campus (thus excluding a sub- 
stantial number receiving parallel training on a nearby sub-campus) was con- 
sidered desirable on the grounds that the comparability of results would be 
greater for those students trained under more nearly typical conditions. Ee- 



ACADEMIC ADJUSTMENT OF STUDENTS 139 



garding the group chosen for study, the f ollowing information was obtained 
by examination of some results of the questionnaire analysis: About 30 per 
cent of the veterans were under tventy years of age; 60 per cent had completed 
two years or more of active duty; and roughly 15 per cent -were married. About 
75 per cent would probably have attended college without the GI scholarships. 
Fifteen per cent of the veterans and 30 per cent of the nonveterans reported 
that their fathers had completed college. Housing arrangements were diversified, 
with no predominant pattern. About 15 per cent of the veterans owned or were 
renting a house or apartment. 

Two sets of intercorrelations were calculated, involving a larger number of 
variables than was used for the analysis of covariance. These intercorrelations 
are shown in Table 25. Th statistics at the right of the diagonal are for non- 
veterans, and those below the diagonal are for veteran students. 

The predictors studied include total score and part scores on the ACE 
Psychological Examination. Scores on this test were reported as percentiles, 
which were converted to a standard scale having a mean of 13 and a standard 
deviation of k. The English and mathematics tests were placement tests. The 
Mathematics Test was prepared at Midwest Tech, and the English Test was pre- 
pared by the United States Armed Forces Institute . The criteria, quarter 
average grades for the three quarters of the freshman college year, are based 
on a five-step scale with the numerical values k to 0. 

Comparison of the means of veterans and nonveterans on the predictive 
measures shows that the average veteran excels on only one test, the English 
Test, although the differences in means are generally small. The greatest ad- 
vantage of the nonveterans appears to be in mathematics, where the difference 
in means amounts to about half of a standard deviation. Turning to the means 
for quarter averages and course grades, we find that the veterans were higher 
in every case . The mean number of credits is greater for nonveterans than for 
veterans in each of the three quarters. 

The correlation of number of credits for each quarter with average grade 
for that quarter is low and positive for both subgroups, indicating a tendency 
for the better students to take heavier course loads. 

In general, the best predictors of quarter average grades are total score 
on the ACEE and High School Average Grade. There is very little difference at 
Midwest Tech in the predictive value of high school standing for veterans and 
nonveterans, although the correlations with average grades tend to be slightly 
higher for the nonveteran subgroup. The ACTE tends to have slightly higher 
validity coefficients for the veterans than for the nonveterans, except in 
the case of First -Quarter College Average Grade. Here the correlation is 
higher for the nonveteran subgroup (.65 as compared with .53 for veterans, a 
difference which is significant at the % level). As might be expected, the 
best predictors of English grades are the English placement test and the L- 
score of the ACFE. The Mathematics Test is the best predictor of mathematics 
grades for both veterans and nonveterans, the validity coefficients for non- 
veterans being somewhat higher. 



140 



ADJUSTMENT T COLLEGE 



a *s 



m 



co of) oo no oo 



o 

H^JHH ?H 



vo 



O OO LfMA 
OJ O 



OJ OJ Al <M CVI CM CM 



TOT * 
SOT ' 

TOT - 




8V 



3 



TV 



80*1 

96- 

6' 

6L- 

69' 

el- 

69' 




S'fit 



E'T 
E'T 
S'T 



-g - 



b-SHOY 




SI'S 

e'vc 

8'T 

2/rr 

VET 



39' 
V 
V 



S'E 



edits 
redits 
edi 




ege Average 
lege Averag 
ege Average 



l 
ll 



r Coll 
er Co 
r Co 



rade 



e 
ade 
1 G 
2 G 



rad 



r 

10 



ri 7 T nd 1 



Firs 
Seco 
Thir 
Engl 



ish 101 G 
ish 102 G 
Mathematics 1 
Mathematics 



ACADEMIC ADJUSTMENT OF STUDENTS 141 



The inter correlations of the quarter averages range from .68 to .78 with 
a median of about .73- Assuming a reliability of .73 for the quarter averages, 
use of the Spearman-Brown formula would predict that the reliability of average 
grades based on three quarters would be 089. When it is recalled that this 
method of estimating reliability tends to give an underestimate (because the 
three quarter grades are not random thirds of the freshman average grade),, this 
figure may be considered satisfactorily high. 

The results of the analysis of covariance are shown in Table 26. The 
multiple correlations of the ACPE in combination with High School Average are 
unusually high: .72 and ,76 for veterans and nonveterans respectively, and .70 
for both subgroups combined. As has already been noted, the veterans on the 
average make lower ACPE scores and have a lower High School Average, but achieve 
higher First -Year College Average Grades in the College of Engineer ing The 
hypotheses of equality of errors of estimate and slopes of regression planes 
are not disproved, but the difference in intercepts is significant at the 1% 
level. There is, then, a marked tendency for veterans to earn higher grades 
than nonveterans of equal ability as measured by the predictors used. The 
difference amounts to .30 grade units or .70 error of estimate units. Seventy * 
six per cent of the veterans exceed the average nonveteran, which is the largea 
difference found in the study of twenty -five different groups, 

Middle State Univer a ity . At Middle State, inter correlations of a number of 
predictors and criteria were also studied. Here, American Council Psychologica 
Examination total score, High School Rank, Year of High School Graduation, 
First-Year College Average Grade, and course grades in drawing and mathematics 
were analyzed. A unique feature of "the analysis of Middle State data is the 
fact that veterans were compared with nonveterans not only on the basis of 
First-Year Average Grade, but also on their grades in mathematics and draw- 
ing. In all three comparisons, differences in High School Rank and in ACPE 
total score were taken into account. 

Middle State is a large state university located in a midwestern city. 
Students of engineering are drawn from graduates ranking in the upper half 
of their high school classes. Chemistry, English, mathematics, and drawing 
are required courses for engineering students. 

The group studied may be defined as follows: students who were admitted 
to the College of Engineering as freshmen in the fall of 19^6, who completed 
three quarters of work, including eleven or more hours each quarter, and who 
completed a questionnaire. Transfer students were excluded, as usual. From 
the questionnaires, it was learned that about 35 per cent of the veterans were 
under twenty upon entrance, that about 55 per cent served two years or more 
on active duty, and that approximately 10 per cent were married. Somewhat 
over 75 per cent would probably have attended college without the GI Bill. 
Ten per cent of veterans reported that their fathers had been graduated from 
college," the corresponding figure for nonveterans was roughly 15 per cent. 
About 65 per cent of the veterans and about 75 per cent of the nonveterans 
were living at home or with near relatives at the time the study was made. 



142 



ADJUSTMENT TO COLLEGE 



Table 26 

COMPARISON OP AVERAGE GRADES EARMD BI "VETERAN AMD HONVETERAU MALE STODEKTS 
Mlcbreat Technological University, College of Engineering, Freshmen 

I. Correlations, Means, and Standard Deviations: 







Correlation with: 








Variable 


Sub- 




H. S. 


Ist-Tr. 


Mean 


3D 


w 




group 


ACPE 


Average 


Average 














Grade 


Grade 








1, ACJE (19^5) Total 


MV 




M 


.60 


13.7 


3.5 


271 


(converted score) 


m 




.37 


.61 


Ik.k 


3.5 


128 


2, High School 


m 


M 




.62 


2.75 


.62 


271 


Average Grade 


m 


.37 




.53 


2.96 


.56 


128 


3 First -Year College 


m 


.60 


.62 




2.39 


.63 


271 


Average Grade 


m 


.61 


53 




2.24 


.63 


128 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



.e 



Male Veterans 
Male Nonveterans 
Combined Group 



Multiple E 



-72 
.76 
.70 



III. Analysis of Covariance Results; 



hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 


0.601 


1 


Between ,30 and .50 


B. Equality of slopes 


1.718 


2 


Between .30 and .50 


C. Equality of intercepts 


39.780 


1 


Less than .01 



IV. Difference between Subgroups with Ability Held Constant : 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 

0.30 

0.70 

76 

1* 



ACADEMIC ADJUSTMENT OF STUDENTS 143 



Hie predictors eigployed in the study included total score on the 1937 
edition of the American Council Psychological Examination (expressed in terms 
of raw score) and Etgh School Bank* The ACPE was given to high school stu- 
dents in their senior year throughout the state a It was administered at the 
university in cases where a student lacked a score on this test* She princi- 
pal source of data was the high school testing program, but in a few cases 
scores obtained at the time of college entrance were used. The previous 
studies at Adams and Stewart indicate that the time of testing, within the 
limits encountered here, is of little importance, The other predictor used, 
rank in high school class, was converted from percentile ranks to a standard 
scale with a mean, of 13 and a standard deviation of ^. 

The principal criterion employed is the First-Tear College Average Grade 
(actually the point-hour ratio), based on a four-category scale having the 
equivalent numerical values of 3 to 0. First-quarter grades in engineering 
drawing and mathematics were used as additional criteria. For these variables, 
a five-step grading scale was used, the values ranging from 3 to -1 Still 
another variable, Tear of High School Graduation, was used. The data for 
this variable were taken from, questionnaire item 6(b) "When were you, last in 
full time attendance in high school or preparatory school?" The item was 
preceded in such a way that 1 before . 19^-0, 2 * 19^-0, 3 19^1, etc. Ordi- 
narily the response to the item would indicate year of high school graduation; 
the item was put into the form shown above because results for year of high 
school graduation would be misleading 'in the case of veterans who were granted 
high school diplomas, after war service on the basis of OSAFI examinations or 
credit for military service. 

The inter correlations, means, and standard deviations of these variables 
are shown in Table 27* The mean Tear of Sigh School Graduation indicates that 
most nonveterans graduated in 191*6, while the typical veteran last attended 
high school in 19%. (The item code from the table less 2 gives the last 
digit of the year.) Veterans, of course, showed much greater variability 
with respect to this measure. Jfonveterans had higher means on both ACPE and 
High School Rank, while the veterans were superior, on the average, on all 
three criterion measures employed. 

In the case of nonveterans^ the correlations of Tear of High School 
Graduation with other variables are somewhat meaningless, because of the 
narrow range and highly skewed distribution. Whatever correlation is found 
is due to the few nonveterans -who graduated prior to 19^6* For veterans the 
correlations are more meaningful; they range from .20 (with High School Rank) 
to -.01 (with Drawing Grade). Except for the r of -.01, all the correlations 
are positive and above .08, which may indicate a very slight tendency for the 
more recent high school graduates to be superior. Such a relationship could 
easily result from admissions policy or a process of self -selection and does 
not necessarily reflect a general tendency o 

The best predictor of First-Tear Average Grade ia High School Rank; the 
r f s are *51 and .53 for veterans and nonveterans respectively. The ACPE 
yields a validity coefficient for veterans which is considerably higher than 
for nonveterans; the coefficients are A2 and .28. As might be expected, the 



144 



ADJUSTMENT TO COLLEGE 



Table 27 

OP AMERICAN COUNCIL PSYCHOLOGICAL EXAMINATION TOTAL SCORE, 
HIGH SCHOOL RANK, TEAS OF HIGH SCHOOL GRADUATION, 
AM) FIRST -YEAR COLLEGE GRADES 

Middle State University. College of Engineering, Freshmen, 1946-19*17 





Male Nonveterans (N=98) 


05 


CD 








3 


? f * 








* s 

EH & 
| 


fc $ 

43 oj 43 
03 H CtJ 
H P 


9 
<u 


R 
CO 




ACEE -Total (raw score) 


\ A9 .Ik 


.28 .18 .28 


,3.! 


22.3 




High School Rank 


.k6 \ .01 


.53 .14 .40 


16.4 


3.3 


(r 

n 




Year of High School Graduation 


.10 .20 \ 


-.14 .10 -.09 


7.8 5 


.71 


First -Tear College 
Average Grade (4-step scale) 


42 a "51 o OQ 


\ -35 *75 


1-32 


.60 


I 


Drawing Grade 
First Quarter (5-step scale) 


.23 o27 -.01 


.50 \ .19 


0.53 


.91 


i 


Mathematics Grade 
First Quarter (5-step scale) 


,34 o4i .10 


.76 .30 \ 


0.39 


1.10 


Mean 


88,1 15.6 5.15 


1.40 0.71 0.49 






Standard Deviation 


21.5 3-1 1.67 


.66 .83 1.10 







ACADEMIC ADJUSTMENT OF STUDENTS 145 



Drawing Grades are not veil predicted by either of the predictive measures 
used, Validity coefficients for Mathematics Grades are somewhat higher and 
are of about the same magnitude for veterans and nonveterans. 

Three analyses of covariance were computed for the data at Middle State 
University, one for each of the criteria employed* The results for First - 
Tear College Average Grade are shown in Table 28. The multiple correlations 
are .56 and .53 for veterans and nonveterans respectively, and ,53 for both 
subgroups combined. The hypotheses of equal errors of estimate and equal 
slopes are not disproved, but the hypothesis of equal intercepts of the gener- 
alized regression planes is disproved at the 1$ level of significance The 
.08 advantage of the veterans in raw mean average grade is increased to .19 
when the effect of ability is taken into account. In standard error of esti- 
mate units, the veterans' advantage is ,3!*; 63 per cent of veterans excel the 
average nonveteran. The advantage of veteran students in First-Year Average 
Grade at the College of Engineering at Middle State University is highly 
significant* 

Table 29 presents the results where Drawing Grade is used as the cri- 
terion- This criterion is less predictable than either of the others; the 
three multiple correlations are .30, .19, and .26. The intercepts are sig- 
nificantly different at the 1% level, the difference in intercepts being .25 
grade units. In error of estimate units, the difference is .30, and 62 per 
cent of the veterans are found to exceed the average nonveteran when the 
measures of ability are taken into account . 

The results of the analysis of covariance in which Mathematics Grade is 
used as the criterion are shown in Table 30 Here the three multiple corre- 
lations are very similar in magnitude, .t, Al, and .^3. The differences 
in errors of estimate and slopes of the regression planes are no greater than 
would be expected by chance, but the intercepts are significantly different 
at the 5$ level. The advantage of the veteran subgroup, which is 10 in raw 
grade units, becomes .24 when allowance is made for the difference in ability* 
Fifty -nine per cent of the veterans excel the average nonveteran 

For all three criteria used at Middle State, then, the veterans proved 
to achieve significantly higher grades than nonveterans when allowance is 
made for the difference in ability,, When Mathematics Grade is used as the 
criterion, however, the difference is significant at only the 5$ level, 

.Midwest City University. The analysis of data for engineering students at 
Midwest City was limited to a comparison of their relative first-year grades 
after allowing for differences in total score on the American Council Psycho- 
logical Examination. 

Like the students in arts and science, Midwest City engineering students 
are drawn from the upper two -thirds of their higjh school graduating classes* 
As freshmen, they take English, drawing, descriptive geometry, chemistry, col- 
lege algebra, trigonojmetry, analytic geometry, and mechanics. The engineer- 
ing program at MtLdwest City is well known for its practical emphasis. 



146 



ADJUSTMENT TO COLLEGE 



Table 28 

COMPAEISON OF AVERAGE GRADES EARNED BT VETERAN AND NONVETERAN MALE STUDENTS 
Middle State University, College of Engineering, Freshmen, 1946 -194 7 

I. Correlations , Means ? and Standard Deviations: 







Correlation withs 








Variable 


Sub- 
group 


ACEE 


H. S. 

Ro-nV 


lst-Tr. 
Average 


Mean 


SB 


N 










Grade 








1. ACHE (1937) Total 


MV 




.46 


.k2 


88.1 


21.5 


352 


(raw scor'e) 


m 




JL.Q 


.28 


93.1 


22.3 


98 


2. High School Barik 


m 


.46 




51 


15.6 


3-1 


352 


(converted score) 


m 


.49 




.53 


16.1). 


3-3 


98 


3. First-Year College 


m 


.42 


.51 




1.40 


.66 


352 


Average Grade 


m 


.28 


53 




1.32 


.60 


98 


(4 -step scale) 

















II, Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Male Veterans 
Male Nbnveterans 
Combined Group 



Multiple E 



.56 
*53 
.53 



III. Analysis of Covariance Eesults: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. .Equality of intercepts 


0.913 
4,702 
8.816 


1 
2 
1 


Between .30 and .50 
Between .05 and .10 
Less than 01 



IV. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 

0.19 

0.34 

63 



ACADEMIC ADJUSTMENT OF STUDENTS 



147 



Table 29 

COMPARISON OF DRAWING GRADES EARNED BY VEEERAN AND NONVETERAN MAIE STUDENTS 
Middle State University, College of Engineering, ffreshmen, 1946-1947 

I. Correlations, Means, and Standard Deviations: 







Correlation withs 










Sub- 










Variable 


group 


ArPTH 


H. S. 


Drawing 


Mean 


SD 


N 








Rank 


Grade 








1. ACEE (1937) Total 


MV" 




M 


.23 


880 1 


21.5 


352 


(raw score) 


m 




49 


.18 


93.1 


22.3 


98 


2, High School Rank 


m 


.46 




.27 


15.6 


3.1 


352 


(converted score) 


m 


.49 




.14 


16.4 


3.3 


98 


3. Drawing Grade- 


m 


.23 


o27 




Oo71 


.83 


352 


1st Quarter 


m 


as 


.Ik 




0.53 


91 


98 


(5-step scale) 

















II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Male Veterans 
Male Nonveterans 
Combined Group 



Multiple R 



.30 
.19 
.26 



III. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


2.197 
6.849 


1 
2 
1 


Between .10 and .20 
Between .30 and .50 
Less than .01 



IV. Difference between Subgroups with Ability Held Constants 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonvetdran 

Level of significance of difference (from IIIC above) 



Male Veteran 

0.25 

0,30 

62 



148 



ADJUSTMENT TO COLLEGE 



Table 30 

COMPARISON OF MATHEMATICS GRADES EARNED BY 
"VETERAN AND NONVETERAN MALE STUDENTS 

Middle State University, College of Engineering, Freshmen, 1946 -1947 
I. Correlations, Means, and Standard Deviations: 







Correlation with: 










Ch-Th 










Variable 


group 




H. S. 


Math. 


Mean 


3D 


N 








Rank 


Grade 








1. ACHE (1937) Total 


MV 




.46 


.34 


88.1 


21.5 


352 


(raw score) 


m 




.49 


.28 


93.1 


22.3 


98 


2. High School Rank 


m 


.46 




.41 


15.6 


3-1 


352 


(converted score) 


m 


.49 




.4Q 


16.4 


3*3 


98 


3. Mathematics Grade- 


m 


.34 


.41 




0.49 


1.10 


352 


First Quarter 


m 


.28 


.40 




0.39 


1.10 


98 


(5- step scale) 

















II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 


Multiple R 


Male Veterans 


.44 


Male Nonveterans 


.41 


Combined Group 


.43 



III. Analysis of Covariance Results s 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
"would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


Oo0566 

0.725 
4.342 


1 
2 
1 


Between .80 and .90 
Between .50 and .70 
Between .02 and .05 



IV. Difference between Subgroups with Ability Held Constant j 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IHC above) 



Male Veteran 

0.24 

0,24 

59 



ACADEMIC ADJUSTMENT OF STUDENTS 14 



The sample upon which the analysis was based may be defined as follows: 
students entering as beginning freshmen in the fall of 191^6, -who completed 
the full academic year, 1946-19^7, and who returned a questionnaire. Further 
information which may help to describe the group -was determined from the ques- 
tionnaires. The veterans in this group were relatively old,, only about 10 
per cent being under twenty years of age. They had extensive military ser- 
vice | some 80 per cent served two years or longer. Slightly over 20 per cent 
were married, a figure larger than that for any other group composed entirely 
of freshmen who entered in 19^6. About 65 per cent probably would have at- 
tended college without federal aid Just over 10 per cent of the veterans 
and about 15 per cent of the nonveterans indicated that their fathers were col- 
lege graduates. Slightly over 50 per cent of each of the two subgroups lived 
at home or with near relatives, and approximately 15 per cent of veterans were 
renting or owned their own housing. 

The res-alts of the analysis are shown in, Table 31. The means for the two 
subgroups are only slightly different on the ACEE and First-Tear College Aver- 
age Grade, the veterans being a trifle higher on both. Although the validity 
coefficients differ by .09, the errors of estimate, slopes, and intercepts of 
the regression lines do not differ significantly. The nonveterans are found 
to have an advantage which amounts to only .02 grade units, when allowance is 
made for* ACPE. scores; k$ per cent of the veterans excel the average nonveteran. 
In the College of Engineering at Midwest City the veteran and nonveteran stu- 
dents are unusually similar with .respect to college achievement in relation to 
aptitude 

Southern Technological University, Analysis of data for engineering students 
at Bouthe:m Tech was limited to a comparison of the relative achievement of 
male veteran and nonveteran students after allowing for differences In total 
score on the American Council Psychological Examination. In order to secure 
enough cases to warrant analysis, it was necessary to combine students who 
entered In the fall of 19^5 with those who entered in the fall of 



Southern Tech is the land-grant college for white students in a state 
which also supports a state university. Students are admitted upon completion 
of an appropriate high school course. Engineering students carry a freshman 
program which Includes chemistry, EnglJjdx, algebra, trigonometry, analytic 
geometry, drawing, descriptive geometry, and shop. 

The group studied may be defined as follows : students who entered as 
beginning freshmen either in the fall of 19^5 or the fall of 19^6; who com- 
pleted three quarters of engineering work (including eleven or more credits 
per quarter) during the academic year in which they entered; and who completed 
a questionnaire. Students who received ten or more credits in specific sub- 
jects for training received while in the armed services were not considered 
to be "beginning" freshmen and were excluded. J'urther data upon the group 
was obtained from selected questionnaire items, as follows : somewhat over 
60 per cent served two years or more in the armed services; approximately 15 
per cent were married. About 65 per cent of the veterans would probably have 
attended college without the GI Bill of Bights. Slightly more than 10 per 
cent of veterans and about 25 per cent of nonveterans reported that their 



150 



ADJUSTMENT TO COLL1CE 



(Table 31 

COMPARISON OF AVERAGE GRADES EAREED BT VETERAN AHD NODTVETERM MALE STUDENTS 
Midwest City University, College of Engineering, Ire shmen, 19^6 - 

I. Correlations, Means, and Standard Deviations? 







Correlation "with: 








Variable 


Sub- 
group 




Mean 


3D 


u 


APPJP 


First-Tear 








Avg. Grade 








ACPE Total 


m 




^5 


118 


19 


167 


(raw score) 


m 




.36 


U5 


20 


171 


First-Year College 


m 


.45 




3-3^ 


7* 


167 


Average Grade 


m 


.36 




3-32 


77 


171 



Validity coefficient for combined group: 



II. Analysis of Covariance Results: 



Hypothesis 


CM- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
ould arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


1.184 
0.646 
0.101 


1 
1 
1 


Betveen ,20 and .30 
Betveen *30 and .50 
Betveen ^0 and .80 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Nonveteran 

0.02 

0.03 

k9 

Not signif ioattt 



ACADEMIC ADJUSTMENT OF STUDENTS 151 



fathers had completed college. Almost 40 per cent of the veterans and some 
75 per cent of the nonveterans in this group were living in rooming or board- 
ing houses. In addition, nearly 20 per cent of veterans rented or owned an 
apartment or house . (Age of veterans is not reported for Southern Tech since 
the method of determining age at entrance is not adapted to groups whose 
members did not enter at a fixed time.) 

The following information may aid in interpreting the results presented 
in Table ^>2.i the criterion, First-Year College Average Grade, is based on the 
work of the three quarters of the freshman year. Grades are based on a five- 
step scale of 4 to The only predictor used was the total raw score on the 
American Council Psychological Examination. 

The results of the analysis indicate that the ACPE means are almost the 
same for the two subgroups, but the veterans achieve a higher freshman average 
grade than do the nonveterans. The test had reasonably high predictive value, 
the validity coefficients being .48 and Al for veterans and nonveterans 
respectively and .46 for both subgroups combined. The hypotheses of equal 
errors of estimate and equal slopes are not disproved, but the difference in 
intercepts is significant at the 2% level. The difference in mean grades, ad- 
justed for differences in ACEE score, is .25 (which is almost the same as the 
unadjusted difference since the two subgroups are nearly equal in ability). 
In standard error of estimate units, the difference is .42; 66 per cent of 
the veterans exceed the average nonveteran in freshman average grade. 

Si Ternary. In three of the four engineering schools studied, veterans were 
found to earn higher freshman average grades than nonveterans, when the influ- 
ence of ability differences is eliminated. The difference found at two of the 
universities would be expected to occur by chance less than once in a hundred 
times, and at the third institution less than twice in a hundred. At the re- 
maining college there was practically no difference between veteran and non- 
veteran students, although what difference occurred was in favor of the non- 
veterans. At one university, the veterans were also found to earn better 
grades than nonveterans of the same ability level in two specific courses, 
drawing and mathematics. 

The tendency was found at Midwest Tech for the more able students, as 
judged both by predictor variables and college grades, to take the heavier 
course loads. Nonveterans tended to take slightly heavier course loads than 
the veterans. 

High school standing has usually been found to predict grades better for 
nonveteran than veteran students, but at Midwest Tech the opposite conclusion 
was reached and at Middle State the difference in validity was small. 



152 



ADJUSTMENT TO COLLEGE 



Table 32 
COMPARISON OF AVERAGE GRADES EAMED BI VETERAI AID tiOENEXBRKB MALE STUDETOS 

Southern Technological T3niversity ; College of Engineering 
( Students who entered as freshmen In fall/ 19^5 / or fall, 

I. Correlations^ Means, and Standard Deviations: 



Variable 


group 


Correlation with: 


Mean 


SD 





ACPE 


First -Year 
Avg. Grade 


ACEE Total 
(raw score) 


MV 

m 




:5 


111 
109 


21 
21 


120 

50 


first -Tear College 
Average Grade 


m 
m 


!to 




2.50 
2.22 


.70 
.60 


320 
50 



Validity coefficient for combined group: 



.46 



II. Analysis of Covariance Eesnlts: 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
wuld arise by chance 


A. Equality of errors of estimate 


0.956 


1 


Beteen .30 and .50 


B. Equality of slopes 


0,885 


1 


Between .30 and .50 


C. Equality of intercepts 


6.110 


1 


Beteen .01 and .02 



III. Difference between Subgroups with Ability Held Constant; 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from HC above) 



Male Veteran 

0.25 

OA2 

66 

2)1 



ACADEMIC ADJUSTMENT OF STUDENTS 153 



Grades of Yeteran and Jtaiveteran Students in Agriculture 
and Business Colleges in State "Universities 

Midwest Technological IMiversity* For students in the College of Agriculture 
at Midwest Teen, a fairly extensive analysis was undertaken. Xntercoirela- 
tions of the following variables were studied* part and total scores on. the 
American Council Psychological Examination, score on an English placement test, 
High School Average, academic load carried for each quarter of the freshman 
college year, average grade for each quarter, and course grades in English. 
For the comparison of relative achievement "between male veterans and male non- 
veterans, allowance was made for differences in. total score on the ACPE and in 
Etgh School Average Grade 5 the criterion was First-Tear College Average Grade. 

Eesidents of the state may "be admitted to the College of Agriculture at 
Midwest Tech upon the completion of an appropriate high school course. English 
is the only course required for all agriculture students, although various 
other courses are required for certain large subgroups of these students. 
Even during the freshman year, a definite emphasis upon agricultural subjects 
is present, and students preparing for different specialties take somewhat dif- 
ferent programs* 

The group upon wnich the analysis was based may be defined as follows: 
students who entered as begi.Ttn.lTig freshmen in the fall of 19^6 and who com- 
pleted three quarters of work in the College of Agriculture (including ten or 
more hours of work each term) during the academic year 19^-6-19^7. Veterans 
who received ten or more credits in specific courses for training received 
during their service with the armed forces were excluded * The following addi- 
tional information about the group was taken from their questionnaire responses: 
Teterans In this group were relatively old, only about 20 per cent being under 
twenty at time of entrance, almost 60 per cent had two years or more of active 
duty; and 15 per cent were zoarried. About 70 per cent of the veterans, indicated 
that they would probably have attended college without government aid. Slightly 
over 10 per cent of the veterans and about 15 per cent of the nonveterans had 
fathers who were graduated from college. As with the Midwest Tech engineers, 
the housing arrangements for these students were quite varied . Somewhat over 
10 per cent of the veterans were living in houses or apartments which they 
owned or rented . 

It may be noted, in Table 33, that veterans tended to take slightly 
lighter course loads than the nonveterans, the difference amounting to about 
half a credit -hour dumber of credits correlated positively with all the other 
measures, both predictors and criteria, but no systematic difference is apparent 
between veterans and nonveterans in the magnitude of the correlations* As in 
the other analyses where information on this point was obtained, the better 
students tend to take the heavier course loads. 

Comparing the means of the predictors^ we find that the nonveterans are 
higher only on the ACPE Quantitative score and on High School Average. Without 
exception the veterans exceed the nonveterans on the criterion variables,. 



154 



ADJUSTMENT TO COLLEGE 



on 
on 



a s 



g 3 

fi 3 



201 

TOT 



-epos: 



^ 
C 



S 3 







bO 
0) 



s 



l 

r-i 
03 
O 

H 

to 

O 

1 



45 
CQ 



03 - 



-g - 



T-SJOV 



D-VO 

on on on on 



ir\vo 



OJ 

H 



H H r-i OJ 

H H r-i 




H 

OJ 



O H O H 

ro no en o 



O CU OJ 8 



on o cvico o> 
oj on coal 



LTXVO 




H 



O <MOD 



kOCDI^OrHOJONCA 




C? OJ 




Secon 
Third 



S 



tOM/N IT\ 

ir\ tr\co co 



o 

ON 



OJ OJ OJ rH H 



04-^too u-\ H 
ononn oj OJ 



b- LT\ O OOVO 
04 OJ H O 




6l'S 



99 



9* 



SI 
ffC 
ifC 



6'TE 



6*11 



0*3 
9*1 
T'T 



ACADEMIC ADJUSTMENT OF STUDENTS 155 



Somewhat different re stilts are obtained for veterans and nonveterans 
with regard to the validity of the predictors e For nonveterans, High School 
Average is unquestionably the best predictor of quarter average grades, while 
for veterans the English Test appears to be generally superior. For predict- 
ing grades in English, the English Test is again better for veterans than 
for nonveterans. 

The analysis of covariance results are shown in Table 3^. The ACIE and 
High School Average were selected as the predictors, and the criterion was 
First -Year College Average Grade. The multiple correlations proved to be 
quite high for both subgroups (.65 for veterans and .70 for nonveterans). 

The standard errors of estimate were significantly different at the 2$ 
level; grades of nonveterans tended to be more predictable. The equality of 
the slopes of the regression planes was not disproved, but the intercepts vere 
found to be significantly different at the 156 level. The veteran subgroup was 
superior, the difference in intercepts being .23 grade units. In error of esti- 
mate units, the difference was .5*4-, which means that 71 per cent of the veterans 
excelled the average nonveteran students. The test of the significance of the 
difference is ambiguous, in view of the fact that a difference in errors of 
estimate greater than that obtained would be expected by chance less than two 
times in one hundred. 

Midwest State University. Analysis of academic data for students in the Col- 
lege of Business of Midwest State University was limited to a comparison of 
achievement after allowing for differences in scores on the Ohio State Uni- 
versity Psychological Examination. 

Midwest State is a large state university located in a large city. Al- ' 
though it is coeducational, only male students were included in this analysis. 
Students who are residents of the state are admitted upon the completion of an 
appropriate high school course of study. Students of business are required to 
take courses in English, economics, business, and geography, and to meet 
certain broader requirements during their freshman year. 

The group to be studied is defined as follows: students who entered as 
beginning freshmen in the fall of 19^6; who completed three quarters in the 
College of Business, including at least ten hours of work in each quarter , 
and who returned a questionnaire. Students who were granted ten or more hours 
of credit in specific courses for training received during service were ex- 
cluded from the group. The questionnaires provided the following additional 
information: About 30 per cent of the veterans were under twenty years of 
age; almost 60 per cent of them served two or more years on active duty; and 
slightly more than 10 per cent were married. About 75 P e r cent of the veterans 
would probably have attended college without the GI Bill. About 10 per cent 
of the veterans and roughly 25 per cent of the nonveterans indicated that 
their fathers were college graduates. With respect to housing, about 35 P e ^ 
cent of the veterans lived at home or with near relations; ^0 per cent lived 
in boarding or rooming houses, and an additional 10 per cent lived in houses 
or apartments which they rented or owned. Slightly over 50 P e *" cent of the 
nonveterans lived at home or with relatives, and roughly 25 per cent lived in 
boarding or rooming houses. 



156 



ADJUSTMENT TO COLLiGi 



Table 3^ 

COMPARISON OF AVERAGE GKADES EAKKED BY VETERAH AH) NOlTVE'nERAN MAUE STUDENTS 
Midwest Technological University, College of Agriculture ? Freshmen, 1946-1947 

1. Correlations, Means, and Standard Deviations: 







Correlation with: 








Variable 


Sub- 
group 


ACPE 


E. S. 
Average 


Ist-Yr. 
Average 


Mean 


SD 


N 








Grade 


Grade 








1. ACPE (1945) Total 


MV 




.38 


.52 


11.9 


3-9 


140 


(converted score) 


m 




.49 


.48 


11.5 


3-5 


102 


2, High School 


m 


.38 




.56 


2.52 


.64 


l4o 


Average Grade 


m 


^9 




.68 


2.75 


59 


102 


3. First -Tear College 


MV 


.52 


.56 




2.30 


.60 


140 


Average Grade 


m 


.48 


.68 




2.15 


51 


102 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 



Sample 


Multiple E 


Male Veterans 


65 


Male Honveterans 


70 


Combined Group 


.64 



III. Analysis of Covariance Ke suits? 



Hypothesis 


Chi- 
square 


Degrees 
, of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


An Equality of errors of estimate 


5.425 


1 


Between .01 and .02 


B. Equality of slopes 


2.357 


2 


Between .30 and .50 


C. Equality of intercepts 


16.137 


1 


Less than .01 



IV. Difference between Subgroups with Ability Held Constant; 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIIC above) 



Male Veteran 

0.23 

0.54 
71 
Ambiguous 



ACADEMIC ADJUSTMENT OF STUDENTS 157 



The results for Midwest State "business students are presented in Table 
35. The Ohio State IMiversity Psychological Examination (OSPE) proved to be 
a good predictor of freshman grades for both veteran and nonveteran students ; 
the validity coefficients were .59 s&d .55. oo^e OSPE scores were converted 
to a standard scale having a mean of 13 and a standard deviation of k. Since 
high school standing was reported in very coarse groupings (upper, middle, and 
lower third of the high school class), it was not thought worth-while to in- 
clude it as a predictor o The criterion, which as usual was the First -Tear 
College -Average Grade, was based on a scale of five categories having the 
values of 4 to Oo 

The nonveterans were superior, on the average, with respect to OSPE score, 
but their freshman average was slightly lower than that for the veteran stu- 
dents. The errors of estimate and slopes of the regression lines did not 
prove to be significantly different, but the intercepts were significantly dif- 
ferent at the % level. The difference in intercepts amounted to .18 grade 
units. In error of estimate units, the difference was .33; 63 per cent of the 
veterans excelled the average nonveteran 

Littletown State University. For students in the College of Business at Little - 
town State, analysis was limited to a comparison of the achievement of male 
veteran and nonveteran students, taking into account differences in total 
scores on the American Council Psychological Examination* The criterion again 
is First-Year College Average Grade* 

Residents of the state were admitted to the College of Business upon com- 
pletion of an. appropriate course of study in high school. Those with poor 
higji school records were admitted on somewhat of a probationary basis. Stu- 
dents of business were required to take English and a survey course in business 
and to meet certain broader requirements during their freshman year. 

The group selected for study may be defined as follows, s zaale students 
who entered as beginning freshmen in the fall of 1<&6; who completed two semes- 
ters, including eleven or more houra of credit In each; and (since question- 
naire returns were excellent) who returned questionnaires* The following 
information, obtained from the questionnaires, includes additional descriptive 
materials About 25 per cent of the veterans were under twenty years old; just 
over 60 per cent had two or more years of active duty; and slightly over 10 
per cent were married. These figures closely resemble those for the students 
of business at Midwest State. Somewhat over 70 per cent of the veterans would 
probably have gone to college without a government scholarship. About 15 per 
cent of the veterans and roughly 35 per cent of the nonveterans reported that 
their fathers had been graduated from college Approximately k$ per cent of 
the veterans and 85 per cent of the npnveterans were living in college dormi- 
tories; fewer than 10 per cent of the veterans were living in an apartment or 
house which they rented or owned. 

The ACFE gives higher validity coefficients in the College of Business 
than in Arts and Science; the r's were .55 and .57 for veterans and nonveterans 
respectively (see Table 6)* The two subgroups were very similar with respect 
to mean ACPE scores, but the male veterans tended to earn slightly higher grades. 



158 



ADJUSTMENT TO COLLEGE 



Table 35 

COMPARISON OF AVERAGE GRADES EARNED BY "VETERAN AND NONVEOTAN MALE STUDENTS 
Midwest State University, College of Business^ Freshmen, 

I. Correlations, Means, and Standard Deviations: 



Variable 


group 


Correlation withs 


Mean 


SD 


I 


OSFE 


First -Year 
Avgo Grade 


OSFE 
(converted score) 


m 
m 




59 
-55 


12,0 
13,2 


3o2 

4,0 


232 
58 


First -Year College 
Average Grade 


m 
m 


-59 
55 




2o26 
2.22 


.65 

.70 


232 
58 



Validity coefficient for combined groups 



.57 



II, Analysis of Covariance Results; 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise bjr chance 


A, Equality of errors of estimate 
B. Equality of slopes 
C Equality of intercepts 


1.201 
1.189 
5-029 


1 
1 
1 


Between a 20 and .30 
Between -20 and -30 
Between -02 and ,05 



III. Difference between Subgroups with Ability Held Constants 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 

0.18 

0.33 

63 

5* 



ACADEMIC ADJUSTMENT OF STUDENTS 



159 



Table 36 

COMPARISON OF AVERAGE GRADES EARNED BY VE'IERAH AUD HOJSVE!EEEAH MATE STODEHTS 
Littletown State University, College of Business , Freshmen, 19^6-1947 

I. Correlations, Means, and Standard Deviations: 







Correlation -with: 










Sub- 










Variable 


groiip 


ACPE 


First -Year 


Mean 


SD 


IT 








Avg. Grade 








ACPE Total 


m 




*55 


106 


23 


ite 


(raw score) 


m 




57 


107 


21 


65 


First -Tear College 


M? 


55 




2.31* 


,6 5 


ite 


Arerage Grade 


m 


*57 




2.23 


.61 


65 



Validity coefficient for combined group: 



.56 



II. Analysis of Covar lance Eesults; 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a ralue 
greater than obtained 
would arise by chance 


A- Equality of errors of estimate 


0.3V7 


1 


Between .50 and .70 


B. Equality of slopes 


0,0191 


1 


Between .,80 and 90 


Co Equality of intercepts 


2^36 


1 


Between .10 and 20 



= Difference between Subgroups with Ability Held Constants 



Superior subgroup 

AdYantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of reterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 

Ool2 

0,23 

59 

Not significant 



160 ADJUSTMENT TO COLLEGE 



No significant differences were found, however, in the analysis of corarian.ee j 
the difference in intercepts was significant at only the 20$ level* The mean 
difference in grades of veterans and nonveterans of equal ability (as measured 
by the ACPE) was 12* Fifty-nine per cent of the Teterans exceeded the average 
nonveteran student in the College of Business at Littletowa State 

Summary* Male veterans proved to be superior to male nonveterans with respect 
to college achievement (in relation to ability) at the single agriculture col- 
lege and in both business colleges studied. The difference was significant at 
the 5$ level in one instance (Midwest State College of Business ) The differ- 
ence was not significant in the case of the business students at Littletowa 
State,, and was ambiguous for agriculture students at Midwest Techa 

The tendency for veterans to take lighter course loads than nonveterans 
was again found for agriculture students at Midwest Tech;, and again the 
tendency was noted for the more able students to take heavier loads within 
each subgroup. Eigh school standing was again found to yield better predic- 
tion of grades for male nonveterans than for veterans. 

Grades of Interrupted Teterans and Uninterrupted Nonveterans 

In the studies so far reported, the first-year average grade has been used 
as the primary criterion, and ability to do college work has been assessed on 
the basis of one or more tests of scholastic aptitude, tool skills, or achieve- 
ment in high school subjects, used wherever possible in combination with a 
measure of high school standing. In the studies of interrupted veterans, how- 
ever, the veteran is in a sense compared with himself as a nonveteran* The 
interrupted veteran, as defined here, is a student who attended college for a 
certain period of time (usually one year) as an ordinary civilian student, 
after which he entered military service j upon completion of his military ser- 
vice, he reentered the same institution and completed a defined amount of aca- 
demic work (usually one year) For such students it seemed that the best 
basis for evaluating the effect of the student's intervening experiences on 
college achievement was a comparison of his achievement as a civilian student 
(before war service) with his achievement as a veteran student* 

Logically, it might seem that the most* appropriate control group would 
be one made up of students -who followed the same pattern of college attend- 
ance as the interrupted veterans but whose academic careers were interrupted 
by something other than military service* Such a group, however, probably 
would not be representative of students generally. Furthermore, a sufficiently 
large group of such nonveterans could not be found in the colleges studied. 
Consequently, the interrupted veterans were compared only with nonveterans 
whose study was not interrupted* Care was taken, however, to base all com- 
parisons on parallel stages of each student's academic career; thus, if sopho- 
more grades were used as criteria and freshman grades as predictors for the 
veterans, grades for these same periods were used for the nonveterans. 



ACADEMIC ADJUSTMENT OF STUDENTS 161 



In planning tlie analysis, it was thought likely that grades earned during 
the term just "before tlie student entered the armed forces might not be typical, 
since the prospects of leaving college shortly might lead the student to slacken 
his efforts in academic work. It was also supposed that grades for the first 
term following the return to college might not be typical, partly because back- 
ground information and study skills might need refreshing before a student re- 
gained his original effectiveness, and partly because students returning to 
college might be temporarily fired ith an enthusiasm which would not persist. 
In three of the five groups studied, the first-term grade was used as the pre- 
dictor and the fourth-term grade as the criterion, leaving out of consideration 
the doubtful second and third terms* The analysis of covariance method was 
again, used, in order to compare the second-term sophomore grades of veterans 
and nonveterans, taking account of ability as indicated by grades earned in 
the first term of the freshman year* 

Adams University. A relatively elaborate study was made of the group contain- 
ing interrupted veterans at Mams I&iiversity in order to obtain a more complete 
picture of the relationships of a number of pertinent variables In this study, 
the typical definition of groups WHS made; veterans completed one year before 
war service and one year afterward; nonveterans entered in the fall of 19^5 
and completed two consecutive years of college work. Students not enrolled 
in the arts and science curriculum, and veterans given advanced standing for 
college training during their service were excluded 

From the questionnaire, the following descriptive material was obtained s 
About 35 per cent of the returning veterans almost as large a proportion as 
of entering freshman veterans at Adams were under twenty years of age on re- 
sumption of their college careers; more of them had served at least two years 
of active duty (slightly over 50 per cent as compared with about 45 per cent 
for the entering freshman group); and again only 3 per cent were married , 
Virtually all of them, as might be expected, would have returned to college 
without the aid given by the GI Bill. As with the previous Adams group , about 
60 per cent both of veterans and of nonveterans had fathers who were college 
graduates. The overwhelming majority, almost 90 per cent of veterans and 
about 95 per cent of nonveterans, were living in college dormitories, 

!Ihe intercorrelations obtained for the two subgroups are shown in Table 
37- 03ie statistics above and to the right of the diagonal were obtained for 
the nonveteran subgroup, and those below the diagonal for the interrupted 
veterans 

Ere hypothesis that First -Semester College Average Grades would furnish 
a better prediction of Fourth-Semester Grades than tests plus hi^h school 
standing was verified. For veterans, the Predicted Grade, which was the 
best of the ordinary predictors, correlated .51 with Fourth-Semester Average; 
the correlation of First-Semester Average with Fourth-Semester Average was 
59* For the nonveteran subgroups, however, the difference was much less 
marked j the analogous correlations were o$f and .66* 

It is of interest to note that the validity coefficients of the two 
Scholastic Aptitude tests against the fourth- term averages are rather similar 
for veterans and nonveterans, but the validities for Adjusted School Rank and 



162 



ADJUSTMENT TO COLLEGE 



Table 37 

IMERCOREEIATIONS OF SCHOLASTIC APTITUDE TEST SCORES, ADJUSTED SCHOOL RAMC, 
PREDICTED GRADE, AHD FIRST- AID SECQSD-IEAE COLLEGE A"V1RAGE GRADES 

Adams University, College of Arts and Science 

(Veterans ho completed one year before "war service and one year after war ser- 
vice! nonveterans ho entered in the fall of 1945 an3 completed two years ) 





Male Nonveterans (3)1=111) 


H 
O T$ 
O ctJ 


(D 

JL, fc 
0) 


(D 








CQ 


<3j <lj 


5 








-P 

** 
J> SJ CQ -H 

B EH 'o 

CQ CQ <J fi 


o * 

CQ CO 
P ftf 


O 

CO CO 
E - i 

oo -d" 


1 


ra 




SAT-Verbal 


X. .33 -26 ,72 


.60 .53 


,47 .46 


559 


107 


x-x 


SAT-Mathematics 


.28 \, ,24 ,55 


.42 .23 


.32 .22 


574 


93 


on 

& 


Adjusted School Bank 


.31 .29 NV n 62 


.49 .41 


.46 .48 


74.7 


8.2 


CQ 




Predicted Grade 


.67 .46 .73 \, 


.70 .63 


.64 .64 


7^.5 


7-3 


1st -Semester Average 


,60 ,34 ,43 ,63 




.72 .66 


75.6 


7.8 


<tf 

1 


2nd -Semester Average 


.54 .22 .49 -66 




.73 -71 


76.4 


7.4 


3rd -Semester Average 


.46 ,39 .37 -55 


.55 .52 


\.83 


76.9 


6.8 




S 


4th-Semester Average 


.48 ,22 .31 -51 


.59 .57 


.80 N. 


78.0 


6.7 


Mean 


547 576 74.7 72*8 


72.9 72.8 


77-6 78.6 






Standard Deviation 


101 78 7,5 6.6 


8.1 7.8 


6.0 6.3 







ACADEMIC ADJUSTMENT OF STUDENTS 163 



Predicted Grade are noticeably lover after the interruption of training than 
during the initial college year* This finding is consistent with the previous 
observation that high school standing ordinarily has less predictive value for 

veterans than for nonveterans, 

It is also worthy of note that the intercorrelations of term grades wliich 
cross the time of interruption (i.e., all except first -semester vs. second- 
semester and third-semester vs. fourth-semester) are considerably lower for 
veterans than for nonveterans This suggests that military service or other 
related experiences has affected the academic performance of the veteran stu- 
dents o 

The means for the veterans are slightly lower on most of the predictive 
measures., and their First- and Second -Semester Average Grades are lower than 
for nonveteransa The average grades for the third and fourth semesters, on 
the other hand,, are higher for veterans It will be noted that the mean Pre- 
dicted Grade for veterans was 72.8 and the mean First -Semester Average was 
almost the sames 72 9* The iionveterans exceeded the prediction slightly 
morej the predicted mean grade was 7^*5 and the mean of the First -Semester 
Average Grades was 75 6. In the second tern, the veterans' mean was prac- 
tically unchanged, while the nonveterans 1 mean grade increased slightly. 
There is then some slight evidence for believing that the anticipation of 
being drafted did lower the relative achievement of veterans to some extent. 

The analysis of covariance results, comparing the regressions of Fourth- 
Semester Grades on First -Semester Grades for interrupted veterans and ordinary 
nonveterans, are shown in Table 38. As has previously been observed, the 
veterans on the average earned lower grades than nonveterans in the first 
semester of the freshman year and higher grades in the second semester of 
the sophomore year. The hypotheses of equal errors of estimate and equal 
slopes are not disproved, but the intercepts of the regression lines are 
found to be significantly different at the 1% level. The difference, allow- 
ing for differences in ability as measured by First-Semester Grades, amounts 
to 1.92 on the hundred-point grading system used at Adams. In error of esti- 
mate units the difference is .38; 65 per cent of the interrupted veterans excel 
the average nonveteran in Fourth-Semester Grade when adjustment is made for 
ability differences as measured by First-Semester Average Grade- 

This result is particularly striking in view of the results previously 
reported for the freshman veteran and nonveteran students at Adams; for the 
freshmen, no significant difference was found, although a relatively small 
difference would have been significant because so many students were included 
in the analysis* In the freshman group, only 52 per cent of the veterans ex- 
celled the average nonveteran, 

Stewart University , At Stewart s the only variables taken into consideration 
were the average grades earned during the first four semesters of college 
work* As at Adams, the study involved a comparison of nonveteran, students 
with veterans who had completed two full semesters of college work in the 
arts and science college, who were interrupted for military service, and who 



164 



ADJUSTMENT TO COLLEGE 



Table 38 

COMPARISON OF AVERAGE FOURTH-SEMESTER GRADES EARKED BY IEEEKRI3PTED VETERAN 
AM) IJIOOTERKUFTO NOHVETEEM MALE STUDENTS 

Adams University^ College of Arts and Science 

(Veterans who completed one year "before war service, one year after war 
service; nonveterans who entered in the fall of 19^5 and completed two years.) 

I. Correlations, Means, and Standard Deviations : 





GfiK 


Correlation with: 








Variable 


group 


First-Sem, 


FourtJi-Sem. 


Mean 


SD 


K 






Avg. Grade 


Avg. Grade 








Fir s t ^Semester College 


MV" 




59 


72.9 


8.1 


13* 


Average Grade 


m 




M 


75-6 


7-8 


111 


Fourth-Semester College 


m 


.59 




78 6 


6.3 


13^ 


Average Grade 


m 


.66 




78.0 


6.7 


111 



Validity coefficient for combined groups 



.61 



II, Analysis of Covariance Results: 



Hypothesis 


square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


1,822 
8,298 


1 
1 
1 


Between .90 an< ^ *95 
Between .10 and .20 
Less than .01 



III. Difference between Subgroups with Ability Held Constants 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 

1.92 

0.38 

65 



ACADEMIC ADJUSTMENT OF STUDENTS 165 



subsequently completed the two semesters of the sophomore year following 
their return from military duty. Hie nonveterans chosen for comparison 
entered in the fall of 19^-5, aid completed two full years of college work* 
Transfer students and students given advanced standing for military training 
were eliminated. The predictor used was the First-Semester Average Grade, 
and the criterion was Fourth-Semester Average Grade* The grading system in- 
volved seven categories which were given the values of 7 to 1 for purposes 
of analysis. 

Additional descriptive material is available from the questionnaires; 
Be turning veterans at Stewart were older than the comparable group at Adams 9 
only about 20 per cent being under twenty years of age Likewise they had 
served longer tours of duty; nearly 80 per cent had had two years or more of 
active duty a About 5 per cent of veteran sophomores at Stewart were married. 
A larger proportion of fathers who had graduated from college was reported than 
for any other group in the study; 75 per cent of the veterans and 70 per cent 
of the nonveterans fell into this category. All of the nonveterans and most 
(just under 90 per cent) of the veterans were living in college dormitories; 
about 10 per cent of the veterans lived In apartments or houses which they 
rented or owned* 

The intercorrelationa, means, and standard deviations of the four semes - 
ter average grades for nonveterans and interrupted veterans are shown in Table 
39 33ae intercorrelations are uniformly lower for veterans than for nonveterans, 
but the differences are especially large for First~Semester vs. Third -Semester 
and Second-Semester vs* Third-Semester Average; the correlations are much lower 
for the interrupted veterans than for the nonveterans, presumably because of in- 
fluences related to the period of wax service. It is interesting to note, how- 
ever? that the correlations of First -and Second-Semester Averages with Fourth- 
Semester Average, which also bridge the period of interruption for veterans, 
do not show nearly as great a difference. 

The means of veterans 1 grades are lower for both first and second semes- 
ters, but higher for third and fourth semesters than the grades of nonveterans, 
Both subgroups show a decrease in mean grade from the first semester to the 
second, and the difference is nearly the same for both subgroups. Apparently, 
either the anticipation of induction did not result in a let-down for the 
veterans, or the let-down was equally influential during both terms of the 
freshman 



The magnitude of the intercorrelations between adjacent semester averages 
suggests that the grades are reasonably reliable. 

Turning to Table IfO, it wiH be seen that the correlations between the 
two semester grades employed as predictor and criterion were high: *6j for 
veterans and .68 for nonveterans* The nonveteran subgroup earned a higher 
mean First-Seiaester Average Grade, while the interrupted veteran subgroup was 
superior with respect to Fourth-Semester Average Grade. The analysis of co- 
variance shows that differences in errors of estimate and regression slopes 
are not significant, while the difference in intercepts of the regression 
lines is significant at the 1$ level* 



166 



ADJUSTMENT TO COLLICi 



Table 39 

INTERCOEKELATIONS OF FIBST- AM) SECOKD-IEAR COLLEGE AWAGE GRADES 
Stewart University, College of Arts and Science 

(Veterans who completed one year before war service and one year after 
wax service; nonveterans who entered in the fall of 19^5 an ^ completed 
two years, ) 





Male Nonveterans (N=70) 


0) CD 
bfl tt) 


1 SP 








o3 cd 


03 oj 
JH JH 
<U CD 








% % 


CD 1 
CO CO 


9 


o 




DQ 5 
H CXI 


on jt 


1 


CQ 




First -Semester College Average. 


X. -79 


.73 .68 


5.04 


99 


tf\ 
fd LT\ 

0) II 

p & 

S CQ 


Second-Semester College Average 


7^ N^ 


.77 .72 


4.91 


.85 


Third-Semester College Average 


.50 M 


N. .77 


4.94 


92 


Ii).tei 
Veterar 


Fourth-Semester College Average 


,67 .64 


70\^ 


5-13 


.82 


Mean 


4.88 4.77 


5.29 5-38 






Standard Deviation 


.86 ,95 


.71 .68 







ACADEMIC ADJUSTMENT OF STUDENTS 



167 



Table kO 

COMPARISON OF AVERAGE FOIffiTE-SEMESTER CJRADES EARKED BT 

VETERAN AM) UNIKTERRUPTEI) NONVETERM MALE STUDENTS 



Stewart University, College of Arts and Science 

(Veterans who completed one year before war service, one year after war 
service j nonveterans who entered in the fall of 19^5 and completed two years., ) 

I, Correlations, Means, and Standard Deviations^ 



Variable 


Sub- 
group 


Correlation with: 


Mean 


3D 


N 


First-Sem. 
Avg* Grade 


Fourtfc-Sem. 
Avg. Grade 


First -Semester College 
Average Grade 

Fourth-Semester College 
Average Grade 


M 
MN 

M7 

m 


.67 

.68 


.67 
.68 


^.88 
5.0!* 

5-38 
5-13 


.86 
.99 

.68 

.82 


55 
70 

55 

70 



Validity coefficient for combined group: 



.65 



II. Analysis of Covariance Results; 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


1,756 
11. 1C* 


1 
1 
1 


Between .10 and .20 
Between .80 and .90 
Less than .01 



III. Difference between Subgroups with Ability Held Constants 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from I 1C above) 



Male Veteran 

0.32 

0.57 

72 



168 ADJUSTMENT TO COLLEGE 



The difference in intercepts is ,32, about one^third of a grade unit* 
In standard error of estimate units,, the advantage of the veteran subgroup 
is ,57; 72 per cent of the veterans exceed the average nonveteran in Fourth- 
Semester Grades -when appropriate allowance is made for differences in First- 
Semester Grades c 

Eastern City University* At Eastern City,, the analysis was concerned only 
with First- and Fourth-Semester College Average Grades in arts and science* 
The groups studied were defined in the same manner as at Adams and Stewart a 
tfninterrupted nonveterans who entered in 19^-5 and who completed the sophomore 
year comprised the nonveteran subgroups the interrupted veterans completed the 
freshman year before military service and the sophomore year after military 
service. 

The questionnaires provide the following additional information; About 
25 per cent of the returning veteran sophomores were under twenty years of 
age; approximately 55 per cent served two or more years of active duty; and 
only 3 Per cent were married. In the latter two respects this group resembled 
that of veteran sophomores at Mams* With respect to father's education the 
sophomores at Eastern City diverged widely from the two previously discussed 
groups; only 1 per cent of veterans and about 5 per cent of nonveterans re- 
ported that their fathers had graduated from college. Living arrangements, 
too, differed; while the earlier discussed groups tended to live in college 
dormitories, nearly fti.T of the Eastern City sophomore students lived at home 
or with close relatives* 

For the analysis of covariance, First -Semester Average Grade was used 
as the predictor j Fourth-Semester Average Grade served as the criterion. The 
grading system involved five categories having values *2 to -2. Transfer stu- 
dents and students enrolled for fewer than twelve credits in any of the four 
semesters were eliminated. 

The relationship between First- and Fourth-Semester Grades was not as 
marked at Eastern City as at Adams and Stewart j the correlations were ,^9 
and .60 for the two subgroups, as shown in Table 4l A similar pattern of 
mean grades was found, however; the nonveterans were higher for First -Semester 
Grades and the veterans higher for Fourth-Semester Grades . The results of the 
analysis of covariance are also similar to that of the other two institutions* 
The hypotheses of equal standard errors and equal slopes are not disproved, 
while the difference in intercepts is found to be significant at the 1$ level - 
The difference amounts to 29 in grade units or *59 in standard error of esti- 
mate units o Seventy-two per cent of the interrupted veterans are found to ex- 
ceed the average nonveteran student; it will be recalled that the same figure 
was found at Stewart, 

KldwestJPechnological Ifaiversity, College of Engineering* At Midwest Tech, 
the nonveteran comparison group was composed of prewar students rather than 
a postwar group as in the three studies juat discussed. In this study, it 
was possible to increase the size of the interrupted group considerably by 
introducing some flexibility into the defined pattern of interruption, In 
the three studies just discussed, the interruption occurred between the 



ACADEMIC ADJUSTMENT OF STUDENTS 



169 



Table 4l 

COMPARISON OF AVERAGE FOUR1H~SEMES!EER GRADES EARNED BT 

"VETERAN ATO3 IMJUNTfclRHlJPTED HOITVETERAIT MALE STUDEFTS 

Eastern City University, College of Arts and Science 

(Veterans who completed one year "before war service, one year after wr 
service j nonveterans who entered in the fall of 19^5 and completed two years . ) 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


IT 


First-Sein* 
Avg, Grade 


Fourth-Bern* 
Avg* Grade 


First -Semester College 
Average Grade 


m 

Mff 




A9 
.60 


0.19 
0.29 


.68 
65 


70 
99 


Fourth-Semester College 
Average Grade 


MV 
Mff 


^ 9 
.60 




0-57 
0-33 


.58 

.60 


70 
99 



Validity coefficient for combined group: 



-53 



II. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees 
of 
* Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 


0.357 


1 


Between 50 and .70 


B. Equality of slopes 


1.37^ 


1 


Between .20 and .30 


C. Equality of intercepts 


13.779 


1 


Less than ,01 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 

0-.29 

0.59 

72 



ADJUSTMENT TO COLLEGE 



second and third semesters. At Midwest Tech a veteran whose college career 
was interrupted at any time after the second quarter and "before the seventh 
quarter was considered to fall -within the defined pattern, provided that he 
completed at least eight quarters. As in the other studies of interrupted 
veterans, students who were given advanced standing for college training re- 
ceived as members of the armed forces were excluded. The nonveterans entered 
.n the fall of 1939 and completed eight quarters of work; they were, of course, 
tot interrupted. Transfer students and students who changed divisions within 
he university during the eight quarters were excluded from the study. Stu- 
Lents who enrolled for fewer than 10 hours of work in any quarter or fewer than 
2 hours in the first, second, or eighth quarter were also excluded . 

A summer term counted as a regular quarter, provided the student was 
egistered for twelve or more credit hours. Students were also rejected if 
:hey had not taken freshman mathematics or if they lacked a record of high 
school standing. The same procedures described here apply to Midwest Tech 
"ollege of Agriculture students; the study of the agriculture students will 
)e described in the following section. 

An analysis of questionnaire data (which could only he obtained for 
veteran students, since the nonveteran group to which they were equated had 
jompleted its college study long before the questionnaire administration) 
revealed the following: This group was older than any previously discussed 
'the entire group was at least twenty years old upon returning to college 
ifter service); the aggregate tour of duty was longer than for any other 
jroup in the study (fully 95 per cent had served at least two years); and 
:hey included the highest proportion of married students (almost half the 
p^oup). About 20 per cent of these interrupted veterans reported that their 
'athers had graduated from college. As with other Midwest Tech groups, hous- 
.ng arrangements were varied; the largest number, nearly kO per cent, owned 
>r rented a house or apartment; the next largest group, 20 per cent, lived 
.n fraternity house s 

In the analysis of the data, First- and Second -Quarter College Average 
rrades were pooled and used as a single predictor; Eighth-Quarter Average 
trades were used as the criterion. The results of the analysis are shown in 
!able k2 . The validity coefficients tend to be somewhat lower than were found 
'or the studies involving interrupted veterans at Adams and Stewart, probably 
Because the criterion grades were earned at a time which is more remote from 
he predictor grades. The veterans were again found to earn lower grades 
.uring the initial two quarters (the predictor) than the nonveterans; on the 
lighth-Quarter Grades (the criterion) the relative position of the two groups 
,s reversed. The errors of estimate and slopes of the regression lines are 
.ot significantly different. The difference in intercepts is also found to 
e within the range of chance expectancy, the difference amounting to only 
11 grade units or .18 standard error of estimate units. Fifty-seven per 
ent of the interrupted veterans excel the average nonveteran. This engi- 
.eering group at Midwest Tech is the only group among the five involving 
nterrupted veterans where comparing college achievement in relation to 
bility did not yield a statistically significant difference in intercepts. 



ACADEMIC ADJUSTMENT OF STUDENTS 



171 



Table 42 

COMPARISON OF AVERAGE EIGHTH-QUARTER GRADES EARNED BT IMEEffiUFEEOD 
VEIERAN AID H2OMERRIJPTED NONVE'llEKAB STUDEM'S 

Midwest Technological University, College of Engineering 
(Teterans who completed two to six quarters before wax service and two to 

siz quarters after war service for a total of at least eight quarters; 
nonveterans who entered in the fall of 1939 asd completed eight quarters.) 

I. Correlations, Means , and Standard Deviations: 







Correlation with: 










^iih.. 










Variable 


group 


1 & 2 Quarter 


8th-Quarter 


Mean 


SD 


N 






Avg. Grade 


Avg. Grade 








First- and Second-Quarter 


MV 




57 


2.26 


.72 


140 


College Average Grade 


ME 






2.37 


.64 


215 


Eighth-Quarter College 


MV 


57 




2.43 


74 


140 


Average Grade 


MN 






2.38 


67 


215 



Validity coefficient for combined group: 



.50 



II. Analysis of Covariance Results: 



Hypothesis 


Chi- 
square 


Degrees\ 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


CL0731 
1.496 
2.640 


1 
1 
1 


Between 70 and .80 
Between ,20 and -30 
Between .10 and .20 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 
0.11 
0.18 

57 
Not significant 



172 ADJUSTMENT TO COLLEGE 



Midwest Technological University College of Agriciature> For Midwest Tech 
students of agriculture, the definition of the veteran and nonveteran groups 
and the plan of the analysis were in all respects parallel to those for the 
students of engineering described in the previous section. Questionnaire 
descriptive material lifceid.se showed the two groups to be very similar , the 
agriculture veteran sophomores differing from the corresponding engineering 
group in no respect by more than 5 percentage points . 

Results of the analysis for this group are shown in Table k3 For the 
interrupted veterans,, the correlation of the predictor with the criterion is 
the same as that found for the parallel group of engineering students (.57). 
For the nonveteran subgroup, however, the correlation is considerably higher 
for the agriculture students (062 as compared with ,45 for nonveteran engi- 
neering students ) * The differences in means show the same trend as did the 
preceding groupa involving inteiTupted veterans? the veterans earned lower 
grades on the predictor and higher grades on the criterion measure . 

The analysis of covariance showed that the difference in errors of esti- 
mate is significant at only the 10$ level;, and that the difference in slopes 
is even smaller than would be ordinarily expected to arise by chance. The ad- 
vantage of the interrupted veterans is found to be significant at the 1$ 
level o The difference in grades, when the effect of ability as measured by 
First- an} Second Quarter Grades is eliminated, amounts to .26 grade units, 
or half of the standard error of estimate Thus 69 per cent of the inter- 
rupted veterans excel the average nonveteran at the Midwest Tech College of 
Agriculture with respect to Eighth-Quarter Grades when differences appearing 
in grades for the first two quarters are taken into account. 



In all five comparisons of interrupted veterans with male nonveterans, 
the veteran subgroup was found to be superior to nonveterans in achievement 
relative to ability,, In these five comparisons, ability was measured in 
terms of freshman college grades In four of the five comparisons the dif- 
ference was significant at the 1$ level. The differences were generally 
greater in comparisons involving interrupted veterans than in the comparisons 
involving freshman students where ability was measured by tests and high 
school standing. 

In all five of the studies it was noted that the iaterrupted veterans 
had lower mean scores for freshman grade than the nonveteran male students . 
Thia fact mist be taken into account in interpreting the analyses of covari- 
ance results For example, if the lower freshman grades are the result of 
a tendency to let down on effort in anticipation of being drafted, the effects 
of the interruption would be primarily a matter of depressing effectiveness 
before service rather than enhancing effectiveness after service. 

In addition to the possibility that the freshman grades of the inter- 
rupted veterans were low because of a feeling that college work was unim- 
portant in view of the war emergency, particularly since their college career 
might he interrupted before the end of the college year, there are other 
reasonable interpretations. One possibility is that there might have been 
a tendency for faculties to give higher grades in 19^5, when the nonveterans 



ACADEMIC ADJUSTMENT OF STUDENTS 



173 



Table 43 

COMPARISON OF AVERAGE EIGE3E-QUARTER GRADES EARNED BY INTEBRIJPTED 
VETERAN AND UN1OTERRIJPIED NONVETERM MALE STUDENTS 

Midwest Technological Unl7er8ity ? College of Agriculture 
(Veterans who completed two to six quarters before war service and two to 

six quarters after war service for a total of at least eight quarters; 
nonveterans who entered in the fall of 1939 and completed eight quarters.) 

I. Correlations, Means , and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


N 


1 & 2 Quarter 
Avg. Grade 


8th-Quarter 
Avg. Grade 


Pirst- and Second-Quarter 
College Average Grade 


m 
m 




:S 


2.16 
2.30 


.63 
.60 


57 
106 


Eighth-Quarter College 
Average Grade 


m 

m 


57 
.62 




2.62 
zM 


.69 
59 


57 
106 



Validity coefficient for combined group: 



57 



II. Analysis of Covariance Results; 



Hypothesis 


Chi- 
square 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of Intercepts 


2-901 
0.0000 
9-255 


1 

1 
1 


Between .05 and .10 
Greater than *99 
Less than ,01 



III. Difference between Subgroups with Ability Held Constant: 



Superior subgroup 

Advantage expressed in grade units 

Advantage expressed in standard error of estimate units 

Per cent of veterans excelling the average nonveteran 

Level of significance of difference (from IIC above) 



Male Veteran 

0.26 

0.51 

69 



174 ADJUSTMENT TO COLLEGE 



ere freshmen, than In the period two or three years earlier when the veterans 
were freshmen. Another possibility is that selective factors of some sort 
were operating which tended to lower the means for the interrupted veteran 
subgroups who returned without advanced standing gained in the various college 
training programs sponsored by the armed services. This seems a likely possi- 
bility , since students who were known to have had college training while in 
the service were excluded from the study. 

No direct evidence is available bearing on the hypothesis of a let-down in 
anticipation of being drafted except the slight suggestion from the separate 
semester means at Adams that there might have been a let -down in the second 
terms the second- semester mean grade increased slightly over the first -semes- 
ter mean grade for nonveterans, while for veterans there was a slight decrease. 
A similar result was not found at Stewart , however. It would appear that if 
there was a let-down, it affected both terms of the freshman year to about the 
same extent. 

Evidence from Eastern City and Stewart University records shows that mean 
freshman average grade is rather stable from year to year, which casts doubt 
on the hypothesis of grades drifting upward In addition, it has been shown 
that the relationships of the mean grades are the same at Midwest Tech, where 
the nonveterans entered in 1939 > as at the other three institutions. The 
hypothesis of grades drifting upward over the year is not necessary to account 
for the differences at Adams. The percentile equivalents of semester averages 
were obtained, based on the entire class for that term; the percentile equiva- 
lents of the First- and Second -Semester Averages for the median interrupted 
veteran were about 35, while for the nonveterans the percentiles were about 55. 

The most probable explanation of the lower freshman grades of the inter- 
rupted veterans is that selective factors were operating. Students with the 
best academic records are somewhat more likely to have been rejected in our 
editing procedure because they were retained in ASTP or V-12 college training 
programs or because they were selected for other types of military training 
which resulted in their being given advanced standing. 

The Predictive Value of the American Council Psychological 
Examination and High School Standing for Veterans and Nonveterans 

It may be profitable to consider the assembled data on the predictive 
value of the measures most frequently used in this study for forecasting col- 
lege grades --total score on the American Council Psychological Examination 
and high school standing. Table kh is a summary of the validity coefficients 
obtained for these predictors. 

The validity coefficients for the ACPE showed considerable fluctuation 
from one subgroup to another, much of which is presumably ascribable to sampl- 
ing error, since some of the subgroups are rather small. The coefficients 
range from .28 to ,6l, but the median correlation for the twenty-four sub- 
groups of male students where ACPE total score was used is .^7. 



ACADEMIC ADJUSTMENT OF STUDENTS 



175 



Table 44 

CORRELATIONS OF AMERICAN COUNCIL PSYCHOLOGICAL EXAMINATION SCORES AHD 
MEASURES OF HIGH SCHOOL STANDING WITH FIRST-YEAR COLLEGE ATCRAGE GRADE 



Group 




Validity Coefficients 


Measure of 
High School 
Standing Used 


See 
Table 
No.: 


ACPE Total Score 


H. S. Standing 


University 


Curriculum 


m 


M 


FN 


m 


m 


m 


Adams 


Arts '46 











55 


.60 





Adjusted rank 


3 


Stewart 


" <46 








.. 


53 


.62 





Adjusted rank 


5 


Douglas 


" '46 


<>35 


-55 


51 


52 


.65 


.46 


Average grade 


6,7 


Harris 


11 '46 





__ 


.. 


-49 


.65 


.- 


Rank 


11 


Miller 


" '46 


.41 


-39 


.. 


.43 


.58 





Rank 


12 


Evans 


" '46 


51 


.45 





.. 





-.. 




13 


Western State 


" '46 


.46 


51 


.45 


53 


.59 


.62 


Average grade 


15,16 


Central State 


" '46 











.61 


.61 





Average grade 


20 


Central State 


" '45 


._ 








.33 


.65 





Average grade 


-_ 


Littletown State 


" '46 


.49 


o34 





._ 










22 


Midwest City 


" '46 


o57 


29 


.. 













24 


Midwest Tech. 


Engr '46 


.60 


061 





.62 


53 





Average grade 


26 


Middle State 


11 't6 


A2 


.28 


-- 


.51 


.53 


.. 


Rank 


27 


Midwest City 


11 '46 


A5 


o36 
















31 


Southern Tech. 


" # 


o48 


.41 
















32 


Midwest Tech. 


Agri '46 


52 


48 


. 


,56 


.68 





Average grade 


34 


Littletown State 


Bus '46 


55 


o57 





.. 


.. 


.. 




36 



The group at Southern Tech. was composed of 
the fall of 1945 or the fall of 1946. 



students who entered as freshmen in 



176 ADJUSTMENT TO COLLEGE 



Ihere is a slight tendency for the test to yield higher correlations 
with grades of male veterans than with those of male nonveterans; the median 
validity coefficients for veterans and for nonveterans respectively are about 
48 and A3. la eight of the twelve groups, the validity was higher for 
veterans -fefrgji for nonveterans, and in only four cases was the validity higher 
for nonveterans o It is apparent, of course, that an advantage for one type 
of group in eight out of twelve comparisons might readily arise by chance. 

The correlation of high school standing with first-year grades varies 
from o33 to *68 In the twenty-two male veteran and nonveteran subgroups where 
it was employed as a predictor. The median validity coefficient is 57o The 
tendency for high school standing to yield higher correlations for nonveterans 
than for veterans has previously been mentioned* 33he median coefficient for 
veterans is *53 and for nonveterans 6lo When the coefficients are compared 
in each of the eleven groups, the coefficient for the nonveteran subgroup is 
higher in almost every instance* So consistent an advantage (nine times out 
of eleven, with one tie) would be expected to arise by chance less than ten 
times in a hundred trials 

The universities which supplied information on Mgh school standing used 
various measures of relative performance* The three types of measure employed 
may be designated as rank in class, adjusted rank in class, and average grade. 
Of these, average grade suffers from the disadvantage that different secondary 
schools use Tn^ypV-r-ng systems which diverge markedly in form. Rank in class over- 
comes this difficulty, and is presumably preferable to average grade. The 
various secondary schools mey, however, differ greatly in the calibre of their 
students o In order to overcome this difficulty, adjusted rank in class pro- 
vides for a system of corrections to the ranks on the basis of past experi- 
ence with graduates of the various schools. 

The results obtained by the use of various indices of high school success 
are summarized in Table H, Direct comparisons of the validity coefficients 
for the different types of measures are seriously limited by the fact that 
some colleges use high, school standing in selection while others admit all 
graduates of approved high schools. Students who had poor high school records 
were not admitted in any of the five groups where adjusted rank or rank in 
class were used$ except for Douglas, such students -were admitted in the groups 
where high school average was usedo 

Without a detailed analysis of the degree of selection which has taken 
place in the various xiniversities, the relative merits of the different meas- 
ures cannot be decisively appraised from these findings. On the basis of the 
findings for the male nonveterans, however, it may properly be concluded that 
each of the universities is utilizing high school records in a form which is 
quite effective as a predictor of first-year college grades for its own stu- 
dents o 

In six of the groups, validity data are available for both the ACPE and 
high school standing. For male veterans, the median validity of the test is 
.H and of high school standing about .52 for these six groups, while the 
analogous medians for male nonveterans are about .50 and about .58. In none 



ACADEMIC ADJUSTMENT OF STUDENTS 177 



of the six possible comparisons for veterans does the test have a higher 
validity coefficient than high school standing, while for nonveterans the 
test is superior for only one group,, The results thus show that a measure 
based on high school grades tends to be a better predictor of college success 
than the single test of scholastic aptitude, the American Council Psychological 
Examination. This result is in line with the findings of previous studies in 
which a single test was compared with a. suitable index of high school achieve- 
ment* It should be understood, of course, that these two predictors should be 
considered as joint members of a predictive team, rather than as competitors, 
in practical problems of prediction*, 

trends in the College Grades of Veterans and Nonveterans 

As has previously been mentioned, it is possible that veteran students 
are initially handicapped because of such factors as forgetting and deteriora- 
tion of study skills, which result in temporary underachievement* Another 
possibility is that the veterans return to college fired with enthusiasm and 
good intentions which leads to temporary overachievement. It is also, of 
course, possible that both these effects occur, in the same student or in 
different students , and that they tend to cancel each other. It was accord- 
ingly thought worth-while to examine the trends in grades of veteran and non- 
veteran students. 

Average grades by terms were obtained for six groups, including the groups 
at Adams and Stewart which involved interrupted veterans. The average grades 
earned in successive terms by veteran and nonveteran male students are shown 
in Figures 12 and 13* It will be noted that the entire scale of grades has 
been shown on the graphs for all universities except Adams. This fact should 
be kept in rni^ in interpreting the graphs, since the variation among average 
grades for large groups of students is not likely to be great relative to the 
complete scale of grades in use 

The results for interrupted veterans at Stewart and Adams, shown graphi- 
cally in Figure 12, have already been described in the preceding discussions 
of the academic data. At Adams, the veterans do not show the slight increase 
in second- semester average grade which is present for the nonveterans, which 
suggests there may have been some let-down in effort in anticipation of being 
drafted- At Stewart, however, no such trend is found. Apparently any reduc- 
tion in effort resulting from the imminence of withdrawing from college for 
military service had about the same effect on both semesters. At Adams, the 
line showing average grades for the semesters following return from war ser- 
vice is almost exactly parallel to the line for nonveterans. At Stewart there 
is some suggestion that the veterans did relatively better in the third term, 
than in the fourth. The absence of any marked trend in the second -year mean 
grades of interrupted veterans suggests that forgetting or deterioration of 
study skills was not an important determiner of postwar achievement If such 
deterioration did occur, it was apparently more than counterbalanced by influ- 
ences favorable to success. 



i ivmnt i iw 



0) If 



I 



90 



80 



I 



60 




Before var 
service 



Veterans 
(N = 55) 

Nonveterans 
(N = 70) 



j_ 



2 

Semester 
Stewart University 




Before war 
service 



Veterans 
(IT = 134) 

Nonveterans 
(N = 111) 



I 



2 

Semester 
Adams University 



FIGURE 12. TKEKDS IN GEADES OF IMIEREUPTEB VEO^RAITS 
UBINTEBEUPTED NONVETEIRANS . 



ACADEMIC ADJUSTMENT OF STUDENTS 



179 



90 



80 



50 



Veterans 

(N = 531) 

Nonveterans 

(If = 6910 



Semester 
Adams (Arts) - 



I 



, Veterans 
(N = 433) 
. Nonveterans 
(N = 222) 



_L 



1 2 
Quarter 
Western State (Arts) - 1946 



3 - 



1 



Veterans 
(N = 271) 
Nonveterans 
(N = 028) 



123 

Quarter 
Midwest Tech. (Engr.) - 1946 



SL 2 



I 



Veterans 

(N = 140) 

Nonveterans 

(N - 102) 



. 

Quarter 
Midwest Tech. (Agri. ) - 



FIGURE 13- OKEKDS IN GRADES OF VETERAN AND NOHVETERAN MALE STUDENTS. 



180 ADJUSTMENT TO COLLECi 



Mean term grades were computed for groups of veteran and nonveteran 
males who entered as freshmen in the fall of 19^6 at three institutions 
Adams, Western State, and Midwest Tech; at the latter institution, such data 
were available for both engineering and agriculture students. These means 
are shown graphically in Figure 13. The lines for veterans and nonveterans 
are nearly parallel in all four cases. At Midwest Tech there is a slight 
tendency for the difference between veterans and nonveterans to "be greatest 
in the first of the three terms. There is again no consistent evidence that 
veterans are more handicapped at the start because of such factors as for- 
getting; if anything, the tendency is for veterans to do relatively better 
in the first term after war service than in subsequent terms. 

Summary and Discussion 

In comparing veteran and nonveteran students with respect to academic 
success, the twenty studies in which male veterans who entered college after 
their war service were compared with their nonveteran classmates provided the 
basic information. 

In these twenty comparisons, it was found that nonveterans were superior 
to veterans of the same ability in only four instances; in none of the four 
was the difference greater than might reasonably be expected to arise by 
chance. In three of the comparisons in which the veterans excelled, the ad- 
vantage of the veterans was so great as to be significant at the 1% level. 
In two additional instances, the difference was significant at the 2$ level, 
and in one case the difference was significant at the 5$ level. In a total 
of siz of the twenty groups where veterans and nonveterans were compared, 
then, the difference in freshman grades, when the effect of ability is elimi- 
nated, favored the veterans to an extent which would be expected by chance 
fewer than five times in one hundred. In one other comparison, the veterans 
exceeded the nonveterans by a considerable amount; the significance of the 
finding cannot properly be evaluated, however, because the ability measures 
were related to freshman average grades differently for veterans and non- 
veterans. Among the twenty comparisons based on entering freshman students, 
we would expect, by chance, that the veterans would excel in ten. Actually, 
the veterans excelled in sixteen of the twenty comparisons. Such a result 
would be expected to arise by chance less than five times in one hundred. 

On the whole, it may therefore be concluded that veteran students of 
the kind included in these studies tend to earn higher grades, in relation 
to their ability, than do nonveteran students. The actual magnitude of the 
difference is small, however. In terms of a grading system based on the 
letters A, B, C, D, and F it would amount to only a quarter or a third of a 
letter grade even in an institution where the difference was highly signifi- 
cant. In the most extreme case, the average difference would be equivalent 
to about the difference between a Cf and a C, 

Another way of representing the amount of difference is in terms of the 
overlapping between frequency distributions of grades for veterans and non- 
veterans of equivalent ability. If the two distributions are exactly alike, 



ACADEMIC ADJUSTMENT OF STUDENTS 181 



50 per cent of the veteran subgroup would excel the average nonveteran stu- 
dent. In the four cases where veterans were inferior to nonveterans of equal 
ability, the per cents of veterans excelling the average nonveteran were 39, 
48, kg, and 50 (rounded) . In the remaining sixteen groups, the per cents 
ranged from 52 to 76 < En the median group about 56 per cent of the veterans 
excelled the average nonveteran student. 

At one engineering college, course grades in drawing and mathematics were 
used as criteria as well as freshman average grade, Teterans were superior on 
all three criteria. 

An important supplement to the twenty basic comparisons is a series of 
five studies in which veterans whose college careers were interrupted by war 
service were compared with nonveterans whose careers were not interrupted. In 
these groups, the veterans had completed at least two quarters of academic work 
as ordinary civilian students before war service, and after military service 
they returned to college and completed at least two more quarters of college 
work* The nonveterans were students who entered at a designated time and who 
completed the same total number of tenoB as the vetterans. For these groups 
the problem was to determine whether sophomore grades (or Junior grades for 
two of the groups) were higher for veterans or for nonveterans of equal ability, 
and the measure of ability which was used was freshman grades earned In college* 
In general, grades earned in the terms which for veterans just preceded and 
immediately followed war service were not used in the computations. In all 
five of these comparisons, veterans .were found to be superior to nonveterans 
who earned equivalent freshman grades* 0!he difference was significant at the 
1% level in four of the five cases She per cents of interrupted veterans ex- 
celling the average nonveteran ranged from 57 to 72 with a median of 69 per 
cent. 

ae interpretation of -the differences found between veterans and non- 
veterans in these five groups is particularly difficult. At first glance, 
it would appear that the veterans had gained greater maturity and a capacity 
for more intensive and prolonged effort during their absence from the campus; 
accordingly, they showed a marked gain over their initial performance in 
contrast to the relative stability of performance shown by a nonveteran group 
whose careers were not interrupted. Such an interpretation would overlook two 
ia$K)rtant complications : first, the possibility that shifts in grading stand- 
ards over a period of time might influence the results 5 and second., the fact 
that in all five groups the veterans earned, lower freshman grades than did 
their nonveteran controls. OSie hypothesis that grading standards shifted is 
weakened by the fact that in three of the groups the veterans were compared 
with a postwar nonveteran group (students entering in 19^5) while in the 
other two the nonveteran comparison group was a prewar group (students 
entering in 1939)* The advantage of the nonveterans in their freshman grades 
constitutes a more serious problem- Among the hypotheses which may be advanced 
to account for this finding, two appear to be most plausible: first, the 
members of the veteran group, realizing that their academic careers were 
likely to be interrupted by war service, may have slackened their efforts; 
and second, the necessary exclusion from the comparisons of veterans who con- 
tinued their academic careers in V12 or ASOJP may have removed a dispropor- 



182 ADJUSTMENT TO COLLEGE 



tionate number of superior students, and thus lowered the average IB the re- 
maining group. With these limitations in -mind., it must be recognized that 
the findings are not as conclusive as the significance tests might indicate. 
The findings, in spite of the limitations, may properly be considered to 
favor the veteran group- 
In addition to the comparisons of male veterans and male nonveterans , 
study as made of a number of problems which were considered pertinent to the 
interpretation of the findings or which were by-products of the plan of analy- 
sis followed in making the basic comparisons . These problems were studied in 
varying numbers of groups; no attempt was made, however, to exhaust the pos- 
sibilities of the available raw data in seeking answers to them* 

One question, studied in three groups, was the possible significance of 
academic load, as zneasured by the number of credit hours carried. In all 
three instances, it was found that veterans tended to take a slightly lighter 
load. This might merely be a reflection of the fact that veterans were not 
required to take physical education or military science. It was also found 
that correlations between work load and measures of ability and college achieve- 
ment are consistently positive students who take the heavier load tend to get 
the higher grades in college. This tendency was consistently found both for 
the male veteran subgroups and the male nonveteran subgroups. This finding 
tends to discount the hypothesis that veterans did better because of a reduced 
course load, although it does not provide an adequate basis for rejecting it. 

Some interest attaches to comparisons of female students with each of 
the two male groups The procedure used in comparing male veterans with male 
nonveterans was applied, in two universities, to these comparisons. At one 
university, the women students were found to be slightly superior in grades, 
relative to ability, than both the male veterans and male nonveterans; but 
the obtained difference was no 1 greater than might reasonably be expected by 
chance. At the other institution, no comparison of grades in relation to 
ability for women students and male veterans could legitimately be made be- 
cause the two subgroups differed significantly with regard to the standard 
error of estimate . Grades could be predicted from ability measures more accu- 
rately for the female subgroup. In this institution, the corresponding dif- 
ference between women students and male nonveterans was practically zero. 

It is important to study the relationship between the time when apti- 
tude tests were taken by veterans and the scores which they earned. In two 
institutions where the College Entrance Examination Board Scholastic Apti- 
tude Test was used, some veterans had applied for admission and were tested 
before leaving for war service and some were tested after discharge* This 
variable, date of testing, was correlated with all the measures of ability 
and college achievement. The correlations tended to be quite low, indicat- 
ing that date of testing, within the range encountered in these studies, is 
of little importance. This finding is reassuring in the interpretation of 
the results, since in most of the groups the veterans were tested after war 
service, several years after high school graduation, while the nonveterans 
were tested a few months after graduation from high school ;o 



ACADEMIC ADJUSTMENT OF STUDENTS 183 



As an essential part of the procedure followed in studying veteran -non- 
veteran differences, considerable attention was given to various predictors 
of academic success. Validity coefficients based on the total American 
Council Psychological Examination score were available for twelve groups. 
Another commonly used predictor of freshman grades was a measure of high 
school standing; such a measure was used as a separate predictor in eleven 
cases. In six groups both measures were used. The median validity coeffi- 
cient found for the ACHE total score was .^7; the correlation coefficients 
tended to be slightly greater for male veterans than for male nonveterans. 
The median validity of high school standing was .57> and this variable tended 
to yield higher correlations for nonveterans than for veterans. The median 
validity coefficient for veteran subgroups was .53 a ^ for male nonveterans 
,6l. A comparison limited to the six groups where both ACHE score and high 
school standing were used confirms the superiority of high school standing 
as a predictor of freshman grades. 

It is pertinent to inquire whether any difference in outcome of analyses 
of covariance is apparent in the groups for which both high school standing 
and various test scores were used as compared with those groups for which 
only test scores were used. As it happened, there were ten groups of each 
type among the twenty groups of entering freshmen. The general outcome of 
this comparison is shown in Figure 14. The median value of the per cent of 
veterans excelling the average nonveteran is 56 in both sets. However, the 
range in per cent of veterans excelling the average nonveteran is 48 to j6 
in the groups where high school standing was used, and the range is 39 to 66 
in the groups where tests alone were used. This suggests that the inclusion 
of high school standing as a predictor is associated with findings which are 
slightly more favorable to the veteran group. 

Finally, attention was given to possible trends in grades which veteran 
students earned in successive terms after war service. Examination of the 
mean grades of veterans and nonveterans shows no evidence that veterans were 
seriously handicapped at the beginning of the college work because of for- 
getting or deterioration of study skills nor that there was a period of 
initial enthusiasm which led to marked temporary overachievement * In four 
comparisons of this type, no clear indication of a consistent shift in dif- 
ferences of grades was found; if forgetting does play a role in grades earned 
by veterans immediately after their entrance to college, it is apparently 
counterbalanced by other factors such as a more enthusiastic approach to the 
study situation. 



184 



ADJUSTMENT TO COLLEGE 



Per Cent of Veterans 
Excelling the 
Average Nonveteran 



Predictors Used 



High School Standing 
in Combination with 
a Test or Tests 



A Test or Combination 
of Tests vith6ut 
High School Standing 



78 or higher 
73 to 77 
68 to 72 
63 to 67 
58 to 62 
53 to 57 



OQOX 



@00 

o 



48 to 52 



00 



ooo 



k3 to 47 
38 to 42 
33 to 37 
28 to 32 
23 to 27 
22 or lover 



Difference significant at or beyond 5$ level 
o Difference not significant 
K Outcome of analysis ambiguous 



FIGURE It. DISTRIBUTION OF PER CENT OF VETERANS EXCEEDING THE AVERAGE 
NONVETERAN WITH EESPECT TO ADJUSTED AVERAGE GRADE FOR THE TEN GROUPS IN 
WHICH HIGH SCHOOL STANDING WAS USED IN COMBINATION WITH A TEST OR TESTS 
AND FOE THE TEN GROUPS IN WHICH TESTS ONLT WERE USED* 



185 



Chapter IV 
SOME CHARACTERISTICS OF VETERAN STUDENTS 



This and the remaining chapters "will be concerned "With the results of 
the analysis of the items contained in the Student Opinion Questionnaire. 
Among these items is a set intended for veteran students only; they deal 
with certain background information on military service and its effect on 
ability to do college work. Certain of these items which were answered 
only by the veteran students will be discussed in the present chapter. 



Interpreting the Results of the Questionnaire Analysis 



Before discussing the results dealing with characteristics of veteran 
students, certain general questions regarding the interpretation of the 
questionnaire findings may well be considered. These comments regarding 
interpretation apply, of course, not only to the findings of the present 
chapter but to the following chapters as well. 

A relatively detailed account pf the plan followed in presenting the 
results of the questionnaire analysis is provided in the last section of 
Chapter II; the present discussion will be concerned with general con- 
siderations which may affect the interpretation of the findings to be pre- 
sented. 

As was observed in Chapter II, the primary emphasis in the discussion 
which follows will be upon the summarized results for twelve basic college 
groups. In interpreting the results which follow it is relevant to note 
that: (l) each group contained at least 75 veterans and 75 nonveterans 
who completed questionnaires; (2) each of twelve colleges and universities 
is represented once (and only once) in the group; and (5) each group is 
composed of students who entered college as beginning freshmen in the fall 
of 19^6. There are nine groups which may be considered as liberal arts 
groups; the other three are engineering groups. Half of the groups repre- 
sent privately-supported universities; the other half are state and municipal 
universities. Broad geographical regions are represented as follows: East, 
three; Middle West, seven; South, one; and Ear West, one. On the basis 
of the classification used by President Walters in his enrollment surveys 
reported in School and Society, nine belong to the category of "university 
and large institutions of complex organization," two (Evans and Harris) 
are "colleges of arts and sciences," and one (Midwest Tech) is an "inde- 
pendent technical institution." All but three had chapters of Phi Beta 
Kappa; of these three, two had chapters of Sigma Xi. Seven of the twelve 
are located in cities of more than 100,000 population; the remaining five 
are located in small cities or college towns. All but Adams and Stewart 
are coeducational. 



186 ADJUSTMENT TO COLLEGE 



Perhaps the first question which arises in interpretation is whether 
or not substantial bias has arisen from incomplete returns on the question- 
naires. Among the twelve groups which receive most of the attention in the 
discussion which follows, adequate evidence was available to indicate that 
80 per cent or more of the students "belonging to the defined group had com- 
pleted questionnaires in eight of the groups in several of these , the re- 
turns were close to 100 per cent. In four groups considered too important 
from the viewpoint of securing adequate diversity to omit from the basic 
twelve groups, somewhat less confidence can be placed in the results. In 
one group, Western State, the per cent returns for veterans was 63; for 
nonveterans, it was 66. Examination of the Adjusted Average Grades of the 
students at Western State who returned questionnaires suggested that they 
were overachievers to a slight degree. (They were less than one tenth of 
a standard deviation above the total group.) In three other basic groups, 
the procedure for collecting data did not permit an adequate estimate of the 
per cent returns. These groups were Central State, Evans, and Miller. It 
is accordingly necessary to interpret the results for these groups with 
caution* The inclusion of these four groups in the basic twelve was con- 
sidered justifiable on the assumption that the basic tendencies in these 
four groups would contribute substantially to the over-all picture, while 
minor biasses in the detailed results for. these groups would have little 
effect on the general results. [Examination of the questionnaire findings 
for these groups indicated that the likelihood of a serious bias in the 
returns of any of the four groups was negligible. 

It will be noted that both positive and negative results are treated 
fully in the ensuing pages. No apologies are made or are needed for this 
procedure; in some respects it is as important to know that veterans and 
nonveterans are alike in a certain way as to know they are different, or to 
know that a particular item is unrelated to AAG as to know that it is related. 
An objective of this study was to determine whether or not veteran-nonreteran 
differences could be accounted for by differences in background, attitudes, 
and experiences as identified through questionnaire responses- It was not 
expected that all the hypotheses tested would result in positive findings ; 
negative or suggestive findings may prove to be just as useful to future 
investigators of noneognitive factors in relation to college achievement. 

Comments regarding interpretation of the questionnaire results will 
ordinarily be limited to the findings for the twelve basic groups which are 
shown in the figures. If results for other groups depart considerably from 
those of the twelve groups, this fact will be noted in the discussion of the 
item. 

The results for those questionnaire items dealing with the background 
and attitudes of veteran students will be described in the balance of this 
chapter. These items differ from most questionnaire items in that they were 
answered by veteran students only; hence no comparison with nonveteran stu- 
dents can be made, and the graphs contain points pertaining only to veteran 
subgroups . 

Aspects of Service Experience 

In this section, the results of the analysis of various items dealing 
with some of the more objective and factual aspects of service experience 



SOME CHARACTERISTICS OF VETERAN STUDENTS 187 



wl 11 "be consi4ere<if Factors in service experience such as "branch of ser- 
vice, length of service^ overseas service, and rank ill be dealt with in 
order to find "what characterized the veteran students in the various col- 
lege groups and to -what extent the attribute measured by each item was 
associated with Adjusted Average Grade. 

Branch of Service , Questionnaire Item 8 (a) asks, "In which of the follow- 
ing did you serve?" For purposes of analysis the veterans were divided 
into four groups i (A) Army, (B) .Navy, (C) Marine Corps, and (D) Coast 
Guard, Merchant Marine, and Field Services. Each of the three groups com- 
bined in D was quite small. The results of this item will indicate "whether 
a classification of veterans with regard to branch of service is related to 
the achievement relative to ability of veteran students. The general results 
are shown graphically in Figure 15; the detailed findings are shown in Appen- 
dix Table 8(a), 

It was found that the great bulk of the veterans 
in this study had served in the Army or the Navy, but 
in general a greater proportion of the veterans who 
entered college as freshmen in l$&-6 came from the Na-vy 
than from the Army. The reverse is true for the inter- 
rupted veterans* There was no clear tendency for 
veterans who had served in the Army, or any other 
branch of the service, to earn higher grades in rela- 
tion to ability than students from other branches of 
the service. 

The proportion of veterans who served in the Army varied from about one- 
fifth to a little over half in the twelve groups, and the proportion serving 
n the Navy varied from about one-third to two-thirds. In the median group, 
a little less than lj-0 per cent of the veterans served in the Army and about 
55 per cent served in the Navy Comparatively few of the veterans had served 
in the Marine Corps and only a scattering in the Coast Guard, Merchant l^arine 
and Field Services , 

TSie generally larger proportion of veterans in the basic twelve groups 
who had served in "the Navy is interesting in view of the fact that the Army 
was decidedly larger than the Navy during the recent war. As of June 30, 
19^5, (according to Statistical Abstract of the United States, 19^-8) the 
contained more than twice as many men as the Navy* Since the rate of decrees 
from 19k5 to 19l|-6 was somewhat greater for the Army than for the Navy, almosl 
three times as many veterans came from the Army than from the Navy in the 
period from June 30, 1945, to June 30, 19^-6. It is apparent that the twelve 
groups in this study are not typical of the general veteran population with 
regard to branch of service* Appendix Table 8 (a) reveals that the inter- 
rupted veterans (who were older) came more often from the Army than the 
Navy. It appears that the Havy contributed a much larger share of the 
veterans who entered college in 19^6 than did the Army, relative to the 
number of veterans from each of these branches of the service The reason 
is not clear, but it may be related to different recruitment policies in 
the Army and Navy. 



188 



ADJUSTMENT TO COLLEGE 



100 



tJ 

I 



1 

o 



50 



25 



Median 



Army 



Navy 



Other 



Marixie 
Corps 

FEE CENT BEPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



120 



Army Wavy Marine Other Total 

Corps ( group 

MEM AAG OF MEDIA COLLEGE GROUP FOE SPECIFIED CATEGORIES 



FIGURE 15 o BRANCH OF SERVICES I2EM 8 (a) 



SOME CHARACTERISTICS OF VETERAN STUDENTS 189 



The median group with. Army service earned a mean AAG of about 135, 
while for the Havy and the Marine Corps the corresponding figures are 
slightly lower. (Hie small proportion of students in the other services 
showed a median value slightly higher than that for the Army,, ) Differences 
in mean AAG favored the Army over the Havy in no more of the groups than 
would be expected by chance. The F-test showed one category mean ithch 
was significant at the 1% levels veterans at Evans who had served in the 
Army earned AAG's which were significantly higher than those of veterans 
from other branches of the service It is to be expected, however, "when a 
large number of significance tests are made, that by chance some of the 
comparisons will be found to be significant. Since sixty-four tests were 
made, it might reasonably be expected that one of the comparisons would be 
significant at the 1$ level, even if no relationship existed. 

Length of Service. Item 8(b) asks, "How many months were you in service 
(on active duty, whether in training or in duty assignments)?" This ques- 
tion was included in order to test the hypothesis that those veteran stu- 
dents who served the longest, and therefore had the greatest opportunity to 
be affected by service experiences, would show a greater tendency to over- 
achieve than those veterans whose length of service was brief. For analysis, 
the veterans were classified as follows: (A) less than one year, (B) one to 
two years, (C) two to three years, and (B) three years or more of service. 
The results are shown in Figure l6 

Most veterans in the study were in service for 
from one to three years . There is a very slight and 
generally insignificant tendency for veterans who 
served longest to over achieve more than veterans whose 
length of service was brief , 

Categories B and C obtained the most frequent responses > indicating 
that the length of service was typically between one and three years. Among 
the twelve basic groups, three years or more of service was reported most 
frequently by the groups of engineering students at Midwest City. Groups 
which included students who entered in 1<&5 (Central State and Taylor) and 
the interrupted veterans at Midwest Tech were also high on this category. 
Fewer than 10 per cent of the veterans entering in 19^6 at Stewart, Adams, 
and Douglas had served three years or more. 

For students having three years or more of service, the median value 
of the twelve mean AAG's is slightly higher than for students having had 
less military service i otherwise, no trend is discernable. In only one of 
the basic groups, Western State, is a difference found which is significant 
at the Ijt level; here the "three-year" veterans earned a mean AAG of about 
150, which is significantly greater than the mean AAG for veterans choosing 
other categories o In nine of the twelve groups the mean AAG for veterans 
who served three years or more is greater than the mean AAG for all veterans; 
however, this finding cannot be regarded as statistically significant. 



190 



ADJUSTMENT TO COLLEGE 



GQ 

1 



I 






s 



100 










50 

25 





* 








k 


m 

,p 





L 









Ir 


Less than 
12 months 


12 orp to 
24 mcoiths 


2k up to 36 months 
36 months or more 



Median 



PER CENT KEPCRTING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 



130 



3 120 



110 















{ 






^ 

^^ 










































Less 
12 m 


than 12 u 
onths 2k m 


p to 2^ u 
onths 36 M 


p to 36 H 

tcaiths or 


lonths Toi 
more grc 


,al 
>up 



MEAN AAG OF MB2>IAN COLLEXxEJ GROUP FOR SPECIFIED CAMGORIES 



FIGURE l6, LENGTH, OF SEKVTCEs ITEH 8(b) 



SOME CHARACTERISTICS OF VETERAN STUDENTS 191 



Highest Rating, Bank, or Grade The results for Item 8(c), "What was the 
highest rating, rank, or grade you held while in service?" are shown in 
Figure 17- This was a free-answer item; the categories used in classify- 
ing the responses are shown in Appendix Table 8(c) , 

The great majority of the veterans in this 
study had held enlisted ratings of sergeant or 
petty officer third class and higher. There was 
some tendency , although not a significant one, for 
higher rank to be associated with higher AAG* These 
results are consistent with the hypothesis that char- 
acteristics leading to promotion in a military organiza- 
tion are to some extent the same as those influencing 
overachievement in college. 

The analysis of the item shows that in the twelve basic groups the 
great majority of veterans were in the higher enlisted brackets: i.e., 
sergeant, or petty officer third class, or higher. The frequencies in 
this category typically ranged between 70 and 85 per cent; in three other 
groups, however, the per cent was much lower. The Central State 19^5 group 
had only about 50 per cent in the higher enlisted ratings, and had an un^ 
usually large per cent in the lower commissioned ranks (up to major and 
lieutenant commander) , The same tendency characterized the interrupted 
veterans both in engineering and agriculture at Midwest Tech, where the per- 
centages in the higher enlisted ratings were about 50 and 30 and the percent- 
ages in the lower commissioned ranks were about kO and 55 respectively. The 
higher proportions of commissioned officers tended to appear in groups which 
entered college in 19^5 *" earlier (the interrupted veterans). Otherwise 
only about 5 Per cent of the students had held commissions. Kb veteran was 
found among these students who had held rank of lieutenant colonel, commander, 
or higher. The percentages in the lower enlisted ratings ranged from about 
5 to 20 per cent for most groups; but at Eastern City almost 30 P** cent of 
the interrupted veterans were in this category. 

For students who had held lower enlisted ratings, the median value of 
the mean AAG's was somewhat less than 130, while for the commissioned officers 
the mean AAG for the median group was iV}* The mean AAG of veterans in the 
lower enlisted ratings was lower than the general average for eight of the 
twelve groups, and for the commissioned group it was higher for eight of 
the twelve groups. The veterans in the two other categories tended to be 
intermediate, so far as the medians are concerned. There does seem to be 
a slight tendency for Adjusted Average Grade to be positively related to 
rank held in military service. "When significance tests were made comparing 
each category against the other three categories combined, only one of 
forty-eight such comparisons for the twelve basic groups turned out to be 
significant at the 1$ level; the relationship between rank in the armed 
services and AAG is generally insignificant. 

Duty Outside the United States. Item 8(f ) asks, "Did you serve outside the 
United States, either during or after hostilities?" The response categories 
ares (A) no service outside U\ S., (B) served on sea duty, and (G) served 



192 



ADJUSTMENT TO COLLEGE 



100 



75 



<D 

1 

02 



1 

O 



5 





- 








[ 














\ 







, 

f * 


Corporal, Higher Cadet, Commissioned 
seaman 1/c, enlisted midshipman, ranks 
or lower ratings warrant officer 



= Median 



PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COIiEGE 



150 

13*0 

s 

g 130 

<J 
d 

<D 
43 



3 12 

110 








/ 










._ . 


/ 
/ 
/ 
/ 
^ 




i 


< 


r-'^ 






















Corporal, Higher Cadet, Commissioned Total 
seaman 1/c, enlisted midsliipman, ranks group 
or lower ratings warrant officer 



MEAN AAG OP MB2DIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 17. HIGHEST RATING, BAM OR GRADE HFIJ.T) WffTT.E IN SERVICES ITEM 8(c) 



SOME CHARACTERISTICS OF VETERAN STUDENTS 193 



in land areas outside the U. S. It was necessary to provide a fourth 
category (D) for veterans who had served both on sea duty and in land 
areas outside the United States , since some students checked more than 
one response. The responses to the question, "What areas?" were not 
coded. This item was included to test the hypothesis that those veterans 
who had had duty outside the United States and hence who had had broader 
experiences, travel, and possibly combat experience, would achieve higher 
grades, in relation to ability, than veterans who did not serve outside 
the United States . The results are shown in Figure 18. 

Most of the veterans had served outside the 
United States, more often in land areas than on sea 
duty These veterans with the broadest experience, 
travel outside the United States, and possibly combat, 
did not prove to be superior students in relation to 
their ability. Sea duty, on the contrary, seems to 
be associated with underachievement to a slight extent . 
None of the differences^ however, appears to be 
statistically signif leant * 

Nearly 40 per cent of the veterans, in the median group, had served 
in land areas outside the States, more than 25 per cent had served on sea 
duty, and more than 25 per cent had had no service outside the United States. 
Less than 10 per cent had had both land and sea duty outside the States . 
The largest proportion of veterans who had served outside the country was 
at Evans (about 85 per cent) and the smallest proportion (about 65 per cent) 
was at Stewart, Douglas, and Southern Tech. 

Veterans with no overseas duty and those who had served in land areas 
outside the United States were about equal with respect to AAG, while those 
with sea duty and those with both sea duty and duty in land areas outside 
the States tended to be lower with respect to AAG. For nine of the twelve 
basic groups, the mean AAG for those with no overseas duty was higher than 
the general mean The difference was significant at the 156 level for one 
group, composed of engineering students at Middle State University. On 
the other hand, for nine of the twelve groups the mean AAG of those report- 
ing sea duty only was lower than the general mean. 

Length of Service Outside the United States . Item 8(g) asks for length 
of service outside the United States. The categories used in the analysis 
were (A) no service outside the States, (B) less than six months, (C) six 
to twelve months, (D) twelve to eighteen months, and (l) eighteen months 
or more. Another category (F) was added; this category (service outside 
the U. S.; amount not specified) includes veterans who did not respond to 
Item 8(g) but had answered the preceding Item 8(f ) by checking Served on 
sea duty or Served in land areas outside the U. S. The purpose of this 
item is similar to that of the preceding one; to observe the relation to 
Adjusted Average Grade of a variable which is associated with travel and 
breadth of experience. The results are shown only in Appendix Table 8(g) . 



194 



ADJUSTMENT TO COLLEGE 



100 

75 

CQ 
P 

1 

! 

S 25 














>> 




,,,,,, 





^ 























i ......... 

r III 


Hone Sea 
duty 


Land areas Both land 
outside U. S. and sea 



= Median 



PER CMT KEPCKTHTG DESIGRATEDD CATEGORIES IN EACH COLLEGE 



150 

([) "j liQ 














a 














s 

<D ^ OA 




k. 


x" 


\ 


< 


) 


^ 130 
< 

1 

o 

rrj r\f\ 




> 


^ 




1 




<5 120 

no 














Hone Sea Land areas Both land Total 


duty outside II. S. and sea group 



MEAN AAG OF MEDIAN COLL1SE GEOTIP PQR SPECIFIED CAOIEGORIES 



FIGURE 18. SERVICE OOTSIDE 3HE UNITED STATES; ITEM 8(f ) 



SOME CHARACTERISTICS OF VETERAN STUDENTS 195 



Of those with service outside the United States, 
the majority had served at least a year. The hypothe- 
sis that greater experience, travel, and possibly coin- 
bat (as indicated by length of service outside the 
United States) is associated with higher achievement 
in relation to ability is not borne out; there is, 
rather, a tendency for the opposite relationship to 
appear. 

Category A (no service outside U. S.) is logically the same as Cate- 
gory A of the preceding item, and if all students had been perfectly con- 
sistent in their responses, the two distributions of per cents would be the 
same. As it turned out, the distributions are very similar but not identical; 
about 30 per cent of the veterans in the median group had not served outside 
the United States according to this item. Fewer than 10 per cent had served 
outside the country for less than six months, and for each of Categories C, 
D, and E the median proportion was about one -fifth. 

The mean AAG for the median group of veterans with no overseas duty 
is about 135 > and is no higfrer for any of the other categories. The value 
is about the same for the "less than six months" group and the "18 months 
or more" group; all others are lower. 

In only a few instances is the category mean AAG for a particular group 
significantly different from the mean AAG for -all other categories , At 
Middle State the mean AAG for veterans with no service outside the United 
States is significantly high and for those with eighteen months or more 
overseas significantly low, at the 1$ level. At Adams the mean AAG for 
veterans spending between twelve and eighteen months overseas is signifi- 
cantly low at the 1$ level . In the twelve basic groups, a total of seventy- 
two significance tests were made on the six categories of this item; this 
fact makes it necessary to discount somewhat the presence of three signifi- 
cant differences. 

The mean AAG for veterans with no overseas duty is higher than the mean 
AAG for all veterans for ten of the twelve basic groups; one would expect 
this amount of consistency in the direction of the differences less than 
five times in a hundred trials . The mean AAG for veterans reporting six 
to twelve months of duty outside the U. S. is lower than the mean AAG for 
all veterans for ten of the twelve groups; again the proportion of differ- 
ences in one direction would be expected less than five times in a hundred 

Separation Date- Item 8(h) asks for the year in which the veteran was 
separated from the service. It was necessary to combine all years up to 
19^6 into one category because of the low frequencies of response; the two 
resulting categories were (A) prior to igk6, and (B) 1J&6. This item should 
indicate whether the veterans who enrolled as soon as possible after separa- 
tion achieved higher grades, in relation to ability, than veterans who al- 
lowed a greater amount of time to elapse before enrolling in college. De- 
tailed results for this item are given in Appendix Table 8(h). 



196 ADJUSTMENT TO COLLEGE 



The great majority of the veteran students were 
separated from the service in 19^6, the same year they 
enrolled in college. Veterans who were separated at 
an earlier date tended to earn about the same Adjusted 
Average Grade as those separated in 



It turned out that the great majority of veterans who entered college 
in the fall of 19^6 were separated from the service in the same year; in 
the median group almost 90 per cent were separated in 19*1-6. There was ap- 
parently a strong tendency for the veteran students to enter college as 
soon as possible . Of the twelve basic groups, the highest proportion of 
"before 19^-6" responses was found at the Midwest City College of Engineering 
(30 per cent) . The proportion separated before 1<&6 was of course greater 
for groups which included veterans who entered or returned to college in 



The mean AAG for the median group is almost exactly the same for those 
separated in 19^-6 as for those separated in 19^5 or earlier, and none of 
the differences are significant for the groups considered separately. The 
mean AAG's for the "before 19^6" subgroups are below the group means about 
as often as they are above the group means . Year of separation from the 
service is apparently unrelated to Adjusted Average Grade. 

Smrpnary. The results indicate that most of the veteran students had served 
in either the Army or the Navy for from one to three years and held enlisted 
ratings of sergeant or petty officer third class and higher. More than a 
quarter had not served outside the United States; of those who had, most 
served more than a year outside the United States. The great majority of 
the veteran students in this study entered or returned to college in the 
same year they were separated from the service . 

None of the items was markedly related to AAG. There were slight 
tendencies for high AAG to be associated with greater length of service and 
higher rank. The hypothesis that veterans who had served outside the United 
States and therefore had broader experiences, travel, and possibly combat, 
would excel other veterans in grades relative to ability was not supported; 
in general, the veterans who .served in the Zone of the Interior did better. 

Education Received During Service 

College Training Programs. The purpose of Item 8(d) ("While in service, 
how many months did you spend in college training courses such as V-12, ASTP, 
CTD, or Pre-Flight?") is to determine whether or not such training was related 
to overachievement in college and whether or not the superiority of veteran 
students might be a function of such training. Of course, since many of the 
veterans who received such training were given advanced standing and hence 
not included in the study, the students included had, in general, received 
only a small amount of college training, if any, while in service. The 



SOME CHARACTERISTICS OF VETERAN STUDENTS 197 



categories used in the analysis were (A) none, (B) one month up to six 
months, and (C) six months or more- Results for this item are shown in 
Appendix Table 8(d) . 

When veterans who received advanced standing 
are excluded from consideration, there remain only a 
few who had college train ing while in service. The 
present analysis indicates that the tendency for 
veterans to achieve higher college grades than non- 
veterans of equal ability cannot be ascribed to this 
college training. 

The great majority of the veterans included in the study, after the 
preliminary editing of the data as described in a preceding chapter, had 
received no college training; the median percentage in this category was 
about 80. The median per cent in the "one to six months" group was 15; 
in the "over six months" group, 5, The single striking exception was the 
group of interrupted veterans in the College of Engineering at Midwest Tech, 
where about 70 per cent had received college training. 

The median AAG values shifted upward very slightly with amount of col- 
lege training. The differences in mean AAG were not significant at the 1$ 
level in any of the twelve groups, and the trends In individual groups were 
not consistently in either direction <, 

USAFI Courses. The veterans were asked in Item 8(e) whether they had taken 
any courses from the United States Armed Forces Institute, because it was 
thought that taking such courses might be an indication of academic inter- 
est which would be reflected in college achievement . It is realized of 
course that opportunities for taking such courses were not equally favor- 
able for all servicemen. Only the Yes -No answers were analyzed; the response 
to the further question of "What courses?" was not coded because of the 
small proportion of Yes replies. The findings related to this item are 
shown in Appendix Table 8(e). 

Even fewer of the veterans included in the study 
had taken USAFI courses than had taken college train- 
ing courseSo There was no significant tendency for 
those who had taken USAFI courses to earn higher AAG's 
than veterans who had not taken USAFI courses. 

For the median group, more than 85 per cent of the veterans had taken 
no USAFI courses^ and the range was comparatively small among the twelve 
basic groups o The smallest proportion of No responses among all twenty - 
five groups was at Eastern City (about 70 per cent) and the largest at 
Southern Tech (about 95 per cent) . The median group giving a Yes answer 
did obtain a slightly higher mean AAG than the median group giving a No 
answer, but the difference is negligible . In only one group (Adams, 19^6) 
was the difference between the two means significant at the 1$ level. In 
the twelve basic groups the No-category veterans were lower than their group 
mean in five cases, and the Yes -category veterans were higher in seven cases. 



198 ADJUSTMENT TO COLLEGE 



SiiTTimftry. The great majority of the veteran students had received no col- 
lege training in such programs as the AS13? and V-12 (since such students 
had been deliberately eliminated in most of the groups) and had taken no 
TJSAET courses , The tendency of veterans in this study to earn higher grades, 
in relation to ability, than nonveterans presumably cannot be accounted for 
on the basis of college training while in service, since AAG was not 
significantly related to amount of such college training. Veterans who 
had taken USAFI courses on the average earned AAG f s which were not signifi- 
cantly higher than those of veterans who had not taken such courses. 

Judgments Regarding the Effects of Service Experience 

The items so fax discussed are concerned with relatively objective 
matters related to service experience. In this section a group of items 
will be discussed which have to do with the veteran's own opinions regard- 
ing the relation of his military service to collegeo More specifically, 
the items deal with opinions concerning the effects of service experience 
on eagerness to attend college, on ability to do college work, and on the 
quality of college work actually done,, 

Influence of Service Experience on Eagerness to Attend College. Item 8(p) 
is concerned with the influence of service experience on eagerness to attend 
college. Category A included veterans who reported that, on the whole, their 
experience while in service made them "more eager to go"; Category B was "Did 
not change my feelings about college"; and Category C was "Made me less eager 
to go." The item was included in order to provide some basis for an estimate, 
however crude, of the extent to which military service experiences contributed 
to an increased motivation for college attendance., The analysis of the item 
should also provide an indication of the relation of change in attitude about 
attending college to Adjusted Average Grade. The results are shown graphically 
in Figure 19. 

The veterans tend to agree in testifying that 
service experience made them more eager to attend 
college. Those who reported less eagerness as a 
consequence of military service tended to earn lower 
grades, in relation to ability, than those veterans 
reporting greater eagerness It is possible, however, 
that this relationship results from rationalizing 
the grades already obtained by the students when they 
answered the questionnaire item. 

About 75 P r cent of the students in most of the groups reported that 
military service made them more eager to attend college,, In the median 
group, almost 25 per cent reported no change in attitude, and less than 5 
per cent reported less eagerness to attend college . The proportions tended 
to be fairly uniform in the various basic groups . It should be kept in 
mind, of course, that these figures are based only on veterans who entered 



SOME CHARACTERISTICS OF VETERAN STUDENTS 



199 



I 

<D 



100 



75 



@0 






1 ^@ 

r 








$ 
** 






99 

*** 


L~ 




Increased Ho Decreased 
eagerness effect eagerness 



Median 



PER CEHT CHOOSING HES3X3KA3ED CATSGORIES Hff EACH COSLLBGE 



150 



130 



120 



no 















^--v_ 




t 


' 






"\ 


\ 














* Increased ITo Decreased Total 
eagerness effect eagerness group 



MEAJT AAG- OF MEDIAH COLLBGEE: GROUP FOR SPECIFIED CA3JBGORIES 



FIOOEffi 19 IKFLHENCE OF SERVICE 



OH EAJSERHESS TO A1CEKHD COURSE : ITEM 8(p) 



200 ADJUSTMENT TO COLLiGi 



college; and this finding might not hold for a random sample of veterans. 
Among the interrupted veterans, where there had been college experience 
before war service, there was a greater tendency to report no change. 

The relationship of the item to mean AAG is somewhat more marked 
than for items previously considered. The mean AA0 for the median group 
who reported greater eagerness is about 135; the corresponding figure for 
those who were less eager is about 125 - Ik students whose interest was 
increased were on the average superior to other veterans in nine of the 
twelve basic groups and inferior in only one group. The opposite tendency 
was found for the other two groups of students,, Those who reported no 
change were on the average inferior in ten, and those who reported less 
eagerness were inferior in nine, of the twelve groups. 

In interpreting these results , it must be remembered that the stu- 
dents responded to the questionnaire during the second term of the aca- 
demic year, after they had considerable knowledge of their academic success. 
Since AAG is rather closely related to obtained grades, the possibility that 
the responses are rationalizations reflecting knowledge of academic success 
must not be overlooked. 

Influence of Service Experience on Ability to Do College Work, The purpose 
of Item 8 [gl v as to determine whether, in the opinion of the veteran stu- 
dents, military service experience had increased or decreased ability to do 
good scholastic work in college. The three categories of response were (A) 
increased ability to do good scholastic work, (B) decreased ability, and 
(C) no effect on ability to do good scholastic work* The results are shown 
in Figure 20. 

The majority of veterans thought that military 
service either increased their ability to do good 
scholastic work or had no effect; only about a fourth 
thought that ability was decreased by service experi- 
ences The relation of the item to AAG is highly 
significant and in the expected direction. The inter- 
pretation of these findings is in doubt, it is true, 
because of the possibility that the opinion of the 
veteran was influenced by the grades he actually had 
earned o 

Slightly less than ^0 per cent of the veterans, in the median group, 
reported Increased ability, and approximately an equal number reported no 
effect. Fewer than 25 per cent, in the median group, felt that military 
service experience had decreased their ability to do good scholastic work 
in college o Most of the college groups cluster rather closely about the 
median. One group which differed appreciably with regard to responses to 
Item 8(q) was Eastern City, where kj per cent of the veteran students thought 
that their ability to do academic work was decreased by service experience- 



SOME CHARACTERISTICS OF VETERAN STUDENTS 



201 



100 



I 

^ 



50 



















/ 

f^* 





**** 


* 




r- 




Increased Decreased No 
ability ability effect 

?ER CENT CHOOSING- DESIGNATED CATBG-CRIES 



= Median 



150 






120 



no 



4 


\ 

\ 
\ 










\ 
\ 
\ 
\ 


; 


1 < 






\ 
\ 
\ 
\ 

V 


/ 
/ 
/ 
/ 
/ 








\ 
\ 

\ 


/ 
/ 

/ 







Increased Decreased No Total 
ability ability effect group 

MEAN AAG OF MEDIAN COURSE ORODP FOR SPECIFIED CATEGORIES 



FIGURE 20. IHFLDMCE OF SERVICE EXPERIENCE ON SCHOLASTIC ABILITY: ITEM 8 



202 ADJUSTMENT TO COLLEGE 



The relation of this item to Adjusted Average Grade is very striking, 
in comparison with other questionnaire items. The mean AAG (for the median 
group) of those who felt that ability was increased was about 1^5 and of 
those who felt that ability was decreased was about 110. In the case of 
each of the twelve basic groups, the mean AAG of those students reporting 
increased ability was higher than that of other students in the group, and 
the mean AAG of those reporting lessened ability was lower than that of 
other students in the group. Of the twenty -four differences just mentioned, 
twenty-one were significant at the 1% level There can be no doubt of the 
strong tendency for Adjusted Average Grade to be associated with judgments 
about the influence of war service experience on ability to do scholastic 
work in college. 

The interpretation of these results is again dependent on the fact 
that the questionnaire was filled out late in the academic year when each 
student had rather definite knowledge of his degree of academic success . 
Since AAG has a reasonably high correlation with grades, it appears likely 
that those students whose grades were high tended to attribute their success 
to service experience, while those with low grades blamed their poor achieve- 
ment on experience while in the service. 

Influence of Service Experience on College Work. Item 8(r)(l) was included 
in an attempt to ascertain whether, in the opinions of the veteran students, 
they were doing better or worse in their college work than they would have 
done if they had gone on with their schooling instead of going into the ser- 
vice, regardless of the reasons for their answer. The three response cate- 
gories were (A) now doing better, (B) doing worse, and (c) doing neither 
better nor worse than would have been done. The findings on this item are 
shown only in Appendix Table 8(r)(l). 

Almost as many veteran students felt they were 
doing worse as a result of their service interruption 
as felt they were doing better; about one fourth felt 
they were now doing neither better nor worse. As with 
the preceding item, the responses are significantly re- 
lated to AAG. 

Although it was found in the previous item that only about one fourth 
of veterans attributed a loss in scholastic ability to their service experi- 
ence, the percentage who felt that they were doing worse than they would have 
done if their schooling had not been interrupted rose to about 35* There was 
a corresponding reduction in the neutral response, from almost kO per cent 
to about one fourth. The percentage who felt that they were doing better was 
about the same as the percentage who judged that their scholastic ability was 
increased. A plausible hypothesis is that some veterans felt that their 
present work was hampered by their interrupted schooling, even though their 
service experience had not lowered their basic ability to do college work. 
A related possibility is that, since the college record was relatively con- 
crete and definite, they expressed a definite positive or negative attitude 
concerning it; the more nebulous concept of "scholastic ability" tended to 
elicit neutral responses 



SOME CHARACTERISTICS OF VETERAN STUDENTS 203 



In this item, as in the item previously described, the relation of the 
item to AAG was clearly significant; the mean AAO for the median group re- 
porting that they were doing better was about 150 and for those who said 
they were worse, about 115. Again twenty-one of the twenty -four differ- 
ences between category means and the mean AAr r s for the total groups were 
significant at the 1$ level, Again it must be pointed out that the hypothe- 
sis that the results are due to rationalizations must be seriously considered. 

Seasons for Influence of Service Interruption on College ork Item 8(r)(2) 
is an open -end question, "What is the most important reason for your answer?" 
which followed Item 8(r)(l). The purpose was to discover the reasons (or 
rationalizations) given by veteran students as justification for doing better 
(or poorer) work than they would have done if they had continued schooling 
rather than entering the service. 

Two categories were used in coding the responses of those students who 
had chosen the "now doing better than I would have done" response, and two 
different categories were used for coding the responses of students who chose 
the "now doing worse' 1 response. The no response cases included those who 
chose the "neither better nor worse* 1 category as well as those who failed 
to make any response to Item 8(r)(2). 

The two 8(r)(2) categories used in coding the responses of students 
who thought they were "doing better" may be characterized as follows: (A) 
more mature, more responsible, broader experience; and (B) improved attitude 
toward education, clearer objectives and better concentration. Reasons given 
for "doing worse" were classified into two more categories : (C) 'impaired 
ability to absorb new information, have lost knack of studying, have for- 
gotten background knowledge; and (D) restlessness, nervous tension result- 
ing from wartime experiences, changed sense of values, tendency to place 
extracurricular activities above academic achievement. The results for this 
item appear only in Appendix Table 8(r)(2). 

The analysis of the responses seems to show that 
veterans who feel they are doing better in college 
than they would have done if they had continued their 
schooling tend to attribute their doing better to broad 
factors such as maturity and experience slightly more 
often than to more specific attitudes concerning educa- 
tion and educational objectives. Those who report that 
they are doing worse feel that it is due to loss of 
specific skills or information more often than to emo- 
tional or attitudinal factors. The reason given is 
unrelated to AAG, although, as noted before, those who 
thought they were now doing better earned significantly 
higher grades than those who thought they were doing 
worse. 

The reasons given for doing better which were classified as Category A 
were given slightly more often than the Category B responses; the median per 
cents were about 20 and 15 respectively. The tendency was to ascribe the 



204 ADJUSTMENT TO COLLEGE 



better achievement to maturity and the like slightly more often than to 
more specific attitudes and motives . The Category C reasons for doing 
worse were chosen considerably more often than the Category D reasons y the 
percentages for the median groups being almost 25 and about 10 respectively. 
There seemed to be a definite tendency to ascribe poorer achievement to 
loss of specific skills or information rather than to emotional or atti - 
tudinal factors. 

The most striking tendency, so far as mean AAjG is concerned, is for the 
students who gave reasons for doing better to earn higher AAG's than the 
students who gave reasons for doing worse. This effect is of course the 
same as the one described in connection with Item 8(r) (l) . The type of 
reason given for doing better is unrelated to AAG-; similarly, the type of 
reason given for doing worse is not associated with grades relative to 
ability. 

SiMmarr* The great majority of the veteran students claimed that service 
experience made them more eager to attend college, while very few reported 
less eagerness. Less than half thought that ability to do college work was 
increased as a consequence of military service, while about a fourth thought 
ability was decreased. With regard to the question of the influence of 
service experience on the quality of work done in college, opinion was almost 
equally divided between "doing better" and "doing worse" than would have been 
done without the intervention of military duty. "Doing better" is slightly 
more often attributed to such broad factors as maturity and experience than 
to more specific attitudes concerning education and its objectives. "Doing 
worse" is blamed on loss of specific skills or information more often than 
on emotional or attitudinal factors. The results for these attitudinal items 
are undoubtedly related to the fact that the questionnaire was filled out 
after the students had considerable knowledge of their academic success in 
college. Students with high grades presumably tended to attribute their 
success in part to service experience, while students with low grades tended 
to blame service experience for their poor standing. ^Nevertheless, the rela- 
tionships found throw significant light upon the process by which students 
evaluate their past experiences in relation to present status. 

Service Experience and Educational Plans 

Time of Decision to Attend College. Many veteran students had planned from 
the beginning to attend college, and for these students the war merely post- 
poned or interrupted college attendance. For other veterans, college had 
not been seriously considered until experience related to employment or war 
service and possible financial assistance through the educational provisions 
of the GI Bill influenced the decision to attend college. The purpose of 
Item 8(J) was to investigate the variation in tioae of decision (defined in 
relation to high school attendance, employment, and service experience), and 
to study the Adjusted Average Grade of students who decided at these various 
times to attend college. The item as stated was, "When did you first decide 
definitely that you would go to college?" The responses were as follows: 
(A) before graduating from high school; (B) after working awhile, but before 
entering the service; (C) while in service; and (D) after discharge from the 
service. The results of the analysis are shown in Figure 21. 



SOME CHARACTERISTICS OF VETERAN STUDENTS 



205 



CO 

-P 



I 

CO 



-p 

I 






100 



75 



m 











>~ 

















:. 
>r 






I *** 








\ 9999* 


While After work- While After dis- 
in high ing, ^bef ore in charge from 
school service service service 



f = Median 



PER CENT REPORTING- DESIGNA.1EID CATEG-CRIES IN EACH COLLEGE 



150 



130 



120 



110 















4 


^ 

X 


'v. 
^"> 


^--^ 






























While After work- While After dis- Total 
in high ing,, "before in charge from groiip 
school service service service 



MEAN AAG OF MEDIAN COLraGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE. 21* TIME OF DECISION TO ATTEND COLMGE: TTOC 8( j) 



206 ADJUSTMENT TO COLLEGE 



Veteran college students generally decided to 
go to college wnile still in nigh school; at some 
universities this tendency was especially marked 
Bie remaining students generally decided while in 
service. The time of decision to attend college 
bore essentially no relation to Adjusted Average 
Grade* 

Considerable variability among groups was found with respect to the 
proportion of veterans who had decided to go to college before graduating 
from high, school At several colleges more than 90 per cent had decided 
while in secondary school. The proportion was highest at Stewart (over 95 
per cent); at Adams and Douglas the per cent was also 90 or higher. In the 
remaining basic groups, the percentage of responses in this category varied 
from about ^5 at Evans to about 75 for Harris University. In the interrupted - 
veteran groups, practically all had decided while in high school to go to 
college* For the interrupted veterans in the two groups at Midwest Tech, 
the percentages went down to below 90; the remainder of the students had 
generally worked a while before deciding to go to college. 

Except for interrupted veterans, the time of decision was, if not before 
high school graduation, generally during military service. Comparatively 
few veterans said that they first decided to go to college after discharge 
from the service. 

Inspection of the mean AAG's for the median groups reveals that time 
of decision to attend college is not strongly related to Adjusted Average 
Grade. Students who did not decide to go to college until they were in 
service earned almost exactly the same mean AAG as those who decided while 
in high school. A category mean AAGr was significant at the 1% level in only 
one instance: at Harris, the group of students who decided to attend after 
working but before war service was significantly superior to al 1, other stu~ 
dents in the veterans subgroup, 

Interruption of Educational Career > Both veterans and nonveterans were 
asked to answer Item 6(b)V "When were you last in full-time attendance in 
high school or preparatory school?" This item was considered to indicate 
year of high school graduation; the phrasing of the question was designed 
to prevent confusion in cases where veterans were granted diplomas after war 
service on the basis of military training. Year of high school graduation 
was almost invariably 19^6 for the nonveterans in the twelve basic groups; 
therefore, detailed results are presented only for the veteran students, 
(it should e noted that^ for this item? only those college groups composed 
of freshmen entering in 19^6 will be considered.) 

Five categories were used in analyzing the responses: (A) prior to 
, (B) 19^1-19^, (C) 19^3, (D) 19W*-, and (E) 19^5-19^6. Since the 
veterans had been away from formal school for varying periods of time, it 
was thought that any trend which might appear in AAG's of these groups would 
be useful in understanding veteran-nonveteran differences. The results are 
shown in Figure 22. 



SOME CHARACTERISTICS OF VETERAN STUDENTS 



207 



s 



s 



100 



75 







Median 



Prior 
to 19i|.l 



19l*l- 
1942 



1911-3 



19W- 



19^*5- 
1$*6 



PER CENT EEPCEITING DESIGNATED CATEGORIES IN EACH COLLEGE 



l^U 





\ 












1**0 




\ 












g 




\ 












i 




^ 


< 




__^ 


t < 




i 13 

rr-t 


















43 

_j - OA 
















g 120 

IT A 

















Prior 19^1- 19^3 19W- 19^5- Total 
to 19^1 19^2 19^-6 group 

JMEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CA03EGORIES 



FIGURE 22- TEAR OF LAST FULL-TIME ATTENDANCE IN SECONDAET SCHOOL: ITEM 6(b) 



208 ADJUSTMENT TO COLLEGE 



The amount of service interruption varied con- 
siderably; with more veterans in the two-year category 
than in any other. Those who had finished Mgh school 
at least six years before they entered college tended 
to obtain relatively higher grades than tho.se ith 
less interruption. 

Inspection of the arrowheads representing medians shows that typically 
about 30 per cent of the veterans last attended high school in 19W-; most 
of them, presumably graduated in the spring of 19^, two years before they 
entered college. Nearly 20 per cent had finished school more recently than 
19i-, and the remainder had been out of school for more than two years. About 
10 per cent had last attended high school prior to 19^1, at least six years 
before college entrance. The various college groups differed considerably 
with respect to year of high school graduation; at Stewart, for example, no 
veterans were found to have graduated before 19kL, while at Midwest City 2^ 
per cent were in this category. 

There is a tendency for those veterans who had graduated six or more 
years previous to college entrance, the "before 19^1" group, to excel in 
Adjusted Average Grade. The mean AAG for the median group in this category 
is almost 1^5, while for students who had graduated in 19^3 or later, the 
corresponding value is about 130. In all eleven of the basic groups con- 
taining veterans who reported graduation before 19*H, the mean AAG of these 
veterans was higher than for the other veterans. In six of the eleven 
groups, the difference was significant at at least the 5$ level of confi- 
dence . The recent (19^0 graduates, on the other hand, were lower in mean 
AAG than veterans with a greater or with less interruption in ten of the 
twelve basic groups. The superiority of the veterans whose high school 
attendance was most remote may perhaps be attributed to selection; such 
veterans, being older, probably do not choose to return to college (or are 
not admitted to college) unless there are special factors of motivation 
which are later responsible for the tendency to over achieve. 

SiKTimR-ry. Veteran students generally had decided to go to college while they 
were still In high school; but colleges varied considerably in the proportion 
of veterans who had decided at that time. If the decision was not made 
while in school, it was almost always made while in the service. Most 
veterans had last attended school two or more years prior to college 
entrance. Time of decision to attend college was unrelated to AAG; there 
was, however, a tendency for those veterans who had completed high school 
before 1$A-1 to excel in Adjusted Average Grade. This tendency might, of 
course, he accounted for on the basis of selective factors: the older 
veterans probably do not choose to return to college unless they are strongly 
motivated to do so and have special incentives for college work. 



SOME CHARACTERISTICS OF VETERAN STUDENTS 209 



Hhat About the Married Veteran? 

Marriage; Status and Plans, Item 3^ inquires as to the marital status of 
the students. The presence of numerous married veterans on the campus 
during the postwar period provided an opportunity to study how the married 
veterans compared with the unmarried in college achievement adjusted for 
ability differences. The analysis to be reported is limited to veteran stu- 
dents, since practically none of the nonveteran freshmen were married. 

!Ehe item as it appeared in the questionnaire contained four choices: 
(l) single, not engaged to be married, (2) single, engaged to be married, 
(3) married, and (k) widowed, divorced, separated, For purposes of analy- 
sis, the very. few students in (4) were combined with those in (3) to make 
a new category: married, now or previously. The results are shown in 
Figure 23. 

The veterans in this study were typically 
single and not engaged. A significant tendency was 
found for married veterans to excel the single 
veterans in Adjusted Average Grade. It is of 
course not necessary to assume that students improve 
scholastically as a consequence of getting married. 
It is more likely that the superiority of married 
students is the result of selective factors; a 
married student would not ordinarily be expected to 
continue in college without unusually compelling 
motives to do soo 

The great majority of. the veteran students in the basic groups were 
single and not engaged to be married The median percentage was almost 80. 
For the per cent engaged, the median was about 10 per cent; the median for 
per cent married was slightly greater than 10 per cent. The percentages 
did not vary greatly among the various colleges, although at Adams and 
Stewart more than 90 per cent were single and not engaged, while at Midwest 
City, the per cent in this status was less than 70. The per cent married 
was, logically enough, higher in the older groups; it reached a high of 
about if5 per cent in the interrupted group at Midwest Tech, At the other 
extreme were Stewart and Eastern City, with no married freshman veterans 
in the group studied. 

The married veterans tended to earn higher grades, in relation to 
ability, than the single students. In the median group the mean AAG- of 
married students was about 1^0; for single students not engaged, the corres- 
ponding median AAG was a little over 130, and for engaged students it* was 
slightly under 130, In each of the eleven basic groups which contained 
married veterans, the married students were superior to single students; 
obtaining eleven differences, all in one direction, would be expected by 
chance less than once in a hundred times. Considering the groups separately, 
however, in only one case (Miller) is the mean AAG for married veterans 



210 



ADJUSTMENT TO COLLEGE 



J.UO 

a 75 

I 

CQ 

S 50 

-P 
1 

CM 25 






t> 
M 



































k e*e* 

r @ 


,:. 

1 a 

r M 
i * 


Single, Single, Married 
not engaged engaged 



= Median 



PER GEM? EEPORTIHG- DESIGNATED CATEGC3RIES IN EACH COLLEGE 



I 

s 



8 

1 



130 



110 






SJbagle, Single, 

not engaged engaged 



Married Total 

group 

MEAN AAG OB 1 MEDIAN COUU3GE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 23 MAETEAL STATDS: CTEM 



SOME CHARACTERISTICS OF VETERAN STUDENTS 211 



significantly greater at the 1$ level than the mean AA& for students fall- 
ing into other categories j it should be kept in mind, of course, that the 
number of married veterans in most groups was rather small . 

The remaining questionnaire items concerned with marital status were 
to be answered only by students who had answered Item 34 by checking 
married . The percentages and mean AAG's are therefore based on much 
smaller numbers of students than hare typically been available. In order 
to minimize the number of statistics based on extremely small samples, the 
responses to these "married only" items have been classified into two cate- 
gories in aH cases. Since there were no married veterans in the Stewart 
group, the findings are based on only eleven of the basic groups. 

lumber of Years Married. Item 35(a) asked simply, "About how long have you 
been married?" In order to get two categories with approximately equal 
numbers of responses, several choices were combined to form Category A, one 
year or more. Category B was, accordingly, less than one year. Analysis 

of this item should reveal whether or not length of time married is related 

to Adjusted Average Grade, as well as to find out something about how long 
veterans who are in college have been married. Results are presented in 
Appendix Table 35 (a). 

A majority of the married veterans had been 
married for at least a year. So far as our data 
show,, length of time married is unrelated to Adjusted 
Average Grade. The data are, of course, not adequate 
to furnish a good test of the hypothesis. 

Since the number of married veterans in the eleven basic groups having 
some married veterans varied from 62 to only 3 (at Littletown State), the 
results are extremely unreliable. The statistics for the median groups may 
furnish a more stable reference point, however It was found that almost 
two thirds of married veterans had been married for a year or more. The 
mean AAG's were approximately the same for the median groups in the two 
categories, and the number of groups in which the "year -or -more" mean AAG 
was higher than the "less -than -one -year" mean was no greater than would be 
expected by chance* 

Number of Children. Item 35 (b) asks, "How many children do you have?" 
The two response categories used were (A) none; and (B) one or more. The 
hypothesis to be tested is that veterans with children earn higher grades, 
in relation to ability, than veterans without children. Results are given 
in Appendix Table 35 (b). 

Among the married veterans in this study, only 
about one fourth had one or more children; the avail- 
able data showed no tendency for the presence or 
absence of children in the family to be related to 
AAG. 



212 ADJUSTMENT TO COLLEGE 



Die married veterans in the eleven basic groups under consideration 
typically had no children- Only one fourth of the median group reported 
having one or more children. Again the mean AAG f s for the median groups 
in the two categories were approximately the same* The number of groups 
where the veterans with, children excelled in mean AAG exactly equalled 
the number of groups where the childless veterans obtained a higher mean 
AAG, 

Satisfaction with Living Arrangements . Item 35(d) asked, "How well satis- 
fied are you with the living arrangements you and your wife have at the 
present time?" The number of categories was reduced to two by letting 
Category A represent those- who were satisfied and Category B those who were 
dissatisfied with their living arrangements. Results are presented in 
Appendix Table 35 (d). 

Roughly two thirds of married veterans reported that 
they were satisfied with their living arrangements; no 
evidence was found that attitude toward living arrange- 
ments was related to AAG. 

So far as the medians show, about two thirds of the married veterans 
were satisfied and about one third dissatisfied with their living arrange- 
ments. As might be expected with so few students in the groups,, this per- 
centage varies greatly from college to college. No difference between the 
satisfied and dissatisfied veterans was found as far as AAG was concerned, 
and no consistent difference occurred in the various college groups con- 
sidered separately. 

Judgments About the Relation of Marriage to Studies. In Item 35 (e) the 
married students were aaked^ "In general, do you feel that as a married 
student you are handicapped or benefited, relative to single students, in 
your studies?" The three choices were reduced to two categories by combin- 
ing "handicapped by being married" and "neither handicapped nor benefited" 
into one category (A); Category B includes only students who reported that 
they were benefited. Results for this item are shown in Appendix Table 
35(e). 

Roughly two thirds of married veterans thought 
that they were benefited in their studies by being 
married; there appears to be some tendency for higher 
AAG and reported benefits from marriage to go together . 

In the median group approximately two thirds of the married veterans 
thought they were benefited in their studies by being married. Again, the 
considerable variability among groups may well be a function of the large 
sampling errors which result from the small numbers. There is some tendency 
for the students who felt they had benefited from marriage to excel in Ad- 
justed Average Grade; the mean AAG's are about 1^5 &a& !35> and in nine of 
the eleven groups their mean is above that of their less happy colleagues. 



SOME CHARACTERISTICS OF VETERAN STUDENTS 213 



It appears likely, however, that the underlying consideration is that 
high -achieving students are more likely to view their marriage favorably 
in this respect y while low achievers tend to take the opposite view. 

Sii*nft.rj_. in. the typical group of beginning freshmen about four fifths of 
the veterans were single and not engaged and about one out of ten "was 
married. 33ie proportion of married students varied from none to almost 
25 per cent. Married veterans tended to earn higher Adjusted Average 
Grades than unmarried students. 

Responses to items intended only for .married students shoved that, 
among these freshmen, about one third of the married veterans had been 
married for less than a year, about three fourths had no children, about 
two thirds were satisfied with their living arrangements, and about two 
thirds thought they were benefited in their studies by being married. Hone 
of liiese characteristics was found to be related to AAG except the last 
there was some tendency toward over achievement on the part of veterans who 
thought marriage had helped them in their work. "Whether this finding re- 
flects anything beyond a more favorable attitude toward marriage among the 
overachievers (and vice versa) cannot be determined from, these data. 



Conclusions 



The "typical" veteran student who entered college in the fall of 
might j insofar as the findings of the present study are representative, be 
described as follows: 

He was on active duty one to three years, held an enlisted rating of 
petty officer third class (or sergeant) or higher, and was more likely to 
have served in the JTavy than in the Army. He served outside the United 
States, more often in land areas than on sea duty, for six months or more; 
he entered college in the same year ha was separated from the service; he 
had had no college training and had taken no "OSAFI courses. (Veterans who 
had received sufficient college training while in service to give them ad- 
vanced standing were excluded from the study.) 

0!he typical veteran believed, according to his questionnaire responses, 
that his service experience made him more eager to go to college. He did 
not feel that his service experience had decreased his scholastic ability. 
With respect to the effect of the interruption of his schooling, he was 
about as likely to feel that he was doing better as that he was doing worse 
in his college work than he would have done had he gone on with his school- 
ing instead of going into the service; whether this opinion was favorable 
or unfavorable seemed to depend in part on how well he was succeeding in 
college at the time he filled out the questionnaire* 

He had decided to go to college while he was in secondary school, and 
two or more years had elapsed between school and college. He was not 
married and not engaged to be married at the time he filled out the ques- 
tionnaire . 



214 ADJUSTMENT TO COLLEGE 



The majority of the questionnaire items answered by the veteran stu- 
dents shoved no significant relation to Adjusted Average Grade. Among the 
few items which were related to AAG were several which dealt with opinions 
regarding the effects of service experience on college vork and on ability 
to do college work. For these items the relationship between opinions and 
AAG which was found might best be interpreted as evidences of rationaliza- 
tion ; since grades in college were known by the respondents when they filled 
out the questionnaire,, it is quite possible that students with low grades 
tended to blame service experience, while students with high grades tended 
to attribute their success to that factor. 

There was a tendency for those veteran students who had completed 
secondary school six or more years before starting college tp earn higher 
AAG-'s than students who had finished school more recently. This tendency 
might be the result of selective factors; older men probably do not choose 
to return to college without unusually compelling reasons for doing so, 
reasons which are related to greater motivation for college achievement. 
It was also found that married students tended to excel with respect to 
AAG, and again the hypothesis of selective factors may be invoked to account 
for this differences perhaps the married veterans who choose to return to 
college are those with stronger incentives,, Whether the superiority of 
older veterans and married veterans should be attributed to personality 
change 3 associated with age an^ -the responsibilities of marriage or to 
selective factors which become operative by virtue of increased age and 
responsibility for a wife unfortunately cannot be definitely determined . 



215 



Chapter V 
AGE AM) GEHERAL BACKGROUND OF TOTERAH AND NONVETERAIiF STODEOTS 

To the college faculty, perhaps the most salient characteristic of 
the veteran students was their greater age. Less easily observed but 
equally worthy of consideration were possible differences between veterans 
and nonveterans in work experience and in family, community, and secondary 
school background. The present chapter will deal with the questionnaire 
items which fall in this general area, taking particular account of the 
light they may throw upon veteran-nonveteran differences in Adjusted Average 
Grade 

Age 

It is very tempting, in speculating about reasons for the veteran 
superiority in grades relative to ability, to ascribe the difference to age, 
to assume that the mere fact of being two or three years older gives the 
veteran a greater maturity which accounts for his greater achievement. It 
unfortunately appears to be impossible to settle this problem from the data 
here available. Age and veteran status are inextricably bound together; it 
is impossible to be a veteran without spending some time at it. A "young" 
veteran could not have spent much time in the service and hence cannot be 
representative of veterans generally. Similarly an "old" nonveteran is 
older than the typical nonveteran because he delayed going to college for 
some reason, which reason is likely to make him atypical not only with 
respect to age but also other characteristics. He therefore cannot reason- 
ably be used to represent nonveterans generally. 

An attempt was made, when the data were being collected, to find a 
group of nonveterans whose college work had been interrupted by something 
other than military service. It was intended to employ such a group as a 
control group with which to compare the interrupted veterans. It was not 
found possible to find such students in sufficient numbers. Perhaps it is 
just as well. Such a group might have been composed mainly of those classi- 
fied as t-F, have differed in desire to attend college, or have been unrepre- 
sentative in some other way of nonveteran students generally. 

Age, then, is a characteristic which, so far as this study is concerned, 
is almost synonymous with veteran status. As will be seen, sorting students 
into two age groups is almost the same as classifying them with respect to 
veteran status. 

Item 32, which deals with age, asks simply, "When were you born?" Mne 
choices ranging from "before 1923" to "1930 or later" were provided. In 
order to have frequencies of reasonable size, it was found necessary to use 
one set of three categories for veterans and a different set of three cate- 



216 ADJUSTMENT TO COLLEGE 



gories for nonveterans The categories were in the order of "older" to 
"younger" in each. case. The categories were as follows: 

Veterans Nonveterans 

A. Before 192^ A. Before 1928 

Bo 192^ or 1925 Bo 1928 

C. Later than 1925 Co Later than 1928 

Because of the use of different categories for veterans and nonveterans, 
it is necessary to present the results in two figures instead of one. 
Figure 2^ shows the general findings for male veterans and Figure 25 for 
nonveteran male students in the twelve basic groups * 

The typical veteran who entered college as a 
freshman in 1946 was born in 1925; the typical non- 
veteran was born in 1928. There is little overlapping 
in the age distributions of veteran and nonveteran 
college students. The oldest subgroup of veterans 
shows a significant tendency to -overachieve as com- 
pared with younger veterans, while the younger non- 
veterans exhibit a similar but less marked tendency 
to earn higher Adjusted Average Grades as compared 
with older nonveterans. These tendencies can prob- 
ably be accounted for by selective factors. 

So far as the median groups are concerned, about 35 per cent of the 
veterans were born in 192U or 1925 and about 50 per cent later than 1925; 
most of this latter group were born in 1926. In the median nonveteran sub- 
groups, about ?0 per cent were born in 1928, almost 25 per cent later than 
1928, and only a little more than 5 P*" cent prior to 1928. There is thus 
very little overlapping in the age distributions of veteran and nonveteran 
male students. Female students were quite similar to the male nonveterans 
with respect to age distribution, although there was a tendency for women 
to be slightly older Variability in age is understandably greater for 
veterans than for the male nonveterans. Veteran students at Adams, Stewart, 
and Douglas tended to be younger than veterans at other institutions. 

Inspection of the median values of the mean AAG's reveals that age is 
related to Adjusted Average Grade, both for veteran and nonveteran students, 
but that the direction of the relationship is different for the two subgroups. 
Among veterans, the older students tended to earn high grades in relation 
to ability, while for nonveterans the younger students tended to earn the 
highest AAG T So The mean AAG- for the oldest veterans (born before 192^) is 
significantly higher (at the 1% level) than for veterans in other age groups 
in three of the twelve basic groups. It is also significantly higher at 
the Midwest Tech College of Agriculture, among the four additional groups 
for which AAG's were computed . These oldest veterans excelled other veterans 
in ten of the twelve basic groups . The trend among nonveterans for younger 
students to earn higher AAG ? s is not so striking- 



AGE AND GENERAL BACKGROUND 



217 



100 



p 



o 



& 



50 



25 



Median 



Before 192^ or Later 
192^ 1925 than 1925 

HER GEM 1 REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



yj 
1 3A 














\ 








0) 




\ 








SP 

m 1 "3H 




N 


i < 


. < 
> 




8 

P 














**~O 

__r -I <or\ 












^ 120 
i n A 












Before 1924 or Later Total 
192^ 1925 than 1925 group 



MEM AAG OF MEDIAN COLEEXzE (iRODP POR SPECIFIED CAOIEGQRIES 



PIGURE 2^o 1EAR OF BIRTH^ MALE "VETERAN STUDENTS: ITEM 32 



218 



ADJUSTMENT TO COLLEGE 



100 



75 



-P 

co 



-p 
fl 


O 



25 





o 






oo 






ooo 






ooooo 






o 








000000 






o 
ooo 


oo 




00 


> ooooooooo 
o 






Before 1928 Later 
1928 . than 1928 



Median 



PER GEM? KEPOKTING DESIGNATED CATEGORIES IN EACH COLLEGE 



Adjusted Average Grade 

- M \-> H H 
- 1 fO C>J 4r- Vji 
D O O O C 






















< 


^^ 


-""""' 


s 















1928 



Later Total 
than 1928 group 



Before 
1928 

MEAN AAG OF MEDIAN COLLEGE GBOUP FOR SPECIFIED CATEGOBIES 



FIGURE 25. ZEAB OF BIRTH, MALE NONVETERAN STUDENTS: ITEM 32 



ACE AND GENERAL BACKGROUND 219 



It seems reasonable to account for the trends on the basis of selective 
factors. The veterans "born before 1924 entered college when they were 23 or 
older; the typical nonveteran was 18. The older group of veterans would then 
be at least 27 at the time of graduation. Such students would not be expected 
to begin a college career unless there were special incentives or unusually 
strong motivation for undertaking academic work-, The superiority of the 
oldest veteran subgroup probably can be attributed to such selective factors. 
It may be surmised that this group of veterans may have contributed dispro- 
portionately to the favorable impression created by the veteran group upon 
teachers and administrators . 

In the case of nonveterans, selective factors of a different sort were 
probably at work. It is well known that a negative correlation is usually 
found between age and intelligence for students within a particular high 
school group; this can be accounted for in terms of acceleration of the best 
pupils and retardation of the poorest. The youngest nonveterans, then, are 
those pupils whose school progress was accelerated because of more rapid 
achievement. These students are continuing to show in college the same 
characteristics which caused their arrival at college at an earlier age. 

The question as to whether or not the superiority of veteran students 
is merely a function of greater age cannot be rigorously answered. There 
is very little overlap in the age distributions of veteran and nonveteran 
students; and even if subgroups of veterans and nonveterans alike in age 
could be found^ their comparison would not settle the issue y since such 
"young" veterans and "old" nonveterans would presumably not be representative 
of veterans and nonveterans generally. Thus, controlling age in a study of 
veteran-nonveteran differences would be somewhat analogous to controlling 
depth of voice in a study of male-female differences in college students. 
In this study, preference was given to the comparison of veterans and non- 
veterans chosen to be as typical as possible of their groups; under these 
circumstances the two groups necessarily were quite different in age. It 
should be noted that, to the extent that greater age is associated with 
greater maturity, more direct evidence regarding maturity of attitudes and 
motivation will be found in later chapters of this report. 

From one point of view, however, the findings for this item suggest 
that greater age, in and of itself, can not account for veteran -nonveteran 
differences in AAG Eliminating the oldest group of veterans virtually 
destroys any correlation between age and AAG in the veterans group, while 
leaving the "younger" veteran group superior by a substantial (though re- 
duced) amount to the nonveteran group . Although the argument is not 
rigorous, it makes less attractive the hypothesis that the superiority of 
veteran students is primarily due to their greater age. 

Work Experience 

Work experience may reasonably be thought to have a maturing effect 
on young employees, the possible relation of such experience to veteran- 
nonveteran differences in AAG was studied in the analysis of Item 9 Ob). 



220 ADJUSTMENT TO COLLEGE 



This item asks, "If you worked full-time before entering military service 
or college, how long were you employed?" In the analysis, the categories 
were (A) did not work full -time, (B) worked less than six months, and (c) 
worked six months or more* Item 9 (a) > which also dealt with work experience, 
was used to identify individuals without work experience who omitted Item 
9 (l>); such students were included in Category A. The results of this analy- 
sis are shown graphically in Figure 26 

It was found that substantially more of the 
veteran students had had work experience than was 
true of nonveterans; half of the male veterans and 
three quarters of the male nonveterans had not held 
a full-time job. As might be expected, even fewer 
of the women students had worked Generally non- 
significant differences in AAG were obtained, al- 
though veterans who had worked six months or more 
earned slightly higher AAG's than other veterans, 
and nonveterans with work experience of six months 
or more earned lower AAG 8 s than other nonveterans . 

A rather striking difference is found between veteran and nonveteran 
male students with regard to amount of work experience. In the median 
veteran subgroup, about 50 P** cent reported no full-time employment, while 
for the median male nonveteran subgroup the percentage was somewhat more than 
75. Conversely for those employed six months or more the median percentages 
were about 30 for veterans and just over 5 for the nonveteran subgroup . Women 
students reported work experience slightly less often than the nonveteran 
males. 

There was considerable variability among colleges with respect to work 
experience of students . At Adams, Stewart, and Douglas (of the twelve basic 
groups) 90 per cent or more of the nonveteran men reported no work experience, 
while at Evans, Western State, and Midwest City fewer than 75 P*" cent had 
not had a full-time job* The range was even greater for veterans: 85 per 
cent or more at Adams and Stewart had not worked, while at Evans and Midwest 
City the percentage was only about 35 

The relation of amount of full-time employment to AAG proved to be 
negligible. Judging from the medians as plotted in Figure 26, the relation- 
ship is slightly positive for veterans and slightly negative for nonveterans, 
although neither trend is significant. When the relation between this item 
and AAG was considered for each subgroup separately, a significant asso- 
ciation was found in two subgroups of veterans (at Adams and Western State), 
where the students employed six months or more earned a mean AAG which is 
significantly higher (at the 156 level) than the mean AAG of students in 
other categories. In none of the nonveteran subgroups were differences 
found which were significant at the 1$ level , Although veterans %nd non- 
veterans differed noticeably in amount of work experience, this factor obvi- 
ously cannot account for veteran superiority in AAG because of the lack of 
any marked relationship between AAG and amount of work experience. 



ACE AND GENERAL BACKGROUND 



221 



100 



1 
I 





ooo 


> 00000 






I 

p Medi 
Vete: 
o Nonv 


an 
rans 
3terans 




ooo 



















m 




< 





















oo 










^ 


> ooooo 














oo 













000 





; oooooooo 
oooo 


W MN W MM" W MN" 
None Less than Six months 

six months or wiore 



PER CENT BEPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 



130 



120 



110 













< 




___^-- -"" 






< 


>-^ ^ 




\ 












Veterans 
o Nonveterans 




None Leas than Six months Total 
six months or more group 



MEAN AAfi OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 26 . LENGTH 0!F FULL-TIME EMPLOYMENT BEFORE ENTERING WAR SERVICE OR 
COLLEGE: ITEM 



222 ADJUSTMENT TO COLLEGE 



Size of Community 

The size of community from which a student came may be pertinent to 
his college adjustment! accordingly, attention was given to differences 
between veterans and nonveterans in this respect. For both veterans and 
nonveterans, the relation of size of community to AAG was studied, and the 
possible value of this information in accounting for veteran-nonveteran 
difference was assessed. Size of community might be expected to throw some 
light on the problem because of the relations of community size to cultural 
opportunities available and to general quality of secondary education 

Item 31, which dealt with community size, was stated s "How large was 
the community in which your home was located during the time you were in 
high school? (If your residence was a suburb or town in a metropolitan 
area, check the population of the larger area.)" En five choices in the 
questionnaire were reduced to three categories, in order to avoid extremely 
small frequencies | the three categories were (A) less than 2,500 population, 
(B) 2,500 to 100,000 population, and (C) over 100,000 population- The re- 
sults of the analysis of this item are shown in Figure 27- 

Although there is considerable variability from 
college to college, in general it appears that a 
slightly greater proportion of veterans than non- 
veterans came from rural areas and from towns and 
cities of under 100,000 population . Almost half of 
the male students and a somewhat larger proportion of 
women had lived in cities of over 100,000 people when 
in high school For veterans, grades appear to be un- 
related to size of community o Uonveterans from the 
larger cities tended to obtain higher AAG's than other 
nonveterans . 

As is shown in Figure 27, the differences between veteran and nonveteran 
male students with regard to size of community are relatively slight Compari- 
son of the medians reveals a slight tendency for a larger proportion of 
veterans to come from the towns and cities of less than 100,000 population. 
It appears that in the median groups only about 15 per cent of the students 
were from rural homes or from villages of less than 2,500 people. The dif- 
ferences among colleges are very great, however, as would be expected on 
the basis of the characteristics of the various institutions. At Eastern 
City, for example, less than 2 per cent of the students had lived in commtmi- 
ties of less than 2,500 during their high school years, and more than 95 
per cent came from cities of more than 100,000; while at the Midwest Tech 
College of Agriculture more than two thirds of the students were from the 
communities of less than 2,500o The proportion of women reporting that they 
came from small towns or farms was smaller than that for men in practically 
all of the nine groups where women ! s questionnaires were analyzed. 

The relationship between size of commmiity and AAG is slight, and the 
nature of the relationship appears to be different for veterans and non- 
veteranso For veterans, the students from communities of 2,500 to 100,000 



ACE AND GENERAL BACKGROUND 



223 



100 



CQ 



1 

o 



50 











^ Jfedia 


a 










Yeterans 










o Honve 


terans 










_ 


o 
o 













o 






* 


o 


^ 


OO 





o 


f" 


ooo 

00 
00 





I 000 









o 




OO 


L 




>ii! 


o 
^ ooooooo 










00 


@ I OO 












mm mm mm 

Less than 2,500 to Over 

2,500 100,000 100,000 

PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



o 
d> 



S 130 
8 



120 



no 



Yeterans 
o lonveterans 



Less than 2,500 to Over Total 

2,500 100,000 100,000 group 

MEM AAfi OF MEDIAN COIJJJGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 27. SIZE OF HOME COMMDNITX' WHILE IN SECONDARY SCHOOLS ITEM 31 



ADJUSTMENT TO COLLEGE 



were lowest, so fax as the median AAG's are concerned, but the difference 
is not significant. For nonveterans, the students in the first two cate- 
gories were similar in mean AAG, while the students from cities of over 
100,000 were superior; the mean AAG ! s for the first two categories are 
about 120 and for the large cities about 13Q> in the median groups. In two 
of the twelve basic groups the large-city students were significantly 
higher (at the 1$ level) in mean AAG than students in the other two cate- 
gories, and the difference was significant at the 5$ level in three more 
of the groups. In ten of the twelve groups the students from large cities 
were on the average superior in AAG to students in the other categories; 
this proportion of the differences in the same direction is significant at 
the 5$ level. 

Size of community cannot help to account for veteran superiority in 
mean AAG because the difference between veterans and nonveterans in fre- 
quencies is not great and because the item is not, for veterans, signifi- 
cantly related to AAG. 

Among nonveterans, then, there does seem to be a tendency for students 
coming from cities of over 100,000 to earn higher grades in relation to meas- 
ured ability than students from smaller communities. One possibility is 
that students coming from cities were graded more severely during high school, 
so that their tendency to overachieve might merely reflect this relative 
undermeasurement . Fortunately, four of the groups had AAG 'a in which only 
scholastic aptitude tests were used in allowing for ability. In these four 
groups, the city students were significantly superior at the 1% level in one 
instance, at the 5$ level in another, were superior in a third, and equal in 
a fourth to their f ellow-nonveterans . In general, these results are so 
similar to the results for all twelve groups that the hypothesis of relative 
undermeasurement in high school grades may be rejected* The difference prob- 
ably cannot be attributed to differences in grading standards between larger 
and smaller schools . 

If it is assumed that students from the cities have a broader back- 
ground and generally better preparation for academic work, it migiht be 
thought that students from smaller communities would gradually overcome 
their disadvantage as they experienced the richer opportunities offered by 
the college. Therefore, their grades in college would be higher than was 
predicted on the basis of ability measures they would earn AAG's above 130 
providing, of course, that college would tend to equalize the differences 
in background. Such an assumption is not warranted by the find ings reported 
above. It must be remembered, however, that perhaps one year of college is 
not sufficient to overcome background differences appreciably. 

The findings suggest the hypothesis that size of community is not 
associated with any systematic tendency either to depress or to raise ability 
measures (defined, for most of the groups, as a combination of aptitude test 
scores and high school record), but that there is a tendency for city boys 
to acquire certain characteristics perhaps more effective study habits or 
greater motivation which result in higher initial achievement in college 
relative to measured ability. It would also appear, since the relationship 
of AAG to community size disappears for veterans, that whatever advantage 
the city boys had did not persist through the period of war service. 



AGE AND GENERAL BACKGROUND 225 



Father's Income 

Item 43 as&s, "Approximately what was the annual income of the head 
of your family while you were in high school?" The five choices presented 
on the questionnaire were reduced to four categories by combining the two 
choices at the upper end of the income scale The four categories result- 
ing were (A) $6,000 or more, (B) $4,000 up to $6,000, (C) $2,000 up to 
$4,000, and (D) under $2,000. Direct comparison of veterans and nonveterans 
with respect to father 3 s income is unfortunately made ambiguous by the inclu- 
sion in the item of the phrase "while you were in high school." Nonveterans 
were usually in high school until 1946, but the veterans typically left high 
school two or three years earlier . During these two or three years, incomes 
in the United States increased considerably* The difference between veterans 
and nonveterans is undoubtedly affected by this general change . The question 
was presented in the form used in order to get information pertaining to a 
-time when the student was presumably living at home and more directly affected 
t>y family influences,, The results of the analysis of the item are shown in 
Figure 28. 

About half of the nonveteran mule and female 
students estimated their father *s incomes at $4,000 
or more for the period of their high school attendance, 
while only about a third of the male veteran students 
reported this high an income. It should be noted that 
this difference may have resulted from the upward trend 
of incomes between the times the two groups were in 
high school o Considerable variation among colleges 
with respect to family income was found. Both veteran 
and nonveteran students from the "under $2,000" families 
tended to earn higher grades relative to ability than 
students who reported higher family incomes 

Inspection of the arrowheads representing the median groups in Figure 28 
shows a tendency for nonveterans to report higher incomes than the veteran 
students. Honveterans were more likely than veterans to report that their 
father 's income was $4,000 or higher * The greatest difference between 
medians occurs in the "under $2,000" category^ for the median, group of 
veterans more than 15 per cent reported a family income of under $2,000, 
while for the median group of nonveterans lesa than 5 per cent fell in this 
category. Although fewer women than men attempted to estimate an income, 
those who did closely paralleled the male nonveterans. The veteran-nonveteran 
differences must be attributed, at least in part, to the general trend of 
incomes during the war period,. 

The variation among college groups was wide a At Stewart, for example, 
incomes of $6,000 or more were reported by almost 80 per cent of the 
veteran and almost 70 per cent of the nonveteran students, while at the 
Midwest City Engineering College about 10 per cent of the veterans and 15 
per cent of the nonveterans fell in this category. 



226 



ADJUSTMENT TO COLLEGE 



100 



75 



g 

O 



25 
















| Median 
Veterans 
o Nonveterans 





O 
















OO 

























o 














k I oo 






m 


o 






I ooooo 














00 




' 








i OOO 





OOO 





o 







m 


oo 





> ooooo 











<8-$ 





\ 


oo 




o 


> @ 










*e 







oo 
























ooooo 

















ooooooo 



150 





s 
s 



iao 



110 



W 



$6^000 
or more 



m 

$lj-,000 up 

to $6^000 



w m 

$2 ,000 up 
to $4,000 



m m 

Under 
$2,000 



PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 















4 






^^^ 


\ 

) 

i 




' 


Y"' 




^ 














% Veterans 
o Nonveterai 


is 


$6^000 $4,000 up $2^000 up Under Total 
or more to $6,000 to $4,000 $2,000 group 



MEAN AAG OF MEDIAN COLLEGE GROUP FOE SPECIFIED CATEGORIES 



FIGURE 28o HEAD OF FAMILY'S INCOME DURING STUDENT'S SECONDARY SCHOOL 
CAREER? ITEM 43 



ACE AND GENERAL BACKGROUND 227 



The lower part of Figure 28 shows a tendency among both veteran and 
nor**eteran students for lower family income to "be associated with higher 
AAG The difference "between the students in the two extreme categories is 
almost 10 AAG units. In only two of the veteran groups is such a tendency 
significant, however; at Adams and Evans the "under $2,000" students are 
significantly higher in mean AAG than students in other categories (at the 
l4 level of significance) B Veteran students in this lowest income group are 
higher in mean AAG than other veterans in nine of the basic groups and lower 
.a one, which is significant at the 5$ level. The small numbers of students 
in this category among the nonveterans make it impossible to discern any 
trend for these students. 

Although the bearing of income differences upon the explanation of 
veteran-nonveteran differences in AAG is difficult to evaluate from the 
information available , it may be judged that father's income is not an 
important factor in accounting for the differences in overachievement be- 
tween veterans and nonveterans. 

The slight but fairly consistent tendency for students from lower income 
groups to attain higher grades , in relation to ability, than students from 
the upper income groups has interesting implications for those interested in 
scholarship programs or other means of subsidizing higher education Caution 
must be exercised, of course, in generalizing from this finding; the low 
income students are in many instances those who have been awarded scholar- 
ships and they may, therefore, have been selected with unusual care. The 
great similarity in the trend for veteran and nonveteran students suggests, 
however, that more is involved than careful selection; scholarships were not 
ordinarily awarded to veterans. The analysis does show that sons of low in- 
come families tend to achieve higher grades relative to their ability than 
the sons of the more well-to-do families; there is considerable variation in 
the apparent strength of this tendency in the various colleges. 

Father ! s Education 

Item kk was included in order to make possible studies of the differ- 
ence between veterans and nonveterans with respect to amount of father *s 
education, the relationship of father's education to Adjusted Average Grade, 
and the extent to which veteran-nonveteran differences can be explained by 
differences in father's education. The item as stated in the questionnaire 
was, "How much formal education did your father have?" The six choices 
offered were analyzed in terms of three categories: (A) not a high school 
graduate, (B) graduated from high school, and (C) graduated from college. 
The results are shown graphically in Figure 29. 

The fathers of male nonveteran students tended 
to have had more schooling than the veterans' fathers; 
about a quarter of the former group were college gradu- 
ates, whereas fewer than one in six veterans 1 fathers 
had completed college. An even larger proportion of 



228 



ADJUSTMENT TO COLLEGE 



100 



! 
1 



p 



o 



75 



50 



25 





























@ 


00 


















L 


o 















o 






















o 




o 


9 


> oooo 
o 




I 
r 


oooo 
i oooo 

r o 




Q 

o 

00 












oo 


... 


oo 












k... 


ooo 





000 








f .... 















1 





Median 
Veterans 
o Nonveterans 



MV MN 

Not a high 
school graduate 



m m 

High school 
graduate 



MV MN 

College 

graduate 



PER GEM 1 BEPOETBTG DESIGNATED CATEGORIES IN EACH COLLEGE 



Adjusted Average Grade 

J H H H h 

- 1 ro ix> 4=- ^ 

D O O O C 












< 


, 1 


^ -^____ 


I < 


i 


( 


"^ -.^ 


^^ ^ 
) " 


) ( 


> 








Veterans 
o Nonveterans 


Not a high High school College Total 
school graduate graduate graduate group 



MEAN AAG OF MEDIAN COLLEGE GROUP ?OR SPECIFIED CATEGORIES 



FIGURE 29 o FATHER'S EDUCATIONS ITEM 



ACE AND GENERAL BACKGROUND 229 



women students f than of male nonveterans f fathers 
had received a college degree, father l & education 
was not related to Adjusted Average Grade to any 
marked extent, although there was a slight tendency 
toard under achievement among nonveterans whose 
fathers had graduated from high school but not from 
college . 

As shown "by the position of the arrowheads which indicate median groups, 
the majority of the fathers were not college graduates; the fathers who had 
not graduated from high school constituted the largest of the three groups. 
The fathers of veterans tended to have less formal education than the fathers 
of nonveterans; in the median group half of the veterans* fathers had not 
graduated from high school, while less than kO per cent of the nonveterans 1 
fathers had not graduated. Except for Adams and Stewart, the percentage of 
veterans whose fathers were not high school graduates was higher than the 
percentage of nonveterans in the same category for all the "basic groups. 
Similarly, the percentage of veterans whose fathers were college graduates 
was smaller than the percentage of nonveterans for all the "basic groups except 
Adams, where the percentage was the same. A still higher percentage of 
women students 1 fathers were college graduates than were the nonveteran males' 
fathers . 

Amount of education of the father is apparently unrelated to Adjusted 
Average Grade. JOT none of the "basic groups are any of the category mean 
AAG's significantly different from the means of other categories. There is 
a tendency for nonveterans whose fathers were graduated from high school "but 
not from college to underachieve; in ten of the twelve "basic groups the Cate- 
gory B mean AAG is lower than the mean for all questionnaire respondents. 
Differences in education of the father apparently do not help to explain the 
veteran -nonveteran difference in achievement relative to ability. 



Secondary School 

Students were asked, in Item 6 (a), "What kind of secondary school did 
you last attend before entering college?" Choices offered were (A) private 
preparatory school, (B) public high school, and (C) parochial school. Tabu- 
lations of responses made it apparent that in most of the colleges in this 
study, the great majority of students were drawn from the public schools. 
Consequently, this item will be considered only for three colleges Adams, 
Stewart, and Douglas which draw heavily from private secondary schools; 
presentation of the findings will be limited to a discussion of the results 
for these three schools. 

At the three colleges with reasonably large 
private school groups, the proportions in the private 
school category varied from somewhat under a half up 
to about three quarters of the male students. At 



230 ADJUSTMENT TO COLLEGE 



only one of the three colleges was there a smaller 
percentage of veterans than of nonveterans from 
private schools. In two of the three institutions 
the public school graduates did better in relation 
to measures of ability than did students from private 
secondary schools. However, the tendency for public 
school students to overachieve may result from the 
previously noted superiority of students from the 
lowest income families and the likelihood that 
scholarship students axe drawn from this level, 
rather than any difference in preparation provided 
by the two types of secondary school in question. 

At Adams about two thirds, at Stewart about three fourths, and at 
Douglas somewhat less than one half of the male students in the freshman 
group had attended private schools. At Adams and Stewart, the freshman 
veteran group included a greater proportion who had attended private schools 
before entering college than did the freshman nonveteran group; at Douglas, 
this difference was reversed. Among the interrupted veterans and nonveteran 
sophomores at Adams and Stewart, the private schools contributed a smaller 
proportion both of veterans and nonveterans than was true for the freshman 
groups . 

In studying the relationship between type of secondary school attended 
and MG, four groups were used: these included freshmen from Adams, Stewart, 
and Douglas, and sophomores from Adams. Since the number of students in these 
groups who reported attendance at parochial schools is very small, and since 
only one student in all these groups omitted the item, only the results for 
private secondary schools need be considered. 

Since each of the four groups included both veteran and nonveteran 
students, eight comparisons of private school graduates with graduates of 
other types of schools were possible. In seven of the eight comparisons, 
the private school graduates earned lower AAG's, on the average, than the 
public school students. In two of these instances the difference was signifi- 
cant at the 1$ level; these two instances involved the veterans and the non- 
veterans at Adams who entered in 1946. In another case (nonveteran students 
at Stewart), the difference between private and public school graduates was 
significant at the % level. The only subgroup in which the public school 
graduates were found to be inferior in AAfi was the one containing veterans 
at Douglas; this difference was not significant. Considering the over-all 
results, obtaining seven out of eight differences in one direction is sig- 
nificant at only the 10J& level of confidence. These results, then, indicate 
some tendency for the kind of student who prepared for college in private 
schools to underachieve in college. The findings at Adams and Stewart should 
be interpreted in the light of the fact that students who reported a high 
income for the head of their family also tended to be underachievers . The 
fact that consistent differences were found in the two colleges employing 
College Board tests may also be of some importance in interpreting the re- 
sults; possibly the findings may come about because the private schools 
attempt more than public schools to prepare their students for the College 



AGE AND GENERAL BACKGROUND 231 



Board tests and thereby produce test scores which, slightly overestimate the 
student's ability,, Still another possibility is that public school students 
were selected more stringently on qualities other than those used as ability 
measures in this study. 

Although this item could hardly be expected to contribute much to the 
explanation of veteran-nonveteran differences in AAG for colleges in general, 
it should be noted that there is no indication that the type of secondary 
school attended aids in accounting for veteran-nonveteran differences in these 
four groups . 

Evaluation of College Preparation 

Although the student's evaluation of his preparation for college may 
tell more about the student than it does about the school which prepared him, 
such an evaluation may throw some light upon the success of the secondary 
school in preparing its students for college. Item 26 (a) asked, "When you 
first enrolled in this college or university, how well do you feel that you 
were prepared, by virture of your previous education and experience,* for 
getting the most out of your courses?" !Ehe choices offered were (A) very 
well prepared, (B) fairly wen prepared, and (C) poorly prepared. Results 
for this item are given in Figure 30. 

In general the typical veteran believed that his 
college preparation was only fairly adequate; the 
typical nonveteran (male and female) held a slightly 
more favorable view. There is some tendency for higher 
AAG's to go with acre favorable attitudes toward prepara- 
tion,* the trend is clearer for nonveterans than for 
veterans. The difference between veterans and non- 
veterans in AAG seems to be little affected by the 
characteristics involved in this Item* 

A majority of both veterans and nonveterans in the typical college 
group considered themselves "fairly well prepared." Eowever, nonveterans 
showed, a greater tendency than veterans to consider themselves "very well 
prepared," perhaps because of the recency of their secondary school training. 
Women students closely resembled male nonveterans In their estimates of how 
well they were prepared for college. The fact that only about 10 per cent 
of nonveterans considered themselves "poorly prepared" (in the median col- 
lege group) may be taken either as an endorsement of the secondary school 
or as an indication of the confidence of youth. 

Por both veterans and nonveterans there appeared to be a fairly con- 
sistent relationship between AAG and evaluation of previous preparation. 
Nonveterans who reported that their college preparation was poor earned 
lower AAG's than the other nonveterans in eleven out of the twelve basic 
groups; in the twelfth group, they equalled the general average. The 
veterans who reported that their preparation was poor were below the other 



232 



ADJUSTMENT TO COLLEGE 



ra 
P 



P 
co 



100 



75 



50 



















- 
















t> 


k OOOOO 
" 








ooo 


. 


000 
























> oo 




















@ 




I 


000 








oo 


* 


o 






o 


ooo 


eo 








@ 


oo 












OOOOO 



\ Median 
Veterans 
o Nonveterans 



w m mm mm 

Very well Palrly well Poorly- 
prepared prepared prepared 

PER CENT CHOOSING DESIGNATED CATEGOERIES IN EACH COLLEGE 



a 
s 



Q 130 


-P 

I -o 



110 













I 
< 


^ 




i 


> 




\, 


^-^^ 


> 








"\^ 


o Veterans 
o Nonveterau 


as 


Very well Fairly well Poorly Total 
prepared prepared prepared group 



MEAN AAG OF MEDIAN COLLEGE GEOUP FOR SPECIFIED CATEGORIES 



FIGURE 30. EVALUATION OF PREPARATION FOR COLLEGE: ITEM 26 (a) 



AGE AND GENERAL BACKGROUND 233 



veterans in AAG in nine of the twelve groups. The median of the mean AAG's 
is in agreement with this trend. For veterans, the median of the "very well 
prepared" group is almost 1^0 as compared with 125 for the "poorly prepared" 
group. The corresponding figures for nonveterans are about 130 and 115. 
Students who thought themselves very well prepared for college were signifi- 
cantly higher in AAG, at the 1$ level, in two veteran and one nonveteran engi- 
neering school groups . This finding suggests that the association may be 
stronger for students in engineering than for those in liberal arts. 

The usefulness of this item for explaining veteran-nonveteran differences 
in AAG turned out to be negligible when evaluated by the sign test. This was 
true even though., on the average, the veterans were inclined to select some- 
what more often than nonveterans the response which is associated with under- 
achievement . 



Housing 

There is much reason to believe that a student *s college experience is 
influenced in various ways by his living quarters. To determine whether living 
arrangements might be related to AAG, students were asked in Item 30 "Where 
are you living at the present time?" Choices offered included (A) with 
parents or near relatives, (B) college dormitory, (c) fraternity house, (D) 
rooming or boarding house, (!) apartment or house (self -rented or owned) and 
(F) other arrangements. Results for this item are shown only in Appendix 
Table 30, 

While types of living arrangements varied widely 
from college to college, the two predominant cate- 
gories were "living with parents or near relatives" 
and "living in dormitories," Among male students, 
the former was reported more frequently by veterans, 
the latter by nonveterans j women students named both 
more often than either male group. T^pe of housing 
had no marked relation to AAG; however, for veterans, 
renting or owning one ! s own home seemed to be favor- 
ably related to academic achievement, while living 
with parents or relatives was associated with under - 
achievement. For nonveterans, dormitory residents 
tended to earn slightly higher AAG*s than students 
with other housing arrangements. 

The diversity of the sixteen colleges included in this study comes out 
clearly when the replies to this question are examined. For this reason 
material based on this item was used in Chapter III to help describe the col- 
leges. When the basic groups are viewed as a whole, it appears that the most 
frequent arrangements were "living with parents or near relatives" and "living 
in dormitories," These two plans accounted for the majority of men students 
in all of the basic groups except Harris, Central State, and Midwest Tech. 



234 ADJUSTMENT TO COLLKE 



At Harris, the majority of the freshman men were living in fraternity houses; 
at Central State and Midwest Tech the living arrangements vere quite diverse . 
When median groups are compared we find that proportionately more veterans 
than nonveterans lived with relatives or in an apartment or house which they 
rented or owned, and that a greater proportion of nonveterans than of veterans 
lived in college dormitories or in fraternity houses. Women students' living 
arrangements vere not so varied; nine out of ten lived either with relatives 
or in dormitories, more frequently the former. 

General results on the relationship between various types of living 
quarters and AAG are complicated by the fact that some arrangements are 
virtually nonexistent in certain groups. Accordingly, when the number of 
students in a group who report a particular plan dropped below ten,, that 
group was excluded from the comparison. Medians of the mean AAG's were not 
computed for this item. No clear-cut advantage appears for any type of living 
quarters in the twelve basic groups There was some suggestion that living 
with parents or relatives was an unfavorable arrangement for veterans , since 
veterans giving this response made lower AAG's than other veterans in eight 
groups out of ten. For nonveterans, dormitory life seems somewhat favorable, 
since nonveterans living in dormitories were above other nonveterans in AAG 
in five instances, tied in three instances, and below them in none e In the 
eight groups of veterans who owned or rented their house or apartment, superior 
AAG's characterized six groups out of eight with one tie. Although none of 
these trends can be considered statistically significant, each of them seems 
plausible, particularly the last, which fits the hypothesis that greater 
personal responsibility may go with higher" AAG. 

When the sign test was applied to the data to determine whether this 
item might aid in accounting for veteran -nonveteran differences in achieve- 
ment relative to ability, the results were found not to be significant. 

Conclusions 



The findings on the various background factors considered in this 
chapter lead to a number of statements describing various characteristics 
of veteran and nonveteran students. These students typically had never had 
a civilian job on a full-time basis; but the proportion who had held a full- 
time job was considerably greater among the veterans than the nonveterans. 
There was considerable variation among the various colleges, especially for 
the veteran subgroups, with regard to the proportion of freshmen with previous 
work experience. The students typically came from small towns or from 
cities; relatively few reported having lived on farms or in villages of 
less than 2,500 population while attending high school. The head of the 
family was likely to have had an annual income between $2,000 and $4,000 
while they were in high school; there was, of course, marked variation in 
the average income from one college to another and within a particular 
college Veterans tended to report lower family incomes than nonveterans; 
this finding must be discounted somewhat, however, in view of shifts in 
average income during the war years. Almost half of the students reported 
that their fathers had not completed high school; fathers of veterans had, 



AGE AND GENERAL BACKGROUND 235 



on the average, less formal education than fathers of nonveterans * Marked 
variation occurred from college to college in the proportion of students 
whose fathers were college graduates. Except in three of the colleges, the 
great majority of students were products of the public high school, The 
typical veteran considered himself fairly wen prepared for college, while 
the typical nonveteran took a somewhat more favorable view of his prepara- 
tion. 

A number of background items showed some relationship to over achieve- 
ment,, There was a tendency, especially among nonveterans, for high AAG to 
be associated with residence in. a large city while attending high school. 
There was also some tendency for high AAG to characterize students coming 
from relatively low- income families The relationship of age to AAG is 
different for veterans and nonveterans; the older veterans excelled in AAG-, 
while among nonveterans there was a tendency for the younger students to 
be the overachievers . There is some 'indication that private school students 
tended to underachieve in college, although the evidence is insufficient to 
justify a definite conclusion Amount of full-time civilian work experience, 
amount of formal education completed by the father, and type of living ar- 
rangement at college had no apparent relation to AAG. A rather clear-cut 
and statistically significant relationship was found between the nonveterane 1 
evaluation of their preparation and AAGj for veterans, the trend waa in the 
same direction but was less distinct,, 

None of the background characteristics discussed in this chapter can 
be said to help account for the general tendency for veterans to achieve 
higher grades relative to ability than nonveteran students. Nhile veterans 
and nonveterans differ with regard to certain of the characteristics, these 
characteristics are not related to AAG in such a way as to permit the 
interpretation that the veteran-nonveteran difference in AAG would be 
noticeably changed if the two subgroups were alike with respect to these 
characteristics 



236 



Chapter 71 
SOME FACTORS RELATED TO MOTIVATION 



The hypothesis that the superiority of veteran students in college 
grades relative to ability may be accounted for by increased motivation, 
which in turn results from greater maturity or from military service , 
deserves serious consideration. Veteran students may, for example, have 
more definite vocational objectives or greater realization of the import- 
ance of college for advancing their careers. It is obviously impossible 
to measure differences in motivation in any precise way by means of such a 
crude technique as a questionnaire, nevertheless f a number of items were 
included with the hope that any gross differences in motivation would be re- 
vealed. These items have to do with such areas as reasons for going to 
college, vocational plans, plans with regard to acceleration of progress 
through college, and adjustment to the demands of college work. 



Beasons for Going to College 

Item 10 asks, "What would you say were the chief reasons for your 
Doming to college?" Eight reasons were listed, and the student was in- 
structed to indicate the one which best expressed his most important 
reason by 1, and the second and third most important reasons by 2 and 3.. 
3nly the first choice was used in the analysis, "~ 

For purposes of analysis, the following categories were used: (A) 
! I wanted to prepare myself for a better-paying job than I would otherwise 
3e able to get," (B) "A college degree is necessary in order to enter the 
profession I have chosen," (C) "I wanted to increase my general knowledge," 
and (D) other reasons. The other reasons were the remaining choices offered: 
"wanted a chance to enjoy college life," "wanted to make social contacts and 
develop my social skills," "wanted a chance to find out what line of work 
I would be most interested in, " "my family and friends expected me to come, " 
and "coming to college just seemed the logical thing to do," The results 
are shown graphically in Figure 31 

Veterans seemed to be motivated more often than 
nonveterans by a desire to prepare for a better -paying 
jobj nonveterans more often said they wanted training 
for a profession. Women students much more often than 
men said they wanted to increase their general knowl- 
edge, and more often gave other reasons, many of which 
are related to social motives. Male students who went 
to college for necessary professional training or for 
general knowledge earned higher grades in relation to 
ability than those who went for a better-paying job or 
for other reasons; however, veterans who considered 



SOME FACTORS RELATED TO MOTIVATION 



237 



100 















\ Medi* 


in 














Veterans 














o Nonv< 


sterans 























00 














> *0 






000 













o 




o 



















> oooo 













> ooo 


> 90 
















ooo 


@ 


oo 




oo 










eee 




ee 






ooo 




o 




o 


' ** 


000 

^ ooo 


90 


ooo 
oooo 













oo 
oo 





oo 



-p 
G 
0) 



P 
CQ 



o 
-p 



25 



MV MN 
Preparation for 
better -paying Job 



MV" MN 
Necessary profes- 
sional training 



m MN 

General 
^knowledge 



MV MN 
Other 
r eas ons 



PER CETO CHOOSING DESIGNATED CATEGORIES HT EACH COLLEGE 



JL^V 
l^U 














3 














O 




^^ 


- 1 


^ s 











****' 










SP 




^-^ J 


)_ _ 1 


K \ 


( 




130 


< 


^" s 




V N^ 






5 

8 




/ 




X XN 


( 


t 


-P 




X 










3 




X 




v 






^ 120 








T[ 






3 ll ~ 


c 






















o Veterans 


Tin 










o Nonvetera 


is 


Preparation Necessary General Other Total 
for better- professional knowledge reasons group 



MEAN AAG OF MEDIAN COLLEGE GROUP FOE SPECIFIED CATEGORIES 
FIGURE 31- CHIEF SEASONS FOR ATTENDING COLLEGE: ITEM 10 



238 ADJUSTMENT TO COLLECI 



getting a better -paying job tiieir primary reason 
did not underachieve nearly as much as did non- 
veterans "who put that reason first. 

There are noteworthy differences "between veterans and nonveterans 
with respect to reasons for going to college, according to these findings. 
The reason most commonly reported by veterans was preparation for a better- 
paying job. In eleven of the twelve basic groups this reason was given 
more frequently by veterans than by nonveterans. Finding eleven out of 
twelve differences in the same direction would be expected by chance less 
than once in a hundred times. For nonveterans , the most commonly given 
reason was to obtain necessary training for professional work; this reason 
was given by nonveterans more often than by veterans in ten of the twelve 
groups. Possibly the veterans, being older, were less willing to spend 
additional years in professional graduate schools. To increase general 
knowledge was given infrequently by both subgroups.; in the median group 
the percentage is about 15. The "other reasons" were given more often by 
nonveterans in eleven of the twelve subgroups. 

The desire to qualify for better -paying jobs was more popular with 
engineering students than with those enrolled in liberal arts; the top three 
dots in the figure, both in the veteran and nonveteran columns, represent 
the engineering schools. Students at Adams and Stewart gave this reason 
less often than students at other colleges; they tended more often than 
other groups to give "general knowledge" as the chief reason for going to 
college. Women students apparently did not resemble men very closely in 
their motivation for going to college; they far more often gave "general 
knowledge" as their chief reason, and they also gave "other reasons" more 
often than the men. 

Reference to the lower portion of the figure shows that the item bears 
a fairly close relation to Adjusted Average Grade. Students giving pro- 
fessional training or general knowledge as their reasons for attending col- 
lege earned higher grades In relation to ability than do students who 
attended in order to prepare for a better -paying job or for other reasons. 
The differences are not significant for most of the groups, although both 
veterans and nonveterans at Adams who gave professional training as the 
chief reason are significantly higher (at the 1% level) in AAG than other 
students, and those giving "other reasons" are significantly lower. At 
Stewart nonveterans giving better-paying jobs as the reason are significantly 
lower than other nonveterans . In eleven of the twelve nonveteran groups, 
the mean AAG for the" professional training category is higher than the mean 
AAG for students in the remaining categories. The relationship of the item 
categories* to AAG is very similar for veterans and nonveterans, except that 
preparation for a better-paying job is not associated with low AAG for 
veterans to as great an extent as it is for nonveterans. 

The item does not help in accounting for the tendency for veterans to 
earn higher AAG f s than nonveterans, since the veterans are not more numerous 
than nonveterans in item categories which are associated with high AAG's. 



SOME FACTORS RELATED TO MOTIVATION 



Vocational Plans 



Several of the questionnaire items are related to vocational objec- 
tives. These items have to do with the kind of vocational object ive, the 
certainty of this vocational plan, and the importance ascribed to college 
graduation and to college grades in relation to vocational opportunities. 

Vocational Objective. Item 11 is a free -answer item: "What kind of work 
are you planning to do after you finish your studies? (Describe the job 
as specifically as you can.)" Four categories were used in coding the 
responses; the instructions which were prepared for the coders are shown 
in Appendix C3. The four categories may be briefly described as follows; 
(A) the profession named is one requiring graduate study; (B) the profes- 
sion named probably requires a college degree but not necessarily any 
graduate training; (C) other professions or occupations not classifiable 
under (A) or (B) were named; or broad fields such as business, agriculture, 
civil service, or politics were mentioned; or the respondent stated that 
he was not planning to work; and (D) markedly different alternatives were 
being considered or the respondent said he was undecided. The results of 
a study of the accuracy of the coding were given in Chapter II. A summary 
of the Item 11 analysis is shown in Figure 32. 

The distributions of responses for veterans and 
for nonveteran students were very similar, although 
there was a tendency for nonveterans to name occupa- 
tions requiring graduate training more often than 
veterans. Women students named jobs requiring gradu- 
ate work considerably less frequently than did men. 
For both veterans and nonveterans^ higher AAG was 
associated with choice of professions requiring 
greater amounts of educational training. 

The most common type of response, both for veterans and nonveterans, 
was the rather miscellaneous category C. In the median group more than 35 
per cent of both veterans and nonveterans were in this category. Jobs re- 
quiring graduate study and jobs requiring a college degree but not graduate 
study were chosen about equally often; the median percentage in each of these 
categories was about 25. Less than 15 per cent were in the undecided cate- 
gory. There are in general only minor differences between veterans and non- 
veterans with respect to the medians, as indicated by the positions of the 
arrowheads in Figure 32; but nonveterans named occupations requiring gradu- 
ate training somewhat more often than did the veterans. 

Variability among colleges is especially great with respect to per 
cent giving professions requiring graduate training. At one college, Miller, 
almost 65 per cent of the nonveterans are in this category, while in the 
engineering schools the percentage drops to about 10. Women students con-' 
templated professions requiring graduate study far less often than did men. 



240 



ADJUSTMENT TO COLLEGE 



100 



75 







-p 

s 

o 



ifi 



W MN 
Profession 
requiring 
graduate work 















| Mediae 


i 














Veter* 


ms 














o Nonve" 


serans 




O 
















O 

















O 


9 


O 


* 

















, 


ooo 









oo 




O 


* **d 


oooo 








k 




O 










k O9 


ooo 


1 * 


oo 




ooo 






f ** 






> oo 

















** 


O 






[ 




6* 


O 




ooo 






f a 


k 000000 


9 


O 




O 






f 


f OOOO 














i 


oo 



m m 

Profession 
requiring 
college only 



MV MN 
Other 

occupation or 
broad field 



MN 
Undecided 



FEE GEM 1 REPORTING DESIGNATED CAIEGORIES IN EACH COII^GE 



150 

















^ 


^s 

\ 
\ 




4 






^^< 


> 

^ 

^"-^^ 


^ 


( 












* Veterans 
o Nonvetera 


as 


Profession Profession Other Undecided Total 
requiring requiring occupation or group 
graduate work college only broad field 

MEAN AAG OF MEDIAN COUGE GROUP FOR SPECIFIED CATEGORIES 



130 



lao 



no 



FIGURE 32. VOCATIONAL OBJECTIVE: ITEM 11 



SOME FACTORS RELATED TO MOTIVATION 24| 



The relationship of the item to Adjusted Average Grade is shown in 
the lover part of the figure. For both veterans and nonveterans there is 
a tendency for the lines to slope downward to the right. Categories A, 
B, and C form a continuum roughly ordered with respect to amount of train- 
ing required for the occupation, and hi$ier AAG is associated with greater 
amounts of required training. Individuals who are undecided about voca- 
tions tend to earn low AAG s s Differences are highly significant at "Adams 
for all of the first three category means , and significant differences 
(at the 1$ level) are also found for veterans planning careers requiring 
graduate study at Stewart and Midwest Tech. The mean AAG for students 
choosing professions requiring graduate study is higher than the mean AAG 
of other students in ten of the twelve groups for veterans and in nine of 
the twelve groups for nonveterans . The Category C students (other profes- 
sions or occupations, or broad fields) tend to be lower than other students; 
they are lower in mean AAG for eleven groups of veterans and all twelve 
groups of nonveterans. Such consistency of results is highly significant . 
Category C tends to be a "catchall" category, because of the inclusion of 
"broad fields"; if it contained only jobs which did not require college 
training, the difference in mean AAG might be even greater. 

In interpreting these results, the possibility should be considered 
that the relationship between kind of vocational objective and AAG is 
influenced by the grades the students have obtained. Students may have 
modified their vocational plans in a manner which is consistent with the 
college grades they had already earned in their freshman year. 

The item does not help to account for the tendency of veterans to 
earn higher AAG'So Since nonveterans more often possess the characteristic 
which is most strongly associated with overachievement (preferring a job 
requiring graduate study) and veterans more often possess the character- 
istic associated with underachievement (jobs least likely to require col- 
lege training), the item would on the contrary lead one to expect the non- 
veterans -to excel in grades relative to ability. 

It would probably be more realistic to conclude that vocational choice 
poses different problems for veterans than for nonveterans; the sacrifices 
involved in extended professional training would be substantially greater 
for veterans, who are starting college at the age when students usually 
are completing college, than for nonveterans 

Certainty of Vocational Choice . Item 12 asks, "How sure do you feel that 
you will actually do this general kind of work?" The categories used in 
the analysis are the same as the choices as printed in the questionnaire: 
(A) I am almost certain; (B) I probably will, but may do something else, 
and (C) I am not at all sure what I shall do. The results for this item 
are shown graphically in Figure 33 . 

More male veterans than nonveterans were "almost 
certain" of their vocational objectives, although an 
almost equal proportion of each were "not at all sure" 
what they would do. Women students tended to be less 



242 



ADJUSTMENT TO COLLEGE 



CD 
-P 

I 



-P 



O 



100 



75 



& 



T3 
CD 
43 







'25 















^ Median 




o 














o 


@ 








Veterans 


L * 








ooo 






o Nonveterans 





oo 


06 


> OOOO 








* 


\ OOOO 


> ** 


oo 








99 




96 


o 










ooo 





o 





oo 












L @@ 


L OOOO 












^ 90@ 


r OOOO 














oo 




MV MW 
Almost 


MV MN MV MW 
Probable Not at 


certain all sure 



PER CENT CHOOSING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 



S3 130 



110 













( 


, 


>v -^ 






f 


^^ 


""""" - . . 












Veterans 
o Nonveterans 




Almost Probable Not at Total 
certain all sure group 



MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 33 . CERTAINTY OF VOCATIONAL CHOICE : ITEM 12 



SOME FACTORS RELATED TO MOTIVATION 243 



sure than men. Students who were fairly sure of the 
kind of work they would do earned higher AAG's than 
those who were uncertain about their objectives. 

Inspection of the arrowheads indicating medians in figure 33 shows 
that the veterans are "almost certain" slightly more frequently than are 
the nonveterans, and nonveterans fall more frequently in the middle cate- 
gory. About 20 per cent of both veterans and nonveterans are not at all 
sure of what kind of work they will do. This proportion is higher than 
the undecided category on the previous item,, probably because some of the 
undecided students wrote in the names of broad occupational fields in 
response to Item 11. 

Variability among the college groups is not extreme. It is interest- 
ing to discover that the engineering school groups are at or near the bottom 
with respect to percentage of "almost certain" responses. Women students 
tended to be less sure of their vocational objectives than the men. The 
women at Douglas were most extreme in this regard: over half said that they 
were not at all sure what they would do. 

Certainty of vocational aim tends to be associated with higher Adjusted 
Average Grade | the median AAG for students who are certain is nearly ten 
points higher than for those who say they are not at all sure. The mean AAG 
of veterans who are not at all sure is lower than the mean AAG of veterans in 
other categories in eleven of the twelve basic groups; for nonveterans it is 
lower in nine of the twelve groups , with a tie in two cases. 

There is no evidence that the superiority of veterans in grades relative 
to ability would be reduced if veterans and nonveterans were alike with re- 
gard to certainty of vocational objective. While veterans are "certain" 
somewhat more often than nonveterans, it happens that the veterans in the 
"probable" category earn AAG f s which are as high as for those who are certain. 
The net result is that the sign test turns out not to be significant. 

Importance of College Graduation. The next item in the questionnaire, Item 
13, asks, How important is it for you to graduate from college in order to 
do the kind of work you are planning to do?" Two categories were used in 
the analysis of the responses: (A) absolutely necessary, and (B) not abso- 
lutely necessary. The students in the first category are those who checked, 
"I can't do that kind of work unless I have a college degree." The second 
category includes students who said a college degree was not absolutely 
necessary or that it wasn't at all necessary. The results are shown in 
Figure 34. 

The majority of male students considered col- 
lege graduation absolutely necessary to their future 
plans, with a slightly larger proportion of non- 
veterans than veterans expressing this opinion. 
Women students more often considered a college degree 
not absolutely necessary to their planned occupation. 



244 



ADJUSTMENT TO COLLEGE 



100 



I 

03 

I 

I 



75 



50 















O 






* | OOOO 

> F ooo 






p- Median 


9 


OO 


1 @ 





Veterans 
o Nonveterans 


* 





:.... 


oo 
ooo 









oooo 










o 










o 













W MK JflT MN 
Absolutely Not absolutely 
necessary necessary 

HER CENT CHOOSING DESIGNATED CA03SGORIES IN EACB COLLEGE 



1 

*TD 

3 



no 



Veterans 
o Nonveterans 



Absolutely Not absolutely Total 
necessary necessary group 

MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 3^. IMPORTANCE OF COLLEGE GRADUATION FOR VOCATIONAL 



SOME FACTORS RELATED TO MOTIVATION 245 



Students who considered college graduation essen- 
tial obtained higher AAG r s than those who did not. 
The judged Importance of college graduation is 
significantly related to AAG; but the frequencies 
of the questionnaire responses were such as to lead 
to the expectancy that the nonveterans, rather than 
the veterans , would excel in grades relative to 
ability. 

College graduation was considered absolutely necessary by a majority 
of the male students. Women students considered graduation essential much 
less often, although there was one exceptional group: women at Eastern City 
chose the first category slightly more often even than the men. Honveterans 
considered college graduation necessary somewhat more often than did the 
veteran students* In eight of the nine liberal arts colleges among the 
basic groups, the nonveterans chose the "necessary" response more often than 
the veterans; while in engineering schools the veterans and nonveterans were 
more similar with respect to the importance attached to college graduation. 

There is a rather marked tendency for students who regard college gradua- 
tion as essential to be overachievers in comparison to students who do not 
consider graduation absolutely necessary. The difference in mean AAG between 
the two category means amounts to about ten points for both veterans and non- 
veterans, judging from the median groups. In three of the veteran and two 
of the nonveteran groups, the difference in mean AAG between those who thought 
graduation was essential and those who did not was significant at the 1$ level. 
In all twelve of the veteran groups and in eleven of the twelve nonveteran 
groups, those who said graduation was not essential earned a lower mean AAG 
than students in the other category; such a proportion of differences in one 
direction would be expected by chance less than once in a hundred times. 

On the basis of this item, one would expect the nonveterans rather than 
the veterans to be superior in Adjusted Average Grade, since nonveterans 
possess the characteristic associated with overachieveaent to a greater ex- 
tent than do the veterans. Although the results of the sign test are consist- 
ent in direction with this statement, the results are by no means significant. 
Here, again, the possibility that this question meant one thing to veterans 
and another thing to nonveterans must be considered* A nonveteran who con- 
sidered that graduation was not essential may have been low in academic 
inclinations; the veteran, being older, may have adopted the same view as 
a realistic adjustment to the fact that he might be unable to complete his 
college work. 

Importance of College Grades. The next questionnaire item has to do with 
the importance attached to college grades. The statement of Item Ik is, 
"How important do you think college grades will be in relation to the kind 
of opportunities that will be available to you after college?" The three 
choices were (A) very important, (B) fairly important, and (C) hardly 
important at all- The results for the twelve basic groups are presented 
in the usual fashion in Figure 35- 



246 



ADJUSTMENT TO COLLEGE 



$ 
g 



100 



75 



-P 
PI 

O 

I 




























OO 












, ".. 


o 






| Median 











{ 






Veterans 




OO 







^ oooo 






o Nonveterans 


& 


o 


















OO 



















\ ooooo 















' 


o 














.... 


O 











o 














i 


oooo 














1 


1 OOOOOO 














r 







MV MN W JMN MV MN 
Very Fairly Hardly important 


important important at all 



HE CENT CHOOSING DESIGNATED CATEGOEIES IN EACH COLLEGE 



j.pu 

<D ]AO 

3 

s 

0) 

8P 

o 130 

< 

tJ 

(D 
-P 



5 120 
<! 

110 












i 


*\ 

c\ 




^ 


i 




"""-N 


^-^^ 


i < 








1 


e Veterans 
o Nonveterai 


is 


Very Fairly Hardly Important Total 
important important at all group 



MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 35- IMPORTANCE OF COLLEGE GRADES FOR VOCATIONAL OPPORTUNITIES: ITEM Ik 



SOME FACTORS RELATED TO MOTIVATION 247 



Veterans tended to put slightly less stress on 
grades than did male nonveterans; both groups most 
frequently felt college grades would be only fairly 
important in relation to vocational opportunities , 
Somewhat fever women students felt grades would be 
"very important" than the members of either male 
group. Mean AAG's obtained were quite closely re- 
lated to the amount of iioportance the students 
attached to grades 

As indicated by the medians in the upper part of the figure , half of 
the male nonveterans and somewhat more of the veterans judged that grades 
would be only fairly important. About a third of the students felt that 
grades were very important, Nonveterans believed that grades were very 
important slightly more frequently than veterans and that grades were hardly 
important at all somewhat less frequently. !Ehe variability among colleges 
with regard to the frequencies of responses was moderate. Students in the 
engineering colleges tended to report that they felt grades were "hardly 
important at all" less frequently than the students in the liberal arts 
colleges. In general, a somewhat smaller proportion of women students con- 
sidered grades "very important" in relation to future opportunities than 
did either male group 

The tendency for students who considered grades very important to earn 
higher AAG's than other students is rather marked, and the relationship is 
very similar for the veteran and the nonveteran students. The superiority 
in mean AAG of the students choosing "very important" is significant at the 
1$ level in five veteran groups and one nonveteran group. Students in this 
category were superior to other students in ten veteran and eleven nonveteran 
groups o 

Again we find that the difference between veterans and nonveterans would 
not be reduced if they were equated with respect to questionnaire response. 
Veterans and nonveterans are too similar with regard to the importance attached 
to college grades to support the hypothesis that the veteran -nonveteran differ- 
ence in AAG- can be accounted for on the basis of this variable. 

SnT^nary. Male students, both veteran and nonveteran, most commonly reported 
that they were planning to get a job for which college graduation or graduate 
study is essential. Fewer than 15 per cent gave responses which indicated 
that no decision had been made as to the kind of work they planned to do. 
Women students planned to do work requiring graduate study far less often 
than men, and were more often undecided. A somewhat greater proportion of 
veterans were certain of their vocational objectives than was true of non- 
veterans. College graduation was considered necessary by a majority of the 
male students; the male nonveterans considered graduation essential somewhat 
more often than the veterans Women much less frequently thought that 
graduation was essential. Male nonveterans tended to ascribe slightly 
greater importance to college grades than did the male veterans or the 
women v 



248 ADJUSTMENT TO COLLEGE 



High Adjusted Average Grades tended to "be earned "by students who 
planned to enter a profession requiring college or graduate training,, 
who were certain of their vocational objectives , who considered college 
graduation essential for their work, and who "believed that college grades 
were very important in relation to vocational opportunities. The tendency 
for veterans to excel nonveterans in grades relative to ability cannot be 
accounted for on the "basis of any of the items concerned with vocational 
plans. In fact, on two of the items (nature of vocational aim and import- 
ance of college graduation) nonveterans more often than the veterans 
possessed the characteristic which is associated with higher AAG. On the 
"basis of these items, one would expect the nonveterans rather than the 
veterans to excel in grades relative to ability. 

Acceleration of College Progress 

An indication of the feeling of urgency with regard to completing one's 
education and getting on with his career may be given by the response to 
Questionnaire Item 21 This item asks, "Are you planning to take your degree 
in less than the usual amount of time spent (either by attending summer 
sessions or by taking a heavier than normal load of courses)?" The response 
categories used in the analysis were (A) yes, and (B) no. (A few students 
who planned to take more than the normal amount of time and who did not plan 
to graduate were also included in Category B.) The results of the analysis 
are shown in Figure 36. 

Veterans apparently do experience a greater 
feeling of urgency to get about the business of 
earning a living, as is evidenced by the much 
larger proportion planning to finish in less than 
the usual time. Those veterans who plan to acceler- 
ate tend slightly to earn higher grades relative to 
ability than veterans who do not, but among non- 
veterans intention of acceleration is not related 
to AAG. 

It is apparent that a much larger proportion of veteran students planned 
to accelerate their progress through college than was true of the nonveteran 
students. In the median groups, about kO per cent of the veterans and less 
than 10 per cent of the nonveterans planned to take their degrees in less 
than the usual amount of time. Even fewer of the women students planned an 
accelerated program. The variability among colleges was not great for the 
responses of nonveterans; but among veterans there was considerable varia- 
bility. At Miller more than half of the veterans planned to accelerate, and 
the proportion was even greater at Turner, Eastern City, and Central State 
(for the group which entered in 19^5) . On the other hand, at Stewart and 
at the Midwest City College of Engineering fewer than 10 per cent of the 
veterans planned to graduate in less than the normal time . These varia- 
tions are probably the result of different regulations in effect in the 



SOME FACTORS RELATED TO MOTIVATION 



249 



02 
-P 

a 



1 

ca 



-p 

I 



100 



75 










oooo 











> oooo 








o 


000 








o 


o 








>: _ 














\ Indian 











@ "Vetersns 



* 








o Honveterans 





o 

, oooo 
\ ooo 
oooo 








MV MN W MN 


Less than The usual 


the usual time time 



FEE OEM 1 REPORTING DESIGNATED CA1EGC3RIES IN EACH COLLEGE 



J-^w 

o 140 
I 

| 130 










< 


' " ~ 





i 


< 


>~ , 1 


. < 




-p 

5 120 














Yeterans 




110 






o Nonveterai 


is 



Less than The usual Total 
the usual time time group 

MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 36, LENGTH OF TIME PLACED FOR EARNING DEGREE: ITEM 21 



250 ADJUSTMENT TO COLLEGE 



various colleges* Probably the median percentage of veterans planning to 

accelerate would be even higher if university regulations at all institu- 
tions freely permitted acceleration. 

For nonveterans there is no consistent tendency for plan to complete 
college in less than the normal time to be associated "with higher grades 
relative to ability. The mean AAG (for the median groups) is almost ex- 
actly the same for both of the item categories. .Among veterans,, however, 
there is a slight but generally nonsignificant tendency for students who 
wish to accelerate to earn higher AAG's than students who do not wish to 
accelerate . Although veterans differ considerably from nonveterans with 
respect to the intention to complete college quickly, the relationship of 
the item to AAG is too weak to permit us to accept the hypothesis that the 
veteran superiority is due to any characteristic assessed by this question- 
naire item. 



Adjustment to the Demands of College Study 



It is not unreasonable to suppose that a measure of motivation for 
academic work in college might be obtained by finding out something about 
the student's work habits, how difficult he finds it to keep up in his 
work and to do his assignments on time. Three items were included in the 
questionnaire which it was hoped would at least reveal any gross differences 
which might ezist between veterans and nonveterans with regard to this aspect 
of his adjustment to the demands of college. 

The Difficulty of College Work. Item 20 asks^ "Have you found it more or 
less difficult to keep up in your work this term than you had expected it 
to be?" Three response categories were used: (A) more difficult, (B) 
about as expected, and (c) less difficult. These categories were formed 
by combining certain of the five choices as given in the questionnaire. 
The results for the twelve basic groups are shown in Figure 37 

Male veterans and nonveterans are quite similar 
in their judgments of the difficulty of keeping up in 
college work. Almost half of both groups found it 
more difficult than they had expected; relatively few 
thought it less difficult. The opinions of women stu- 
dents agree closely to those of men in this regard. 
The relationship of Adjusted Average Grades to judged 
difficulty of college work is highly significant, those 
feeling the work more difficult than expected obtaining 
low AAG's, those feeling it less difficult obtaining 
high AAG's. This, of course, may result from rational- 
ization of known achievement* 

Almost half of the students reported that they found it more difficult 
to keep up in their college work than they had expected, while only about 
15 per cent thought that it was less difficult. The differences between 



SOME FACTORS RELATED TO MOTIVATION 



251 



I 



-p 



o 

I 



100 



75 



50 



















o 










a 












\ Median 





ooo 










Veterans 




k 00 





oo 






o Nonveterans 




oooo 


> &. 


ooo 











o 


* 


> ooo 











o 


0* 


oooo 



















O 

t oooooo 












k OA 


oooo 












-3 


o 




MV m m m m m 


More difficult About as Less difficult 


than expected expected than expected 



EEE CENT CHOOSING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 



140 



g 130 



g 120 



110 









i 






> 
/ 


t 


< 




i 










< 


/ 




o Veterans 
o Nonveterans 


3 


More difficult About as Less difficult Total 
than expected expected than expected group 

MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGOEIES 



FIGURE 37, DIFFICULTY OF COLLEGE WORK: ITEM 20 



ADJUSTMENT TO COLLEGE 



reterans and nonveterans in judgments about the difficulty of college work 
are very slight, and women students appear to be quite stmilax to men in 
this regard. Insofar as judgments about the experienced difficulty of col- 
lege work reflect interest or motivation, there is no indication that 
veterans are more strongly motivated than the nonveterans. 

The judgments about the difficulty of college work are quite closely 
related to Adjusted Average Grades. Students who reported that their work 
was less difficult than they had expected earned AAG's which on the average 
were nearly 20 points higher than the AAG's pf students who felt that the 
work was more difficult than they had expected. The relationship is highly 
sigaif icant . In six of the twelve veteran groups and five of the twelve non- 
veteran groups, students who said "more difficult" earned significantly 
lower mean AAG's (at the 1% level) than students choosing other responses. 
In all twelve groups, both for veterans and nonveterans, the mean AAG of the 
students reporting "more difficult" was lower than the mean AAG for students 
giving other responses. Such consistency in the direction of the differ- 
ences would be expected by chance less than once in a hundred times. 

In considering how this finding should be interpreted, we again must 
remember that the questionnaires were filled out after the students had con- 
siderable knowledge of their academic success as measured by grades, and it 
is quite possible that the relationship between item and AAG was enhanced 
by a tendency to rationalize their grades on the basis of the difficulty of 
the work. 

At any rate, it is clear that judgments about the difficulty of college 
work, whether they are interpreted as rationalizations or as evidence of 
strength of interest and motivation, do not provide an explanation of the 
higher grades relative to ability which are earned by veteran students. 
Although the item is significantly related to AAG, the veterans and non- 
veterans are too similar with respect to the proportion choosing each cate- 
gory to permit the interpretation that the characteristic measured by the 
item accounts for the veteran-nonveteran difference in AAG. 

Effort. Item 28 attacks the problem of assessing motivation in a very 
direct .manner; the question is, "In general, would you say you usually 
exert strong effort to do good work in your courses, or do you tend to 
do just enough to get by?" Only two categories were employed in the 
analysis: (A) usually exert strong effort, and (B) usually do not exert 
strong effort- Ibis second category included students who checked either 
"I work fairly hard in some courses, not so hard in others" or "I usually 
tend to do just enough work to get by with fair grades." The results are 
shown in Figure 38* 

There is close agreement among male veterans, 
male nonveterans and female students in the amount 
of effort claimed; about one fourth of each group 
said they usually exerted strong effort. The Ad- 
justed Average Grades are significantly higher for 



SOME FACTORS R!LATEI> TO MOTIVATION 



253 





1 

CQ 



100 



T5 



50 


















... 











m 


L oooo 








,00^0 


ooooo 


















c 




| Median 
Veterans 
o Nonveterans 














00 








^ M 


0000 








* 


ooooo 









o 








W MH" MV MM 


Usually exert Usually do not 
strong effort exert strong effort 



PER GEM? CW3QSIHG DESIGEMED CATEGORIES IK EACH COLLEGE 



J-^v 
o> 140 

OT 

s 


I 


\\ 








x \\ x 






Adjusted Average 

ru LO 
o o 




\ ^\ 


< 






\ VN< 
\ 

\ 

\1 










Veterans 




110 






o Nonveterans 

1 



Total 
group 

MEAN AAG OF MEDIANT COLLEGE GROUP FOR SPECIFIED CATEGORIES 



Usually exert Usually do not 
strong effort exert strong effort 



FIGURE 38, EVALUATION OF EFFORT EXERTED UT DOING COLLEGE 



254 ADJUSTMENT TO COLLEGE 



students who claimed they exerted strong effort than 
for those who did not. As in the previous item con- 
cerning difficulty of work load, these may "be after- 

the-fact rationalizations. 

Again Ye find that veterans and nonveterans are strikingly similar 
with regard to proportions choosing the two categories. About one fourth 
of the students claimed that they usually exerted strong effort. Women 
students were Tery similar to the men in this respect. The twelve "basic 
groups tended to be rather similar to one another; the highest percentage 
of students claiming strong effort was about kO y at Midwest City, and the 
lowest (under 20 per cent) at Evans y Central State and Littletown State. 
Engineering schools do not appear to "be particularly high in the propor- 
tion of students who say they usually work hard. 

The relationship of amount of effort to Adjusted Average Grade is 
marked; it is similar to the preceding item with regard to significance 
tests. Again the possibility that rationalization is involved must not 
be overlooked, however; students with low grades may have tended to feel 
that they could have earned higher grades if they had really tried. 

Because of the marked similarity between veterans and nonveterans in 
proportion giving each response, the superiority of the veterans in AAG 
cannot be attributed to amount of effort as assessed by this questionnaire 

item. 

.Keeping Up -to -Pat e . Another questionnaire item designed to get at differ- 
ences in motivation is Item 29, which asks, "In general, how well do you 
keep up-to-date in your study assignments?" The three categories used in 
the analysis are (A) keep ahead, (B) up-to-date, and (C) behind. This item 
represents another rather direct approach to assessing motivation, but puts 
the judgment on a slightly more objective basis than the preceding item. 
Since the analysis of this item gives results which are in general very 
similar to those of the two preceding items, no figure is presented; the de- 
tailed results are shown in Appendix Table 29. 

Again both male veterans and nonveterans showed 
a similar pattern of response; about three fourths 
said they usually kept up-to-date in their study assign- 
ments. An even higher proportion of women students fell 
into this category. The relatively small proportion of 
students who generally kept ahead in their studies 
earned significantly higher AAG t s than those who fell 
behind. This item appears to aid slightly in accounting 
for veteran-nonveteran differences in AAG. 

Approximately three fourths of the male students in the median group 
gave the second response they claimed that they usually got their assign- 
ments done on time. About 15 per cent were so eager that they completed 
assignments before they were due, and the remaining 10 per cent were 



SOME FACTORS RELATED TO MOTIVATION 255 



usually late getting assignments done,, according to the questionnaire 
responses. There is a slightly smaller proportion of the veterans who 
would admit lateness: veterans were less likely to admit "being laggards 
in eleven of the twelve groups . Somewhat more than tnree fourths of 
the women students reported that they usually got their assignments done 
on time. Variations among the colleges are again not extreme, and there is 
no tendency for engineering students to "be characterized "by greater or less 
promptness in doing assignments than liberal arts students . 

The relation of "keeping up-to-date" to Adjusted Average Grade is 
again highly significant. Students "who claim to complete assignments "before 
they are due earn AAG f s which are on the average about 30 points higher than 
those earned "by students "who are usually "behind in their work* 

In 12 of the Ik subgroups which showed poorer-than-average AAG for 
"both veterans and nonveterans, the nonveterans were more likely than veter- 
ans to "be represented; this tendency is significant at the 5$ level. From 
thiSj it would seem that the veterans 1 advantage would have been reduced 
if veterans and nonveterans had "been equally prompt in meeting assignments. 

Summary. The three items discussed in this section,, which presumably are 
related to such factors as interest and motivation, show that veteran and 
nonveteran male students are strikingly similar. Almost half of the student 
felt that it was more difficult to keep up in their work than they had ex- 
pected. Only about a quarter of the students claimed that they usually ex- 
erted "strong effort" in their work. A"bout three fourths said that they 
usually got their assignments done on time, the remainder "being either ahead 
of time or "behind time. Women students do not differ markedly from the men, 
although a somewhat larger proportion of them were in the middle category 
with respect to keeping up-to-date on assignments. 

All three items difficulty in keeping up in college work, amount of 
effort exerted, and keeping up-to-date in study assignments are signifi- 
cantly related to Adjusted Average Grade. The relationship may, however, 
"be enhanced by the tendency of students to rationalize, since they knew 
their first-term grades, at least, at the time they filled out the question- 
naire. Only one of the items appears to help in accounting for the veteran 
superiority in AAG? in all three of these items veterans and nonveterans are 
quite similar with regard to the proportions choosing the various response 
categories. 

Conclusions 



seems to be some difference between veterans and nonveterans 
with regard to motives for attending college. The reason for going to 
college most often given by nonveterans was to get necessary training for 
entering a profession; veterans most often said they wished to prepare 
themselves for a better-paying job. Possibly one reason for the difference 
is that the veteran students, being older, were not willing to spend addi- 



256 ADJUSTMENT TO COLLEGE 



tional years in professional graduate schools. Women students differed 
considerably from the men; they much more often said that they wanted to 
increase general knowledge, or they gave other reasons which are related 
to social motives. 

The analysis of a number of items having to do with vocational plans 
showed only slight differences between veterans and nonveterans. Students 
in both groups were typically planning to get a job for which college gradu- 
ation or graduate work is essential, and fewer than 15 per cent indicated 
that they had made no decision as to the kind of work they would do. 
Veterans tended to express certainty as to their vocational choice somewhat 
more often than nonveterans. Hbnveterans more often considered college 
graduation essential for their vocation and tended to ascribe greater 
importance to college grades than the veterans*, Fewer women students than 
men were planning to go into a profession requiring graduate work; women 
were more likely than men to be undecided about a vocation. They tended 
to ascribe less importance to college graduation than did the male students. 
The sex differences with regard to items pertaining to vocational plans do 
not seem surprising in the light of the different roles ordinarily played 
by women in our society. 

^be greater feeling of urgency on the part of the veteran student is 
shown by the responses to a question about acceleration of the college 
program. A much larger proportion of veterans than of nonveterans were 
planning to graduate in less than the usual amount of time. 

Veteran and nonveteran students were strikingly similar in their 
responses to a series of items designed to investigate how the students 
are adjusting to the demands of college life. Almost half of the students 
felt that it was more difficult to keep up in their college work than they 
had expected* Only about a quarter claimed that they usually exerted 
"strong effort" in their work, although about three fourths said that they 
usually got their work done on time. Women students did not differ markedly 
from men in their replies to these questions . 

Those students tended to earn higher grades in relation to ability who 
went to college for necessary professional training or general knowledge, 
planned to enter a profession requiring graduate training, were certain 
of their vocational choice, considered college graduation essential for 
their future work, and believed that college grades were very important in 
relation to vocational opportunities. Many of these trends seem to indi- 
cate a realistic view of the situation on the part of the students, in view 
of the importance of the undergraduate record for admission to a graduate 
or professional school. Planning to accelerate progress through college is 
slightly associated with higher. Ad jus ted Average Grade for veterans but not 
for nonveterans. The three items dealing with adjustment to the demands of 
college the experience of having difficulty in keeping up in college work, 
amount of effort, and keeping up to date on assignments are all signifi- 
cantly related to Adjusted Average Grade; but it appears likely that the 
relationship may be enhanced by a tendency on the part of the students to 
rationalize ("Tes, my grades are low, but I didn't really try very hard"). 



FACTORS RELATED TO MOTIVATION 257 



The superiority of veterans cannot "be explained on the "basis of the 
aspects of academic motivation dealt vith in this chapter, even though 
these Items vere rather clearly related to AAG. The veterans and nonveterans 
vere very similar vith respect "to the proportions choosing the various 
response categories to many of the items. On tvo of the Items (those deal- 
ing vith the kind of vocational objective and the importance of college 
graduation) the nonveteraas possessed the characteristic associated vith 
high AAG somewhat more often than the veterans. Only one item shoved any 
clear tendency for nonveterans to choose the responses associated vith lev 
AAG more often than did veterans. On this item, vhich dealt vith keeping 
up-to-date on study assignments, the differences In proportions "betveen 
veterans and nonveterans vere small; the tendency vas sufficiently con- 
sistent, hovever, to be significant at the 5$ level. 



258 



Chapter VII 
THE WORRIES OF COLLEGE STUDENTS 



Another hypothesis which should be tested is that the tendency for 
veteran students to achieve higher grades in relation to ability than non- 
veteran students is due to differences in such characteristics as emotional 
stability, anxiety, or feelings of insecurity. "When the veterans began re- 
turning to college, concern was expressed by some college administrators 
with regard to the psychological adjustment of the veteran student to col- 
lege life. Although such concern "was later felt to be unjustified, there 
might still be a difference between veterans and nonveterans in general 
quality of adjustment. "While it is usually assumed that well-adjusted stu- 
dents earn better grades than poorly-adjusted students, such a relationship 
does not necessarily exist for all kinds or sources of worry. It therefore 
seemed worth-while to find out, so far as is possible by means of a question- 
naire, how veterans and nonveterans compare in their tendency to worry about 
various types of problems, to study the relationship between worry and Ad- 
justed Average Grade, and to find out if the tendency for veterans to excel 
nonveterans in Adjusted Average Grade is due to the more frequent possession 
by veterans of worry-characteristics which are associated with higher AAG. 

Sixteen questionnaire items deal with worries Item 39 is & general 
question, "Do you sometimes feel worried and anxious or upset?" which was 
intended to reveal gross differences in neurotic tendencies. Hie next item, 
Item 40, is really thirteen related items, each of which is intended to show 
the extent to which students are worried or anxious about some particular 
problem area* There is also one open-end item, inquiring about sources of 
worry not mentioned in the questionnaire. The last of the items discussed 
in this chapter requires the student to judge to what extent any of his prob- 
lems has interfered with college work* 



The Frequency of Worry 

The question (Item 39) w & s "Do you sometimes feel worried and anxious 
or upset?" and the answers were (A) yes, frequently, (B) occasionally, and 
(C) seldom or never. Figure 39 shows the distribution of the response 
frequencies for the twelve basic groups and the relation of the item to 
Adjusted Average Grade. 

The analysis of this very general question 
about feelings of worry and anxiety shows that 
veterans and nonveterans were essentially alike in 
their answers. Women were more likely to report 
that they worried frequently than were men students. 
There is a tendency for greater amount of worry to 
be associated with lower Adjusted Average Grade; 



WORRIES OF COLLEGE STUDENTS 



259 







.", J 
\ \ " 


i 




p Median 








Veterans 


, > 




j 






o Nonvet 


srans 






.. 


oooo 












- MM** 


t oooo 






50 









ooo 












o 


































o 


25 





oo 








ooo 


i *@ 


\ ooooo 








ooo 




" 


oo 









ooo 


n 





ooo 








o 



m M? MS 

Frequently Occasionally Seldom 

"worry "worry "worry 

PER GEM? CHOOSING DESIG1M15D CASSGOBIES IN EACH COLLEGE 



s 
s 



43 

CQ 



130 



120 



110 



* Veterans 
o Uonveterans 

1 



Frequently Occasionally Seldom Total 

worry worry -worry group 

MEAH AAG OF MEDIAH COLIJBX3E GROUP FOR SPECIFIED CATEGORIES 



FIGURE 39. 1EHDENCY TO WORRY; ITEM 39 



260 ADJUSTMENT TO COLLEGI 



this tendency is significant for veteran students. 
The superiority of veterans in AAG is, however , 
not due to a difference "between veterans and non- 
veterans in amount of worry. 

Most students reported that they felt worried and anxious occasionally; 
in the median group almost 60 per cent of the students, both veteran and non- 
veteran, fall into this category. About 20 per cent said they "worried fre- 
quently, and about 25 per cent seldom or never; there is a very slight 
tendency, as shown by the arrowheads in Figure 39 which indicate median 
values , for the veterans to worry less than the nonveterans . Women worry 
more than men, according to their responses to this questionnaire item; at 
eight of the nine colleges where women's questionnaire responses were tabu- 
lated, the women chose the "yes, frequently" response more often than male 
nonveterans . finding eight -out of nine differences in one direction would 
be expected by chance less than five times in a hundred. 

The lower part of the figure shows that there is a tendency for those 
students who worry most to earn the lowest grades relative to ability. 
The difference between the median values of the mean AAG for the "worriers" 
and those who worry least is about 10. The relationship appears to be 
slightly- greater for veterans than for nonveterans . The mean AAG for the 
"frequent" worriers is lower than for other students in eleven of the twelve 
basic groups, when veterans only are considered; this proportion of the dif- 
ferences would occur by chance less than once in a hundred times. At Adams 
the veterans who checked "yes, frequently" are significantly lower (at the 
1$ level) than other veterans, and those who checked "seldom or never" are 
significantly higher. At Stewart, also, the veterans in the first category 
are significantly low. 

Tendency to feel worried and anxious, as measured by this item, cannot 
account for the superiority of veteran students, since veterans and nonveterans 
are very similar with respect to the proportions choosing each category of 
this questionnaire item. 



What Students Worry About 



The item previously discussed deals with worry and anxiety in a very 
general way. In the various parts of Item ^-0 the students were asked to 
report on tendencies to worry about specific problem areas, in order to 
find out what are the major problem areas and to find out if the relation 
of AAG to worry depends at all upon the kind of worry. The item is as 
follows: "Below are listed some sources of worry and anxiety which seem 
to be bothering a good many students at the present time. For each problem 
check the appropriate category to show how much you have been bothered by 
the problem during this term." The three categories provided were 
bothered very much, bothered some, and little or not at all. The results 
of the analysis of Items to(a) to ^-O(m) are presented below. The order in 



WORRIES OF COLLEGE STUDENTS 



which these sub-Items are discussed is the order of their izgioQrtazice , for 
male ponTeterans 9 as sources of worry. (The ranking Is "based on the median 
of the percentages for the tweiTe "basic groups.) 

Cmcentrat ion . The source of worry mentioned In Item ItO(g) Is "being unable 
to concentrate," The results of the analysis are shewn In Figure kO. 

Yeterans -were "bothered to a greater extent "by 
Inability to concentrate than were nonTeteran stu*- 
dents , according to these cpie st loimaire findings* 
Women if ere also more concerned about concentration 
than nonreterans. Tendency to worry about concentra*- 
tlon "bears a marked relationship to Adjusted Average 
Grade: students who "worried least earned the highest 
Adjusted Arerage Grades. This Item also showed a con- 
sistent tendency for nonTeterans to give responses 
associated with higher Adjusted Average Grade more 
often than veterans did. 

The general pattern of results for this Item Is rather similar 9 so far 
as percentages are concerned, to that for Item 39 ? the general Item on worry 
and anxiety. About 55 per cent of the students, Teterans and nonveterans 
alike, fall In -the middle 'Bothered some" category, as shown "by the arrow- 
heads which Indicate the median values. Veterans, howerer, tend to be 
"bothered Tery much" somewhat more often and "little or not at all" slightly 
less often than the nonTeterans, In eleren of the twelye basic groups a 
larger proportion of veterans than nonyeterans said they were "bothered rery 
much, "by being unable to concentrate, omen tend to worry about concentra- 
tion somewhat more than the male 



The tendency to worry about "being unable to concentrate "bears a close 
relationship to AAG, as is shown in the lower portion of Figure 1*0. The 
difference In median miues of mean AAG between those who were "bothered 
rery much and those who were "bothered little amounts to more than 25 points 
for both Teteran and nonreteran students. The direction of the relation^ 
ship Is the expected one ouch worry Is associated with low AAG. Veterans 
who were "bothered Try much are significantly lower in laean AAG (at the 1> 
10T61) than other Teterans in nine of the twelve "basic groups, and those 
bothered little are significantly higher In mean AAG in eight of the 
twelTe groups. The nuriber of Instances of significant differences is 
smaller for nonTeterans, Jor "both Teterans and nonveterans, students in 
the "little or not at all w category earned higher AAG T s than other students 
In mil ttrelTe groups. The relationship may of course be enhanced by the 
fact that the students knew their grades at the time they responded to the 
C[ue st lozmaire , 

Among 13 responses which were associated with better-than-aTezrage 
AAG for "both Teterans and nonreterans, nonTeterans were more likely than 
Teterans to gire the preferred response in ten comparisons, with two ties? 
this result is significant at the 556 leTel. Thus, If Teterans and non- 
Teterans had "been equal in worry about inability to concentrate, the adTan- 
tage of the Teterans in AAG would presumably hare "been enhanced- 



262 



ADJUSTMENT TO COLLEGE 



100 



m 
-P 



75 



50 



s 





















o 













oo 

rt 






* 




I a I OOOOOO 










fr 


00 
















ooo 










@ 


i ooo 


k $* 









@@ 


r oo 





ooo 









oo 


* 


> ooooooo 






0^> 


oo 





o 










W OT MV MET. W MM 
Bothered Bothered Bothered 


very much some little or none 



Median 
Veterans 



PER CMS CHOOSHG DESIGNATED CATEGORIES IN EACH COII^GE 



-L^V 

<D 140 

I 

o 

jf 

s 130 

<D 
p 
m 

120 

no 






/ 








J 
X 


x>H 


i 






X 
X X S 


/ x/ 






i 
( 


/ /' 

/ 
S 




Veterans 
o UoBveteranf 


5 


Bothered Bothered Bothered Total 
very much some little or none group 



AAG OF MEDIAH COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGT2RE 



WOKBY AND A1CXIETY" ABOXJT CONCEIJTBATION: ITEM 



WORRIES OF COLLEGE STyDENTS 263 



Getting Accustomed to College Study . The second most common source of worry 
among nonveteran males was "getting accustomed to college study. " Item 
kO(f} ? like the preceding one, emphasizes a problem of academic adjustment. 
The results of the analysis are shown in Figure Ul. 

Male veterans, male nonveterans, and female stu- 
dents were very similar in amount of worry about 
getting accustomed to college study; two out of five 
students were bothered little or none by this problem. 
Those bothered least earned significantly higher Ad- 
justed Average Grades than those bothered very much. 

Veteran and nonveteran students were very similar in their tendencies 
to worry about getting accustomed to college study. About 15 per cent re- 
ported that they were bothered very much, 45 per cent bothered some, and 40 
per cent little or not at all. Women students were typically very similar 
to men in this respect, and, there was relatively little variability among 
colleges with respect to the proportion of students in each category. 

The relationship of this worry item to AAG is even more striking than 
that of the preceding one. The difference between the median values of mean 
AAG for the two extreme item categories is almost 35 I* 1 "fcbe case of all 
twelve groups, for both veterans and nonveterans, those bothered very much 
were lower in mean AAG and those bothered little were higher in mean AAG 
than the remaining students in the same college groups. Thirty-one of these 
kS differences were significant at the 1$ level. 

Again it is found that the factor assessed by the item does not help 
to account for the superiority of veterans in mean AAG; although amount of 
worry is closely related to AAG, veterans and nonveterans are very similar 
with respect to amount of concern expressed about getting accustomed to col- 
lege study. 

Deciding What Course of Study to Follow. A pertinent finding for persons 
interested in student adjustment is the fact that when the sources of worry 
are put in rank-order (based on the responses of nonveteran students), the 
top three sources of worry are primarily concerned with academic adjustment. 
"Being unable to concentrate" and "getting accustomed to college study" were 
the most common reasons for worry, and the third is "trying to decide what 
course of study to follow/' The results of the analysis of Item 40(k) are 
shown in Figure 42. 

Yeterans appear to have worried somewhat less 
about choosing a course of study than did nonveterans. 
Women students were more apt to be worried about this 
problem than male students. There is a slight but 
consistent tendency for superior AAG to be associated 
with lack of worry about choosing a course of study. 
This item, may account to a slight extent for veteran 
superiority in mean AAG, although the sign test results 
are not significant. 



264 



ADJUSTMENT TO COLLEGE 



100 





! 

CQ 



-P 



O 



75 



50 



'H 
<D 
* 25 




























o 






o 








@ 


00 







o 











> 0000 




> 


00 








t><D@ 




ooo 







f- 

:.- 


k 

r ooo 

00 


* 



































eeo 


oooo 














' ** 


. oooooo 

































MT MN 
Bothered 
little or none 



Median 
Veterans 

o Nonveterans 



w m m 

Bothered Bothered 
very much some 

PER CENT CHOOSING DOSSIGMTOD CATEGORIES IF EACH COLLEGE 



150 



o 130 
? 

rrf 12 

I 

S no 

100 






; 

s t 


i 








X X 


4 






X s 


X 

X 
y 


' 




4 


< X / ' x 








4 


y 




Veterans 
o Nonveterans 




Bothered Bothered Bothered Total 
very much some little or none group 



MEAN AAG OF MEDIAE COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE lH. WORRY AHD A3JXIETY ABOUT GETTING ACCUSTOMED TO COLLEGE 
STDDY: ITEM 



WORRIES OF COLLEGE STUDENTS 



265 



100 



m 

I 



p 

g 

o 


















1 Median 
Veterans 
o Nonveterans 












% 
% 

* 
^@ 


o 

o 

ooo 












** 


> 0000 






%% 









ooo 










oooooo 














&&4MMMI 
* 


ooooo 

























%&@ 


ooooooo 





o 










* 


ooo 














r 

















mm mm m MM 

Bothered Bothered Bothered 
very much some little or none 



PER CMT CTOOSH3G ZESIGHA.3ED GA3EGOBIES HJ EACH COLLEGE 



i^U 

o> 140 
S 

g 130 














^ 


^^ 








^ 


^^^^ 






Adjusted 





^ ^- 


^*" 








^^^ 








110 


e 


f 




Veterans 
o Nomreterans 




Bothered Bothered Bothered Total 
very much some little or none group 



MEAN AAG OF MEDIAN COLLEGE GROUP ?OR SPECIFIED CATEGORIES 



FIGDEE 42x WORRY AMD AWTTKW ABOUT DECIDING WHAT COURSE OF STUDY TO 
FOLLOW: ITEM 



266 ADJUSTMENT TO COLLEGE 



Again the similarity of veterans and nonveterans in amount of worry is 
rather marked, although there is a slight tendency for the nonveterans to 
worry somewhat more than veterans about what course of study to follow. The 
difference is greatest on the "little or not at all" category, where the 
median value for veterans is about 55 per cent and for nonveterans between 
^5 and 50 per cent. In ten of the twelve groups fewer nonveterans than 
veterans checked this response. Only about 15 per cent of the students 
reported that they were bothered very much about deciding what course of 
study to follow. These results are consistent with the previous finding 
that more nonveterans than veterans are uncertain as to the kind of work 
they will do. Engineering school students apparently worry less about 
What course of study to follow than liberal arts students, presumably be- 
cause they have already committed themselves to a course of study within 
which there is less freedom of choice, omen students, who reported much 
more often than men that they were undecided about vocational choice, worry 
more than men about what course of study to follow. A smaller percentage 
of women reported little or no worry than did male nonveterans in eight out 
of nine of the groups where comparisons can he made. 

Tendency to worry about course of study is associated with lower Ad- 
justed Average Grade, but the relationship is less marked than for the two 
kinds of academic worries previously discussed. The difference between the 
median values of mean MG for the extreme categories of the item is less 
than 15 points. lonveteran students who reported little or no worry were 
superior in mean AAG to other nonveterans in all twelve basic groups s 
veteran students checking this category were superior in eleven of the twelve 
groups* However, only three of these twenty-four differences are significant 
at the 1$ level. The hypothesis that the relationship is merely a matter 
of students f rationalizing grades earned seems less plausible for this item 
than for certain others which have been discussed. The fact that the re- 
sults are consistent with those found for certainty of vocational choice 
tends to confirm the findings for both items. 

On the whole, since veterans are slightly less likely to worry about 
what course of study to follow, and since freedom from worry about choice 
of study shows some association with Adjusted Average Grade, it seems pos- 
sible that this item would aid in accounting for veteran-nonveteran differ- 
ences. Application of the sign test does show a tendency in this direct ion; 
the tendency, however, is by no means consistent enough to be statistically 
significant. The results for this item cannot be considered to help in 
accounting for the veteran superiority in Adjusted Average Grade. 

Making Ends Meet . The source of worry included as Item ^O(a) was "making 
ends meet financially." For this and the succeeding sources of worry, only 
two categories were used in the analysis: (A) bothered some, or bothered 
very much, and (B) little or not at all. The combination of the "some" 
and "very much" responses was necessitated by the fact that the more ex- 



WORRIES OF COLLEGE STUDENTS 267 



treme response was infrequently chosen by the students. The results of the 
analysis for Item 40(a) are shown In Figure 43. 

A considerably greater proportion of male veteran 
students than of male nonveterans were bothered about 
finances; women students were least concerned about 
making ends meet financially . Ehis item tends only 
slightly to be related to AAG; students who worried 
earned somewhat lower AAfi r s than those not bothered 
about making ends meet. 

Financial worries y which for nozsveterans were fourth in order of import- 
ance, ranked second for veteran students. Ihis is the only one of the sources 
of worry studied where a striking difference between veteran and nonveteran 
students was found. Die median value of the percentage of veterans who were 
bothered souse or very much was about 65 > while for nonveterans the median per- 
centage was about 45- Veteran students chose the "bothered some or very much" 
category more often than nonveterans in all of the basic groups except Stewart, 
where the proportions were the same. There was considerable variability among 
the colleges, however, with respect to the proportion of veteran students who 
reported being bothered by financial considerations* Yeterans were most con- 
cerned about finances at Evans and Little town State, where about 75 per cent 
reported being bothered some or very much about making ends meet. The smallest 
proportion in this category was found at Stewart, where only about ko per cent 
were bothered. Bie variability is much less when only nonveterans axe con- 
sidered . 

Superficially, it might appear strange that students who are given fi- 
nancial assistance through the educational provisions of the GI Bill worry 
about money more than those who are not given assistance. The reason for 
the paradox is probably related to the fact that, as has previously been 
reported, veterans come more often from families whose economic status would 
preclude college attendance without some financial assistance. Apparently 
the educational provisions of the GI Bill have encouraged a number of 
veteran students to enter college, even though the amount of financial 
assistance is not sufficient to enable them to face with confidence the 
financial problems entailed. 

Making ends meet financially is one of the few sources of worry which 
concern women students less than men. In *vH of the nine groups where 
comparisons can be made, women students indicated that they were bothered 
"some or very much" less frequently than male nonveterans. 

Biere is a slight and generally insignificant tendency for worry about 
finances to be associated with lower grades relative to ability. Veteran 
students who worry "little or not at all" are higher in mean AAG than those 
who are bothered "some or very much" in, eight of the twelve groups, and non- 
veterans are higher in nine of the groups* 

In spite of the marked difference in proportions between veterans and 
nonveterans reporting financial worries, the Weak association of this worry 



268 



ADJUSTMENT TO COLLEGE 



ca 
43 



1 

CQ 



P 

I 



100 



75 














t@@0> 
:. 








000 

> 000 


i 





00000 


k oooooo 

F OOO 



@@ 







00 


M 






o 

















W MET MF MN" 


Bothered some Bothered little 


or much or none 



Jfedian 
Veterans 
o Homreterans 



H2R CEUT CHOOSING" BESIGUAIED CATEXrQRIES IN" EACH COLLEGE 





s 

0> 



130 



3 



no 



* Veterans 
o Nonveterana 



Bothered some Bothered little 
or much or none 



Total 
group 

MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 43. WORRY AND AMXIETT ABOUT MAKING ENDS 

MEET: ITEM; 



WORRIES OF COLLEGE STUDENTS 269 



with AAG prevents the Item from having any important bearing on veteran- 
nonveteran differences in AAG; this conclusion is confirmed by the outcome 
of the sign test. 

Feelings of Inferiority, The source of worry included as Item 40(j) is 
"feelings of infer iority, inability to compete with others or to live up to 
your own standards." Tfriis item, like the following one on nervousness, is 
related to personality or general quality of psychological adjustment. 
These two items come together in rank order of importance behind the sources 
of worry concerned with academic and financial problems. The results for 
Item 40(j) are shown in Figure 44. 

lonveterans were bothered by feelings of in- 
feriority more often than were veteran students, and 
women tended to be bothered by inferiority feelings 
more frequently than the male nonveterans. Among 
veterans, students who worried about feelings of in- 
feriority tended to earn slightly lower AAG r s than 
those not bothered. 

Honveteran students indicate by their item responses that they worry 
more than veterans about feelings of inferiority. The median value of the 
percentage checking "bothered some or very much" is about 30 for veterans 
and 40 for nonveterans. In all twelve of the basic groups, the percentage 
for nonveterans exceeds that for veterans, 'which would be expected to occur 
by chance less than once in a hundred times. This finding is reasonable in 
the light of the greater age and experience of the veteran students. The 
difference in amount of concern about feelings of inferiority is not neces- 
sarily due to greater experience im^ng veterans more confident; it is pos- 
sible that the presence of veteran students has affected the self-confidence 
of the younger nonveterans. The eighteen-year-old freshman student, who is 
perhaps away from home for the first time, may feel quite inadequate in com- 
peting with the numerous older veteran students. 

Women students are again found to exceed the men in amount of worry as 
indicated by questionnaire responses, although the difference is not great. 
The median value for women who worry at least some is about 45 per cent as 
compared with 40 for nonveterans. The percentage of women who reported 
concern about feelings of inferiority is greater than that for nonveterans 
in six of the nine groups containing women. 

The relationship to MG of worry about feelings of inferiority is in 
the same direction as has usually been found greater worry goes with lower 
Adjusted Average Grade* This time the association appears to be somewhat 
closer for veterans than for nonveterans, although it is generally insignifi- 
cant. In. only one group is a significant difference (at the J.% level) found - 
veterans at Harris who worry about inferiority are significantly lower in 
mean AAG than those who do not. In ten of the twelve groups the veterans 
who worry earn a lower mean AAG than those who do not, and in one case 
there is no difference. Since veterans on a campus with nonveteran fresh- 
man students would have generally less reason to indicate that they are 



270 



ADJUSTMENT TO COLLEGE 



100 
75 














\ 


ooo 











\ oooo 


,0 




o 




oo 

00 




ooo 




o 




* 


oooo 









* nir* 


oo 














MV MK MF MN 
Bothered some Bothered little 
or much or none 



I Median 
Veterans 
o Honveterans 



PER CENT CHOOSING DESIGUMED GATSGCSRIES UST EACH 



150 





a 
s 



g 130 

8 

-P 



^ 120 



110 











4 




i 




<; 


, - c 










Veteran* 
o Nonrete] 


3 

trans 


Bothered some Bothered little Total 
or much or none group 



MEAH AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CA2EGOKIES 



FIGIHE 



WQiRRT AKD AUXIECT ABOUT FEELINGS OF 
: ITEM 



WORRIES OF COLLEGE STUDENTS 271 



bothered by feelings of Inferiority unless the feeling Is "based on a more 
fundamental personality characteristic,, the Item may "be touching on a more 
basic problem for veterans than for nonveterans. 

Again the Item does not help in accounting for the observed superiority 
of veterans in mean AAG, although they more often than nonveterans possess 
the characteristic which Is associated with higher achievement relative to 
ability. The validity of the Item Is too low,, especially for nonveterans, 
to permit a significant result. 

Nervousness. The source of worry Included as Item 40(d) was simply "nervous- 
ness." It was intended that this item might furnish a crude measure of 
anxiety or neurotic tendency. The results of the analysis are shown in Figure 



Veterans tended to be bothered by nervousness 
only slightly more often than male nonveterans, but 
neither group reported this worry as frequently as 
did women students. Among the nonveterans there was 
a slight tendency for the worriers to earn lower Ad- 
justed Average Grades than those not bothered about 
nervousness. 

Again, only a slight difference is found between veteran and nonveteran 
students in tendency to be bothered by nervousness. In the median group of 
veterans, about 40 per cent are bothered some or very much, and the proportion 
is slightly lower for nonveterans. The greater tendency toward nervousness 
on the part of the veterans Is not significant; in only nine of the twelve 
basic groups are veterans bothered by nervousness more often than are non- 
veterans . 

Women students show a greater tendency to worry about nervousness than 
men. In the median group about 55 per cent of the women students report 
that they are bothered some or very much, and in all nine colleges where 
women students were studied, a higher proportion of women than of male non- 
veterans reported being bothered by nervousness; there was, however, one 
college In which the male veteran subgroup exceeded the women in this 
respect. 

Variability among colleges was not great for most of the items on 
sources of worry, and this Item Is no exception. The largest proportions 
of worriers about nervousness were at Central State and Evans, where more 
than 50 per cent reported that they worried some or very much about nervous- 
ness. The smallest proportion was at Middle State, where less than 25 per 
cent of the students were in this category. 

As is shown in the lower portion of Figure k5, there Is a slight 
tendency among nonveterans for worriers to earn lower grades relative to 
ability than those who report little or no worry about nervousness. The 
association is not significant, however, and for veteran students there 
is practically no relationship. Because of this insignificant relationship 



ADJUSTMENT TO COLLEGE 



100 



CQ 
-P 

I 









m 














O 










00 








M 


> 0000 









P 


O 


41 


GO 




* 


oo 


** 


O 






O 


* 

' * 

90 




1 00000 
00 





















Median 

@ Veterans 
o Non-veterans 



Bothered some 
or much 



W MN 

Bothered little 

or none 



PER CENT CBDQSJBG DESIGHA2ED CATEGORIES IN EACH COLLEGE 



H 

8 



110 











4 


>__ < 


i 
< 


> 





^--"^ 


* < 


s 






9 Veteran* 
o Nbnvetei 


J 
ans 


Bothered some Bothered little Total 
or much or none group 



MRAN AAG OP MEDIAH COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 



WORRY AND AUZTET5T ABOUT MEKTOUSJSESS : ITEK 



WORRIES OF COLLEGE STUDENTS 273 



"between the two variables,, the tendency for veterans to earn higher AAG f s 
than nonveterans obviously cannot "be accounted for on the "basis of the dif- 
ferential in amount of worry about nervousness. 

Getting to Know People Socially, Following the worries about academic ad- 
justment, finances, and neurotic tendencies in order of importance come two 
sources of worry which have to do -with social adjustment. The first of 
these is "getting to know people socially," which is Item 40(h) . The re- 
sults are shown in Figure k-6. 

Getting to know people socially bothered non- 
veterans considerably more than veterans. Women 
reported concern more frequently than male veterans 
but less than male nonveterans. Concern over getting 
to know people is associated with high Adjusted Average 
Grade. Honveterans shoved a rather consistent tendency 
to choose the responses associated with better-than- 
average Adjusted Average Grade more often than did 
veterans . 

Veteran students are apparently less concerned about getting to know 
people socially than the nonveterans. The median percentage of students who 
reported being bothered some or very much is about 25 for veterans and 35 
for nonveterans. In eleven of the twelve basic groups, proportionately more 
nonveterans than veterans expressed concern about getting to know people 
socially. Women students typically worried more than the veteran men but 
less than the nonveteran men about this aspect of social adjustment. The 
fact that women are less concerned than the male nonveterans, who are about 
equal in age, perhaps reflects the greater rate of physical maturation in 
girls . 

In the case of every source of worry so far considered, greater worry 
is associated with lower AAG. Worry about getting to know people socially, 
however, is associated with higher AAG. The difference between the median 
values is more than 10 points of AAG, both for veterans and nonveterans. 
In all twelve groups, veterans who reported worry earned higher AAG's, on 
the average, than those who did not, and in eleven of the twelve groups the 
nonveterans who worried were higher. Five of these twenty-four differences 
were significant at the 1$ level, and an additional five were significant 
at the 5$ level. At least in their own estimation, it would appear that 
the overachievers are somewhat less successful in their social relation- 
ships than are the underachievers . One possibility is that those who worry 
are those who spend less time in social activities, being less successful 
in social relationships, and who therefore have more time to devote to aca- 
demic pursuits. 

We find, then, that nonveteran students possess more often than veterans 
the characteristic which is associated with high Adjusted Average Grade : a 
tendency to worry about getting to know people socially. This item would 
therefore lead to the expectation that nonveterans rather than veterans 
would excel in grades relative to ability. The sign test showed that in ten 
out of eleven instances where both veterans and nonveterans had high mean 
AAG ! s, the veterans were more likely to be represented than the nonveterans; 
this tendency is significant at the 5$ level 



ADJUSTMENT TO COLLEGE 



100 



1 

CD 

13 

-P 
Pt 

CD 

O 







I.. 


o 










00 

1 oooo 
ooo 










00 


p Median 










Veterans 










o Wonveterans 




oo 
L oooo 
f ooooo 








>- 



* 


o 








Bothered some 
or much 


MV MKT 
Bothered little 
or none 



HER CENT CHOOSING EESIOK/ITED CATEGORIES UST EACH COLLEGE 



JU^W 

<!> llfO 

1 


i 


k 






c 


\ 


< 


i 




x \ 


< 




<D 
-P 




^ x 






g 120 




N 










* Veterans 




no 






a Uonveterar 


is 


Bothered scaane Bothered little Total 


or much or none group 



MEAH AAG OF MEDIAE COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 46. WQRKT AND ARXJETT ABOUT GETTIHG TO KNOW EEOELE 
SOCIALLY: ITEK 



WORRIES OF COLLEGE STUDENTS 275 



Trying to Make Up a Deficiency in Preparation. Another source of worry 
which Is concerned with the academic adjustment of students was stated, in 
Item 40(1), as "trying to make up a deficiency in preparation for some 
course." This is the only item relating to scholastic problems which was 
not near the top of the list in order of importance. The results of the 
analysis are shown only in Appendix Table 



About a third of the students, whether male 
veteran, male nonveteran, or female students, were 
bothered about trying to make up a deficiency in 
preparation for some course. Students concerned 
about this source of worry earned significantly 
lower AAG f s than those not bothered by trying to 
make up a deficiency. 

Approximately one third of the students, regardless of veteran status, 
reported that they were bothered some or very much about making up a defi- 
ciency in their training. Being away from school for two or three years 
did not, according to this evidence, produce any marked feeling of concern 
on the part of the veterans about being inadequately prepared to undertake 
their college work. Women students were not found to worry more than men 
about making up a deficiency. 

Variability among colleges was not unusually great, but it is of interest 
that the smallest proportion of students who reported worry about making up 
a deficiency (exclusive of interrupted groups) was found at Turner. This 
may reflect the fact that Turner has a rather elaborate system of placement 
tests in the initial classification of students. 

As is true of the other academic worry items, worry about making up a 
deficiency bears a rather close relationship to Adjusted Average Grade. 
The difference between the median values of the mean AAG's of those in the 
two item categories is about 20 points, both for veteran and nonveteran stu- 
dents, In none of the groups, either for veterans or nonveterans, did the 
students who were bothered little or none earn lower grades than those who 
worried some or very much. In 15 of the 2k subgroups, the difference was 
significant at the 1$ level. The highly significant relationship is con- 
sistent with the results found for other items dealing with worry about 
academic adjustment. Because of the high degree of similarity between 
veterans and nonveterans in the proportion who worry, the item obviously 
cannot help in accounting for the veteran superiority in AAG. 

Relations with Members of the Opposite Sex. The second source of worry 
dealing with social adjustment is stated in Item 4o(m) as "relations with 
members of the opposite sex." The importance of this source of worry, as 
indicated by the proportions of students who report that they are bothered 
some or very much, is about the same as that of the other social adjust- 
ment item ("Getting to know people socially."). Concern about social ad- 
justment, according to these findings, follows academic adjustment, finances, 
and neurotic tendencies in order of importance. The results for this item 
are shown only in Appendix Table ^O(m) . 



276 ADJUSTMENT TO COLLEGE 



More nonveterans than veterans expressed con- 
cern over relations "with members of the opposite sex; 
little difference existed between male nonveterans 
and female students Again tendency to worry is 
associated with high AAG, but to a much lesser degree 
than was found for worry about getting to know people 
socially. 

The median values of the percentage reporting that they were bothered 
some or very much were about 25 for veterans and 30 for nonveterans. The 
results are thus consistent with those of the previous item on social adjust- 
ment in showing greater concern on the part of the nonveterans,, although the 
difference is not as marked. The percentage reporting worry is greater for 
nonveterans in nine of the twelve "basic groups* 

The median percentage of women students who worry about relations with 
members of the opposite sex is the same as for nonveteran males , when the 
comparison is based on the nine groups where women were studied. In ;fchese 
nine groups, the percentage is greater for women in six cases. The sex dif- 
ference in worry about this aspect of social adjustment appears to be very 
slight. 

The relation between amount of worry and Adjusted Average Grade is 
again the opposite of the one usually found greater worry is associated 
with higher AAG. Both of the items dealing with concern about social adjust- 
ment have this characteristic in common. In the case of this item on worry 
about relations with members of the opposite sex, however, the relationship 
to AAG is much less marked than for the more general one on getting to know 
people socially. The worried veterans were superior to the unworried ones 
in only eight of the twelve groups (there was no difference in one group ), 
and the worried nonveterans also excelled in eight of the twelve groups . 
In only one group was a difference found which is significant at the 1$ 
level. The relationship to AAG thus cannot be considered significant, and 
the superiority of veteran students cannot be ascribed to any characteristic 
measured by this item. 

Health Problems The importance of health problems as sources of worry among 
college students, is relatively low, according to these results. The source 
of worry included as Item 40(e) was "health problems (e.g., eyes, sinus 
trouble*)" The choice of relatively minor health problems as examples was 
made deliberately because of the assumption that serious physical handicaps 
would occur rarely among college students. The results of the analysis are 
shown only in Appendix Table kO(e) 

Only about a quarter of the students indicated 
that they were bothered by health problems. The 
proportions bothered were about equal for male 
veterans and male nonveterans, and only slightly 
higher for women,. This item bears, no marked rela- 
tionship to Adjusted Average Grade. 



WORRIES OF COLLEGE STUDENTS 277 



Between 25 and 30 per cent of the students, whether veteran or non- 
veteran, reported being bothered some or very much by health problems. 
It is worthy of note that variability among the colleges is very small 
with respect to percentages in the two questionnaire categories; apparently 
health problems are, by and large, unrelated to the characteristics of the 
various institutions studied. Women students tend to worry about health 
only slightly more often than the male students. This greater concern ex- 
pressed by women might result from acceptance of the popular stereotype 
that women are the "weaker sex" rather than from any actual difference in 
health. 

There is no marked tendency for concern about health to be associated 
with Adjusted Average Grade, either for veteran or for nonveteran students* 
Hone of the differences are significant at the 1$ level. It can reasonably 
be concluded that concern about health is unrelated to grades relative to 
ability. Obviously the superiority of veteran students in AAG is not re- 
lated to concern about health. 

Illness or Death in Family* The second item concerning health pertains to 
the student's family rather than to the student himself. The source of 
worry, as stated in Item 40(c) was "illness or death In your family." Be- 
suits for this item are shown only in Appendix Table ^O(c) . 

Eelatively few students reported illness or 
death in the family aa a source of worry. Veterans 
tended to be bothered on this score sligjhtly more 
often than male nonveterans, and female students 
more frequently than either male group. As with 
the student's own health, family health appears to 
bear no significant relationship to Adjusted Average 
Grade. 

Only about 15 per cent of the nonveterans and a slightly higher propor- 
tion of veterans indicated that they were bothered some or very much about 
illness -or death in the family. In ten of the twelve groups fewer non- 
veterans than veterans were concerned about this source of worry. The 
slightly greater amount of worry on the part of the veterans might be re- 
lated to the fact that, on the average, their parents would be older and 
somewhat more susceptible to ill health. Women students consistently ex- 
press greater concern about family health than male nonveterans. Again we 
find relatively little variability among the various institutions in amount 
of concern expressed. 

Being bothered about the health .of members of the family has little re- 
lation to AAG. Among veterans there is practically no relationship, while 
among nonveterans greater worry tends to be associated with lower AAG. None 
of the differences is significant at the 1$ level* Concern about illness or 
death in the family bears no relation to the higher grades relative to 
ability of veteran students. 



278 ADJUSTMENT TO COLLEGE 



Strained Personal Relations. Another source of worry which is related to 
social adjustment was stated in Item kQ() as "strained personal relations 
with close relatiyes or friends ." The results are shown in Appendix Table 



Strained personal relations were a source of 
worry for relatively few students; little difference 
existed between veterans, nonveterans. and female 
students in the proportion claiming to be bothered. 
There is a very slight tendency for students who re- 
ported worry over personal relations to earn lower 
AAG's. 

Only a few of the students about 15 per cent reported that they were 
bothered some or very much. There was no significant difference between 
veterans and nonveterans in tendency to worry about strained personal rela- 
tions y but women students reported concern somewhat more often than men. 
There was cang>aratively little variability among the college groups,, except 
that Eastern City veterans expressed concern considerably more often than 
students in other colleges. This tendency was shown by the interrupted 
veterans as well as the freshmen, 

A slight and generally insignificant tendency is shown for greater 
worry to be associated with lower Adjusted Average Grade. Wo differences 
were found which were significant at the 1% level. The greater mean AAG 
of veteran students cannot be explained on the basis of a differential 
amount of worry about strained personal relations. 

Housing. The last of the sources of worry to be discussed is one referred 
to in Item ^O(b) as "lack of adequate housing accommodations." The results 
of the analysis of this item are shown in detail in Appendix Table 



In a typical group only about 10 per cent of 
nonveterans and a slightly greater proportion of 
veterans expressed concern about lack of adequate 
housing, and even fewer women were worried about 
housing accommodations. In some institutions, how*- 
ever, a third or more of the veterans were 'bothered 
by the problem* The relationship between worry about 
housing and Adjusted Average Grade was negligible. 

Slightly more than 10 per cent of the nonveterans said that they were 
bothered some or very much about housing, and the percentage of veterans 
was only about 5 per cent greater, in the median group. It would appear 
that the importance of the housing problem wets not great in the typical 
university, althougih at some institutions a considerably greater amount 
of concern was expressed. At Littletown State almost a third of the stu- 
dents, both veteran and nonveteran, were in the "worried" category, and 
at Southern Tech almost kO per cent expressed concern. Women students 
were less concerned than men about housing, presumably because dormitory 



WORRIES OF COLLEGE STUDENTS 279 



provisions are usually more adequate for women and the overcrowding of 
housing facilities was ordinarily much greater for men. 

The relationship between concern about housing and AAG is negligible, 
although it is interesting to discover the direction of the relationship: 
greater worry is associated with higher rather than lower Adjusted Average 
Grade o In only one group was a difference found which is significant at the 
1$ level This item does not, of course,, help in accounting for the veteran 
superiority in AAG because of the small differences in amount of worry and 
the insignificant relationship with AAG. 

Other Sources of Worry. Item 1*1 was a free-answer question designed to 
elicit statements about sources of worry not included in Item 40 The 
question was, "Are there any problems not mentioned in the previous ques- 
tion which have been bothering you in the past six months?" Yes and No 
responses were presented for checking, and the additional question, f or 
those who said Yes, "What general sort of problems?" The Yes-No responses 
were not tabulated. The responses to the free-answer question"~were coded, 
and the detailed results are presented in Appendix Table 4l. 

Relatively few students added anything to the 
list of possible worries presented in the previous 
items; fewer than one in ten veterans mentioned an 
additional source of worry, and a somewhat greater 
proportion of nonveterans and women cited further 
problems which bothered them. Worries mentioned were 
quite varied, with "worry about examinations, fear 
of flunking" the most frequently given. 

Only a small proportion of the students replied to the free-answer part 
of Item 4l, and fewer veterans than nonveterans responded. Less than 10 per 
cent of the veterans wrote in a response; the proportion of male nonveterans 
and of females responding was almost 15 per cent. This difference does not 
necessarily mean that the veterans had fewer things bothering them; veterans 
may have been somewhat less willing to go to the trouble of writing out 
their worries. 

Eight categories were used In coding the responses to Item tl. The 
number of students in each category is necessarily very small since less 
than 15 per cent in a typical group made any response* The categories 
most frequently chosen are as follows: 

(A) Tensions or conflicts concerning contemporary social or economic 
institutions, and/or worry about economic, national or international situa- 
tions at time of graduation. About 2 per cent of the students gave responses 
coded as A. 

(B) Indecision regarding type of future work for which to train, 
and/or whether or not to plan on post-graduate training. About 1 per cent 
of the students gave responses of this type* 



280 ADJUSTMENT TO COLLEGE 



(C) Worry about examinations, fear of flunking. This is the most 
commonly given type of response. In the median group, only about 3 per cent 
gave answers classified as C, but in some groups as many as 8 per cent indi- 
cated worry about examinations or flunking. 

(E) Insufficient time or faulty division of time. Less than 1 per 
cent of the veterans and about 3 per cent of the nonveterans gave responses 
in this category. 

[Responses which were classified as belonging in the remaining categories 

occurred even more rarely These categories were as follows: 

(B) Indecision regarding continuing work vs. leaving to take a job* 

(E) Homesickness. 

(F) Religious or moral conflicts 

(G) Parental or family conflicts indirectly involving the respondent. 

The number of cases in any one category is insufficient to expect satis- 
factory reliability of the mean AAG values . It might be mentioned only that 
students who worry about examinations or flunking tend to earn low AAG's, 
which of course is consistent with the results of other items dealing with 
concern about academic problems. 

The Influence of Worry on College Work 

The study of the worry items in relation to AAG, as described above, 
indicates that in general greater worry is associated with lower grades in 
relation to ability. Item lj-2 of the questionnaire is concerned with the 
students 8 judgments as to the effects of worry on academic work. The ques- 
tion as stated is, "How much would you say that any of the problems mentioned 
on the previous page either the ones listed in Question 40 or any other 
have interfered with your college work in the past six months?" The response 
categories were (A) have not interfered at all, (B) have interfered a little, 
but not much, and (C) have interfered a good deal. The results are shown. 
graphically in Figure kj 

Most students felt that their work had been 
affected by worries to some extent^ although only one 
in five thought the interference had been consider- 
able Veterans, nonveterans and women students were 
very similar in their responses to this question. 
The Adjusted Average Grades bore out the students 1 
opinions; those who felt that worries had interfered 
a good deal did in fact earn much lower grades than 
those who reported no interference* 



WORRIES OF COLLEGE STUDENTS 



281 



J.UU 

CQ 

S 

CD 

I 

CQ 

% 5 
-p 

1 


fc 25 
















1 Median 
Veterans 
o Nonreterans 






f r_. 


. ooo 
^ oooooo 
ooo 








OOOO 









o 


( 


o 
* 000060 








r 


oooooo 

OOOO 

o 


MT MN MV m mm 

Not at all A little, A good deal 
but not much 



HER CENT CHOOSING DESIGNATED CATEGORIES IN EACH COLLEGE 



iyu 
lij-0 



1 
I 130 

^ 

(D 

5 

-^120 

^s 

CQ 

ft 

3 110 

100 


< 


K^ 

\ 
^ 








( 


^-^ ^""^ 
"^-^ ^^< 


k 


< 


> 






r^\ 


< 


> 






x \ x ^ 










> 


Veterans 
o Nonveter* 


ans 


Not at all A little, A good deal Total 
but not much S rou "P 



MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE kj e EXEENT TO WHICH WORRIES AND ANXIETIES HAVE TNTWEWETD WITH COLLEGE 
WORK: ITEM 42 



282 ADJUSTMENT TO COLLEGE 



About one fifth of the students in the typical college group reported 
bhat their worries had interfered a good deal with their college work. 

15 per cent felt that the problems had not interfered at all; the 
majority (about two thirds) took the intermediate position ("inter- 
fered a little"). Male veterans, male nonveterans, and women showed roughly 
the same proportions, in general, choosing each category. 

The relation of the item to MG indicates that the testimony of the 
students is consistent with the findings previously reported that greater 
worry is usually associated with lower grades relative to ability* Those 
students who felt that their problems had interfered a lot tended to earn 
low AAG ! s (the median values are about 120 and 110 for veterans and non- 
veterans respectively), while students who reported no interference tended 
to earn high MG's (medians are about 150 and 1^0 ). In all twelve basic 
groups the "interfered not at all" students, both veteran and nonveteran, 
were superior to students choosing other categories; similarly in all twelve 
groups the students choosing "interfered a good deal" were lower than other 
students in every case, both for comparisons involving veterans and non- 
veterans. Of the forty-eight comparisons involved, twenty-four were signifi- 
cant at the 1$ level of confidence. The high relationship may, of course, be 
accounted for in part by what we have previously referred to as rationaliza- 
tion: students who are doing poorly may attribute their failure to worry 
It almost seems that students are willing to attribute their success or 
failure to 'any plausible reason which is presented to them. 

Because of the high degree of similarity between veterans and non- 
veterans in the proportions choosing the various categories, it is apparent 
that the higher standing of veterans in AAG cannot be attributed to what- 
ever characteristic is assessed by this item. 

From a methodological standpoint, it is noteworthy that although almost 
one fourth of the students in the typical group reported that they seldom 
or never worried, only 15 per cent reported that worries had not interfered 
at all with their college work. Whether this resulted from the fact that 
one item preceded and the other followed the check -list of worries, or 
whether it resulted from the difference in wording cannot be answered from 
these data 



Conclusions 



In response to a general question about tendencies to feel worried and 
anxious or upset, only about one fourth of male college students (in the 
median group) reported that they seldom or never worried. About 20 per 
cent, at the other extreme, said they felt worried and anxious frequently., 
There is no tendency for veterans to worry .more than male nonveterans ; but 
female students report worries more often than men in most of the colleges 
where their responses were studied. 

On the basis of the questionnaire responses of nonveteran men, it seems 
that problems related to scholastic adjustment are the most common sources 
of worry and anxiety . "Concentration, " "getting accustomed to college 



WORRIES OF COLLEGE STUDENTS 283 



study/' and "deciding what coarse of study to follow" were the three most 
common sources of worry e More than half the nonveteran students, in the 
median groups , reported being bothered some or very much about each of these 
problems . 

"Making ends meet" was the next most common source of anxiety The 
most marked difference between veteran and nonveteran students was with re- 
gard to finances, veterans showing a definitely greater tendency to worry 
about making ends meet* The median percentage of veterans who said they 
were bothered some or very much was about 65, as compared with a median 
value of about 45 for nonveteranso 

Concern about personal adjustment appears to be next in order of import- 
ance o The sources of worry as stated in the questionnaire were "feelings of 
inferiority" and "nervousness " In a typical college group a third of the 
male students showed some concern about such symptoms of neurosis, although 
veterans were bothered somewhat less often than nonveterans about feelings 
of inferiority* 

Next in order of importance as sources of worry seems to come social 
adjustment, as judged by the frequency of worries about "getting to know 
people socially," and "relations with members of the opposite sex." About 
a fourth of the veterans and a somewhat larger proportion of nonveterans 
indicated concern about these problems. 

Health problems, involving either the students themselves or their 
families, are low in importance as reasons for worry, according to the 
median values found . Housing is also low on the list, as is "strained 
personal relations . '* Belatively few students took advantage of an oppor- 
tunity to report other worries in a free -answer item*, 

Women students as a general rule reported being bothered by these 
problems more often than male students. Whether this finding is the result 
of a greater tendency toward neuroticism among women or a general acceptance 
of the stereotype that women are more emotional cannot, of course, be told 
from, these data* It is interesting to note that the two sources of worry 
that bother women much less than men are "making ends meet financially" and 
housing. Such a finding might have been foretold on the basis of the role 
of women in our society,. Women exceed men most in frequency of being 
bothered by nervousness, according to these results. 

When asked if any of the problems had interfered with college work, 
about two thirds of the students answered "a little, but not much, " and 
about a fifth said "a good deal/ 1 There was a marked tendency for students 
who thought their problems interfered a good deal to earn low grades rela- 
tive to ability . This finding may merely be another example of the tendency 
for low-achieving students to blame their failure on any plausible reason 
which is suggested to them*. 

With respect to the relation of sources of worry to Adjusted Average 
Grade, the most sweeping conclusion that can be drawn is that worry is 
associated with lower grades relative to ability., There are several inter- 



284 ADJUSTMENT TO COLLEGE 



esting exceptions, however. Those students who worry most about social ad- 
justment ("getting to know people socially" and "relations -with members of 
the opposite sex") tend to earn higher grades than those vho worry little 
or not at all. It would seem that satisfaction with one T s social adjustment 
is slightly detrimental to scholastic achievement ; an obvious hypothesis is 
that the more studious freshmen feel that they are missing the social side of 
college life . The other source of vorry where greater worry tends to go 
with high AAG is housingj but the relationship here is very slight and cer- 
tainly not significant. 

The relationship between amount of worry and AAG is most marked for the 
sources of worry concerned with academic adjustment. These relationships are 
highly significant, in a statistical sense; but the interpretation of the 
responses- as rationalizations for grades already earned must again be con-* 
sidered. It is quite possible that the relationships of the items to grades 
would be even closer than to AAG. 

On the basis of the sign test results, two of the items may be con- 
sidered relevant to veteran-nonveteran differences in Adjusted Average 
Grades. Both with respect to worry about inability to concentrate and worry 
about getting to know people socially, nonveterans showed a greater tendency 
to give responses associated with higher Adjusted Average Grades. It would 
appear, then, that if these worries had been equally prevalent among veterans 
and nonveterans, the advantage of the veterans in AAG would have been en- 
hanced. It should be understood, of course, that these items, although sig- 
nificant at the 5$ level, may only be symptomatic of underlying differences 
between the two groups j there is no intention to Imply that these worries 
caused the veterans to do worse than they would have done had their pattern 
of worries matched that of nonveterans. 



205 



Chapter VIII 
HOW COLLEGE STUDEM'S SPEM) THEIR TIME 



Among various hypotheses as to the factors responsible for the higher 
grades relative to ability earned by veteran students , the hypothesis that 
veterans studied more and spent less time in "frivolous" activities "would 
seem veil worth investigating. A number of items included in the question- 
naire were concerned with the number of hours in a typical week spent by 
students in studying, athletics, extracurricular activities, social affairs, 
and other activities which ordinarily consume a significant proportion of 
the college student's time. As usual, the analysis was directed at finding 
the frequency with which the various questionnaire categories were chosen, 
the relationship between the characteristic assessed by each questionnaire 
item and Adjusted Average Grade, and the influence of the characteristic in 
producing the observed difference between veteran and nonveteran students 
in grades relative to ability* 

Most of the information on disposition of time was derived from one 
question, which had a number of parts* The question (Item 22) was, "During 
the past week, how many hours did you spend at each of the following activi- 
ties? (If the past week was not typical, indicate the number of hours for 
a typical week.)" Nine activities were listed, including attending class, 
studying, athletics, extracurricular activities, social activities, attend- 
ing lectures and concerts, bull sessions, paid employment, and "other non- 
routine activities." In order to reduce the number of response categories, 
the numbers indicating hours spent which wer<e written into the questionnaire 
by the students were coded, ordinarily three categories being employed in 
the analysis of the item. 

One problem which was encountered in the analysis of certain of these 
items was the presence of an unusually large proportion of students in the 
"no response" category students who failed to enter a number opposite an 
activity,, This tendency was especially marked on items where the amount 
of time usually spent was very small; for example, for Item 22(f ) (attend- 
ing public lectures, concerts, and other cultural activities), about a fifth 
of the students gave no response, and less than a third reported spending 
more than one houro (Cf . Appendix tables for the proportion of students 
in the "no response" category. ) Presumably the students who did not write 
in a response were those who had spent little or no time in that activity; 
it seems unlikely that students spending a substantial amount of time in 
an activity would omit the item,. Therefore the "no response" group was 
merged with the category representing the smallest number of hours, which 
was also the modal response of those who did respond. 



286 ADJUSTMENT TO COLLEGE 



Disposition of Time 

Attending Classes, The first activity listed, in Item 22(a), is "attending 
classes, labs, regularly scheduled course conferences." The purpose of in- 
cluding this item is to investigate differences "between veterans and nonveterans 
with respect to work load and to discover the relation "between work load and 
Adjusted Average Grade. The results of the analysis are shown in Figure kQ. 

Veteran students reported spending less time 
attending classes than did nonveterans; this presum- 
ably is a result of veterans being excused from the 
usual physical education or military science require- 
ments,, Considerably more time was generally spent in 
class meetings at engineering colleges than in liberal 
arts colleges. Students who spend more hours in class 
meetings (relative to their own college group) tend to 
earn higher AAG ! s than students who spend fewer hours 
in class. 

Preliminary tabulations of the responses to this item showed that the 
colleges differed considerably with regard to the number of hours which stu- 
dents reported they spent attending classes and laboratories. In one engi- 
neering college (Midwest Tech), the median number of hours reported by non- 
veterans was almost 31, while in one of the liberal arts colleges (Turner) 
the median for nonveterans was less than 15. It was impossible, for this 
reason, to use a three -category code which would be appropriate for all col- 
leges. The solution was to choose a code for each college which was suitable 
for the range of responses there obtained. The generalized categories for 
number of hours in classes and laboratories may be described as follows : 
(A) relatively few, as compared to own college group; (B) moderate number, 
as compared to own college group; and (C) relatively many, as compared with 
own college group. The "own college group" of course includes both male 
veterans and male nonveterans. In the case of Midwest Tech engineering 
students, Category A represents less than 29 hours, B 29 to 3! hours, and 
32 or more hours; while for Turner A represents less than 14 hours, B 
Ik to 16 hours, and 17 or more hours. For these two extreme groups there 
is thus practically no overlap between the distributions. 

Since the item categories are defined relative to the particular col- 
lege group, Table k 5 is included to show the median value, for each of the 
twenty-five groups, of the hours spent attending classes as reported by 
students. It is apparent from the table that engineering and agriculture 
students generally spent more hours in classrooms and laboratories than did 
liberal arts and business students; the reason undoubtedly is the greater 
amount of time spent in laboratory work by students in the applied science 
courses. It is also apparent that nonveterans consistently spent more time 
attending classes than did the veterans. In nineteen groups containing 
veteran and nonveteran freshmen, the median of nonveteran students is higher 
for all but Turner; at Turner, there is no difference. This finding agrees 
with the results reported in Chapter III for Douglas and Midwest Tech, where 



HOW COLLEGE STUDENTS SPEND THEIR TIME 



287 



w 
















4 


1 
I @ < 


j 
> 







^^^ Median 






.^aSfc^ 








i> Veterans 














o Ifcxaveterane 


m 


















X *"* 

00000 

JL*^ 


o 







10 15 20 25 30 35 


Soiors Per Week 



HOURS PER HEEK ICR EACE COLLEGE 



Adjusted Average Grade 

H H H t J- 

S 8 o ^ 














^^-^" 


, 


) < 






> 


^^ 


( 




< 


^ 




Veterans 
o Uonveterans 




Relatively Mpderate Relatively Total 
ev hours number of hours many hours group 



, AAG CF MEDIAN COLUSGE GERODP FOR SPECIFIED CATEGORIES 



FIG-TIRE 48. SOURS PER WEEK SPENT ATMHOTCr CLASSES s ITEM 22(a) 



288 



ADJUSTMENT TO COLLEGE 



'CABLE t$. MEDIAN HOURS PER WEEK SPENT ATTETOMG CLASSES ^ LABCSIATORIES, 
MD OTHER ITOULABLI SCHEDULED COURSE COBDFERMCES: HEM 22 (a) 



College Group 


& 

Median Hours Per Week 


Male 
Veterans 


Male 
Nonveterans 


Females 



GROUPS IHGLHDITO VETERANS WHO ENTERED COLLEGE AFTER WAR SERVICE 



1. Central State, Arts, 19t6 


16.9 


20.9 


19-2 


2. Evans, Arts, 19t6 


ISA 


17.6 


17-6 


3. Western State, Arts, 19t6 


15-7 


18.0 


.17-9 


If. Miller, Arts, 19t6 


17oO 


19-8 


17-8 


5. Stewart, Arts, 19t6 


17-5 


20.5 




6, Harris, Arts, 19t6 


16.5 


18.9 


18.6 


7, Adams, Arts, 19t6 


19-t 


20.7 




8. Douglas, Arts, 19t6 


19o8 


20.4 


18,7 


9. Littletown State, Arts, 19t6 


19*2 


21.6 


19-3 


10. Midwest City, Arts, 19t6 


16.6 


19. t 




11. Eastern City, Arts, 19t6 


19-9 


20.3 


20. li- 


12. Ttirner, Arts, 19^5-46 


It. 9 


It. 9 


lt. 7 


13 o Central State, Arts, 19^5 


17.9 






li. Midwest Tech., Engr-, 1J&6 


26-9 


30-9 




15. Middle State, EngTo, 19^6 


20-3 


20.6 




16. Midwest City, Engr., 19^6 


27-3 


28.2 




17. Southern Techo, Engr., 19^5-^6 


25-8 


30.0 




l8o Midwest Tech., Agri., 19^-6 


23-9 


26.5 




19. Midwest State, Bus,, 19^6 


17-3 


20.1 




20. Littletown State, Bus., 19^6 


17-6 


20.2 





GROUPS INCLUDING VETERANS WHO EETURNED TO COLLEGE AFTER WAR SERVICE 



21 o Eastern City, Arts (MH-19t5) 


16.8 


20.2 




22. Adams, Arts (MN-19t5) 


17*5 


16.0 




23. Stewart, Arts (MGT-19t5) 


15.8 


16.5 




2t. Midwest Tech., Engr. (MSf-1939) 


22.9 






25 c Midwest Tech., Agri. (MN-1939) 


23-9 







In interpreting these figures it should be remembered that veteran 
students may have been exempted from physical education and military 
science courses which nonveteran students were required to take. 



HOW COLLEGE STUDEHTS SPEND THEIR TIME 289 



data on course load were obtained ftcm transcripts of offical records. This 
difference between veterans and nonveterans in class and laboratory hours 
probably may be attributed to the exemption of veterans from, the usual 
physical education or military science requirements. 

Female students tend to spend slightly less time in classrooms and 
laboratories than male nonveterans. Since all of the female groups were 
enrolled in Arts and Sciences , this difference may depend, in part, upon a 
tendency for them to avoid laboratory sciences, or to postpone taking them* 

The over -all findings for this item are shown in Figure 48. It will be 
noted that median hours spent attending classes is shown in the upper part 
of the figure* The median of the median values is about 18 hours per week 
for veterans and over 20 hours for nonveterans. 

The lower portion of the figure shows the relation of time spent in 
classes to Adjusted Average Grade? but here the categories A, B, and C, de- 
fined in relation to the student's own college, are employed. There is a 
tendency, as shown by the median values of the mean AAG f s, for higher achieve- 
ment relative to ability to be associated with greater amount of time spent 
in classes. The difference between the median values for nonveterans spend- 
ing relatively few and nonveterans spending relatively many hours in class 
is about 15; for veterans, however, the difference is only about 5 AAG units. 
Veterans spending relatively many hours in classes were significantly higher 
(at the 1$ level) than other veterans in AAG only at Evans; for nonveterans 
a similar significant difference was found at Miller. In all twelve basic 
groups nonveterans in the "relatively many" category were superior to other 
nonveterans. The direction of the relationship is consistent with that found 
at Douglas and Midwest Tech, as reported in Chapter III, and is presumably 
due to a tendency for the best students to take heavier course loads. 

Since nonveterans are found more often than veterans to possess the 
characteristic associated with high AAG spending more time in classes and 
laboratories one would expect nonveterans to excel in grades relative to 
ability, rather than veterans. Indeed, the sign test indicates that non- 
veterans show a greater tendency than veterans to choose responses associated 
with better -than-average AAG in ten instances out of eleven, with one tie. 
Statistically, this would be significant at the 1% level Such reasoning 
may be fallacious in this instance, however, since the tendency for veterans 
to take lighter loads is presumably related to an entirely different set of 
factors than is the tendency for high-achieving students in general to take 
heavier loads or more laboratory courses. When account is taken of the dif- 
ferent factors involved, it makes more plausible the hypothesis that veterans 
earn higher AAG*s because of their lighter loads . Unfortunately a crucial 
test of the hypothesis is not available. 

Studying* The activity listed in Item 22 (b) is "studying in your room, the 
library, or elsewhere " The purpose of this item is, of course, to compare 
veteran and nonveteran students with respect to time spent in study and to 
observe the relationship of amount of study reported to AAG The results 
are shown in Pigure ^9- 



290 



ADJUSTMENT TO COLLEGE 



100 



75 



-P 
fl 

5 

JH 



50 



25 





















. 

















I L o 







o 




oo 


| A| oooooo 







o 





o 


r 


r oo 







o 


0S 


ooo 




00 




, 


oo 





o 








@* 


oooo 





oo 
















o 










oo 






o 










Q 



m m 

l4 hours 
or less 



m m 

15-2l- hours 



m MGT 

25 hours 
or more 



I Median 
Veterans 
o Honveterans 



PER CENT REPORTING DESIGNA1HD CATEGORIES IN EACH COLLEGE 



Adjusted Average Grade 

H H H H l- 
H ro UJ fr Or 

o o o o c 














^s* 


*""""" ^-^^ 


\ 
> 

4 


i 


i 


X 




( 


j 


c 






Veterans 
o Nonveterai 


is 


14 hours 15-2^ hours 25 hours Total 
or less or more group 



AAG OF MEDIAN COLLEGE GROUP FOE SPECIFIED CATEGORIES 



FIGURE if9. HOURS PER WEEK SPENT STUDYING: ITEM 22 (b) 



HOW COLLICI STUDENTS SPEND THEIR TIME 291 



Veteran students were found to spend slightly 
more time In study than nonveterans, A greater number 
of hours devoted to study Is associated with higher 
grades re la tire to ability. 

Although there was considerable -variation among colleges with respect 
to number of hours spent In studying, it was possible to use the same coding 
system for all groups. The categories employed were (A) Ik hours or less, 
(B) 15 to 2k hours, and (C) 25 hours or more. nevertheless, in order to make 
the results more meaningful, the median number of hours spent In studying was 
determined for each of the twenty-five groups | these medians are shown in 
Table If 6. 

Comparison of male veteran and nonveteran students shows that the 
veteran students reported studying slightly more than the nonveterans. The 
median of the median values for veterans is slightly more than 20, while for 
nonveterans It Is about 19. In nine of the twelve basic groups, the median 
number of study hours reported is greater for veterans than for nonveterans, 
and in one group there Is no difference. There is relatively little differ- 
ence between number of study hours reported by male nonveterans and by female 
students . 

The lower portion of Figure k$ reveals a tendency for more study hours 
to be associated with higher grades relative to ability. The difference In 
median values of the mean AAG Is roughly 15 points. Those who study 14 hours 
or less earn a mean AAG which is significantly lower (at the 1$ level) than 
that of students studying more than ih hours in three of the basic groups of 
veterans and also in three of the basic groups of nonveterans. Both for 
veteran and nonveteran students, In ten of the twelve groups, students who 
studied 25 hours or more earned higher AAG r s than, those who studied less 
than 25 hours per week. 

Since veterans, on the average, study more than nonveterans, and since 
there Is some association between hours spent in study and Adjusted Average 
Grade, it would appear that this question might help to account for the 
veteran superiority in Adjusted Average Grade. The difference in study 
hours Is so slight, however, that little weight can be given to this find- 
ing; the results of the sign test turn out not to be statistically signifi- 
cant. 

Athletics. Item 22 (c) pertains to the activity described In the question- 
naire as "athletics and physical recreation (not counting physical education 
courses)." Since excessive participation in athletics is often supposed to 
be detrimental to good scholarship, the hypothesis to be tested is that a 
veteran-nonveteran difference in amount of participation might be responsible, 
In part, for differences in Adjusted Average Grade. The results are pre- 
sented graphically in Figure 50* 

A slight tendency was found for nonveterans to 
spend more time in athletic activities than veteran 
students. Women spent considerably less time In 



292 



ADJUSTMENT TO CQLLiCi 



TABLE 46. MEDIAN HOOKS PER WEEK SPENT STUDYING IN EOOM^ IN LIBRAE!, 

ITEM 22(b) 



OR 



College Group 


Median Hours Per Week 


Male 
Veterans 


Male 
Nonveterans 


Females 



GROUPS INCLUDING VETERANS WHO ENTERED COLLEGE AFTER AE SERVICE 



1. Central State, Arts, 1946 


17-5 


16.5 


14.7 


2. Evans, Arts, 1946 


20.4 


18.7 


18.5 


3- Western State, Arts, 1946 


21.1 


19.1 


22,0 


4. Miner, Arts, 1946 


19.1 


16.3 


13-5 


5. Stewart, Arts 7 1946 


22-7 


21.6 




6. Harris, Arts, 1946 


25.6 


22,7 


23.1 


7. Adajns, Arts, 1946 


23-5 


23.5 




8. Douglas, Arts, 1946 


19-0 


19-9 


18.7 


9. Littletown State, Arts, 1946 


18.7 


I8o8 


21.3 


10. Midwest City, Arts, 1946 


14.0 


16.8 




11. Eastern City, Arts, 1946 


18.4 


14.5 


14,4 


12. Turner, Arts, 194546 


25.8 


26.3 


24.2 


13- Central State, Arts, 19^5 


18.6 






iL Midwest Tech., Engr., 19^6 


21.4 


20.4 




15- Middle State, Engr,, 19^6 


20.2 


17-7 




16. Midwest City, Engr., 19^6 


19.0 


15-1 




17. Southern Tech., Engr., l^-WJ 


17.5 


14.2 




18. Midwest Tech., Agri*, 19^6 


22.1 


22.2 




19- Midwest State, Bus., 19^6 


17-2 


12.7 




20. Littletown State, Bus., 1946 


18.9 


16,5 





GROUPS MattDING VETERANS WHO RETURNED TO COLLEGE AFTER WAE SERVICE 



21. Eastern City, Arts (MN-1945) 


15-4 


14,2 




22. Adams, Arts (MN-1945) 


24.5 


20.4 




23. Stewart, Arts (MK-1945) 


23-5 


20.2 




24. Midwest Tech., Engr. (M-1939) 


27-4 






25. Midwest Tech., Agri. (MN-1939) 


18.7 







HOW COLLEGE STUDENTS SPEND THEIR TIME 



293 



100 



$ 

CO 

50 



O 

I 25 

















.. 


o 












k t 


OO 












J 99 


oooooo 












o 


.... 


oo 
o 




I o 

AAAA 


ooo 
loooo 

000 


.. 


i O 

I oooooo 













:** 


' 000 














o 



m m 

3 nours 
or less 



i Median 
Veterans 
o Nonveterans 



W 



-7 hours 



m m 

8 hours 



PER CEIflT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



o 





120 



110 













4 


. __ ^ ^ 





i ^ 






"~~"~~~~- 


, -< 


' 










Veterans 
o HonveteraruE 


3 


3 hours 4-7 hours 8 hours Total 
or less or more group 



AAG OF MEDIAE COLLEJGE GROUP FOR SPECIFIED CAOEGORIES 



FIGURE 50. HOURS PER WEEK SPENT IN AUBUETICS : ITEM,22(c) 



294 ADJUSTMENT TO COLUEG! 



athletics and physical recreation than male students. 
Amount of participation in athletics had essentially 
no relation to Adjusted Average Grade. 

In coding the responses, the following categories were used: (A) 3 hours 
or less; (B) 5 to 7 hours; and (C) 8 hours or more. More than kO per cent of 
the students, in the median group, reported three hours or less of athletic 
activity, while about 25 per cent spent eight hours or more,, The median per 
cent in the "8 hours or more" category was about 5 per cent greater for non- 
veterans than for veterans . The amount of variability among colleges does 
not appear to be large. As would be expected, women students generally 
spent much less time in physical recreation than men; in the median group 
(from the nine colleges where women f s questionnaires were analyzed), only 8 
per cent reported spending eight hours or more in athletics It may be 
appropriate to recall that the questionnaire was administered in the spring; 
different results might be obtained from questionnaires filled out at some 
other time of the year. 

The relationship of amount of athletic activity to AAG is very slight. 
The difference between the median of the mean AAG values for the extreme 
categories amounts to no more than 3 points. In only one of the basic 
groups was a category mean AAG found to be significant at the 1$ level; 
nonveteran students at Littletown State who were in the middle category 
were found to be significantly low. The hypothesis that a considerable 
amount of participation in athletics is detrimental to scholarship is not 
borne out by these findings; different results might, of course, be obtained 
from questionnaires administered^ say, in the fall term. Because of the 
lack of relationship to AAG, obviously this item cannot account for the 
superiority of veteran students. 

Extracurricular Activities Another type of activity which is of interest 
from the standpoint of possible veteran -nonveteran differences is partici- 
pation in extracurricular activities. In Item 22(d) of the questionnaire 
they were described as "other organized extracurricular activities (except 
social affairs)." The results of the analysis of this item are shown in 
Figure 51. 

Nonveteran male students consistently reported 
spending more time in extracurricular activities than 
veterans, and women spent more time than nonveteran 
males in this type of activity. Amount of participa- 
tion was unrelated to grades relative to ability. 

In coding Item 22(d), the following categories were used; (A) 1 hour 
or less, (B) 2 to 3 hours,, and (C) k hours or more. The amount of time 
spent in extracurricular activities, as defined in Item 22 (d), appears to 
be rather small. More than 60 per cent of the nonveterans and 70 per cent 
of the veterans, in a typical group ^ spent one hour or less per week in 
extracurricular endeavors, while only about 20 per cent of nonveterans and 
15 per cent of the veterans spent more than three hours in such activity. 



HOW COLLEGE STUDENTS SPEND THEIR TIME 



295 



100 



-p 

I 
















@m 












1 @@@@ 


o 










I ooooooo 










: 


'00 












o 
o 

















o 












oo 




000 









00 





oooo 








ooooo 




ooooo 








oo 


@ 





Median 
Veterans 
o Nonveterans 



m m 

1 hour 
or less 



m m 

2-3 hours 



m m 

ij. hours 
or more 



PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 



1^4-0 



130 



120 



110 













( 


-""""""""" """*"" 


^^"^ _ 


\ 




< 


*""""" -** 


^^-*~~"~"~~*~~ 


) 










m Veterans" 
o Nonveterans 




1 hour 2-3 hours 4 hours Total 
or less or more group 



MEAN AAG OF MEDIANT COLLEGE- GROUP FOR SPECIFIED CATEGORIES 



FIGURE 51. HOURS PER WEEK SPENT IN ORGANIZED EZTRACUBRICULAR ACTIVITIES: 
ITEM 22(d) 



296 ADJUSTMENT TO COLLEGE 



There is little doubt that greater participation is characteristic of non- 
veterans^ however; in eleven of the twelve baaic groups a larger proportion 
of nonveterans than of veterans spent more than three hours per week in 
extracurricular activities, and in the twelfth group there was no difference. 
Such consistency in the direction of the results would be expected to occur 
by chance less than once in a hundred times. Women report a still greater 
amount of extracurricular participation than nonveterans; considering only 
the nine groups for which women were studied, the median percentage in the 
"4 hours or more" category is 20 for male nonveterans and almost 30 for 
women. 

The lower portion of Figure 51 shows that amount of participation in 
extracurricular activities has no consistent relation witii Adjusted Average 
Grade. The median values of the mean .AAG are almost identical for the ex- 
treme categories. The suggestion from the figure that a moderate amount 
of participation is good for veterans and bad -for nonveterans doesn't make 
psychological sense, and probably results from chance f luctuations . In 
only one case was a category mean AAG significant at the 1$ level, and it 
is inconsistent with the above interpretation: nonveterans at Adams in the 
middle category were found to be significantly superior in mean AAG to stu- 
dents who spent a greater or less amount of time in extracurricular activi- 
ties. It must be concluded that this item is unrelated to AAG, and there- 
fore the superiority of veterans in grades relative to ability cannot be 
explained on the basis of amount of participation in extracurricular activi- 
ties. 

Social Activities It might be supposed that nonveteran students would be 
more inclined than veterans to engage in social activities and that greater 
participation in such activities would be associated with underachievement . 
Such a hypothesis was tested by the analysis of Item 22(e); the activity 
was defined as "social activities and recreation dates, parties, movies, 
etc." Three categories were employed in the analysis; (A) 5 hours Or less, 
(B) 6 to 9 hours, and (C) 10 hours or more. The results are shown in 
Figure 52. 

Essentially no difference was found between 
veteran and nonveteran male students in hours per 
week spent in social activities. Women students re- 
ported spending considerably more i^ime than men in 
such activities There was a slight and generally 
insignificant tendency for the least amount of social 
participation to be associated with higher Adjusted 
Average Grade. 

Inspection of the arrowheads representing the median percentage values 
indicates that veteran and nonveteran male students were essentially alike 
with regard to time spent in social affairs. Although the median is slightly 
higher for veterans in the 10 -hours -or -more category, the proportion is 
higher for veterans in only eight of the twelve basic groups, and for other 
item categories no consistent trend is found. About ^0 per cent of the male 
students spent 5 hours or less, and about the same proportion spent 10 hours 



HOW COLLEGE STUDENTS SPIND THEIR TIME 



297 



100 



T3 

3 

CQ 

S 

1 

O 















1 Median 
Veterans 




o 
o 
























o Nonveterans 


@@ 


oooo 









ooooo 







ooo 









oo 







oo 




O 





oo 




d 


o 


*oe 


0000 





o 









09 


ooooooo 




00 




My MN MV MET MT MN 
5 hours 6-9 hours 10 hours 
or less or more 



PEE CENT EEPORTING DESIGNATED CATEGORIES IN" EACH COLLEGE 



-x^u 

r 

(D 

? 

|130 

t* 

4J 

I 

5 120 
110 












< 

c 


'""""""----.^ 




< 






^\ N 


^^ 



< 










* Veterans 
o Nonveterai 


is 


5 hours 6-9 hours 10 hours Total 
or less or more group 



MEAH AAG GP MEJDIAH COLIiEQE GERODP FCQR SPECIFIED CATEGORIES 



FIGTIRE 52. EOORS PER WEEK SPENT IN SOCIAL ACTIVITIES: ITEM 22(e) 



298 ADJUSTMENT TO COLLEGE 



or more on dates, parties, movies, and the like* Colleges vary considerably, 
however. Quhe men's colleges (Adams and Stewart) tended to be low in amount 
of time spent in social activities, presumably because they are not coeduca- 
tional last itut ions ; Douglas and Littletovn State are among the coeducational 
colleges where students reported spending the greatest amount of time in 
social affairs. 

Women reported considerably more time spent in social activities than 
did the men. Considering only the nine groups containing results for women, 
the median percentage for men in the 10 -hour s-pr -more category is about kQ, 
and for women it is over 60. 

Again referring to Figure 52, it appears that the relationship between 
hours per week spent in social activities and AAG is slight . Higher AAG is 
associated to some extent with less social participation; it will be recalled 
that high AAG- was also associated with a tendency to worry about getting to 
know people socially Among the twelve basic groups, the relationship was 
found to be significant (at the 1$ level) only at Adams. 

In view of the great sijnilarity between veterans and nonveterans in 
amount of social activity and the lack of a marked correlation with AAG, the 
tendency for veterans to excel in grades relative to ability obviously cannot 
be a function of amount of participation in social activities. 

Attending Lectures and Concerts . Item 22 (f) was "attending public lectures, 
concerts, and other cultural activities." This activity was included because 
it was felt that attendance at such meetings might be symptomatic of interest 
in scholarly pursuits. The categories used in the analysis are (A) 1 hour 
or less and (B) 2 hours or more* The results of the analysis of this item 
are shown in Figure 53 . 

About one fourth of the male students reported 
spending two hours or more per week attending lectures 
and concerts; there was a slight tendency for nonveterans 
to attend more often than veterans . A higher proportion 
of women than men spent tvo hours or more per week in 
such activity. Amount of attendance at lectures and 
concerts is unrelated to Adjusted Average Grade. 

In ten of the twelve basic groups, the proportion of nonveterans in 
the 2-hour s -or -more category was greater than the proportion of veterans; 
so nonveterans apparently did attend lectures and concerts more often than 
veterans. !Ehe amount of the difference is small, however; the difference 
between the median percentages ia less titan five. Almost three fourths of 
the students, in the median group, reported spending one hour or less per 
week attending lectures and concerts. Of the twelve basic groups, Stewart 
and Littletown State students reported the greatest amount of attendance. 
Turner students, however, were far ahead of all others in this respect; 
half of its students reported spending two hours or more per week in attend- 
ance at lectures or concerts. 



HOW COLLEGE STUDENTS SPEND THEIR TIME 



299 



100 



75 



I 



1 

o 



25 






O 










OO 








<>> 


i OOOOOO 











ooo 
















I Median 
Veterans 










o Uonveterans 









oo 











oooo 










' 000 











oo 








r 


o 





MJT 



MT 



1 hour 
or less 



2 hour's 
or more 



EEB GEM? EEPORTraG EESIGKATED CATEGCEIES IN EACH COLLEGE 



JL^V 

| ^ 


















i< 










S 




4 


) 




> 










S 

<D I QA 


4 


^ '""" " 


4 




s> lo^ 

'Gf "i OA 


< 


^"~~^~~~ --^ 


) 




<j oJciU 






Veterans 




1 1 A 






o Honveterane 


j 


1 hour 2 lioiirs Total 
or less or more group 



MEAN AAG OF 



COLLEGE GRODP FOR SEECIFIED CAEEGORIES 



FIOTBE 53. HOURS PER WEEK SB3HT ATTEKDOTG POSLIC 
CQKCEROS^ AKD OTHER CCnOTIRAL ACTIVITIES: ITEM 22(f ) 



300 ADJUSTMENT TO COLLEGE 



The sex difference is found to be much greater than the veteran-nonveter 
difference,. Considering only the nine groups where women *s questionnaires 
were analyzed, 30 per cent of the male nonveterans and 40 per cent of the 
women, in the median group, spent two hours or more per week attending 
lectures and concerts. 

Attending lectures, concerts, and other cultural activities may or may 
not be an indication of academic interest, but it is of no value in predict- 
ing grades. In none of the twelve basic groups was a significant difference 
found, even at the 5$ level, in the mean AAG's of students in the two quest io 
naire categories. In the absence of a relationship between the item and AAG, 
the item of course cannot help to account for the veteran-nonveteran differ- 
ence in AAG-. 

Bull Sessions. The activity listed as Item 22(g) was "bull sessions," a 
type of social phenomenon which is particularly characteristic of college 
students. The purpose of the item was to determine whether or not veterans 
spent less time than nonveterans in this type of social activity, and to 
discover the relation of amount of time spent in bull sessions to Adjusted 
Average Grade. The three categories employed were (A) 3 hours or less, (B) 
k to 5 hours, and (c) 6 hours or more,, The graphic presentation of the 
findings may be found in Figure 5^. 

Bull sessions occupied four or more hours per 
week for more than half of the male students in a 
typical college group. Veterans appear to be remark- 
ably like nonveterans in amount of time devoted to 
this type of social behavior. Women students engaged 
in bull sessions about as much as the men, according 
to their reports Although the relationship between 
amount of time spent in bull sessions and Adjusted 
Average Grade is slight, there is some evidence that 
a moderate amount of time spent in this activity is 
favorable to higher achievement relative to ability. 

The positions of the arrowheads in Figure 5k which indicate the median 
percentages show that veteran and nonveteran students are very similar with 
respect to amount of time spent in bull sessions. In the median group, al- 
most 50 per cent reported spending three hours or less and about 30 per cent 
said they spent six hours or more. Engineering students tended to spend 
less time in bull sessions than the liberal arts atudenta, perhaps because 
a greater proportion of their time is occupied with laboratories and other 
class meetings. Women students reported about the same amount of time 
spent in bull sessions as did the men. Students at Stewart seem to be 
particularly addicted to bull sessions: about half of the students reported 
spending six or more hours per week in this activity. 

The median values of the mean AAG 8 a, plotted in the lower half of the 
table, seem to show a slight tendency for a moderate amount of time spent 
in bull sessions to be associated with higher grades, relative to ability. 
Although the differences are very slight, two of the Category B mean AAG's 



HOW COLLEGE STUDENTS SPEND THESR TIME 



301 



100 



-p 
to 



CD 
O 

s 















| Media] 


i 














* Veterans 














o Nonve" 


berans 




O 

















00 














& 


o 


































oooo 












00 


> 


o 













o 















> 


o 




o 






o 




L 


oo 




o 




e 


o 




r e* 


oo 









**** 


> ooooooo 







o 










oo 







ooo 










o 









3 hours 
or less 



MV MN 

4-5 hours 



MN 



6 hours 
or more 

PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 





! 



130 




-P 
05 



3 120 



110 













< 


__ -""' 


^~^ ^__ 






< 


>-"""" "" 


^ 












o Veterans 
o Nonveterans 




3 hours 4-5 hours 6 hours Total 
or less or more group 



MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 54. HOURS PER WEEK SEENT IN BULL SESSIONS: ITEM 22(g) 



302 ADJUSTMENT TO COLLEGi 



(those for non-veterans at Miller and Eastern City) are significantly greater 
(at the 1$ level) than the means for students in other categories. In ten 
of the twelve basic groups of nonveterans the students in the middle cate- 
gory earned higher AAG s s on the average than students in the two extreme 
categories ; obtaining ten out of twelve differences in one direction is 
significant at the % level. For veterans, no significant mean MG values 
were found; veterans in the middle category were superior in nine of the 
twelve basic groups. It appears that at least for nonveterans a moderate 
amount of time spent in bull sessions is associated with higher achieve- 
ment relative to ability. 

Paid Employment. Part-time employment is a factor which might be expected 
on logical grounds to interfere with successful academic work, taking as it 
does hours which might otherwise be devoted to study. Because of the differ- 
ence between veterans and nonveterans in subsidies through the GI BiH, it 
might also be supposed that nonveterans would need part-time employment more 
often than veterans * If both of these hypotheses are true, one would expect 
veterans to earn higher grades, relative to ability, than nonveterans. It 
is the purpose of Item 22(h) to test both hypotheses. 

The statement of the activity was "paid employment . " Two categories 
were employed in the analysis^ (A) 1 hour or less, and (B) 2 hours or 
more. The results of the analysis are shown graphically in Figure 55. 

Colleges vary considerably in amount of time 
spent in paid employment by their students; in the 
median group, almost a third worked two hours or 
more per week. Somewhat fewer veterans than non- 
veterans reported two hours or more per week in paid 
employment Women students reported less time spent 
in paid employment than men. Hours of paid employ- 
ment has little relation to Adjusted Average Grade, 
although there is a slight tendency among nonveterans 
for those students who work most to earn lower grades 
relative to ability. 

The amount of variability among the twelve basic groups may be shown 
by one or two examples* Students at Evans jnoat frequently reported working 
two or more hours per week, the percentages being 36 for veterans and 57 
for nonveterans. On the other hand, only 8 per cent of veterans at Adams 
and 7 per cent of nonveterans at Douglas were in this category. As implied 
by these examples, the variability among colleges, is somewhat greater 
for nonveterans than for veterans, which is reasonable in the light of the 
subsidies provided veterans by the GI Bill. Konveterans in the two -hours - 
or-more category were more frequent than veterans; the percentage was 
higher in ten of the twelve groups, with one tie. The difference between 
the median groups is, however, less than 10 per cent. Women students were 
less likely to be employed part-time than the men; only at Turner, where 
to per cent of the women students said they worked two hours or more, does 
the proportion of women in Category B exceed that of the male nonveterans. 



HOW COLLEGE STUDENTS SPEND THEIR TIME 



303 



I 



100 



75. 



50 



25 



9 


o 











00 








m@ 1 oo 
I oo 
m [ ooo 














O 


f Median 




o 






m Teterans 








o 


o Nonveterans 









1 oo 

1 OOO 








$$ 














00 








* 


o 




MV m m MN 


1 hour 2 hours 


or less or more 



PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 
s 


















? 

g 130 

8 

43 

3 120 
110 


^ 


i ^ 


t 




< 


"""""""""""- ^< 


. 








Veterai 
o Uonvetc 


is 
araiis 


1 hour 2 hours Total 
or less or more group 



MEAN AAG OF MEDIAJST COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 55* HOURS PER WEEK SPENT IN PAID EMOPL01MEKT: ITEM 22(h) 



304 ADJUSTMENT TO COLLEGE 



The median values of the mean AAG f s, shown in the lover portion, of 
Figure 55, show only a very slight relationship "between hours spent in 
paid employment and Adjusted Average Grade . There is practically no dif- 
ference between the two medians for veterans, but for nonveterans there 
is a slight tendency for greater time spent in paid employment to be associ- 
ated with lower AAG. In eight of the twelve groups of nonveterans, the mean 
AAG for those working two hours or more is lower than that for thos.e working 
one hour or less, and in two groups there is no difference. At Midwest City 
the mean AAG is significantly lower (at the 1$ level) for nonveterans who 
worked most* However, in another instance, at Adams, nonveterans who worked 
two or more hours per week were significantly higher in mean AAG- than those 
who worked one hour or less It must be concluded that the relationship is 
slight at best, and amount of paid employment cannot be considered an impor- 
tant factor in accounting for the general superiority of veterans in grades 
relative to ability. 

Other Non-Boutine Activities . The final item in the list of activities, 
included as Item 22 (i) , is "other non-routine activities." The categories 
used in the analysis were (A) 1 hour or less and (B) 2 hours or more. The 
findings axe shown only in Appendix Table 22(i). 

About one fourth of the students reported spend- 
ing two hours or more in non-routine activities, and 
there was little difference between veteran, and non- 
veteran students. Amount of time reported as spent 
in "non-routine activities" bears no significant rela- 
tion to Adjusted Average Grade* 



IJnrequired Academic Pursuits 

In addition to the various parts of Item 22, which required the stu- 
dent to indicate how he spent his time in a typical week, two other items 
were included which it was hoped would give indications of academic inter- 
est. These items pertained to scholastic activities not required as prepa- 
ration for any course assignments. The hypothesis to be tested is that 
veterans show, by voluntary participation in intellectual activities, a 
more genuine scholastic interest than nonveterans, and that this interest 
is reflected in higher grades relative to ability. 

Voluntary Heading and Study. The first of these questions, included as 
Item 24, was, "About how many hours did you spend during the past seven 
days in reading or studying materials which axe related to courses you are 
taking but which are not a part of course requirements?** The categories 
used in the analysis were (A) less than one hour, (B) one hour up to two 
hours, and (C) two hours or more* The results of the analysis axe shown 
in Figure 56. 



HOW COLLEGE STUDENTS SPEND THEIR TIME 



305 



100 



p 



o 

I 















1 Median 






















4ft 




Veterans 





000 






1 


00 


o Nonveterans 


@@ 1 oo 






! 


oo 




1 ooo 






p 


L OO 




' 


oo 





o 


i 


oooooo 




* 


00 


* 


> oooooo 















ooooo 








W MN MV M MV m 
Less than One up to Two hours 
one hour two hours or more 



PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 

r 



! 

> 130 

< 

*a 

$ 

CQ 

3 

120 
110 












< 









i 





> { 


, 












Veterans 
o Uonveterans 




Less than One up to Two hours Total 
one hour two hours or more group 



MEAH AAG OF MEDIAN COLLEGE GROUP FOE SPECIFIED CA135GORIES 



FIGURE 56. HOURS PER WEEK SPENT BT VOLUNTARY READING AKD STUDY: ITEM 



306 ADJUSTMENT TO COLLEGE 



More than 60 per cent of the students spent 
fever than two hours per week in unrequired study. 
Veterans participated in such study more than non- 
veterans to a very slight extent , and nonveterans 
participated more than women. Amount of reading or 
study not required for any course is unrelated to 
Adjusted Average Grade. 

As is shown "by the arrowheads in Figure 56 , a"bout 35 P** cent of the 
nonveterans and kO per cent of veterans, in the median group , spent two 
hours or more per week reading or studying things not required for any 
course,, However, in only seven of the twelve "basic groups was the propor- 
tion of Category C responses greater for veterans than for nonveterans. 
Women consistently reported less time devoted to such study than did male 
nonveterans . 

The lower part of the figure shows clearly that there is essentially 
no relationship between amount of unre quired study and AAG. Only at Midwest 
Tech, of the "basic groups, are category means found to be significant at 
the 1$ level, and here the tendency is for less unrequired study to be 
associated with higher grades relative to ability. A similar finding was 
obtained for interrupted veterans at Eastern City. The evidence certainly 
does not favor the hypothesis that a greater amount of unrequired study is 
an indication of superior achievement relative to ability. 

Attendance at Evening Lectures. The second item which it was hoped would 
provide some indication of intellectual interest was Item 25: "How often, 
during the past four weeks, have you gone to evening lectures given by 
visiting lecturers or local faculty members but not required by any 
specific course?" Two response categories were used: (A) attended no 
evening lectures, and (B) attended one or more evening lectures. The re- 
sults are shown in Figure 57. 

About three fourths of the students in a 
typical group had attended no lectures, although 
there was considerable variability among colleges. 
Veterans and nonveterans were very similar with re- 
gard to lecture attendance; women attended slightly 
more often than men. There was a 'tendency for those 
who had attended lectures to earn higher grades in 
relation to ability than those who had not. 

An indication of the amount of variability among colleges in lecture 
attendance is provided by the upper part of Figure 57. At one institution 
(Midwest City) more than 90 per cent had attended no evening lectures in 
the four-week period, while at another (Stewart) all but about half had 
attended lectures. The differences among colleges may reflect, in part, 
the amount of opportunity to attend such lectures during the four weeks 
before the questionnaire was filled out. In the median group about 75 per 
cent had attended no lectures. 



HOW COLLEGE STUDENTS SPEND THEIR TIME 



307 



_1_WW 




o 












ooo 













oo 








o 75 

-p 


^ 












oo 






1 





o 








CO 





o 
o 





o 


1 Median 


o 50 










Veterans 
o Nonveterans 











-p 












ti 




o 




o 




s 









o 




JH 






M 


oo 




* 






L . 


L O 








@ 


o 











oo 










@ 


oo 














o 




m MN W MN 


Attended Attended 


none one or more 



PER CENT REPORTING DESIGNATED CATEGORIES IN EACH COLLEGE 



O 
0) 



-p 

3 



130 



120 



110 











4 


, - -"^ 


t 


i 


< 


^-^^^^ 


* c 


d 






Veterans 
o Nonveteraj 


IS 


Attended Attended Total 
none one or more group 



MEAN AAG OF MEDIANT COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 57. ATTEimANCE AT EVENING LECTURES DURING PREVIOUS FOUB 
"WEEKS : ITEM 25 



308 ADJUSTMENT TO COLLEGE 



The similarity "between veterans and nonveterans in lecture attendance 
is very great $ there is a slight tendency for women to report attending 
more often than men. 

Students who had gone to one or more lectures in the four-week period 
earned higher grades than those who had not in nine of the twelve veteran 
groups and eight of the nonveteran groups. Of these twenty-four differ- 
ences, only four were significant, and these at only the 5$ level of confi- 
dence. The association between lecture attendance and MG is not marked, 
although in the expected direction. The superiority of veterans obviously 
cannot "be accounted for "by whatever characteristic is measured by this item. 



Summary 



This survey of how students spend their time indicates that differences 
between veteran and nonveteran students are generally slight. Veterans 
spend less time than nonveterans in attending classes, presumably because 
in most institutions they have been excused from physical education or 
military science requirements. They also tend to spend less time in extra- 
curricular activities, which is consistent with the hypothesis that veterans 
are more serious-minded than nonveterans and are less inclined to engage in 
"frivolous * pastimes. Also consistent with this hypothesis was the slight 
tendency found for veterans to spend more time than nonveterans in studying 
and in voluntary reading and study of materials not assigned by an instructor. 
Nonveterans exceeded veterans sligjhtly in amount of time spent in athletics, 
attending lectures and concerts, and, in part-time paid employment. No dif- 
ferences were found in time spent in social affairs, bull sessions, and 
"other non-routine activities." 

Women students in general resembled tne male nonveterans at the same 
college in amount of study and time spent in bull sessions. They reported 
spending more time than male nonveterans in extracurricular and social 
activities and in attending lectures and concerts . They spent less time 
in athletics, attending classes and laboratories, in paid employment, and 
in voluntary reading and studying. 

Amount of time spent in attending classes, relative to the student's 
own institution, is presumably an index of the course load taken by the 
student. Those students with heavier loads, then, tend to earn higher 
grades, relative to their ability, than students taking lighter loads. 
This finding is consistent with the results of similar studies (reported 
in Chapter III) where course load was obtained directly from the transcripts. 
Since nonveterans take heavier loads, our usual reasoning would lead us to 
the conclusion that nonveterans should be expected to excel in AAG. This 
reasoning is supported by the sign test, which turns out to be highly sig- 
nificant for this item. However, since the lower course load of veterans 
is presumably due to different factors than those which account for the re- 
lationship between load and AAG, the conclusion that nonveterans should be 
expected to excel in AAG is probably not justified. 



HOW COLLEGE STUDENTS SPEND THEIR TIME 309 



In general, amount of time spent in the various types of activity had 
little or no relationship to Adjusted Average Grade . There was a tendency 
for study time to "be positively related to AAG, Very slight tendencies 
were noted for those ho spent the most time in social activities to earn 
lever grades relative to ability. Those vho attended evening lectures 
tended to earn slightly higher AAG f s than those who did not attend. There 
is a suggestion that a moderate amount of time spent in bull sessions is 
associated ith higher grades relative to ability than more extreme amounts 
of time. Time spent in athletics, paid employment, extracurricular activi- 
ties, participation in cultural activities such as lectures and concerts, 
and voluntary reading and study are unrelated to Adjusted Average Grade. 

Only one of the items dealing -with, eipenditure of time yields sign 
test results which are significant. Even for this item, vhich is concerned 
with course load, closer scrutiny suggests that the relationship obtained 
is of little, if any, value in interpreting veteran-nonveteran differences 
in Adjusted Average Grade. It appears, then, that the tendency for veteran 
students to earn higher grades in relation to ability cannot be accounted for 
on the basis of differences in amount of time devoted to any of the activi- 
ties studied by analysis of the questionnaire responses. 



310 



Chapter IX 
THE GI BILL 



The educational provisions of the GI Bill "brought about many changes 
in the educational plans of young men eligible for these benefits. Undoubt- 
edly, some who would have been unable to attend college decided that they 
could go; others who would have attended college in their home community 
went away to school; and still others shifted from a less expensive to a 
more expensive college. In this study, those veterans who were brought into 
college by the GI Bill were selected for more detailed analysis because of 
their unique significance for higher education. Their success in competing 
academically with the students who would have attended without this aid is 
distinctly relevant to the basic question of who should go to college. 
Primary emphasis was placed on college grades relative to ability as the 
measure of academic success in these comparisons. 

Postwar Educational Plans and the GI Bill. In Item 8(0), students were 
asked : "Do you think you would have come to college after completing your 
military service if the financial aid provided by veterans 1 benefits had 
not been available to you?" Jor purposes of analysis, all four of the ques- 
tionnaire responses were considered: (A) yes, I am quite sure I would have 
come anyway, (B) I probably would have come, but I'm not sure; (C) I might 
have come, but I probably would not have come; and (D) no, I am quite sure 
I would not have come to college. Results of the analysis are shown in 
Figure 58. 

The educational provisions of the GI Bill brought 
a substantial number of veterans into college who would 
probably have missed college without this aid. Perhaps 
20 per cent of veterans in the typical freshman group 
belong in this classification. Colleges differed de- 
cidedly in the proportion of their students who would not 
have attended any college without the federal scholar- 
ships The students brought into college by this aid 
were quite similar in academic performance, relative to 
ability, to the remainder of the class; the slight dif- 
ference which appeared favored the group which needed 
the financial aid. 

In the typical basic college group, about 60 per cent of the veterans 
reported that they would definitely have attended college without veterans 1 
benefits, and slightly less than 20 per cent reported that they would prob- 
ably have done so. The remaining 20 per cent divided about equally between 
those who probably would not have attended and those who definitely would 
not have done so Colleges differed widely in the proportion of veterans 
reporting that they probably or definitely would not have come without the 
government^ financial aid; proportions in these two categories combined 
ranged from about one in 100 to more than one in three, in the twelve basic 



THE GS BILL 



311 



100 



1 



I 



1 

o 



75 



ifi 25 
















4 





















tee* 
@ 
i 



& 
. @@ 
** 


~ 

>*- 

r @^@ 
1 


Would nave Probably Probably Would not 
come anyway would have would not hare come 
come have com 



Median 



CENT GHOOSHSKj' 2jQESIO3A5CED CAfEBGrCEIES IN EACH COLUBSE 



JL-2U 














0) lJ,A 














8 llw 














S 








4 






(D 






^ 


r- - - 






se 


i 


^^ 


^^ 




1 


^ 


130 




*^>^^ 


^ 








5 




"S 


r^ 








rtf 














5 












































Sj 120 


























Would nave Probably Probably Would not Total 
come anyway would nave would not nave come group 
come nave come 



MEAK AAG OF MEDIAH COLIiBGE GEROHP FC3R SPECIFIED CATEGORIES 

FIGURE 580 IMPORTANCE OF FINANCIAL AJDD PROVIDED BT VETERANS 1 
BENEFITS IN DECISION OF VEEERAN STDODENTS TO COME TO COLLEGE: 
ITEM 8(0). 



312 ADJUSTMENT TO COLLEGE 



groups. It must, of course, "be remembered that the question dealt only with 
whether or not the veteran -would have attended college at all, whether or 
not he would have chosen the particular college he was attending or whether 
or not he would have needed financial aid in the form of a scholarship or 
student loan was not considered. Thus, although students whose decision to 
attend college depended heavily on GI "benefits constituted a very small 
proportion of the freshman groups studied at Adams, Stewart, and Douglas, 
it cannot necessarily be inferred that the GI "benefits had little influence 
in their attending these colleges. Almost one fifth of the freshman veterans 
enrolled in liberal arts at Evans, Miller and Eastern City and in engineering 
at Midwest City were students who would definitely not have attended col- 
lege without GI educational benefits. Eelatively large proportions of the 
three groups (Turner, Southern Tech, and Central State), which included 
students who entered as freshmen in 19^5, considered the benefits essential. 

It is apparent from Figure 58 that the students whose college careers 
were made possible by the GI Bill performed slightly better, relative to 
ability, than did the rest of the veteran students. The group who definitely 
would not have come without veterans' benefits had the highest median with 
respect to AAG, although its margin of advantage was quite small Only the 
group who probably would have come without veterans ' benefits had a median 
appreciably different from 135 > i^ s median was about 130 , When the pattern 
of differences in the twelve basic colleges is examined, none of the cate- 
gories shows a statistically significant tendency to be above or below the 
general average. When the significance tests were made in each college 
group, it turned out that students who probably would have attended were sig- 
nificantly low in AAG at the 1$ level at Evans and at the 5$ level at Miller. 
Those who would surely have attended were significantly high at Evans, (it 
should be noted here that no significance test was made when less than one 
per cent of veterans in the college group chose the category. ) On the 
whole, these results indicate that the students whose college careers were 
made possible by the educational provisions of the GI Bill performed 
slightly better, relative to ability, than the veterans who would or prob- 
ably would have attended college in any case. 

Some Additional Comparisons. Veterans who probably would not have come to 
college without the GI Bill were compared in several additional ways with 
those who probably would have come without this aid. First, by means of 
analysis of covariance, the appropriateness of using the same prediction 
equations for both groups was tested. Data for Miller University were used 
for this analysis, since the numerically largest group of veterans who would 
not have come were included in this group. Second, by means of the usual 
t-test of significance of differences between percentages, significant dif- 
ferences-between veterans and nonveterans in responses to questionnaire items 
were identified, using Miller and Central State data. Finally, some atten- 
tion was given to the possible effect on the relationship between item 
responses and AAG produced by the presence of the "would not have come" 
group. In the discussion which follows, veterans who reported on Item 8(0) 
that they definitely or probably would have attended college without veterans 
benefits will be called the "would have come" group or Subgroup A; those 
who reported that they probably or definitely would not have attended without 
this aid will be called the "would not have come" group, or Subgroup B. 



THE GS BILL 313 



IB. predicting first-year grades, the same regres- 
sion equation may properly "be used in predicting grades 
for yeterans who would have come "without veterans' "bene- 
fits and for yeterans -who would not Jiaye come without 
these benefits, insofar as the yeterans at Miller Uni- 
versity are concerned. 

An analyais of coyariance was done in order to proyide a more rigorous 
check on the "basic similarity or difference of the groups in ability relative 
to achievement* It is apparent from Table Vf, in which the analyais of co- 
variance results are presented, that there is no statistically significant 
difference between the groups for any of the three hypotheses tested. The 
results- for this group of students at Miller indicate that the same predic- 
tion equation would be suitable for both groups , It may also be observed 
from. Table ^7 that there are only slight differences in the mean American 
Council on Education Psychological Examination score and in mean High School 
Rank. The mean is very slightly higher for the "would have come" group for 
both predictors . The "would not have come" group, on the other hand, 
earned average grades which were a trifle higher than those of the "would 
have come" group, 

Analysis of questionnaire responses Indicates that 
the typical veteran who would not have come without GI 
benefits was somewhat older , saw more service, came from 
a family with less educational background, was less se- 
cure financially, was more likely to be married, and was 
less likely to be planning to enter a profession than 
the typical veteran who would have come irithout GI 
benefits. Although these differences were statistically 
significant at both Miller and Central State, there was 
clearly much overlapping between the two groups in these 
respects . 

Students who would have come differed from those who would not have 
come in their response to a number of questionnaire items a In this report, 
only those responses which showed a significant difference between the two 
groups for both the Miller and Central State students will be considered. 
Figures 59 a&d 60 present the results of this phase of the analysis . (In 
general, one response is sufficient to indicate the nature of the relation- 
ship within an itemj more than one response is reported only when the addi- 
tional response adds to the understanding of the relationship.) 

The most conspicuous difference is found in Item 8(m) of Figure 59* 
The "would not have come" group, to a large extent, includes the same 
veterans who appear in the group who probably would not have attended col- 
lege if they had never entered the service j the "would have come" group, 
however, contained only a small proportion who fell in this category 
Other responses indicate that the "would not have come" group, as compared, 
with the w would have come" group, had served longer, had been overseas 
longer, had been separated earlier, were more likely to have decided to 



314 



ADJUSTMENT TO COLLEGE 



Table 47 
COMPARISON' OF AVERAGE GBADES EARMD BY VETERAN MALE STUDENTS 

(A) MO HROBABLT. WOULD HAVE ATTENDED COLLEGE AM) 

(B) WEO EROBABU TOLD NOT HAVE ATTENDED COLLEGE, 

WITHOUT VETERANS f BEMFITS 
Miller Ifaiversity, College of Arts and Science, Freshmen, 1946-1947 

I. Correlations, Means, and Standard Deviations: 



Variable 


Sub- 
group 


Correlation with: 


Mean 


SD 


N 


ACEE 


H. S. Rank 


First -Tear 
Avg. Grade 


1. ACPS (1946) Total 


A 




.29 


A3 


112 


20 


279 


(rav score) 


B 




19 


.39 


110 


20 


146 


2. High School Hank 


A 


29 




A4 


14.9 


3.5 


279 


(converted score) 


B 


.19 




A3 


14.8 


2.8 


146 


3. First-Tear College 


A 


A3 


,44 




1.48 


-58 


279 


Average Grade 


B 


39 


A3 




1.51 


-53 


146 



II. Multiple Correlations (Variables 1 and 2 vs. Variable 3): 



Sample 



Subgroup A 
Subgroup B 
Combined Group 



Multiple R 



54 
-53 
53 



III. Analysis of Covariance Results: 



Hypothesis 


Chi- 
sojiare 


Degrees 
of 
Freedom 


Probability that a value 
greater than obtained 
would arise by chance 


A. Equality of errors of estimate 
B. Equality of slopes 
C. Equality of intercepts 


0.702 
0.724 
0.890 


1 
2 
1 


Between .30 and .^0 
Between .50 and -70 
Between .30 and .50 



U. Difference between Subgroups with Ability Held Constants 



Superior subgroup 

Advantage expressed in, grade units 

Advantage expressed* in standard error of estimate units 

Per cent of subgroup B excelling the average subgroup A veteran 

Level of significance of difference (from IIIC above) 



Subgroup B 
(Would not have come) 

0.05 
0.10 

54 
Not significant 



THE GI BILL 



315 



RESPONSE 



Six years or more 
between high school 
and college : 
Item 6(b); Category A 

Three years or more 

on active duty: 

Item 8(b), Category D 



Sub- 
group 



MTT,T."F?R DNIVERSITT CENOOEIAL STATE 

Per Cent Per Cent 

20 40 60 80 1CX) 20 kf) 60 80 



. 
B 



13 

34 



- 




1CJO 






- 




Eighteen months or 
more on overseas 
duty ; 
Item 8(g); Category E 

Separated from service 
in 19*4-5 or earlier s 
Item 8(h), Category A 



B 36 

A 15 
B 30 





L 



First decided to go to 
college while in ser- 
vice; 
Item 8(3), Category 2 

Decided to attend 
after discharge 
from service; 
Item 8(3), Category D 

Unlikely to have come 
if he had not entered 
military service; 
Item 8(m), Categories C, D 

Married^ now or 

previously; 

Item 3^> Category C 



A 2k 
B 67 



. 
B 




8 
76 







20 

2 
73 

11 
,20 




pH 




LEGEND: 

Difference in 
Difference in 



significant at 1% level 
significant at 5$ level 



FIGURE 59. PER CENT MAIOITG VARIOUS SEIECTED RESPONSES TO QUESTIONNAIRE ITEMS 
AMONG VETERANS WHO PROBABLY WOUID SAVE ATTENDED WITHOUT "VETERANS' BENEFITS (SUB- 
GROUP A) AND THOSE WHO PROBABLY WOULD NOT HAYS ATTENDED WITHOUT TBESE BENEFITS 
(SUBGROUP B). 



316 ADJUSTMENT TO COLLEGE 



attend college while in the service or after discharge^ had been out of 
school longer, were older, were more likely to be married, were more likely 
to have had a full-time job for six months or more between high school and 
college, were more likely to be working part-time, and were more likely to 
worry about mfci.:aig endB inset financially* !0aey were more likely to give 
"preparation for a better-paying job" aa their first reason for being in col- 
lege, and less likely to state as their first reason that they needed a 
degree in order to enter their chosen prof esBion* Ihey were less likely 
to be planning to enter a profession requiring graduate training, and less 
likely to have come from a family in which the father had completed high 
school. The head of the family was .more likely to hare had an annual in- 
come of less than $2,000 during: their high school years . 

Hhen veterans "who would not hare come are compared with non-veteran 
students, using Figure 60, the differences in nearly all respects are 
greater than between the two veteran groups. One notable exception has to 
do with part -time employment , On this item, the veterans who would not 
have come are more like the male nonveteransj the veterans who would have 
come are less likely to be working than either of the other two groups. 

Tlie differences in proportions between the veterans who would not have 
come and the other groups should not be allowed to obscure the fact that 
there was also substantial overlap between those who would have come and 
those who would not have com on all items except 8(m) and 8(j) 

Xaese results fit the hypothesis that the group of veterans brought 
into college by the educational provisions of the GI Bill were predominantly 
men who would have needed financial aid to attend college even if they had 
not spent several years in military service. A distinctly smaller propor- 
tion of these veterans were men who would have given up earlier plans for 
a college career (because of their greater ge and responsibilities) if 
veterans ' benefits had not been provided 

It was thought that the presence of students who 
were brought into college by veterans ' benefits might 
have affected the relationship between questionnaire 
responses and Adjusted Average Grade in the veteran sub- 
groups. 3!he results indicate that this effect was. not 
likely to be a major influence in determining these re- 
lationships., 

In the main questionnaire analysis, a number of responses were found 
to be significantly related to AAG for the veteran subgroup but not sig- 
nificantly related for the nonYoteran #ubgroup 7 in the various colleges * 
To some extent these differences represent chance f luctuations . Moreover, 
in the college groups where the veteran subgroup is larger than the non- 
veteran subgroup, the F-test would be more sensitive for the veterans. How- 
ever, it is also possible that the presence in the veteran subgroup of stu- 
dents who were brought into college by the QI Bill may account in part for 
the different results * In order to make a rough cisjeck on this hypothesis, 



THE CI BILL 



317 



KESPOISE 



Six months or more 
of full-time work 
between nigh school 
and service or 
college ; 
Item 9(b), Category C 



Sub- 
group 



m 



MILLER UNIVERSITY CETORAL STATE UNIVERSITY 

Per Cent Per Cent 

9 2.0 40 60 80 100 9 20 55 60 80 100 



38 

68 

7 

10 




H 

u 



27 

59 

9 

13 




m 



Preparation for a 
better -pay ing job 
chief reason for 
attending college; 
Item 10, Category A 



College degree 
essential to chosen 
profession chief 
reason for attending 
college : 
Item 10, Category B 




Planning on 

profession which 

requires graduate 

trainings 

Item 11, Category A 



m 



32 

63 

16 




Two hours or more 
per week spent in 
paid employments 
Item 22(h) J> Category B 



WsA 



22 
33 

28 





Difference in $ ff s between MVsA and WsB significant at 1^ level 
V777\ Difference in %<s between WsA and WsB significant at jf, level 



FIGURE 60 (PART l). PER CEM 1 MAZING VARIOUS SELECTED RESPONSES TO QUESTIONNAIRE 
ITEMS AMONG: 'VETERANS WHO PROBABLY WOULD HATE ATTENDED WITHOUT "VETERANS 1 BENE- 
FITS (MV:A), "VETERANS WHO PROBABLY WOULD NOT HAVE ATTENDED WITHOUT THESE BENEFITS 
(MV:B), MALE NONVETERMS (MN) AND WOMEN STUDENTS (F). 



318 



ADJUSTMENT TO COLLEGE 



Born before 1924 1 
Item 32, Category A 



Bothered some or 

very much about 

TrujQcing enda meet 

financially: 

Item ^O(a), Category A 



Father's Income 
under $2,000 in 
student 1 a high 
school years: 
Item k$, Category D 



lather not a high 
school graduate: 
Item W-, Category A 



group 



MTf.T.TO UMVKttSITY CENTRAL STATE UNIVERSITY 

Per Cent Per Cent 

2p 40 bO 80 ipp Q 2p >p ^0 ^0 IpQ 



m 

F 



WsA 20 






ft> 


'//////////A 


73 
^5 
^9 




W//////////A 




\ 




| 



LEGEND: 

$m Difference in 
Difference in 



between WsA and W:B aignificant at Ijt level 
between MTsA and WsB significant at % level 



FIGOKE 60 (PAET 2). PER CENT MAKUfa YARIOUS SEI^ECIED EESPONSES TO QUESTION- 
MIRE ITEMS AMOK: TCEERA1S WHO PROBABU WOULD BATO ATTEIDED WITHOUT YETERA1S' 
BEHEFITS (W:A), VETERANS WHO PROBABLY WOULD NOT HAVE ATTEIDED WITHOUT THESE 
BENEFIT (MT:B), MALE NONVETERANS (M) AND WOMEN STUDENTS (j) . 



THE GI BSLL 31$ 



the results of the significance tests Involving association "between question- 
naire responses and AAG were examined for two college groups having a rela- 
tively large proportion of veterans vho would not have attended without GI 
aid. The two college groups, chosen for this purpose were Miller freshmen 
and Central State freshmen. In all, ten questionnaire responses were found 
to have a significant association for veterans in both groups and for non- 
veterans in neither group. (One of these was age, for which a different 
plan of categorizing was used for each subgroup.) Uaing Miller data, the 
veterans group was broken down, using Item 8(0), into those who would have 
come without the GI Bill and those who would not have come without this aid. 
By hypothesis, these responses should show a significant relationship for 
the "would not have come" and total group, but not for the "would have come" 
group. In testing this hypothesis, students who would have come (Subgroup 
A) and who also made a particular questionnaire response were compared with 
the remainder of the veteran students. Similarly, veterans who would not 
have come (Subgroup B) and who chose each response' were compared with the 
remainder of the group o 

Results of this analysis are shown in Table 48. Only three of the 
eleven responses showed the expected pattern, while six showed the reverse 
pattern. Of the three which fit expectation, the association of greater 
age with higher AAG is particularly relevant. However, when age was analyzed 
in the same way, using Central State data, it fell in the "reverse" groups 
i.e., it was significant for Subgroup A (would have come) but not significant 
for Subgroup B (would not have come) . It would appear, on this Bhowing, that 
the presence of students who were brought into college by the GI Bill in the 
veterans group did not lead to major differences between veterans and non- 
veterans in the way item responses are associated with AAG. It was judged 
that the labor involved in making a more precise evaluation of this effect 
was not warranted. 

Military Service and Educational Plans. A different approach to assessing 
the influence of the GI Bill upon college -going vas provided by Item 8(m), 
which asked: "Eegardless of how you felt about going to college when you 
left high school, do you think you actually would have gone to college if 
you hadn ! t entered military service?" This item represented an attempt to 
abstract "college -going tendency" from the complex of confusing influences 
introduced by the war. Thus, the student was asked to eliminate from con- 
sideration the influence of his service experience, the influence of being 
well above the usual college entrance age when discharged, and the influence 
of the educational benefits of the GI Bill on his plans. In contrast to 
Item 8(0), this item classified as "college -goers" men who required fi- 
nancial aid as discharged veterans but who would probably have attended col- 
lege if they had never entered the service. On the whole, it seems likely 
that eligibility for GI benefits was the outcome of military service which 
was decisive for many of the students who would not have attended if they 
had never entered the service. On the basis of responses to Item 8(m), stu- 
dents were divided into four groups: (A) would definitely have attended, 
(B) probably would have attended, (C) probably would not have attended, and 
(D) almost certainly would not have attended. Results are shown in Figure 6l 



320 



ADJUSTMENT TO COLLEGE 



Of. 



1 




H 

1 



H <D 

II 



* 

ffi Q 







ctJ O 
O tfl 



3 



O CO 



vo co 

a a 



1 



o ^4- 
tQ CV1 (D 
OS H 



H 

<D 

I 



ON cvj ^-- ' 

OJ CO O 



, : , : t i 

VO ^ |s_ -(- H o\ 

SI 1 2i d 



O ON O| ir\ r-4 co 
-=i- CM ir\ oo co co 



88 



S 3 



3 H 



VQ .it O CO 
co <N CM H 



3 

O 

53 




3 -P -P 

s a -5 

O -H O 

O O tH 

h S S 

r i 3 "13 

-P d 

^ S h 



^Q cq to ^ 

oj CD (D O 

<P -H 

4 H *d y 

O CQ f g| 
(D -H 

S 1 ^ ^ 

<D CQ ft TJ 

TJ t _ J^ <D 

<D 



i 

-p 40 

a | 

CQ od 

1 1 



I 



2 



VU T* I I T~ 

L <D fl OJ O 
M H CQ PQ 



pq 







a 



t CO -^ 
CO CO O 



1 




S> 



o 

-p 
o 

H 

SI 

CQ 
S 



s? 



X 



O 

-P 



P 
H 

d 



S 



a 



-p -p 

03 0} 

-P -P 

CO O O 



*H B CQ 
O H -H 







H -P 



H S S 
co ^ 

H 05 CO 

4? -P 



ft 



43 .p 

CQ CO 

I <D <D 

CQ -CQ 

0) <D 



S 1 

(D 0) 
jQ ^) 



O 

h 





d b ^ 

CD CD Q) 



s 



THE GI BILL 



321 



CD 
P 



! 

CO 



100 



75 

















>r 


























::~ 


>:::: 


Wo-uld Probably Probably Would not 
have come would would not have come 
anyway have come have come 



Median 



PER GEM? CHOOSING DESIGNATED CATEGORIES DT EACH COLLEGE 



150 



<D 
-P 



3 



130 



120 



110 















4 




^ 

X" 

X* 


" -^ 


4 


I 


























Would Probably Probably Would not Total 
have come would would not have come g^oup 
anyway have come have come 



MEM AAG OF MEDIAE COIIEGE GROUP FOE SPECIFIED CATEGORIES 

FIGURE 61. UZELIHOOB OF STDI3EKT f S ATTEflDlHG COLLEGE IF HE HAD 
HOT ETOERED SER?ICEs ITEM 8(m) . 



322 ADJUSTMENT TO COLLEGE 



When veterans were asked whether they thought they 
would have attended college if they had not entered the 
service, about 60 per cent were reasonably sure that they 
would have attended,, about 20 per cent said they prob- 
ably would have attended, about 10 per cent thought they 
probably would not have attended and somewhat less than 
10 per cent thought they would not have attended. The 
two groups who considered their college attendance un- 
likely tended to earn higher Adjusted Average Grades 
than the other veterans* Those who definitely would not 
have attended without military service excelled the re- 
maining veterans in all seven ,of the college groups 
which contained ten or more such veterans. Statistically 
significant differences on this item were found in three 
of the college groups 

In the typical basic group, slightly more than 60 per cent believed 
that they would have attended college if they had never entered military 
service, and about 20 per cent believed that they probably would have at- 
tended. About 10 per cent thought that they probably would not have gone, 
and somewhat less than 10 per cent were almost sure that they would have 
missed college had it not been for their service experience * These figures 
are quite similar to those for Item 8(0) . With one except ion- -the small 
group of liberal arts freshmen at Eastern City the same college groups 
which had relatively few students who considered GI benefits essential also 
had relatively few students who would probably have missed college if they 
had not entered the service. 

Item 8(m) appears to be somewhat more closely related to AAG than was 
Item 8(0) . The two groups who probably would not have gone show medians be- 
tween 135 and 1^0; the two groups who probably would have gone show medians 
just over 130. When the pattern of differences in the twelve basic groups 
is considered, there is a reasonably clear trend in favor of the two groups 
of students who would not have attended. It may be noted that, for Category 
D, among the seven groups which include 10 or more students f all seven are 
above the remainder of the group in mean AAG, a trend which is significant 
at the 5$ level* In one group, Western State, those who were sure they 
would not have gone were significantly high in AAG at the 1$ level; those 
who were sure they would have gone were significantly low in AAG at the 1$ 
level. Two differences, in other college groups, were significant at the 
5$ level* OH the whole, the slight association with AAG was favorable to 
the students for whom military service and presumably the veterans f bene- 
fits deriving from this service led to a decision to attend college. 

Although primary emphasis was placed on achievement relative to ability 
in analyzing responses to Items 8(0) and 8(m), it was thought desirable to 
secure some evidence on the relationship between responses to these items 
and ability and achievement considered separately. Accordingly, Table 49 
presents mean scores on predictors and mean average grades for engineering 
freshmen at Middle .State and for liberal arts freshmen at Central State . 
Differences tend to be small. The group who definitely would have come to 
college without GI benefits is consistently above the over-all average in 



THE 01 BILL 



323 



!Table i# 

BilWEffl COLLEGE PLAHS AS EEPOROTJ OK OTfe 8(0) A2JP 8(m) 
AID YAEIOOS MEASURES 0? APTJZEODE Affi) ACiLLKVJilMEIT 



I. Middle State University, 352 I'reshman in Engineering 



College 
Plans 


Item 


ACEE 
Total 
(raw score) 


High 
School 
Bank 


College 
Average 
Grade 


AAG 


Per Cent 
of Group 


Definitely -would 
have come 


8(0) 

8(m) 


90.1 
90.3 


15.8 
16.0 


1.44 
1.47 


134 
135 


53 
53 


Probably would 
have com 


8(0) 
8(m) 


87.5 
88.0 


16.1 
15-7 


1.48 
1.38 


136 
131 


24 
27 


Probably would 
not have come 


8(0) 
8(m) 


82.3 
78.9 


15-3 
14.0 


1.27 
1.12 


128 
126 


10 
11 


Definitely would 
not have come 


8(0) 
8(m) 


86.1 
86.8 


14.3 
15-1 


1.21 
i.4o 


127 
136 


12 
9 


Total Group 




88.1 


15.6 


1.40 


133 


100 



II. Central State University > 466 Freshmen in Liberal Arts 



College 
Plans 


Item 


Composite 
Test 
Score 


High Sch. 
Average 
Grade 


College 
Average 
Grade 


AAG 


Per Cent 
of Group 


Definitely would 
have come 


8(0) 
8(m) 


12.8 
12.7 


2.52 
2.53 


2.23 
2.22 


134 
134 


56 

a 


Probably would 
have come 


8(0) 
8(m) 


11.8 
12.3 


2-il-l 
2^5 


2.11 
2.16 


133 
133 


22 
20 


Probably would 
not have come 


8(0) 
8(m) 


12.0 
12.2 


2JJ-3 
2AO 


.16 
2.22 


136 
140 


10 
9 


Definitely would 
not have come 


8(0) 
8(m) 


12.9 
12.2 


2^0 . 
2.26 


2.22 
2.09 


137 
136 


12 
10 


^. 
Total Group 




12.5 


2.48 


2.20 


134 


100 



Includes two students who did not answer 8(0) and 8(20,). 



324 ADJUSTMENT TO COLLEGE 



both predictors in both schools, but it is- BOt always the highest of the 
four groups. The differences among the other three groups in the measures 
of aptitude favor the "probably would have come" group to a slight degree. 

pisabled Veterans . In the questions previously discussed, no distinction has 
been made between disabled veterans drawing benefits under Public Law l6 7 
which is limited to the vocational rehabilitation of veterans having a ser- 
vice-connected disability, and the veterans whose benefits derived from 
Public Law 346 or various other laws* Item '8 (n) asked: "Are you now drawing 
(or have you applied for) veterans 1 educational benefits from the Veterans 
Administration?" Responses were classified into three categories: (A) yes, 
under Public Law 16 (and any others); (B) yes, under Public Law 3^6 and/or 
any others except Public Law 16; and (C) no, I have not applied for veterans' 
educational benefits. Eesults of this analysis are shown in Figure 62, 

Veterans drawing benefits under Public Law 16 
constituted less than five per cent of the veteran 
subgroup in the typical basic group. In general, 
they were earning about as high grades, relative to 
ability, as the veteran group as a whole. 

Students who were drawing benefits under Public Law 16 accounted for 
less than five per cent of the veterans in the typical basic group. In the 
twelve groups, the highest per cent was eight and the lowest was one* The 
disabled veterans have a higher mean AAG in five college groups and have a 
lower mean AAG in seven groups, toe median of the mean AAG's is slightly 
below the over-all average It is clear that the difference between the 
disabled veterans and the remainder of the group is not statistically sig- 
nificant. When the various college groups were considered individually ^ 
one instance was found in which the mean AAG of the disabled veterans was 
significantly lower, at the 5$ level, than the mean AAG. of the remaining 
veterans. In the absence of any clear trend in the mean AAG's between dis- 
abled veterans and the group as a whole, it appears that the disabled group 
held its own in the academic competition with the other veterans, insofar 
as AAG is concerned. 

Veterans Hot Drawing Benefits, 

O^ypically, veterans not drawing benefits amounted 
to less than five- per cent of the veteran subgroup in 
the twelve basic groups. They showed some tendency to 
earn higher AAG ! s than the other veterans, but their 
advantage was not statistically significant. 

Figure 62 also shows the results for the small group of veterans who 
reported that they were not drawing benefits. These veterans amounted to 
less than five per cent of the veterans in the median basic group. The ex- 
act proportion would depend, of course, on whether or not men who served 
in the merchant marine and field services were classified as veterans or 
as nonveterans. In any case, it appears likely tliat a portion of the group 
not drawing benefits were saving them to use for later professional train- 
ing | it is of course possible that a few of these veterans were discharged 
under conditions which made them ineligible for benefits* 



THE Gl BILL 



325 



Students 

3 a I 




W 










! 










m@ 
> @@@ 




@ 


Public Public 
Law 16 La^r 3*4-6 


None 



= Median 



PER GEM 1 REPORTING DESIGNATED CATEGORIES ZW EACH C03XEGE 



I 



S 130 



110 



Public Public None Total 
Law 16 Lav 3^6 group 

MEAN AAC OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 62. "VETERANS 1 EDUCATIOHAIx BE3MPITS APPLIED 
POR: 



326 ADJUSTMENT TO COLLEGE 



There was some tendency for veterans not drawing "benefits to be above 
average in AAG Although the median of their Bean AAG's was 140, not much 
reliance may be placed on this difference in view of the small number of stu- 
dents involved. In the various- college groups, no significant association 
with AAG was found. 



Conclusions 

O^ypically, in the basic twelve college groups, about ten per cent of 
veterans would not have attended college without the educational benefits 
of the GI Bill, and another ten per cent would probably not have attended. 
The proportion of veterans who reported that they probably or definitely 
would not have attended without veterans ' benefits varied widely among the 
colleges. In most of the groups, they represented a substantial minority 
of the veteran subgroup. 

On the whole, the students who would not have attended without veterans 1 
benefits earned very slightly better grades relative to ability than did 
those who would probably have attended in any case. Ho significant trend 
was apparent for likelihood of college attendance with GI aid to "be related 
to Adjusted Average Grade a It should be remembered, of course, that the 
veteran subgroups tended to be superior to the nonveteran subgroups in Ad- 
justed Average Grade. 

Veterans who were brought into college by the GI Bill differed in a 
number of respects from those who would have attended without this aid. In 
studies of veterans in two colleges y it was found that the typical student 
who attributed considerable weight to the GI Bill in his decision to come 
to college was older, had been out of school longer, had served longer, had 
been overseas longer, came from a family with less educational background, 
was more likely to be married, was probably less well off financially, and 
was less likely to be planning to enter a profession than were the other 
veterans. There is, to "be sure, much overlapping between the two groups in 
nearly all of the specific characteristics. Perhaps one fourth of the stu- 
dents brought into college by the GI Bill were students who would have at- 
tended college if they had never entered the service; less than ten per cent 
of those who would have attended college without GI aid were led to attend 
by their military service. 

4 

A brief exploration was made of the contribution of veterans who would 
not have attended college without GI aid to the association between various 
item responses and Adjusted Average Grade* The results indicated that the 
presence of these students in the veteran subgroups was not a major deter- 
miner of the relationships . 

Veterans were asked whether they thought they would have attended col- 
lege if they had never entered the service. The proportions of students 
indicating various probabilities of college attendance were quite similar 
to those for the question on the GI Bill. There was a slight tendency for 
the veterans who had not originally planned to enter college to outdo the 
other veterans in Adjusted Average Grade. 



THE GI BILL 327 



Little evidence was found that a consistent difference in ability level 
existed "between those who probably would have and those who probably would 
not have attended college. (Caere was some indication that those whose col- 
lege plans had been quite definite were slightly higher in ability measures 
than the remaining veterans. 

Disabled veterans drawing benefits under Public Law 16 included perhaps 
five per cent of the median basic group. !Ehe differences between these stu- 
dents and the veteran group as a whole, considering the small number of stu- 
dents in this group, were too small to indicate any clear superiority or in- 
feriority for the disabled group. 

Veterans who were not drawing benefits at the time of the study amounted 
to less than five per cent of the veterans in the median basic subgroups in 
this study. These veterans showed some tendency to earn better Adjusted 
Average Grades than the other veterans, but their advantage was not statis- 
tically significant, 

!Ehe evidence presented in this chapter indicates that when students are 
selected according to the criteria used by the colleges in admitting veterans, 
there is no distinct difference either in ability or in grades adjusted for 
ability between those who would have come and those who would not have come 
without financial aid. This in turn supports the view that a substantial 
pool of effective academic talent could be tapped by lowering economic bar- 
riers to higher education. 



328 



Chapter X 
THE STUDENT VIEWS HIS COIIiEGE 



How did the postwar college look to its students? Did veterans ap- 
proach college with a "chip -on -the -shoulder" attitude toward an institution 
accustomed to dealing with younger students? Do attitudes toward the col- 
lege, and toward its program, faculty, and facilities help to explain why 
some students do "better work than would be expected on the basis of measures 
of their ability? Are differences in attitudes toward college relevant to 
the problem of accounting for veteran-nonveteran differences in Adjusted 
Average Grade? Answers to these questions were sought in a series of ques- 
tionnaire items, including one which allowed the student to suggest any 
changes which he felt might help him to get more out of his college experi- 
ence. 

Certain features of the procedure used in administering the question- 
naire have a special bearing on the interpretation of the results and should 
be reviewed at this point. Among the relatively favorable features are: 
first, the fact that the students were assured that the questionnaires would 
be studied only by an outside organization and that their individual answers 
would not be available to anyone at their college ; and second, the fact that 
virtually all students took the questionnaire near the end of their first or 
second year of college, which were opportune times for stock-taking by the 
student on the basis of a reasonable amount of college experience. The re- 
sults cannot necessarily be taken at full face value, however. Students may 
have "pulled their punches" a bit because of loyalty to their colleges, or, 
on the contrary, may have made numerous disparaging comments because of some 
transitory irritation Moreover, the sample studied is probably not truly 
representative of freshmen and sophomores in American colleges; the results 
presented in this chapter should, therefore, be taken only as a reasonable 
first approximation of the opinions of lower -division students in the spring 
of 191*7. 



Evaluation of the Educational Program 

Interest in Courses. Discussion of student attitudes toward the college may 
well begin with the rather specific question raised in Item 37s "Of the 
courses you are now taking, how many would you say you are really interested 
in?" Responses were divided into two groups? (A) interested in half or 
fewer of these courses, and (B) interested in most or all. This division 
split the students into two groups of roughly equal size Eesults for the 
twelve basic groups are shown in Figure 63. 

The majority of the freshman students were 
interested in most or all of their courses. Veterans, 
nonveterans, and women were in close agreement on this 



THE STUDENT VIEWS HIS COLLEGE 



329 



J_UU 









o 




75 












i */ 
-p 






















o 




1 









oo 


k 


<u 50 

3 


, . M . 


^ oooo 





L 

o 


| Median 
Veterans 
o Nonveterans 


::: 


1 00 






ooooo 








ooo 








o 

OR 












A< e^> 




o 





















o 












W MN MT MN 


Half or Most or 


fewer n~\i 



PER CENT CHOOSING DESIGKA.1ED CATEGORIES IN EACH COLLEGE 



xpu 
o 












> 
/^ 


\ 




sr 




/ L 


) 




S? 1^0 




-^ 






TJ 


4 


S ^^ 





> 


CO 




^^ 






3 




jf 






*~3 

tJ -ion 


c 


Y^ 












Veterans 










o Nonveterans 


110 










Half or Most or Total 


fever all group 



MEAH AAG OF MEDIAE COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 63 . PROPORTION OF COURSES KEAT.T.T OF OTTEEEST TO 
STUDENTS ITEM 37 



330 ADJUSTMENT TO COLLEGE 



point. In the various college groups, students 
who made the more favorable evaluation shoved a 
clear tendency to earn higher Adjusted Average Grades. 

It is clear that the majority of the students "were interested in most 
of their courses, even during the freshman year "when required courses pre- 
dominate. In none of the twelve basic groups did the per cent expressing 
the more favorable view fall below 40; the percentage of favorable responses 
was almost 90 for veterans in engineering at Midwest City. When veterans, 
nonveterans, and women students are compared with respect to attitudes toward 
their current studies, the differences are remarkably slight. The opinions of 
sophomores in this study appeared to be somewhat more favorable than, those of 
freshman groups from the same college. 

The relationship between attitude toward college studies and Adjusted 
Average Grade was quite marked; in all twelve of the groups, for both non- 
veterans and veterans^ less favorable attitudes were associated with lower 
AAG'a. Ifedian AAG's were as follows: "favorable" veterans, about 1^0; "un- 
favorable" veterans, about 125; "favorable" nonveterans, about 130; "unfavor- 
able" nonveterans, about 120. Jh six of the twelve veteran subgroups and 
three of the nonveteran subgroups, the mean AAG was significantly lower (at 
the 1$ level) for students giving bhe less favorable response. 

Although the relationship of this item to AAG is clear enough, it is 
by 210 means apparent what Interpretation should be put upon the finding. 
Since AAG- is rather closely related to actual grades, it appears likely 
that expressed lack of interest reflects dissatisfaction on the part of 
the student with a program in which he is not doing well. If this hypothesis 
is accepted, it might be Inferred that veterans 1 attitudes are slightly more 
responsive to such influences than is true for nonveterans. 

The similarity in the attitudes of veterans and nonveterans on this 
item suggests that it will not be helpful in accounting for differences be- 
tween veterans and nonveterans In grades relative to ability. 3&is hypothe- 
sis was confirmed by the negative results when the sign test was applied. 

Enjoyment of Studies. Approaching the problem of satisfaction in another 
Way, Item, lo 1 asked the students, w la general, are you enjoying your studies 
In college this term as much as you had expected to?" Besponses were divided 
into (A) enjoying studies less than, anticipated, and (B) enjoying studies as 
much or more than anticipated. In the analysis, attention was focussed on. 
the "disappointed" group. Results are given in Appendix Table 18. 

Students who found courses less enjoyable than 
they had expected made up almost one fourth of the 
typical group; little difference was found between 
veterans, nonveterans, and women in this respect. The 
disappointed students were consistently below the 
general average of their subgroup in Adjusted Average 
Grade. 



THE STUDENT VSIWS HSS COLLEGE 33! 



Students who were not finding their studies as enjoyable as they had 
expected constituted a substantial minority of the student groups , amounting 
to almost one fourth for both veterans and nonveterans* Little difference 
in attitude was found among male veterans, male nonveterans, and women stu- 
dents. Although there was rather little variation from college to college 
with respect to this item, it may be noted that the Midwest City engineers 
showed the lowest per cent of "disappointed" students, among the twelve "basic 
groups. Sophomores tended to have more favorable opinions than the corres- 
ponding freshman groups in the same college. 

A marked association between unfavorable views and low AAG is apparent. 
In all but one of the basic subgroups, the students who expressed disappoint- 
ment were lower in mean AAG than students in the other item category. The 
median values of the mean AAG r s are about 120 for the "disappointed" veterans 
and 1^4-0 for those who were better satisfied; the corresponding figures for 
nonveterans are about 110 and 130 The analysis of each of the twelve college 
groups separately yielded eight differences significant at the 1$ level for 
the veterans and five such differences among the nonveterans, when students 
who gave the less favorable response were compared with students in the other 
category. The finding that underachievement in college work and disappoint- 
ment with college studies at the end of the year go hand in hand is reason- 
able; the data available do not permit any clear inference regarding cause 
and effect relationships. 

This item does not help in accounting for veteran-nonveteran differ- 
ences; this outcome is reasonable in view of the slight differences between 
veterans and nonveterans in the per cent choosing each response category. 

Value of College Studies . In Item 38, the student was asked, "Do you ever 
feel that the things you are studying in college are not really worth the 
time spent on them?" Three categories were used in analyzing responses to 
this item: (A) yes, frequently; (B) sometimes; and (C) seldom or never. 
In contrast with the itams previously discussed, which emphasized interest 
and enjoyment, this item asked the student to make a more practical evalua- 
tion of his college program* Results are shown in Jigure 6. 

Almost one half of the students reported that they 
"sometimes" doubted the value of their current studies, 
about one third seldom had doubts on this subject, and 
only about one fifth had frequent doubts. No appreciable 
differences were found in respons.es to this item between 
veterans, nonveterans , and women* High frequency of 
doubts was significantly associated with low Adjusted 
Average Grades for veterans; the relationship for non- 
veterans was less marked although in the same direction. 

Almost half of the students in the BBdian college group reported that 
they "sometimes" doubted the value of their courses . Koughly one third of 
the students reported that they seldom had misgivings about the value of 
their work, and about a fifth reported that they had such feelings frequently. 



332 



ADJUSTMENT TO COLLEGE 



100 

co 75 

-p 




























-P 
-p 

! 

& 25 








oo 
o 








o 





oooooo 
ooo 


>'_ 




o 
o 
ooo 
ooo 


r'"* 


ooooooo 











oooo 







oo 















o 










W MN MV MN MV MN 
Frequently Sometimes Seldom or never 
feel studies feel studies feel studies 
not worth-while not worth-while not worth-while 


EER CENT CHOOSING DESIGNATE.:) CATEGORIES HI EACH COLHEG3 



Median 
Veterans 
o Nonveterans 



150 



s 
s 



130 



120 



110 







S 
s* 


> 






_ji 


X 

X 


< 




< 

c 


^ ^^^^ 


--""" 


< 










Veterans 
o Nonveterans 




Frequently Sometimes Seldom or never Total 
feel studies feel studies feel studies group 
not worth-while not worth-while not worth-while 



MEM AA.G OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGOEIES 



FIGURE 6k. EVALUATION OF COLLEGE STUDIES; ITEM 38 



THE STUDENT VIEWS HIS COLLEGE 333 



No clear difference between veterans and nonveterans appeared in the relative 
popularity of the three alternatives and, no clear sex difference vas found. 
On this item, the engineers from Midwest City again had the most favorable 
opinions, among the twelve basic groups, however, students from Turner Uni- 
versity excelled them in reported satisfaction, with more than half of the 
students in all three subgroups reporting that they seldom or never doubted 
the worth of what they were studying. 

The relationship of this item to Adjusted Average Grade is quite distinct 
for the veteran groups. The most favorable response is associated with higher 
AAG in all twelve basic groups and the least favorable response is associated 
with lower AAG in all but one. For nonveterans the trend, although in the 
same direction, is not sufficiently consistent to be statistically significant. 
The medians of the mean AAG's agree with this observation; for veterans, the 
median of the mean AAG's for those who seldom doubt the value of their studies 
is above ikO; for those who often feel doubts, about 125. The corresponding 
figures for nonveterans are 150 and 120. Examination of the significance tests 
in the separate college groups confirms the relationship The veterans who re- 
port that they seldom doubt the value of their studies show a superiority in 
AAG which is significant at the 1% level In six of the twelve basic groups; 
the corresponding results for nonveterans show two differences significant 
at the 1% level. These figures are rather similar to those for the expres- 
sion of interest in the various courses. The hypothesis that veterans are 
slightly more prone to question the value of their program than are non- 
veterans, when they are not doing veil, is suggested by these findings. How- 
ever, the sign test revealed that this item did not make any significant 
contribution in accounting for differences between veterans and nonveterans 
in grades adjusted for ability differences. 

Satisfaction with Education, A further question, Item $6, asked for an over- 
all evaluation: "On the whole, how well satisfied are you with the kind of 
education you are getting?" Responses were divided, for analysis, into: (A) 
very well satisfied, (B) fairly well satisfied, and (C) somewhat or very much 
dissatisfied. Results are given in Figure 65. 

About half of the students reported that they were 
"fairly well satisfied" with their education; the remain- 
ing students were divided about equally between "very 
well satisfied" and "somewhat or very much dissatisfied, 11 
Male veterans, male nonveterans, and women students 
showed little difference in attitudes on this point. 
A clear association was found between greater satis- 
faction and higher Adjusted Average Grades. 

Responses to this item showed a pattern similar to that for the preceding 
item (38), which deals with the worth-whilenese of college study. Somewhat 
more than half of the students, in the median group, gave the answer, "fairly 
well satisfied," and somewhat less than one fourth of the students fell into 
each of the other two categories. As in Item 38, Turner had the highest propor- 
tion who reported that they were very well satisfied; the engineers from Mid- 



334 



ADJUSTMENT TO COLLEGE 



100 



to 
p 


<D 

1 

CQ 



5 

JH 
(S 



75 



50 



25 





















o 













00 










@ 


oo 










1 *M 


> oooo 













oo 








ooo 


9 


o 


@ 







o 









00 




o 






e 


o 


e* 


o 






aee 


ooo 


e 


oo 






00 


oo - 


**6O 


ooo 









ooo 











9 





Yery well 
satisfied 



w m 

Pairly well 
satisfied 



MV MN 
Somewhat 
or very much 
dissatisfied 



Mectian 
Veterans 
o Nonveterans 



PER CEEFT CHOOSING DESIGHA.TE3D CATEGORIES IN EACH COLLEGE 



150 



SP 
S 130 



no 



4 


< 








( 


\ 

\ 

\X 




1 


i 




*>s 

N 


\ 

O x x 

X \, V N 


^ 


> 






\ 

N 


Veterans 
o Nonveteri 

f 


ans 


Very "well Pairly well Somewhat Total 
satisfied satisfied or very much group 
dissatisfied 



AAG OF MEDIAN COLLEGE GROUP FOE SPECIFIED CATEGORIES 



FIGURE 65. SATISFACTION WITH EDUCATION: ITEM 36 



THE STUDENT VIEWS HSS COLLEGE 335 



west City were nert highest. Ihe sophomore groups did not express greater 
satisfaction than the freshman groups from the same college on this item. 
Ho consistent difference between veterans and nonveterans appeared with 
respect to the attitudes expressed toward their education, nor was a con- 
sistent difference between the attitudes of men and women found. 

As in the other questions in this group, a marked relationship was 
found between responses to this item and AAG* In the twelve basic groups ^ 
the students who reported dissatisfaction were underachievers in all 2k 
comparisons . Kedians of the mean AAG's were about 1^0 for the "very well 
satisfied" veterans and about 120 for the dissatisfied veterans; for non- 
veterans the corresponding figures were about 135 an<i less than 115- When 
the relationship is studied college by college, the dissatisfied veterans 
are significantly low (at the 1$ level) in AAG in six of the basic groups; 
the dissatisfied nonveterans are significantly low in four of the basic groups. 

JJo evidence was found that the characteristic measured by this item 
would help to account for veteran-nonveteran differences in AAG. 

SiiTnmary. In all four of the questions which asked students to evaluate their 
college, certain common trends appeared* Taken as a vhole, the results indi- 
cated that the typical student is fairly well satisfied with his college. 
A substantial minority, representing perhaps one fifth of the students, ap- 
pear to be somewhat dissatisfied and roughly the same proportion are. rather 
enthusiastic about their college program* Male veterans, male nonveterans, 
and women are,, in general, rather similar in their evaluations of their col- 
lege studies. In all four items, a rather clear association between dis- 
satisfaction and poor performance relative to ability is apparent. None of 
the items was useful in accounting for differences between veterans and non- 
veterans in Adjusted Average Grade* 



Attitude toward Present Division 



Still another estimate of a student's satisfaction with his college 
program was provided by Item 16: "10 the school or division (e g, arts, 
engineering) in which you are now studying your first choice, or would you 
prefer to major in some other school or division in the same institution?" 
!Ehe two categories used in analyzing the data for this item were (A) now 
in field of first choice, and (B) would prefer some other school or divi- 
sion. Results of this analysis are shown in Jigure 66. 

typically, almost nine students in ten prefer 
their own division to any other in the same institu- 
tion. Students who reported that they would prefer 
a different division showed a clear tendency to make 
inferior Adjusted Average Grades. 



336 



ADJUSTMENT TO COLLEGE 



100 



75 



50 



f-s 

(2 



25 

















oooo 











ooo 











o 
oo 











o 














| Median 
Veterans 
o Nonveterans 








o 








00 








o 











ooo 











oooo 
o 




MV MN W MN 
Now in field Prefer some 
of first choice other field 



HER CENT CHOOSING DESIGKA.TEID CATEGORIES IN EACH COLLEGE 



150 



s 

<D 



130 



TJ 
(D 

-P 
CO 



120 



110 











< 


- > - c - - 


< 




< 


*x \ 


< 








\ 


Veter* 
o Nbnye" 


ins 
serans 


Now in field Prefer some Total 
of first choice other field group 



MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 66, SATISFACTION WITE PRESENT SCHOOL OR 
DIVISION: ITEM 16 



THE STUDENT VIEWS HIS COLLEGE 337 



It is evident that substantially all students, both veterans and non- 
veterans, prefer their present division to any other. The median percentage 
who are in the division of their first choice is about 90 for veterans and 
slightly less than 90 for nonveterans, in the twelve basic groups . Ihere 
was no consistent tendency for women to be either more favorable or less 
favorable than men in the same college; the differences "Which do appear in 
particular colleges may, of course, have some local significance. 

It is reasonable to expect that dissatisfaction with present divis ion 
and low AAG's would go together ; such is indeed the case. As sJiown in Figure 
66 y the median of the mean AAG's for veterans Is slightly below 135 for the 
satisfied group and slightly above 120 for the dissatisfied group. The cor- 
responding figures for nonveterans are slightly less than 130 for the satis- 
fied students, and about 115 for the dissatisfied ones. Examination of the 
results college by college shows that students in the dissatisfied subgroup 
have a lower mean AAG than students in the satisfied group in 21 out of 2^ 
instances, and two of the remaining comparisons were ties. Such consistency 
would be expected to arise by chance less than once in 100 times. Finally, 
individual subgroups of dissatisfied students in the twelve basic groups 
showed mean AAG's which were reliably different from those of the satisfied 
group, as follows: veterans, two at the 1$ level and two at the 5$ level; 
nonveterans, four at the 1$ level* Ho pattern is apparent in these signifi- 
cant groups except that three out of six of the engineering subgroups show 
significant differences, as compared with six out of 18 for the liberal arts 
subgroups . 

Attitudes toward present division are thus clearly associated with AAG. 
!Ehe similarity of viewpoints between veterans and nonveterans, however, makes 
this item useless in accounting for veteran-nonveteran differences in college 
achievement relative "to ability. 



Attitude toward Faculty 



Only one question was devoted to finding out what students think of 
their college teachers. Item 17 asked: "Sow would you rate, as teachers, 
the faculty members who have taught you this past term?" Choices were 
divided into (A) most, or all, are good teachers, and (B) some, or none, 
are good teachers. On the assumption that relationships with teachers would 
have an important bearing on adjustment to college, it was thought that this 
item might aid in understanding academic success in. college. Results axe 
given in Figure 6? 

A majority of students considered that most of their 
instructors were good teachers, veterans being slightly 
less favorable than nonveterans in their evaluations. 
!Ehere is a consistent relationship between attitudes 
toward teachers and Adjusted Average Grade in the veteran 
groups; in the nonveteran groups, the relationship is 
slightly less consistent. 



338 



ADJUSTMENT TO COLLEGE 



100 



75 



CQ 



1 

o 



50 























000 













a 




*M 1 oo 

f OOOO 


, 

I m I oo 


* 1 


P I ooo 


* 







o 









ooo 





o 


* 


o 











Median 
Veterans 
o Non-veterans 



MV MET W MW 
Most (or all) Some (or none) 
good teachers good teachers 

PER CENT CHOOSERS- DESIGNATED CATEGORIES IN EACH COLLEGE 



150 



"8 

-P 



130 



120 



110 



Veterans 
o Nonveterans 



Most (or all) Some (or none) Total 
good teachers good teachers group 

AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



67. ATTITUDE TOWARD FACXJLTT: ITEM 17 



THE STUDENT VIEWS HIS COLLEGE 339 



A majority of the students, "both veteran and nonveteran, thought that 
most of their teachers were good teachers. Konveterans were slightly more 
favorable than veterans in the judgments expressed; the proportion giving 
favorable responses was higher for nonveterans in nine of the twelve basic 
groups,, -with one tie. No consistent difference between the sophomore groups 
and the. freshman groups appeared on this item. Women had,, on the average, 
much the same attitude as the nonveteran men in their colleges. There was 
some indication that engineers were more critical of their teachers than 
were students in liberal arts; the engineers at Midwest City provided the 
only exception to this generalization. Among all the groups studied, the 
students at Turner expressed the highest degree of approval of the faculty, 
with almost 80 per cent in the favorable category. Within the basic twelve 
groups, students at Harris, Adams, Douglas, and Littletowa State were highest 
in their approval of the quality of teaching. 

Attitudes toward teachers and AAG are related, as shown in the lower 
portion of Figure 67. The difference between the more favorable group and 
the less favorable group amounts to somewhat more than 5 points for both 
veterans and nonveterans when median AAG f s are compared. The trend is more 
consistent for veterans, however. Veterans who reported an unfavorable atti- 
tude toward teachers had lower mean AAG's than the other veterans in all 
twelve groups; nonveterans who gave this response were lower in nine of the 
twelve groups, with one tie. When the relationship between attitude toward 
teachers and AAG is studied in each college group separately, only the engi- 
neering veterans at Midwest Tech and the nonveterans at Adams show a differ- 
ence in AAG which is significant at the 1$ level . In general, attitudes 
toward teachers, as expressed in this item, are apparently less closely re- 
lated to AAG than are attitudes toward college studies. This question did 
not turn out to be related to veteran -nonveteran differences in AAG 



Evaluation of Study Facilities 



One rather specific aspect of the college environment which might be re- 
lated to a student's academic adjustment is availability of a place where he 
can study effectively. To obtain the student's own evaluation of his study 
place, Item 27 asked, "In general, do you have a satisfactory place to study, 
one that is free from noise and distraction and reasonably comfortable?" 
Answers were analyzed under three heads: (A) yes, entirely satisfactory; 
(B) fairly satisfactory; and (C) no, quite unsatisfactory. Eesults are 
shown in Figure 68. 

About one student in ten reported that his study 
facilities were "quite unsatisfactory." Differences 
on this point between veterans and nonveterans were 
quite small; women, however, tended to be more critical 
of their study facilities than the men Lower Adjusted 
Average Grades were typically found for students who 
reported poor study environment. 



340 



ADJUSTMENT TO COLLEGE 



100 



P 
I 











| Median 
Veterans 










o Uonveterans 
















9 





ftftft 


ooooo 






ft 


oooo 


ftftftft 


00 






'ftftft* 





ftftftft 


ooo 






Oft 


ooo 




oo 






ft* 


00 











ft 









ft 


- 












o 










0999 


oo 










m ftftftftft I ooooooo 

F ftft [00 



Unsatisfactory 



W MN W MJT 

Entirely. Fairly 
satisfactory satisfactory 

PER CENT CHOOSING DESIGNATED CATEGORIES IN EACH COLLEGE 



Adjusted Average Grade 

H H H H H 

B 8 3 ^ 















^^"^^-^H 


^^s. 


< 




( 


>-- . . ( ^ 


-^"\, 


< 

t 








1 


) 

ft Vetera 
o Nonvet 


ns 
erans 


Entirely Fairly Unsatisfactory Total 
satisfactory satisfactory group 



MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 68. ADEQUACY OF HACE FOR STUDY: ITEM 21 



THi STUDENT VIEWS HIS COLLEGE 34] 



Only a little more than 10 per cent of the students considered their 
study facilities "quite unsatisfactory," and about 40 per cent considered 
their study arrangements entirely satisfactory, in the median group. 
Veterans were a bit more likely to view their study arrangements as entirely 
satisfactory than were nonveterans j the difference was quite small, however, 
and may merely indicate that the veterans were more tolerant of discomfort. 
With one exception (Eastern City), women were less likely to consider their 
study place entirely satisfactory than either male veterans or male non- 
veterans; here again, the differences were small. 

Satisfaction with study place was associated to some eztent with AAG, 
but the relationship is apparently not very close. For veterans, the very 
well satisfied differed from the quite dissatisfied by only about 10 points, 
judging from the medians of the mean AAG's; for nonveterans the difference 
was aomewhat smaller. Considering all of the 2k subgroups, only five compari- 
sons show a mean AAG for the dissatisfied students which is higher than that 
of the corresponding satisfied groups the relationship is significant at the 
1$ level, when results for veterans and nonveterans are thus pooled. Within 
the various college groups, the dissatisfied group was significantly below 
average in AAG- (at the 1$ level) in three of the veteran groups, but in none 
of the nonveteran groups. Two of the three groups showing significant differ- 
ences were composed of engineering students. J\mong the students who were 
entirely satisfied with their study arrangements, only the nonveterans at 
Adams were significantly high in AAG at the 1$ levelo These results indi- 
cate that the student's opinion of his study place is related in the expected 
direction with AAG, but that the relationship is not close. -Hie relationship 
of this item, to veteran-nonveteran differences in Adjusted Average Grade was 
found to be negligible when the sign test was applied. 



Suggestions for Improving the College 

In contrast to the more structured questions about the college, Item 45 
simply asked the student for a free answer to the question: "Briefly, vhat 
are the main changes you would like to see made in the program or organiza- 
tion of education at this college, in order to help you get wnat you are 
after in a college education?" Students were also invited to make comments 
about Item 36 ("On the -whole, how well satisfied are you with the kind of 
education you are getting?") and about Item 38 ("Do you ever feel that the 
things you are studying in college are not really worth the time spent on 
them?"). Thus, the student was given several opportunities to unburden him- 
self regarding any aspect of the college program which did not fully meet 
his approval. 

In order to reduce the comments of the students to a form suitable 
for mechanical tabulation, a code number was assigned to each comment in 
accordance with a pre-arranged scheme. A full description of the steps 
followed and of the precautions taken in the coding process is given in 
Chapter IIj here it is only necessary to recall one feature T&ich has a 
specific bearing on the interpretation of results* Since coding was com- 



342 ADJUSTMENT TO COLLEGE 



pleted for any one college before work on the next college was begun, com- 
parisons of percentages from one college to another must be made with caution, 
Comparisons of responses between the first three colleges coded (Mams, 
Stewart, and Midwest Tech) and the remaining 13 colleges should be avoided, 
since a simplified code was introduced after the coding of the first three 
colleges was complete. Within any one college, however, comparisons of 
veterans and nonveterana may be made with reasonable confidence j question- 
naires of these subgroups were not segregated before coding. 

Humber of Comments* In analyzing the coded responses, attention was given 
to the number of suggestions made by each student, since it was thought that 
any marked differences between veterans and nonveterans in tendency to com- 
plain would be found by this kind of analysis^ Results on which this dis- 
cussion is based are presented In Appendix Table lj-5-1* 

Veterans, nonveterans, and women were about 
equally likely to refrain from making suggestions for 
improving their college. Ho general tendency was found 
for the number of suggestions made to be related to Ad- 
justed Average Grade- 

Students who gave no relevant response may be considered first. Ho 
clear-cut tendency was found for male veterans, male nonveterans, or women 
to differ in the proportion offering no suggestions* On the average, about 
one student in seven made no response which was codable under the procedure 
used. 

Students who gave no comments in general did not differ in mean AAG 
from the remainder of the group to which they belonged. In the veteran, sub- 
groups, the engineers at Middle State who made no comments were significantly 
high in AAG at the 1$ level! the engineers at Midwest Tech and freshmen at 
Adams who made no comments were significantly high at the 5$ level. None 
of the nonveteran subgroups showed a significant relationship,, 

Consideration of the results for students who gave three or more sugges- 
tions, and comparisons of the average number of suggestions made by students 
in various college groups with their expressions of dissatisfaction on pre- 
coded items indicated that the total number of suggestions made does not pro- 
vide an index of satisfaction with the college and its program* The sign tes- 
based on all four response categories indicated that the number of coded 
responses does not aid in interpreting veteran-nonveteran differences. 

Kinds of Suggestions Most Frequently ;__Magte. Htten coded responses were examine* 
in detail, it was found that four kinds of suggestions were distinctly more 
popular than any others. Indeed, the two most popular codes in every one of 
the 56. subgroups having questionnaire data were found to be among these four 
categories. Accordingly, attention In the discussion will be focussed on 
these four major types of suggestions. The four most popular categories 
were as follows s (A) need for better courses, Instructors, or instruction; 
(B) need for more courses, teachers, or classrooms; (C) need for fewer or 



THE STUDENT VIEWS HIS COLLEGE 343 



more appropriate required courses ? or for more freedom in elect ives; and 
(D) need for changes in general academic requirements, including comments 
on grading systems and examinations. Ee suits for the three colleges coded 
before the consolidation of categories (see Chapter II, pp. 78-79, and 
Appendix C3) are shown in the upper portion of Appendix Table 45-2 (Part l), 
and results for the thirteen colleges coded after the revision, in Appendix 
Table ^5-2 (Part 2). Figure 69 shows the findings graphically for the nine 
basic groups coded according, to the revised plan,, 

Among the diverse suggestions offered by the stu- 
dents for improving their colleges, certain recurrent 
themes were present "more and better teacher s/ r "fewer 
required courses," "less stress on grades and examina- 
tions." More than one student in three wanted better 
courses, instructors or instruction? about one in seven 
wanted more courses, teachers, or classrooms; about one 
in four suggested that fewer (or different) required 
courses were desirable; and about one in six wanted vari- 
ous changes in general requirements, particularly changes 
in the system of examinations and grading. In general, 
there was little evidence of any trend for veterans to 
offer more or fewer suggestions than nonveterans. Women 
were slightly less likely to criticize quality of in- 
structors and instruction than were the men. In three 
of the colleges women were considerably more likely than 
men to ask for more courses, instruct or s, or classrooms. 
Veterans who suggested fewer or different required courses 
tended to be low in Adjusted Average Grade; otherwise, no 
clear-cut tendency was found for an association between 
kind of suggestion made and AAG. Among the comments of- 
fered less frequently, nonveterans were slightly but con- 
sistently more likely to suggest better guidance and 
placement services. 

Cateffory^A; Need for Better Courses, Instructors , or Instruction. The mean- 
ing of Category A may be clarified by considering some of the comments to which 
this code was assigned. The great majority of these comments were concerned 
with the college teachers rather than with method of instruction or course 
content. Many comments merely requested "better" or "more competent" instruc- 
tors; others, however, were considerably more specific. Among the weaknesses 
noted "by students were: lack of interest in students "Very few professors 
seem to care if we pass or fail"; lack of interest in subject "Instructors 
should show some interest in their studies"; lack of age and experience 
"Some. . .know little more than, the student"; excessive age "The old instructors 
who are coasting on past laurels should be shaken up"; inability to get sub- 
ject across "Employ instructors not merely for their knowledge, but also for 
their ability to teach"; lack of contact with student ideas "The professors 
with doctors degrees are rather hard to comprehend"; excessive narrowness- 
"I'm getting the turkey but not the trimmings . .I'd like to meet a few in- 
structors who have a little more than just the course to give"; and excessive 
talkativeness "They gas in an unending stream usually about every irrelevant 
topic save what is pertinent to the course." 



344 



ADJUSTMENT TO COLLEGE 



100 



75 



I 



,p 

rt 

CD 
O 



25 





































O 















* 


OOO 









o 






-9 


oo 






00 






> 


oo 


* 


00 


1 o 









o 


9 




1 o 





oo 






9 




4 


o 


t@ 


> ooo 








00 





o 


- 


00 






f 0* 

? 


ooo 




o 












oo 









oo 










1 


o 





I Median 
Veterans 
o Nonveterans 



W MN 

Better 

teachers 

or courses 



W 
More 
teachers or 
classrooms 



Fewer 
required 
courses 



W MN 

Changes in 

examinat i ons f 

grading^ etc. 



PER CENT SUGGESTING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 















^ 


H-. - - ~"" 4 


' XS " X * S '^ 


^ 

^^ ^ 


{ 






^^\ 

"^ 


^^1 


^-^^ 


( 












Veteran* 
o Nonveten 


i 
*ans 


Better More courses , 3?ever Changes in Total 
teachers teachers or required examinations, group 
or courses classrooms courses grading, etc. 



130 



120 



110 



MEAN AAG OF MEIDIAN COLIEGE OROUP FOR SPECIFIED CATEXK)RIES 

FIGURE 69. MOST FREQUENT SUGGESTIONS FOR IMPROVING THE COLLEGE: ITEM 
(PAET 2)* 



Note that only nine of the twelve "basic groups were used in constructing 
this figure. 



THE STUDENT VIEWS HIS COLLEGE 345 



Comments about method and content included suggestions: that more 
opportunity for student discussion be provided "Too many lecture courses"; 
that less emphasis be placed on details and memorization "Too much emphasis 
on details... particularly english (sic) courses"; that teachers stick more 
closely to the textbook "I -would like to have professors and other instruc- 
tors follow very close to the textbook, regardless of how they feel toward 
the situation"; and that the maturity of students be recognized "Some of 
the courses are conducted on a high school level." 

Some idea of the range of comments coded under Category A may be ob- 
tained from the following examples: 

"The courses given in Economics are not adequate to insure a 
good position on graduation." 

"There a few teachers here which I believe would be benef icient 
if they were discharged . Some don't care about my welfare. I asked 
some questions in class and they refused to answer them/' 

"Higher salaries for teachers so that I know I am getting the 
best possible supervision in my studies." 

"To have the instructors stop trying to influence the students. 
For the instructors to recognize the students as college students 
rather than grade school students," 

"I'd fire four out of five teachers." 

"Actually they [jEnglish teachers] try to tell you how you felt 
under combat conditions/ 1 

"Choose instructors that are enthusiastic, well -trained, and 
interested in putting across their subject effectively and clearly." 

"Totally revise teaching methods so as to embody the methods of 
instruction used by the Army, such as: 1) conferences replacing 
lectures ; 2) use of instructional aids...; 3) eliminate note taking- 
use mimeographed outlines and summaries; i) give teachers instruction 
in how t teach a group of joen; 5) provide a r scope 1 or objective for 
each course." 

The popularity of comments about instructors and instruction is cleaxly 
evident in Figure 69* Somewhat more than one- third of the men in the typical 
basic group had some such comment to make. Ho noticeable difference appears 
between veterans and nonveterans in this respect; but women generally gave 
fewer comments in this area than the men. Students who made comments coded 
in Category A did not differ markedly or consistently in AAG from other stu- 
dents in their college groups; on the whole, however, they are slightly 
superior in AAG. Since comments about instruction were about equally preva- 
lent for veterans and nonveterans, and since the students who made such 
comments were pretty mucja a cross -sect ion of their groups in AAG, the fre- 
quency of such comments does not aid in accounting for veteran-nonveteran 
differences in AAG. 



346 ADJUSTMENT TO COLLEGE 



Category B; leed for More Courses^ Teachers, or Classrooms. Although sug- 
gestions that classes be smaller and college facilities less crowded un- 
doubtedly constituted the largest single group of comments coded in Cate- 
gory B, a considerable number presented requests that the college enter new 
fields or develop further offerings In existing fields She latter group of 
suggestions necessarily varied in" content from college to college, for example, 
requests for a course in business administration would not arise in a uni- 
versity which had such a program in operation- The following comments may 
illustrate the scope of the comments included under this codes 

"Crowded conditions should be abolished*" 

"Smaller classes ." 

"More courses in art and music." 

"More engineering courses. Many who wish to get into that college are 
stopped by crowded conditions." 

"The absence of courses relating to business and business administra- 
tion might force me to attend graduate school Q " 

"Public speaking class." 

"I think there should be a general two,, three, or four year course 
covering subjects of general interest, say, to all womenwomen who intend 
and desire to marry soon after college j courses to improve a girl's mind- 
make her more fit to be a good wife and mother; and, if possible, also to 
prepare her in some way to be able to support heraelf if the need arises," 

"Some specialized training for professions such as radio announcing, 
directing, etc*, selling, business management more practical courses." 

As in Category A, veterans and nonveterans showed rather similar perform- 
ance as far as responses coded in this area are concerned. As shown in the 
upper part of Figure 69, the median proportion is nearly the same (about 15 per 
cent) for both subgroups. In three of the nine groups in which women students 
were included, the proportion of women who asked the. college to offer more was 
distinctly higher than was true in either of the male groups. Further compari- 
sons of men and women students on their attitude toward the adequacy of col- 
lege programs should be made when college facilities are less heavily burdened - 

Students who commented about overcrowding and inadequate offerings did 
not differ greatly in their mean AA f s from the total groups to which they be- 
longed, although there was some suggestion that the veterans who made comments 
coded in Category B were slightly superior, and nonveterans in the same cate- 
gory were inferior, in AAGr* There is no reason to believe that the frequency 
of responses about crowded conditions aids in accounting for veteran-nonveteran 
differences in AAG, 

Category C; .Heed for Fewer or Different Bequired Courses* The numerous 
comments expressing a desire for fewer or different required courses (Cate- 
gory C) indicate that this topic is very much a live issue as far as students 
are concerned. Although many of the comments were merely statements like, 
Required courses not worth-while," or, "Hot require a foreign language," 



THE STUDENT VIEWS HIS COLLEGE 347 



other comments offered certain clues as to the reason for the objections 
A fair number of comments disparaged the objectionable requirements \ they 
were compared in Talue with "the wart on a pickle" and characterized as 
"silly/ 1 "stupid/' "waste of time/' and "irrevelant" (sic) by various stu- 
dents. Other students saw them as obstacles in the way of spending all their 
time on the subject in which they wished to specialize . As one student put 
it, "I think a person should take just the courses that pertains (sic) to the 
profession which he or she wishes to go in. For instance, why take a foreign 
language if you plan to enter Air Transportation just in the IT. S., also 
Botany o" Another young man, who was planning to be a research physicist, 
said, "Some of the courses are just cultural studies not needed in the phase 
of work I intend to do " A few students appeared to object to the principle 
involved in required courses : "I feel that it is not fair to force a person 
to spend his money for something he doesn't enjoy and has no interest in." 

Another said, " University has a supreme contempt for the student's 

ability to know what he wants to learn." One student wrote, "Courses should 
be made to suit the student rather than the university." Others felt that 
the required courses were repeating material that they already knew. In 
other instances, parental attitudes complicated the picture: "I am in 
liberal arts and want to be in business but parents won't permit the change." 
Another wrote, "Wfjr father and I have decided what courses I need." Finally, 
a sizeable group mentioned that required courses were rather difficult. 
Perhaps one example will suffice: "I personally think that people should be 
able to take the courses that are necessary to pursue one's profession instead 
of being required to take harder courses by the university and then making 
the course so darn rough that fellows can't stand the grind and they quit. 
I'll never quit..... 11 

As shown in Figure 69, comments regarding required courses varied 
greatly among the colleges, ranging from less than 5 P*" cent for Midwest 
City engineers to almost one half in s.everal schools. No marked differences 
in attitudes on this matter were found between veterans and nonveterans, the 
median in the nine selected groups being about 25 per cent for both subgroups. 
Although veterans might have been expected to be more critical of required 
courses than nonveterans, the results for this category show a slightly 
higher median for the nonveterans. 

A rather striking relationship of this category to AAG appears in the 
case of veterans | in only one of the nine selected college groups did the 
veterans who commented on required courses earn AAG's as higih as those who 
did not. For nonveterans, on the other hand, no such relationship appears 
These results suggest that veterans' attitudes toward requirements depend to 
some extent upon how well they are doing . It should be added, however, that 
in none of the basic groups considered individually was there a statistically 
significant relationship between frequency of comment on required courses and 
AAG. The results for this category indicate that it is not useful in account- 
ing for veteran-nonveteran differences in AAG. 

Category D; Heed for Changes in General Academic Bequirements . Changes in 
general academic requirements, which were coded in Category D, necessarily 
involved a rather miscellaneous set of comments a Although the specific com- 



348 ADJUSTMENT TO COLLEGE 



ments varied from college to college, the bulk of the comments were con- 
cerned with the system of grading and examinations A number of comments 
about the curriculum as a whole were also included here; to some extent 
these comments tended to overlap in general meaning with criticisms of re- 
quired courses included in Category C. 

Considerable feeling was expressed that too much stress was being 
placed on grades As one student put it, "Emphasis taken away from com- 
mercial-like race for good marks , and place more emphasis on general absorp- 
tion." One young women wrote, "I realize grades are unimportant but too much 
emphasis is placed upon them because they cause one to take 'cinch 7 courses 
instead of useful courses." Students also felt that grading standards varied 
from course to course. For example, "I would like course requirements to be 
more uniformo As it is, some courses require only attendance, while others 
(by far the majority) seem to assume that the student takes no other courses 
whatsoever." Another student said, "Doing the same work for two different 
instructors sometimes means the difference between an A and a Co" One stu- 
dent offered the suggestion that "One Prof, should teach everyone so that 
they are all on the same level/ 1 Numerous other comments on the grading 
system have a familiar sound; "There should be two grades; Passing and 
failing"; "System of grading on a curve is not helpful"; "It is my strong 
belief that in a large class the professor has a quota of students that 
will Q failj "; "Elimination of grading system"; "Grades are unnecessary in 
the case of any student with common sense enough to evaluate his own ability"; 
"Lay less stress on scholastic work thereby enabling more time on extracurri- 
cular activities"; and "Grading by the improvements made by each individual 
during a semester's time, not the class as a whole." 

Adverse coment was also directed toward examinations. In addition to 
the comment that they should be fairer, students seemed anxious to have them: 
less difficult "They are not a test of our ability but some professor's 
crossword puzzle"; less often "It seems I'm always studying for a test"; 
more often "One exam in one subject for an entire grading period is not 
fair"; administered under careful control "Cut out the cheating by jother 
pupils. Although the instructors see this cheating, they never apprehend 
the person"; and frequently revised "Change the tests from time to time be- 
cause ... certain groups have files of these tests." Some students expressed 
a desire for proficiency examinations; for example, "I would establish tests 
to determine a student's foreknowledge so he wouldn't have to take courses in 
college covering material he already knows." It is not surprising that some 
students favored the abolition of examinations for various reasons; one advo- 
cated "Discontinuance of end term examinations as they take two weeks of in- 
struction from the term. " 

A few examples will illustrate the kind of general comments about the 
academic program which were coded in Category D rather than Category C. 

"More time for cultural courses or courses of a general nature." 

"I would prefer to take an even broader course than the B*A. 
offers . " 



THE STUDENT VIEWS HSS COLLEGE 349 



"...take less courses so I could learn more thoroughly the courses 
I do take." 

"Education is too general." 

Comments falling within Category D were more popular with nonveterans 
than with veterans in a majority of the nine selected groups. The middle 
value of the percentages was about 20 for nonveterans and 15 for veterans. 
Ho consistent difference in frequency of response was found between women 
and the two male groups in the nine colleges where both men and women were 
studied*, 

The AAG's for students making comments along the line of Category D did 
not differ markedly from those of the total group to which they belonged, 
and in none of the nine selected groups was there a statistically significant 
difference in mean AAG between students making such comments and those not 
doing so. There is accordingly no reason to believe that this code is useful 
in accounting for differences between veterans and nonveterans in AAG. 

Additional Comments . The four categories next highest in popularity were as 
follows V (I) need for better integration of existing courses j (3T ) need for 
reduction of difficulty of certain courses; (G) need for closer student- 
faculty relationship; and (E) need for better guidance, counselling, and place- 
ment. Detailed results for these responses are given in the lower portion of 
Appendix Table V?-2 (Part l) and in -Appendix Table 1*5-2 (Part 3)- 

Students whose suggestions fell into Category E (better integration of 
courses) made such comments as: "Too much redundancy not enough integra- 
tion o.o I have not learned anything that I hadn't already been exposed to be- 
fore coming to college"; "Separate courses, such as social science and biology, 
could be more coordinated to provide a better insight into the interrelation- 
ship of areas of subject matter." 

Comments relating to course difficulty, Category F, often referred to 
time necessary to fulfill assignments or to excessive amount of material 
covered a Thuss "Some courses require too much work for too little benefit"; 
"Assignments too heavy to lead a well-balanced life"; "Need slowing down on 
speeda little less pressure." 

Of those who wanted closer student -faculty relations, Category G, some 
had in mind the desire for more personal interest on the part of the faculty 
and others the wish for the intellectual stimulation of an exchange of ideas* 
Examples of students" comments along these lines are; "More friendly rela- 
tions between the faculty and students"; "The extremely Impersonal and objec- 
tive treatment of students often proves discouraging"; and "More personal 
contact with instructors and professors opportunities for bull sessions -- 
informal gatherings and chances to talk on no set subject with small groups 
and instructors whose thinking you admire. ." 

Students who expressed a need for better guidance, counselling or place- 
ment (Category E) wanted both educational and vocational guidance.. Tvo of the 
more specific statements were: "A more personal and efficient advisor system 



350 ADJUSTMENT TO COLLEGI 



would help explain opportunities for scholarships, requirements for certain 
careers, etc." and "Complete, specific outlines of courses required and 
usually elected for the various majors with a list of occupations, vocations, 
etc*, closely related to the course of study [[are needed]." 

For these four categories it is necessary to note only that the great 
majority of the percentages are less than ten, and that except for a small 
but consistent tendency for more nonveterans than veterans to suggest better 
guidance and placement, no consistent difference appears between veterans and 
nonveterans with respect to frequency of comment on, these codes 

No consistent relationship with AAG appears for the nine selected groups. 
Of the more than 100 significance tests made regarding the relationship of 
such responses in these four categories to AAG in the 16 groups for which AAG 
was computed, one was significant at the 1% level and four others were signifi- 
cant at the 5% level. 

Preference for Own University 

In view of all the various comments given by students, it is pertinent 
to inquire, as was done in Item 15, "If you could be admitted to (and could 
get housing at) any other university you might choose, do you think you 
would still want to attend the institution at which you are now studying?" 
To isolate out the "hard core" of completely loyal students, only those who 
definitely would still want to attend the same university were put in Cate- 
gory A| all others were assigned to Category B. Thus Category A represents 
those students who would definitely prefer their Alma mater to any other 
college* Eesults for this item are given in Figure 70. ~~ 

The majority of students in the typical group 
prefer their own college to any other; there is marked 
variation among the colleges in this respect, however B 
While male veterans and nonveterans expressed similar 
opinions, women were somewhat more favorable in their 
attitudes toward their college. In general, the stu- 
dents who prefer their own college tend to make slightly 
superior Adjusted Average Grades. 

One striking feature of the upper part of Figure 70 is the spread be- 
tween the lowest values and the highest. Of more general significance, how- 
ever, is the fact that in the typical college among the basic twelve, a 
majority, both of veterans and nonveterans, regard their present college 
as their first choice* Veterans and nonveterans are about equally favorable 
to their own college. There is some indication that sophomores were slightly 
more favorable to their college than the corresponding freshman groups; 
a higher proportion chose the favorable category in seven out of nine com- 
parisons o Some fairly marked differences appear between the attitudes of 
women and men in particular colleges, with the over-all results indicating 
that women are somewhat more likely to prefer their present college to any 
other. 



THE STUDENT VIEWS HIS COLLEGE 



351 



100 



CO 

-p 



I 

CQ 

S 

.p 
fl 



75 






00 









o 






,- 


k i 




k- 




ooo 




eo 


o 
ooo 


H mmmm 
f 


o 


@ 






00 










o 






O 



00 



I Median 
Veterans 
o Nonveterans 



MV m 
Definitely 
prefer present 
university 



m MN 

Uncertain^ or 
prefer other 
university 



HER COTT CHOOSING DESIGNATED CATEGORIES IN EACH COLLEGE 



150 



130 



no 











( 


^"""- ^ .^^^ 


4 




<! 


^ ^^ 
"^"> 


( 








Veterans 
o Konveterai 


is 


Definitely Uncertain,, or Total 
prefer present prefer otner group 
university university 



MEAN AAG OF MEDIAN COLLEGE GROUP FOR SPECIFIED CATEGORIES 



FIGURE 70. SATISFACTION WITH PRESENT UNIVERSITY: ITEM 15 



352 ADJUSTMENT TO COLLEGE 



This item is associated to a moderate degree with AAG, as shown in 
Figure 70. !Ehe group which prefers its own college to any other has an ad- 
vantage in the median AAG value which amounts to slightly less than 5 points 
for veteran students and slightly more than 5 points for nonveterans . The 
consistency of the advantage for the satisfied group is apparent in both 
veteran and nonveteran subgroups in the twelve basic groups. When the mean 
AAG's for the less satisfied students are compared with those of the better 
satisfied students in the same subgroup, 17 of the 2^ comparisons favor the 
better satisfied students, and there are three ties. This result is signifi- 
cantly different from chance expectation at the 5$ level. However, when the 
relationship of over-all satisfaction with the college and AAG is examined 
for each subgroup separately, the difference between the less satisfied stu- 
dents and the more satisfied students is significant at the 1$ level in only 
one instance and at the 5$ level in five other instances. 

The slight differences between veterans and nonveterans on this item 
and the relatively weak association of the item with AAG make it evident 
that this item does not contribute to our understanding of veteran-nonveteran 
differences in AAG. 



Conclusions 



Miale veterans and male nonveterans held quite similar attitudes toward 
their college and its program,. Differences in details appeared; for example, 
veterans tended to be slightly less favorable in their attitudes toward 
instructors. Fairly distinct differences appeared on various items in par- 
ticular college groups, but statistically significant trends did not appear. 
Women showed attitudes which were similar to those held by the men* Here 
again, the trends were not consistent enough to be statistically significant, 
but women were somewhat more likely to regard their present college as their 
first choice, and were slightly more favorable in attitudes toward instruc- 
tors than the men. With regard to study facilities^ male veterans were 
slightly less likely to be dissatisfied, and women were less likely to be 
completely satisfied. Taken as a whole, the similarities between the groups 
are more striking than the differences. This generalization does not neces- 
sarily hold, of course, for each college group; the specific differences in 
attitudes between veterans, nonveterans, and women in a particular institu- 
tion may be quite meaningful in terms of its traditions and policies. 

The degree of satisfaction with various aspects of the college program 
varied from college to college and from item to item. Uaing the median 
value among the twelve basic groups, the following picture emerged; Slightly 
more than half of the students reported that they were "fairly well satis- 
fied" with the kind of education they were getting; the others divided about 
equally between "very well satisfied" and "somewhat or very much dissatis- 
fied." About one third of the students reported that they "seldom or never" 
had misgivings about the value of their courses, while only one fifth "fre- 
quently" had such doubts* The majority of students considered that most or 
all of their instructors were good teachers. A preference for their own 
university to any other was expressed by a majority of students, and nine 
students in ten preferred their present division within the university to 
any other* 



THE STUDENT VIEWS HIS COLLEGE 353 



CM the basis of the middle group among nine selected groups, a number 
of additional points may be made. When asked how the college might be 
changed to help them get what they were after in a college education, only 
one ttudent in seven made no suggestion* Popular types of suggestions in- 
cl'idech better instructors, instruction and courses (one student in three); 
fever, or different, required courses (one student in four) ; changes in 
jcsaeral requirements, especially with regard to grades and examinations (one 
dt- .dent .in six); and more courses, teachers, or classrooms (one student in 



In general, a clear association was found "between satisfaction with the 
college and superior Adjusted Average Grades. This was true whether satis- 
faction was expressed in terms of interest in present courses, enjoyment of 
present studies, frequency of misgivings about value of college studies, 
general satisfaction with the education being obtained, preference for 
present division, satisfaction with place to study, and preference for 
present university. A consistent tendency was found for attitudes toward 
instructors and AAG to be related insofar as veterans were concerned; the 
relationship in the nonveteran subgroups, although in the same direction, 
was not statistically significant. 

The implications of these results are not clear. The fact that AAG's 
are rather closely related to actual grades suggests that poor academic 
achievement may color a student f s outlook, to a greater or less extent, on 
all of these matters On the other hand, the hyppthesis that dissatisfied 
students do not work as hard or as effectively as their more contented fellow- 
students is quite plausible. 

The type of suggestion for improving the college which a student made 
was not, typically, related to AAG. Veterans who suggested fewer or differ- 
ent required courses showed some tendency to earn lower AAG's; however, no 
similar relationship was found for nonveterans and none of the tests for this 
category in the separate subgroups turned out to be significant. It would ap- 
pear that the particular topics used in classifying the suggestion were matters 
of concern both to overachievers and underachievers 

The similarity in attitudes expressed by veterans and nonveterans toward 
their college and its program make it unlikely that these items would be 
helpful in accounting for veteran -nonveteran differences in AAG; application 
of the sign test to the various items confirms this deduction. 



355 
APPENDIX A 

SUMMAET GUIDE TO IMERBRETATION OF APOTDIZ TABLES 6("b) THROUGH 46 

Statement of Items 

The stem of each item Is, in general, quoted directly from the question- 
naire. The response categories shown are in some cases rephrasings of the 
alternatives as printed in the questionnaire; in other cases, two or more 
of the alternatives have been combined to form a response category, "Ho 
Response" being treated as one alternative. The numbers in parentheses fol- 
lowing the response categories refer to the alternatives on the questionnaire 
included in each category. Where no numbers follow the category, the item 
provided for free answers rather than fixed alternatives. 

Headings of Table: 

Ho Response includes all omissions of the particular item by students who 
answered other items of the questionnaire. This percentage is shown sepa- 
rately even for items where it is included in one of the response categories. 

Mean AAG refers to the average Adjusted .Average Grade earned by students in 
the designated subgroup who chose the response in question (Cf. p. 90). 

MV, MN, YN refer to Male Veterans, Male Nonveterans, and female EFonveterans 
respectively. 

Abbreviations Used in Designating Universities (Cf . pp. 67 and 68) : 

CS = Central State H = Harris EC = Eastern City 

E =s Evans A = Adams T = Turner 

WS = Western State D = Douglas MT = Midwest Tech 

M = Miller LS = Littletown State MS = Middle State 

S = Stewart MC = Midwest City ST Southern Tech 

m = Midwest State 
Twelve Basic Groups: 1-9 and l4-l6. 

Time of Entrance of Groups I, 13, and 17: 

Group 12 (Turner University, Arts) and Group 17 (Southern Technological 
University, Engineering) included students who entered in the fall of 194-5 
or the fall of 1946. 

Group 13 (Central State University, Arts) included students who entered 
during the first or second semester, 1945-1946. 

Underlining of Mean AAG * s : 

Underlining of a Mean AAG is used to indicate the level of significance of 
the difference between students choosing a particular response and other 
students in their subgroup. (MV, MN, or FW") . A full underline indicates 
significance at the 1$ level; a partial under line, at the 5$ level. Sig- 
nificance tests were applied to Mean AAG's only in those columns identified 
by underlining "Mean AAG" in the heading. 

Parentheses around Mean AAG*s: 

Mean AAG's enclosed in parentheses were based on fewer than 10 students. 

Dashes and Blank Spaces: 

A dash in the Mean AAG column indicates that no student in the designated 
subgroup chose that response; a blank space in the body of the table indi- 
cates that the analysis plans did not provide for obtaining the information 
belonging to that space. 



356 



APPENDIX A 



H 
O 
O 



O 
H 

8 



a 

H 



' 

vo o 



H TJ 
& I 

"8 



-3- IA 
H H 
ft W 



H 

OJ 

-P ChlfN 
^ ^^ 

o H on 

*H J- j- 

fc 0> ON 

(1, H H 

< PQ 6 



All Questionnaire 
Bespondents 





o 




-^-OJIAHOOOJOJOJO IA ONOOON H 

onooonooooooonooro oj oo oo OJ -=t- 
HHHHHHHHH H HHH H 


1 


Jt'oO 
H H 


i 




VO ooOJ IAO lAiAC OOOOOOCO on C-- OJ t^-OO 
VOOOC OJlAOVOt-OCO 1AONVOVO LAVO OJ-4- 
^fr OJ OJ .4- H H .4- H OJ on H H H 


OJ OJ 

OJ"H 




O LT\ OJ CO on 
t OO oo H IA 


1 Veterans Making Each of the Following Besponses: 


No Besponse 


ii 




i i i i o i I i i i til i 
I I I 1 IT\ 1 1 1 > 1 111 i 

1 1 I 1 H 1 1 1 1 1 I 1 ' * 




if 


Jj | 


i 


8888*8888888888888 


o o 
o o 


m 
8 

1 


8*8 8 8 


Besponse E 


|o 

<B < 
SI < 


cc 


OJOcoOmC OJ !> ON oo [-~OO ON 
ooooojnoooCMooOJoo on -zfjtfiA CVI 
H'HHHHHHHH H HHH H 


CQ 


if 


-p 

C 
CM 


i 
s 

1* 


IAO ONO ONVO onvo ojoncovo o LA o IAONON 

HHHHonHononOJOJOJOOHOJOHO 


vo o 

H OJ 


1 


8888 


Besponse D 


9|d 
IIS 


OC-OOHHOJCOOJ C? b- lAJH lf\ 
roOJooooooononOJOJ on ooOJ Jon oo 
HHHHHHHHH H H HIH H 


\ 


LfNON 
LAOJ 
H rH 


-P 


t- 

6 

q 


CO MD O O O IA-4" ONCOOJ OJ IAO t OO 0\CO-=J- CO oo 


O 

s 


1-3-VQ OJ O O 
H 00 OJ O O 


Besponse C 


1 


< 
< 


S 

(H 


OJ O t>-J-,VOCO H ONOJ VO -^ in LA LA 

on oo oj OJloo OJonojon oj oooooj oo 

HHHHHHHHH H HHH H 


1 


33 


P 
k fl 


6 

CQ 


-4- ro onvo co o vo H LAVO onONooojco OJCOO*NOJ o 


i 


pr> t-H H O 

onoj ooo o 


Besponse B 


1 


i 


S 


VO on OJ H t>~CO LAONOO ON LTNVO^t CO 

oo co-sf- on OJ on ro-=h OJ H on oo OJ oo 
HHHHHHHHH H HHH H 


i 


c-o 

H H 


1 


o 




OJ lACO^l- OJoniAOJLAOJ\OOJ-=j-OONOHlA 
OJOJHOJOHOHHHO on3- OJ H OO OJ H 


S3 





O OJ-3* OOO 



CQ 


1 


1 




t-~lf\VojHl CQ.VO LA.OJ t*- lA-*_^- H 

-^u-KOLd-J voUf oo on oj -=h on on vo 

HV1 HIHJ I ^j,.^^^ -! HHH H 






IAO 

SS 


CQ 

M 


si 




HvOHOOLTsojon LAVO ONco onvo o^- on-* 

HHHOJOOOOOOOOOOOHHOJHOJ 


3S 




s*ss 


UOTSTATCC 


CQ t< H 


1 s 




CQ to >H 




ATUfl 


8 H i SwW <fl2ig & 8|iSgSia 




O Ej Ej 

fx\ <Cj CQ S S 


dno.io 


H OJ m-sfr IAVO i>-co ONO H OJ on-=l- IAVO t-co 

HHHHHHHHH 


ONO 
H OJ 


H OJ oo^t IA 
OJ OJ OJ OJ OJ 



LA 
IA -p 



Ctj 
ft 



CQ d 

H 9 

r O "H 

p 

^i CD 

O 

3 & 



O C! 

f I 

CQ 



+5 J! 

"2 ? 

S r^ 

CO iH 



B 



II 



Cj M 
O 

^ 

-P cn 

cd 





<n -P 

fl 

H 



APPENDIX A 



357 



Table 8(a)* 
Item 8. (a) In which of the following did you serve? (Check any that apply.) 

Aa Army (l) 
B. Navy (2) 
C* Marine Corps (3) 
D. Other (Coast Guard, Merchant Marine, Field Services) (k } 5, 6) 


All Questionnaire 
Beepondents 


c 
1^ 


SaSasassS 1 S3a 3 




^ 


)f 
A brief explanation of the organization of this table and the symbols employed is given on page 355. 

y y 

Students in these groups (except 12, 13, and 17) were freshmen when the questionnaire was administered. 


i 


^o onoj ITNO ifMTM>-mmmco one ojc oocuoj 

^Q CO t OJ ITNOVO C--OCO IfN OAVO VQ lf\MD OJ -3- on~4 
.4 OJ OJ-4 H H-4 H OJonHr-iHOJH 




O ITNOJCO on 
t-i on on H ir\ 

H H 


Veterans Making 
No Response 


3 c 


J O O O i IA l 1 l 110 O 

1 l CO -4 CO 1 VQ ' 1 I l l CQ ,4 
l I -^r-i-^ i H i I l i H H 




l l 
l 1 
I l 


-P 

Ct 


OH M *KMOH| W g g g-ttgHNg H 

J 


H 
b> 


88888 


Veterans Making 
Response D 


aicji 

J9S ^ cr 
SII "Sj 
p. 


i 


i 




3CO\OOHQcQooOm CO -4 j3- O O 

4 on OJ on.4-,4 O H _4lon o -3- OJ H OJ 

3HHHHHOJHOJH H HHH H 


O O 

as 


1J ^ >Vj ** **~ f ' Vl ~"' _ * Vjjr * 




3 

fi r 


i oj oj onoj m H^IOJH onojo-4-oojrooj OJH|OJOJ H 
j 


i 


i 


g-*88* 


Veterans Making 
Response C 


I 

P p 

CD CD i 
P^ I 


1 ^*-^ ^^ ^ *~^ ^ 

1 OJ J- t C O VO t~ t Q OJ CO ITS LTNVD tf\ 

3onOJonH-5t-4-HKOon ol ojonH H 


1 


H H 


) 


ir\vo,it_=t- on-4-vo-=(-co OJj--d- iAon.4- i>-oj rn^ ON 
joooooooooooooooooooo 

\ 


V*w.w 

I 


Veterans Making 
Response B 


5 
' < 


oj c oj co on t j t- ON ^o H ir\on o 
onojonojonojonojoj OJ -=j-onon -4- 

HHHHHHHHH H HHH H 


r 


il 


VOVOVO O ONoncOVOVO H Lf\O\VO ONVO IT\VO H O OJ 
ITX-4 lf\,4 VD IT\ UTNVO IA lf\_4- OJ H LT\V) on lf\ LT\ IT\ tA 


i 


ON ON H -4 H 


Veterans Making 
Response A 


M 


t- ChlH -4-OJVOOJonCTN \D VOCOCXD ^t 

Smw* onononononoj OJ onOJOJ -4- 
HIH HHHHHH H HHH H 




H"H" 


p 
PM O 


covovoj- H oj LT\osonir\ONon ONVO C*-VD H -4; oj MD 
on -4 on LT\ oj -4 onoj on-4-4Mi? tonoj ir\^4- .4 ^4 on 




SR ff 


ao TSWa 


CQ JH -H 

P S) Is CQ 




t& 


i! -S < 




-? 


" AftlQ 


owscow;p3SHEHoSS^S!!3 




^<U coSS 


anojco 


H OJ m-4 IT\VO C CO O\O H OJ on^t 1T\VO DCO O\ O 
HHHHHHHH H.H OJ 




o1Slo7-cSo7 



358 



APPENDIX A 



3 

* V 
<Q CD 
^ 3 

o o 

I 



I 

2 

CO 



J* 
no ir\ 
OJ 02 to 


All Questionnaire 
Respondents 


3 
23 


^ CM IT\ H 00 CJ OJ CVJ O lf\ ON OO ON H 

ro oo oo oo no oo oo oo oo CM " oo on OJ -=h 
HHHrHHHHHH H HHH H 




^vo 
-=t- oo 

H H 


Number 


VQOOOJIAQ lAlA^fOCnoOOOoOl>-OJt s --OOCV 1 lOJ 
VOCO C OJlAOMDt-OCQ ITNONMDVD ITNVO OJ j* m-=J- 
J- CU CVI J* H H J* H CVJOOHrHHCUH 




O IT\ OJCO 00 
t>- oo oo H lA 
H H 


Veterans Making 
No Response 


3 
il 


iiooo'iAii i iri o 
i i CO moo I VD I i i I I -d- 
l t ^-H^- I Hi I I ' I 1 H 




1 i 

i r 


! 

(X, 


OHlM^OHh.O OggOOOOg 0^0 g 

} 


H| 
> t 


88888 


Veterans Making 
Response D 


d 

25 


CO\HHLsl-OVDOOroON ON VQoooo j>- 
^ oo J-IiAroo-^t" oo OJ vojro H ro co no ifNl 
Q H H HIH HHHHH H HHH H 

e 


M 
1 

3 
^ 


PC- 

3 


n3 

ffiS 


s 

jaooHGOon-Bhaooovocv(ifNHVOi-OMr\oj<\| oo\o 

s OJ(nOJ04OHOOHr-lH^-*OJHcvoOJOJ04H 

ti 

!^ 


838liW> 


Veterans Making 
Response C 


Se 
23 



"C 

^ t*- 00 C3\VQ C^- OJ CO C30 O VO OJ 00 VOJ 

-oooJooojooooooojoj on oomoj oj| 

qHHHHHHHHH H HHH HI 


D 

8 


H C- 

* oo 

H H 


-P 

JH P 
(D 
ft* O 


I 
C^hCOOJVOlTNOOVDHOJ t -=* VD lf\ O\ lf\VO O VO lAVQ 

t^j- on^-j- c\j-^- oo oo on oo OJ ^ oooooo-sj-j^- oo oo_4- 

5 

cJ 


EH 


C-C-VO C-ON 
-4- oo tAlA-d- 


^p 43 c*_ 
H g S 

^S ivrT 

to *-_ - 

!** 

lUsS 
3&s 


Veterans Making 
Response B 


SI 
9 c 

05 < 

s|< 


cj 
EOOJOJooOJHjt>-t>- 1A OJOOC^ VO 
00 OO OO OO 00 OO 00 OJ OJ 00 -4" 00 OO -=fr 
QHHHHHHHHH H HHH H 

] 


1 




S H 

33 


43 
h ft 

&& 


s. 

:}j-co oj-ifooo cojooj lAOCQv.oroao ocx) cr\oj 

> 00 OJ 00 OJ VO 00-=h IA IA J- lf\ H H 00-=t H OO 00 OO OO 


g 


SSSI^ 


,4 p Ip Ip 

DJ^tVD 

H? H oj on 
< W O ft 


1 

lA "^ 

S <3> 
Q 


a 
13 






ooco 

S3 


CVJ Oj OOCO CQOJ.POOOO OO ONCQt>- 00 
ITN H H|H CM GNlOJ-5h.=t ON OO OO OO OO 
HHHHH"- 'HHH >-> HHH H 


sl 
l 


-p 

3J QJ 
PH 


OJ-si- IT\OJ H LT\00 ON OOOD ^1- OO OJ OO OO OJ CO OO OO-d" 
OOOOHOOOOOOOOOOOOOOO 




S^888 




uo-FSfA-fa 


P &) i! m 

U=rcss = scEs: = e errs Sh3s 
< W <ffl 




CO JL, -H 
P S) ft 

^ s s R H) 
< &< 




AT^a 


8 w g Em wnaii*8lig8S!3 




w^wlS 




dnoj'3 


H OJ OO-* iPvVO t CO ONO H OJ OOJ- tr\VO CCO ONO 
HHHHHHHHHHOJ 




H OJ oo^t 1A 
OJ OJ OJ OJ OJ 



ir\ <D 
LA P 
KN CQ 



3 






tJ 0) 

I i 



I 
1 ^ 

II 



1 



APPENDIX A 



359 



S <=< 



o 

00 



ng 



ans 
Resp 



ing 



Veterans 
Respon 



ng 



Veterans 
Respons 



Vete 
Re 






dnojf) 



-4 cvj ir\ H on cvj OJ oj o tr\ ON on ON H 

on on on on on on on on on cvj on on oj .4 

HHHHHHHHH H HHH H 



vooncvjiAOifMAt-on ononoo ont-cvic oocucvi 

VOCO C OJ 1T\O VI) C O CO ifN ONVO V) ITN^ OJ -4 on^l" 
-4OJOJ.4HH,-4 H OJonHHHCVJH 



ON-* on cvj O on H i o 
.4;onH.4Cr\onon I H 



O t^-oo t ir\ 
ON ITNCVJH H 

> HHH H 



cjcviojonHcjcuocM IAMD Honcuoncvionroi-iir\ 
oooooooooooooooooooo 



VO VO ON.VO O -* O IA !>-{ 

*" j- lAlonvo .4 on^ col 

HHHHHHHHl 



VO VO H VO 

cvi on irvlcvi 

r-1 HHH 



lAoniAoo H t-vo ir\oncu o ON cvj ON oo IA ir\ t- ir\ cvj 

OOOOOOOOOOOHCXiOOHOOOO 



o-=j-v>ononco-4-oco o o !>--* 
on on cvj on CM ONiTNon-^- H -=t on cu 

HHr-lHH>-^HHi-l H HHH 



feSaS'SsSs^ a 



CO H .CVI ON 

aa'a a 



; on on on c H r-co ^t v 



-=* 
-co 



-* vo ONCO cvi H cvj H cvj j 



-- 
oooo t-co to-oo t>-co c*-vo Jt b-oo t*-c oo oo oo 



.4- CO t-sf t O\ 



-4- ITNHlPv 

3 ^ss 



H cvj on-4- IAVO t-co ON O H cvj ro.4- LT\VO c-co ON o 

HHHHHHHHHHICVJ 



O irv OJ CO on 
c on on H ITS 



H ovo ON on 
O H H on ir\ 



go < 



t-O CVJ CVI 00 
C D-lfNCVI 



cvj on-4 in 

CVJ CVJ OJ OJ 



1 

rcf -r( 

Pa CQ 

O 






t? <D 

fi 



3 



O 

s 

IS) 



360 



APPENDIX A 



3 

3 



<D 

8P 

H 
H 



'S' P 



on 
o? 



ft 



m 6 

fl CQ 



{ OJ 

12 



.35 

i fl rt 



S O CQ 

< pq 6 



stionnaire 
ents 



Qu 



O ffi 
P 525 PnO 




ing 



9 P< 

^ CQ 



-P W 

i? 



HfK^C 
ff 



S 

it 

^ co 



r 



(D (D 
PWO 



-4- OJ IA H on CM OJ CVI O 



vo on oj ir\o ir\iAt^ononono 

VOQOt-CU 1AOMD h-OCO lA 



ITN ON on ON H 

ononcu .4- 

H H H H 



cvi 

H 



o cu c\j 

- 



o cvjoi ur\ 



gonHHCUOHHH-sj-OJHOOJHOOHon-d- 
OOOOOOOOOOOOOOOOOOO 



ir\ o\^t oo t- on ir\ i oj VD D-OJ^- on 
on^- cu H cvi H on i M on on-* on t> 

HHr-lHHHHlH H r-IHH H 



OJ on ONVQ oj^t cvi on .4- ITNHCVJ o 



OJ CM on IACO 00 IAOO OO O H.4- ON-4- VDOO CO on 00 OJ 
HHHHOHHOHOJHOJOJHHOJOHOH 



IT\ HI V> OJ on on H O ^0 -4* H Oj 00 0\ 



00 



3 



H oj on-=f i 



t>oo ON o H oj m_4- IAVO c oo o\ o 

HHHHHHHHHHCU 



J-VD 



o IAOJCO on 



8 8 



3 



on on irvoo on 



H oj on^t ir\ 

CVJ 04 OJ OJ OJ 



I 






I 



Sf 

o 



s 

I 



-P 









(D 

I 



I 



IA tn ju 

<IH in -p 

*^ rt +3 

H DQ 

^ TJ -H 



APPENDIX A 



361 



00 



go 
3 o 



king 
se 



Veteran 
No Re 



g 



Veteran 
Respo 



d. o 



4- OJ IAH onOJ OJ OJ O 

ononmonononononon 

HHHHHHHHH 



IA _..._. 

oj ononoj 

H H H H 



onoj IA 

COt~OJ 
- OJ CVJ -st H H Jt 



onononco 

OCO LAGN 
H 



lf\VO OJ-tf rn 
OJcnHHHOJ 



aO IA OJ IACO t- i lA O O O O O 

j& Ch OS^t on\0 I IA O\ 1A 0> H H 

H H ^ HHHiH N^ H^"H H 



on 
H H 



o IAOJOQ on 
on on H IA 

H H 



HHK.H H gjHg gj H OJ H gH M 4H| N H 



\^- O O\ H lf\ LTN C^-kO CO 

m m on m on rn-4- H CM 

H H H H H H HIH H 



IA CO CO -sf OH 

oo on m on ir\ 

H H H H H 



O H CM m IAt>on 0\CQ onMD IAH 1A-4- 
^HHHHOHOHHOJHHHH 



OOHH 



^ on ir\ H aj CM ON en ON -* OOJCQ 
n on on on m on oj on OJ OJ j- on oj 

HHHHHHHHH r-( HHH 



CO CO C-VO ,4;HD-HH\OOJ-*l> iA^O 0\-* O lA O 
CO CO 00 CO CO O\CO ON CO CO t-COCOCOCO C^-ONOScOCO 



OJ on Jt LAVQ t-CO OA O H OJ on J- 1AVO t^OQ ON O 
HHHHHHHHHHOJ 



o\o vo OJ on 

H H H H H 



CO ONCOCOCO 



9 



OJ on J- lA 
OJ OJ OJ OJ OJ 





H 

P 

to 

fl) 



I 



362 



APPENDIX A 



o 

I 




H 
-P 



I 



il 
of 



CD 
rtf 






I, 



gS 

CQ H CQ 



<< W O O 



is 

CO* 



S +" 

9 

O ff 

H - J 



pq P4 O 



! 



VOfOOJlTNOlfMfNCfno^POCO ont~-C\Jt-OOOJ<M 

vooo t-cviiAOvot^oco tr\o\vovD mvo oj^- m^ 

-^ OJ CU <st H H -=J- H OJfnHHHOJH 



^HHkUO-^OHNg g O g g gHlMg g g g g 



^tajirxHrocuojojo 
on no rn oo no no on ro oo 
HHHHHHHHH 



3 



O Q O ' O IA i i I O I I ' 

m ONCO I CO ' VO ! I I t I I I 

-_~_ i ^ i H I I I H I i I 



I>lt- C*- OVCV1 CO VO GO OJ ^t 00 VO 

H oj roHlinq oj oncu H onoj 

HlHHHHHHHH H HH 



OOVOCXJ OO 

or>aiir\ on 

H H 



CO MD VO 00 lf\-* VO VO VO O OMA CO-=|- IA OJ OJ ^t VO ON 
QOOOQOOOOHOOOOOCDOOOO 



0000 IAOCO OjOJVO CJ M3 

on ooioo on on-* 09 on en cvj 

HHHHHHjHHH H 



- 

no aj q en 

HHH H 



H O\OM>- O\ 



VOt-~ 
OJOJ 



lAOOJ-d"lf\O 



on O-CO-4- CT\C H H . -4- ITvOJkO -=h 

, n ^ i-j^pn W 



oj oo-=t- IAVO t~- co ON OH OJ on^t ir\vo b-co o\o 

HHHHHHHHHHOJ 



-4- 04 on O|-sf 
C5 



O (A OJ CO no 
C no on H tA 



88888 



O b- 

J* CO 

H H 



t-lf\V) lf\O 



C-OJ 

^3 



V4D on H O t 

oj onoooj H 



OJ OJ 0?"OJ 0} 



iTv 
IPs -P 
K*N CQ 



f 



3 

H 

1 3" 



1 9 



APPENDIX A 



363 



bfl O 

a ft 

CQ 

H * 



3 



M 
00 * f ~ l 
-p 



o 

J? 



CO 00 



CQ * ' CQ 

J5 tD 4-4 

P Q O 

ffj a t$ 

" 1,00 

_, l5 

+3 .p 

ctf n o -P 

43 ^ 8 

CQ i > 3 



CQ .CO 



CQ fl -a 

rd 43 o o -P 

3 H -P 

aS O ft 

VO ft 1? 

CQ 3 

fl *H 3 CQ J& 

00 CQ CQ Q ^ 



so 

CO 



stiaire 
ondents 



ing 



P4 



PH 



dno-io 



VJD OOCUIAO lAlAt-OOOO OOOO OOh-CUr-OOOJCU 
^OOO^CVllAOVOI>-OCO IA 0\VO VO IAVO CU ^ m J* 
4-CJOJ^tHH^j- H CUOOHHHCVJH 



B 



, CM IA H OO CU CM CU O tA ONOOC3N H 

oo oo oo m oo oo oo oo oo 04 oooocu -d- 

HHHHHHHHH H HHH H 



l I O > O i IA 1 ' 
i OQ l 00 l VQ 8 
i i ^ i v-* i ,-t i i 



gO^g^gH.vgOOgOOgOOOOOH 



VOtAOJlAmiAirNO O OCOCO V) 

oo o\ co cu H CM tr\-=t H o cu ITS H 

'H> 'HHHr-IHH H HHH H 



H C\|--fCVJm H CVl H rOCVI OJ CU HVOnlWOJ CVJ CVJ -d- m r-l 

oo oooooooooo oooooo 



no ir\o\H ir\o vo ONC 
ro oo ro no-* oo^ C 
' 



in, CD ajjcj 
Hi oo HIOJ 
H H HIH 



JtODHVO\OVO CUVOCVfVO^t-00 
CUHHooOOOJHajOJCVJQJCVJ 



ao o\ov 

21 as 



3 g 



OJ C\J CO CU >- H ChMD oocO-d-O\OOIACOOoOCX>Cr\o 
OJOJHCVJHOJHHCUCMCUHHHHCMHHHCVJ 



O\ H H CO H IT\ OJ ITXC's 
CUOOOOOJOOOJOOHIH 
H'H HHHHHHH 



OO H H 00 



CU ONOOH CT\SO\D HCOCOV>lf\OOO ITNOSOOJ iTNOJ 
CVIHCVJCVIOJCUCUoOHHCUHHCUCUHCUCUaiCVJ 



IA 



00 O t J* CU f- 00 O 
oooo-4-oooooooocooo 
HHHHHHHHH 



OJ IfNOOllTV OO 
oo J-j^-loo IA 
H HH|H H 



J-CO r-GJJd-O O CUooooHj* t-OJ OOlAlAooHO 

cuHCUCurooornaicvicvjcvicv/cucnoocuoocuaicvj 



H CU 00-d- IAVO C-00 O\ O H OJ oo Jj IAV.D C^-OQ a 



N O 
CVI 



O 1ACUOD 00 
t>- OO OO H lA 
H 



OOOO 
O O O 



00 O 00 O H 
CU CU CU CU OJ 



OO CO CJ VO.4- 
CU OJ CU CU CU 



t oo o vo o\ 

O H O O O 



ON CO H CU ON 

oo on cu H 



H CVJ oo 
OJ OJ OJ 



^t IA 
OJ CU 



IA 
LA -P 



O 



!> 

H h 






; i 

! I 
I* 

i! 

CO 
tJ 

^ 



H h 





H 



^ 





-P tQ 

jji 



fl 

Tl 



364 



APPENDIX A 



-p 

g 



05 O 
<D -P 



CD <D 
O 



* 03 , 

<D 

CO* $' 

<D g 



-d- 
on 
a? 
r? 



^ 
3 o 

Q? 
OJ O 

gt 



< ffl 



) -P 

1 o 

: oJ 



00 



O PI 

B 



fn cq 



I 



ft 



h fl 

a 





-P 

o| 



VD oo OJ 



4- cvi 



4-CUlT\HoOOJ(M<MO IA 

m ro on m oo oo no tn oo cvi 

HHHHHHHHH H 



m ro cu 

H H H 



.4- 

H 



POQO rot^CUI>-OOC\JCM 

LA d\vo vo ir\vo aj j* en-* 

H OJroHHHOJ 



I 00 O O IAO I 
I OMACO i V0 IA I 
i i v^H-*-' I H H I 



5-: i 

H I I 



c\j ON m cu O\ 

2 333 3 



lf^oo^o^7 C^coco ON oo OJ on H 
^-5 !?d rJ \^, 7 n ^cncvi IA 

HHHHrHHHHH H HHH H 



?j !!? S ^ 9 JO-^ !!? 10 

HHOJOHOOr-(O 



^ -=t Q <^ o o t^-vo 

Oir\Ch 



HHcnCXlHHH 



pq 



H OJ 00^- invo b-00 ONO H O| mj- ir\VO t^-CO ONO 
HHHHHHHr-lHr-JOJ 



o ir\oj co on 
too oo H ir\ 



88888 



CQ i 

: i 

<P -p 

n CQ 

H 






S 

P 

to 





H 
-P 
0} 
IS! 



SK 



& 



& O 

p 

CD CO 



APPENDIX A 



365 



N ' 3 

co g_ 



tering 



-P O 

5 52 

* 5 

^a- 8 

-P -H -P 



H -H 0} O 0) 

O -p -H bD 

Ti p fl R S 

.p ^ ^ -^ 



09 



^ CQ 

) O J -H 
>: -H tJ 

fc h 
H 
-P -H -P 



< PQ o ft 



CO 

I 

P 



All Questionnaire 
Respondents 


a 




-3-OJUr\HoOOJOJOJO ITy ONPOON H 
00 OO 00 00 OO OO OO 00 OO OJ OOOOOJ J- 
HHHHHHHHH H HHH H 




IS 


Kumber 




VO noOJ IT\O iTNLTNt nononoco ootOJ t- O O OJ OJ 
VOCO t OJ ITNO VO CO CQ LTvONVOVQ 1AVQ OJ -=f oo^h 
-=i- OJ OJ -=f H H -=f H OJ 00 H H H OJ H 




O lAOJCO oo 
t on oo H iTN 

H H 


Veterans Making 
No Response 


a 




i i C-- i O if\ i i l tr~ii i 
i I O\ 1 CO I VQ 1 I 1 ^11 1 
1 1 > i ^-- i ,H i t i r-j i i i 




I i 
I t 
I i 


i! 


, 


H gHgHg g g g g 0-1, 


g 


88888 


Veterans Making 
Response D 


W 


PI 


C\J !> H C^- I 00 OJ 1 OJ OO OJ O VO O 
oo oo UTN no i j* Oj i oo OJ -=l- no oo -=f 


1C 

W 
< 

^ 


i i 
i i 
i i 


jl 


1 


1 CO ir\l>- O OJ HO1AOJ ONVO VOVO^tVO COH OMD 
OHOOOOOOOOOOHOOOHHHO 



* 


88888 


Veterans Making 
Response G 


1 ^ 


COLLEGE 


HCO oj ONQ b- oj.oj-4- vo GOOD o\ v> 

OO OJ J- OJ VO 00 IfNpt OJ OJ 00 OJ OJ -=t- 




i I 
i i 


^ 


li 


1* 


O CO H Q\ OJ ON J- COONOHOJONOOJOCO^tOJt 
ooooOJcnO HOOOJCVJH^t^too 00-* oo3- oo oo 


i 

k 


38838 


Veterans Making 
Response B 


III 


1 


3393:111 I m 1 


r 

D 


iRS 

* *H 


*i 





IP\VO ^J" t~* O OO H OO OJ OJ VO IfN Lf\VQ OO 00 OJ VO t~~^" 

oooooooooooooooooooo 


F 

> 


ooHcuO O IA 
H H 


Veterans Making 
Response A 


IN 




ITMT\ 00 H OJ CO H O O J- CO ITN O CO 
oo oo oo oo oo OJ ooloo oo OJ oo oo oo oo 
HHHHHHHHH H HHH H 




sa 


si 

fM 




t^ j co C h-vo ooQ^- ir\ojto t>-H I>-C>-ONO oj 
ir\3- vo -=1* ON t- ON ONVO t- o--st- oo LT\VO -4- -=t- oo IAITN 




t ON o ONIA 
ON ON o co co 


HOTS wo; 




CQ k -H 

-p. oO ft 2 




03 k -H 


^ [I < W 


'Apia 




SwiscQW<<fta^wB8SSgwSI3 




W^coSS 


dnoxo 




H OJ oo-d- lAVO h-oO ON O H OJ oo-* IAVO I>-00 ON O 
HHHHHHHHHHOJ 




H OJ no-sl* IP\ 
OJ OJ OJ OJ Oj 



o 

R I 



M -H 

H 9 

O -P 

& g 



CQ 

I 



s 






q ~p 

3 PJ 

So 5> 

g s 





H O 
P OQ 



366 



APPENDIX A 



*!% 

H5 



3 H O 
Pfl fl 



wop 



.onnaire 

lents 


ll 




-4- CM 1AH OOCU OJ OJ O 1A ON OO O\ H 




-4-VO 


-P R 
CQ Q 
(D A 

S to 

., 


Number 




<Q oo CM lAO lTMTM>-nonoroCOooC OJ f>-O O OJOJ 
<OCO t-OJlAOVOC OCO 1AONVOVO IAVO OJ -4- 00-4- 
4- OJ OJ -4- H H -4" H OJnoHHHCMH 




O 1ACMCO no 
t-oo oo H LT\ 


s Making 
sponse 


1| 




lA 1 O O O CO ' O IA ' O 

o\ i co HCO IA i tr\ j -* ' | 




( I 

! 1 
1 I 


Veteran 
No Be 


-P 

CO CD 
P-i O 


g 


g.H*g g g 0-*,g g HH| W g g 


?VICE 


88888 


tt) 

5 
3 ^ 

1 CO 

to 


i^ 


1 

M 


^o o-vo f- i o CM"?!? o oj vo co -4- 

ooooVOoo I lA^D oo H -4- J- no oo -sr 
HHHHIHHHH H HHH H 


i 


I I 
i i 
1 I 


Veterans 
Respor 


4^ 

S 


AFOIER W/ 


O^fVO-4O-4-Hj4-CO Lr\HVQ-4-0\O\HCO LfNOJoO 
HHOHOOOOOOHHOJOOHHHHH 


! 


88888 


! 

"3 

S <D 
03 


;>,. 


f 


O 00 -4- CM O O\CO lf\ O CO noVO CO IPs 
4- CO-4- OO\JD mt4- CM O CM -4" CM CM J" 
HHHHHHHHH H HHH H 





i i 

! 1 


3 p, 

fc CQ 

(D (0 

r 


p 

<S 


S 


O\OO O t-H O-sCM oOJt 1AVO CCO OO H VO OO-4- CM OJ 
OHHHOOOOOOOHHHHHHHHH 


ffiTUKNED 


88888 


i 

rA PQ 
S 


ll 


i 


OOlTxCOO-O^t^DlAC^ CO HHH IA 
00 CM OO OJ OJ H OO LT\ OO H OO OO OO ^4 


i 

9 


"o 


^ S 

^s 

sS 


p^ Q 


t 


OCO HCMHCOJ-COfOCO fn-4-CO O\I>- 1AONCO IACO 
OJCMCMCMOOOOCMHHCMHHOJOJHOJOJH 


1 


38888 


CD I 


1! 




^^-OIHOOOH.ONH no OtAt- VO 
no oo oo oo oo oo no CM no OJ Jt oo CM no 

HHHlHHHHHH H HHH H 




53 


3 P- 

b 03 

l& 

t> 


si 

P4 O 




H tr\oot*-c s -cr\oo.4- IAOJCQ oj HCO nocpco CM HVO 




*!!! 


UOJS* 


pATd 




$ J KS 




03 t-t -H 








<q w < pq 




S" " S< 




ITUQ 




Sixigsoaw<;fiagaBr3lg^g|l3 




N:wg 


dr 


ioj:o 




H OJ OO-4- ITvVO [?-CO O\O H OJ OO-4- 1AVO t>-00 Q\ Q 
HHHHHHHHHHCM 




H CM no-4- IA 
CM OJ OJ OJ OJ 



" i 

H 

O fl 



CQ 

I 



APPENDIX A 



367 



*H 

<H 



S 

Ox-x 
CD O 



CD H 
P ft CQ 
CD ft fc 



H 

I 



OJ OtT 



!. 



tfl-H 

a a 



VO-sf P! h 
H CO -H 

aj aJ -H 

Hi Hi TJ H 



H P 

g 





CQ CQ <N * 
H O 
H M^-fe 



l Questionn 
Respondent 



ing 



1 



Veter 
Res 



Veterans Making 
Response B 



H H|c\lO H OH|OJH O H O O OJ H QH|f\HkO OHWO OJ 
fe OOO OOOOOOOO OO OO 



43 



Pk 



-,OJlAr-fOOOJOJCVJO IA CJ\OOON H 

oo oo no oo oo oo oo oo oo OJ oo oo oj j& 

HHHHHHHHH H HHr-) H 



OO C-C\i 

CU CU j* 



IAONVO VO lAVjO CU-^t fn 
H ajroHiHHOI 



CO I oo O CO i O i 

' 



IAO i O 



) O-^ HOO VO O OVO 
I H H H"H H*H"S H 



S 3 ! 

H H H I 



IA.OJ IA H CVJ IA.CVJ O ON t> ON OO ON H 

ooloo oo oo oo no oo oo cvj OJ oo oo OJ -=} 

r-IHHHHHHrHH H H H H H 



^r ^ 

ONON 



O\SP J-OMXDh-b-CVJON t--* ITNOO OM>- OJ t- 



- 

CO ON ON oo ON ON co co i>co co ON ON ON oo ONONCO 



OO 

ON o 



OOOJONOOJ 
J OOJ- OJCO O OJ-tf 



C'-C*- VO CVJCVIO 

-tf CU O CU OJ 00 



OOH t*~ OJ Jd- VO O CO 00 HOO-4-OJ O OJ l>- 1>- 
OOOOOOHOOHOOOHOOO 



OJ 00-^- lf\VO t-CO ONO H CV| OO- 
HHHH 



- IfNVO t^CO ON O 
HHHHHCXJ 



d-MD 
4- 00 
H H 



O LA OJ 00 00 
OO OO H 1TN 



8OOOO 
O O O O 



OJ O O CXI 
H O O O O 



c^ ON ON ON co 



Ixj 



H OJ oO-=t lA 
OJ OJ OJ CVJ OJ 



d 


IA O 
IA P 
KN CQ 
H 
<D fl 

C(J S 



* 

> <D 



13 

S h 

3 H 

CD 

1 1 

CQ t> 



B oo 
H 



II 

?4 O 

I S 

IT 



CQ 
P 
fl 




368 



APPENDIX A 



I 



CQ 

P 

as 

1* 

13 

> 

? 



O -P 

S 



!U 5 a 

O CQ 



i<~>. H 
-'OJ > O 

N-' tf O 

> -3 

3 O 

B k -P -P 

*> p O 

-P t3 O 
OHO 
fi 



O H 
3 



00 


H 



O 
O & 

-p CQ 

O -P 



f> 

5 

S3 

Is 
l& 



^} H 
4J oJ 



O -H 
ft qn 

? 



o 00 

o & g 

gH^^ 

5 -p3 -p 

^l 

H *> k rQ 

ft H 

| :3 

^ O H O 

O > 

H -P 

3 H 

^0 



P y 3 <D 

^ss 
^^5 

SS^i 

c3 -P S 
H ft ja 

^ 15H 
^ J4 -H 

g fta o- 

H H H ft 

< 6 ft 



CQ O 



: 






S ft 

CD 

-P pq ft 



a" 



9S 



k CJ 

Ct 



P 
k fl 


ft O 



IIOJST:AT:Q; 



H|{\H|OJH!WOH|CUH O H H O OJ, OJ O OJH|4-O O H O OJ 
O OOOOOOOOO OOOOO 



041AHOOOJOJOJO 

T on on on co oo oo oo on 
JHHHHHHHH 



ITN ON CO ON 



> cnoj IAO LfMT\t no on mo 

>CO t>-CVj UT\OVOt>-OC30 ITN 



oof OJ ir~o o OJ oj 

VOVO IAVD OJ -=h m-=J- 
CVIroHHHOJH 



S\ oj co i co m i moo ir\ IA t 



b-VO 00 OJ O ' \f\\f\ 

oo oo ir\ co ON tr\.=fr ro c 

HHH'H"* 'HHH 



^t (T\t-C VD 

oj oo OJ oo ir\ 

H HHH H 



OJ OOOCOHlcNO H ITNONCX) tCO ^f^OJONOJlACM 
HOJOH OOOOOHHOJHHHOJHHH 



tPvol\C\J I OJ (V>aQ O t 

co on oo oooo ir\-=t i H on -=! 04 on on 

H 



- 

H" 'H H I H 



- 

HHH H 



onO|O-5f\ocMOJjtVO 



-4" OYH on-4* H OJ H on 

on onion on oo oo on on on 

HHHHHHHHH 



00-4- C 
coonc 

HH 



-d* 
on 

H 



VD as-dh onir\HO\ovo ir\ONi>-ooD--onona\cvi ONO 
iAonvo-4- OM^OO oo vo vo jj- onjd- IA ITN Jt ^4- -4- J- 



H OJ on-4- IPvVO O-OO 



NO H OJ 
HHHHHHHH 



Lf\VO |>-CO ONO 
' ' H H OJ 






O irxojao on 

tycoon H in 

H H 



CVJ HO , 



O on H on ON 
t-QO ONOO t 1 - 



* k 

IA -P 

N^\ CQ 
*H 

| | 

ft 

O CQ 



J | 

HP 
O 



O 
*H 

+3 



45 
ft 



P CQ 
ft 



APPENDIX A 



369 



All Question 
Responden 



-4- OJ IA H no 04 OJ OJ O ITS O\ no ON H 

no no oo oo oo oo oo oo oo OJ oo oo OJ -4- 

HHHHHHHHH H HHH H 



no OJ lf\O IT\ tf\ O oo oo oo CO oo F- OJ O O O OJ OJ 



vo co' o oj' ur\ b'vb' o o 60' 

j4-OJOJ^i-HH-4- H 



OJ.4- ro- 
H H Ojl 



o mojco no 



ing 



no 
H 



iii o trx i i o 

i i i -4- i i no 

ill N^. H 1 H 



8 



an 
Re 



|wOH|cuOH|c\jH O O O CU H OH|CUO O njcy OJ 
O O OOOO OOO OO O 



88888 



J 00 VO IA t- 



t- ON t*- O 



O n 
IA 



no 
OJ 



8 



I-4-J- lT\OOir\HJ- 00 C--VO CAOJOJ-^-^-OjHHOOlOk 
OOOOOOQOOOOOOOOOOOOO 



IT\ O CO H OJ 



t~,CO OJ OD O 



a 



Veter 
Res 



8 



OJ O\^t co oo OOVD b-cx) IACO H jt o\ ONVD o oo vo co 

OJHOJHOJOJoOOJHOJOJOJHHOJOJOJHHfO 



O O 
no no ir\-=r no 



ing 



- OH lAOOVOlb-OS no OlfNH OJ 

% 333 S 



Veteran 
Respo 



H OCO VO O 
^0 VO no lAO 






S 



dnoj-D 



H OJ 00..4* 



OCO O\O H OJ 00-tf lA 

HHHHHHHHHHOJ 



H OJ 00.4" IT\ 
OJ OJ OJ OJ OJ 



370 



APPENDIX A 







fl^g 



<! o 



Questionnaire 
Respondents 



ing 



Veter 
Res 



ing 



Veter 
Res 



PH O 



83S 



O CQ O 



2f7 t ^ !0 9 

Si 333 3 



VD ifMA h- H H cnVO 00 C^ O lA^d- 00 O MD H ^ CO 00 
en m-=t- onj* cn-^- ooooon oo^J-^t cvj-4- ro^=i--4- oooj 



BBBBBBB3I 



S 



mvo oj oj oj t>vo ir\vo ^ <r) H VD ir\ t^cp oj \o 
cn^t cnJ- moooococoooojco on^h on^r^h on 



j s s r 



O IfNOJCO 00 

p- m m HI ITS 

iH rH 



I888S 



OJ Hi 

mir\ 

H HI 



H 04 m-=J- ir\ 

OJ OJ OJ OJ OJ 



3 2 
5 



H 

1 a* 



ff 

O 



t^| 

I 8> 1^ 

. g -P *H 

,">,* 



APPENDIX A 



371 



Ha 



> CQ 



5 



la 
II 

d) 



fc CQ 
<D <D 
-P 



-p pq 



-p 

ss 

PL< U 



- 

fc a 

0) <D 
PH O 





feS 



dnoJCf) 



4-C\JlAHcnC\JOJCVJO lT\ O\on(7\ 

oo oo on <n oo m oo on no oj oorocvi 

HHr-iHHHHHH H HHH 



cob- 

CVJ CVJ - 



- - 

CVJ 00 H H H OJ H 



loo IACU ONVJ 







oo o 

.4" -=t PO 
HH H 



oj HH|C\JHH Hoooojojoj 
oo oooooooo 



^0 H OJ OJ O lAVO ro t 
on oo on ro-4- oo oo oo H 
HHHHHHHHH 



t*- 
OJ 
H 



IACO O 00 
-=h co en ^J- 
HHH H 



^h c\j co -=f- vo on ir\-^- ir\-=j- no m ir\vo LTMA cu o\-if o 
cvicuojcvjcvjcurncncjoJOJcncnojojcVjojojoJOJ 




31 



ir\ m o\vo cj mvo IACO KMD ir\ o vo t*-co m 
m en j- ojajmcnmcvjajojcu 



HRlHIHlHlHTHlH 



MD-4- VO O O l^O) mt-OOOOVOOOO-d- HJ- O 

co-3- on-* on m oo oj oo-=f oo oo oo oo^t -^r ir\-n- j- if\ 



O lAlc 

m ir\i 

H H 



^1 

tr\ 

HI 



CQ 

l = 



EEE2ECS 



H CJ oo J* lf\\Q 



H OJ OO^ IAVO C-CO Ch O 
HHHHHHHHHOJ 



O if\ OJ co on 
c oooo H in 



sssss 



OO C-- O\ H 00 
OJ OJ H CVI OJ 




CJNOD^O O\OJ 



OJ OJ OJ CM CU 





^ I 

03 ti 



-p 

CQ 
<D 

I 
I 



8 oo 

H 

H 

1 g 



i 



IA CD 
<N IA -P 



372 



APPENDIX A 



OJ IAH rnOJ OJ OJ O IA ONonON H 

on on rn oo on on on on on CM on on OJ -4- 

HHHr-lHHHHH 



sO roOJ iT\O lAlAh-oncnonOD nnb-OJ t-O O 
VOCOt-OJ lAOVOb-OOO IA ONVO VO lAVO OJ j* 
.=* CM OJJ- H H-tf H OJonHHH 



IA <M t ON C-CO IfN H 00 CO CO O O O CO VO H VO -* 



Ojojoocnoocwai en-* en en cu cvi on cvi cj 



cncovooo mcovo men ONOO voco c-ONOO-sf oj o H 

OJOJCXI 



O OJ IAOJ.LAH H OJ OJ OJ r _,_. , 

sOUflirx-* on lA/iAl* vo 

HIHIH ri H HIHiH H 



OJJ-OOJVO t- H O cnt-HcOOAHcX) lAjd* O b- 
OJajOJOJHHOJHOJi-JOJHHOJHaJOJOJOJ 




LT\ O 
IfN -P 

i^ m 



bo 
w 





0) 



(D 

6 



IB 



m t 
H H 



P ^ 

$ SI 



; 

5 

1 ? 
P ra 



M -P 

5 S 



APPENDIX A 



373 



a 



K 



ON 



OQ 

tJ CQ M 

H *H 

ft 1-4 CQ 

< PQ 6 



5 



I 



5 



PQ 



5 



PM ! 



dTlOJCJ 



CN OJ CO ITNOO ON ON O O 
OJOJOJOJOJOJOOOO 
HHHHHHHH 



4-OJinHooOJOJOJO 

oooooooooooooooooo 
HHHHHHHHH 



C*- 
no 
H 



IT\ 

C\J 



ON Q 
H oo 
HH 



ON OO ON 

oo oo OJ 
HHH 



Of ON-4-VD oo ooj=r IT\ H 
t~- in !>- H H ON oo COOJ 
OJ H oo OJ CVJ H OJ H 

jd-cocncnvocvio\c*-cvic*- 

- - - 



ON-4- ON oj -4- H H o 

H H 00 H VO H H 



jd- oj CD 
o\ oj ON c 



c^ ur\ 

H H 



o 00 
o invo 



OO OJ mO in tn D OO OO OO CO OO C*~ CU C 

sQcot>-ojino v sQ[>-ocoLn ON^O ^ tnvo 

-OJGJ.4-HH.-d- H OJ oo H 



O OJ OJ 

-=t co-4- 

H OJ H 



!O! 

I CVI I 



I S i I I 



I I i o l 

1 '1 1 O 1 

I i 1 I i i H i 







8gS8 8888888 
88888 8 rf| *8 888588888888 



onOOOOJOlPv 




HOJl>- 



vo 

a 



* 



ONCO ONC H VD OJ OJ-d" O OJ-4- 
OOOOOOOOOOOH 

VO 

oj 



CO VO ONQD C OJ OJ 
OOOHOHH 



- 

o H OJ oj oj LT\VO -4- oj ir\ po-3- oo oo 



00-4- OOOON 



H LT\ OO OO ONKO O OO^t 

aaasi^aaa 



OO 

a 



SS 



iPvONONO.H 



oj ir\,t-to OOOHOJ 



c-- 

a 



vo H o 

aaa 



a 



GO 00 CO <M IAGOVO OCOOJONOOOVOCO OX ON^f 

-3F OOOD ir\ooM5 ir\ t IAOJ H-4* if\ oo-* oo oo-=f 



ITNVO t-co ON O 



O H Oj oo^t ir\vo t~cq ON o 



CVI 

1 a 



O ITNOJ CO OO 

t- oo oo H 

H H 



888 
8838 



OO H OJ 
O O O 



. OOM3VO 
O O OJ 00 



OO OJ 

^sa 
gx- 



ONO ON-4- oo 

O H O H H 



g^i 



in GO 

-4- CO 
H H 



H H OJ 
ON ON oo 



hV>or> ONH 
co oo in in 



OJ OJ OJ OJ 



n3 

s 



555 
ste 



pa 



tf -H 

P 

!>i CQ 

O 

1 & 



I 





CQ 

S 
2 



a 
a 



& 

tao 

S 8 



! 



& 





H 

P CQ 

CO 

S P 



<J CQ 
* $ 



374 



APPENDIX A 




APPENDIX A 



375 



CD 

5 


,Q 
H 

$H 



I 



P 
ft 



(D ol 

^? _^ 

IS 3' 

P O 
CO O -H 

tt)^ 

p -d 

Ctf -H OJ 



ttf O0, 

H p^ *H <d 

St 3 

H|,g 



p B 

-H 




SI 



Mg 



i 



gg 



dnojf) 



Ox OJ CO UTNCO CAONOO 
rHajojojcvjojcvjmm 

HHr-ir-iHHr-iH 



. OJ ITN Hi on OJ OJ OJ O 

ononononononononon 

HHHHHHHHH 



C- 

3 
S 



r-ICT\O 
H 

H 



O H 
m CM 

H H 



>-* IAH 

\ on co oj 

MD J- CO on onVO CM ON t*- CM C*--* 
^D O\-* ON OJ _* H H O t--* ON 



on 
rH 
OJ 



r-i rH on rH VO H H 



J O. t*- tPv O ?RvO 



CO 

-OJ 



OJ-=J-Hr-l-=h 



ir\vo oj j- ro-d- 

onHHHCUH 



8cxi Hiwon 
o o 

8gg8"*g 8ggS8S8 
g-^g-^^S 88Sggg58g S 8 H 8 



GO I 



ool 



ir- 1- o H oj oj vo ON 



" 



ONVO 

OO 



- o OJ oncp o on if\ 

HHiHOHHH 



t t>--4- vo on oncoloj vo c^- 

H r-j Oj H OJ OJ H OJ OJ en 

H H H H H H HlH H H 



vo-sj- 

OH 
H H 



H 

OJ 
H 



ir\o-5t- co 

saa s 



o\r 
OJ o 



\ IfMAon IfN 

C--ONr-|VO ON ON VOODOO ON fr- 

oncviojonononojojHOJOJ 



H p on^ w .ON .on 

co 
on 



ONOVO c o-colo ir\, 
oj CM oj mbt -=H 



3 

VO 

3 



IAOJ 
H 04 
H H 



-ONVOVO, 

333' 



in 
oj 



HonCMrH"3ojHOOJHOJ. 



- O\VOCO H ITNVO C 
onj--=f IPv OJ H H 



i-=t- C Or*,,-* ONlonOJ 

| S^iam 313 2! 



C-VO t O lA-sf tHOOONOJOOVOonOJOJCOcOOJ O\^ 

oJHOJ^i-ojojojmoj-4-onojonHHooooo 



s (x| = s s < pq r 



if\OJco on 
t*- on on H ir\ 

H r-l 



8S8 
0-8SS8 



vo OJCO 
- on 



OM- 

H on 

H H 



iH H 

t-ro 



vo onvovo ON 
on r-4 o o 



W^ 1 



S S 



CM CM OJ CM CM 



3 

a 





I 

a 



376 



APPENDIX A 



H 



-p o " 

- ftkj 

_ CQ 

3 

CQ 



UOTejAfa 



ON OJ CO LAO 
H OJ OJ OJ C 

H H H H r 



ONONO p 

2233 



I <jo H 

I H on OJ 

I H H H 



- CM lf\ H 00 <M CVJ C\J O 
oooooorooofooocnoo 
HHHHHHHHH 



ITN 
cvi 
H 



cnonoj 
HHH 



fi jft-p. H 

cvj H mcvj 



on-=}- - . - 

^3 ^2 



^^lj^^HS 





CVJ O\t>- OJ O-J- CXJ CO H O O CO LTs 
-3- OA OJ -^ H H O t--4- ON CVJ OM>- LTN O lAVO 
H H H 



VQ 
CVJ 



j^3333 

> ON cvj on on ' on o o 



O O 
MD in 



c s -i>-tr\ 
9^!^; 

H rH H 



cnojoocvj 

OOOO 



cvi 

O 



Hcn 

OO 



Hoo 

OO 



HO 

oo 



O 

o 



on O OJH|CAJOJ OJ CVJ H H on OJ OJ 

ooo oooaoooo 
ononojojojoonHHOJOJHLALAHOJ onHjoaoj cvj 

O O O O O O O O O O O O O O O O O 00 



co cu oj ^t H H t 



t lA 

0X1 



H H -=t- 



s-ss 



O 

"* 



ONOO O O 

cu H on cvj 



ON 
CVJ 



voco 

lAOJ 



O t>- O VO O\CO rn O O\ ON-=h CO H H ON CO 00 H OJ 00 
OJHOJHOJHOJOJHHCVJHHOJHHHrHr-iH 



t~-00 t- IAIACO IA O 



_ ontAOcncocvjL,_ 
on OJ ononononononon 

HHHHHHHHH 



VO 

m 

H 



CO-=t"ON 

322 



LA LA f>- 

332 



ON 



IA tAon-sf t-- 
on-4" on on on 

04 O LA CO H H. 



COON 
cvjon 



ONIA O OJ 

cvi on miA- 



on on on on on on-^t oj on cn-4- on on-if 



CVI^t OJ O\CObO IA|H OJ 

a 2 32 s 






t t^ 

33 



2 3 



, C^-b-VO 
on on on-^ 



on 



vq on 
H on 



t^VO VO CVJ O\\O H 

oj cvj 04 -^- ^t on on 



IAVO COtLAHVOCVlonONCVJH 

o") IA cn^h'0n-=l" on tA-4- on cvj on 

LAcp^- o H IAOO ono rr-vo ONt>-t>-H IAC c*- H ON 
1 ^ LA on ~& on LA iA-4- cvj .4- LA on on m on lA^t on 



8 g S CO W <! fi 3 g g E< S S i 



H OJ on-=J- LfNVO tCO ON C^ r-| CXJ 



IH 



o iTxcvjco on 
Con on H LTN 



i ir\ 

8 H 



H CVJ 




VO 



H LfMAOJ H 
CVJ OJ CVJ H H 



^ ONVO 
K ^ m 
"vi H 



t-O tA 



VO ONCO C OJ 

on on cn-=h-4- 



g 



0\CVJ 



on on IA 
on on on 



onvooo oj t~ 



H OJ on^; I 



! 



tA <D 
LA -p 



<D 43 

O* S 



. S 



51 

S I 







* s 



APPENDIX A 



377 



Table 13* 

Item 13 . How important is it for you to graduate from college in order to do the kind of work you 
are planning to do? 

A. Absolutely necessary (l) 
B. Not absolutely necessary (2, 3j No Eesponse) 


All Questionnaire 
Respondents 


1 


ONOJCQlACOONONOO D HONO H 
HCvJCVJCVIOJOJCVIOOOO 00 HHOO OJ 
HHHHHHHHH H HHH H 

-3-OJlAHooOJOJOJO LA ON cn ON H 
OOOOOOOOOOOOOOOOOO OJ OOOOOJ -3" 
HHHHHHHHH H HHH H 


H OJ 
OJ OJ 

H H 

3^00 


*A brief explanation of the organization of this table and the symbols employed is given on page 355 

**Students in these groups (exaept 12, 13 / and 1?) were freshmen when the questionnaire was adminis- 
tered . 




1 


OJ ON-=t VO oo oo-=h IA H 
t-lAC--H H ONOO 00 OJ 
OJHooOJ OJ H OJH 

VO -d- 00 OOOOVO OJ ON t-CVl C^-,4" CVI CO H O O CO LA 
VO O\-=t- ON OJ Jt H H O --3- ON OJ ON t>~ LA O IAVO |a 
H HHOOHVOHH H H H H C 

^O OOCVIIAO IAtAl x -OOoOOOCQoOC*-OJ C O O OJCVJ fc 
VO CO t*~" CVJ IA O VO O CO IA ONVO VO LfNVO OJ ^" OO-3" ' 




O LA OJ CO OO 
t>-oo oo H IA 
H H 


Students Making 
No Eesponse 


<j 

L 




i 0* 
i oo 

<-N , 


OJIOO-4-VOVQOJOD cO O < 

V^ K_-*V_' V-* V^x Q 
,* v 


IA OJ H ONVO CO ONO O OO IAO 1 
oo VO .sf H lA-4" OJ OJ O -4- C^ O P 
HHHHHHHHH H >-' H < 


p 52; 

Sfo 
r 

y 


p 
.3" .4- LA-=f 1A OJ-^t Jt D C 
OOOOOOOOO PI 

OJ O HjcxiH OJ 00 OJ 00-4- VO lA-sh OHOOOOJOh 
00 OOOOOOOOO 0000000^ 

OOHOOOJOOHOOHHCVIOOO OOHlc^HlcviO O O OO H 

ooooooooooooo oooooc 


H LA 


H OO OO O O 

O O O O O 


Students Making 
Eesponse B 


1 


.4- IA I>- ON Okl- HJoo H OJ -=|- O J- H i 
H H OJ H OJlCVJ OJJOJ 00 00 O [H OJ OJ 
H H H H HJH HIH H H H H H H 

co 1 1*~ oj vo ir\j H t**! cvi .^" o oo co cvj co - 

OJJOJloo CVJlCVlloO CVJJOJ|CV! OJ OOlCVI OJ 1 oo - 


c-o 

H OJ 
H H 

+ (i 
^r 1-1 

J- oo 
H H 


"fl So 
l^l 


" -""""" I 
VO f t>-_d- OJ lA-d- CO C*~ 5 

vovovo3" vo t>-vo m-=j- 
a 

00 ONVO !>- OJ f-VO VO ^t OJ O OO CVICO OOVO ONH LA 
-=}-OOOOCVI-tf > -=J"-3"OOOOOO-=fOO IAOO^1AVOI>-VO< 

p 

OJO^-O-=t-LALAOJcX)ONOOOD-si-O OJHOOMAi 
LfN lA-sJ" -=i" LA-sd' IA IA oo OO-^f* IA-^J" -=i" -^ -4* -4" VO VO D-- E 

i 


VO OOCO 

oo-3- -=lr 

O LAVO VO C^- 
OO-* LA OO IA 


Students JMaking 
Eesponse A 


< 
S * 
lfe 


*" 
OO VO CO CO -^1- OJ VO -=f ON O CO IA LTN H C 
OJOJOJOJOOOOOOOOOJ *& HOJOO OJ 
HHHHHHHHH H HHH H r 

O ONCQ-4- C\J OOCO H OO O OOVO IA VO 
4- oo oo oo^t oo 00-4- oo oo -dh oo oo ^t c 
HHHHHHHHH H HHH H g 


00 -* 
OJ OJ 

oo O 
H H 


c 


4- 00 OOVO CO IAVO OJ OO 
OOOOOOIAOOOJOOVOIA g 

t- H-i OOCO oo^--4-VOCO O t>- CO CVI C-J- H CJN LA ^ 

LAVO VD t-iALA LAVOVOVO vovo ^t vo LA^- oo cvj oo P 

BO VO O VO LA IACO OJHOOOJVOOOOONOHtA 
IA IAVO ^h IA-4- -3- VO VO VO LA LA LAVO IA LA^t OO OJ 


4- C OJ 
VO IA LA 


uo, SWC 


CO FH -H 
-P GO f CQ 


CQ h -H 
-P W h 




^" " fl < 


ATW] 


S w g Sra w<nai^B8lii5SIS 


g<.cog| 


dno,i] 


H OJ O0-=t IAVO tCQ ONO H OJ oo^t LfNVO t-CO ONO 
HHHHHHHHHHOJ 


H OJ OO-4- LA 

oj oj rvi oj cu 



378 



APPENDIX A 



I 



K 



^ 



< W O 



Questionnaire 
Respondents 



B 



R 



Cue 

t-lfM*-H 

CU H cnCU CU 



vo .=* co en envo cu ON t- oj c*--^- OJOOHOOOOIA 

MD ON-4- CJN CU .4* H H O l>--=h ON OJ O\ t- IA O ITv^O 
H HHroHVDHH H H H H 

vo oocxj ir\o ifMTv t-*- co en m oo en t- cu k t 
vooo t>cvjir\ovoi>-OGO IAONVOVO mvo 

Jt OJ CU -4- H H -4- H CU on H 



ON CVJ 
HCU 
HH 



IACO ON O\ O O 
CUCUCUCUenoo 
HHHHHH 



-4-CUlAHonCUCUCUO 
enenenenenenenenen 
HHHHHHHHH 



t- 
on 
H 



IA 
cu 
H 



H ON O 



ON on ON 

en on cu 

, H H H 



on cn^f IAH 

H ONon 



CO CU 
CU H 



o cvj cvj 

-^- no^t 

1 H OJ H 




HOcuvoHcoqHcooj envo iAej\oo^-oo cnoqoo 
OJHOJHmOOjOjHOJCUCUoOOOHOcnHH 



MDV 



O\ ONHOJ 

s 333 



OJ CO 00 ^JD C 



OJ 



en 

2J 



VO O 



O. 

3 



O co t--* m H 
in ITS ir\lr\ iA-4- 

ON H H oo ON ON^ o IAVO -=j- en 
^t LrMAm-=J-^-^rN^-^t^3--* 

ITN tr\-=f ifN-sj- iTMA on ir\ if\-=t ir\ IAVD 8 v$ 



en 
VD 



300 CUVOVO.inoOllAOJ 
cu on cu enlon enion on 



H 

,4- 



JhHJ=J- -=t- 

cu cu on H 



t-oo o 
on on 

H H 



HCUC^ 
^t on on 

HHH 



t- 
- on 

H 



-4- 

H 



CU-4-OOCU 

- oj^t- 



O 

on 



t>-l>- 

cu on 



OJ en 

cu en 



OONCUOOVDIA 
cu OJ on-* H H H 



OJ O\^jr t-CUVO H O ITNonen O\ 
onen on-4- onon on^^t-* en en 

CU-^CUJrf-HVOVOO b-VOVO O H ^4- OJ CU VO on lAVD 

cuonononojfnoj^tojcuencuHcuoncuHHCUcu 



|jvo 



>o 



o ir\cuco on 
t on en H ir\ 

H H 



- O 
CU 



33 
135! 



H C-O Q C 
LTN^t IT\VO lf\ 



-4- onH HOO 
" m en H o 



oj oJ o? oj a? 



355 
ste 



age 



s given 
naire w 



loyed 



I 

A 



Ul 

H 

(D Q) 

s* 

OQ t 

IS 

9 

I ^ 
1 3" 

II 

6 5 

<D * * 

5 o 



g ft 

H O 
P OQ 

1^ 

CO 

S 1 

M 

,< ,CQ 



APPINDIX A 



379 



LA 




H <h 
H O 



J3 h 
-H (D 
<M O 

SS 

< pq 



Stud 
No 



ing 



Studen 
Resp 



ing 



Stu 
R 



8888888 
1*88888^88888888888*8 



S 



IT\ 

a 



o 

a 



IA 



co t 
en I 



ir\ 



o\ \o m 
on cvi on 



-4- ir\ 
vo c\J 



IAVD t^-VO CUCO CVJ OJ - 

VD tr\ 1A H on H on IA 1A h- OJ 



ONVOOVDHt-H 
IA on OJ CVJ onVD LA 



CAOJ LAoniAlAO onOJ-d-lAOMDVO HCO LAIAOVO 

vo^t IAH-4-OJ oj IALACO ojon-*onOJononiAon 



IA cvi LA c o\ o ON on on 



HonOLAOCVI-d-CVj 

on on on on on-* on on on 

rHrHii Jrn 



VO JLAO\ VO 

cvj -3-oncvj j- 

r~l p-jp-J, J ,J 



_4- i>- vo IA 
C>-VD on c^- 



cvj on^t on H 
vo^j- ITVJ* vo 

LA-* on-=f-co cvi oo co 
t3-cx)MDaD vo 

HOO LAC-IAIAO t-ODVO LAO-d--=f ONOJ l 
LA on UV^-CO IACO t-^-^- H CO VD LA\O !>- 



H j o -4- ON on o\ 

-^- v> oo t>vo on 



H cvj on-=h LT\V^> t ao a\o H cvj on.4- IA^ o-co ONO 

HHHHHr-iHHHHCV! 



88 



H O O H O 
O O O 




Oj OJ CVJ OJ OJ 



given on pa 



loyed 



ization of this table and the symbol 



13 



ept 12 



q * 

CD tJ 

IS 

P <D 
.CQ -P 



380 



APPENDIX A 





II 



3 



*' 

5 ' 

ll 



D O 
-P 



CQ ft 



S-S 



! 

H 



O ITS ojco on 
on H LTN 

H 




ichool or 
[Response 



H CQ CO 

Sag 
O.HO 

-P h CO 

co O h 

fc 1-3 
H g h 

*h 5 ft 

^00 
O -P -P 

r^J fj CD 

H oJ 



n 
B 



^tcviirsHoocvjojcvjo 
ro m ro en m ro on co on 
HHHHr-IHHHH 



UOTSTATd 



dnojs 



\ C\J CO LTVOD 

CVJCUOJCVJ 

j,_| r _| r q r -I 



O\ O\ O O 
OJOJmoo 
HHHH 



O- H O\ O 

00 Hr-im 

H HHH 

tn ONOOON 

c\j on on CM 

H HHH 



C\J 
H 



- 

t>-ir\c H 

HonOJ 



on 

H 

OJ 



o\ on 

H 



!AH 

oo CM 

<MH 



vjD^too mmvo CM cr\ !>-<M c .* 

VD^N-sF OvCVf-* H H iV^f ON 
H HHonHVOHH H 



OJOO 

CU 0\ 
H 



HOOCOLTN 

t-lf\0 IPv 
HH 



ononOiTNonoooJOO 
mon monco vo CM vo o- 

HHHHHHHH'*-' 



ir\ 
GJ 

H 



H 



- 

0^ Oj gH|0 



88S888 



HCM 



h-oooonaicMonivoi 

HCMHOH^J-CMHO 



ir\|H ( on,OJ on lAon.oj 



s 



oo o 

HOJ 



vo on oj vo 

O OOJ O 



OJ Ch H ^ OJ MD VO -=h 00 I>- O 
OJOJHHOHOOHOHon 



H VO VO CO VD t>- OJ 
OJOOOOOH 



O OJ O CO O\t- ON H IA 



ITM>- OJ on-cf on OJ CM 
ononononononononon 

H H H H H H H H 



H on o H 

-4-onon -4- 

HHH H 



SO H ~d- 
co co co 



t>-00 

ON o- 



^ ON 
ONMD 

" C co co ONOO ON ON co ONOO t- 



N^h-d- oj j- onco 
ONONONON 



NONCO 



S-oo" co oo ON ON ON ON CD oo t~S tr~co ON ON ON ON ONOO 



*===:= srrrrr 



>~CO ONO H CM ro^t lAVO t>-CO ON O 
HHHHHHHHHHOJ 



ON t 
ON 



on oj o 

000 




ON ON 

> oncost ao 
t>* ON co ON ON 



<! CQ 



H OJ m^t 1A 
OJ OJ OJ OJ OJ 



O CQ 

g 

5 2 

&0 -H 

cd 

CQ fl 



.oye 
lest 



CO CD 
CQ -P 

fl 



fre 



a 


O 



ft 

o 5 



anati 
these 



CO 

CD fl 
H CD tJ 

*I5 



APPENDIX A 



381 



<D 
-P 



CQ 

! 



O 

H cd 
EH 



N ' CQ 

CQ 

O 

O 

cd p 




CQ O 

14 Tl O 

O &0 

rC[ O 

o So 

cd ^ 

cd 

43 & 

CQ *N 



O P 

Ps CQ 

O 





eg 




ONOJCO mco ON o\o o t-- HONO H 




H OJ 





3j s 




Hojcuojojojojonon on HHm cu 

HHHHHHHHH H HHH H 




OJ OJ 


k 












H 












cd CQ 


cd 




-4-ojinHonoj-ojojo in ONOOON H 




4" VO 





IS ! 




on on on on on on on on on ou on on oj -=r 

HHHHHHHHH H HHH H 




H H 


O 












H rd 












P fl 


}2; 




ojON-4-vo on on^t inn 






CQ O 


|V[ 




i^int-H H ONon co oj 






ft 


?H 




OJHonOJ CU H OJH 






3 CQ 












Of 
H 


f 




vo^tcoononvocuoNi>-cvjt>-Jt cucoHoocom 
vo ON-4* ONOj-^HHocJd-ON oj ON c m o invo 

H HHOOHVOHH H H H H 


8 


'ON t*"* o 

ON ON -4* 


H 








H 




<! 


| 


h- 


voonou in o inint>-ononooco mt^-ojo-o OCMOJ 
voco h-oj movo toco inONVo vo invo cu-4- on.4- 

^tOJCU-4"HH^ H CUonHHHOJH 




o in oj co on 
c on on H in 

H H 






j> 




CQ 






o g 


CC 


i o i i in 





f 


M 


< S 




VO i i CO 


S 


i 


fl 






I H II ' 






H 




^ 


N_^ 


rrj 




r^ CQ 

JS o 


1 


D 


O O O O I 

o m o vo t 


pq 


s 


ft 


s J 


P 


H H H H I 


^ 




CQ CQ 




tr 


**-* ' **~s -^^ 






P 

fl rt 




p s 


% 


So o o o oo oo 
ooo o oo oo 


1 




TJ 













5 s 




g 


8O O HjoiO OOOOOOCU OOOOOOO 
00 OOOOOOOO OOOOOOO 


O 


888 


CQ 


^ 






o 






si 





n|j*O O HJ4-O O HJjtO HOOOOOOOOOOO 

oo oo ooooooooooooo 


g 


H H 
O O O O O 


S 1 


ji 


I 


moj.ON.4- on.oj cujONin in H invo ON 

HHlOJOJOJlCUCUOJon CU HHOJ H 
HHHHHHHIHH H HHH H 


1 


H(H 

H H 


H ^ 












jjd pC) 


a 


j_. 




E 




ro 


8 


p 


OU H HCO VO VO ONCO.OJ in -4" H CO CO 


pa 


on c*- 


QJ 

cc 


1; 





on on onoJOJOJOjHloj oj on on OJ on 


PH 


in on 

H H 


CQ p 




{n 




O 




P O 
fl ft 


P Ja; 


** 


4* o o in in co oj o H 


S 




CO 


fl PR 


01 


4" -4" m-4" OJ -4* -4" VO OJ 






n3 









CQ 




3 pq 




< 


o D c o in in in cu -4- o -4- o oincocooon^t 




^t in in 


P 





CC 


in -4- -4- tn -4- on on -4- on in in oj t in on in m m m 


<q 




CQ 


h 


E 


r*- HCV n\ H H-t-nsrno H n\ 


s 






*S 


s 


in in in in-4- on^f- on-4- inincu invo vo^d- invo in4- 


p 


on-4-^vovo 


a 1 


3 


p 


onoo-i>-ojojonot>- H H-4-on on 
ojonojojononononoj -4- Hcuon cu 

HHHHHHHHH H HHH H 


g 


ovo 

m oj 

H H 


H 

rM <! 

JgJ CD 
CQ 
CQ 3 


ife 


INCLU 


t>-j^- ON-4- co vo -4- co in in covoo vo 
on on on on on on on on on cu -4- on on -4- 

HHHHHHHHH H HHH H 


I 


CO VO 


P 




cc 








fl ft 
CQ 


"8 


fc 


vooom in cuco OON 
invo m in i>- in in -si- t- 







r d 


CQ 





1 


oononoininincovoovoo omojojo tvo 
in m m m invo vo mvo in-4- co on-4- vo _4- m^t -4* 


o 




VQ m in 
.4- vo in 




SI 




on co ON co mHONVocooJONVOHOJOVoonONt>-H 

^h -4" -4* -4- mvo invo in^i-^ t>-^4- on-4" m-4* on^4- in 




H ONVO invo 
vo in in on on 








CO k -H 




CQ h -H 


UOfSTATQ 




p C30 S-J CQ 






^ H < pq 


ATllfl 




8 H g s m H < o a i i ft 8 S I S g S 1 ^ 




g<:oa|l 


duo^ 




H oj on-4- invo tco ONO H oj on-4- mvo t*-oo ONO 

HHHHHHHHHHOJ 




H oj on-4- in 

OJ OJ OJ OJ CU 



in 
in 

K\ CQ 



O ra 



n3 -H 

!>S CQ 

O 

H 3 

ft Q* 



CQ P 

H 

O fl 





I 






cd OJ 

ESI H 



CQ 

! 



382 



APPiNDSX A 



I 



CO 



0) 
H 



CQ 



-P O 



CD CD 



u 
s 



n 
R 



'o| 



R 



ud 
Re 



HOI SI AfCC 



CFN OJ 00 IAQO ON ON O O t 

Hojojojojojojonon on 

HHHHHHHH H 



,4-OJlAHonOJOJOJO tA 

on on on -on on on on on on OJ 

HHHHHHHHH H 



HO"\O 

HHOO 

HHH 



CT\ on 

ononc 

HH 



oj 

H 



OJ ON-4- VO on 
C*-tAt~H H 
OJ H on OJ OJ 



on-=f IAH 

ONon CO <M 

H OJH 



H H on H VO H H 

> on oj ir\o lAiAtonononcoontojt o O cxi OJ 
) CO t~ OJ LA O VO CO LA 0\VD VO 1AVD CVJ J- fO 
- OJ CVJ -=b H H J- H 



- 
OJonHHHOJH 



CO' 
H 



lOi' 
ON I 



oo 



LA 

I H 



H 



' ' O 

1 ' OJ 



\c\\r\ ' 



88 H ' w 8g8 

OHCMO OJ H OHlOHlCMg O O O OJ 



O O\OJ HVD OJCO-* 

onojononononojpn 
oocooo IACO t>-cocovo 

333333333 



O\ VOOMD OJ 

on Hcucv*) c\i 



IA j- vo H on 

2i 333 3 



coco-4-t>- 
r-ooco i> 



on 

oo 



coco o o\ 



O-=h O CO CO tT-OJ H O VD OCO 

co t-covo t t>cococo t-t-co 



IA OJ t- OJ H -4- IA 

t-t-coco covoco 



t>-l>-I>-ooao O-cOOO 



O ON H H 0\ 
OJ H 



|O\-=i- 

HlO H on HI 
HHHHIHlHH 



H 
H 



3^ S SiSi 

o o oj cxi mco o\o j 

CVJOJOJonOJOJHHOJ 



o OJ 

H 



OJm 



IAOO onoo CT\VO ir\ 

OJOJHHHonH 



H CM on J- ir\vo i>-co o\ o H oj m^t IAMD t-co OA o 

HHHHHHHHHHOJ 



o IA c\i c 
t*- on on 



C7\ 



88 



O 

o 



CXI CXI 
H H 



LA ON 
on 
H H 



. IA 

IT- CO 00 



on OJ H CO LA 

CO COCO t^ 



VO.OJ 

S'S 1 



t-CO ON 
H H H C 



cxi on-=f LA 

OJ OJ OJ OJ 



IA 
LA 



a 

g 



0) -P 

c? S 



CD CD 

p 

CQ C*- 

i 
$ s 

3 

H 

03 CX? 

9 & 

C30 CD 



^ -P 



,<f 



CD 

JH 



APPENDIX A 



383 



TJ 

$ 

O 



A 

I 



CQ 
H 

5 

^ 

I 

5 

*8 I 

p. 

o S 1 

g jg 

EH o 
P 

^ 



CQ 

S 
i 



"- NlA 

CM 00 



^ ^J,' T:JN ^ 

O 43 o 4J 

fa H H 

i lei 

H (D -H 
P k qn 

" S 3 S 

f4 3 
O S 03 

o h o oa 

H ,Q ,0 <D 

s s *<^ 

& 10 < W 

3 

ri -P 



(D 
cd m 






PQ 



5 



4- OJ IA H 00 CXI CM CVI O 



lAOO ON ON 

C\jc\ic\j[c\J 

HHHH 



ON ON O O 

oom 

HH 



H O\O 

HHoo 

HHH 



ONOO ON H 
oo oo CM -4- 
HHH H 



CM ON-4-VQ OO 00-4- IT\H 
D-tf\f>-H H ONOO OD CM 
CMHOOCM CM H CMH 

^O .4- 00 oo OO VQ CVJ ONtCMt 
O ON.4- ON CM ,4" H H O C-4- 
H H H OOHVO H H H 

) OOCM LTNO LfNlAtoOOOOOCO oOt>-C\J t-O O CM CM 

>co I>-CM ir\o vo t>-ooD IAONV0VO ifNV0 cvj 

h CM CM J- H H-4- H -. . . 



-4- CM 00 H O O CO tf\ 

ON CM ON t- IfN O ITNMD 

H H H 



MD VD irsvo CM jt- oo^t 

CM H 



CM oo H H H 




O H 000 

00-4- -4- oo 



-4- H C CO t*--4- ON 
~~ ' h CM OO oo CM 



.4" H ON H OOCO UT\ ON-4; O LT\VO 
oo oo oo oo^t OO-4- 00-4- -4- ^t -4- 

H CM CO VO H 1A-4- OO ON ITNO O-VO t H C O ON CVJ O 
14- oo oo 00-4- -4- -4- -4- oo OO.4- oo^t-4" oo^i"^*- OO-4- OO 



LTNVO 
H 

H 



- - . 

-4- OO oo OO.4- OO oo oo 
HHHHHHH 



ON CM 
oo oo 



53 



VO 00 OOVOJtfNl 
H CM CM H H C 
H H H HIHI 



CM 
OO 
H 



H t 00 CM t-t*- OM>- lAONLTNCO -OOb-H-4-J00 
4- ITN^t IA-4- ^4- 00-4- J- -4- -4- -4- -4- oo lf\-4- -4- -4- OO-4" 



b-OO ON O H CM 00^4- lA^O t>- 
HHHHHHHH 



cO ON O 
HHCM 



O ITNCMCO OO 

ooooH ir\ 



O CO CM 

,4- 4- oo 



-4; t 
lA 



ON ino 
oo\o 



CM CVJ CM CM CM 



IfN 

IfN -P 

IA OQ 

<D 



13 -H 

(D -P 

P*> CQ 

O O 



-P ^^ 

to t 



1 3 



CH +3 
a> fl 

H O 



384 



APPENDIX A 



fP cT v m 



pq 



Of _CD 

H 



*8 
Ili 

CQ CQ 
P 

J "l' 

3i 




1 



3 

cvl 



_ O LA OJ 00 CO 
CC f co on H IA 
H H 



OOO'O' 



O 



IA 411 

Sill 



1 : 



8 



H-^-H-4-cjajONcn 



VOOJ 



Ch 
o\ 



en VO c 



I*-.* H O\ON CO O\t*-OJ-CO H - 0> ONOO (AOO c 

co ON o^ b~ 0*\ 0\ ON CO 0^ t 1 " " ~~ 

O O VO IA 



o\co 



co 

1A 



[>- t>- IA O LA 

ON c^n E^~ IA LA 



t-o 

r-i H H H ! 
1>- LAOO H ( 



l 

HI 



o 

3 



lACOOl 

t>-q mi 



-=t, 

* 



O'OH O OO OJH 

rr\MD ON HnI(Mt>-H ro O VO OJ ON en H 
HOOOJ OOHHOJOJH O H 

LTNCOVO CVJ ONOJ ONVO \O 



HjCMOJ )A OJ OJ 
OjOHH 



O\ O H OJ rn^f LA\D D-<X) O\ O 
H-HHHHHHHHHOJ 



O\ tO 

- 



H t- OO IA IA 
CX) CO 1AITN 



<! CQ 



H OJ ro-d- LA 
OJ OJ OJ OJ OJ 



O CQ 




& & 



O fl 

II 

CQ 



j3 S 

vj H 



H 



I 



* I 

a a 

o 

H 

-P CQ 



^ 

q^J 

P 6 

P 



APPENDIX A 



385 



II 



CD CO (D 
!> -P |> 



! PQ o 




a 



II 



-P O 
A A 
CD CQ 



H O gg H,H 

H W H VO H|OH|W o Hp* O O -d- H| on 

13: OO O OOO O 



HICUH gHC.0 



\CVJ) tf\Cp O\O\O O t- HO\O H 

OJCMCVjOjojCMonon on HHon KJ 

HHHHHHHr-J H HHH H 



-4-CMlAHonCMCMCMO 

ononononononononon 

HHHHHHHHH 



B 



o\ on ON 
on on CM 

HHH 



CM cr\-4-vo 

[ lAh-H 
CM H on CM 



on -4- 
ON on 



t< 



XQ.4-CQ on onVO CM O\i>-CM t- 
O\xi- ON CM J- H 



H O 
HHonHVOHH 



j* CMCOHOO 
0\ CM O\ [>- Lf\ O 
HHH 



> lAirNC-onononc 
toov.of-oa 

H H -4- H 



ont^CMto o CMCM 
\\> vo ir\vo cvj^t on^i- 

CMonHHHCMH 



ROCMOITN 
on onoo ON 



O 

H 

H 



OOO 

o\co ir\ 



OOQOOI 

a\ j- ir\-^ i 



O 

co J- o H 



Ot-O 

- o 



O OOO 



CM g g OJHCUO OH, W g 



-=h o t-,cvi o cvj-=t- D^H. 

M i!? ^^ 50 ^9 cr ?-= 1 " -*^ 
HHHHHHHH 



CM,C\l 

j-lin 

HH 



^ 

jOCO-*.H' r ?J- CM H M3 

'Ra3. f 333 s 



J- IA on 00 

j^-mcM <r> 

HHH H 



HCMon O HCO O t- 
CMmCM H HO HH 

HO\CMoncOHOlAOC-lACM 
- HoniAlAHOJH-5l--^HCM 

IA H D O\ H on ON IA^1 ONonO OMDOO HCOVOVO IA 
"HHCMonOOOHCMCMCMonOHCMOonOO 



on co -=t- oo o\ CM ao 

jf H -=J- cr> IA H O 



OMA O VD t 



IT-VDVO lAon 
5^^ 1 5 OC M 

HHHHH 



CM on IAO 

MD 1AIAH 
9 "3 H H 



vo 

J- 



VO O t^J- 

CM^t IALA 



onoonococucMtsi 
on IA if\ on -CM l>-co H 



CM^t V.O CM CM b-IAiAon-d-J* UT\ 

ON ON cnvo t-co o H CM ON CM IA\O ON on H IAVO t- on 

CM H \O CM CM jf -* lAoniAoniAonH-d" t-jj- CM iAVO 



VDVDCplAO-HHV 



IA 

a 
& 

H 



fO 

3 



t- ON on 

3^3 



cocMcoco 

HonOCM 



CJ\ 

CM 



on CM VD CJNCMVO o iAi>-t>-h-^t -=t- CM-4- CM onvn H 
HonoHCMHononcMOoncM CMonoonHHH 

VDOOlAHOH-d-CMCM IAJ* on LA ONCO CO 00 CO CM 

' >- CM -d- -d- IA iA.3- IA H -=t- CM on t>- on o 3- on on on 



H CM on-=t" IAVO t CO O\O H CM CO-4- IAVO t-CO ON 
HHHHHHHHHHC 



I! 

;8 



0\jfr-sf CM CVJ 

o oj on IA on 



H O 

as 



SI 



-=f ono CM on 



Of 
Pi 



}> 

a 



. I 



} I 

JH o 



386 



APPENDIX A 



<D 
H 
-p 
H 

5 

-P 

S 

M 
5 
g~ 

3^5 

O 
M O 

CD 

S3 

O 

*! 

A& 



JH 

P O 
cd <H 



CU ft O 
CU CQ ^j 

H O O 



d 

3' 



week 
ical 



CD 

1 



a a 

5 



II 



-4- LAIA 
H H CM 



< pq 



! 



P 
CO 



15 
^{ ^ 



o o 

0) 

I 



S3 <u 

CQ 
CQ fl 

s s 

rt P 

CQ 



a^ 

<D F 



:g 



fl 

8 5 



HOT:ST:AT:Q 



HlCVHlCUgg 



IACO ON ON O O 

33 



ON C\| 00 IACO ON ON 

3333323 



CU tAH rnCU CU CVJ O 



H ON O H 

333 si 



ON m ON H 



f^LAtH H O\cn 
cu H m cu cvj H 

>.4-co mmvo cu CM>-CU . 
) OY4- 0\cvj-d- H H Q C-.* 

i H H cnHVO H H 

VOmCV IAO lAlAton on onCO cnt-CUb-OOCUCU 

k " CU IAO VO CO 00 LAONVDVD 1AVO CU -sf rnJ- 

' H CU rn H H H CU H 



jd- CJ 00 H O O CO IA 
ON CV] ON t IT\ O lAVO 



O 1 O\ 1 

f a 1 



lH 



HH<M 



HIM 



g H| W H 



> H O CU 

> o o o 

- H O H O CU CVJ on 

> o o o o o o o 



8 8 8 



CM irvvo 
cvj 0( q 

H H H 



ON 
lcn m 

IH H 



ONONO 
o cu uM 



t-\O 00 MD CO MD 

s's a tfns 



a 



CO IT\ VOQO 

Hen H-d- 



V> 00 VO H 

HCVJOOH 



^HiAONcvjojirvHvofnoofnONoo ONVO cu ON IA ON 
CM en <n CM jt tA-=h cvjojHcvjiACMcncvjcjcvjmcvjcvj 



H H ONO o 



O CO CU ON -3- 

on O cu oi cu 

CU -sKONON LA 

3 3'33 S 



81 



LA 00 

-4- on 



CUC-CUHCUOJCU-d-Hb-rncn 

mJ- J- j*JtJt onj--* cnrn 

l>-t-ONHOJ-onV)O-d-CUC 

mm cn^-^^-jd- J-- 1 - j 



ONQO 
" cu m 



LA ,0 

:j rj' 

r-{ H 



faa 



H H m j* 
3Sc\f m 
HHH H 



- C- OH 
CVI IAH 



H -sf H oj o vp m cu ONCQ o m 
mm^i- aj H H cnaj miAH 



o- c- cu vo ONCO 
j mil- IA H IA en 



ONCUVO ocqj- H-- 

mrncuooHO HmmiAcnHrncvj rnrnmcumm 



mvoao OHHCUOOVDOOOJ 



H CM cnj- LAVO ivoo ON o H CM cn_ 

HHHHHHHHH 



IAVO b-co ONO 

CU 



O lACVJOO rn 



8 



H ONt-QQ VO 
--* tAOJ 



goat 



HMJOO 

-* on m 



ooao 

33 



H CU m-4" IA 
J CVJ CU OJ 



IA o 

LA 43 



I. 

& o 1 



5 8 



8 cJ 

N H 



<D a 

ri Q) 



APPENDIX A 



387 



I 



352 

O 03 O 

<n o 2 



*s 

g "M 



O T3 
*Cv7 CD O 

& ^ 
s ^ 

H Q 

ts a 



1 



S2 

Q OJ 

^J o 



I'd 

15 -a 



5 <d S, 

"5 I 

JR 

CD -P fl 

CD CO 

^8 
-p 

S 
P< 

i 1 

43 aj 

CD 

tt) !* *-* 

q o 

H HP ' 

IE 



o CQ o 

} 

5S 



4^ 
CD 



pq 



1 



5 



dnoiQ 



I 



oocjiHt^cocpmrH rovovo ojv> vo r-v> en HMD,* 

OHrHOOOOOOOOHOOOOOHOO 



SS'S'g g ?I8 



<M IT\ H 00 OJ CJ CM O 
en en 0-) en ro co rn m 

r-\ e~i r~i r- 1 rH rj rH cH 



tTN O\cnO\ H 
OJ ro en OJ ^- 

r- i rH iH rn r-J 



- rn cn_=t 

t-lAC*-H H CJNcn 

ojHcnoj cvj H 



H cncvj Cvj 
CO OJ 

-rHH^ " 

rncvj IAO ifMPv t-- m, en 



V5 CV| O> I>- CU C ~=h CVJ 00 H O O CO IT\ 

^ H HO t^jd- Os cu o\ t- tr\ o I?\AO 

HMD H *-4 H H H H 



H H 
C-Q O OJ < 



ON 

3 



o cnvo t- ovo 



cnO-=J-ir\cviojOH 



CO 



a 



-=i" m ir\ m 



NCO o 

J9 3 



I oo no oo oo cn cn oj 

H CO O t if\ t-^4- OJ ONCO VD V> ( 
mOjrnojAjcnmmcnojcncvjc 



o cu t-vo o co cvi 
_=t-mco <n^4- OJ en 



1 



a^a s 

CO OJ- rH 

33S ^ 



) t O H-d-VO ON CO IT\ 

oo-4--=l- men cn envo 

i>-ojrnir\ir\ovocovoHrH ir\t 
1 ir\^t ir\ en-* en en m IA tr\vo u 



enH O^J; H tTvO 
^4-^1-^^ U-N-4-.4- 

.^^^^^^ 



03 
P 



r crccrcEress 




I 





a 



a 
la 

S) * 
o 8 

CD "* 



388 



APPENDIX A 



P 

s 

p 



f ft OJ 

i -P 

I td 8* 

'* s 






<D 
CO CQ 

q 

s a 

*H CO 

>r Q2 



11 1 

f-p S 

s 

I 9 J3 



ff^* 

. ! PQ o 



as^ 



$ 



J? 



B 

p 



&fe 



4! 



^^^tr'DCy.ir'Q o\ O Oj g g\ 

HI H HI r-i 

WVfiD^va^oOcQco on t^fcrQ 



* -4- HVO CVJ CVI CO t*- COO 

Hcyonooon cvi on cvj cvj cvj 



jONO H 

ss 21 



if CM IT\ i-( CO C\J CVI CM O 



H 



_ on on-=f IA H 
i>-ir\t-H H o\ on oo cvj 
cviHoncvi cvj H cviH 



^0^-00 rnoOVO CVI 
'sD 0\^; ON CU ^; H 
Hi HHcnHVO 



C7\ t-CM t^ 
^ C*-.* 
HH H 



OJ O\ t*- IA O IAVO 



roao cnt>-c\j c-^o o 
ir\ o\vo vo IT\VD a} ^ r. 

QJfnHHHCMHl 



_ on-d- H on IA on o ur\ 

ij^iSi^SiSllIJSJ 00 

iTvonco IACO vo b cvj t 



on cvj o cvj 



OJ IA 

3 9 



CO 



s 



o o q LT\CJ\ 



ir\jt v ;j 

33 | 



on O H 1T\ 



ir\ H,CT\O on 

3 ^3 ^ 



QSPlDCvj q\ HOJ oovg 



oncvjcvicn on 



^|--*I>-HlCVI 

333 



o\ o\ t b- on cvi 

oj OJ on o cvi on 

HlOJt>-HQ-3- ^t VO,.=i-CT\ 

333333 1 3 3"3a 



OOOM3 00 

-- on on 




J on-sf IAVO tCQ ON O H CVI on^t IAVO C--CO O\O 
HHHHi-iHHHi-iHCVJ 



o\o 

33 






O ITNCVJCO 

i>- on onH tr 



^ 
^i^S!^ 



C VT O\OVO 
CVJ O CVI 



^-oco 

A LfMA 

H VO CVJ on CM 

tr\ t>-vo on 



cvj on-=t- ir\ 

CVI OJ CVJ OJ 



UJ 

LA O 
LA +3 





5* 



, 





tin 43 
H 
S ^ 

^< J 
* $ 



APPENDIX A 



389 



<D 
H 
P 



Q 



Iti 

O 

;! 



81 

5 



g 



r-i 
O 
H 






CO 



I O 

H 



Questionnaire 
espondents 



i* 



4-OJtAHooOJOJOJO 

333333333 



O\ ON O O ^ r-l 

! 3 213 a 3 3 



O\O 

33 



IA O\ PO O\ H 



O\-sj-\D on ro-3- ir\H 
C^-IAC^H r-l Chm CO CVI 
OJ H onOJ OJ H OJ H 



-d-co mrovo CM ost-cy t 
ON-* o\ cu -* H H o i-* 

HHcnHVDHH H 

oooj ir\o ir\ if\t m co mco mt>-oj t 



cvi oo H o o CD ir\ 
cvj cr\ c IA o TrSvo 

H H H 



O JAOJOO m 
t>-on on H 




a 

iA 2 

IA -P 

K*N CQ 

. c 



O CQ 
flj 

* 

{> 0) 

tt 5 



* 5 

P -e 



f I 






H H 

5 



OJ OJ 

N H 
H 

-P 



43 CQ 

H & 
O O 



390 



APPENDIX A 



> 

3 



*" 2 

I 



I 



CQ , 



CQ 
P 



I 

o 

~ 

P CQ 
O S3 



CQ 

O 

O ? 

o H 

> 

CQ -H 



S 29 

ft A A 



i s 

4P 



OJ 
OJ 



in' 



dnojf) 



O\ C 

H 



4- CVJ 1A H CO OJ OJ CVJ O 

333333333 



<ocncvjiAoif\ir\i>-oocn men not cucoocvjcvj 
sOcot>-CVJlf\OVO t^-OcO lAONVOVO 1TN\D OJ -=f- onj- 
Jt CVJ OJ.it H H^h H OJ m H H r-i CVJ H 



CO monvovo o cu t*-oj 
H OJ OJ on oj on m c\l ir\ 

HHHHi-JHHHH 



IACO ON ON O O 

ic\jaja|C\Jcoon 



t 

oo 



H O\ O 

HHCO 



oj 



lf\ ONfOON 

3 333 



OJ ON-st-vo 



a 

CVJ 



HH 



vo 

on 

H 



OJ HVOVO ONfOt H O 



o OJ on 

HHH 



CMTN C 

on oo OJ 

HHH 



vo H onn vo 
H oncvj H H 



&a 



s& 



ON 

a 



t>-ir\cq ON 



ONCVJ ITvOVOVO IfNOJ t- 
H^H H H*H H H H H 



CO CVI C^-OJ 

21 ^35 



H 
"3 



VQVOCQ IT\ 
=f OJ OOlA 



H 

on 



o o 

on-*- 



oonoooniAOVovoirNHvooj 
ooojojooononcvjojon on-4- in 

t*- \r\ irxop H vp on ONVO cvi oo c- ^ 



o H oj-=t-co ONCO 
on CM H oj cvi CM cvj 



ON on t-t- in ON o 
HojojojoJOJ 

HHHHHH 



H cu 
onon 

HH 



^OJ^tlTNHHHHlAH 

zs onononooononononon 
HHHHHHHHHH 



vo ONH OJ 
on oojon 

H HHH 



oj on on 

H HH 



H 

-it 



=(^ 
3 



on 1A if\ OJ ON^- t 

C-CO t-O-VQ t-t 



hoo^ononojo ONOJCOVOOO 

t-lfMf\t*-GOOOCOCO C- t-VO 



i-jS 



H oj on-d- IAVO o-co ONO 



O H OJ on^t LT\VO t-co ONO 
HHHHHHHHHHOJ 



J-VO 

53 



_ ONH 

& H H 
H H 



O ITNOJCQ on 

l>oo on H ir\ 



00 OO 

as 



H oo lf\ 
00 OJ 



vo oj H C*-ON 






J CVJ CVJ CVJ CVJ 



R . 



>* m 

H S. 



O <D 
CO 5 

3 n 

II 





H fc 



CQ t- 
H H 



3 

H 

P % 

03 CVJ 



fi 



O 

H 

-P CQ 

00 



APPENDIX A 



391 



-p 
3 

P4 



5 

-P 


I 



cH &1 
"^ 

2 S 



<N N 

8-3 



| 

CQ <D 

n h 



! l 

if- 

J *i 
.; 






iE 



jd-aiLfNHooojcxiajo 
co oo ro on ro oo oo on oo 
HHHHHHHHH 



OS CU CO ITiCd OS OS O O 



t H ONO H 

3 333 3 

lA ON on ON H 

3 333 -A 



on 

si 



H HHonHV 



oncvj ir\o 
cot^-oj ifNO 

CVJ <M -4- H H 



t-ooo ir\ oxvo 

- 



onc<Mt-- 
vo mvo 

CVI m H 



o ojoj 
^j- onj- 

t T-l CVJ H 



r-Jocvjonoocoi>-io\ 




ipvcu c-iA 
on on on on 
r-4 H H H r 



onj-vo on 

cvj ir\ cu cvj - 

r Hr 



vo 
cu 
H 



onvoco 

jt CM on 
Hr-fH 



voost-o 

OHHH 



H-4- 

OO 



oco 

OJH 



CM V ir\ f- t H O O-d- 



O 

3 

o\ vovo IA on 



VO H on O OO H-d" ONr-J 

oncvicvion on -=ton Hon 

onoooo IAONCVIVO o i 
on H H oj _-*- J 

00 O 00 O H O OJ IAVO IAJ* VO ON O\VO CO t VO < , 

cvf cvi on IA on^d- cvimcvicviojcvicvicvHonojcvion 



IACO HOOCO o\cy 
cvi H oj oncvi H^t 



VOOO 



t-oo i>-vo on ONCO o 



\D CVI -4- LA COCO Cvl O\H 

SHaHs^s^a 



CVI 

3 



CVI^t IA 

^33' 



ON 

3 



!8J8c5l$ 



cot- on o\ 

333 3 



O\ H on H 

333 3 



CVI 

on 



VQVO 

-4- on 



OO 

vo ir\ 



\ t o\ t- o\ c CM ONonVjo ir\cu m ^t cvi co vo CM ir\ 
- ir\vo-=i* CM^J" on on-4- t-ir\ir\ -=* vo vo on-d- if\ on 



c ~ az c c 



fp 



^ 3 



o ir\cvioo.on 
on on H tf\ 



CM oniAOJ CU 



m 



-, 

3' 



cvi J- 

33 



H Cvl on-d- IA 

OJ OJ CVI OJ CM 



rl 
P 
0) 



I 3 
I t 



-P 


1 



*^| 



392 



APPENDIX A 



CM 

CM 



< pq 



S 

O CD 



CQ O 
<D ft 
3 CQ 



fgj (0 

CD 

CQ fl 
P O 

(D CQ 



II 




UOTBTAia 



ON OJ OD 1ACO ON ON O O 
ojojojojajajmcn 

HHHHHHHH 



^tojinHmojcucuo 
on co on co m oo co no m 

HHHHHHHH 



rovo onco H c\i-=f- o on 
,=f OJ ononJ- ^ coj"-=f 

HHHHHHHHH 



a 



O\ O 
HCV-) 

HH 



cv 

i-f 



IA ON on ON 

2 333 



,-. . on 
r if\ D H H 
CU H onOJ OJ 



vo-=i-co ononvo cu 

VD o\ j- a\ oj ^ H H o 

HHonHVOHH 



lAH 

32 



^t OJOQHO 
o\ oj a\c-tr\ 

HH 



VOcnOJlf\O l 

vocot>cvjir\ovDc ooo I 

OJOJJ-HH-4- H 



cO 

a\vo VD ir\vo oj -4- on.4- 

OJOOHHHOJH 



oJHon 

HHH 



O J-^t fO 

oJOJajc\J 

HHHH 



H-4- 

HH 



OJ LPvt-O\ H 

on OHOO on 

H HHH H 



ON 
on 

H 



VOIAHVD j* 

ojojonH oj 



O ON 

HOJ 



t-O 

Hon 



ONVO oj OJ vo ON o on ON H oj on 

HHOJHOJHOJHHOJHon 



LTN ONOO cu i>- ONCO 

OJHOJOJOJHH 



ONOO 00 CO ON IACO OJCOJhonHCOOOOOJHOVOoO 
HHHHHOJOJHHHOJOJHOJHonOJOJHH 



H H-* O H H H t-Q 

333M3-53J 



on O ON ON 



a 



H 
J; 



>--* ONC-OONt-O HJ-C- 

on LA on-* H-*cu o oniAonoj 



HMD 00 CO CO CO IA 
on on H H-* on H 



ovo^t'voonvoco I>-HO H^OHVOVOO HVO H oj 
ononTnoj HOJOHOJ.3-ojoniAOjc\JHH on-* oj 



on H IACO ON.4- O O 



VOoniAHHonHCUO 

339333333 



ON 



VD CO H ON 

3 333 



00 

a 



SONCO CU t IAC 
VOt^C- ON ONC 



^ onMD H on o H on O ONVO on 

VD J-VO IAOOVO t^O\t^^-VO D* 



ON-=t cu oj OJ 

VO vS CO CO IAV 



6 



IAVO t^co ONO H oj on-* IAVO i>oo ONO 

HHHHHHHHHHOJ 



o LA cu oo on 
!>-on on H IA 

H H 



S 



EHCO I>-LA 
H CU OJ CU H 



H O 

oj on 

H H 



) OJ O 

--* on 



CUCO O 
IAIAC*- 



o on^t vo o 
coco co D-VO 



LA 

IA I 

KN CQ 

H 

s, a 



& 
"Sfl -H 



rfl -H 

P 

>> CQ 

3 



0) 

H ?- 



-P -v 

fl3 OJ 

a H 

S S 

M 

fc o 

O M 



S 1 S 

-P 



APPENDIX A 



393 



H OJ 



ONOJOD lf\CO ONO\O O 

Hojolojojcvjojoncn 

HHHHHHHHH 

4- oj ir\ H m 04 OJ oj o 

333333333 



C*- HQNO H 

ro H H on oj 

H HHH H 

in o\ono\ H 

9 333 5 



CU H on OJ OJ H 

JD .4- CO ononVO OJ O\b-CU 
s> CK-4- 0\ OJ -4- H H O 



- 
HcOHM3iHH 



Sol 

OJ H 

t^J- 
-4* O\ 



CO H O O CO LA 
J ON C- IT\O LHVD 



oncu IAO ifMfMxnromcr) cni>-oj 

cX) C OJ IfNOVD t>OCO lAO^VOVO tPv 
- OJ CV1 -=h H H -4- H CVlcn 



o cuoj 

-=t- fn-* 
cnHHHOJH 



00 VO ON 0\-5f- C 

H 2) S 9 2i 



JCO 

^ 



SOJoomooonm 
r-lHHrHHH 



a\ o\ oj 
oj 04 Jt 

H rH H 



mo 0\ 
cnoo en 
HH H 



cvj oo m o<oo 
vj ir\ 



o oj c 

ir\ifMf\m ir\ cvj i 

OMA o\ o o\ oj m t- o ^ 
ir\j^" m^t-d-^i- in 



3? O OJ O IT\ OJ t-VO ODOJIAOOJOCTXCVICO tOO t- 

4- iA-=f -4- -=f LA J- cn-3- lA-d- IA LAIA 



on LA m IAJ* m 



CMA IAOO t H O\CD 



j* O\0 



4- -4- vo H LA HI m en co 



IA vomo IA 

| 333 



.4-00 
3- m 



^)OOHO\ VO VO 
OJ oj m on H m 

H m OJ .4- -4- t^-j* IA ^ O j* O 0\CO OJ-4- ON-4-VO 

oj mm^t oj oj oj oj OJ-4--4- m HmojmHcncvi 

<DvoojHcncoHmooojojo\ojcoojovo IT\GO 
ojojfnonojHOJHcnrnononoJOJOJOJcnHOJOJ 



H OO ON VO CO IACQ O IA LA OJ O OJ CO 

3 333 SI 

4* OJ LAH OJ mOJ HVO LA O CO ON 0\ 

333333339 21 3321 3 



ONt*-aQ VOVO cnVO LAH VD O H OJ CO VOH VO -4- 

cvo vo LA t- 1- c 1>- h-vo IA b- co vo O-VD co vo t- 



jeccesrrrfisrr 



^-co ox 






3 



_ o LA oj GO on 
a f- m on H LA 
sj H H 

CQ 



- 

- on 

H 



t-OJ 

- LA LA 



t LA 

.4- OJ OJ 

H VO IA-4- 00 

OJ OJ OJ OJ 



OJ OJ CMOJ OJ 



LA 
LA I 

fA 01 



giv 
ire 



ymbols employed 
when the questi 



fre 



f this 
and 17 



organizati 
xcept 12, 



n3 
3 S 

*!* 

ff 



394 



APPENDIX A 



ng materials *jhich 
nts? 



Table 2k 



53 
"g, 
sg 

bO (D 

il 

go 

<M 

5 

OH 

i>> <O 

oj P4 
^ oJ 

N 

5 rt 

tQ 


m S 



g) tQ 



& en 



TJ CQ 
03 CQ 



,3 O S g g 

d +5 ^ ^| 

CD (D d ts 

8 IS 3SS 

^S ^o 



Questionna 
Besents 



ng 



Students 
No Bes 



Stude 
Bes 



ing 



dhoxg) 



^tojiAHmojojojo 

33333333 



vocncviir\oir\ir\D- 

VQCOC OJUTNOVOt^- 
Jf OJ CVJ -* H H -* 



***%% 



ON OJ CO IT\CO O\ CJ\ O O 

3333 



3333 



H Ts p H 

333 si 



ONCOON H 

333 3 



- on 

tirvt H H 

CVJHonCVJ OJ 



o\ on 

i-l 



LA H 
co CM 

CVI H 



vo -4- oo on o->vo cvj o\ t c\j t 

^O ON-d- O\OJ-4- r-i H O t~-3- 
H H H on H VO H H H 



CUCOHOO 

OJOM^IAOI 
H H H 



moo onco on t cvi c*- 

OOOLTN O\ 



VO VO 1 

OJo~)H 



o cviaj 

J -=t CO^h 
HOJH 



lf\ I i 

onso-) 



O 

i>- 



O OO O 

IA I^ON co 

H 8 ^^ r-J 



Q* B on o" "o" 
oS8\oo-)coo-i 



i a 

ir\ 



8 H|W 8 



00 



o0.cv)VO O O\O4-4- OJ en 

ion oncnonooonoooo 

rH rH rH rH rH s 



VD 

oo 



o 
on 



oir\ 
oj on 



cot- 
onoo 



ir\vo o VD o on H H oo IAV 
on -=1-^ -=t on ^t on on on on 



IAOAO OJ IAH C^ 

^-4- IACVJ mcvj 



on oj o o t---=}- cv 
on on on m on on on 



O t-00-4- on C on 00 O\^t ON 

^t-=i--it tr\onoj ononononon 



O\ O 



.4- OJ H 



OHOH^tHt HtT\ 

33333^333 



^tt^VO 
oncvjoj 



H 
on 



ONH IAOO 
cvion CVJH 



[>-,CO VOoniAONOt*-lA OO VOVO 

3^3333333 3 S3 



Honcr\oncovoco.Hoo irv 

3333333'33 a 1 



IT\ 

3 

ON 

3 



VOonOlA 
onon-4-on 



O\ 
on 



O^t 
LAon 



VOlA 
on-4- 



vo \o vo o vo oo vo vo co o\vo o 

on oj on on-4- cvj -4- -* on oj oj on 



cvi t on OJ 
IACVJ^I- on 



O IA H t>--4- 

vj oncvi.3- 



- -" U"\ O t-*-d- CVJ H IA O 

- onoj onononoj-*-^-3--=j- cvi onon 



H OJ on-3" IAVO 0-CO O\ O H OJ 
HHH 



^- tAVO t^- 
HHHH 



OO O\O 
HHCVJ 



33 



j-vo 
-=h on 



? 9J 



O LA oj co on 
t on on H IA 

r~a rH 



oc 
cvi 



on o 
vo cp^ 



H O 
000 



H O 
00 



o on oj 
on on 



33 



VO.jH 

13 



O OJ 

oj oncvj 



VO OJ 

LAon 



VO t-IA 

cvj on-=j- 



b~ o\^f- on t 
on onH 



H OJ on-d" LA 
CVJ OJ OJ OJ OJ 



on page 355 



giv 
ire 



T3 *H 

-p 

>> CQ 

O 

g, & 



the 



and 
fres 



le 
re 



of 



and 



I 3 

P v 

$ 3 

II 

s 

I ~ 

o o 



exp 



j/. 
A brie 

Students 



APPENDIX A 



395 



02 
H 
> 



fl 



H 
W CQ 



O 

o 



O *H 

H -H 

O 

W 

*r4 02 

t^a 



IA 
OJ 




H 




CQ 



t 

fl CQ 



-P 

J* O 
O 

W H 



IA 
OJ 



* 
CQ OO 
& 



o 



N O 
H 



Hi S 

O H 



CO -P -H 

OJ H fl 

ft P J_j 

O fr O 

<d 



fl 



"rjt ^ 

(D 



H 

cd CQ 






ft 

^ 



CQ CQ 
-P 



& 

n3 



PQ 



m 



ft-P 



tlOTSIAJQ 



ON OJCO IACO ONONO O 

-<3!2!3a333 



OJ IA H 00 OJ CM OJ O 
oo oo oo oo oo oo oo oo oo 
HHHHHHHHH 



J>- H ONO H 
oo H H oo OJ 
H H H H H 



IA ON 00 0\ 
C\J oo oo OJ 
H HHH 



VO 00 OO-4" tA H 

_ H H 0\ oo CO OJ 

OJ H 00 OJ OJ H ~' ' 



CO < 
OJ H 



H H OOHVQ H H 
lOJ-^H'HS" H 



OJmH 



- ro^ 
OJ 



o o o 

O\^t ITv 
'HH 



o 

lO 
lH 



ir\o o 

OJlAb- 
lHHH 



O O 

-=f-=f 



o 

1A 
H 



O 

H 



i i ir\ 

1OJ 
IIH 



o 

CX> 
H 



O 00 



ON 1 




oHjwroo 
OOO 



o 

O 



HOO 



OggOHHO 



O.lA^t CO V3 OMAVO O 
xt- oo co OJ oo oo oo oo-=J- 
H H H H H H H H H 



OJ 

3 

ON 

3 



CO.CO IA O\ 



OJ O\ OJ ^ 

-4- OJ -4- -=h 
HHH H 



OJ H-d-^t O CO OJ 
^t- H OOOJ OJ H-^- 

00 00 O 1AVO H CVI JtVO < 



ONVO 

OJ IA 



HLALAOVOlAOJ 



JVO 00 
J H O 



O O\ t^-GQ H VO CO 
OJ H O OJ 00 H OO 

CO-* -5|- CO O- 



IT 
OJ 



- t- tA OJ OO 
HOJOJOJOjOJCVjooro 



OO 

-4- 



t O O 
OOJoo 



HOJVOHONOOOIA 



COO>M3V 
1ACOVO 



O 

CO 



OJCO 

COLA 



OJ t O LA^j- ON CO VO J-C 

t^oo t-co3-vo LAcovbc 

VO *- ON LAIAO-* IACO C 
VO CO VO CO ^t CO VO I>-VD C 



O H 00 OJ 0\-=i- OJ 
COCO ONt^VQ CO VQ 



ON O C-VO VO OJ O MD O\VO 
t>- LAVD t^-CO O\CO VQ t- IA 



m 

S = 



OJ P0-=h IAVO t^-00 O\O H OJ OOJ- lAVO t^-CO 0\ O 
HHHHHHHHHHOJ 



13 



o IAOJCO oo 

C-OOPO H IA 



NPH 



^8888 




1= = 



<j CQ 



H OJ oo J- 
OJ OJ OJ OJ OJ 



IA 

IA I 



m 

a S 



H 
tlO 



o g 



CQ 

11 


H fc 



3 H 

3 



H 
P 03 






- 



396 



APPENDIX A 



! 





' 43 



H 





P -V 

H 

*~ ii 

CCJ *H 

Co s S 
oj 3 ft 

0^0 

) T3 

|9 



-p 

o5 

m o 

H 3 



CQ 





1 



g-H 



%h H' 

^ H 
-P 8. 



1 OJ O 

1-0 . o o 

> 5s <; m o 



H n3 

P C 
03 O 



5 



m p 
P O 
C ft-P 



Hk 8 888 ^^8 8 8 



ON OJ GO IAOD ON O\O O t 

s^masass a 



OJ tf\H onOJ OJ OJ O 



LA 



H 

21 



333 3 



on 

OJ OJ 



vo^" co on onvo OJ OM>-OJ o--^ oj co H o o oo IA 

VO CK-3- ON OJ Jt H H Q c^-^t O\ OJ ONC^IAO lAVO 
H H H on H VO r~t r-| H H H H 

NQ^cuiAQ !0^^-2?ir?fr>^.<r>.5rCV!.trP.Q cvjcv 



8 !. 



*Ot 



O 

LA 



H8SCOOl 
ISICOMD!I 



3 i 



8 88 

8HJJ^O O . 
O O ' 



888838! 

,OgH|W 



t>-cnojir\ir\cY-ojH 

333233221 



33 



3 



o\ o 

S 1 3 



ooojir\ooi>-oir\HHmooo\ ONCXD H o tr\-=f H oj 

OJOJOJHOOJOCVIOJHCUOJCVjiHHH H f OJ OJ OJ 



OOONHOJOjChOOJoo 

^^3 



33333 



O V)O\ro O 

^ S^!3 3 



OOHC30OJcnOOVDI>- ^t lAOjm 

333333333 3 333 



HOJOOJODOJOOOJ 

3333333^ 



VOt cnl 

3'3^l 



-=t-ob~on 



H ON-4- 
ooHOJ 



O 
00 



VD^t 
<r)H 



3% 

-3- O O\lAO\ 
vt r - - - ' 



, - 

OJ 00 ir\^t IfN^J- oororoH 



oj m oo o vo IA 

OJ-4-OJHOJOJ 



33-3 



! Stfi 



H oj ro-4- irxvo 



o H; OJ rn.j^- IAVO t>-oo O\O 
HHr-?HHHHHHHOJ 



33 

Jfr VO 



o\ t>-o 

O\^t- 



O IA OJCO o 
t^- on on H in 

rn i~ 



O O 
H OJ 

Ss 



S H|W 8 8 



33 



IAH LA 
' H OJ 

o c-vq q 

OJ H r-S OJ OJ 



H OJ 

33 



O OJ 

33 



ss 



H - 

oj on on oj o 



oj on^- LA 

J OJ OJ OJ OJ 



LA 
LA 



O 
H 3 

a* & 



s a 

H 

1 3^ 



H 

1 I 



APPENDIX A 



397 



d 



l Questionnaire 
Respondents 



ing 
e 



ents 
Resp 



ing 



Students M 
Response 



Students Making 
Response B 



udents Making 
Response A 






ATXIQ 



PS 



ITNC 

OJ 

HI 



ONON o o 

OJCUonon 

HHHH 



,4- CV| iTv H CO OJ CVJ OJ O 

ononononononononon 

H H H H H H H H H 



on 

H 



in 

cu 

H 



r-i ON o H 

HHro OJ 
HHH H 



ONonON 

on on oj 

H H H 



-4- 

H 



OJON-4-VQ on 

t*~lfM^H H 

oj H on oj oj 



on-4- 
ONon 



VQ .4- ao oo on\o oj ON t- oj r.4 

^D ON-4- ON OJ -4- H H O t jfr ON 



- - 

HHonHVQ 



oo H o o co ir\ 

J ONE*- lA O IfNVO 



on OJ mo IT\ SA Con on onco on tOJ C*-O O OJOJ 
^OaOt-OJIAOVQD-OCOin ONVO VO lAVO OJ -4- 

Jt 04 OJ _sr H H -d- H 



NVO VO l 

OJ en H 



- - 

H OJ H 



R 



OI 



OI8 

ON8I 



O 
CO 



g 



go 



Hk 8 8 S H!W 8 8888^88 
^8 888 8^*8 8 8 g 8 8 H!W 8 8 8 8 Hl(V 8 



OJ 



^J-VOVOCOOONO 



\OO ifN 



on OJ 



OJ 
r-?j 

OJ 





HOM>- 
^ H 

OJ 



OJ 
rj 



ITS 

H 



r-ioo OJO 



ON o\vo co t*- 



ir\ oj ir\ o OMTMA oj vo H t-ojoo ITN 

a a 



O 

ir\ 



=h O\00 

r\ en LT\ 
oj o\ en o oj H oj -4- vo vo o> on 

lAcnif\-4- l"--- 



o\t>- o vo t- 1>- o 

-4- IfN^J- VO 



OJlAHooonCHOVDtr- T\CJ\\O c 
4--q- 1A-4- lTX-4- ITS m J* " " - 



on on on on -4- 4- on on on 

^HHHHHHHH 



-i.oo.o 
.4" on on 



co 
on 

H 



OJ on b-^ 

on -4- on -4- 



lf\ 

on 



O> on 

on on 



-4- on ^ o oo -4- 
-4- ir\^t* on on on H 



on fr-vo a\a\co tr\t>-ovo on-4- 
on^d- on4- OJ on on^4- on-4- on on 

H -d- 1^-4- OJCXD OJcocr>^tOononHJ-co-4" I>- 
4--4- on-4" onononmojLfS on-4- -4- 1TN.4--4- on -4- 



H OJ on^J- if\vo C-oo ON o^ rj Oj on-4; invg 



ON O 



o lAOJoo on 
t>-onon H 

H H 



00 

lf\00 



onvo oj ITNVO 
OJ H OJ 



on o 

955 



-H OJ 

oj-4- on 
on on ir\vo 



i* 1 " 



5H ^^ 10 



I 



ir\ 

ir\ +a 

N"\ ca 

ft ^3 

g 

g * 

5 S 

60 -H 



loy 



5 

P 



1 



II 

10 8 

8 vi 

5 2 

3 IB 

a & 



<H 3 

I 3 

I a 
!l 



a 



H <D 
P QQ 

ii 



^ i 

H O 
h ^ 

<aj CQ 

*"Si 



398 



APPENDIX A 



v 














LA 

LA I 


CQ 
CD 
CQ 

& 


s 


II 




ONOJCO LAOD ONON O O >- H ON O H 

H OJ OJ OJ-OJ ojojonon on HHon oj 

HHHHHHHHH H HHH H 




OJ OJ 
H H 


K^V CQ 
H 
fl 
t5D -H 
03 J 




o 


Oj CQ 







-* CM IA H on oj oj oj o LA ONonON H 




j* VO 


ft r d 




11 


II 




ononononononononon oj ononoj ^d* 

HHHHHHHHH H HHH H 




-^- on 
H H 


CQ 

ctf 


o 


H td 












fl Is 




P fl 














5 


CQ O 
ft 

2 tn 


0* 




t lAt H H ON on 00 OJ 
OJHonOJ OJ H OJH 






> 

bO *rl 


I 


Q? 
H W 


i s 


$ 


^O^tCO onrnVjD OJ ONtOJ I>-^d" OJ CO H O O 00 LA 
^O ON.4- ON OJ ^ H H O t-^* ON CVJ OM>- IA O LAVD 
H HHmHVOHH H HHH 


8 


8^^3 


CQ fl 

^ 1 


s 


3 


, 


8 


VOCQ t^-CU LAOVO C^O SR^VOVO LAVQ OjS on-S 


> 


o LA oj oo on 
D on on H IA 


1 S 


tJ 







H 


^OJOJ-ii-HHJd" H OJonHHHOJH 


^H 


H H 


f>i CQ 


o 






> 




CO 




O CD 


So 






^ 


^--x S~* -^ 


w 


. s 


H 3 
ft O 1 










I f- 1 1 O 1 1 LA 1 








o 


no 


3 




1 .4- i i H I I O I 


J3C 


I LT\ 





"air 


fl 

H 


5 





1 H 11^ 1 1 r-^ 1 


P3 


I H 


CQ -P 


O 03 


M CQ 




Sc 








H 


: i 

Sj CQ 


ft 

CQ CQ 


II 




o o o i i on o o o 
IAOJ t"- i i -* on oo o 

H H H i i H H H H 





I I 
l i 
1 i 






CH tf 
tH 


-p 

fl PS 




"p s 


5 


OHM**, H| HH 00 


M 




CQ 



CD O 


<rj O 














g 
M 


JiS 


o a 


N 


88388888^8*8 88S8888 


D 


OHO 
OOO 


CQ 


fl 

o on 
?-i 


CQ 


fife 


8 


OnlcvHlwO O OnLtO O O O on O H|OJ O HfCM H H[CVJ O O 

o ooo oooooo o o oo 


o 
p 


O O O OJ OJ 

o o o o o 




-p -s 

CQ OJ 











rH 







* ~ 

M ;Q h 


% 


11 


1 


^ooojoo\ooj-4-on LA ONVD ON co 

HOJOJOJHOJOJOJOJ OJ OHH H 
HHHHHHHHH H HHH H 


1 


H H 


1 s 


O M O 


H 














J2 <g 43 . x <IH 


M rn 








H 




CQ C 


3 H G 


"co 


w 


^ 


H Jd- o vo ON r- LA ON^- o t-coco LA 


xj 


H OJ 


H H 


"3 J>5 bO ^ CD 
EH H 


52 
CQ 


n ^ 


o 




W 


-4- on 

H H 


5 


HO -P M 


P O 




g 




O 





SH CD 










3 O O 


m 


"S E 


CQ 


on.^}-_4-VQ H on O ONON 

oot-tr-f- t- c-oo t-vo 


^ 




O 

s 


$ 03 CQ 


T3 
3 W 


o g 




o IAVO t- o en m oj vo \o on o -4- vo vo vo H oo ON 


CQ 

25 


LA^O 00 


3 


3 O 


43 ^ 





M 


CO CO C^ t t*- t- !> [> 0- t CO VD l>- D'-VO CO GO 00 00 


2 


00 C-VQ 


H 


O 3 tiD-P 
>i S3 ^ 


CQ 


Ife 


| 


H o IAOJ on ON_tf- OJ*H HonONoncD ON^J- on j- o 
co co t b- t*-vo t>- coo oo ONVO 5- 1 tv IA tco t*-co 


1 


o on ON cvj oj 

00 D- LAD--VO 


HP -s 
Ct> OJ 
N H 


fe-P 8 8 






F> 




x^ 




H 


03 CQ ^J 
CQ 2 CQ 














"P 
ft 






tt 








T3 -p 
J -P O 
On C r* 


? 


^ 




OJ^MDloJloloboliAonl vol ^o H on! oj 
rnfrn-4-Lt LAiAphUJ-lLA LA H on LA on 


i 


VO ON 
2-3 


tjfl 

Jn o 


O H S 

f>s nd 


3 


<5 


I-S 




3 


r~l r~1 


O M 


rrj O 13 
H -P 


"a! "^ 

IS? 
CQ 


II 


1 


ooKololonloj nn-Htoo ojj -3- -=h oKol ol 
^lo LAptpt^t lAjonVO on -=h LAUh H 
HIH|HIHIH H HIH H! H H HIHI HI 




_, 


^ 


-*-- 




CQ fl 
















-P O 




CQ 








O 


v-P 33 


g& 


g g 


& 


b-VOVO^t ON t-O HH 

HOJOJOJ oj ojoj ojon 


s 




H ^> 


H CQ CQ 


TJ 





O 




D 




o 


<rf 3 ID & 
1^ *rt 


5 

CQ 


J-! 




SLA-4- on O b- t COJ- -d- b- O VD ^h ^- -=i- CJN OJ H 
HOJOJrOOJOJOJOJOJH-d- OJOJonHHHH 





lA^sh OJ 
H oj on 


i 


O 




Sfe 




ONO LAOO^HVOOVO ONONtr-H t-cvi HVD t VD o 
HOJOjojojonojonHHoonojojoj^tojHOJOj 




O t- HOD 00 

oj oj.4- oj on 


i 1 


bond 














H 
















fiS 


tioisTAtrr 




3 k <H 




CQ jLj -H 


' H 




VAW U Urf.* UVJ_ 




< | < cq 




<3 t < 


CQ 
<H -P 














<D pj 


QQ 












*H TJ 


OJ 


ATna 




0[x]SsCQW<p2SHBO^^CQS^ 




o B EH 

H <; CQ g s 


S4 TJ 














-P 














< CQ -P 


H 


Snoao 




H OJ on-4- LAMD I>-CO ONO H OJ on-4" LAVO O-OO ONO 
HHHHHHHHHHOJ 




H OJ on-=h LA 

OJ OJ OJ OJ OJ 


* 1 



APPENDIX A 



399 



P 

03 



o\ 

CVJ 



I 



9- 



I 



H 
H 



<J PQ 



air 



Questi 
Respond 



u 
N 



St 



ts 



Students Making 
Response B 



Students MaMng 
Response A 



CVJ 

t ITX^-H H 

CVJ H ro CVJ OJ 



Hcojoo-^ o bo jcv7|r>-,G"ir\ 

SMrjtfy dRNrJlcM ^ 

OlHlH H HIHh H H ^-" 



mco o\ 0% o o 

Hajcvicvicvjcvfcvfooro 

H H H H H H 



4-cvjJAHoocvjcxscvjo 
o^oorooo^ooonooooo 

HHHHHHHHH 



t- 
'n 

H 



04 
H 



H ONO 
H H en 

H H H 



H 
03 

H 



o\ oo o\ H 
ro oo OJ J^- 

HHH H 



no cn-=fr iA H 
ON PO CO CU 

H OJH 

cvj o\ t- cvi c*--* 

H H o i ^r o\ 





HHfnHsOHH 



jco 

j o\ 



c LT\ o 

H . J 



ooc 
\o\v 



- 

OJCOHHHCVJH 



ornO88iii 
8 e ir\aO B J ! i a 
B B H*-* s a ! a 8 

SOQOOISOB5 



O ss 
ir\ i t 



1 O 1 O 

-=fs O 
SHI H 



OHI^O 00 Hg 

8S H!N 888888- |w g 8888888 
8 g rt|OI 8 8 8 rt|J 8 8 8 8 88 8 rt|J 8 8 H|w 8 rt|tv 



H 

H 



\O [>- O 
VJ OJ H r-j 

HH H 



^t O\VO H CO ON O t 
GO 0-CO CO C^t t*-lf\ 

OCVJVOLPi 
l>O^-VO 

VO C- (A-=J" 
t 



-*^rV 
OO 



-* oo on CU 
HO CV/00 



1^ H CO VO IT\CX) H 
OHOCVIOOJH 



lAC^O O t>~O OA^t- CVJ 
HHHHHHOHHHCOH 

H CO CO 00 OOOD VO CVJ J=r 00*4" H H COVO H CXJ CO CVJ IT\ 
HOOOHOOHOHCXJCVJCVJOOHHOr-io 



fr-HO ON 




,ONlH lf\ 

- 



OJ CVJ OO-=f LOvL,ONlH 

cvl oo co cvj JT H- ookj- 

A H H H HIH HUH 



C , t~00 OOj t*- 

ir\ H H irsl 04 

H H H H| H 



CVJ 

si 






H CVJ CO-=$- IAVO tCO ON O 



^t IT\MD t>-00 OA 



o IACVJ co en 
t oo oo H 



OHO 

o o o 

O H CU 

o o o o 



H lACVJ Ch CVJ 
CVJ H OJ O O 



F-*. 



c-o ONOVO 

ITv^-VO 



H cvj cn^t- IA 

CVJ CV] CVJ CVJ CVJ 



tf\ 

IA -P 
" 



U 2 



t 

CQ 

I 



H 

T3 



II 



H (D 
-P DQ 



400 



APPENDIX A 



I W 



^s 



O 

on 



CQ O 

P 
or 
of 



5 



CQ 



PQ 






S ^ 



1 



lOTSTAia 



dnojf) 



O\ OJ 00 IAOO O\ Os O O 



, cviinHonojojojo 
ononononononononon 

HHHHHHHHH 



J O\ O 



in ON on ON H 

21 333 3 



t in t*- H H 
oj H on oj oj 

v> J-co on on 
vo ON.4- ON oj 



- 

O\ro 
H 



OM>- oj 



ir\ H 
CO OJ 
OJH 



H 
H 



oj ON t- ino invo 



-ocDiA 

H 



-oo ojcvi 
NVO VD ir\vo cvj -4- on^t 

OjmHHHOJH 



o H H vo c ON.OO 



I^I- 

on-* on rovo 

HIH H 



OH 
ITN r-f 
HH 



04 

s 



OJ 
H 



2 JO-5 

33;:! 

^SfcB 

HHH 



oo 
on 
H 



m oj oj 

0000 



o 





H 





^t ON H incvjH|c\]on-=l- 

oj ^t oo vo on o H|r\iON oj 
-OJOHOH OOJ 



H| W g 

^8& 



ONVO m 

OJHOI^roOJ 



o ^t vo 

O 



oj o o on oj o coo mo co 

OOHHonHO 



oj 

H 



H o on 

HHH 



in 

21 



OJ^d- O J- 
O H OJ O 



H|CVI 



O O 

o o 



o o 
o o 



^33d8&8&8 

_ jd- Hn|cyO I>-H|CVJH O 
H O H O in O O 



HOon 
OOH 



oj onvo oj o t^o 

OJ O H H OJ O O 

O H HCO H b-00 onVD O 
OHHHOOOHOO 



onH 

d ,'S 



IVQIOJONOJOJH 



CVJ 



on 



IA ON o ir\1r\Co' 04 vo n 

9 "} ^^ HONrnOJ-5 
H^- 'HHH 



t onoo "o" 

ONCVJOJ H 
HH H 



O ONOQ oj^iwon lAon on oj QNLTN 
HVOVDCO ooonOONONon 



Rg&R8i 
8^& 



8 1 i S 3 



^00 



Ng 



O iTNOJOO 

b- on on H i 






88 



P 



888 



?8 



on 

121 

, ^_ 

13 



21 21 



US 

in 
in -p 

KN CQ 
H 

S | 

& 3 



CQ 



I 





3 I 

-P 

CQ b- 



3 



S -S 

bD 



* | 

; i 

s s 

1 1 



H 





APPENDIX A 



40 1 



3VO 



O 

on 



Questionnaire 
Respondents 



ng 



Students 
No Resp 



Students M 
Response 



Students Making 
Response E 



UOTSJATO; 



Atnn 



4-OJLAHonOJOJOJO 
ononononononnnonon 

"HHHHHHHH 



H_=h 



H|oH|oH|ajO 
O 



lAC 
CVIC 

H 



O\ 0> O O 
OJCUooro 

HHHH 



ro 

H 



H Ch O 
H H on 
HHH 



H 
CXI 
H 



O on O\ H 

on on OJ J3- 

HHH H 



CM O\-=t VO on . 

tlAr-H H O\on 
OJHonOJ CVJ H 



lAH 
CO OJ 
CVJH 



CVJ O. t OJ D .4- 
ON CM .4- H H O ^-sr CTN 
HHonHVOHH H 



OJ CO H O O 00 IA 
OJ CiS h- IA O lA 
HHH 



I>-onronncoro 
COD-OJlAOVDI^OCOin ONVO VD 

OJ OJ - 



- H H -=f 



OJ ^ m-=f 
OJcnHHH'CVJH 



' O 
1 lf\ 



-=f 

H 



o 

H 

H 

> X 


' 


I 
I 


8 


I 
I 
i 


O 

on 


\O 


13 


3 


IA 

S? 




H 


O 
O 

H 





O 



O 

O O 



O O H O O OH4-O OJOOO HOOOHOO 

oooooo ooooo ooooooo 



O OnlnjO H O OJ cnO 
OO OOOOOO 



iO H HH!C\H]OH|W 
OOO 



lOOCVJCOoniAOlA 
1 on-4- on on cvj H O H 

'HHHHHHHH 

H o\ i vSWcTn 



H H c- oj 
HHH H 



no VOiA 



g^S * 83 

on^t ir\H^t H oj.4- HH|WH 

OOOOOOOHO O 

pnCOC\JOmOHCOOJOPOO 
OOOOOHOHOOOH 



cnHjwoo H cvj co 

O OHOO 



OJ _d- H H CO VO O i 
HOOHHOOJ F 



oj 



i cc 
H ir\ i i o i 

HH.IIH''HH 

oj c ir\ IAVO oomoj 
j-^t O^OA m on oo O 
H H H-^rH H H 



o 

.4- 

H 



m 
H 



co ITMA 

-~ H H i 



HmH co 
-4-00-4- IA 
HHH H 



HH|CM H|W HO 
O O O O 



mnJcuO 
O OO 



-H H OHJOJOO 
OOO O 



H H OJ OJ O m O 
OOOOOOO 



H OJ 



UNSO D-CO ^0 H CVJ 



OAg 



o IAOJCO on 

C-- m ro H LA 

H H 






H O 
O O 






OHO 
OOO 



LA on 
Olon 

OJ H 



888 

CO OJ ONCO O 
OOO PQ.4- 



CQ S 



H OJ on-=t IA 
OJ OJ OJ OJ OJ 



IT\ 

LA I 
ro\ CQ 

H 



employed is g 
he questionnai 



s 



the symbol 
hmen when t 



and 
fre 



is table 
17) were 



13 



ation of the organization of 
ese groups (except 12 ? 



to 

<^i +:> 
fl 
-H nd 

^!s 

.<! .CQ 4^> 



402 



APPENDIX A 



H fc 
O 
O tt 



* 
fl 



3 

ll 



55 

43 43 



O -H 
O H 
H O 



3 

& 



& 



S 



>>^ 

-P CQ 



'I 



SI 
3g,' 

I s ! 

t^^ c 



O IA 
..2^ 

^ S 

<8 H O 



m 
111 



O O 
P Q 



H 0?<5 

< m o 



d s 



5 
& 

I 



i 



lACO ON ON O O 



t 

s 



ONO H 

a 



-4-ojiAHonojojojo IA ON on ON 

on on on on on on on on on OJ ononoj 

HHHHHHHHH H HHH 



C\J ONJ-VO (O m-=t IAH 
ttTvt H H O\OO 00 OJ 
OJHPOCU OJ H OJH 



H H onH 

> onoj 



mcnoncp oot>-oj co o oj CM 

OCO mO\VOVO 1AVD OJ ^ fOJ* 
H OJcnHHHOJH 



OO VO"? 

ojcnir\ 



o" 

ur\ 



IJ-IM3II 
i H i H l ! 



3 88 88 



I OJ O Q O H Q O Q O Q O 
' O O O O O O O O O O O 

iSSSSSSSSS-^S! 



3 SI 

VO IA 



H ON 

3 a 



VO OJ OJ H VO VO H VO IA O VO 00 1A 

HGOr-frJHHOJHco OJ onooiA rH 

OJ OJ OJ -=j" co OJ on 



on tAOO J- lAVO lAVO lA OJ Q vo O\-=h OO H -d- IAVO IA H 
IA OJ OJ IA-4- IA-4- OJ IA OJ O CO-sf IA H on-3" OJ on IA Sq 



vo vo tA-sf vo 



vo j* cnvo vo 

a ass 1 3' 






O HV OJ 
o-=t on-st- on 



APPENDIX A 



403 



oj 

on 



too 

I 02 CO 
H CD OJ 
WON 

Sfi S 

ON 03 

CD ^ 

h fc 

3$ 

CD i 



>& 



g 
! 



S 

S 

.o 



^-N^ 

^^3" LT\ 
OJ N ' 

-N KN LA 



O 
^ !H 
O -it CD 

<VH cxi -P 
o ON ofl 
pq H M" 



<J PP o 



3 



9 




&\ 



-4-ajir\Hcnc\ic\jc\JO 
oonnnncY^roooooooon 

HHHHHHHHH 



CQ OH|WO 

O O 



^OJVOOJOonONHLAH 

asassssin^s 



aj co ir\c 
aiajajc 

r-iHH 



o\ ON o o 
cvicvjooon 

f(Hr-4HH 



HONO 
HHm 

HHr-i 



CM 

H 



ir\ 
OJ 
r-1 



o\cncr\ ri 
cnrooj J 
HHH H 



H 

oj H' on oJ oj 

\D ON~ 

H r 



- 

O\on 
H 



IAH 

00 OJ 
OJH 



on onvo cxi O\>-CM 

O\ OJ J* 
HonH 



H r-i O 
VOHH 



icoHO 

I OM>- IA 
H 



ooiA 

IAVO 



vo oncxi LT\O iPvir\t>-monona3ont>-cx! t 
voopt>-cxiir\ovoi>-oco incT 

CXI- 



I I O 8 1 

1 I lf\ 8 



S IA OJ 

H H 



IA I 

on i 

H I 



8 88 88 
883888888888 8888888 
88-*8888S>88888 H 88 



> O O 

O O O H OJ 
PS O O O O O 



o\\o oo 

3i a 



CO 



O OJ J- O, 

H 333 3' 



tA lA-d- O 

on-4- on 



ON CXI IAQO VO H VO j* IAVO on CX! -3" 

vo IA IAVO H H -=f IA cu -4- - 



O\ ON o 



g ri d -=*: It: lOPP ^? J SP 

1 



IA LAVO 



IA 



onJ- H o 

VO t>- I> b-VO VO VO VO 



IAOO .4-HC\iHO 



- a\ oj 

- OJ LA 



cvo H o on-=t on 

VO t-CO IAVO b-00 



C\i 
c\j 



oniAHrniAHO\ovo Os^t ON 
rnoncvioncxioncu-=}-cxiononon 



OOj-^- 



rn o OJ 
jHonrH 



vo 
ion 



vo H IA 
^t oon 



IA CXI OJ O 



H 
on 



O VO O ONIAVO ON-=j-CX)-4-H|(\jH 
HOHOOOOOOO OJ 



ONVO OJONHOVOOQIAO 
HOJOJOJOHOOHH 



VO VD 00 VO IA O 00 
OOOCXIHHO 



H-4: onVO tAOO CO ooOO 
IAVO OJ H on OJ OJ OJ H 



o LA cxic 
on on 

H 




vo on^t o o 
-=} LTN on o o 



CO CD OJ 
VO LTN LA 



H OJ ro.4- IAVO t-CO ON O H Oj on^ 
HH 



LAVO t>-00 ON O 
HHHHHOJ 



_, b- OJ 
OJ on-^- 

onH LAV 

O H OJ 



CO 



tl ' 



<C| CQ 



OJ OJ OJ OJ OJ 



<D CD 



S 5 

$ 



I 
1 a 



-P CQ 



M -P 

^ 



404 



APPENDIX A 



on 

0) 
H 



CD on 
W ^ 



S 
H O 



o ^ 

d) o 



P bO 
O fl R 



O O 
H nj 
P fl 
03 O 
<D ft 
3 03 



*H 0) 

W [Q 

^ ft 

(D O 

-P fc 



J- CQ 

<U 0) 

P W 



<Jfa 



dnojf) 



^tCVIlAHnnCVjCMCVJO 
rnonononmonononon 

r^^HHHHHHH 



i 



ir\ a\ona\ H 
cvi no on OJ -4; 
H HHH H 



vo rn w in O ir\ (A t- on rn rnc 

VOOO^OJiaOMDC-OCO LA 
CVJ (M ^ H H -* H 



t- o o OJ cu 

3 VD m^O CU J; 00-4; 
OJOOHHHOJH 



OO 



TV 

on 

r 



OH !W H g OH|* 



H H^tm 00 

' 



en co-*-* ooo onco onOOO^t-HonvqiAonCM 

HHHHOOOOOHOCVlonHHOJHHHH 



Ocnoo t-cv.o IA^- t- o vovq on oo 

ss^s'a^s^i d ^arf s 



CV| O CTx^D OVOCU-4"OH-shOCOOOOOp 0\-s}" CM X 
HHOOHOHHHHOHOHOOOHHJX 

CO 



CO^f O CO 

3 



H CV/ m-d- 1AVD h-CO Ch O H CM rO-4" lA^D 



- CO O\ O 

HHHHHHr-tr-lHHCVi 



d-vo 
4- en 
H H 



o (A cvj oo on 
t on on H tr\ 



H|WO H CM 

ooo 



on mvo 

J* -* 



on CT\ 
m H 
H r 



.4 CO 
on 



HHH tArn 
CO 



H CM m-4 IA 
CVJ CM CVJ OJ OJ 



CQ g 

H 1 

TJ -H 

O -P 

!>s m 

O <D 

HI 3 

<D 



| g 

*^v 
CQ 1 s - 

*d H 
5 -d 



rt on 

o H 

H 

43 -N 

J CM 

ta H 



I 



s 

e 5 

CD * 3) 

LA 03 JH 

CM LA -P CD 

<D K> fl +5 

H <D CQ 

^ rcJ -rl 

n bD JJ P! 

id -P -H 

,< ft ,CQ S 



APPENDIX A 



405 






M ^^ 



A 
W) 



CQ 

83 



*: 



-P n3 
CQ rt 



a 



i: 



1 

O 

-p a 



3 ft 

^ CQ 




fl 


PH 



Q\OJOJOO 



OJrHONrH COOJMDCOCO 
' 



. 
OJCVJ-d-rnrotHCUrOiH 



8S8S 88888 88! 



HH-*rH0OlT\OO 0O Cr\rHC 



LfNCOOOooCrxCVJONC* OO 

- 



- 

HH H 



ON O- lTMf\ CO 

OJ ro^d- ro vo 

H HHH H 



OLT\_d-ON COOrornO lACOCVJ 



rornO lACOCVJOJODOOHOH 

oo m ir\ oo -* I>-VD vo ir\ t ir\_^- 



888 



LP\ 



IT\ LT\ O r 






H oj m-4- ir\ 

OJ OJ OJ OJ OJ 



I 



0< 

I 



S 

M 



406 



APPENDIX A 



IA <>* 

on <D 







o 


fl 

3 



< pq 



o 

M 



IA 

on 



5 co 
q -p 
o a 

H 0' 
4i ^ 
CO Pi 

3 ft 
G? CQ 


H K 



Veterans 
No Res 



Veter 
Res 



< 

s < 



UOISTAJd 



H O\VD C- OXOJOjcn H H-^-on <X) 

j-ro^t^t -d-J-moo on ^t on on Lf\ 

HHHH HHHH H HHH H 



JHO\H OOCMVOnnoO O OOVO CT\CO 0>H O t-- 
H OJCVJ 



lO'l lll 
IHII ill! 

IHII III! 



1 IQO 

I I&OJ 1 

1 IN-/,-} 1 



8S88 88888 









-p 



H IT\H IA > 00 COO I > 00 OQVOD- ^> 

mcniAfn OMT\IT\ i oo H mj- v,o 

H H H H TJ r H HI ' cHH 




t>O\h-LA LTMAonOCO LAon-4- C- H OO t> b-H 
t^-VOVO> t>t>ODOcO VDOO t>-VO t-VD lACO 
H 









OJ rrjj* IAVO C-CQ O\ O H OJ on^ IT ^X) t>-00 0\ O 
HHHHHr-lHHHHOJ 



CJ-=h CM l 



, 

\s 



8^888 



O IAUMAVD 



H CM en j- LA 

CM OJ CM CM OJ 





H 




H 

" 
CO 
$ 





H 
P 
03 
N 



3 



CQ 
ft 



* -P 

H LA CQ 
P IA CQ -H 



&^ 

uj 

PS CQ cd 

V -P & 

a 

H q 00 

^ 5 -p *3 



APPENDIX A 



407 



(A 
on 



l Questi 
Besponde 



ki 
se 



Veterans 
No Resp 



o 

is 



ing 



Vete 
Be 



ing 



Veter 
Res 



FM 



dnojco 



H o\vo t> o\CMCMon H H j- on 
^ m^t-* j-JtifNco on .4- m on 

HHHH HHHH H HHH H 



CMHONH OOCMVOmCO O onVO 0\CO ON H O t>- 

VD LAonvo H CM CM.4- on m H CM m H 



oo 
ir\ 



O J* I O iili 

monioo iii 

H H I H 1 I I I 



O tQlAO 

i> i o\(M cu 

H ^-rH H 



88882 & 



O\HOOOl>COC\JOO >1A fOOOJ 
on-tf.4" IA CKjt OMA 



TJ 

0) 



Ol OJonOl 
H HHH 



cnoooooo HjqhHooOJ no on c\J -d- -d- oo H Jt CM 



OJ OMA-tf OJ H-^ 1 IA H IAH H -d- 

jtm^-* j-Jd-vo CM on .4- -=f -=1" vo 

HHHH HHHH H HHH H 



onC-CM IAH O J- IACMVD O IA 
CQVOMD VD VD CD IA lA-=f 00 lA 



ChO H 



CM lA^I- 
IACM 



IS 

I H 



8S8S8 



LA 



g~~ 



O 1A O O ON 
O CM OCQ C- 
H H 



H CM on -=f IA 
CM CM CM CM 



& 

H 



I 



M 
fc 
O 

I 



H 

P 
CQ 


& 



$ 


s 



TJ 





| I 



<H 



3 

oj 
N 
H 



3 

P 
ft 

O 



^ g ! 
. &J 

SR S5 

oJ w> PI 

a? n ^ 3 

H ^0 "9 
ft S fl * 



fl CQ ocJ 

^ * 

H a 3 n> 

j^g 35 

H -P OJ 

< to OQ q 
* 



408 



APPENDIX A 



i ^ 
I flS 

ol-P -P 

O -H -H 

n3 O 



IA 
00 



All Questionnaire 
Respondents 


1 < 




H Q\\D C*~ O\CXIO?rn H H-tfoo 00 

tfoo-4-Jt -tf-tfiAoo co ,4*0000 in 




IA'CM 

sa 


N^ ^V-, 


Number 




SHONH COOJVOoOCQ O OOMD ONCO ONH O t- 
lAnovo H CM oj .4- oo oo H cvi oo H 


R 


OJjt OJ IA.4- 
LPvOJ 


Veterans Making 
No Response 


*3J 

S "^ 


'J 

H 

> 


O O 00 liii I IQO Q 
4- H i OJ illi I OSOJ CA 
H H > H illi I " 'H H 


H 
CC 




i "o 
vo 

1 H 

\~s 


p 
h fl 

CD 

{1, o 


1 
1 

si 
O 

H 


ir\v o ir\ ooooo o o\o ur\no C^-IAOOVO 

OOOO OOOOO OOOOOooOOO 

CQ 03 

9 


15 

< 

W 


IAO OCO 
OJ O O O 


Veterans Making 
Response B 


O 

13 


h h 


LTMAM) H roC?C?C? 0H O OJ VO CO 
.3- .4- J- IA> lAoo^D lA |> on IA-4- on ^4- 
r-IHHH HHHH H HHH H 

H3>^^xs^s^ fO 




i 

H: 
o 

g 


,-^^^s 

I! 



S 


HIAC\0\SlfMAt^POCVI 3 1AVO ON-* COCO OJ oo IA 

o ir\ir\vo s c*-t*vo oo\o S ITMAIAVQ IA ITMAVDVO 
5 g 


(X 


o ir\o cvjco 
o cu ovo IA 

H H 


Veterans Making 
Response A 




13 


s 


\T\\Dt~Q\ S"aoO W |A H" OOlAH P 

rofn-4-m miACJOJ oo OJ cvj on vo 

HHHH HHr-lH H HHH H 


S 

1 

1 


Ii o" 
i H 
I H 


-P 
h fl 



(1| O 





4-ONHVQ UTMAml>"CO LfMfN H H lf\ LTN CO 00 ON 
oooo-=hCU CM OJ OOV) oo -4- co-=t OO-4- O-4- oo OJ 


1 

g 


O O OCO OO 

o ir\o oo oo 


CIQTStAId 




CQ JH -H 

P W) H CQ 

L= = srs = = sE = = = fltrr S) p = 
< m -^W 




9 fea 
'- i! 


-Ajun 




CO CQ CQOCJ CQE-fCOO&HEHSCCO 

opqrsScQtEKfiHswEHoSSsrass^ 




^ m si 


dnoa*) 


H OJ 00^- IAV t>00 C3NO H OJ 00~=h IA\D D^CO ON O 
HHHHHHHHHHOJ 




H OJ 00 J" 1A 
OJ OJ OJ OJ OJ 



H ,!A m 
-P IA CO H 
03 fO d 

h^ 



fl CO OJ 

Vl O -P 3: 

! 

H fl 00 



APPENDIX A 



409 



CM> 

I 



-P 

CD 





H 

IS 

O 
T3 







43 



IS 

CQ 



H 
O 



-* 
^C?00 



ft 

CQ i 
C 



O -P 

s^ 

CQ 

N CQ 
CM *H 



Q S 

H CQ 

CH H t>s 

CQ -P & 
H OJ CD 

S S 
S .P 



' PR CQ 

| pq o 



f 



PQ 



5 



-*CMLAHOOCMCMCMO 
OOOOOOOOOOOOOO'"OOO 

HHHHHHHHH 



V> r-1 t- 
H H CM 



ON CM CO LA CO ON ON O O 
H CM CM CM CM CM -" 



Ojpooo 

HHHHHH 



t- H ONO 
no H H OO 
H HHH 



H 

% 



CM OO OO CM -* 
H HHH H 



t lAt*-H H 
OJHonOJ OJ 



00 -* 

CJN OO 

H 



H O t 
HH 



J^HH^ H 



OJCOHOOODLTN 
QJ O\ t- lf\ O IA 

H H H 



CM - 
H 



I _* H H I OO O I 

I 1 H H H I H H I 



O 

H 



scnV 
i -! 



ISOOOlt 

i ! 



iO 



OOO 



o g 88 88 



O HHJOH|OJOH|OJH 
OO O O 



HnJcuH 
O O 



H H O O O CVJ 
OOOOOO 



8 8 HlnH4 8 8 8 g 8 8 8 



888 



m|C\! CM ,VO KO bt 
H CM HlH Op 
HIHHHJHIH 



OJ D00 <n 
OOO H 
HHH H 



liTXCO 

1 



lH 



o co IAD IA 

OO H CM CM H 



MD CM 
H CM 



H OO-* CO O CO H MD H MD t>- CM OHOOCOHOO 
OOCMOOCMHHHHCMCMOOH -* CM H H CM -* CM 

MD CMVO H CM ONLAO ONH O-CM OOMD CM ON oo ON-* MD 

OOCMOOOOHHHCMCM OO-* HooCMCMOHCMCMH 



IHOJoOCMOJ 
H^r-JHHH 



t>-ONH 

CMCMm 
HHH 



o 

-^ 
H 



HHCVJ 
HHH 



CM 
H 



H-=tHHVOHHt--O 

^ ro-^ ooojrnoocMoo 

HHHHHHHHH 



MD tT OO MD LA CO-* HVQ 
LALALALA LA .* LA LACM 

_,VOOenVOenONenC HCVIC O 
LAMD MD LA,* LA.* LA LA LA-* CM LA 

OJ MD CM CO CO CM O t-CO 1 ONOO-* VO CO OO-* C- ON CO OO 
LA LA LA LA-* LA LA-* -* LA-* OO LA LAMD -* MD LA LA LA 



h-CO.O ON, 

d 

H 



d 

H 



VO -d- Jen O\jt>- en 

en-4-W- oout -3- 
H H|H HSH H 



D- 
CM 
H 



lAJCM if\ l>- 
in en oo in 
HIH H H 



VO -=f 
OOCM 



H CM CM H oo oo"C\J HMD 
-=T 9>0 H 01 CM O^ 



O CO -* CO H 
HH-4-HHOH 



CMOJ 
HJCM 



HHOQMA-^roO ON-=J- H in-* l>- O en CO H 
HH-=t-OJcnfnCMHOlAHHH-*OJ HHco 



s s t r 



J S = 



B 



CCO ONO H CM OO-* IT\VO t- CO ON O 
HHHHHHHHHHCM 



O lACMCO 00 
IT OO OO H LT\ 
H H 



I 9 
I I 
t I 



Q 88 

HH|CMO O O 
O OOO 



LA C 

33 



CM OO O 

oo H H 

t>-MD CM CO O 
CM H CM 0000 



o LT\ 

OO 



OO 
,H 



H HMD 
H OO OO H O 



CM CM a? "CM CM 



8 



LA a) 

LA +3 

KN CD 



a 



O CQ 



>> 0) 



O -P 

C? 

H 



-P 

ry 



O 

H 

i 

43 



* 





OO 
H 



-P -. 

g a 



fc O 

O K 





8 * 

H 

P tQ 

oj 



CQ 

<H -p 



I 



410 



APPENDIX A 



PQ 



-p 

H 



All Questionnaire 
Bespondents 


1 




ON CM CO LACO ON ON O O C^- HONO H 

HojOJCMOjojojonon on HHon CM 

HHHHHHHHH H HHH H 
* CM LAH on OJ OJ OJ O LA ON on O\ H 


H CM 
CM CM 
H H 

J- VD 

-4- on 

H H 


^ 


I 


1 


oj ON.* vo on on.* LAH 

D-tAtH H ONfO CO CM 
OJHmCM OJ H OJH 

^DJ^-CO oncnvo oj ON> oj tj- OJCQHOOCOIA 

vQ ON.J* CNCM.* H H O D---* ON CM ON !>- Lf\ O IAVO 
H HHonHVOHH H HHH 

vo onojLAO LAiAt>-pnononcoonc>-CMh-o o CM CM 

,* OJ OJ -* H H -* H OJonHHHOJH 


ON t O 
> O LA CM CO on 

d t- on on H IA 
3 H H 
n 


Students Making 
No Be spouse 



1 


a 


cc 
ffi 

i 
1 




C 
^ i i 
5 1 1 


oooo oo oo 
i co LAH H ono onH 

1 N-/H H H H H I I H H 


I I O I on I I O O O O 
I 1 on I LA 1 I t>- .* C-OJ 
I I H i H 11^ HHH 


LI CM CM 
^ H H 


-P & 

o s 


COIJkEGE AF 


<HlCVH|OJO O O O H|(MCM 
O O O O O 

O H onnJwH OH|J-H O HH|CMO O HnJcMO O O O 
OOO OO OOO O'OO OOOO 

8 8 H|W 8 S888S888 8 r1|tvH|AH| ~8 888 


B 


a o o o 

3000 


Students Making 
Bespoiise B 


U | 

L 


VETERANS WHO ENTERED 


^o on ON co CM CM LA on ^t on LA.* IA on 
ojonojojononononon Jt HOjon CM 

HHHHHHHHH H HHH H 

-t on-=f on^t-* -4- on on CM -* on on -* 

HHHHHHHHH H HHH H 


i HCO 

3 on CM 

q H H 

q LACM 
q J--* 
H H 


p 

CD 


1 


VD OJCQ LA CO H OJ LA_=h 

LAOJ ir\OJ LAIN- ONOD on ONO i>- oo ONCMVO LAONCO 

-* LT\V) VD IA LA-* -*VO_*-*[>- ^--rj-COVO^sDOJon 

^- IAO LAO OJ t"-CO\> OJOcOOJIS-CMONoncOONO 
^sr LA^ MD IA LTN-* -* LA LAJ* IA LA LTsCO t VD MD LA 


3 
t 

5 VO ONCO 

5^- vovo 
: o^ ono vo 


I 

S CD 
CD 
Q C 



1 


g 


1 


o 


^r ONliA H on/* ro|t*- CM ON CO LA O t*- 
n CM CM OjfOJ OJ CM CM H| O H H H 
H HlH H H H H|H H H HHH H 

OiUt-lt 14-IlA H CMKO VO H IA H f- -* 
OJJCM OJKM CMlOJ OJ CM OJ OJ on on OJ CM 
-ilHJHlHlH HHIHH H HHH H 


"I H H 


Student 
Beajpo 


PH 


GBOUPS 


-* CO CM LA OJ CNCO LAVD 

ir\-* on on LA Lr\.d- LA CM 
LTNCOlACOlAroHCMI>-HOon CMHCO-*LAHOJ 

LTN-sr on on-*-* LA lAnn LAVO oj IALAH on on c vo 


)^- H CM 
H LA on on 

onvo t--o-* 
-* on^t on CM 


HOT ST AT (3 




& t B 


CO ^ *H 


^ l~ " " ^1" 


Atao 




SHggoo^^oaig^sii^ssis 


lUcaSI 


dno*: 




r"" CM on,-T IAVO C^-CO ON O H CM on_* Lf\VD D CO ON O 
HHHHHHHHH HOJ 


H CM on-d- LA 

CM CM CM CM CM 



IA 
LA 





tao 



o 

H 



CQ 
H 

O 
ft 



CD 
H 

8 

-P 



ts 

Nl 





H 
P 



H 

! 

CD 



3 



O 
H 
4J 

CD 




l>- 
H 



on 

H 



& 

O 



CD TJ 

TJ CD 

-P CD 



APPENDIX A 



411 



13 



-P 
(D 



-p 

s 



H 





I 



s 

H 
<D 

p 
+2 



H 

<D 
CD 



co 



P A J> 

fl OJ CD 

CD > fl 

cr 1 CQ ^ 

<D 0) O 



CD O 
H CO 



l rt o 



OJ CO 

O CD 

H 1$ 

-p fl 

CQ O 



p a 

r 
Ife 



"5 

I 



CD J> 



UOf ST Aid 



AIUQ 



-4-CXIlAHonCXICXJCVJO 

onononononroonnnon 

HHHHHHHHH 



xo^tco monxo cxi ONt-cxi i 

VO ON-4 O\ CXJ -3- H H O t-- 
H H onH VO H H i 



ioo IACO a\ a\o o 



H H on 

HHH 



3 



IA 
CXI 

H 



& 



a\m 
H 



LAH 
CO CXJ 
CXJ H 



OJ CO H O O CO lA 
cy 



VO LT\VD 
OJmH 



OOIA 
J-VO H 



3 



i on 
i cxi 

I H 



LAO 

-4 O 
H H 



8 

H 



O O 

on H 

H H 



O 

3 



OH|tynko 

o o 



H|C\J 



O O 

o o 



HJOJH 

o 



OHICUO ggO 



gOOOgO gH~OH<MO 



\O 

a 



3 ^HOO H 
H HHH H 



CXJCXllACXJ Jt 

cxi on m oo cxi 



a .4- onvo co 
CXI 



OCX) 

cxi cxj 



^OO 

cxixo 



- - 

on en on oo ro cxi -* oo 



O -^ VO CO t- CXI J 

oo cxi .4- cxj cxi H 



H t>--=|- HcnHOOONCXIHtHH b-XO O CXJ 00 IT\ 

cxi cxi onooonnnonnncxi oocxi ITNCXI oncxi-* oooncxi cxi 



cxi c*- on H H 



IA o cxi if\ o ON o xo -d- 
onononcxioncxicnonm 

HHHHHHHHH 



& 



co H on ON 
oooncxi en 

HHH H 



o 
ir\ 



m tr- 
- 



.4-xo 
tr\ cxj 



in ir\ t>- cxi co cxi co ON ON -4- -4- ON 

--4--4--=t--4--4- -4--=J- IA CXJ 



ON on cxj co ON mco 

" " " - IT\ 



O-4-vocp 

-- 1 



cxiHHirNOon o 

- 



ON ON IA 



O C-lOJ t-CO 

H ONfon onHcxi 

H HHHH 



CXJVO 

ioom 

HH 



H 

oo 

H 



NONHJ 

NHO 



& 

H 

ON 

a 



_. o i>co -* 

CXJ OJ H H CXI 



t>- LA 
CXI CXJ 



ONXO 

H O 



O O CXI m.4- XO 00 CO CO O t 
ICVICVJCVJHCVJHCXJOHCXJO 



H CXI CXJ -4 j* on IA 
CXJCX1HCXJCXJ-4- 



ON C5N O H t IA rnXO ONXO COCOOVOO O-4" ONonO 
CVJCVICVlCXJHCVICXJCXJOJCVICVJO-d-HCXJHHHCXiCX] 



CXJ on-4- IAXO h-CO <>> O H CXI m.4- 1AXO f-CO ON O 
HHHHHHHHHHCXI 



Ost-O 
ONON-4- 



O LA CXI CO on 
O-onooH IA 



888 

HH|C\JO < 



LAlA 



oncost LA LA 

on-4- -4- CXI H 



on co 
.4 CXI 



.4- HCO 
LA-4 on 

O b~-4 
-4 on-4- 



HXO 

H H 
H H 



ft: 



- LfN CXI 00 -4- 
CXI H H H CXI 



H CXJ on-4 IA 
CXI CXI CXJ CXI CXI 



CD 

* h 
LA CD 



CQ fl 



P 
CQ 



I 



3 

s , 

ccS CXI 

N H 



S 



ft 

W 

a) 



1 



* r 



412 



APPENDIX A 



ca O H 
(DOG) 

H O CO 



ON 

on 



in 



u 
R 



k 
B 



R 



UO?SJAT(I 



AttlQ 



dnojf) 



ON CM CO IAO 
HCMOJOJC 
HHHH 



-3-ojiAHoncMOJOJO 
monononononononon 
HHHHHHHHH 



H H O O\onCM D-OlonVO 

O-=t on CM on on-=t*-4-W' fo 

- H H H H H H HJH H 



t- 
on 
H 



if\ 
01 
H 



HONO 
H H on 
HHH 



ON m 
on on C 
HH 



H 
CU 

H 



CM ON-4-VO on 
t-lAt>-H H 
CM H m OJ CM 



ro-sf LAH 

ONon 00 OJ 

H CMH 



VO-=t co on onvo CM O\ t- CM C -=i- 
VxD ON.4- ON CM -4- H H O C -3- ON 
H HHcoHVOHH H 

O iTNlAt^-oOrnoO 



op H O o CO lA 
ON t IA O lAVO 
H H 

POt>-CVj t~- O O CM CM 

vo LT\VO cvi^t onj- 

OJOOHHHCMH 



oom 



oi 

O 
H 



BOO 

IPOH 
I H H 



OOIT\LT\O 
mPOlA-3-tC 4 - 
HHH HI*- 



H|cuOHw O 



H|cuO 




O O 
00 



HHOHICMOHJCMH 



O O 
00 



HHlcug 



HH,cvg O O O 
H OHICMO H OHlcvg O O O 



H ON ONCY^ O m-4- CVJ ON 
CMHCMmoooorn-^on 
HHHHHHHHH 



ON^t O 
OOJPO 
H 



CO 
on 
H HH 



oo if\ o .on -3- 
CM j^-^tlon on 



HHH 



lf\-4- 00 C CM j* VO on O- CM H C^- ONHcnOHHO 
Hj-ir-jHooCMCMCMCMCMCMCM CM^hCMCVJCMOnCM 

O m^- OVOCOVOlArOOONHcXDOOOVOHONCMH 
ajCU(NJOJCXJHCMC\lCMcv^O 



00 LOvOQ OO t ON ON ON H 



^tOOONHOMTNCMoOlA 



O O t OJ IT\ 

3 333 2 



a 



CQOO 



t-MD 
Lf\ LT\ 



CM^J-^d- 
fMTMA 

VOJ-HOirNHOQMDCMCM(Y-)iA on t*-^* VO on t*- CM 
IT\VO VO VO LT\ lA LT\ IA 1AVO VO IA IT\ j=h VO IT\ VO jfr VO 

ON C*-VO 00 VO O ON H H ONOO -4- 
ITMfMTS ITMTNVO ITMT\VO 



CO CO m ON Ov ON 
TN Lf\VO ^j- lT\ LT\ 



onVD[>-oonooMA 

HCUCMonCMOJHH 
HHHHHHHHH 



ON 

CM 
H 



l>-CMro 

HHCM 
HHH 



00 

! 

H 



,H ON^f JCM CO JOO O 

tl :rlB^; 

IHHIHH 



- 

H 
H 



-=* 

^ 

H 



1AVO C^j* 

on on CM on 



on H H 

CM on on 



CM ON 

on on 



OJ on-4- iTvVO t-CO ONO H OJ on-* IAVO t 00 C7N 



HH 



- 
HHHHHHH 



O 
CM 



o\co 




G 



O IACM 00 on 

t on on H m 



O 
O 



8 8 



VO CM 00 LTNVO 
OJ CM OJ CM CM 






^3 






Lf\LT\ 



on t-vo 



co vo 

ITNVO 



CM 
CO 



IH LA 
^3 



on H IA 
CM OJ CM 

H HVO t>-OO 
OJ CM H H 



CQ 

-P 



H CM on^t lA 
CM OJ OJ CM CM 



355 
s te 



s given on 
naire was 



I 3 

!>a CQ 

O CD 

3 & 



s table and the 
7) were freshme 



<M Cd 

o 

a 

H 

4^ ^ 

cd CM 

ts3 H 
H 

ft 

W 

g 

5 ~ 

- a 

n ? 

> 

H 

P CQ 



V 

A brief expl 

y y 

Students in 



APPENDIX A 



413 



* 9 ^ 

*-> <o ,0 i 



-J .-5 3 

S3 3~ 
**3*3 



* P ^ d 

s* 3 



a co 



O CO , .d -P 

CQ CQ -P -P 

^ 2 IS S"1 

2 fe e HI 



i m 



3 tn 
C3? CO 



m m 

-P CO 
CO 

1 



] ffl 



CD CQ 

3 

m 



"AfUQ 



3 



H H rnHVO i 

_ . . .O KM 

VDO3 t-CVI 1AOV 

d- cd CVIJ-H S. 



g 



IW 



cu inn mcvj CM CM o 



3 



s 



o\cnos H 

3321 3 



l^c 

C-OCO i 
H 



OJt^- 

U^VO 
ojfOH 



1 1A O O LTNCO O 
l H C-vZ5 CVJ H IT\ 
I H^-'H HHH 



IAQO > O i 00 

o5 no ; oo j 3 

^-'H 1 H 1 H 



I IT\ O 
i .4- C- 
i H H 



O t-CO O IT\ 

q ON.* H if> 

H v^H H H 



OOOO 



OO 



OO 



3'3^S3333 

0000 enH^ IACM CM ON 

333333393 



H cnncu m 

3 dSin 9 



HVO CM 

533 



if\cnon H HVOVO a\cuoo o\o\ 

lAlTMTMAVO lA-iAVO (A LTMA^t 



o en CM VD ^ 

VD VO UT\ m^ 



fr~lA 
> 1AIA 

LA-=h H cnHCO-=f-sfVO t*-CVJ cnent*-H lA-4" 00 ^t IA 
en CM -4- envo en IA en CM en^t en cn-=f ^t en CM 3- CU en 



C\J H f- H O O CM VO VO O VOHOO 

33333333?) 3 333 



CTsOJOO 

cn^t cn-4- 



IA 

H 



VOH 

CM en 



HQ 

en IA 



1A t>- t- ON O\- 



LAVO 0\t- O\ C 
VO t-lAVO enV 



t\D eniAc: 



-=h encO t-t-enOMAVO CMVO IA 
t-VO lAVOVD lAlAVO t-lAt- 



H CVJ en.4- lAVO b-00 CT\ O H CM en-=|- lAVO tCO ON O 
HHHHHHHHHHCVJ 



J- V> 

33 



s 



O IACMOO en 
t men H IA 





L 



8888 



ON CO 
en H 



o eno 

lAlAXO 



UMA^t J* 



la 

3 



H CM en J- IA 

CM CM CM CM CM 



R ,. 

H 
CO ^ 

"d 
"y 

. 

g * 



II 



f 



o 

3 

H 

P -. 

5 3 



(D 
P m 



414 



APPENDIX A 



II 



^f- 03 in H en oj cvj oj o tr\ ONPOON H 

si 332 3 



o\ CM oo tr\co o\ o\ o o 



\co 



s 



HO\O H 



OJrHoOOl OJ 



POCQ oo c~- oj t~~ o o oj oj > 
lAONVOvo IAVO cu-* rn-3- E 



ONON5 



O IA OJCO CO 

C~- CO CO i-l IA 

H H 



T3 -H 
<D -P 
l>3 03 
O <D 

f 3. 

O 













w?S 



I J23 O 



OJ rOHjwrtJcuOJ OOOHHOJOJ^I-OJHOJOJH POHJoMcx 



OO 



^- 
OOOOOOOOOOOO 



I 



O\ OJ IT\ IT\V) ON O\ ON 
HOJOJCVJOJOJOJOJ 
HHHHHHHH 



H-4- ONOO-4- H ONO 



O OJ ON O 
-=h HHno 



OJ 

j oj 



J wo 



oo ON ON "ON ON ONCQ co ON 

co o cntAt>-co o i o ONt>-cn ONHOO oj oj onoo 

I>- ON ON ONO- CO ON CO l>- ON CO CO 00 ON CO CO ON ON N- 
^- OO^t OO^t- O ON^i- t-CO HVQVO IAIAO fOH IAVO 

tcocococo ON ir- ON vo co co t-tco cocovo co t-vo 



O OOlA 
ON ON ON 



ONMD H oj H 

1>-CO CO CO CO 



$&' 

-ilH i 



CO 00 

rn oj 

HH 



a^H 

IA C H ON.OJ IA lA-=h CO 



VQ 1AONIA 

oj o H on 

H HHH 



1 

CQ 



H tA-4" VO OO ^h ^O OO O 

HOOO O OH HH 

cuqi>~iAooojqooQHp' % )i>- HONOJCOOOIT-OJ 

OJHOOOJHHHOOOHH HOHHOOOJ 

sO 0-VQ f~\O O H VO CO CU ON J* _=h IA UTN O !> ON lA^ 
OJHHHHHOJOooHHIOJOJHHOJooHOJrO 



O C ITS 
HOO 



-p 

i rj 

H 



nOISTTATO: 



LTNVD t-00 ONO H OJ 



rn 
r-f 



^t IfNVO t~CO ON 
HHHHHi-l 



O 
OJ 



OJ OJ OJ OJ OJ 



APPENDIX A 



415 



r! 



O 03 

O 

bOO 

0} 

9$ 

H bQ 
J4 Iff 
CD -P 

A J 



P 

P 

CD CO 



^6 

CD CD 

s* 

m 

M -H 

o ,d 

4J 





iH 



i?i 

*^f ' 

o 

3 a 



5^ 

P tJ 



H > 

I 6 



*t 



,11 

>s 
I ,0 



i +3 S 

) 

"S3 
IEJB 

) (D 



P , 

CQ 
H 
H - 



a. 



J-j -v 

CD 00 

O H 



s 



-P _ 
ft 






fe 



oxojc 
Hoj 

H H 



C\J IA H m c\J OJ CM O 
on on rn on en on rn m on 
HHHHHHHHH 



H o\o 
HHm 

HHH 



ITN 

oj 
H 



O^t IAH 

^ g 



CO oo OOVO OJ ON C OJ !>-.= 

HHOOHMDHH 7$ 
") OJ 1A O LA IA t*~ 00 CO OOC( 

l&SSSS'-S 1 *"^ 



VOXO LAVO OJ 

oJc^HHHOJH 



1AVO t- 

HHOj 

HHH 



O CO CO O 1 1 

vocno^-i i 

HHHHi! 



en j* 
H- 



o i oco i 

OJ I OOVO I 

H I H H i 

O CO H 00 CO 

H CO LAOO IA 

H * 'H H H 



S8S8 S 88 8S 



ggg 



oo O\lf\C 



O on 

cn oo 

HH 



H H H H r^ H 
LAOOPO O H LAO H H 



OJ H O\ 

H oj oj 

HHH 



^-00 t- co 
, 0000 00 OJ CO ^ VO I , _ . 

QHCOCOCOCOCOCOCO c-cocococo 



OMAtT-O t^-H 
C-CO I>-COCOCO 



OJtVOJOJ 

CO C-COCO 



VO Ch O\VO I> 00 LA 

cooocococoooco -, 

P! 



VDOJ O\j^- H C"- 



C-CO t-CO ONCOCS 



-=t- oj IA ON c vo cr\ o 



HtOJooHcOO.OJ 



oj 

a a 



j* 

l 



ir\ 

a 



H OOOJ OJ 

3 3H3 ^5 



}3S ^iriS^3& 

>s?8nsas?ss*asisHS& 






H OJ 00-^" IAVD t-CQ OXO H OJ oO^t lA^D C"-CO 0\ O 
HHHHHHHHHHOJ 



H OJ 

as 



O IAOJCO 00 

t- 00 OO H (A 

H H 



I>CO CO 

D-CO t 



H OJ OO -4 1 IA 
OJ OJ OJ OJ 0! 



03 

a S 



S 



O <D 

d 2, 



i 



ts g 

M H 



S^ 

-P ca 



i 



416 



APPENDIX A 



<J W 



1 



IS 

O CD 
H 13 
P P 
CQ O 






CQ CQ 
-P CD 
fl W -P 



5; 



i ^ 
! O J 



CQ 
T3 CD 



8 fe 



dnojf) 



t LT\I>- H 
OJ H ro OJ 



3 

OJ 



LAH 

?33 

& 8lEiR8 

H HHH 

fxJVQooOJlAO LA LA C 1 - OO OO OO 00 00 C-- OJ C O O OJOJ 
OVOC01>-OJLAOVQC--OCOLA O\VO VO IAVO OJ ^t OO- ' 
'OJOJ-*HH-sf H OJ 00 H H H OJ i 



H|tiH|<\lgH|fti 



ONOJOO inco o\o\o o 

H 01 CVJ CVJ CVJ CM OJ CO no 
HHHHHHHHH 



-sJ-OJlAHcnCUCVICMO 

oo ro on oo oo ^o cn on oo 

HHHHHHHHH 



"A 

OJ 

H 



HONO 

HHcn 
HHH 



ChooO. 

oo ro OJ 

HHH 



H 

OJ 
H 



ONOO 
H 



CO no oo \> 

^J- ON OJ ^t H HO 

HHCOHVOHH 



O Hj(UO 
o o 

8388388 

H HnJCVHlcuoO OooOHOJOooOOJH-4-Hoo Hnlw 
OO OOOOOOOOOOOOOOO 



ON IAC oo ON oo ON^J- H 

HOJOJOJOJOOOJOOOO 
HHHHHHHHH 

LA1AOO H 00 t-CU t- 



I 



3 



LACO IA 
H H oo 



H 

ir\ 



-ON o jh j* o jf 
- invo VD 



o on oo oj H IA oo 

4- IAVO r- 1AVD 



O-4 O VO 
LA3- lA^l- 

'5S&^ 

0>H LAO\00 OJ IA OJ ON O\ OJ IA 
-sf LA IS-VO LT\ LA IA-* 



^Ogc 

H f 



HOO^J"O\ 



LA VOlAOO 

w i o oj oj 

H HHH 



OJ 

oj 

H 



IAOJH OJ 
oooooo rol 
HHH H 



O VO 

IALA 



O -4 1 
LT\Tr\ 



LAVO IALA 

5|H OVOVO OVO H IAVO O O VO HHI 
C3 LALAOO OOOO-3* CO OO O0-=h 00 00 -4- OJ < 






H OJ 00-* LAVO t>00 0% O -H CVJ OO^t IAVO C 00 ON O 
HHHHHHHHHHCVJ 



2I 



IA 
OJ 



888 



- - OJ 

LAVO LA^flA LAVO 



j* oo oo oj o 

VO IALA 1AVO 



OJ 00, 

oj oo 

H H 



H OJ 00-4- LA 
J OJ OJ OJ OJ 



3 

-P 
CQ 
CD 

& 
CD 



r 
ft 



O CD 
H ^ 





CH a 

a, oo 

O H 
H 

-P <N 

03 OJ 

M H 



S 

& 

H CD 

P CQ 

cd CD 



.si 



APPENDIX A 



417 



W 



j .p 

O 



ft 
3 CQ 
& 



H 

oj H' in oj oj 



>_=f IAH 

\en co oj 

H OJ H 

l vo.=i*co encnvo oj OMT--OJ t -d- 

\O ON^t- OS OJ -=f H H O !>--d- ON 
H HHcnHVOHH H 



fnojir\o LfM 

VOCX)D-C\!Lr\OVOt>-OOOlA 
^-OJOJ^-HHJ- H 



O\OJCO IfNCD ONONO O C HONO 



OJ IAH enOJ OJ OJ O IA ON en ON H 

, ^ . , m en OJ ^t 



OJeX) H O OOO IA 
OJ ON t~~ IA O lAVO } 



IVOIAOOJODO 
1 H HVO on O J- 



O\OJOlAmi_d-OO 
OJOC300JJ-iooO\t>- 
H H H H H i H N-XX^- 



1 < IfMfN 

I I CV)VO 

i HH 



O lACO 
H OlA 
H H H 



88 



S 8 8 & 1 g 8 8 S H|nj g gig 8 8 



ON H 1A-=J- t-0 

H oj oj OJ OJ 

HHHHHr 



OJ O\^t O\ 

HHTn H 

HHH H 



- 

33 



o t t>-o 
t-vo t>-i>- 

I- CX) OJ 1A-=1" ON LA ONIA H VO en H O IT-VQ O ^t ON 

) \) (v_ |v_ t>-V) (v>vo JV-CQ |s_ j^. IV-QQ t^_ t^. t_ ^^ 

) HCDCO LAC^-IACOCO 
-t>-VO t>-t--t>-I>-VDV5 



ON en LA O OJ O LA.H VO CO ON O ON VO 

H OJ cn en en en enloj cu oj o OJ H oj 



OOONH^J- ONOO.ONC 
mc\j-=i-rocnHlojc 



cn 



VO 
oj 



VQ tr\o ir\ o 
oj mcnoj en 



V) t-OJ H OJ 
en en en en en 



O en 
en en 



en O 
OJ en 



OJ CO LT\VO 

oj 



ITS H ITS ON.4- C 



o- ONO en-3- OVQ H 
OJ OJOJOJOJenojen 

\P W 2*> tr <Z~} ^ u^u^Q \P-rt O?>WOJ lAeniAOJOJ 



H oj 



IAVO c^oo ONO H oj en.* mvo t>-oo o\o 

HHHHHHHHHHOJ 



O ITNOJOO 

t on en H IA 



^gg 

enoj o H oj 
o o o o o 



s 



ONH oj 
vo t-vo 



t^ONOCD 
tN-CVO t VQ 



^8?S 

O en iH O OJ 

en OJ en en en 



$ 
r = 



H OJ m 
OJ OJ OJ 



-4- LA 
OJ OJ 



a l 

!i 



-S 

-p 

>> CQ 





(Lj m 

6 

3 

H 



tf S 





418 



APPENDIX A 



r* 



? 



fc Q 
9 -^ 



03 
P -H 



s *| 

rt H jj 
O -P 

T* m 

* ^3 

*, 45 

*H >S^ 
> .p 60 

o <D 

-4" -H -H 

d H 3 

r" O 

jg & g > 
S P( o ? 

44 ^a -P 
g o m 

S a 

o M 

t* JH Q 

H 



O 



!>a O 

CQ * ,b 

S|^ 3 

|5| T, 

D5 P iCJ 

<0 -P 

1 a - a 



SO O O -H 
^ 03 pq pq Hi 



: 



i 



1* 

CQ 



SI 



HOTSTAja 



O\O\OO 



J-OJiAHcnWOJOJO 
) oo no oo no oo 

I H H H HH 



00 

s! 



HHcnHVOHH 



H 

Hi 



i IA i m i i 
i OJ I IA I 1 
I H I H I I 



oor\ o oo oj 
cnoovovoj-^- 

HHHHHH 



m 
o 

H 



ir\ 
ITS 

H 



83* 



83 



H-IOHWC.H g,0 H 



H H 



88 
88 



lHIHlHJHVJlH rHlH 



volcvjl 
cu oo 
r-tlHI 



lcvji i>- 
Uj- OJ 
IHl H 



- 
m ro -4- ir\ on ir\ on 



OOC30 

--=h-4" 



ir\ tnHOJ ON 

a sns? a 



ONHHMDCOCMfniPvH CO HVOO O 

cvinncncucMmcuoncvj oj tncvicvi m 

HHHHHHHHH H HHH H 



tn H CO IA H 



LA-* -d- cn IA m j 
H >- ON OJ ON J- ON ro H ON t> t-vo < 



co 
- 3F 



"H HlJ d'p'HlH'H 1 



, 

O 



oo 

O 



-=KVOiOlmKQ VDlCOLt.OJ M mlOllAl 

CVjlH H H H OlONlHlrj O\| HO ONI 



ONVOOOOJ t~ voo\ \OVQ 

HHHH H CVI H OH 



s w 



1 S ! ! 3 



r^(!\i(n ^nr\\D t cb ON o 

HHHHHHHHHCVl 



o (A 04 co m 

t>- OO OO H IA 






m c\j 

00 



OO OJ VO OJ C\| 
O O O 



O H O O\ 

oo PO.4- OJ H 



IAOD 

Hi '> 



nd 

^g 

LA 4^ 



tlD -H 

5 | 

d -H 

P 



a 

3 



I & 

S 8 









CO 



APPENDIX A 



419 



A ccJ 

-P O 

* $ 

(D OJ 

T! 
- 



12 |j" 



h ^- 

H 03 J^ 
J-l 0) ft 

< P^ -P 

CQ ft 
<D 
p 1 T3 



I 



I 



3 



5 II s 

3 >J -= 



a> -P 
PQ ra 



o 

-4- 



13! 

O CD 

H d 
p a 

CQ O 
CO ft 

2 CQ 

cy CD 



* 



5 
& 

T3 CO 

r 



5 



*;& 



HOT si Aid 



Apnn 
dnojf) 



HICV.W H HHK,g g g H g g gg HNH g 



C H^O H 

3 333 3 



~ 

333333333 21 3321 3 



CMHnoCM CM i-4 CM H 

vO-jfrGOoOOOVOCMONt CM t .4" CM CO HI O O QQ IA 
VOCA^tONCM^HHOt-^tuN CM ON C*~ IA O IA VO 
H H H PO H V> H H H H H H 

VO OOCM tAO lAlAt-ronoootp not-CUt O O CM 
<D CO C^-CM IAOVO t~-OOO lAONVOVO IAMD CM^t <n 
J-CMCMjd-HH^t H CMroHHHCVJ 



< o"ir\ j 

2121 ! 



HHHHHIH 



8 

CM 



S^l^ ^rsrsr 



ro,VO| 



|ON|O|O -d- 1A|O\ cnl 
l5-LtLt3-_d-lrn ir\ 
lHMR H HlH Hi 



VQOMDtr- CM (Ot- b-00 
HCMCMCM CM CMCM CMCM 

CT\ONO roiAO\t^-oo H,O\ ro CM 

CO O\CO VQO O\OOO POOOPOH^ 
p-iHHCMrOCMooHCMCMCMrOi 



' tA O -* t-^jh 
i H O--* VO .4- ON 



COONJ^ 



NIAC 



\OmO\CMC*-lAoOr-IO\ 
moompomonmnnOJ 

HHHHHHHHH 



27 



rnoncxj 

HHH 



O 

-=j- 

H 



VDt-O\cv-i tA 
lf\ IA 



OO 
lf\ IA 



lACMOOOxaOCMVOt-VOVO 
\0 VO LfMA-d* 



IA| 

si 



r4 



a 



OOroKcnaO^O\O\VD 
CMCMCMCMCMCMHCVJCM 



O LA IAO ON H -4" 
CM H H CM H ro no 

CM rO O\ H C*--4" H 
CM CM H H H CM CM 



CQ 



e c3 a s s 



* M < 



S 9 i B 1 



sa- 



H ON CVJ HCO 

CM c\i ro on 



O\ IA 

33 



CM IA 

321 



S3S 






CD -P 
>a CO 
Q CD 



il 




g a 

1 1 



g a 

H (D 

+3 CQ 

03 0) 



420 



APPENDIX A 



03 CQ 

O 
H T3 



ft 



CQ CQ 
JJ 

K 



-p o 
ri ft-p ! 

OQ P i 



tQ 

co S3 

P O 

fl ft 

CQ 

d 



dnojf) 



, _ . . , ONF--OJ C*--4- 

VO ON3 1 ONOJ-4- H H O t-Jt ON 
H HHcrjHMDHH H 

VO onOJ IA 
VOCO t>*CU 



N OJ CO 1ACO ONON 

cu OJ ai cvj cu cu 

HHHHHH 



OO 

m on 

HH 



H H m 

HHH 



CU IAH mOJ CU OJ O IA ONonON 

" ~~ " n on on OJ on on OJ 



f333SS3t 



VQ m 

-H H 

CVJH(T)OJ OJ 



IAH 

ooaj 

OJH 



CM co H o o co I 

ajOMXAOlAV 
HHH 



foojtnoir\iAi>onmmcp mir-oj 0-0 o CM CM 

CO t-CVl lf\OVO C-OCO lf\ ON \JDVO 1AVO CAI^t OO^J; 
OJ CM -d- H H J- H C\J 00 H H H CVJ H 



H cu ?voncu ovS i 
H H H H H H H I 

IA t IACO co 8 on o 
ono^o H cu i on ON 

H H H H H i H*-^ 



! & 



I O VO O 

i on^o co 

I H H H 



ono 

OOJ-IA 
>-' H H 



d 



88 



OJOlOJHrnOcnHHCMO-d-O 

ooooooooooooo 



HCXIOJHHH 

oooooo 



l^< 



53 Si 



o IA cuco on 
on on H IA 

H H 



on cu o 
o o o 



ono on on cu 

o o o o o 



U VO O IA 



IA 



CT\CQ C 

333 



H CO m O ON CU H CO C- OJ 

333333333 SI 



VD O O\ CO 

333 S 



OJHCOCO 





t*- COCO 

f t*~ ^0 lA 



CO H CU O t~-cO ON O -* Jt O lA 
IAVOVD t^VO IAVO t^-V.0 t-C^-VO 

^HgNHpO^^tjCUVOp 



IA ON ON-4- -* H ON 
VO ^^O t>-VO IV 



aJOJOt>-OMAC>- 
CO C*-VO O- O- 



I t>00 0\KOCO\OfcQCO 

I OJ on on on on onRu on 
i H HHH H HIH H 



vo on on ITMA.CO 

ro-4- on on ir\on 

JHJH H H H H H H 



H 

.4- 

H 



cu on 

HH 



CO 

cu 

H 



COONOJCU 
cu cu OJ cu 



oj 

on 



mo 
oj on 



ojoj 

on.4- 



OJ ONCO O rOOJ H O V) VD on lA 

j* on ro on on^t on on on OJ cu on 



IAHHVOVOONH 

on cu on OJ on cu on 



vo ONH OMAO on-sf vo oncOjd-ococoo onHLAon 

CUCUCUHOJOJCUHCUHCUOJonOJOJCUOJonCUCU 



H CU on^j- (AMD t^GO ON O H CU on-^ LAVO t>CO ON O 



LA CO 

4- cu 

iH H 



IfSVO OJ 
IAC-- 



-5 M3 CO 00 



cu.vo 
on cu 

H H 



38 

_ O H O ON 

OJ on oncu H 



CQ 



id 

| <3j CO S 



H OJ on-4- ITS 

OJ CU CU OJ OJ 



"5 

5 l 

r^ rj 



CQ C- 

a: 



I & 
i? s 



a 

Q. 



H 

^ 
OQ 

H T3 

,b 3 54 
43 

< CQ -P 
* * 



APPENDIX A 



421 



O? 03 



ONOJCQ IAOQ ON ONO O 
HOJOJOJOJOJOJonon 
HHHHHHHHH 



-d-OJiAHonOJOJOJO 

ononononononononon 

HHHHHHHHH 



on on.4- UTNH 

H Ch on CO OJ 
H OJ H 



OJ ON.4-VQ 
t>lAI>-H 

OJ H onoj oj 
vD-rtco on onvo cu . 

VD ON.4- ON OJ -=j- H H O 
H HHonHVOHH 



cot-oji 



O- 
on 



LA 

3 



ON O 
H on 
HH 



ON on ON 
on on OJ 
HHH 



H 

-d- 



CM oo H o o co in 

OJ ON !>- !A O tA 



OLAiAr~onononooont s ~ojt > -- 
iAovococo LT\OVO VD ir\vo 

HHJ- H OJfOr-l 



H OJ 
OJ OJ 
H H 



33 



ONt-O 
ON-=J" 



o moj oo on 
t>- on on H ir\ 



IA 

LA I 

DT\ CQ 

H 



OOJOOOCOt- 
04 H I>-\O -4- O CU 



t-t-O on t-covo oco 
onocoojoj-3-onONOj 

HHHHHHH-~- H 



> rOV) 



VDCO on LA 

o -3" on VD 



SK 



H OJ OJ H 
OOOO 



g 



5H|w 



g 



H|OJV)HlCMH OjHlcviH 
O OO O 



O onnlcvjO 
OO O 



H H CVJ J- O OJ 
OOOOOO 



CU OJHJCMH OJHonHirNOJOjroonCUOJ-zfH HH|W 
OO OOOOOOOOOOOOOOO 



OJ OJ O 

o o o 

oj o on oj oj 
ooooo 



? 




5 



ONonONOJco ONO ono 
HOJOJOJOJOJononon 

ononLAO^t^t oionH 
ononononononononon 

HHHHHHHHH 



oo 
on 



H O ON 
H OJ OJ 
HHH 

ON J- ON 

332 



OJVD 

-^ on 

H H 



OVOCO H t-COVO O ... 

co co t-oo co co oo ONOO I 



O-=h ON OJ b--^- O 
ONCOCO ONOO t-co 



ir\-4- o o -=i- on^-vo j* H o IAOJ ONJ- onoo j- H H 
co co 00 oo oo ON. co oo co oo VO oo ON co oO oo oo oo oo co 



ONVD HO 

VD t*- ONO 



S 

&0 CD 



fl A 

CQ 



I>-VO J- fr- ON-=j- CVJ^t H 

H H oj onloj OJ oj olon 

HHHHHHHHH 



CO D-tAtA^-o 1 oniALr\ 

onojononojolonojoj 

HHHHHHHHH 



- 

on 

H 



H 

-4- 

H 



H HJ^ 

HHH 



COC-OJ 

onojon 



ITN 

cxj 



OJ 

on 

H 



e j- 







-co 

OJ OJ 
H H 



00 ON 



\o Ch 

HH 



OJVD 

oj on 



oj on oj on vo 

OJ OJ OJ OJ H 

O-=f OJ ONonOJ^t- OCQVQVOJ3- 
OJHOJHHHHHHOJOJOJ 



OMDHOOonVOO 
HHHOHOJOJ 



VO VO 1A 

oj oj on 



LAVD O O VO t>*VO jf VD ONO lACO H VD t- OJ VD ON ON 
HHOJOJHOHHHHJ-HOHHHHHHH 



ONOJ LA 
j o H H 



XIOTeTATCt 



'ATOQ 



dnoj;f) 



i^co ON o H oj on j- IAVO ooo ON o 

HHHHHHHHHHOJ 



H OJ on-4- LA 
OJ OJ OJ OJ 



422 



APPENDIX A 



0) ft 



& n 



I 

-P 



cd CQ 

S "^ 

O 
H TlJ 

ft 
G? 



a 1 

/ 



IS 

03 



ft-P, 



$$ 
g&^i 



O\ (M CQ \T\CQ O\ O\ O O 

Hojojc\jojcvjojor>fn 

HHHHHHHHH 



CM LA H on CM CM CM O 

ononononononononon 

HHHHHHHHH 



HO\O H 

HHOO c\j 

HHH H 



ON on ON H 
on on CM .4- 



on-* 
o\on 



.. o^vo on 

t-lAr~H H 

CM H on CM CM H < 

}^tcoononvocMOC i CMt 

) ON-4- ON CM -* H H O C*-- 
^ H H onHVO H H 



CM ON tr- IA O IAVO 



SI 



| 



^ O O t-CQV.0 O O 

^t o co H xxi vS on ON D- 

H H H H H H H '-'^-' 



-<]r\jCM CMH|fVj H 
O O O 



1(33 VO on CM IA O U , , ,. ,-,.-,..-.,. ~. 
lOVOCO C- CM LAOVO [-- O CO LA ON.VO VO IAVO < 
' ^t H CM on H i 



IAO OV) i O I O 

o lAonon ' IA i H 

HHHH'H'H 



i O CM 

i on^o 

I H H 



IA ^t\OMD 

o o-st-on 

H HHH 



O HjwCM 
O O O 



H| W g 



g H| W 



gj g 



CO IA t- 

cMCMCM 

HHH 



- CM 

CMononcn 

HHHH 



H 

H 



IAVQM) onmO fO H O 



pr>t- VO O\CO HCO 
IA1A IA VO LT\ IAon 

IA IA cn on i> o .4- CM ONCO oncooo vo CM t>iA 

j* VO IAVO VO IA IAVO VO VO LA IA IAVO VO VO VO VO VO 




ONHCOCO-=t'O 
VO I>-VO t>-VOVO 



t-VO t>-I>-COVDVO IA 



- H mtAo o o-onvo 
oncMmoncMCMCM 

HHHHHHH 



on OHCM 

H HH 



on-4-,4- i>-mon|o CMCO 

sg^gssbna 



CO CM HC 

CM Jt on ( 
H H H i 



3 



on 

H 



O O Cm 
- 



CM 

n^t- 



H OMfMf\ t~ t- on O VO CO H CM 



C-- CM CM ^r CO on IA 

3" on on on on on on 

C-O\CO onHOHO\CMCMVOOOJ-ONONO-4-onH 

oncMonononononCMonCM on-4- on on CM CM CM on on^ 



CM m^ IAVO t*-GQ ONO H CM on-d" LAVO t>CO OvO 
HHHHHHHHHHCM 



CM CM 



i 



O IACM 00 on 

c-onon H IA 

H H 



, 
CD 



|88 

HH|OJO CM CM 
O O O O 



IAon 
CM H 
H H 



,-, C-CV1 

SK on 



O onco 

VO IAIA 



lAt^ H O 
LAIA-4-VOVQ 



\O CM 

H on 

H H 



ONH 
H"3 



on iAon ON o 
S1- iAon j- 



H CM on-^ IA 
CvJ CM CM CM CM 



CD -P 
>> CQ 



C^ 




CQ f- 
H H 



a ft 
w 

s g 
*t 

ft 

O 

a Si 
g * 



w 

& 



I 



APPENDIX A 



423 



rl 



a 

TJ rd 



o 

f 9 

CD CtJ 

fi *H 

O ft 

a ft 

CD & 
CQ 

<D 

A A - 

o -P 



'X (D -P 

5 t>^ 



3 25 
8 tj 8 s 

H 3 ft 

Co rH 
h> O O 



I: 



P -p h 
o a) 
ffl -P fl 



1> 
5 



1-S 

^ a 




IA CD 
LA -P 
N"\ CD 



CH 
0) 

5 



1 a 
ll 

8 



424 



APPENDIX A 





g 




-a S 




CQ 




O 




bQO 




d 




^ 




H bQ 




J-l CD 




*P 




t c{ flj 




P O 









,Q O 




-P 




CD OJ 




^U 




ft 




P 




S ft 




CD ft 




CO td | 




O CD 

S3* 




H CQ 




S33 




CD -P 




!, 




-P M 




CD H 




H S -H 


*-N 


S5^ 


3 

o" 


Co Xi TJ 

o 8 a 

S ft o 


-=h 


9 H 


5 

a 


h-S 
CS& 







*j feS 




o 




> 




03 ft 




CD CD 




B rcJ 




H -3 CD 

p -p h 









CQ -P ,d 



CD 

ft S 

ft o 

13 CD 

<D ^3 

"P rCj 

CQ -P CD 

3* 

03 ^1 



5| 

fl W 



15 CD 

39-8 



,3 <D 

I 



i!3 

ov 



fc a 

<D H 



^Ili 

58- 



driojf) 



-d*CMlAHonCMCMCMO 

on onononononononon 

'HHHHHHHH 



-CooDD-CUlAO^Ob-OCOLrN ONVO'^0 lAVi) i 
H-d-CMCM-^-HH^ H OJonHi 



H 



CTs CJ CO 
HCMCM 



ITNCO C^ ON 
OJCMCMOJ 
HHHH 



O O 
OOon 



H ON O 
HHfO 
HHH 



IT\ ON oO ON 



t~lT\t-H 
CMH^nCJ CM 



m 

H 



ONOO 
H 



lf\H 
CO CM 
CUH 



OD cnmvo CM ONt-cvi !>-=* 

jd- 0\ CM ^t H H O fr--* ON 



CMCOHOOOOITN 

CM ON t- IA O IT\VD 
HHH 



-=h on-=l- 
HCMH 



OOOOO 
IACVJ !>VQ CM 
HH^-'HH 



OOOODO 



HOlA 



OJOOO l> 

333 3 



d 



H P" 
OO 



c\!H CMH|wOJHO-=t-HO 
OO OOOOOO 



838S8S8 



ggg g 



CMCMOOONLAOJVOOOON 

~" nnojcu ononononon 

HHHHHHHH 



OCOCO b-CM-^" b-CO O 

^t on on on-=h ^ on on on 

r^ H H HHH HHH 



ON 
CM 

H 



O-3- tA 
CM CM on 
HHH 



VOONIA ON 

-4-mon -4- 

HHH H 



IA OOVQ CU CU 
\O vo VO IS" iS- 



H VO ON ON\O CM ON oncO ON CM on O CM CU CU on^t C*- 
VOlAb-VOb-VOVOb-VOVOVOCO VOt t>-t>-IAtS-lA 

N LA O [*- IAVO -^OlAt>t>-O-*OoniAIAHI>-ON 
NVO b-VO h-lAC-VO IAIAIACOVO b- t>- 1- t> VO VO IA 



H HlH 



|O 

H 

IHIH 



w|cM|aJ"!H|blHrHlcMCM 

HHIH HIHlHlr ' ' 



on 
CM 
H 



O 

H 

H 



felSfii 



on ON CM cu 
^t- CM H on 



IA 
on 



cu^i- 
onon 



coco 
CMCM 



ONJ- H H.4-CQ H C^CU H 00 t- 

on3- cu on CM on on OJ m on on H 



O 00 CO 00 b-VO on O 
-d- OJ CU CU J- CM-4- " J 



H LA O on ITS J- ^OOlAonrOOVOOt-lALfNONonH 

- 



IAVQ t-CO ONO H CM on_=f IAVO OCQ ON O 
HHHHHHHHHHCM 



ONt>-O 

ONON^t 



O IA CM co on 
t~- on on H IA 



ONjtOO 
lAt>- LA 



ono H o cu 



O ONOOO 

CM on o on cu 



H cu on^t LA 

CU CM CM CM CM 



T3 -H 

CD -P 

>> CQ 

O CD 



<D 

. 5 

H 

O fl 

II 

CQ 

o 



+3 
CQ 



S ft 

bO CD 

?H O 

s 

CD N-" 

5 g 

^ 



% 13 

H CD 13 



APPENDIX A 



425 



a 

I 



3*3 



CVJ 



A 
B 



a 
s 



ng 



Students 
No Res 



u 
R 







P K 



dtio.r{) 



O\OJCO ITvOD CT\ChO O 



t 

a 



O\O 



4* cj IA H m cxj oj cvj o 
ononon cnonoo mmro 

HHHHHHHHH 



tA ONCOON 
cvj cncnoj 

H HHH 



H 

a 



H 
-=f 

H 



- 

irM>-H H 
CVIHC004 CVJ 

jD-3-ao oo mvo 

^0 O\^t 



ON CVI jf 



- 

OSfn 00 CVJ 
H CXIH 

i ONtr-oj o-=t 

H O t>--4- O\ 
HH H 



CVJCOHOOCOIA 

CVJ O 



IT\ O lf\VO 
H 

no oj ir\ o ifMAC^-moomco mt OJE OOCXJCM 
vocot>-cvj lAOvot-ocoirs o\vo vo invo cvi ^ cn^t- 

JtCVICVU-HH-=i- H CVJfOHHHCMH 



0*0 o o ir\coc 
ta ON-4- o -=f 



sDlAOCVJO 
4-OVOHJ" 
H H H H H 



HOO 



CVJ 

H 



cn i 



CUCOO 
H-4-^t 
HHH 



CVJ CVI CVfH|w 
O O O 



O O HlaiCVI 
O O O 



-i[CV!cr>H](\HlOJCVjH|cvH|aiO O-4- 
O O OO 



- 
OO 



HOJ 
OO 



IOCVIO 
OOO 



CUWHHH 
OOOOO 



c^cnHOJO-3-OCVJGJCvjHHH 
OOOOOOOOOOOOO 



00 -4" .4- O t> O- ON CVI 



4. HVO H H CVJ OOH ON 

m m cn CY^ en en on oo cvi 
HHHHHHHHH 



1A CO CVJ C- 

cvi on on cvi 



CVJ 

-=h 
H 




HO 

vo 



vo 



VD CO 

vo In 



HVD OOHCO ONO\VO 
t-VO \D I>-VO C-VO VO 



H 00 CVJ J- LA t- OJ 
t>- t- C- h- IT\ IAVO 



i>HVO IAH lAO\H.d-VO lAHlAOJCOGO OMACVJ 
t- t>- CVQ CO MD t>- t*-CO VO VO CO t-b- t-t-VQ 



H^t- on volH cvj OMACVJ 
cu H on on on on cxj cvj on 

HHH HIH HHHH 



cvi voc 
-* o 

H H 



vo 
i m 

H 



O onchvo cvj oscvi o 
ononmcvioooncvicv^cn 

HHHHHHHHH 



vo 

cvi 

H 



oir\o\ 
^j-mon 

HHH 



o\, 
c\j| 

H 



co 
on 

H 



CVJONCOVO 

m cvj cvj cvj 



IA 

on 



ONO 

on^t 



-=f CVJ 

on-4- 



4- ON o -4- CK-3- 1 o\ cvi H H -4- 
on H -4- on cvi on on cv on cvj on on 



o\ cvj covo LA onco _ 
oj cvi cvi cvi^-4- onpg 



- IAOMAH ONVO^h lAOMACQ CVJ 

ojcucvjcufnHrncMOJHfnonHajcvjcvjc 



H 1AOO 

fncvjoj 





Jj C 



H cvi 



- IAVO f>-oo O\o H cvi on^t ir\vo O-OD O\o 

HHHHHHHHHHCVJ 



o lAcvjco on 
t^ on m H IA 

H H 



388 



f>~VO O 
VO 1AIA 



H H 



-3- o 
.4- ir\ 

cvi co oo .4- 
-4- .4- mH cvi 



H oj m-=r 

OJ CVI CVI CM CJ 



LA 

IA 1 

K\ O 

(D fl 

u -d 



given on 
ire was 



the symbols employed 
hmen when the questio 



of the organization of this table 
oups (except 12, 13, and 17) were 



ati 
se 



lan 



03 

% -s . 

H T3 

h tf 

rQ p JH 

P 

< CO -P 

* J 



426 



APPENDIX A 



<D 

$ 



I 



3 

O 
,0 



<D 

! 



i 



I 



J- 0) 

a 

3 a! 

a 

fc 
0> ft 

* 

s 
g t 

o 



college 



C. Fear of academic 
D. Indecision about 
E. Homesickness 
following page) 



ture economic, 
ations 
of future work 
ries F, G, and R 



pj 43 CH 4^ CD 
CD 14 -Hp, 

HO Jn O K 



n over present 
al or political 
sion regarding t 
(C 



!5 

> 3 



S^ 

< CQ 



I'S 

<i PQ 



ing 



Students Mak 
Resse D 



Students M 
Resnse 



Students Making 
Response B 



Student 
Respo 



I O I i 
i H i i 

1 OJ 1 I 



O 
ro 

H 



O O 
tr\ o 



8888 *" 38 88 



888888888 



888*888888 



i i o o H i i o 
i i o\ oo- 



OOIO 

c\jivo 

,-j 1 H 



\ oo-sj- i i en 



PO'O' 1 

vJcr\iH>> 

<* ' 1 H 1 1 



o 
H 

OJ 



OIIT\ o 
O i -=r if\ 



OOl 

rnni 

HHI 



88d Him 



85 



r H H H r-f|H H 



HOHHlTNOOOfO 

-=t- CVj OJ on QJ 2\lJ" CM 



"ir\ 
O\ 
I v-r- ^ 

IT\ r-l-4"00 O. 

rn OJ o H 1 



O O iPvOJbO I OJ O O 
O CO -=f b-K^S 1 H CO I>- 



O 
H 

H 



ir\if\m i o 



8 d 8 8-3^ g S rt|w 8 8 



LTsO OJ O OOVOOO O 1 

o CVj -^ on cv-i on ir\ i 



3 



S 



s = = = = r c = = r = = 



) = 



mvo ooo o\ o H oj c 

HHH 



i>-oo ON o 

iHHHOJ 



888 
888 



o o 
o o 









8 



355 
ste 



page 



given 



<D -p 
>* m 

5 

& 

^ 

!! 

fl) 
P 9 



tab 



3 



12 



_ 

OJ OJ OJ OJ OJ 



II 

S 1 

17 



lanation 
these g 



# 
A brief 

Students 



APPENDIX A 



427 



I 



s 

M 




5 



O 
H 
-P 

CQ 


& 



-P 

M fi 

cT 

P! no 



I 



P 

cT 





fl H 



> a 

H -H 
O -P 



-P -H 
O CQ 



& CQ O -P 

<H -H CQ H 
HO -H H -P 3 
P J4 <H O Cd 

SO o "H J-i 

-P 05 H O 

OH O 03 O 

PS CtJ N^ JL, g 

CQ S H *-P 

1 J4 l! -P 



g -g 

H CO 



O i 
H H 

bD i 



i O W 



X 

h -H 



cd CQ 





a 1 



S 



dnojf) 



ON oncO-4-VQ co ON H O 
H OJ OJ OJ ojloj oj on on 

HHHHHHHHH 

PC} -=} OJ 1TN O lAJ- OJ O H 

pjononononononononon 



ON CM CO ITNOO ON ON O O 

Hcuajcviojojojmm 

HrHHHHpHHrHH 



H ON P 

Hr-i?n 

HHH 



monrnooonoocnoooo 

HHr-ir4HHHr-}H 



ir\ 
cvj 

.-1 



ONCO COCU 
H C\JH 



- 

N-4" ON OJ -4- H rH O t>--=t ON 
HHroHVOHH H 



JCOHOOCOLTN 
ON t~- IA O ITNVD 



C OO CXJOJ 



VQmCMlTNO irN 

\OCX3t-CVjLfNOVOI>-OOOtA ONV) VD ITNVO CM J- 

CM 



-=i-OJC\i-4-HH-=f 



- 
ooHHHOJH 



HHHH 



HHHH 



VO H CM.O H 

on H CM Ion CM 

H HHH H 



on ON on ON on, 
CM onon CM -=H 

H HHH H 



CM t cvj 
N ONCO ON 



on UTN 
co co 



LTNOO 
cb t>- 



. _ _, vo mvovo i>- oj-3- ITNVO c 
00 ONCO co CO oo co co ON ON t*-co c 

CMVOONOJOJCMO-d-OJOonOJOC 

ON ONCO ONCO ONON ON ON ONCO co ONC 



H vo -3- ONCO LTN 

ON ON ONCO CO "~ 
iTMTMfN on LTN OJ 

ON ON ON ON ON ON 



ot-oo 
onHCMOJ^tojonocM 

HHHHHHHHH 



o oo 
c HO 

*- ' HH 



i IAO O O CO IfN I 

i onvo irx^- co on i i 
i H H H H H H I I 



S 



g 

H H H CM 



- oncM-=fonHOVOH 
oooooooooooo 



CMIAOCMHOCM 
ooooooo 



8HOJCM HHJIMH HOOH-4-C\IC\JCMHCXJ 
00 OOOOOOOOOOOO 



8 I CO I ITNIAH b- I 

1 r^ I ^j,^^j,^j, 

o i on o ur\ i o i i 

H i H on CM i o " i 

H i H H H i H i I 



S 

H 



O t O ' 

H LT\I>-i 

H HHI 



Og 



H 



g g O g O O g 



HNg 



H, N g 







H g 



OQIAOONVOO 
ON on-sj- -=t -4- -* 
N^-HHHHH 



Rg^ ' ^ 



H 

onOOCMVO| 

H H I - 'H H 



O i IO O 

VO i i ON on " 



8 



CXI 
VOl 

H 



8 



OJ O Hr 

OOO 



28! 



H OJ 
O O 



HH]CVHJC\JO H H CM 
O OOOO 



H H CM UTN O HJCMO Hlw O 
OOOOO O O 



J O 
O 



^ 

bO 



H oj rnj- Lf\v> t>-co ONO H oj on^t ir\vo t>-oo ONO 

HHHHHHHHHHCM 



Jt VO 

V3- on 

H H 



ON t- O 

ON ON -3- 



O ITN CM CO on 

c on on H ir\ 

H H 



CO 



j co on on 
co co co co 



O 
on 



83888 



i 



888 



88 



g g 



oj on-* if\ 

CM CM OJ OJ 



5 

-P 

>* CQ 

O 

ft CT* 



3 

I 



I 



o 

H 
-P 



R 





H 



428 



APPENDIX A 



-p a 

!H -P 



<B h 

ss 







m tn 
oJ H -P 



5-? 



rf 



15 8 
^ 

CQ ^J- 



3 
aT> 

< 4J H" 



-P H 

CQ -p -p 
03 -P 



-P 
CO O 
ft 






a 



1 



& 



fife 



HO I SI Aid 



-|<MH|cuC\J H|*H OHOiHCVI^HCUOJ H|CVH|CUH O H Hj 
O OOOOOOOOOO OOO 



OJ O\-3" VQ no CO-=J" IT\H 
t U"\ fr-~ H H ON no 00 OJ 
OJ H ro OJ OJ H OJ H 



- 

^ l 

OJ 



04C 

oo^t 

OJH 



ACO a" i aT i 

V>OiPiroro-d- 



< 

Ch 



HJC\| H OJ 

O O 

H O O H H H 

O O O O O O 



|H.t*-,c?N|iA cnl 
9 9 



rol VDloO ml O\ 

'I ^ o ^ 

H| H H 



LT\ 

H HM o 

H 



j'VDjsJ-HVOt-CV.OOJO OJ OVOOO lf\ O O\OJ 
cnOJHajHCVICVJOj^troOJOJHHHOJHa 



OHHOSH V> 



O\ OJroO O\ 

SI ^33 3 



3 



si 81 



= = = 6 S = = 



* 



g g s s g a 



^ 
HHHHH 



COO\O 
HHOJ 



o ir\ cvj CXD co 






S3 

331 



is 



H OJ CM 



>> CQ 

3 

& 0< 



CQ 



P 

CQ t*~ 



3 

Isf 



APPENDIX A 



429 



!> O O 
H -P O 



* O 

00 O i 

^ -s 



H .d 

" 



on 
o 






03 CD 

fl -p 

O 

H T3 

-P C 

ra o 

ft 



|j 



8 



g g 

& 



a 



s S 

$ 

& 

nzJ 

-P 
CO 



1 I 



I 



UOTSfAja 



dnojcf 



ONOJCO JACO ONONO O 
HojojojojojoJonrn 



H ON O 
HHH 



IA 
OJ 



a 



f 2! 

vo -4- co on oovo oj ON t* oj c^-4- 

VO CN.4- ON CU -4- H H O C-4- ON 
H H H m H VO H H H 




CvJ C 

3' 



! O O CO LA 
- LA O LA' 
I H 



vo oooj LTNO ir\ir\! 



-oooorocX) rol>-c\j !>-OO OJ oj 
o\vo vo mvo oj- 

OJfOrHr-i 



-4-ooDVDOjoorocor- 

OOiHHHCOiHOJ^-r-J 

HHlHiHHHHrHlH 



ON ON O\VD CX) 
~d"OroOJ-^- 

Hr-JrHHH 



CO H 
GJOJ 

rHH 



O OJ 
-=i-ro 

r-JH 



ir\ 

H 
rH 



O 

H 

r-l 



coOr 

O\C\r 



- CM 
noHOJ 

H 



I 

B 



h VO - 
O 



lAVQ 
OO 



S 



^S'S'SS^'S&S^^ 88!S8 

jjgg 



,4-^t- 00-4- mOJVOHIlAt-- OJOJ LA-4- HI -4" iH 

ooooooooooooooooo 



c\j o\ mjco H 
on-4- ro flrn m 

H HIH H H H HIiH H 



rol 



800 o 
oo on 

H H 



o\moj t*- 
on rn on ^d- 

H H H 7-1 



VO -4- -4" 
OOO 



8S 1 

! OJ v 



in on t- ON oj vo on oj oj vo o o -4- -4- t-- oj oo LAOO 

OOOOOOOOOOLAH 00000 00 

CO t ONVO OO-4- OJIAO tonOOJOJVOO-4-CO ONOO 
HHHHOHOOHOOJononOJOJOJOJHHO 



H oj 

HH 



OJ t oj H OJ.CK 
oj OJ on -4- IA on 

HHHHHH 



H on on oj o H lAlcvi co 
rnononon-=t'onJ-Uj-OJ 
H H H H H H HIH H 



H 
cu 
H 



IW 

ONIAON 

on on OJ 
HHH 



ON 
o 

H 



816 



-4-S 1 IA-4- on OJ 04 



LAONlAtOOO C H LAH LAO 

ononononHOJHHonono^t 

H ON CO OOVOVO OJVO^- H ON ON 00 CO OJ VOCO ONLAOJ 



oj on oo -4- OJ IAON.O on 
cv| 04 on oj on on on on OJ 
HHHHHHHHH 

OJ-4- on ON^d-^4- D-H t>- 



I 
I 



OOVO LA 



VO 

a 



vo IA IAVO 

OJOJOJOJ 



vo 



c\J oo 

OJOJ 



ON on 

OJOJ 



t>- IA-=1" -4" LA on LA H H ON 

ojonojojHOJHCU 



VO t CO ON on VO H CO -4* 

onHojoj ojojojojoncuon 



H t- IA-4- HVOOJO LALAHONHt>-OJ ONOO^-OD O 
OJHHHHOJHonOJOJOJOHHHHHHHOJ 



onvo -4- vo vo H onl-4", 
ojoojojcyojojloj 

HHHHHHHlH 



- H LA on H o 
on on OJ on -n 

HHHHH 



HOon 

HHH 



on 
oj 

H 



OJ 

21 
S' 



ooonHon 
ojcnoj 



oniA 
-4-oj 



HOJ 
ojoj 



VOONIAHON!>~OHO ONOOH 
OJHOJOJVOOOVOVOOOOOHOJ 

VO on LA OJ 00 OJ ONOO VO ONVO O LAOO CO H ON LAVQ I 



ft^^SS' - 

_ . . _ . OJ ONOO VO ONVO O LAOO CO H ON LAVQ f- 

HHHHCOJVOonHHOHOOOHOHHOJ 



c = = = r = = Er = = = fx] r s r < pq c 



H OJ on-4" IAVO I>-CO ON O H OJ m.4" IAVO b-OO ON O 
HHHHHHHHHHOJ 



-4- VO 

-4- on 

H H 



a: 



ON 
ON 



t>-O 

ON-4- 



o LA ojco on 
t on on H LA 



83S 



a: 



t oj o onvo 

H O OJ OJ 



CO VO O 
IAH H 

onCO OJ LA H 
LAH H -4- LA 



ONO 
H 00 
H H 



t-VO OJ LA 
O LAl>-r-i H 



<= = fxj 



< CO 



H OJ 00-4- LA 
OJ OJ OJ OJ 



^ 

& 



I 



H 
13 



-P 
H 



430 



APPENDIX A 



o. 
H <D 



-P *v v 

oj H oof"^ 

<^j -s_xv_^vjO 

3 4J O IA 

& 3~ 

Tj O (D 

tJ 53 oa tiD 

S fe A 5 



fl 1-1 -rl 

3 8* 

la -S8 

"h 



<H 

^s 



2 -s-$ 

*n 

4^ S S 



<c| pq O 



w 

3 

43 







NONOO 



ON 



4- CVJ IA H CO OJ CM C\J O 
oo oo oo on oo en ro on co 
HHHHHHHHH 



<O-4- co m mvo OJ ON t CM t- 
Co o\,SF ON OJ j* H H o b~-=f 

r4rHl H 



H r-i H OOHVO 
<O 00 OJ IT\O tTMA 

vg cp - oj ir\ 



H OJ H OJ 
0000 



t 

^ 



H ON O 

H H 2 



IT\ ONOOON 
OJ en on OJ 

H HHH 



H 

-=* 



CO 



00~=1- IT\H 



s op H o o an ir\ 
o\ t*- ir\ o irSvo 



C30 t-CVJ lAOVO 
- CVJ CVI^sf H ri^- 



b-oncnoo 
t-OOO 



c30 cnt-OJt 
ONVOVO IAVD 



O OJ OJ 

JJ- OO^h 

HCVJH 



lTMTxa 



onCVJ 



O 



C0^ 
H 



O 

H 
H 



lf\ 
LT\ 
H 



H 

o 



O 04 IT\O4 

oo oo 



HVO too HOD vo oj 



' Q VO I 
> OJ OJ i 



ON 

^ 



^ 



.=!- l>- IA H 



H IT\ 



U \\*J U \ 1 I 

dda a 1 

CO ONH CO 

' " " a 



HC30 
OJ CVJ 



- . r-JlAOJt-HVOH 

* OOOOCOOJOJCOOJ^OOOOHOO 



roroooojoocucn 



ooooc>iOJoocoOJ 



VO 
lA 



OJ-= COCO ONONOJ 
CO CO CO OJ CO CO OJ OJ"OO CO'CO CO 



" t-O 



OJJ 

si 



op 



OJ 1AOO OJ(BO CO^t O 00 



rf o?3 SS 

t iKS -=J- ^t oo 

H O CO tAOO ON OJ O OO COOO OJC7NQOJOJOJONOOON 

~ .vo ^t vo o mHojj-^tiA ir\ -d- IAVO vo ir\^ -^ 



OMAH CVJ-=t-^l- IfNONCO lf\ 
- oo IA HI oo H H co ooVQ oo 



H OJ oo^t LTNVO t>-00 ONO H OJ oo^t IfNVO C-~ CO ONO 
HHHr-JHHHHlHHOJ 



o ifsojcq oo 

C OO OO r-j IA 



i 



H OJ 
O O O 



> ir\ 

la 

d'lfN 



VO ONO 

o ir\ r~~ 

O'^PSS? 



1 



r = 



L *" 

i! 



H OJ 
OJ OJ 



oo^t lf\ 



ir\ <D 
ir\ -P 

KN. CQ 









2 







H S 







P "v 

s a 



g & 

P CQ 



5 



APPENDIX A 



431 



*z pp 



in 
-* 



fl ft 

CQ 
r^J 



id CD 

l" 



CG 
CQ H 

"S ft 

(D CQ 

nd 
CQ 



W) 

.3 



CO UMAOJ OJ 
HOJOJcnOJ 



OVQCOH t in ON H O HCO t>-ON-* oOCOj*CO OJ GT\O LA-* CO 
OJOJOJooOJenOJinHHHHHHenCJenHOjOJHHHO 



i S oj 



UOf STATCI 



A-run 



OJ vo en 
en on oo 

HHH 



O\ 
OJO 



CVI O 

on oo 

H H 



ON O C 
H OO 
^HI 



OOOJ LA 

333 



H ON 
CVJH 
HH 



IT\ OJ' t- O 
OJ Oj rO ro 
HHH H 



H OJ LA ON 
OOOOOJ OJ 
HHH H 



OJ t*- OOO OJ 
VQ ^ro^" H 
IALAQ OJ 



1 



& 
3&ft&S38N8 



00-* 
H 00 
CM H 



en-* in 

ONt-CO 

en oj 



vo ONH 

H LAOJ 
OJ H H 



o 

LTN 



\o 

OJ 



cx)O(nLAcnajaf 
HJ- tr\o O-4- ooi 

HH HHHCVJr 



-i>- 

C\J 



jco 

<M 



CVJIACVI D- 

en on PO o 

HHH H 



-OO -*-d-C7\ 

n oj cu m cu 

H HHH 



-* VO 

en oo 
H H 



OO 

3< 



O OO-* 
en en en 

HHH 



-=i- IA on H n 

oj H ojojj- 

HH HHH 



O en O t .* VQ 
-* en oo-*|oo OJ 
H H HHH H 



_*VO 
H H 



LA ON CM 

oj oj en 



VD-* 00-^- O OOCO H en-4- 
HHOJenOJHOJ en-=f H 



OJ OJ oO 
OOH H 

O I>-0-* C-0 
OOOJ H H O-* 



i>-cr\Lr\ 
OJ OJOJ 



o 
H 
H 



oj 
H 
H 



ojjoocxjl en 

HHH H 



oj H 

cv|-=t 
HH 



O-=J- 
oo oo 
HH 



CJNCO ooco ir\ 
HH enHen 
HH HHH 



en 
oo 
H 



O\HO 

OJ-d-j- 
HHH 



COCO -=fHO\ Lf\ 
^t oo oornOJ OJ 
HH HHH H 



H 00 VO 
-* oo en 



CT\ 
OJ 



LP\ ON ON 
OOOJ OJ 



ooVD en O H 
ojcncnoo-4- 



o\ oo if\ co o\ oo co en c ON 
OJ oo en oj oj ^t- oj on oj cu 



t- oj H co en en-* VO^DOOO oooo D-OJ^-O^J-CO IACO o o vo 
OJ oooooj oo^-oooj OJ ooooenenenoj cnooen-3-ojoj H oooj CM 



VD 

H 

H 



O OJOO ON 

oo oo en oo 
HHH H 



IACO 
OO OJ 
H H 



ITNO OJ 

Hoocn 

HHH 



H VO .* 
OO OJ 00 
HHH 



LT\ OO 

HCU 

HH 



OJ CO 

ojen 

HH 



ON H VO 00 OO 
.* 00 OJ 00 OJ 
H H HHH 



\o t> 
oOJOJ 



CONOIAH 
ro H rn rn rn 



H vo co co o ON ONVD ir\ oj 
OJooOJOJ-* 



en -4-ooOjOJ^t 



CO^OVOLAO 
OjooooooooOJ 



OOCOJ-ONOOOH ON-=t- 

* 



CO 

oj 



I>- O OJ OOVO ON ON H CO H OO t- 

- 



ON oj o O 

OJOJOJ H 

HHH H 



H tr\ 

enOJ 
HH 



ON ON CM 

HOJH 
HHH 



ir\iA -* 

cnH CV1 
HH H 



co 

JOO 



oj 

00 
H 



, 

LTN 
H 



t>-O 
OJ O 



OOC 

OO 

H 



j* OJVO 
OHO 



OOHOJOJXO 
OJOJHHO 



in 

O 



. - 

HHHHHHOH 



2fiS 
rf^^^^l 



oo m oovo o cxj LTN cu ir\ oo rnvo en ON o cp ONVO -d- en-* ON-* -* oo 

HHOHOOOHHHHHOHOOHHHOJOJOJOOH 




P O 0) O 

CQ -P O *H 

CQ CD fj *H -P 

H d CD > CQ 

O< H CD 

fd CQ <o 3 

0) CD 2 00 O 1 



432 



APPENDIX A 



p O ft 



<! PQ 



CQ 

0) Pk 

-1 



rcJ 
^ 
CQ 



!S^ : 



! lfe 



00 C--C 

cvj on 



^O co! 
_ O -P 

fi H g 



-P 
CQ 



SJ: 





OJ v> on ON 
on on on on 

H H H H 



l&SJSSi 8 



IAIAO CM r*-ao o on 
on tA on^ H ^t ITN 



t- t O OJ 
CVJ H CM D- H 
r-i H H H H 



on f>-on 
on onon 

H H i-l 



Oj 

on 



O O 
CVJ 



H OJ H OJ 



33 S 

H H H 



OJ 

on 



cu onlcV S 

H H H H 



CO 00 ,00 IA 00 
OJ^J- CVJ -d- -=f 
H H H H H 



C-CO C-OJ C- 
H OJ H OJ H 



& 



O\QO OCO -_ 
HCUOJCVJHOJHOJ 



OJ CM OJ 

on on on 
1 H H 



3 

H 
CO 

3 



O> ON O IA J>- on 

onon c-vo t> oo 



I s - oj irxfo-* j- co ir\ 

OJ CVJ H OJ H OJ 



G 

H 



estio 
pond 



Qu 



ng 




CQ 
CQ P 

*a 

CQ 


? K 

CQ 



ud 
R 



i g 



!H 
I 



00 CQJ 
S3"!' 

& TJ 

20 y 
PM -P { 
fr CQ : 








! if 

a 1 54 T3 

P 

fc PW -p fe 

fi CQ S 



85i 
M 
SfSB 
**S! 



o o CQ ! 

s i 

h <a 
**l! 



on 
H 



vo on os 
oo on on 
H H H 



ITNIAO OJ t>-00 O on 

en if\ mvo H -4- iA 

H H CVI H H 



O.-4- ro 
OJ O-4- 

H H H 



O 

.4* 



VD t- C OS CVJ -=f 
OOOOOO 



S3&BS8&8 



H 

3 



vo ON on oj ON 

O O H O 



is 3 



ON OJ 

oj o 

r-i rH 



CD 



01 in en.* Jt co ir\ 
W OJ H CVJ H W 



CQ 

5 



,0 





-P +5 
<S fl 
O 



gj -p 

A ^ 54 , 

CJ -H fl 

O H 



Rl t30 
o as 

CD PJ 
g-H g 



q? o ^ 



CQ tJ . x 
H O rl CO 
O O Q d 



!0 
-P 

0) 



: 
,0 



P H -P 
O pj cos 

CQ O J-j 



O -P cfl 

P! * O 
O A ^ 



CO CQ 

S P *d ' 
^^^ 



:f*-> o 
s on 
^ o ITS 



OJ 

OJ 



APPENDIX A 



433 



OJ 

j, 
OJ 
IT\ 

H 
&1 



% 



W 



2 

H 

03 CQ 

O 

i-i n^ 

-P fl 

CQ O 

CD ft 



I 



CQ 
P 

fl - 

CD tQ 
T? CD 



ONO 
CM on 

H H 

OJ O 

as 



si 



^-vooo r ON o t>-onHoni>-H iA_=t vo H H 

rHHOjHOHOJHOJOJHOJHOJHHH 







iE 



O\OOD 



c\|iH 

rH 1-4 



o 
on 



HOJir\ o\ 
mcocu cu 

r-5Hr-f H 



S3 



- ,-^ IT\ OJ VO ON H 

ON coo t H ir\ oj 

on OJ OJ OJ H H 



ir\ooc\i ojcu i>-c\jcno 
o o^t on 



HHHOJrn 



O-ITM^V 

OJ 



GJ oo 

-3-CU 



oo rot o 
ONOOVQ oj 



ITN.O 

on on 

H H 

ON co 
on on 

H H 



O C H _ 

on on on H H 

HHH H H 



ON t IT\ 

133 3 



-3- O CO _ . , _ . . __ 

c\i oo on ir\ on ojonH 

HHH HH HHH 



H 
OJ 



CM CO ^t 
OICXJ01 



VD 
OJ 



H OJ ITN O ON IT\ ON ITNVO Lf\ 
CVJOJOJHOHOJCVJmOJ 



LT\0 O 
H H on 

o-on t^H ^ oj 

H OJ OJ H O H 



oo OJ 
cxion 

HH 



HO 

ooon 

HH 



O 
on 

H 



co 
H 

H 



on ir\vo,c\ o" 
OJ oj onpt m 

H HHH H 



ojojoj onon 

HHH 



O 

on 

HH HHH 



OM>- 

ojH 



\O 

oj 

H 



^-S^^^SSJRS^ 



oir\OC7\ ON^- co 
-^ H OJ O H O O 



ON LA ON oj \o oj i 



21 



OOO I>~O\ 



LA O H oj 

on^d" on on 



IAVO 

HH 



ON OJ OJ OJ 
OJ H 



OJ on 

H"HOJrOHH onOJ"HO 
H H H on 



-=|- O IT\ 
-^"OJO 



OJooHO 
cnOJOJH 



O CO ... -. ,, 

on o o H H on 

H t <r^\o OJ oo ir\ 
OJ OJ H H H H OJ 



O LTv 

33 



OJ H 

33 



OJ H O VO OJ 

H oo on OJ H 



3: 



on oj ON ONCO 
on-if on ^t on 
HHH H H 



OJ CO CO H 

3'33 3 

o on oj -=f 

333 a 




VsO o OJ vo vo ir\vo on on-=t- 
oj-=t en en ^t on -st- -=*-* 



- 
ononononv 



3 S S o S 



8 



> ONO o\ ir\co 

OJHH 



HOJ 



oJ 

H 



! o fl c 

O *H -H 
CD -P 03 



uu 

,3 



51 



434 



APPENDIX A 



-P 

03 

I 



+> 

CQ 

I 



=H 

3 



I 



fi ^ 



CD f-4 O <in 

mi 430 

d co .rj a) 

co fl OQ -P 

00 -H OQ 

^ $ o f 

O <v-f cd 

O +5 

a -p -P 

' " -H 'H 

I -P -P 



I 



I 






IS 



;g 



ft 



., OJlAHonOJCVJCMO 

saassas^s 



IA 

3 



o\ooO\ H 



3 

<M 



CM H 00 CM 

H H H on H V> H i 



fi?> 

H 



H Og gg 



g g g g 



g H g 



g S-* 



I i I i O IA ' 

i i i i co i ON i ' 

I i i t HI i %_*, i i 

i i i o IA i cn ' * 

i i i -3- ON i m i i 

l l I 7-1 ' I H I 



8888 8 88 88 
8888^8^88888 



888888 
88888 



8^8 



ro ' i Q IACO lA O O 

H i 00 cnt-oo i m o 

i H H r-J^-' i H rH 

IAOO i on m IA o 

r-4 H l H H i H H 



H|(M 



g MHH g g-*g g^ g g 



W V- VJ *J VJ 

88d-*S8 




ON 

3 



^ ^ oj 

OJ IACO* O*. 

3^3 ' 



lAroiACUoocnfOOJco 



3 



co ON ON "ON ON ON ON cooo 

j\ ONCO ON ONOO ON ONCO t*-oo ON "ON ON o\co oo oo 
ONCAONONONONONONCOOO C^ONCKONONONONONONON 



- tAVD l>-00 ONO H OJ OOJ- lAVO t 00 ONO 
HHr-lHHHIHHHHG] 



33 



o IACVJCO cn 
t>- on m H IA 



838 
83888 



88 
s-^SSS 



3~3 



I3sao "&\ UN ON 



'S **"* 

>> CQ 

5 



S 



1 3" 

rl 

S -P 

S> g 1 



o 

H 
-P 



4S 



**$" 



APPENDIX A 



435 



oo 
P 

}> 

CM 
UTN 


H 

I 



O -H 

*H P 

-P Cd 

tf y 



ClQ 

O 

ft H 

O O 



P O 
O 43 



CO 

BE 



H 



O 



CM 

in 



p 
fl 



o 

*H ft 

1 S 



H 
ft H 




o o 





8P- 

ft 



4J 
fltJ 

TO ^H 



H O 

Tf 

33 

O 



fl 43 P 

ft ft 
000 
CQ 4^ 
ft O 4> 
ft H 

S"j 



cd CQ 

CQ 

-s CQ 

O h CQ 

p h 

PQ O O 



H -H 

CQ 4> C8 

CQ -P 

H *H JH 

J4 M 
000 

9H ^ 
POO 
CO 

O fl >* 

^^-H S 

4> *3 

03 O 



a* 

r-i 



P fl 

CQ O 

g* 

C? 



I M 

SH 
CQ 

42 O 

fl ft 

CQ 

T3 

w 



j? 



o 



5 
8, 






PC, 







IR 

o 
ft 


Pk 



Aftra 



o\o 

CVJ 00 
H 



O\OOD 

H 00 CJ 
HHH 



a\ 

H 



OJ 
Hr 



o 

CO 



CVJ 



VO t- LPv 

j* o vo 



. .. . , LTN CM 

ON c co t>- 
00 CM CM 

ONCO I>-ONVO 



rH LA CVJ 
OJrHH 



CM rH 



R 



iPvonOJ CM CVJ I s CVJ ooO 
O O-=h ooiTNt t"-ir\t>- 
HrHCVJon CVJ 



rot^-O 
VO 
H 



cOVO OJ 
H 



CM 

,g 

2s 



ir\OCO 
00 LT\ OJ 
rHHH 



Cvl CO 
CVJ or> 
HH 



, 

i-lOJ 
HH 



OOrOCVJ 
rHiHH 



oo CM VO 
H OHO 

H OO OOVQ VO b- t 
H O O O O O O 



O 

2J& 



000 

&38S 



cvj a\cp on ir\ iTN-4- ON t vo ITNMD cvj co vo H o 
ooooooooooooooooo 



H H 

"?0 



HOOJ 




33 

O 00 
*sfr OO 



vd CM 



cnHLTN 



^ 

s 



ir\co CM tncT\cp 
oc3oooc5 



t~- ir\ vo VD t>- oo ir\ o 
oo OOOOHH 



''? 



cp"oo" 
CM ro 



r-ioc 
oo vo CM 
HHH 



ocvi' 
J?-d*. en 

HHH 



o\ 
oo 
H 



L CQ 

i-p 









a a 



8 s 



s 



VO ON p O\ IT\CO 00 H H H 00^*" CXI CVJ O VO t^- 

CMHHI HcM H HHHH 



tJ -P "s 

P fl 

CQ XI 8 O 

H ^ -H 

!> -x 43 

T3 CQ OS 



IT\ , ft 

LfN Q P ^~ 

42 ffl ,Q TJ 

CD -H cq oJ 

tiO ft 

- - ft UO * 




Td 



S>> o3 ^J 

O o o JH 

H -H 

ft ^ 43 O 

S CQ 3C - .H 

fl {> 45 

O fl O CQ 

Cd ftH -H 

H CQ ,Cj t> p 

o ft 4= ft a* 

M TJf O CQ 

ft CM . fl 



CQ ^Cj O 
4^ 



3 

-P 

OQ 

I 






I 



at ' 



4J 4> O 0? fl 

- "P , M J-l * 

nfl[ <u fl fl o 

00 J3 

" ^ ) cq CQ 

o 



ft to 



H O H 4n 
CQ S H H 
co bo O 



H 

O & s43 -v 

o CQ 

p 



CM 



o 

H H 



J3 ^ 
^ 
CQ tQ 



< CQ CO -N 

11^ 
'" gjj 



CVJ 



H O CM CQ 

CQ^ ^ &f 

En fl O J-i 

1 ^ & 

d 'S -9 

H -H ncJ 

<H CQ 

-H J-i H Q 




436 



APPEITOIZ Bl 
AMALYSIS OF COVARIAHCE EROCEDURE 

The design and general theory of the analysis of covariance pro- 
cedure used in this study have been described by Gulliksen and Wilks (42.). 
The specific operational procedures employed in making the calculations 
were developed by Dr. Ledyard R Tucker and Mrs, Judith E. E. Aronson. 
These procedures are presented here because it is believed that they will 
prove to be useful in further applications of the procedure to the many 
analogous problems which arise in psychological and educational research,, 

Notations In the discussion which follows, the following notation will 
be employed: 

k = number of groups 

h = number of predictor variables 

I = criterion variable 

X = predictor variable 

CL, Y = covariance between X and X 

CL. = variance of X. 

b. = regression weight 

t = square of the standard error of estimate 

General Procedure for Determining the Variance Error of Estimate . In carry- 
ing out the analysis, it is necessary to determine the variances and covari- 
ances for all pertinent variables in each group separately. The regression 
coefficient for each predictor is also determined for each group separately. 
Four decimal places are carried in all recorded entries. 

In order to solve for t, the variance error of estimate, it is neces- 
sary to compute g. in each of the following equations; 

g 1 = b C y Y + b 9 C Y + * o o + b. 

JL JL A- A- c. A~A,~ 

g 2 = b.^ z + b 2 C x x + . . . . + b. 
g = TxjCg x + bgC^ x + . . . . + b h C. 



g = b n C y _ + b 9 C T _ +....+ b, C T Y 
n l A_^ A, d A^A, h X, X, 



APPENDIX Bl 437 



for the criterion, 



The values so obtained may be used to determine a value j, 
where : 



Then, t = C_ - This is the value of the square of the standard 
- Li J 

error of estimate for the group, that is, the variance error of estimate . 
The multiple correlation coefficient may of course be determined from the 
same constants, as follows; 

__ 

% / 1\ 

"T ( C ) Tllis va ^ ue as routinely computedo 

\yy/ 

Test lag hypothesis A The test of Hypothesis A depends essentially upon 
the ratio of the variance error of estimate in each subgroup to the weighted 
mean of the variance errors of estimates, where each variance is weighted 
according to the number of cases in the subgroup. The ratio for each sub- 
group Is expressed as a natural logarithm, the result is multiplied by the 
number of cases in the subgroup, the sum Is calculated, and the sign is re- 
versed. The resulting value is distributed according to the>? distribu- 
tion with KXL degrees of freedom, If N Is reasonably large. In the actual 
calculations, the weighted sum of the ratios of the individual group vari- 
ances to the weighted mean variance was computed, as a check step This 
value should equal the total number of cases . 

Testing Hypothesis Bo The test of Hypothesis B requires the calculation 
of the variance error of estimate for the total group on the assumption 
that the mean of each variable is the same in all subgroups This error 
of estimate may readily be obtained by determining the weighted average 
of the variances and covar lances In the various groups, and then proceed- 
ing to compute the variance error of estimate in the manner described 
above This variance error of estimate is designated asT,, 

The test of Hypothesis B depends essentially on the ratio of T to the 
weighted average of the variance errors of estimate in the various sub- 
groups, which was computed in connection with the test of Hypothesis A, 
In making the significance test, the ratio of these values was computed, 
the result was expressed as a natural logarithm, the resulting logarithm 
was multiplied by the number of cases, and the sign of the result was re- 
versed. The value so obtained, Gg, is distributed as"X? with (K-l)h 
degrees of freedom when N Is reasonably large 



438 APPENDIX Bi 



Testing Bypothesis C. The testing of Hypothesis C retires tlie calcula- 
tion of the variance error of estimate for the total group from the raw 
scores 7 without regard to subgroup means and standard deviations. Once 
the variances and coyariances for the total group have been obtained in 
this way, the variance error of estimate may "be computed in the usual 
jjay. The variance error of estimate for the total group is designated as 



*& ~^s 

The test of Hypothesis Q depends on the ratio of t to t, -which was com- 
puted in connection with the testing of Hypothesis B. In making the sig- 
nificance test, the ratio of these values was computed, the result was ex- 
pressed as a natural logarithm, the resulting logarithm was multiplied 
"by the number of cases f and the sign of the result was reversed. The 
value so obtained, Or c , is distributed as "X with K~l degrees of freedom 

Note on Natural Logarithms , In the calculations involved in the analysis 
of covariance, the use of natural logarithms is required. The following 
procedure was employed in determining these value ss first, the common 
logarithm (logarithm to the base 10) was determined, using the tables of 
the Handbook of Chemistry and Physics , which -gives seven~place mantissas 
for 5~disi"k numbers between 1 and 2 without interpolation, For numbers 
which did not begin with 1, five digits were used, by linear interpola- 
tion in the fifth digit, to obtain five-place mantissas The common 
logarithm so obtained was multiplied by 2,30258509 to obtain the natural 
logarithm. 

Example. There follows an example of a two-predictor problem completely 
worked out This example also illustrates the form used in the calcula* 
tions. In this example, the measures are expressed in the transmuted 
units used in the original calculations of the variances and covariances, 
since a linear transformation of all scores for a particular variable will 
not affect the outcome of the analysis of covariance tests 



APPENDIX B1 



439 



GROUP: 



ANALYSIS OF COVARIANCE 



PROJECT; 
DATE; 



-( 



CRITERION (Y) r*<<L^-(^ 
PREDICTORS (X 1 ) jS/ff 7~-~ }/ 

OU 






SAMPLE 1 



SAMPLE 2 >^ 

IU3TERENCE FOR DATA: 
forms. 

GENERAL INFORMATION: 



SAMPLE 3 



SAMPLE 



Accompanying intercorrelation and multiple correlation 



a ) 
b) 



c) 



_ = t error of variance of prediction (generalized formula). 
n 

Covariances to be used in solving for t, IT, and t, to be obtained 
from accompanying intercorrelation forms "where the C entries are, 
respectively, computed for each sample, computed from the weighted 
averages of the C entries for the sample, and computed after total- 
ling sums and n's over the samples. 

t, V, and t to be solved for by using accompanying multiple corre- 
lation forms "where covariances (above) are used instead of corre- 
lation coefficients. Then: 

1) For each predictor computes 



etc. 
2) For the criterion compute: 



3) Compute: 



k) Compute: 2 

Sy 

t = C YY " T 
d) Notation: 

C = sample designation 

h = number of independent variables (predictors) 
k = number of samples 



440 



APPENDIX ii 



HSPOTHESIS A; EQDAHTI OF ERRORS OF ESTIMATE 



Sample^) 



53J 



69+ 



/a,/, 



Degrees of Freedom 

k ' tec 

A oc * "^^ "h 

= JL. OQ79 



HYPOTHESIS B: PARAIIEL HEGRESSIOT HABES 



4- 999 r 

"t 



In 



p * > , 5tf < . 70 



H 


1 


2 


3 


2 


1 


2 


3 


3 


2 


fc 


6 


k 


3 


6 


9 



Degrees of freedom 

G B = -N In -~- 



HYPOTHESIS C: EQUAUTT OF INTERCEPTS 



t 



. 739/ 



,9990 



Degrees of Freedom 



,30 



APPENDIX B1 



441 



ANALYSIS OF COVARIANCE 
(Supplementary Form) 





^*^ \^rwtMLAS /^? ix_*ix^u ^iujeuui 
MOITIPEE CCBBEIATIOW COMPDTIHG FORM 


ojai/e 

Formulas 
J*y _ 3"^. .S*"^ ^"X 


Predictor Variables (X) 


Criterion .Variable (T) 


Off*. 




2 .S/4T-A) 


j 


' :$ 


^ ' \ ' >J 7 L 77'' 


3 




Cw 

> D-*^ " I t \? * & *c-"O f 




H= S-.3/ 






\ R TABLE (*) 


3 


\ 


x, 


X x 


X 3 


Y 


s 


?7.o^?y 


/y./^/ 




^.7^7 


/5-X. W/rf 


01, ^.7^5-^ 


x* 


/?. /**/ 


^9. 7^^ 




J3.5^0 


/0/.J2<<^ 


J 


^ 3 












J* 


.Y 


+A.7*'? 


J-5-.5^0 




/4*7./^5- 


2 A7.i5-/A 


3 ^3.5-^9 


2 


/S-/- ?6/6 


/^/.^.^s- 




^2-7-^5V-3 


52,0.^^5-3 


ftrf //J.3/7 


j 




^S,^,.^,. 


JJ // 3. ^ 3X 7 


3^,^^/7 






\ A TABLE 


\ 


X, 


Xa. 


X 3 


Y 


CK 


s 


9..J506 


/. f^ J 




^V 


>* aW 


,*** 




^, 1} & 


? , /^7 




^W 


/,^ 


,,,*^ 
















b 










jj,/ n ^ 


35-7P 





Covarlancee 



442 



APPENDIX il 



ANALYSIS J COYAEIANCE 
(Supplementary Form) 



Group: 



Projects 



Date r, 



CCEKEIAITIOK" COMPUTING gCBM 



Predictor Variables (I) Criterion Variable (T) 

1 3VgT-l/ 

2 f 

3 , 



X, 



Y 



Z 




E TABLE (*) 






Y 



S 



/x 1X32 6 



3, 



A TABLE 



X a 



/.3 3 a/ 



X 3 



,3233 

Xn 



Y 



Ch 



. 7579 



, 75-73 



Covariances 



APPENDIX Bl 



443 



ANALYSIS OF C V A K I A ff C E 
(Supplementary Form) 



Group 


: (Jj?(&&-CvCGi>uLfL Tb-lUsiiajkfaJ^ Pro.leot: Date; 




MULTIPLE CCEEEIAIION COMPUTING FORM 


Formulas 
?/ S^3. 5-5-2, <? 


Predictor Variables (%) 

i SftT~-V 


Criterion Variable (Y) 




yyl^* 


2 .5V7T-A1 


J " 

^ j -\ 7"6 5^ 


3 


T"*&^ R^ 


Cyy , /^A ^/^ 


1 








\ B TABLE (*) 


g 


\ 

X, 


X, 


x a 


v, 


Y 


2 


7 '9.7699 


/+.4069 




^.^^5- 


A3 *.</? 


9 f ^^ 


X* 


/*/-. &O(>3 


6,23? 




3/.7J<5J 


/ /<*f * -iStSc? / 


q j A 77,3-7 


** 












3a 


Y 


v+.v-zss- 


?/. 7-3^^. 




/: MSS- 


^ 3 / 5 7 ^7 


3 


Z 


,3?,2<>'7 


f/Jt.S-S&l 




a*/.37*Z 


V-a.79tf 


r cf /^f. 7^/J- 


J 




-,VM.- S , S 




33.^5-^X 






\ A TABLE 


\ 

x, 


x, 


X a 


*3 


Y 


ch 


s 


* 73/* 


A^^ 




y.^ar 


^^v^ 


7575^7^ 


x 




7.f7Ai 




^,^7/ 


/,.**** 


x^.^^3 


x s 














b 








,??? 


.s^?^ 





(*) Co variance s 



444 



APPENDIX BI 



Group: 



ANALYSIS OF COYAKIANCE 
( Suppleafentary Form) 

% Projects - 



Date; 



MULTIPLE CORRELATION COMPUTING FORM 


Predictor Variables (X) Criterion Variable (Y) 


Formulas 

V vj2^ 




i SAT--V 


2 5>?T-A) *^f*iJ*-ujLa*JH>jJL. 




.*^ 


/ 


^ r?-(w^ 





\ 



9, 



E TABLE (*) 



47,0030 



33,. 3073 



JL 



A TABLE 



7 



Y 



5-, 



57 /7&3L 



Ch 



Ch 

s 



33. 



//A 



Covariances 



445 



APPENDIX B2 
DETERMINATION OF ADJUSTED AVERAGE GRADES 



In computing Adjusted Average Grades, it -was desired to determine the 
differences between observed and predicted grades in such a way that the 
resulting distribution would have a mean of 130 instead of and a stand- 
ard deviation equal to kO rather than equal to the standard error of esti- 
mate. An efficient procedure for determining the needed constants was 
worked out by Dr. Ledyard R Tucker and Miss Henrietta Gallagher. 

The following example illustrates the two-predictor case; the general- 
ization., however is obvious: 

Let 

S = Adjusted Average Grade, 

Y = Criterion Measure, 

X = Predictor, 

b Multiple regression weight, 

A, B, C and D = Constants to be determined, 



Then, 



a = Standard Error of Estimate of T. 

y 



H = AY - B X - C X 2 + D 



The following relations will hold: 

7 

B = A b 

C - A b 2 

D SB 130 (A My - B M-j^ - C M^) 

(in practice, D was increased by 5, and the units digit of the computed 
AAG was dropped . ) 

The Mean of 2 will be" 

S = A ML - B My -CML +D = 130 
X ^ A 2 



446 APPENDIX B2 



If C_ represents a variance and C~ x represents a covariance, 

1 ' 2AC V, + 

The following very useful checks on the constants "were employed, 
using variances and covariances from the original correlation table, and 
the value of the multiple correlation coefficient, K, previously deter- 
mined* 



0. 



r s -vT C% "^A VJ " 

**, -- - u . - - /AC BT ' 



Then: r^ + E_ - 

CM. X *J! 



In all, MG's vere determined for 16 college groups In nine of these 
groups the same students ere included hoth in the analysis of the aca- 
demic data and in the analysis of the questionnaire data. In all nine of 
these matching groups , the mean AAG when calculated fell between 129 and 
1J1. The standard deviations in all "but one instance fell between 59 and 
kl when calculated directly from the two-digit AAG*s. (in the one excep- 
tional instance, Miller, an error was discovered in three cases when the 
distribution of AAG f s was inspected. The preliminary analysis was re- 
worked to make it exact. It was judged, however y that the mean of 129 
and standard deviation of 58 obtained when the corrected scores were sub- 
stituted into the initial equation were so close to the desired value as 
to make recomputing of the AAG f s unnecessary. ) 

In six of the groups where some of the students who were included in 
the academic analysis were excluded from the questionnaire analysis, the 
following values were obtained for the mean and standard deviation of 
Adjusted Average Grades 

Group Mean AAG SI) of AAG 

Western State 133 38 

Stewart 1^0 kO 

Adams 130 kO 

Turner 131, kO 



APPENDIX B2 447, 



Group Mean AAG SD of AAG 

Midwest Tech (Ingr.) 130 40 
Eastern City (Interrupts) 1JO 4l 
Adams (interrupts) 130 4l 

It -would appear then that excluding the students who failed to complete 
questionnaires did not have a very great effect upon the mean or standard 
deviation of the total group. In most instances, as reported in Table 2 
of Chapter II, the number of students so excluded vas rather small. 

In the seventh group (Midwest Tech Agriculture students who entered 
in 1946) the mean AAG was 133 and the standard deviation 40. This devia- 
tion from the desired value of 130 was ascertained to be the result of a 
small difference between the mean used in computing the constant term in 
the equation and the effective mean of the grades used in computing AAG. 
It will be noted that this error does not affect in any way the correla- 
tional properties of the resulting AAG values. Questionnaires were avail- 
able for all but two of the members of this group. 



448 



APPENDIX B3 
TESTING SIGNIFICANCE OF RESPONSES BY THE F-TEST 

In order to determine the significance of the relationship of an item 
response to Adjusted Average Grade, using the F-test, the following pro- 
cedure was applied: 

Let 

M = mean AAG in total group, 

J T - standard deviation of AAG in total group, 

Nm = number of cases in total group, 

PA P er cen " t cases in Category A, 

M. = mean AAG of cases in Category A, 

F O = value of F needed for significance at 5$ level for 1 and 
^ Mm - 2 degrees of freedom, 

F . = value of F needed for significance at 1$ level for 1 and 
N T - 2 degrees of freedom. 

Then, it may be shown that if 




the difference between M. and the remainder of the group is significant 
at the % level. A similar relationship can be written for the 1% level. 

Using these relationships, and values of cr for each group answering 
the questionnaire, the value of M A - ]VL needed for significance for 
each value of p^ was tabulated. These tables were then used to deter- 
mine the level of significance of selected response categories for each 
item. It should be added that the test was undeniably rather coarse for 
large values of p. 

The following values of a^ were employed in making these tests: 

College Group or for Veterans cr T for Nonveterans 

Central State, Arts, 19^6 3-9^6l 3-9289 

Western State, Arts, 19^6 3,9275 3.5^29 

Miller, Arts, 1946 3.7036 3-9518 

Evans, Arts, 19^6 4.0192 4.0370 

Stewart , Arts, 1946 4.0932 3-9357 



APPENDIX B3 



449 



College Group cr^ for Veterans cr for Honveterans 

Harris, Arts, 1946 4,1470 3,7922 

Adams, Arts, 1946 3.8410 4.1067 

Douglas , Arts, 1946 4.0646 4.0314 

Littletown State, Arts, 1946 3.8299 4.1728 

Turner, Arts, 1946 3.5232 4.3171 

Midwest Tech, Engr., 1946 3.7875 3-6992 

Middle State, Engr., 1946 3-9469 3-" 7193 

Midwest City, Engr., 1946 3,8038 4!l6ll 

Midwest Tech, Agriculture, 1946 4.1981 3-5017 

Eastern City, Interrupted 3*9970 3.8299 

Adams , Interrupted 4.0450 4.1349 



450 



APJMDIX Cl 

DIRECTIONS FOR AMOTISTERIHG THE CABHBGIE 
FQT3TOATIOH VEOERABS SOOTY QUESTIOOTAIKE 



Before distributing the questionnaires, read the following paragraphs 
to the class. All material not enclosed in boxes or brackets is to be read 
to the atudents exactly as it is printed. 



At the request of the Carnegie Foundation, the College Entrance Examina- 
tion Board is making a study of factors related to s.cholastic success in col- 
lege* 

Part of the study will consist of relating various data on students.' 
preparation and background to their grades in college. Another part, equally 
important, will consist of obtaining from, a large cross-section of students 
information as to their activities, interests, and views on aspects of their 
college experience, 

Dais (class, group) has been selected as part of the cross -section of 
students whose views are desired. The questionnaire which you will receive 
is being administered to several thousand students at a number of universities 
and colleges* You are asked to answer it as fully as possible and to be com- 
pletely frank in your answers. 

In order that the data contained in the questionnaire can be matched 
with grades and other data available from the [Registrar's Office, it is 
necessary to ask your name, A sheet inserted in the questionnaire booklet 
has a space to print your name. The sheet is numbered the same as the ques- 
tionnaire booklet. After your grades have been obtained from the Registrar, 
you will be identified only by the number on your booklet. 



APPENDIX Cl 451 



The questionnaires will be sent immediately to Princeton, N. J.; no 
one at this university will work with the questionnaires. No one using the 
questionnaire to prepare statistical tabulations will know whose question- 
naire he is working with. Everything you write in this questionnaire will 
be held in strictest confidence. Please feel perfectly free to report your 
experiences and views without regard to what you think might be expected of 
you. 

It is hoped that the information gained from this study will even- 
tually result in improved methods of teaching and better educational oppor- 
tunities for college students . No individual participating in the study 
will be directly affected by it, but each participant has a chance to de- 
scribe his experiences and express his views to an influential group of edu- 
cators. You are asked to express yourself as fully and as frankly as possibl< 
There are no correct or incorrect answers to these questions; what is wanted 
is a report of your experiences and your opinions. 

In just a moment,, the questionnaires will be distributed, Please read 
the brief instructions on the front cover and then read the questions on the 
first page without answering them. There are one or two definitions that we 
might agree on in order to insure comparability of answers to these questions. 



Distribute the questionnaires . 



If you will turn to question one, you will note that it calls for 
college, school, or division. For {name of university) students this means 



452 APPENDIX 1 



In questions two and eight (l), by "terms completed" is meant the 
number of (semesters or quarters] of full-time study for which you received 
grades, regardless of how many courses you passed in any term* If you com- 
pleted a session of summer school or took work at some other college with a 
(quarter or semester] system rather than a [semester or quarter] system, 
please note this in the space for comments. 

In answering question three, give the number of credit hours as 
indicated in the college announcement, [Explain if necessaryT] 

Now print your name on the inserted sheet and fill in the other blanks * 

Check to make sure that the number on this sheet is exactly the same 
as the number on the questionnaire. If it is not the same, please raise your 
hand. 



If any case is found where the two numbers do not agree , collect the 
booklet and inserted sheet and give the student a booklet and inserted 
sheet whose numbers are the same* 



Collect the inserted sheets after they have been filled out* 



Now you may begin work. 



Collect the questionnaires at the end of the period* 



453 



APEENDIX C2 



Please print: 



Name 



Last First Middle 



Date of first entry at this university . 



Month Year 



Name of University Date_ 



Please check to make sure that the number on this sheet 
is exactly the same as the number on the front of the question- 
naire. If it is not the same, report to your supervisor that 
there is a discrepancy. 



APPENDIX 2 455 



COLLEGE ENTRANCE EXAMINATION BOARD 

Study of Scholastic Achievement 

sponsored by 
Carnegie Foundation for the Advancement of Teaching 

Student Opinion Questionnaire 
Directions 



1. Please read each question through carefully before giving your 
answer . 

2. Answer questions in the order in which they appear. 

3. Answer every question. If the suggested answers do not quite fit 
your views, check the one that comes nearest to what you want to 
say, and then explain your views in the margins or comment space 
provided. 

4. Make no marks in the "Code" boxes which you will find near cer- 
tain questions. 

5 Raise your hand and ask questions about any item whose meaning 
is not clear to you* 

6* Remember, honest and frank reports of your views and experiences 
ire the only "correct" answers. 



Inside the front cover of this booklet you will find a sheet of 
paper on which is stamped the same number as appears on this 
booklet. Please print your name, last name first, in the space 
provided on the sheet, and also fill in the name of your college 
or university and the date. 



It is necessary to ask you to record your name in order that 
your college grades can be obtained from the registrar's office. 
After the grades have been obtained, you will be identified only 
by the number which appears on this booklet. No one working with 
the questionnaire will know the name of the person who filled it 
out. 



456 



APPENDIX 2 



1. In what college, school^ or division (e.g., li