(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 "Educational psychology"

jMyRSftnfiTC* 



Digitized by the Internet Archive 

in 2008 with funding from 

IVIicrosoft Corporation 



http://www.archive.org/details/educationalpsychOOstaruoft 



fcj.isiUU^U 



.QLISTJUNl ^92S 



EDUCATIONAL PSYCHOLOGY 



ih^y^^ 



THE MACMILLAN COMPANY 

NEW YORK • BOSTON ■ CHICAGO • DALLAS 
ATLANTA • SAN FRANCISCO 

MACMILLAN & CO., Limited 

LONDON • BOUBAY • CALCUTTA 
UELBUURNE 

TTTE MACMILLAN CO. OF CANADA. Ltd. 

TORONTO 



EDUCATIONAL PSYCHOLOGY 

1^ 



BY 
DANIEL STARCH, Pn. D, 

UNIVERSITY OF WISCONSIM 



A 






"P 



Nftn fork 
THE MACMILLAN COMPANY 

1920 

Al rights reserved 



Copyright, iqiq 

By the MACMILLAN COMPANY 

S«t up and cUxlrutypcU. Published June. io«<) 



/ 



TO MY WIFE 



yll 



PREFACE 

The preparation of this book has been carried out according to 
two fundamental purposes: First, to present that material which 
seems to be most useful and relevant to the problems of educa- 
tional psychology; and second, to maintain a strictly experimental, 
scientific viewpoint in discussing these problems. The result of 
these aims has been a considerable reduction in the amount of 
space usually devoted in texts on educational psychology to cer- 
tain topics such as, instinct, fatigue, and imagery, and the inclusion 
of new topics such as tests of intelligence, studying, transference 
of training in school subjects, the assignment of marks, and much 
of the material in Part III which has as yet not found a place in 
text-books. 

The space devoted to the discussion of instinct has been ma- 
terially reduced for two reasons: In the first place, while the in- 
stincts are fundamental in human life, too much time has usually 
been devoted to their consideration for the amount of direct 
benefit gained. The actual use in school work that can be made 
of a detailed knowledge of instincts, which in our present stage 
of information is largely analytical and theoretical, is relatively 
small when it comes to dealing face to face with concrete school 
problems. In the second place, a great deal of experimental 
and statistical material has accumulated in recent years which is 
more immediately valuable in solving the problems of the psy- 
chology and pedagogy of learning. 

It would have been desirable to include a discussion of the 
l)sychology of more of the high school subjects; but this is impossi- 
ble at the present time. The discussion of the school subjects in 
Part III has been confined to tangible, scientific investigations. 
Obviously there is little or no material of this sort on most of the 
high school subjects. The consideration of educational tests in 
the chapters of Part III is perhaps brief; but a detailed treatment 
of the theoretical and statistical principles underlying their con- 
struction belongs rather in special treatises. Chapter XII on 
How to Study is not altogether satisfactory, because of the scar- 
city of definite or substantial material '^in this field. It was, how- 



v-iii rKKF-ACE 

ev'cr, included because the topic is exceedingly important in school 
work and because it was hoped that its inclusion would stimulate 
discussion of it by teachers and prospective teachers. 

I take pleasure in expressing my obligations to the persons who 
luve assisted me in various ways in the preparation of this book; 
namely, to Dr. Helen Hubbert Caldwell and Mr. A. O. Hansen, 
who have read the manuscript and ofTered many helpful sugges- 
tions, to Mr. W. R. Ames who has j)rei)ared the drawings, and 
e>IK-eially to Dr. C. L. Hull who has critically examined every 
[jortion of the manuscrij)! and has offered many suggestions which 
have Ix-en incoqxjrated in the book. 

Daniel Starch. 

•Madison, Wisconsin, 
October 5, 191S. 



i^ 



CONTENTS 

Chapter page 

I. Problems and Scope of Educational Psychology i 

Part I. The Native Equipment of Human Beings 

II. The Instinctive Elements of Native Equipment 9 

III. Variation in Human Capacities 26 

IV. Correlation Among Human Capacities 49 

\'. Sex Differences 63 

VI. The Inheritance of Mental Traits 73 

VII. The Measurement of Mental Capacities 97 

Part II. The Psychology of Learning: A. In General 

VIII. Analysis of Problems 115 

IX. The Reception of Stimuli: A. Sensory Defects 121 

X. The Reception of Stimuli: B. Perce ption and Obsgrvation of 

Sensory Material 132 

XL The Rate_andProgress of Learning 141 

XII. How to Study 176 

XIII. Transference of Training in Special Mental Functions '191 

XIV. Transference of Training in Abilities in School Subjects. . . . 217 

Part III. The Psychology of Learning: B. Of School Subjects 

■ XV. The Psychology of Learning School Subjects 2^j>^-« 

XVI. Reading 261 

XVII. Handwriting 297 

XVIII. Spelling 322 

XIX. Language 349 

XX. Arithmetic 374 

XXI. History 416 

XXII. Marks as Measures of School Work 426 

Bibliography 45i 

Index 465 

ix 



EDUCATIONAL PSYCHOLOGY 



EDUCATIONAL PSYCHOLOGY 

CHAPTER I 

PROBLEMS AND SCOPE OF EDUCATIONAL PSYCHOLOGY 

What is Education? The problems and the scope of educational 
psychology are necessarily determined by our conception of what 
education is. If we conceive education to be primarily self-devel- 
opment, our problems will be of one sort; if we conceive education 
to be fundamentally social adaptation, our problems will be of 
another sort. In the former case, education would mean the 
complete training of the mental and physical capacities irrespective 
of environment; in the latter case, education would mean the 
training of those capacities which will adapt the individual most 
adequately to the social and physical environment in which he is 
to live. For our present purpose it is not necessary to define in 
complete detail the aim and meaning of education. It will be 
suflScient to state in the simplest terms what education is as a 
psychological process. 

In the broadest sense, education is the production of useful 
changes in hmnan beings, i These changes may be classified into 
three divisions: changes in knowledge, in skill, and in ideals. 
Through education the child is to acquire useful knowledge; he is 
to acquire skill, both motor and intellectual, in the use of his 
muscles and in the manipulation of ideas and concepts; and, 
finally, he is to acquire the right ideals of life which will actually 
function in his behavior. Probably all changes wrought in human 
beings which in any sense are educational, fall under these three 
heads. Obviously then, education is the most momentous, as well 
as the most essential, business of the human race; for the welfare 
of the race depends upon education as it depends upon nothing 
else. 

' Thorndike has defined the purpose of education thus: "The aim of education is, as 
we have seen, to change human beings for the l)etter, so that they will have more humane 
and useful wants and be more able to satisfy them." ('12, p. 52.) 



,, EDUCATIONAL PSVCHOLOCJV 

Which changes arc ustful and which arc not is a (luestion that 
cannot \)c answered as easily as it would seem at first glance. 
U-aming to read or to figure are obviously useful modifications; 
l.ut it is not so easy to say whether the study of a given drama, or 
I III- kii..wli-<lj,'i- of certain facts of histor>', or the understanding of 
.1 certain theory of matter, or ability to read a given foreign lan- 
„'Uiige, are useful, or sufliciently useful to be included in the com- 
i school, in the high school, or in the university. The term 
111 should not Ik- limited narrowly to the things which are 
iiiunediatelv applicable in making a living, but should include 
ill cluinges wliich will broaden and enrich the life of the in- 
dividual. 

The Problems of Educational Psychology. In accordance with 
our detinltiiMi, the lund.tineiilal problem^ that we must raise 
voncerning education are as follows: 

1. What changes arc to be made in human beings? 

2. What are the agencies by which the changes may be broug\ic 
.iljout? 

V What are the capacities which human beings possess for 
icquiring the changes? 

4. What are the most economical methods by which these 
t hangeN nuiy be brought about? 

'I'he first problem is j)rimarily a jtroblem for philosophy anti 
4K iolog)'. What changes are ultimately to be made de[K*nds upon 
our ideals of life and our views of society. The nvxlifiaitions 
ihiit haye been sought by dilTerent nations and ditlerent sys- 
tems of c<lucation have varied fron\ century to century and 
from race to race. The ultimate aims of education sought by 
the ancient Greeks or by the mediaeval monks were very 
different from those sought by the modem Americans or 
I',uro|H-ans. 

' < und problem dc*:ds primarily with the value of school 

uid exercises in bringing about the changes that are to be 

iii.KJe. To whxit extent will the study of arithmetic, the study of 

U'r.iinnuir, or the study of physics or Latin be able to produce the 

ir.iininL' that philosophy and sociology dictate? This problem is 

' partly p.sychological. It is .sociological in 

it ion of txiucational agencies depends upon 

'h' i lul social environment of mankind; it is jisychological 

iit .-«. ..I. .. li necessitates a study of mental processes affected or 



PROBLEMS AND SCOPE 3 

brought about by these agencies. This latter phase of the problem 
merges into problems three and four. 

Problems three and four are fundamentally psychological and, 
together with the second phase of problem two, constitute the 
main scope of educational psychology. It is a psychological prob- 
lem to determine what capacity and eciuipment human beings 
have for acquiring the changes that are to be made. Likewise, it 
is fundamentally a psychological problem to discover the most 
economical methods of learning. Accordingly, the field of edu- 
cational psychology is divided into two large divisions which we 
may designate as: 

I. The native equipment of human beings; 

II. The psychology of learning. 

Psychology and Teaching. Methods of teaching rest funda- 
mentally upon the psychology of learning. Since the experimental 
analysis of learning processes will have to reveal the principles ac- 
cording to which the human mind learns, and learns most economic- 
ally, it follows that the methods of teaching will have to be based 
upon these discoveries. This may be illustrated in the case of 
reading. If the child learns to read most economically by the word 
method, it follows that the most economical way of teaching read- 
ing would be ])y the word method. Likewise, if a child learns to 
spell homonyms more easily by studying them together, or memo- 
rizes prose or poetry more readily by wholes than by parts, it follows 
that these exercises should be taught accordingly. Evidently the 
fundamental principles of teaching must be based upon the psy- 
chological laws and principles of economic learning. 

Waste in Education. Exact information concerning the proper 
procedure in educational matters is exceedingly rare. Definite, 
scientific knowledge of the proper methods of learning and teaching 
school subjects and of the efficient administration of our schools 
is surprisingly small, and the field of educational psychology in its 
broadest sense opens up endless problems for the future to solve. 
We know relatively little in a scientific way about the learning of 
any single school subject. For example, we do not know with any 
definite assurance what is the most economical amount of time to 
devote to any one of the school subjects. From such investigations 
as have been made, we may infer that there is an enormous waste 
in our educational practices which is indicated by such facts as the 
following: It has been found by recent tests and measurements 



4 KDrcwrrowL psvcholoc.v 

tlut some schools obtain just as gtXKl results by devoting on! 

onc-hulf as much time to writing as other schtK>ls do. Similar fac 
^^ have been brought out in the case of reading, arithmetic and oth( 

'-* school subjects. Sch(H)Is which have devoted as much as ic 

''^ minutes a week, or 20 minutes a day, have obtained no better r 

^^ suits tlun other schools devoting 50 minutes a week, or 10 minuti 

'' a day, to the s;mie subject. If these facts actually represent ll: 

f^ rial |)ossibilities, it seems quite obvious thiit there is an enormoi 

"^ wa>te in our schcM>ls and this waste is far greater than we reali; 

"- until we nuike delinite calculations of the possible saving of tim 

'" If by some means it were possible to save one minute a day k 

al ever)' school day during the eight years of a child's school lif 

*'' Wf would be able to save one entire week of school tinie. If \\ 

could s;ive four minutes a day for the same length of time, ^^ 
"^ would Ik- able- to save one month; if we were able to sa\e 18 mil 

utts a day, wt- would be able to .save one-half of a school year; an 

if by more t'coin)mical methods of learning and distribution < 

time we were able to save 36 minutes a day for eight years, \\ 
;jj would be ablr to save an entire school year. Such a .saving is m 

imiK)ssibk-; indeed, by a better use of time and more elTecti\ 
;i, methods of learning, it is highly probable. Eighteen minutes 

day would mean a reduction of only ^yi minutes in each of foi 
, \ subjet Is; 3^ minutes a day would mean a reduction of only 

minutes a day in each of four subjects. This time could be d< 

voti-d with greater advantage to other and possil)ly more advance 
** school subjects and school exercises. 

"' The Specific Topics and Problems. In order that we may I 

projR-rly oriintaled with reference to the problems tliiit will l 
" discussed under the two large divisions of educational ])sycholog} 

'•^ the fuilowinL'ordi Tnf tuDJcs will be considered: 

I. i he ii:iti\e ((luipiiienl of human beings. 

a. What d(H'S it consist of? 

b. To what extent does it N'an*? 

1. Ajnong individuals. 
'" (a) In single traits. 

^'' 0>) In combinations and relationships of traits. 

2. At different times of life in the SiUne individual. 
I*" 3. Hetween the sexi*s. 

*^ C. To what extent is it inherited? , 

: <l. How may it be measured? 



PROBLEMS AND SCOPE 5 

II. The Psychology of Learning. 

a. The psychology of learning in general. 

1. Observation and perception. 

2. The rate and progress of learning. 

3. Transference of training. 

b. The psychology of learning school subjects in particular. 

1. The psychological processes involved in each subject. 

2. The measurement of ability and progress in learning 

each subject. 

3. The most economical methods of learning the material 

of each subject. 



a* 
d 



M 
Ol 

U 
ti 
fr 
tl 
<1 



PART I 
THE NATIVE EQUIPMENT OF HUMAN BEINGS 



CHAPTER II 

THE INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 

Reflexes, Instincts, and Capacities. The equipment with 
which human beings start in hfe may be divided into three types 
of inherited responses and abihties: Reflexes, instincts, and capaci- 
ties. The distinction among these three is primarily one of definite- 
ness and degree of complexity. An instinct may be defined from 
the neurological side as an inborn neural connection between sense 
organ and muscle. It may be defined from the functional side 
as an inborn capacity of responding in definite ways under definite 
circumstances. These responses are prior to experience and train- 
ing, and need not be learned. To close the eyes when an object 
suddenly approaches them, to get food when hungry, to strike 
when struck, and to be afraid of thunder and of the dark, are illus- 
trations of instinctive responses. The reflexive and instinctive 
responses are inherited in the sense that there is present in the 
nervous system, either at the time of birth or later on as a result 
of growth, a set of nervous connections already formed for the 
carrying out of a particular action in response to a given situation. 
If the child closes his eyes when an object suddenly approaches, 
it means that the motor impulses travel from the retina to the 
visual center of the brain, from there to the motor center which 
controls the movement of the eyelids, and from there out to the 
muscles of the eyelids to cause the contraction. In the case of 
inherited responses, the connection from the sensory to the motor 
centers is already present and ready to operate in carrying out 
the action. In the case of acquired responses, such as habits, 
these nervous connections must be formed as a result of effort 
and trial on the part of the individual. 

The difference between reflexes and instincts is largely a differ- 
ence of complexity. Both are inherited types of responses. Re- 
flexes are simpler forms of reaction usually involving a limited set 
of muscles and occvu-ring in response to precise stimuli. The 
contraction or expansion of the iris, the closing of the eyelids, the 
knee jerk, are illustrations of reflexes. Instincts are complex re- 



lO KDUCATIONAL PSYCHOLOGY 

actions involvin;:; tlu- use of large groups of muscles or, in many 
instances, the entire muscular system of the body. They may he 
aroused either by external stimuli or situations or possibly by 
internal stimulation. To make movements in the direction of 
getting f(MKl when hungry, to seek shelter when cold, to offer re- 
sistance when hemmed in, to s])it out what tastes bad, and the 
like, are instinctive reactions. Capacities are distinguished from 
reflexes and instincts in being general mental abilities rather than 
specific motor responses and in referring primarily to the native 
mental e(juipmcnt, such as the powers of sensation, j)ercei)tion, 
retention, attention, imagination, and all the varied complex 
I)sychic processes. 

Classification of Instinctive Responses, The older classifica- 
tions of instincts usual!}' dixided thcni into three or four large 
groups of responses, such as individual, racial, and social, and re- 
garded them rather as general tendencies than as specific responses. 
The present conception of instincts is to regard them as specific 
responses with inherited neural mechanisms which will be set 
into action by specific stimuli or situations. On this basis the 
classification consists of an enumeration of as many definite, identi- 
fiable, unlearned reactions to specific situations as can be observed 
and as can be recognized in human beings prior to training and 
habituation in each particular ty]>e of activity. Accordingly, 
Thonidike ('14, I) enumerates forty or more different tyjx's of in- 
stinctive reactions as follows: 

1. Food getting and j)rotective responses. 

1. Eating. 

2. Reaching, grasping, and putting objects into the 

mouth. 

3. Acquisiti(jn and possession. 

4. Hunting. 

5. Collecting and hoarding. 

6. Avoidance and repulsion. 

7. Rivalry and cooperation. 

8. Habitation. 

9. Response to confinement. 

10. Migration and domesticity. 

11. Fear. 

12. Fighting. 
I J. Anger. 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 1 1 

II. Responses to behavior of other human beings. 

14. Motherly behavior. 

15. Gregariousness. 

16. Responses of attention to other human beings. 

17. Attention getting. 

18. Responses to approving and to scornful behavior. 

19. Responses by approving and scornful behavior. 

20. Mastering and submissive behavior. 

21. Display. 

22. Shyness. 

23. Self-conscious behavior. 

24. Sex behavior. 

25. Secret! veness. 

26. Rivalry. 

27. Cooperation. 

28. Suggestibility and opposition. 

29. Envious and jealous behavior. 

30. Greed. 

31. Ownership. 

32. Kindliness. 

^^. Teasing, tormenting, and bullying. 

34. Imitation. 

III. Minor bodily movements and cerebral connections. 

35. Vocalizations. 

36. Visual exploration. 

37. Manipulation. 

38. CleanHness. 

39. Curiosity. 

40. Multiform mental activities. 

41. Multiform physical activities. 

42. Play. 

Relation of Education to Native Endowment. The inherited 
equipment of the human being is the foundation upon which educa- 
tion must build; it consists of the faculties and capacities which 
the child has for reacting to his environment. It is the utilization, 
the training or the curbing of these endowments which education 
attempts to accomplish. In much of the writing and thinking con- 
cerning educational problems, there has been a relative overem- 
phasis, in space and time, upon instincts and an underemphasis upon 
the mental capacities. Education in the sense of schooling has as 



12 EDUCATIONAL PSYCHOLOGY 

much if not more to do with the latter than with the former. The 
direct appeal to, and use of, instinctive reactions in actual concrete 
instances in school work are not as frecjuent and specific as is com- 
monly implied. The number of instincts enumerated in the pre- 
ceding list which may be directly and concretely appealed to in the 
learning of a school subject is relatively small. The best way to be 
convinced on this j)oint is to take the various instincts one by one, 
and to determine to what extent each one may be aj)pealed to or 
used in teaching the various sxibjects. The number of specific 
applications is much smaller than one is likely to anticipate. Two- 
thirds or three-fourths of them are probalily never immediately 
but only indirectly concerned in school exercises, and most of the 
remaining ones, such as rivalry, cooperation, collecting and hoard- 
ing, are serviceable chiefly as general motives. As such they are, 
to be sure, highly important. 

We must, of course, not minimize the j)lace and importance of 
instinctive reactions in behavior as a whole. They furnish the 
general motives and mechanisms for doing and learning, but the 
mental capacities are more directly and concretely involved in the 
acquisition of knowledge and skill in school subjects. 

The instinctive elements in learning any school subject are for 
the most part simple reflex actions or undeveloped connections. 
Take learning to read as an illustration. The chief instinctive 
elements ])robably are the reflexes in the control of the eyes, the 
neural mechanism for receiving and transmitting \isual impulses 
to the brain, the capacities for attentiveness and retentiveness, 
and partial motor control of the speech organs. The process of 
learning to read assumes these and uses them; but, what is 
more important from the practical side of getting the meaning 
of the printed word is the establishment of countless new con- 
nections. 

Perhaps the most important role of instinct in education lies 
in motivating and energizing the learning processes. There can 
be no education except through the activity of the child himself; 
and no activity can take place which does not ultimately depend 
ujjon native tendencies. They are the origin of effort, the springs of 
action. The skillful teacher j^lays upon them and aj^peals to them 
in countless ways. The ability to do this is an art which is not 
rasily learned from books; it is acquired rather by jxitient practice 
and by sympathetic contact with children. 

The energizing power of instinct makes itself felt largely through 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 13 

its control of the attention processes. Owing to the peculiarities 
of our inherited nervous organization, certain impressions have a 
potency over others in attracting the attention and interest of the 
child. A flash of lightning, a hoHday parade, one's name in the 
newspaper, or a moving picture make certain instinctive appeals to 
the attention of a young girl which the study-lamp, the doing of 
errands for mother, the seeing of a stranger's name in the newspaper, 
or the reading of the history lesson do not make. The great im- 
portance of attention for the learning process lies in the fact that 
associations, analyses, and indeed all mental processes are carried 
out much more effectively when they occupy the focus of attention. 
Ebbinghaus found that, after inattentively reading over nonsense 
syllables until many successive persons had learned them per- 
fectly, he himself could repeat very few of them. Impressions 
must occupy the focus of consciousness in order to be retained 
effectively. 

Considering the three main t)^es of attention, passive, active, 
and secondary passive, the most simple and the one most directly 
related to the instinctive life is probably the first. Passive atten- 
tion is such as one gives spontaneously to any curious or interesting 
sight or sound. Active attention is such as one gives perhaps to an 
inherently distasteful task which requires an efifort of will to keep 
the mind upon it. While such a task itself does not supply the 
stimulus for vigorous instinctive reaction, it is in some direct way 
connected with one that does. A little girl will apply herself to the 
disagreeable task of learning a spelling lesson, not because the 
words in themselves have any charm for her but because she has 
the instinctive craving for the approval of her teacher. The third 
type of attention, secondary passive or derived attention, is at- 
tention which has become passive only after having passed through 
an initial active stage. It is illustrated by the common experience 
of becoming absorbed in a task which at first required a distinct 
effort. In the beginning the motivation lay outside the task, say 
in a sense of duty or social obligation; but after a time an adequate 
stimulus for activity was encountered in the work itself. 

Apparently back of every act of attention lies somewhere a more 
or less primitive, innate tendency to action. To focus the atten- 
tion of a class upon various associations involved in learning the 
multiplication table, a skillful teacher may on one day appeal to 
curiosity in the novelty of the combinations; on another day she 
may appeal to the native pleasure in rhythm by making the table 



14 EDUCATIOXAL PSYCHOLOGY 

into u rhyme or sonp;. Or the tendency to play will he utilized by 
making the number combinations into a game. More than likely 
the game itself will dejiend upon instinctive rivalry and emulation. 
The love of social ai)proval is appealed to by giving distinctions, 
murks, gilt stars, and the like. Future ad\'antage may be used as 
an intlucement for present api)lication. The instinct of pugnacity 
may be utilized in wanting to succeed in a hard task. The teacher 
herself, in standing position with face and body in animated 
attitudes, may api)eal to the fundamental interest in change and 
action. Lastly may be mentioned the more doubtful negative 
moti\es of deprivation from coveted privileges and, biologically 
l)erhaps the most fundamental and powerful of all motives, physical 
pain. It may be worth noting that animal psychologists have 
found pain in some cases a more potent motive to learning than 
])leasure. Hoge and Stocking ('12) found that rats when rewarded 
b\- food alone had by no means learned j^erfectly certain sensory 
habits in 610 trials; when they were punished for failures they 
learned the habits jjcrfectly in this number of trials, but when 
they were both rewarded for successes and punished for failures 
they learned the habits perfectly in 530 trials. 

Educational Doctrines Based upon Instincts. Some very far- 
reaching s])ecu!ations and theories with regard to the nature and 
value of instincts for education have been spun out, some of which 
an' largely imaginary and questionable, and are based upon analogy 
rather than fact. For the most part these educational doctrines 
have centered around three concepts. 

The first is that instincts are the great dynamic forces of human 
nature which determine the actions, desires, and achievements in an 
indivitlual's life. Hence the injunction to the school has been to 
work with nature rather than against her or apart from her. We 
shall call this the dynamic theory of instincts. 

The second is that these instincts are highly transitory; that they 
burst out at certain times in the growth of the individual with 
more or less dramatic force and suddenness, and that if they are 
not allowed to manifest themselves, they will disaj^pear never to be 
revived again. From this assumption has been derived the peda- 
gogiail application of the ma.xim, ''strike while the iron is hot.'' 
'ihis is the theor}' of the transitorincss of instincts. 

The third is that instincts appear in the growth of the child in 
t he order in which they appeared in the evolution of the race. From 
thi^ assumption has been derived the jjcdagogical maxim, "teach 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 15 

the child his activities in the order in which the interests for 
them appear." This is known as the recapitulation theory of 
instincts. 

Critique of the Dynamic Theory of Instincts. To work with 
nature rather than against her is undoubtedly a sound principle. 
The fundamental instincts of man are the driving forces of human 
life that determine ultimately the motives and causes of behavior. 
They are so deep-seated in the human psychophysical organism 
that we may almost say that to work apart from, or against, nature 
is a futile task. If, through the instinct of multifonn activities, a 
child manifests a tendency to draw, the school should take ad- 
vantage of this original impulse and build upon it. All the original 
manifestations of a child's nature should be used in the acquisi- 
tion and training of those exercises which education considers 
valuable. 

This dynamic theory of instincts, however, involves on the one 
hand a difficulty and on the other a questionable assumption. The 
difficulty is that the principle is general and as such does not point 
out the particular ways in which the school may cooperate with the 
inborn forces of child nature. It is easy enough to say "work with 
nature," but just how is that to be done in teaching a pupil how to 
make the letter "a," or to learn the reading of a printed word, or to 
acquire correct speech, or to learn the grammatical rules of a foreign 
language? Ultimately, the concrete use of the principle must be 
determined experimentally. Our definite knowledge of the tech- 
nique of learning in the case of school subjects is appalUngly limited. 
Only by careful and painstaking experimentation can this principle 
be made useful in the concrete work of the school in anything more 
than an offhand impressionistic manner. 

The questionable assumption is that the instincts are infallible 
guides of human life. It may be argued that since instincts are 
such powerful springs to action as to have maintained the individual 
and the race for numberless generations, they must necessarily be 
dependable in producing action and interest of the right sort. But 
the question may fairly be raised: Are the native tendencies always 
right so that we should always cooperate with them and never coun- 
teract or curb them? The theory of the infallibility of instincts is 
based on the belief that for countless ages nature has found by ex- 
perimentation and natural selection what is best for the individual. 
Whatever the child is inclined to do by virtue of his natural pro- 
clivities is right and good for him; or, if apparently not useful, it is a 



1 6 EDUCATIONAL PSYCHOLOGY 

necessary precursor or a necessary accompaniment of useful tenden- 
cies. On the theory of immunization or catharsis, the undesirable 
tendencies prepare the ground for the proper development and 
growth of the desirable ones. However, the theory of catharsis 
is highly questionable and runs counter to the law of use and disuse 
which in general operates to the effect of making permanent the 
functions exercised. The belief is that if a boy has proclivities 
toward thieving or lying or dishonesty, and is allowed to exercise 
these unconstrained, he will purge himself of these tendencies and 
he the more honest and truthful later on in life. Concrete data, 
however, in addition to the general principle of use and disuse, seem 
to point in the opposite direction. Experimental and statistical 
inquiries show that the relative strength of the various traits 
remains fairly constant throughout life; that if a child manifests 
certain abilities and tendencies even during childhood, these abili- 
ties and tendencies will remain relatively dominant during his adult 
life. Early interests and intellectual capacities are very certain, on 
the whole, of being prophecies of similar interests and intellectual 
abilities later in life. As will be pointed out in a succeeding chapter, 
scholastic ability remains fairly constant in any given child all 
through his educational career. 

Furthermore, we must remember that nature, in the develop- 
ment of instincts and in securing adaptation to environment, 
works on the whole in a very slow and prodigal manner; and that 
conditions of life may change long before the organism through its 
behavior appropriately adapts itself to the surroundings by fur- 
nishing the necessary native equipment within the organism. On 
this account a great deal of our native equipment is out of date and 
has adapted man to primitive conditions of uncivilized life. As a 
result of this, we manifest many tendencies which are not par- 
ticularly useful at the present time. We would be better off if in 
l)Iace of them we had instinctive capacities for meeting situations 
with which we are to-day confronted in civilized life. As Thoni- 
dike has pointed out: 

"The imperfections and misleadings of original nature arc in fact 
many and momentous. The common good requires that each child 
karn countless new lessons and unlearn a large fraction of his natural 
birthright. The main reason for this is that original equipment is archaic, 
a(!ai)ting the human animal for the life that might be led by a family 
group of wild men in the woods, amongst the brute forces of land, water, 
wind, rain, plants, animals, and other groups of wild men. Tiic life to 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 17 

which original nature adapts man is probably far more like the life of 
the wolf or ape, than like the life that now is, as a result of human art, 
habit and reasoning, perpetuating themselves in language, tools, build- 
ings, books and customs." ('14, 1, p. 280.) 

That these primitive tendencies persist with great strength is 
shown by the ready manner in which the veneer of civilization comes 
off and by the fact that men and women in strained circumstances 
will easily revert to their primitive, brutal instincts. The chief ad- 
vocate of the theory of infallibility and catharsis of instincts has 
been G. Stanley Hall, who has said: 

"In education, don't cut off the tadpole's tail." 

"Rousseau would leave prepubescent years to nature and to these 
primal hereditary impulsions and allow the fundamental traits of savagery 
their fling till twelve. Biological psychology finds many and cogent 
reasons to confirm this view if only a proper environment could be pro- 
vided. The child revels in savagery; and if its tribal, predatory, hunting, 
fishing, fighting, roving, idle, playing proclivities could be indulged in 
the country and under conditions that now, alas! seem hopelessly ideal, 
they could conceivably be so organized and directed as to be far more 
truly humanistic and liberal than all that the best modern school can 
provide. Rudimentary organs of the soul, now suppressed, perverted, or 
delayed, to crop out in menacing forms later, would be developed in 
their season so that we should be immune to them in maturer years, on 
the principle of the Aristotelian catharsis for which I have tried to sug- 
gest a far broader application than the Stagirite could see in his day." 
(Hall, '08, p. 2.) 

"He should have fought, whipped and been whipped, used language 
offensive to the prude and to the prim precisian, been in some scrapes, 
had something to do with bad, if more with good, associates, and been 
exposed to and already recovering from as many forms of ethical mumps 
and measles as, by having in mild form now he can be rendered immune 
to later when they become far more dangerous, because his moral and 
religious as well as his rational nature is normally rudimentary." ('08, 
P- 23S-) 

The Theory of the Transitoriness of Instincts. This concep- 
tion of instincts contains two elements: First, the suddenness of the 
appearance of instincts, and second, the unrevivable disappearance 
of instincts. The most conspicuous advocate of the former idea in 
this country, has been G. Stanley Hall, who states it as follows: 

"But with the teens all this begins to be changed and many of these 
precepts must be gradually reversed. There is an outburst of growth 



i8 



p:i)Ucational psychoix>gy 



tlial iiL'cds ;i large pari of the total kinetic incrgy of the IxkIv. There is 
a new interest in adults, a passion to be treated like ones elders, to 



180 
170 

leo 

E 

^ 160 
c 

Z 140 

u 
o VJi) 

120 

110 

























_ 


























r^ 


rW 






















^?5 


y- 


— 




















.'f' 


^^ 






















y 


^ 






















^ 


^ 






















^y 


^ 
























^^ 



























5.5 0.5 7.5 8.5 9.5 10.5 11.5 12.5 135 14.5 15.5 16.5 17.5 185 
Age in Years 

Fig. I. — Height of boys and girls measured in centimeters, based on measure- 
ments of 45,151 lx)ys and 43,298 girls. After Boas Cgb-'gy). 

make plans for the future, a new sensitiveness to adult praise or blame. 
The large muscles have their innings and there is a new clumsiness of 
body and mind. The blood-vessels expand and blushing is increased, 
new sensiilions and feelings arise, the imaginatit)n lilossoms, love of 



AlU 


















^ 






190 


"^^^ 


r 












/ 


Boys 




•" 
X 


180 
rtlTO 














/ • 



















^ 


















A' 


f 
















140 


i^' 





















lU 11 



12 13 
Ages 



15 10 



18 



Fir,. 2. — Rale of tap|)ing. Number of taps made with right hand in 30 
seconds. Based on tests with 395 boys and 495 girls. After Smedley 
('00- 01). 

nature is born, music is felt in a new, more inward way, fatigue comes 
easier and sooner; and if heredity and environment enable the individual 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 19 

to cross this bridge successfully, there is sometimes almost a break of 
continuity, and a new being emerges." ('08, p. 236.) 

The correctness of the theory of sudden appearance is primarily 
a question of fact. Thus Hall describes the social instincts at the 
time of adolescence as follows: 

"The social instincts undergo sudden unfoldment and the new life of 
love awakens. It is the age of sentiment and of religion, of rapid fluctua- 
tion of mood, and the world seems strange and new. Interest in adult 
life and in vocations develops. Youth awakes to a new world and under- 
stands neither it nor himself." ('04, Preface p. XV.) 

The advocates of this viewpoint maintain, therefore, that there is 
a nascent period for motor activity, for memory and habituation, 
for reason and logical thinking and the hke; that the school should 
seize these opportunities to teach those activities which will exer- 
cise the particular capacities that occupy the stage of youth at that 
period; and that more can be accomplished at those periods in a 
given length of time than can be accomplished in several fold as 
much time later on. 

The facts do not seem to warrant an interpretation of such 
marked suddenness but indicate rather a gradual waxing of in- 
stincts. There appears to be no special time during which the child 
suddenly begins to reason or to reason very much more than he had 
done theretofore. The same description seems to be true of 
memory, motor ability, the collecting instinct, and other capaci- 
ties, as indicated in the accompanying graphs. 

A great deal of the dramatic bursting forth of instincts is chiefly 
a dramatic bursting forth of descriptive words. The actual facts 
seem to justify more nearly an interpretation of gradual unfold- 
ment instead of a sudden bursting forth. Growth in height and 
weight proceeds by a very uniform increase even diu-ing the adoles- 
cent period. Motor capacity grows steadily and uniformly without 
particularly sudden leaps or bounds. Memory ability increases 
steadily for both rote and logical material up to adulthood, during 
which it probably remains fairly constant mitil senility sets in. 
There is no memory period diu-ing which the child memorizes very 
much more readily than he did before or than he ever will later. 
The memory of the average adult for both mechanical and logical 
material and for either immediate or permanent retention is superior 
to the memory of the average child at any age. Even reasoning 



20 



EUUCATIONAL PSYCHOLOGY 



ability, which is usually described as appearing suddenly at the 
dawn of adolescence, is a matter of gradual development . To argue 



100 
96 
90 
b5 
M) 
75 

•^70 

o 

*'66 
•i 

<u 
ex. 60 

&5 

£0 

45 

40 

a5 



































































































^,' 


'\ 


>/ 














.^' 




"' 






^V 














1/ 
























-i."/ 




H- 


\ud 


tor; 


: — 





^ 












/ 
/ 




1 




















/ 




J 






















v 


/" 
























/ 






















^ 
























^ 


























n 


















•1 1 


7 1 




Q •» 



Ages 

Fig. 3. — Memiiry for diRils based upon tests of 9,^7 |>upils. After Smedley 
('cx)-'oi). 



2.6 

2.0 

•S)1.5 

10 
.5 




















/ 








A 


/ 






y 


/ 




^^ 





















\ 5 

Grades 



Fir.. 4.- Development in arillimetical reasoning as measured by llie Courtis 
test No. 8, Series ;\. The verlica.1 a.xis shows the number of problems done cor- 
rectly in si.\ minutes. 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 21 

that school exercises which consist mainly of memorizing should be 
placed at the "memory age" on the ground that the pupil will 
learn them more readily at that time than at a later time in life, 
is fallacious. It may be advisable to begin the study of foreign 
languages earlier than is customary, but not for any reason of more 
rapid memorizing at an earlier age. If rapidity of tapping, Figure 2, 
is any indication at all of endurance or of quickness of becoming 
fatigued, it does not seem to be true that "fatigue comes easier 
and sooner" during the adolescent stage. The graphs for tapping 



11 

10 

w 7 
6 
6 










/ 










J 


/ 










/ 










/ 


/ 










/ 










/ 












/ 























4 5 6 7 8 

Grade 

Fig. 5. — Development in arithmetical reasoning as measured by the arith- 
metical scale A. After Starch ('16). The vertical axis represents the scale step 
or the number of problems done correctly. 

do not drop but tend to rise gradually even during the years from 
eleven to fifteen. There is practically a level at the age of eleven 
but no drop. 

The unrevivability of instincts through disuse has been advocated 
chiefly by James as follows: 

"This leads us to the law of transitorincss which is this: Many in- 
stincts ripen at a certain age and then fade away. A consequence of 
this law is that if, during the time of such an instinct's vivacity, objects 
adequate to arouse it are met with, a habit of acting on them is formed, 
which remains when the original instinct has passed away; but that if 



22 



EDUCATIONAL PSYCHOLOC;\ 



no such objects are met with, then no habit will be formed; and, later 
on in life, when the animal meets the objects, he will altogether fail to 
react, as at the earlier epoch he would instinctively have done." ('90, 
II, p. 398.) 

"Leaving lower animals aside, and turning to human instincts, we 
see the law of transiency corroborated on the widest scale by the altera- 
tion of different interests and passions as huniiin life goes on. With the 
child, life is all play and fairy-talcs and learning the external properties 
of 'things'; with the youth, it is bodily exercises of a more systematic 
sort, novels of the real world, boon-fellowship and song, friendship and 
love, nature, travel and adventure, science and i)hilosophy; with the 





























4 










/ 


\ 


















/ 






\ 














m Q 
1 






/ 






\ 


^ 


/ 


\ 












/ 
















"^ 


\ 




I2 




/ 
























/ 






















1 













































































7 S U 10 11 1-' 13 U 15 16 17 
Ages 

Fig. 6. — The number of collections made by children at various ages. After 
C. V. Burk ('00;. 

man, ambition and policy, acquisitiveness, responsibility to others, and 
the selfish zest of the battle of life. If a boy grows up alone at the age of 
games and sports, and learns neither to play ball, nor row, nor sail, nor 
ride, nor skate, nor fish, nor shoot, probably he will be sedentary to the 
end of his days; and, though the best of opportunities be afforded him 
for learning these things later, it is a hundred to one but he will pass 
them by and shrink back from the effort of taking those necessar>' first 
steps the prospect of which, at an earlier age, would have filled him with 
eager fielight. The sexual passion expires after a protracted reign; but 
it is well known that its peculiar manifestations in a given individual 
dejK'ud almost entirely on the habits he may form during the early 
IH'rio<l of its activity. Kxposure to bad company then makes him a 
lo*»sf livL-r all his days; thastity kept at first makes the s;ime easy later 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 23 

on. In all pedagogy the great thing is to strike the iron while hot, and 
to seize the wave of the pupil's interest in each successive subject before 
its ebb has come, so that knowledge may be got and a habit of skill 
acquired — a headway of interest, in short secured, on which afterward 
the individual may float. There is a happy moment for fixing skill in 
drawing, for making boys collectors in natural history, and presently 
dissectors and botanists; then for initiating them into the harmonies of 
mechanics and the wonders of physical and chemical law. Later, intro- 
spective psychology and the metaphysical and religious mysteries take 
their turn; and, last of aU, the drama of human affairs and worldly 
wisdom in the widest sense of the term." ('90, II, pp. 400-401.) 

Facts seem to support this aspect of the nature of instincts more 
than the theory of the sudden appearance. The law of disuse of 
functions is necessarily in general support of the theory. Any 
fiinctions will, as a rule, be strengthened through exercise. The 
assumption, however, that instincts, if not exercised when they first 
manifest themselves, will become dormant beyond the possibility 
of reawakening, or that they actually become dormant, is question- 
able. James gives isolated instances in favor of his viewpoint. 
Experimental and comprehensive observations are missing at the 
present time. Isolated illustrations of the opposite viewpoint, 
however, also are to be found. Thus it frequently happens that 
through the change of circumstances of life, instincts apparently 
long dormant or never given opportunity to manifest themselves, 
will quickly appear for action. For example, the writer has a friend 
who as a boy had never developed the instinctive tendencies in- 
volved in fishing. About the age of thirty, through the opportuni- 
ties of a new environment, the instinct appeared so strongly that 
he will go to great lengths at any time of day or night to follow this 
sport. But isolated instances are dangerous bases on which to 
generalize, and future inquiries will have to solve the problem. 
Many instincts apparently are dormant only because no opportu- 
nity of expressing themselves are at hand, or because other more 
dominant interests prevail, but may, when appropriate circum- 
stances arise, rapidly appear for action. 

The Recapitulation Theory of Instincts. The principle of re- 
capitulation was formulated by biologists to account for the develop- 
ment of animal organisms in the early stages of their growth. 
The theory of recapitulation holds that the individual retraces in 
its growth the successive stages of development of the entire animal 
series. Thus Hall says: 



24 EDUCATIONAL PSYCHOLOGY 

"Holding that the child and the race are each keys to the other. I have 
constantly suggested phyletic explanations. ..." ('04, I, p. \'III.) 

"The best index anil guide to the staled activities of adults in past 
ages is found in the instinctive, untaught, and non-imitative plays of 
children. ... In play every mood and movement is instinct with 
heredity. Thus we rehearse the activities of our ancestors, back we know 
not how far, and repeat their life work in summative and adumbrated 
ways. It is reminiscent, albeit unconsciously of our line of descent, and 
each is the key to the other. . . . Thus stage by stage we enact their 
(our ancestors') lives. Once in the phylon many of these activities were 
elaborated in the life and death struggle for existence. Now the elements 
and combinations oldest in the muscle history of the race arc re-presented 
earliest in the individual, and those later follow in order." ('04, I, 
pp. 202-203.) 

And Puffer says: "We are by turns vertebrates, gill-breathing verte- 
brates, lung-breathing vertebrates (we make the great change at birth), 
little monkeys, Uttle savages, and finally civilized men and women." 
('12, p. 77) 

The e\'i<U'Tices ;:jiven for the principle of recapitulation are largely 
embrj-olo^^ical and structural. Vestigial organs such as the vemai- 
form appendix, gill slits, etc., are further cited as evidences of the 
remainder of structures once useful. The facts seem to be that 
such recai)itulation as takes ])Iace is very brief and confined almost 
wholly to the prenatal ])eriod of an individual's develo])ment. 
Davidson, after a comprehensive review of the biological evidence 
for the theory, concludes thus: 

"The history of recapitulation is an instructive one. A principle of 
limited application within the field of its origin was elevated to a position 
of wide generality, and so gave rise to a conception in the main mislead- 
ing. Carried into a new territory without a sufikient examination of 
its merits, it was applied broadly as an explanatory principle and thus 
distributed its misleading influence beyond its own borders." ('14, p. 99.) 

"A more thorough consideration of the facts has led to a view of de- 
velopment essentially contradictory to this recapitulatory one. Ontogeny 
represents the ancient life-cycle which as such has been transmitted from 
the beginning. The chronological secjuence from egg to maturity is not 
a rehearsal of a like historical series of events throughout the i)hylogeny 
of species; it is but the recurrence of an order which has been repealed 
in the lifetime of each individvial from the beginning. In general, the 
effect of the modifications induced by germinal mutations and selection 
in the successive ontogenies, to make them over, and to destroy the 
resemblance of later ones to their predecessors." 



INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 25 

The recapitulation theory with its pedagogical corollate, the 
culture epochs theory, has been developed largely as an analogy 
with many of the analogues missing. Its usefulness for educational 
thinking seems to the writer to be greatly exaggerated. It has 
built pedagogical mountains out of biological molehills. It is 
primarily an anatomical principle proposed to account for the 
embryological development of biological organisms, and has been 
brought over into human behavior to explain on the one hand, the 
order and dates of appearances of instincts, and to furnish on the 
other hand, a basis for the order and dates of teaching subjects in 
the school curriculum. The former assumption is more or less 
dubious, since most, if not all, of the demonstrable recapitulation 
occurs before birth; and the latter assumption is quite certainly 
dubious, since the anatomical and probably also the functional 
recapitulation has long ceased when the child begins his definite 
education. 



CHAPTER III 

VARIATION IN HUMAN CAPACITIES 

How May They be Measured and Represented? Differences 
among human beings are quantitative rather than qualitative. 
That is, all human beings have the same reflexes, instincts, and 
capacities; all have the powers of perception, discrimination, at- 
tentiveness, retentiveness, reasoning, and so forth. Ail persons, 
consequently, have the same general qualitative make-up. The 
variations from person to person are, therefore, primarily differ- 
ences in the strength of the various abilities that each individual 
possesses, and in the manner in which amounts of the various 




Fig. 7. — Distrihiilion of numory ability of 173 University students. The 
test consisted in diiliilin^ ten monosyllaljic nouns. The [K-rsons then recorded 
the words that they remembered. The horiz<jntal axis indicates the numlx-r of 
words and the vertical a.xis indic;ites the number of persons ha\ ing each memory 
ability. 

36 




-> 



40 



50 



60 70 80 

Letters per Minute 



90 



Fig. 8.— Distribution of ability in the A-test. Based on 164 University stu- 
dents. The horizontal axis represents the number of A's canceled in one minute; 
the vertical axis represents the number of persons of each ability. 




3 4 



13 



5 6 7 8 9 10 IL 
Eigures per Minute 

Fig. 9. — Distribution of ability in canceling a certain geometrical figure in a 
page of figures. Time allowed, one minute. Based on 164 persons. 



a8 



EDUCATIONAL PSYCHOLOGY 



trails combine in the same [)erson. The ditTerences are quaUtative 
only in the sense that combinations of \arying amounts of diverse 
traits occur. 

The most convenient manner in which to rej)resent and deter- 
mine the amount of variation in a given trait is by means of the dis- 
tril)ution curve, or the surface of fret|uency. The distribution 
cur\'e is a curve designed to represent how fre(|uently each amount 




20 30 40 50 

Associations per 15 Seconds 

Fjc. io. — Distribution of ability in K'i^'nk' associations in response to a 
stimulus word. The horizontal axis f^ises the number of words >jiven in 15 sec- 
onds. The vertical a.\is gives the number of persons of each ability. Based on 
13s i)crsons. 

or strength of a given trait occurs in a given group of persons. The 
range of abihty fron\ a small amount to a large amount is repre- 
sented along the base line, or .\ axis, from left to right, and the 
number of times each ])articular ability occurs is rejiresentcd ver- 
tically along the ordiiiates, or y axis. (.See Figures 7, S, 0, and 10.) 

How Wide are the Differences? The investigation of this 
problem in recent years has brought out the fact tiiat the differ- 
ences among human beings are very much greater than has com- 
monly been thought. If we measure a group of jjupils in a given 



VARIATION IN HUMAN CAPACITIES 



29 



class or grade, we find that on the average the best pupil is able to 
do from two to twenty-five times as much as the poorest pupil, 
or is able to do the same task from two to twenty-five times as well 
as the poorest pupil. The accompanying table shows the range of 
differences between the highest and the lowest in a series of tests 
made upon fifty university students. 



TABLE I 

Range of differences between the best and the poorest in a series of mental 
tests. Based upon the writer s Experiments in Educational Psychology, 
page 8, which may be consulted for the nature of the tests. 



Best 
Record 



Poorest 
Record 



Ratio 



Memory span 
Memorizing. . 

E Test 

Er Test 

Opposites. . . . 
Genus-species 

Addition 

Subtraction. . 

Average 



8 words 
I min. 

25 sec. 
I min. 30 sec. 

30 sec. 
45 sec. 

31 sec. 
20 sec. 



4 words 
4 min. 

I min. 30 sec. 
3 min. 25 sec. 

1 min. 50 sec. 

2 min. 5 sec. 
2 min. 

I min. 30 sec. 



I : 3-35 



What is the Nature of the Variation? From the general ap- 
pearance and form of the distribution curves of mental traits, we 
note that abilities range without break from the lowest to the 
highest. Our common terminology of dividing groups of persons 
into various classes as dull, mediocre, and bright, on the assumption 
that they may be divided into distinct classes with gaps between 
them, is psychologically incorrect. The fact rather is that all 
grades of ability, varying by infinitesimally small amounts from 
the lowest to the highest, are found in the human species. 

The next conspicuous feature about the nature of the distribu- 
tion of mental abilities is the general shape of the curve. This 
indicates that the large majority of individuals cluster about the 
center. In the accompanying illustration it will be noticed that 
if the entire range of abilities is divided along the base Hne into 
three equal sections so that we may designate the one at the right 
as the superior section, the one in the middle as the medium sec- 
tion, and the one at the left as the inferior section, we find that 



30 



i:i )l tA riONAL rSYCHOLOGY 



approximately two-thirds, or db'/'o of all persons fall into the middle 
third; one-sixth or if'o fall into the superior one-third, and the 
remaining one-sLxth, or if/i fall into the inferior one-third of the 
range of abilities. In other words, the normal distribution curve 
is a symmetrical, bell-shaped figure, having its mode in the center 
and dropping at first rather gradually, then very rai)idly and finally 
\ery slowly. The statement attributed to Lincoln that "God must 
have loved the common people because He made so many of them" 
is i)sychologically true. If the middle third of the entire range 
of abilities represents the common people, then two-thirds of all 
persons are common people. 




32 to 40 45 48 

Inches 

Fig. II. — Distribution of chest measurements of English soldiers. 



The third interesting fact to be noted is that psychological and 
bi(jlogical traits vary universally in the same manner in conforniity 
with the normal, bell-shai)ed curve. Note for exam|)le the dis- 
tril)ution of such biological traits as chest measurement, height, 
girth of head, and so forth, as represented in the accompanying 
graphs, Figures ii, 12, and i,^. The number of men who are ex- 
tremely tall or extremely short is very small, and the number less 
tall or less short is larger and larger as the median is being ap- 
proached. 'I'his uniformitN throughout organic nature is an in- 
teresting and significant fact. Ai)parently nowhere are there traits 



VARIATION IN HUMAN CAPACITIES 



31 



which are discontinuous so that gaps would exist within the ranges 
of the traits, nor do we find that, on the whole, traits are distributed 



55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70" 
Inches 

Fig. 12. — Distribution of the height of 1,052 women. 

in a skewed manner, so that the great majority of individuals 
would lie either in the upper or lower range of abilities. 




51 52 53 54 55 56 57 58 59 60 61 
Head Girths in Centimeters 



62 



Fig. 13. — Distribution of the head girth of 1,071 boys, 16-19 years of age. 

Finally, the variation in both psychological and biological traits 
occurs apparently according to the law of chance, that is, according 



32 



EDUCATK )\ \l. PSYCHOLOGY 



lo the frecjuency of occurri-iuc- of ;i c haiur tvint. Consequently, 
on this assumption, the statistical treatment of the distribution of 
mental abilities becomes subject to the mathematical j)ro[)erties of 
the probability integral. What we mean by the statement that the 
variation occurs according to the law of chance may be illustrated 
in the following manner: If we toss up ten pennies at one time, 
count the numi)er of heads up and keep a record of it, then repeat 
the tossing a thousand times and keep a record each time, it will 
be found that the number of times no heads are up will occur very 




Fio. 14. — Distribulion of the nunihorof heads u[) in tossing; ten pennies i.ooo 
times. The horizontal axis >,'ives the number of |)ossil)le heads up in each tossing; 
the vertical axis gives the number of times each number of heads was U|). 

rarely, likewise, the number of limes all ten heads are uji will occur 
very rarely, the number of times one head is up or nine heads are 
up will t)ccur less rarely, and as you ai)j)roach from either side 
toward four, five, and six, the occurrences will be more and more 
frequent. The actual records of a thousand such tossings are 
represented in Figure 14. It would seem as though nature, in the 
production of her creatures, aimed at a target. The largest num])er 
of trials strikes somewhere near the bull's-eye, a smaller numl)er 
strikes within the ne.xt circle, and a still smaller number within the 
next circle, and so on. The corresjumdence between the actual 
tlistribution of abilities and the values of the i)robabilily integral is 
exceedingly useful in permitting statistical treatment of series of 



VARIATION IN HUMAN CAPACITIES 



33 



measurements of any trait. Figure 15 gives the mathematical 
or theoretical probability curve. 

Variation in Abilities in School Subjects. The differences in 




Fig. 15. — The theoretical probability curve. 

abilities in school subjects are fully as wide as in special psycho- 
logical capacities. They are probably due primarily to native 



Grade 8 




• • • • 


• • • • • 


• •«• • 








Grade 7 


• 


• • ••• 


• • • 


^ , 


• 






Grade 6 


• • 


.: : 


.:.: 


• •• 


• • • 






Grade 5 




• , 


• 

• • 
• • • 


• • • 


«• • 


• • 




Grade 4 




• • « •• 




. 








Grade 3 


. . 


• • I • 












Grade 2 


• . 


•• •••• 


• • 











1.0 



3.0 



4.0 



5.0 



6.0 



7.0 



8.D 



Fig. i6. — Distribution of pupils (in one school) in grades 2 to 8 in reading 
ability as measured by the author's tests. The horizontal axis represents speed 
and comprehension combined in terms of speed, i. e., words read per second. 

ability rather than to differences in opportunity, training, or en- 
vironment. Table 2 shows the range of difference in ability 



Grade 8 










.:•:! 


... I 

• • •• •• 


• •• • 


•• 


• 


Grade 7 




• 




.. : 


. t 

••• •• 


••••• 


••• ••• 


• • 


• 

•• 


Grade 6 






•• 


• 
• • 


•• • 


• • ••• 


•«• 


• 




Grade 6 




• 


• •• 


• 
• • 


• 
• 
• •• • 


• • 








Grade 4 






•• •••• 


• • 

• • • • 
••• •••• 


• 
• • 


• 








Grade 3 


• 
•••••• • 


:. 


• 














Grade 2 


• 


. :. 


• 















90 



100 110 120 

writiii;,'. S[)cc(l and 



20 yO 40 50 60 70 80 

T-'iG. 17. — Dislrihution anrl overlap[)ing of puf)il 
quiilit}' combined into one score as explained in tlie author's Educational 
Measurements. The numbers along the horizontal axis represent speed and 
C|ualit_\' in terms of speed, i. c., letters per minute. Quality was measured by the 
Thorndike scale. 



Grade 8 








• 




. . : 


. i . . 




Grade 7 






• 




• 


• • 


. : 

• • • 


:» . : 


• 


Grade 6 








• •• 


•: 


••••• •• 


•• • » 


1.: .: 


: . 


Grades 








• 
• • • 


••••••• a 


• • 


• 

•• •• • 


• 




Grade 4 




• • 


••• • 


• 1 
• • 
• • • • 


• 


..:.!?: 


• 


• 




Grades 


• 

• • • 


•:• •: 


• • 

• • • •• 


1 :. .. 


• 
• 


• 








Grade 2 

• 
• •• 


• • • • • 


•ff •• •• 


' • • 


• • • 













10 20 aO 10 00 60 7Q tft) W 100 

Fig. 18. — Distribution and ovcrlapi)inK of [)upils in spelling as measure*! by 
the author's test. The numbers along the horizontal a.\is are the numbers of 
words sfKlled out of a list of 100. 



VARIATION IN HUMAN CAPACITIES 



35 



in various school subjects as found in a class of 36 eighth-grade 
pupils. Abilities in reading, arithmetical reasoning, spelling, 
grammar, and history were measured by the author's tests ('16). 





Grade 8 

• 


• • 

• • 

• • 
• • • 


' • • • 

• • • 

• • • • 
I • • • • 






Grade 7 


• 
• 

• 

• • 

• • 

• • 

• • • 

• • • • 


' • • • 
• • • • 






Grade 6 


• • 

• • • 

• • • 

• • • 


• 

• • • • 






Grade 5 

• 
• 
• 


• • • • 

• • • • 

• • • • 

• • • • 

• • • • 








Grade 4 










Grade 3 


• • • 







10 



15 



Fig. iq. — Distribution and overlapping of pupils in ability to solve arith- 
metical problems as measured by the author's Scale A. The numbers along the 
horizontal axis are the steps on the scale. 

Quality of writing was measured by the Ayres scale, the four 
fundamental operations in arithmetic by the Courtis tests (series 
B), and composition by the Hillegas scale ('12.) 



36 



EDUCATIONAL PSYCHOLOGY 



TABLE 2 

Ranp;es of difference between tlic IkjsI and the poorest in a class of .^6 eighth 

grade pupils. 



Best 



Poorest Ratio 



Reading — Sjx-cd 

Reading — Comprehension 

Writing — Speed 

\\" riling — Quality 

Arithmetic — Reasoning 

Arithmetic — Addition (Rights) 

Arithmetic — Subtraction (Rights) 
Arithmetic — Multiplication (Rights) 
Arithmetic — Division (Rights) .... 

Spelling 

Composition 

Grammar (Scale A) 

History 

Average 



6. 

76. 
io«. 
90 
15 
15 
17 
17 
16 
90 
70 

13 
104 



I 


8 1:3.7 


22 


1:35 


57 


1:1 6 


60 


':i 5 


2 


17-5 


I 


1:15 


2 


1:8.5 


I 


1:17. 


2 


1:8. 


45 


i;2. 


30 


1:2.3 


6 


1:2.2 


4 


1:26. 



1 :7.6 



In the accompanying diagrams, Figures 16-26, the complete 
distril)Ution of the abilities of the puj)ils in each grade in the sub- 
jects of reading, Avriting, spelling, etc., are shown as determined 
by methods of measurement described else^vhere. These graphs 
show that the range from the lowest to the highest ability in any 
given subject within any given grade, is aj)pro.\imately as great 
as that found for special mental functions referred to in a ])receding 
section. The best j)upil in reading or spelling or any school subject 
is from one and a half to twenty-five times as capable as the poorest 
pupil. As a result of this wide range of abilities, there appears an 
enormous amount of overlapi)ing of the abilities ])ossessed In- the 
pupils in other grades in the same school. Thus it will be noted 
that the best pupil in arithmetical reasoning in the third grade is 
as capable as the poorest i)upil in the eighth grade. All j^upils 
had been tested l)y the same set of problems. The Siime statement 
a})plies with practically identical details to any school subject. 
Putting the situation in a little different statement, it has been 
shown that 60% of the best pui)ils in any given grade could be 
e.xchanged with the 60% of the poorest pupils in the ne.xt higher 
grade, with the result that there would be no change in average 
ability of the two grades. 



VARIATION IN HUMAN CAPACITIES 



37 



The question next arising is this: Granting that the range of 
ability in any one subject is as large as the results of the tests show 
it to be, may, however, a given pupil not be two or three years 
ahead of his grade in arithmetic, two or three years behind his 
grade in spelling, up to the average of his class in reading, etc., 
and may he not be placed correctly, after all? The facts, however, 



• • i 

• • 

• • • 


I • • 

< • • 

• • 

1 • • • 

I • • • • 


Grade 

• 
• 


8 

1 


• 
• 

• • 

• • • 

• • • 

• • • < 

• • • • « 

• • • • < 


( 

e 

1 • • • 


Grade 

• 

• • 


7 


• • • 
• • • • • 


• • 

• • 


Grade 

• 


6 






• 
• 
• 
• 
• 

• • 

• • • 

• • • 


• 


Grade 

• 


5 




> • • • 


a-CA^ 




Grade 


4 


t I I I I \ 


> • 


« O 9 • • < 


> e • 



5 10 15 

Fig. 20. — Distribution and overlapping in addition as measured by the 
Courtis test. Ttie numbers along the horizontal axis represent the number of 
examples done correctly. 

seem to be as represented in the accompanying illustration, Figure 
27, in which a combined score was obtained for each pupil as fol- 
lows: In reading and writing in grade i; in reading, writing, and 
spelling in grade 2; in reading, writing, spelling, and arithmetic 
in grades 3 and 4; and in reading, writing, spelling, arithmetic, 
language, and composition in grades 5 to 8. Even when the varia- 
tions in abilities in different subjects possessed by the same pupil 



38 



EDUCATIONAL PSYCHOLOCY 



arc- C()uiitt'rl);ilaiuc{l aiul avi-ragcd, the range of abilities and tin 
overlappitij^ is ]tra(li(ally as large. It Avill he noticed, for e.\- 
amjjle, that the hist i)U])ils in the second and third grades in these 
three subjects coiiihined, are almost U|) to the abiiit)' of the poorest 
pupils in the eighth grade. The fact is that in every eighth grade 



Grade 8 


• * • 


• • • • 


* • • 


* • • • 


• * * 


Grade 7 






• • • 

• • • 




• • % 


Grade 6 


• • • 

• • • 4 




• 
• 


Grade 5 


• • • 




• 



b 10 15 

Fig. 21. I )i^tril)iiti()n and ovcrlapjjinK in use of correct Knglish as measured 
by the author's (irammalual Scale A. The numbers along the baseline are 
the steps on that scale. 

one pu|)ii in nine is fully ef|ual in ability to the average ability of 
the pu|)ils in the second year of high school and could do the work 
e(|ually well if he had been allowed to go on rai)idly enough to be 
in thi- second year of high school. Two pu|)ils in every nine are 
equal in ability to the average pupil in the first year of high schot)l, 
three of tin- nine pupils are correctly placed in the eighth grade, 



VARIATION IN HUMAN CAPACITIES 



39 



two are equal only to the average seventh grader, and one is equal 
only to the average sixth grader. Thus by proper promotion or 
classification, one pupil in every nine could save two years in eight, 
and two pupils in every nine could save one year in eight. 

Expressing the same facts in a different form for the school 
population as a whole, we may say that: 



I pupil in loo could finish the 8 grades in 4 yrs. or at 10 yrs. of age. 



2 pupils 




9 

21 " 


C ( 


33 "■ 
21 " 




9 " 




2 " 




I pupil 





II 






12 






I.S 






14 






15 






16 






17 






18 







Grade 7 


« • • « 


• 
• •• 


• • 

• ••• • 


• •• ••• 




• • 








Grade 6 




• • 

• •• • • 


• 


• • • • 




• • •• • 








Grade 5 


: :::: 


• • • 

















10 20 30 40 50 60 70 80 90 100 

Fig. 22. — Distribution and overlapping in geography as measured by the 
author's geography test. The numbers along the horizontal axis are the scores 
in the test. The situation in the case of history is very similar. 

The last two groups are composed of pupils so retarded that 
they probably never would or could complete the elementary 
school. The variation in ability is so great that the children of 
any given age are spread out over about nine years of maturity. 
For example, children ten years old range in ability all the way 
from fourteen-year-olds to six-year-olds or less, and the numbers 
of pupils at each age of mentality are approximately those given 
above. These facts are further borne out by recent tests of in- 
telligence, (See Chapter VII.) 

This enormous range of ability and the resulting overlapping of 
successive grades, is probably the most important single fact discovered 
with reference to education in the last decade. The import of it is so 
significant of the situation as it exists in our schools to-day and of 
the possibilities in the direction of the proper reclassification or 



40 



EDUCATIONAL PSYCHOLOGY 





Uradc s 




1 M U ^ 

• mi • 


• 




Grade 7 




• 
• 


1 

• 










Grade C 


• • 1 


• ' 


• • 

• • 

• • •■ 


• 

• • 

• • 

• • 


• 
• 






Grade 5 

• 


• 




• 
• 
• 


• • 

• • 


• 







;X) 



40 



60 



80 



Fio. 23. — Distribution and ovcrlappinp in ability to write a coni|)osition as 
rated by the Hillegas scale. The numbers along the base are values on that 
scale. 



Grade 8 








Grade 7 

• • 

• • 


: . : 

• • • 

• • ■ 

: :.: :. 


! ■ • • • 




Grade 6 


• 
• 

: : 

• • • • 
••••••• 


• 

• • • 




Grade 5 


• • 

• • • 

• • • 

• • • • 

. : ::; 


• 

• • 

• • 




Grade 4 


: .:: : 


:: 

• • 
> • • • 




Grade 8 

• 

: 

• : : 

• • • 


• * 






Grade 2 

• • 

• • 

• • 

X : : 


I : : . 







i) 10 lb 

Fig. 24. — Distribution and overlappinR in drawinR ability. The numbers 
along the horizontal .i.\is arc the units of Thorndike's drawing scale. 



VARIATION IN HUMAN CAPACITIES 41 

readjustment of pupils according to ability that we have scarcely 
begun to realize how great the differences are or in what manner 
the readjustments may be made. 

Provisions Made in the School for the Variations in Abilities. 
Experimental work has drawn renewed attention to the possibilities 
of taking account of the enormous ranges of abilities such as are 





Year 4 

• 

• • • • • 


m 
6 


Year 3 \ \ 

• • • 

• • • • 


CO 

§ 

2 

(U 


Year 2 \\ , , 
• • • • • • 




• • • 

Year ll ,\ *,\\ ', 

• ••••••• • 

••••••••• • • 

1 1* .1 







20 



40' 60 

Scale Values 



100 



Fig. 25. — Distribution and overlapping of pupils in a high school in ability to 
write an English composition. The numbers along the horizontal axis are 
values on the Hillegas scale. 

found even in an ordinary class of supposedly homogeneous pupils. 
To keep an ordinary class of pupils together is no doubt very 
wasteful in time both for the gifted as well as for the stupid pupils. 
The gifted must listen to questions and explanations designed 
chiefly for the benefit of the dull pupils, but which the bright 
pupils already understand. The dull pupils, on the other hand, 
waste time by being dragged along too rapidly in the endeavor 
to keep the bright pupils occupied. 



4-^ 



EDUCATIONAL PSYCHOLOGY 



Tlu- plans which havr bt-en proposed for meeting the varjing 
abilities of pupils fall into two general classes: First, those which 
attempt to keep the pupils of a gi\en class together but vary the 
manner of instruction for the i)upils of difTerent capacities; second, 
those which keep the manner of instruction uniform but promote 
or retard pupils according to their achicxcments. 













































IB 


- 












10 


High School 
Seniors 






































m 
S 6 

D 
.5 


- 










• • 

• • 


I'm 
^ 






• 






• • • 


^16 


High 


School 










lU 


Fre 


shmen 






































6 


- 






























• 








• • 


fi 


1 1 










• • 

I 1 1 I 1 



1 -' 3 4 5 6 7 8 <J 10 11 12 la 11 15 
Steps -Grammatical Scale A 

Fin. 2Ct. — Distribution and ovcriuppinj; of pupils in a lii^;li school in ability 
in flis^riminati^^,' iHlwcen a)rrcct and incorrect Kn);lish. The- numbers along 
tlie horizontal axis arc the steps on the author s CJraniniatical Saile A. 

The princi])al schemes of the first general method which have 
been tried in various schools are known as the individual instruc- 
tion or Pueblo plan, the monitorial group plan, the extra-work 
j)lan, and the sujjervised study or Bataxia ])lan. The individual 
instruction plan was en\ployed by Suj)erintendent P. W. Search 
in Pueblo, Colorado, and consisted in the abolition of all class 
instruction and the substitution of indi\idual teaching according 
to the needs of the pupils. The monitorial group j^lan is carried 
out by dividing a class into several groups, usually three, according 



VARIATION IN HUMAN CAPACITY 



43 



to the abilities of the pupils, and by appointing a monitor for each 
group from among the members of the class. The extra-work 
plan consists in having recitation and class instruction chiefly for 
those who need it, and in assigning additional work to the capable 
pupils to be done at their desks. The supervised study plan de- 



Grade 


8 








• 


• 
* • 


• 

• • 

• • • i 


.:.. 


• 
• 
• 






• • • • 1 


• • • 


• • • 


Grade 


7 










• 






• 


• 






• • • 


' • • 


• • • • 


• • 


Grade 


6 






• • 


• 
• • • • 


• 

• • • 

• • • • 


« « • 


< • • • 


• 




Grade 


5 






• 


* 

• • 

• • 


• 

• • • • 

• • • • 


• • 

• • • • 








Grade 


4 


• 


• ' 


• 

• • 
• • • 


• • • • 

• • • • 

• • • • 


• 
• • • 










Grade 


3 




• 1 

• • • • 1 

• • • • 1 

• • • • ' 


• 
• 

• • • 

• • • ' 


• 


» • • 










Grade 


2 




• 


• • 














• • • 


• • • • ( 


• • 


Grade 


1 

• 
• • 1 


• ( 

• • < 

• • • • < 


• • • < 
• ••• 


• • 


• 













0.5 



1.0 



1.5 



2.0 2.5 3.0 3.5 
Average Scores 



4.0 



4.5 



50 5.5 



Fig. 27. — Distribution and overlapping of pupils when their attainments in 
different subjects are averaged. Reported in a thesis by Helen Craig in the 
library of the University of Wisconsin, 1918. 

votes a part of the class period to the usual recitation and instruc- 
tional work, and the remainder to study done under the supervision 
of the teacher. Sometimes the class period is considerably length- 
ened and no home study is done; at other times, the class period 
is kept at the usual length with some assignments for home study. 
These plans have been in operation in various schools during 



44 KUUCATIONAL rSVCFIOI.OdV 

the ])asl thirty years witli \arying amounts of success or failure. 
Most of them have been successful ^vhen carried out under the 
immediate sui)er\ision of the persons uho devised them. The 
difikulty, however, has usually been that when others have at- 
tempted to use them they have not been so satisfactory. Some 
of the schenu'S ha\e been objectionable on other grounds also. 
For example, the individual instruction i)lan is in part unsatisfac- 
tory because it removes a larj^e share of the social stimulus and 
interaction that is derived from class instruction. 

The one tv])e of pkm ^vhich is being adopted on an extensisc 
.scale and is ])roving to be generally applical)le, is some form of the 
supervised study plan. The methods with which this plan is 
carried out differ considerably and great care must be taken to 
avoid formality in the di\ision of the time between recitation and 
study during the class jjeriod and in the order and nianner in whidi 
the super\ision is carried out. A more detailed discu.ssion will 
be given in the chapter on "How to Study" where this subject 
])roperly belongs. 

The dilTerent schemes coming under the second general proposi- 
tion, namely, that of keeping the manner of instruction constant 
and varying the rate of promotion, have been applied widely, 
and many dilTerent plans designed to produce greater flexi- 
bility in the rate of ])romotion have been worked out in \arious 
school systems. As illustrations, two ])lans will be mentioned 
because they have been in successful operation for many years. 
In Cambridge, Mas.sachu.setts, there has been in operation a plan 
for some twenty years, in which the work of grades three to eight 
is laid out in three dilTerent courses of study. Pursuit of course A 
pennits the completion of the remaining six grades in six years; 
pursuit of cour.se B permits the completion of the work in five 
years; and the ])ursuit of course C makes jiossible the completion 
of the six grades in four years. Transfer from one course to another 
may take place at any time. 

In the St. Louis schools a method of |)romolion has been in force 
for a great many years which consists in dividing the school year 
into four quarters of ten weeks each. Promotion can be made 
at the end of each f|uarter. Pupils who have made a grade of 
excellent may be promoted to the next higher class at the end of 
any ten-week period, and pujjils who l)a\e made very low grades 
or practically failed, must repeat tluir work beginning with the 
various ten-week periods. 



\'ARIATION IN HUMAN CAPACITIES 45 

The effect of this plan in shortening the time of a considerable 
proportion of pupils is shown in a study made by W. J. Stevens. 1 
This investigation shows the length of time required by each of 
1,439 pupils in four elementary schools in St. Louis to complete the 
eight grades. 

TABLE 3 

The average attendance per grade of 1,439 pupils, graduates, required to com- 
plete each of the eight grades. Forty weeks is assigned in the course of 
study for each grade. 

Number of Average Ntjmber of Number of Average Number of 
Pupils Weeks to Complete Pupils Weeks to Complete 

Each Grade Each Grade 

1 17 33 4i 

2 18 49 42 
I 20 29 43 
I 21 27 44 
8 22 19 45 
8 23 20 46 

13 24 IS 47 

17 25 9 48 

19 26 5 49 

25 27 4 50 

46 28 4 51 

43 29 2 52 

52 30 2 53 

83 31 2 54 

103 32 I 55 
99 33 2 56 . 

109 34 2 57 

92 35 2 58 

no 36 I 59 

87 37 I 60 

104 38 2 62 
95 39 I 63 
87 40 2 70 

Median 35 weeks 

Total average time of attendance 288 .weeks 

Xo do 320 weeks' work 

Double promotions 17% 

Normal promotions 67% 

Repeaters 16% 

It will be noticed from this table that thirteen pupils completed 
the eight grades in an average of twenty weeks to do forty weeks 

1 Reported in a thesis (1914) in the library of the University of Wisconsin. The study 
was carried out under the direction of Professor E. C. Elliott. 



46 EDUCATIONAL PSYCHOLOGY 

of work, that is, in half of the prescribed amount of time. In other 
words, about i^^ of the pupils required approximately four years, 
6.3% five years, 22.8% six years, 34.6% seven years, 24.9% eight 
years, 7.6% nine years, 1.7% ten years, and 1.3% eleven to thirteen 
years to complete the eight grades. These results agree quite 
closely with the figures suggested on i)age 39. 

Promotion by subjects is a i)lan adopted in various schools. The 
])rogram must be arranged so that all grades recite in the same 
subject at the same ])criod in order that a pupil may do his work 
with the ]>articul;ir class to which he belongs. For example, a fifth- 
grade pupil might recite in spelling with a seventh-grade class, in 
reading with a sixth-grade class, in arithmetic with a fourth-grade 
cla.ss, and so on. 

In high school work there is equal need for flexibility in progress. 
Plans should be dc\ised whereby a class could be divided into 
three sections, a rapid, a normal, and a slow section. For example, 
an algebra class, after some early tests, could be divided into three 
divisions. Section A could easily do the year's work in two-thirds 
of the year and then pass on to geometry or more ad\'anced algebra 
or even some other subject. Section B could do the normal work 
in the year, and Section C could take a year and a third to do the 
normal year's work, or could cover only two-thirds of the ground 
in the year and receive only two-thirds credit. Differences in 
ability are sufliciently great to make ])Ossible as much dif- 
ference in progress as is here indicated. The more capable 
] )upils could easily shorten their high school course by half a year 
or a year. 

A plan of flexible promotion that can be administered success- 
fully has in niany respects distinct advantages over any plan which 
merely varies the instruction for the reason that it allows the 
capable pupil really to gain the advantage of his ability; because 
he is able to shorten his elementary school period, which is one of 
the aims striven for at the present time. The elementary school 
( ourse is considered too long. Any plan which varies the method 
of instruction so as to require more work of the cajnible jiupil no 
doubt occui)ies the time of these pupils and gives them the benefit 
of the additional work achieved, but it does not give the pupil the 
full benefit that lie de.servi'S in accordance with his capacities. In 
practical life the capable man performs several tinies as much 
work or makes several times as rapid |>rogress in the same perio<l 
of time as the incapable man, both having equal opj)ortunities. 



VARIATION IN HUMAN CAPACITIES 47 

Why should not the schools permit progress according to ability 
and achievement? Greater flexibility in promotion or retardation 
is an advantage both to the more gifted and to the less gifted pu- 
pils. The former will be able to step forward whenever they are 
ready and the latter will not need to step back so far whenever a 
part of the work must be gone over again. Promotion once a year 
works to the disadvantage of both types of pupils. The bright 
pupils cannot well jump an entire year and so will not be able to 
progress as rapidly as their abiUties warrant, while the slower 
pupils will have to repeat an entire year when a quarter or half of 
a year would be sufficient. School progress is determined too 
much by the calendar and not enough by capacity. The most 
capable one-third of pupils are advanced too slowly, and the least 
capable one-third are advanced too rapidly. A saving of half a 
year or a year on the part of a fourth or a third of the pupils would 
be of inestimable value to the pupils themselves and to the com- 
munity at large, either in getting an earlier start in their life 
work, or, preferably, in securing more advanced and thorough 
training. 

Finally, one of the most important, if not the most important 
aspect of the principle of progress according to performance, is its 
appeal to the individual to do the best he can. Few incentives are 
as strong as the personal impulse of going forward as rapidly as 
possible and of putting forth the best that is in one. If a child 
knows that, if he can spell as well as the pupils in the grade above 
him, he will be put with them, he will be stimulated as he would 
be in no other way to reach that degree of attainment. Likewise, 
if he knows that he is likely to be put back to recite in speUing 
with the pupils of a lower grade if he falls behind, he will put forth 
his best efforts to hold his own. Dawdling could hardly be en- 
couraged more than it is in many of our schools. Rewards in 
adult life are more nearly according to ability and performance. 
The same conditions would work to the advantage of school 
life. 

The schools have given special attention to the backward pupils 
by organizing separate classes for them and by giving them extra 
help, but they have given little or no attention to the advanced 
pupils. Society would be compensated far more for paying at 
least equal attention to the gifted pupils since they primarily will 
determine the future progress of mankind. The leaders of society 
will come from the right end rather than from the left end of the 



48 EDUCATIONAL PSYCHOLOGY 

ilistrihution ciirvr. Wi-^doni would dictate thai \vc tlcxoU- at 
Irast as much care and ihou^dil to thi-in, that we surround thcni 
with an atmosphere of hij,'h asi)iratioii and achievement and stim- 
ulate to the full their powers of originaHty and discovery. This 
would make for maximum i)rogress based upon ability and per- 
formance, not upon birth or social caste. 



CHAPTER IV 
CORRELATION AMONG HUMAN CAPACITIES 

Problem. Any given single trait varies over an enormously 
wide range among the members of the human race as a whole. 
The problem, however, before us now is: To what extent is a given 
amount of any capacity accomi)anied in general in the same person 
by equal, larger or smaller amounts of any other ability? To what 
extent is a good memory in the same person accompanied by an 
equally good capacity for reasoning or attention or perception or 
judgment? To what extent is poor or mediocre ability in memory 
accompanied by poor or mediocre ability in other directions? If 
all mental abilities were measured on a scale of o to lo, the con- 
crete problem would be: To what extent would a memory ability 
of 7 be accompanied in the same person by a perception ability of 
7, or a judgment ability of 7? If it is not accompanied by the 
same amount of other abilities, by how large or small an amount 
of any other ability is it accompanied? 

Educationally the problem is important and takes the following 
form: To what extent may we expect pupils, who are excellent, 
mediocre, or poor in one subject to be excellent, mediocre or poor in 
other subjects? To what extent is a statement such as the following 
true in general: "I simply cannot learn languages or mathematics, 
although I get along very well in my other studies" ? To what 
extent is freedom of electives in studies justifiable on the basis of 
variation in the combination of capacities in the same individual? 
To what extent are mental and physical traits correlated? To 
what extent are abilities similar at different times of life in the 
same individual? To what extent is ability in childhood or youth 
a forerunner of ability in adulthood? 

Methods of Measuring Combinations of Traits. The extent 
to which various amounts of abilities accompany one another is 
measured or expressed definitely by the coefficient of correlation. 
The value of the coefficient of correlation ranges from i.oo through 
o to -1.00. A coefficient of correlation of i.oo means a complete 
agreement. If the coefficient of correlation between ability in 

49 



50 EDUCATIONAL PSYCHOLOGY 

Latin and ability in German were i.oo, it would mean that the 
best pupil in Latin would be also the best pupil in German, the 
second best pu[)il in Latin would be the second best pupil in Ger- 
man, etc., dowTi to the poorest pupil in Latin who would also be 
the poorest in German. As the correlation drops farther and 
farther belo\v i.oo toward o, the closenos of this agreement be- 
comes correspondingly less until o is reached. If the coefficient 
of correlation between ability in Latin and ability in German 
were —I.oo, it would mean that the best pupil in Latin would be 
the poorest pupil in German, the second best pupil in Latin would 
be the second poorest in German, etc. As the correlation rises 
above —i.oo toward o the reversal becomes less and less until o is 
reached. A coefficient of o means that no relationship exists. 
A pupil might have any amount of ability in one subject and any 
other amount c)f aljility in the othiT subjcrt.' 

The Correlation Among Specific Mental Abilities. The early 
investigations in this field found surjirisingly small correlations 
even among apparently vcv}' similar or closely related capacities. 
Thus it was thought that a person might ha\e a good memory' for 
words but not for numbers or faces; he might ha%-e a keen percep- 
tion of words but not of letters or geometrical figures and the like. 
As typical of the earlier results on correlations we may cite a few 
from Wissler ('oi) as follows: 

Auditory memory of figures and visual mcmor>- of figures .:() lo .39 

" " " " " auditor}' '" " pas.>iagc .04 

" " " " " memory of length of line .00 
"passage" " " " " " -.07 

" " " " " quickness in naming colors .03 

" " " " " reaction time .12 

" figures " " " .17 

The significance of these coefficients may be interjireted approxi- 
niately as follows: A coefficient of o means that no correlation 
exists, and roughly speaking, a coefficient of .30 or less is small 
and practically means very little agreement. Correspondingly, a 
coefficient lying between .,^0 and .50 nuans a moderate amount of 
agrcLinent, a coefficient between .50 and .75 means a considerable 
correlation, while a coefficient above .75 indicates a ver>' close 

' For mcthofLs of rompiitinrr the cocfTK-icnt of correlation, consult the author's Fx- 
perimnUs in Educalional Psychology, (1017 Edition) Chapter I\': Whipple, MdniuU 
of Menial and Phyui'il Tfsls. ('h.i|)iiT III: Thorndikr. Stmlal and Social Measure menu, 
(hapttr XI; and Rugg, II. O., Stalislical Methods Applied to liducalion, Chapter IX. 



CORRELATION AMONG HUMAN CAPACITIES 51 

relationship and, as it approaches i.oo, indicates practically perfect 
agreement. From the table, it appears that many coefficients are 
very low and imply little or no agreement. The coefficients so 
low as to indicate practically very little correlation are stated to 
exist between auditory memory of figures and visual memory of 
figures. This would mean that a person might have a very 
good memory for figures seen but a poor memory for figures 
heard. 

The difficulty with these coefficients is that they are based upon 
unreliable and incomplete measurements of the traits concerned. 
Many of the measurements of the early investigations of correla- 
tions were derived from group tests which had been made but 
once. Measurements thus obtained have been shown by subse- 
quent investigations to be rather uncertain indications of the real 
amount of a given trait possessed by the individual. In order to 
obtain a fairly accurate measurement of a given capacity, it is 
necessary to repeat several times under favorable circmnstances 
the measurement of the trait in question. You cannot obtain 
anything like an accurate measure of any amount or quantity 
by a single measurement made under distracting conditions. If 
we should wish to measure the memory capacity of a given in- 
dividual, we should not consider the result very trustworthy if a 
single test were given consisting of eight lines of poetry learned 
in three or four minutes. We ought at least to repeat the test 
with several similar passages, preferably on different days, and 
derive therefrom an average. This is, in fact, the sort of procedure 
that has been followed in subsequent researches. Inaccurate 
measurements, as indicated by recent analyses of correlations, 
tend to reduce very materially the computed coefficients below 
the actual amounts of correlation. 

Recent researches have shown that among many traits qmte 
close, and among other traits very close, correlations exist. An 
investigation by J. A. Stevenson ('18) showed remarkably close 
correlations between various types of sensory discrimination. 
The plan of the investigation consisted in making extensive and 
repeated measurements of discrimination in lengths of lines, in 
intensities of sound, in degrees of brightness, in shades of gray and 
in pressures on the end of the first finger. The correlations com- 
puted on the basis of these measurements with ten persons were 
as follows: 



52 



EDUCATIONAL PSYCHOLOGY 



TABLF 4 After Stevenson ('i8). 

Lines and intensity of sound go 

Brightness and intensity of sound 90 

Pressure and intensity of sound 36 

Pressure and lines 3Q 

Lines and hrij,'htness 92 

Pressure and brightness 41 

A similar investigation in the field of memory, conducted hy 
Miss N. F. Bennett ('16), showed on the basis of numerous and 
repeated tests with nine subjects fairly close correlations among 
the capacities to remember various kinds of material such as 
syllables, numbers, nouns, prose, and faces, between visual and 
auditor}' presentations of the material, and between mediate and 
immediate learning. Her conclusions are stated thus: 

" I. There is a high correlation between mediate and immediate re- 
tention if a siiflicicnt number and variety of measurements for each 
l3pe of menior)' are taken, and the results amalgamated to determine 
ranks. 

"2. There is a high correlation between the memory span, or imme- 
diate retention for disconnected materials, and the ability to learn the 
siime." 

Holling\vorth made a study to determine the increase in the 
coefhcients of correlation among six dilTercnt capacities with the 
increase in the number of measurements made uj^on each aipacity. 
His results are set forth in the following table. They indicate a 
ver)' marked rise in the coeflicients with the increase in the number 
of tests. 



TABLE 5 

The average correlation of each test with all others at various |>oints in the 
curve of practice. After llollingworth ('12). 



Trial 


Addi.nc 


Opposites 


Color 
NxyiNG 


DlSCRtMI- 

NATtON 


Co-ordina- 
tion 


Tapplsc 


Final 
Average 


I 


10 


. 10 


•15 


-.07 


- 15 


• ' 7 


• 06s 


5 


4" 


.26 


15 


•35 


.21 


•32 


.2S0 


25 


SO 


•35 


4.S 


• 27 


•03 


35 


320 


80 


■ sa 


4i 


.V? 


.st 


.iS 


•34 


• 3QO 


205 


.48 


.62 


61 


• " 


.M 


■52 


.490 



CORRELATION AMONG HUMAN CAPACITIES 



53 



As an illustration of a series of correlations among special mental 
functions based upon measurements repeated several times but 
not as frequently as those in the preceding tables, we may cite the 
coefficients obtained by Simpson. These coefficients are unusually 
high because they are based upon tests performed on two ex- 
treme groups of subjects, the one a highly intelhgent group and 
the other distinctly unintelligent. 



TABLE 6 
Correlations among certain mental abilities. After Simpson ('12). 



1 . Ebbinghaus test . . . . 

2. Hard opposites 

3. Memory of words. . 

4. Easy opposites 

5. A-Test 

6. Memory of passages 

7. Adding 

8. Geometrical forms. . 

9. Learning pairs 

10. Completing words . . 

11. Drawing lines 

12. Estimating lengths . . 













cr 






























a 
H 


H 


Q 



H 





< 

< 




OS 



a 


Q 
a: 



< 

u 


S 


i 




> 


a 



0. 
0. 



> 






> 

a 

a 




z 

Q 


H 
W 

a 



Pn 

z 

z 




g 

H 
Id 
h) 

a 


W 


X 


S 


< 


< 


§ 


^ 


w 




1-1 



u 


I 


2 


3 


4 


5 


6 


_7_ 


8 


9 


10 


92 










92 


92 


















75 


81 


68 
















68 


76 


70 


71 














91 


86 


89 


69 


60 












71 


74 


56 


70 


67 


66 










54 


64 


67 


54 


94 


60 


44 








72 


72 


82 


43 


44 


63 


46 


40 






50 


70 


51 


50 


84 


38 


77 


61 


34 




26 


25 


06 


53 


27 


12 


27 


30 


04 


17 


52 


55 


59 


56 


57 


5« 


17 


35 


54 


22 



22 55 



Burt obtained the following correlations (Table 7) from a va- 
riety of tests of specialized mental functions made upon forty-three 
pupils. The test designated as dotting was regarded as a measure 
of voluntary attention; the tests designated as spot pattern, mirror 
and memory were designed to measure memorial and associative 
capacities; the tests called alphabet and sorting referred to sensori- 
motor capacities; dealing and tapping to motor functions; and 
the remainder to sensory discrimination. 



54 



EDUCATIONAL PSYCHOLOGY 



Dotting 

S[Kit pattern. 

Mirror 

Memory. . . . 
Alphabet. . . . 

Sorting 

Dealing 

Tapping 

Sound 

Lines 

Touch 

Weight 



TABLE 7 
After Burt ('09). 



•38 
•63 
.67 
•05 
•74 
.66 

•55 
•3S 
30 



•73 
58 
.12 
.40 

• 23 
.16 
.06 

• 14 



•57 
• 17 
.26 
-.08 
. I 



.09 



29 



•23 
.00 



49 



The import of the researches up to the present time seems quite 
certainly to prove that the higher mental capacities are on the 
whole rather closely correlated. The coefficients lie for the most 
part above .50, and some of them reach uj) to .So and .90. The 
same statement holds approximately for sensory cai)acities among 
themselves and also probably for motor capacities among them- 
selves. The cross-correlations among traits from these three levels 
is, so far as we can judge at the present stage of our knowledge, 
lower than among the traits within a given level. This seems to 
be particularly true of the correlation of motor cajiacities with 
hitellectual capacities. 

Correlations Among Abilities in School Subjects. The develop- 
ment of knowledge concerning this aspect of our problem has had a 
history similar to that of the special mental functions. The early 
correlations among abilities in school subjects were computed 
upon relatively uncertain data. About 1903, coefficients obtained 
by various investigators, were as follows: 



CORRELATION AMONG HUMAN CAPACITIES 



55 



TABLE 8 

Summary of coefBcients of correlation between abilities in high school subjects 
as reported up to about 1903. (Thorndike '03, pp. 26, 30-31). 

B = After Burris, based on nearly 1,000 pupils. 

P = After Parker, based on 245 pupils. 

Br = After Brinckerhoff, Morris, and Thorndike. 



History B . 

P. 

Br 
Science B . 

P. 

Br 
Algebra B . 

P. 

Br 
Drawing B . 

P. 

Br 
German B . 

P. 

Br 
French B . 

P. 

Br 
Latin B . 

P. 

Br 
Mathematics B . 

P. 

Br 



.40 
.62 




■41 




■41 


.40 


.58 


.56 


.26 


.61 



•55 

15 
.20 

•65 
•30 

•49 



.62 

•SO 
•39 

.09 



•38 



.16 

•49 
.42 

•58 

.43 
•43 
•44 
•33 

.26 



.40 



■33 
■30 

.62 
•58 



.44 
•54 
•35 
.41 

.07 



■52 



•54 



.06 



•30 



.40 



■33 



•38 



.48 



.40 
■31 



In the case of grammar school subjects, A. G. Smith (Thorn- 
dike '03 p. 13), computed the following correlations: 



English and Mathematics 39 

" " Geography 43 

" " Drawing 15 

Mathematics and Geography 36 

" " Drawing 14 

Geography " " 12 



56 EDUCATIONAL PSYCHOLOGY 

These coefficients for the most part indicate only a moderate 
amount of correlation. Thus Thorndike interpreted them in 1903 
by the following statement: "For our purpose the most striking 
thing about these figures is their small amount. It is safe to say 
that in a grammar or high school student a deviation from the 
average ability in any one subject implies by and large a deviation 
in any other not more than half as great. The most talented 
scholar in one field will be less than half as talented in any other, 
The most hopeless scholar in one field will in another be not so 
very far l)elow mediocrity." ('o.^, pp. 37-38). 

The coelVicients here quoted were based usually upon marks of a 
single teacher in any given subject. Recent studies have called 
attention to the unreliability of marks and the difTerences in stand- 
ards of marking employed by dilTercnt teachere. Sec Chapter 
XXII. This necessarily produces a considerable reduction in co- 
efficients based upon them. 

A computation based upon the average mark of each pupil in 
each subject in grades five to eight yielded the following coefficients 
(Table 9): 

TABLE 9 

Correlations among abilities in school subjects. After Starch ('13). 

Arithmetic and languapc S5 

" geography 83 

" " history 73 

" " reading 67 

" spelling 55 

Language and gcograi)hy S5 

" " history 77 

" reading 83 

" spclHng 71 

Geography and history 81 

" " reading 80 

" spelling 52 

History and reading 67 

" " spelling 37 

Reading and spelling 58 

These coefficients are almost twice as high as those previously 
quoted and represent very close correlations. They would warrant 
the interi)retation that the pupil who is g(X)d, mediocre, or poor 
in a given subject, is good, mediocre, or pcwr to very nearly the 
same, but not equal, degree in all other subjects so far as his abili- 



CORRELATION AMONG HUMAN CAPACITIES 57 

ties are concerned. Such lack of agreement as does exist is due 
probably to a difiference of interest and industry on the part of 
the pupil in different subjects at different times and to a real 
difference in abilities in the various fields. Thus spelling ability 
correlates apparently less closely with ability in other subjects 
than abihties in these other subjects correlate among themselves. 
The up-shot of the whole problem concerning the variation in the 
combination of traits, or the extent to which different amounts of 
mental traits accompany one another, may fairly be stated as 
follows: 

First, no negative correlations exist either among the abilities 
involved in school subjects or among the special mental functions 
measured by special tests. Popular and "short-story" psychology 
is false in the assumption and description of antagonisms of mental 
traits. They apparently do not exist among desirable and useful 
traits. Advice, given to prospective wives, such as "if he is good- 
natured, he may be lazy; if he is scholarly, he may be cold; if he is 
thrifty, he may be sting}^; if he is generous he may be wasteful," 
may produce caution, but it is not true psychology. Good-natured 
men are probably on the whole no more lazy than ill-natured men 
are, and scholarly men are probably on the whole no more cold- 
hearted than stupid men are. In fact the opposite is more likely 
to be true. And such statements as "Johnny is very bright in read- 
ing, but he simply cannot get arithmetic" is a soothing salve for the 
feelings of parents, but not apt to be sound psychology. 

Most of the opinions of students who state that they "simply can- 
not get" mathematics or language or history are in part probably 
due to a relatively small discrepancy in abilities, that is, to some- 
what less ability in mathematics, language or whatever the sub- 
ject may be; but to a larger extent they are illusory, because, when 
the actual facts are obtained or when more careful measurements 
of the abilities in various directions are made, the abilities corre- 
late much more closely than the student's statements would lead 
one to believe. As a concrete example the following case of a col- 
lege freshman, brought to the author's attention, may be cited: The 
student claimed that he had always had great difficulty in learning 
foreign languages but that other subjects were easy for him. He 
stated that in high school he never was able to obtain a grade in 
languages higher than about 75 but that in other subjects his grades 
were always high, as high as 95. Since his trouble seemed to be 
language it was thought that he might have a defective memory 



58 EDUCATIONAL PSYCHOLOGY 

or an abnormal type of imagery. Some memorj- and imagery 
tests revealed the fact that he had normal memory and imagery of 
average ability. This at once led to an inquiry into his actual 
high sch(X)l record to ascertain his grades. The various grades for 
any given subject are final grades in dilTerent courses as follows: 
Knglish, Sj, 80, 78, 81; History, 88, 75, 83; mathematics, 80, 87, 
77; science, 87, 87; Latin, 77, 79, 75; German, 75, 75. When these 
marks are compared there is little or nothing to ex])lain. His 
marks in Latin and German were somewhat lower than in other 
subjects, which is probal)ly largely e.\])hiined by his o\m statement 
that he "hated" languages, but they were not much lower on the 
whole. The highest grade in any subject was in the first year of 
history, 88, but he also had a grade of 75 in the second year of 
history and 77 in the 3rd course in mathematics. There was no 
grade of 95 in the entire list. This record was corroborated by the 
grades which he received at the end of the first eight weeks of his 
freshman year in college: Spanish, Fair; Geolog}', Fair; English, 
Fair; Mathematics, Fair; History, Poor. His abilities are pretty 
uniformly mediocre in all respects. 

Excejjtions do occur such as that of a boy seventeen years of age 
in the second year of the high school who was able to carrj' his work 
satisfactorily, but was able to read no more fluently, either orally 
or silently, than the average ])uy)il can at the end of the first grade. 
He was a normally intelligent boy. Such cases occur perhaps once 
among one or two hundred pupils, and may be regarded as ab- 
normal. 

Second, intellectual and scholastic abilities are for the most 
part closely correlated. Barring certain e.xceptions, which are 
rarer than is generally suj)posed, abilities are combined in fairly 
similar amounts. Intercorrelations between the different levels, 
intellectual, sensory, and motor, seem to be smaller and in some 
traits, jiractically zero. Some of the motor abilities, such as hand- 
writing, ha\-e practically no correlation with intelligence or general 
mental abilities. 

The wider bearing of the facts about the combinations of 
mentid capacities, together with the distribution of mental traits 
according to a continuous, bell-shaped curve discussed in the 
])rece(ling chapter, are deeply significant for the j)rol)lem as to 
whether there are distinct mental t)pes. Mankind apparently 
cannot be divided into three or four separate tjix'S. The ancient 
cla.ssification of tenii)eranunts into sanguine, choleric, melancholic. 



CORRELATION AMONG HUMAN CAPACITIES 59 

and phlegmatic, may be conveniently analogous to the four seasons 
of the year, spring, summer, autumn, and winter respectively, but 
there are no mental types that correspond to such superficial 
characteristics and none that are marked off sharply or even vaguely 
from one another. If all members of the human race were to be 
exhibited in a distribution curve whose base line represented from 
left to right different amounts of "sanguine-melancholic, or choleric- 
phlegmatic" natures, the curves would in all probability not be a 
series of four distinct curves separated from one another, nor even 
possess four modes with depressions between them, but would very 
likely be single continuous distribution surfaces of the usual normal 
form with one mode. The human beings who even remotely ap- 
proach any one type are very rare. The rule is that each person 
possesses more or less of all different traits, and within certain limits, 
roughly similar amounts of the various traits. Persons in whom 
the divergences are large are the exceptions rather than the rule. 

Correlation between Special Mental Capacities and General 
Intelligence. So far as definite data are available on this point, 
the inference may be drawn that many special mental functions 
are correlated anywhere from moderately to very closely with 
general intelligence. Men of intelligence have, on the whole, keen 
powers of perception, observation, and attention, remarkable re- 
tentiveness, exceptionally rapid and varied association processes, 
as well as unusually incisive powers of analysis and soundness of 
judgment. We may note here in passing, by turning to Chapter 
VII, the amounts of correlation of certain capacities with general 
estimated intelligence as found by Simpson, Burt, and others. 

The usefulness of the facts that many specific mental capacities 
are reliable symptoms or essential constituents of general intelli- 
gence will be particularly important in the future in the develop- 
ment of tests and methods of measuring intelligence. The value of 
this to mankind, not only in education but in all fields of human 
endeavor, can hardly be foretold at the present time. Further con- 
sideration will be given to it in a later chapter. 

Correlations between Mental and Physical Traits. In the case 
of adults, the correlations between mental abilities and such physi- 
cal characteristics as height, weight, size of head, lung capacity, or 
strength of grip, are either very low or zero. In the case of children, 
the situation is somewhat different, B. T. Baldwin made an elabo- 
rate study of 861 boys and 1,063 S^^^^ i^ the University of Chicago 
elementary and high school, the F. W. Parker school of Chicago, 



6o EDUCATION.\L PSYCHOLOGY 

and the Horace Mann School of Columbia University. Measure- 
ments of various physical characteristics were obtained at yearly 
and half-yearly intervals on two proups of pupils. One group was 
followed continuously through the ages from six to twelve, and the 
other from twelve to eighteen. A parallel comparison between 
the physical measurements and the school records of the same 
pupils was then made. From these results, Baldwin has deri\ed 
the following conclusion: 

Taller, heavier children mature physically in advance of the shorter, 
lighter ones. Those whose physiological age is accelerated complete 
the last grade of the elementary s( hool at 12 years, 9 5/6 months of age 
with an average of 84.39c, and those below average or of retarded phys- 
iological development, complete the elementary school work at 13 years 
7 4/13 months of age, with an average of 81.7%. (Bulletin of Bureau of 
Education No. 5S1, 1914. Page 82.) 

Correlations Between Early and Later Mental Abilities. The 

jirobiem here is, to what extent will a given pupil maintain his 
record of excellence, mediocrity, or stupidity all through his educa- 
tional career or all through his life? Will the puj^il who has high, 
medium, or low ability in the clementar}- school also have high, 
medium, or low ability in high school and in college? The first 
extensive study in this field was made by W. F. Dearborn (09) who 
traced through the high school and through the university the 
scholastic records of various groups of students, varj-ing in size 
from 92 to 472, and coming from eight large and four small high 
schools in Wisconsin. He divided the pujiils into four quartilcs 
according to their marks in high school, and then ascertained to 
what extent the pupils remained in the same riuartiles during their 
university course. His records showed that the pupils maintained 
the same records with remarkable consistency. He states his con- 
clusion in the following words: 

We may say then, on the basis of the results secured in this group 
(472 pupils) which is sufllcicntly large to be representative, that if a 
puj)!! has stood in the first quarter of a large class through high school, 
the chances are four out of five that he will not fall below the first half 
of his class in the university. . . . The chancer arc but about one in 
five that the student who has ilonc poorly in high school — who has been 
in the lowest quarter of his class — will rise above the median or average 
of the freshman class at the university, and the chances that he will 
prove a superior studciU at the university are ver)' slim indeed. . . . The 



CORRELATION AMONG HUMAN CAPACITIES 6 1 

Pearson coefficient of correlation of the standings in the high schools and 
in the freshman year, for this group of 472 pupils is .80. ... A little 
over 80% of those who were found in the lowest or the highest quarter of 
the group in high school are found in their respective halves of the 
group throughout the university. . . . Three-fourths of the students 
who enter the university from these high schools will maintain through- 
out the university approximately the same rank which they held in 
high school. 

F. 0. Smith made a similar study of 120 students entering the 
College of Liberal Arts at the University of Iowa. He traced their 
records from high school through the entire university course and 
found almost the same situation. Expressed in terms of coefficients 
of correlations, the results were as follows: 

TABLE 10 
Correlations. After Smith. ('12.) 
H. S. average and Univ. Freshman Average 



H. S. Average and Univ. Sophomore Average . 
H. S. Average and Univ. Junior Average .... 
H. S. Average and Univ. Senior Average .... 

ist and 2nd Year High School 

ist and 3rd Year High School 

ist and 4th Year High School 

University Freshman and Sophomore 

University Freshmen and Junior 

University Freshmen and Senior 



T. L. Kelley compared the marks of 59 pupils as they passed 
from grade five up into the first year of the high school. The extent 
of the agreement of their records in successive years is shown in the 
following coefficients of correlation: 

TABLE II. After Kelley. ('14)- 

Correlation between marks in the grades and marks in the first high school 

year. 

First Year of High School and 7th Grade 72 

First Year of High School and 6th Grade 73 

First Year of High School and 5th Grade 53 

First Year of High School and 4th Grade 62 

He then states: 

"The net conclusion which may be drawn from these coefficients of 
correlation is that it is possible to estimate a person's general ability in 
the first year (H. S.) class from the marks he has received in the last four 



62 EDUCATIONAL PSVCHOLX)GY 

years of elementary school with accuracy represented by a coefficient of 
correlation of .jSq, and that individual idiosyncrasies may be estimated, 
in the case of mathematics and English, with an accuracy represented 
by a coeflicient of correlation of .515. . . . Indeed, it seems that an 
estimate of a pupil's ability to carry high school work when the pupil is 
in the fourth grade may be nearly as accurate as a judgment given when 
the pupil is in the seventh grade." 

A study of the permanency of interests was made by Thorndike 
('12) by comparing the relative strength of interests and abilities 
during each of three periods of a person's school aireer, during the 
elementary school, high school, and college. These comparisons 
were made by asking one hundred individuals to estimate in ret- 
rospect, their relative interests and abilities in mathematics, 
history, literature, science, music, drawing, and manual work. 
Such data are necessarily subject to the errors of memory and 
judgment, but they are practically the only results available so far 
as strength of interests is concerned. Thorndike inferred from these 
estimates that early interests are not passing whims, but rather 
prophetic, with a fair degree of certainty, of later interests and 
abilities. He concludes that "A correlation of .60 or .70 seems to 
be appro.ximately the true degree of resemblance between the 
relative degree of an interest in a child of from ten to fourteen and 
the same person at twenty-one." "Interests are shown to be not 
only permanent but also symptomatic to a very great extent, of 
present and future capacity or ability. Either because one likes 
what he can do well, or because one gives zeal and effort to what 
he likes, or because interest and ability arc both SNTiiptoms of 
some fundamental feature of the individual's original nature, or 
because of the combined action of all three of these factors, interest 
and ability are bound very closely together. The bond is so close 
that either may be used as a sym[)tom for the other almost as well 
as for itself. The importance of these facts for the whole field of 
j)ractice with re.spcct to early diagnosis, vocational guidance, the 
work of social secretaries, deans, advi.ser, and others who direct 
students' choices of schools, studies and careers is obvious." 

The impression gained from all these investigations is that human 
nature is not a medley of capricious ca])acities which vary from 
year to year, but rather a fairly consistent combination of abilities 
throughout lifi . 



CHAPTER V 

SEX DIFFERENCES 

Educational Significance of Sex Differences. If we may judge 
fairly at the present time concerning the nature and amounts of 
differences between the sexes in mental characteristics, it would 
seem that the differences are so small in native intellectual abili- 
ties that they are almost wholly negligible in the education of boys 
and girls. That boys and girls ought to be educated differ- 
ently may very probably be desirable, but for reasons other than 
differences in ability. The professional, business, and domestic 
life of men and women makes it necessary to have different train- 
ing for boys and girls. But so far as the native abilities involved 
in school work are concerned, boys and girls might as well pursue 
the same courses from the first day of school to the last. 

Popular vs. Scientific View of Sex Differences. Probably more 
fallacious psychology of sex has been spread abroad by novelists 
and journalists than has been disseminated on any psychological 
question of popular interest. Occasional and extreme differences 
in individuals of either sex have been seized upon and exaggerated 
by descriptive phraseology and represented as though they were 
the normal divergences between men and women. Up to less than 
two decades ago, there was practically no scientific knowledge of 
the nature of sex differences available, and the statements of popular 
beliefs about such differences were hardly exaggerated by the sort of 
differences implied in the Sanscrit myth of the creation of woman. 

"In the beginning, when Twashtrai came to the creation of woman, 
he found that he had exhausted his materials in the making of man, and 
that no soUd elements were left. In this dilemna, after profound medita- 
tion, he did as foUows: He took the rotundity of the moon, and the 
curves of the creepers, and the clinging of tendrils, and the trembling of 
grass, and the slenderness of the reed, and the bloom of flowers, and the 
Ughtness of leaves, and the timidity of the hare, and the vanity of the 
peacock, and the clustering of rows of bees, and the joyous gaiety of sun- 
beams, and the weeping of clouds, and the fickleness of the winds, and 
the softness of the parrot's bosom, and the hardness of adamant, and the 
sweetness of honey, and the cruelty of the tiger, and the warm glow of 

63 



64 EDUCATIONAL PS^'CHOLOGY 

fire, and the coldness of snow, and the chattering of jays, and the cooing 
of the kokila, and the hyixicrisy of the crane, and the fidelity of iht- 
chakrawiika, and then coniiH)unding all these together, he made woman 
and gave her to man. But after one wtck, man came to him and s;iid: 
Ix)rd, this creature that you have given me makes my life miserable. 
She chatters incessiintly and teases me beyond endurance, never leaving 
me alone; and she requires incess;int attention, and takes all my time up, 
and cries about nothing, and is always idle; and so I have come to give 
her back again, as I cannot live with her. So Twashlrai said: Very well; 
and he took her back. Then after another week, man came again to him 
and siiid: I>ord, 1 find that my life is very lonely since I gave you back 
that creature. I remember how she used to dance and sing to me, and 
look at me out of the corner of her eye, and play with me, and cling to 
me; and her laughter was music, and she was beautiful to look at, and 
soft to touch; so give her back to me again. So Twashtrai said: Very 
well, and gave her back again. Then after only three days, man came 
back to him again and said: Lord, I know not how it is; but after all I 
have come to the conclusion that she is more of a trouble than a pleasure 
to me; so please take her back again. But Twashtrai said : Out with you, 
Be off. I will have no more of this. You must manage how you can. 
Then man said: But I cannot live with her. And Twashlrai replied: 
Neither could you live without her, and he turned his back on man, and 
went on with his work. Then man said: What is to be done? For I 
cannot live either with or without her. (Thomas, Source Book of Social 
Origins, p. 512.) 

Such )X)]iular beliefs have been in part justified by the pro])ability 
that many ()]>vi()us difTerences are clue to the work, and the result- 
ing variation in experience and environment, of women as con- 
trasted with those of men. Thus men know more about business, 
])olitics, current events and machines because their occupations 
bring them much more in contact with these things; but it does not 
follow that women could not, or would not, know us much about 
them if their occujiations were as much concerned with them. 
Women know more about cooking, social events, and household 
utensils because their occupations bring them much more in con- 
tact with thiin; but it does not follow that men could not, or would 
not, aCfjuire as much knowledge or skill in these directions if their 
occupations required it. 

The difTerences between the sexes are probal)Iy quantitative 
rather than (|ualilative. Both men and women have the same re- 
flexes, instincts, and capacities with the exce])lion of certain as- 
pects of the sex instinct. Tliese are probably similar in the main 



SEX DIFFERENCES 65 

and differ chiefly in their manner of expression. The differences 
due to sex life and the rearing of children, with the consequent 
differences in occupations and experiences, will account for many 
of the superficially observable differences between men and women. 

What are the differences that have been scientifically measured 
and compared? In order to produce a complete picture of mental 
differences between men and women it would be necessary to 
measure each trait in a very large number of persons and to com- 
pare the measurements with regard to both the averages of the 
abilities and the manner of the distribution of each ability. This 
has been done in part only with a few traits and only upon small 
groups of persons. 

Differences in Average Amounts of Mental Abilities. There 
are two methods by which abilities of two groups may be compared. 
Either we may state the actual average or median of each group, 
or we may state how many members of one group reach or exceed 
the average or median of the oth6r group. The latter method is 
preferable in many respects to the former in that it makes possible 
a comparison of groups of various sizes and indicates the relative 
differences more nearly true to fact. The two methods may be 
illustrated in the case of a memory test consisting of the oral pres- 
entation of ten words at the rate of one word per second and of 
asking the subjects to record immediately the number of words 
remembered. He ma), then state that the number of words 
remembered on the average by men was 6.9 and by women 7.2. 
Or we may state that 43.6% of men reached or exceeded the me- 
dian of the women. The latter method of comparison represents 
probably more true to life the amount and kind of difference or 
similarity that actually exist. The differences, hastily inferred 
from a comparison of averages only, would lead to the conclusion 
that in regard to memory women are distinctly superior to men. 
The implication would be that all women have a memory superior 
to that of men, whereas the fact is that the number of women 
having a memory superior to that of men is really small and that, 
in these few women, memory is better only by a very small shade. 
If 43% of men reach or exceed the median of women, it means 
that if the 7% of women having a slightly superior memory were 
omitted, the remaining 93% of the women would have a memory 
ability identical with that of the men. A difference of 7% in the 
distributions between two groups is represented by the curves in 



66 



EDUCATIONAL PSYCHOLOGY 



Figure 28. The difference is so small that the groups could hardly 
be distinguished. 

By the method of amounts of overlapping in the distrihntion 
of one grouj) over the other, the following results have been ol> 
tained from students in the University of Wisconsin in a series of 
tests on memory as just stated, on perception consisting in the 
cancellation within one minute of as many of a certain geometrical 
figure as possible, on motor ability consisting in tapping with a 
pencil upon a card as rapidly as possible for thirty seconds, and on 
mental addition as described elsewhere.' 




Fk;. 28. — Distribution cun'cs representing a difference of 7% between the 
medians of the two groups. 

TABLE I-' 
Percentage of men reaching or exceeding the median of women. 



Perception of geometrical forms . 

Memor\' of words 

Motor ability 

Mental addition 



193 men 


200 women 


54-5% 


55 men 


77 women 


43-6% 


25 men 


50 women 


72 0% 


21 men 


46 women 


66.7% 



In the interpretation of these percentages of overlapping it 
must be remembered that if 50% of one group reaches or exceeds 
the median of the other, it means of course that the two groups 
are identical in ability and distribution. If the percentage of men 
reaching or exceeding the median of the women is over 50% it 
means that the men are superior by the number exceeding 50%. 

Helen Thompson Woolley made a series of tests as indicated in 
the following table upon twenty-five men and women at the Uni- 
versity of Chicago, on the basis of which Thonulike has computed 
the followiivj ixrc cnfaL'es of men reailiiiii: or i \i itiiiiiir the median 
for women: 

' Ejcp<rinu:nU in Uiudiumal J'iyckjliisy, rcvi.-A:J edition, chapter 16. 



SEX DIFFERENCES 



67 



TABLE 13 

Percentages of men reaching or exceeding the median of the women. After 
Woolley as computed by Thorndike ('14, III, p. 178). 

Reaction time 68% 

Tapping 81% 

Sorting cards, speed 14% 

Sorting cards, accuracy 44% 

Thrusting at target 60% 

•Drawing Hnes 72% 

Threshold of pain 46% 

Threshold of taste 34% (22) 

Threshold of smell 43% 

Lifting weights 66% 

Two-point discrimination 18% (43) 

Memory (syllables and learning) 32% (46) 

Ingenuity 63% 

In a similar comparison made on the basis of 100 boys and 100 
girls from results obtained by Gilbert, the percentage of boys 
reaching or exceeding the median of girls was as follows: 

TABLE 14 

Percentages of boys reaching or exceeding the median of the girls. After 
Gilbert ('94) as computed by Thorndike ('14, III, p. 182). 

to 14 years 15 to 17 years 

Discrimination of weights 48% 58% 

"colors 39% 58% 

Reaction time 57% 76% 

Resistance to size-weight illusion 55% 68% 

Rate of tapping 64% 73% 

Thorndike ('14, III, p. 183) reports measurements in which the 
comparison of the percentages of boys reaching or exceeding the 
median of girls for persons 8 to 14 years old, were as follows: 

TABLE IS 

Associative tests, opposites, addition, multiplication, etc 48% 

Perception, A-test, etc ^^% 

Memory of words 40% 

The writer has made comparisons in the case of school subjects 
on the basis of abilities measured by means of tests and scales. 
Speed of writing was measured in terms of letters written per 
minute. Quality was rated by the Thorndike scale. Attainments 



68 EDUCATIONAL PSYCHOLOOV 

in the remaiiniij^ subjects were measured by the author's tests in 
these fields. The following percentages of boys, reaching or ex- 
ceeding the median of the girls, were obtained: 

TABLt: 10 

Speed of handwriting, about i loo boys and i loo girls 47% 

(^)uality of handwriting, " 1100 " " 1100 " 39% 

Arithmetical reasoning, " 1250 " " 1250 " 60% 

History, " 429 " " 526 " 72% 

( icograpiiy, ' " 447 " " 472 " 48% 

Figures of a similar sort computed by Thomdike ('14, III, p. 
183) on the basis of teachers' marks showed the following percent- 
ages of boys reaching or exceeding the median of girls: 

TABLE 17 

High school pupils 

English 41% 

Mathematics 57% 

Liitin 57% 

History 60% 

College students 

English 35% 

Mathematics 45% 

History and economics 56% 

Natural sciences 50% 

Modern languages 40% 

The difficulty with many of the measurements is that they are 
based on too small a number of j^ersons. Comparisons based on 
twenty-five ])ersons from either sex may be indicative but not 
final. Sunmiarizing, we may say that women and girls are supe- 
rior in sensibility, in memory, in most forms of perception, in 
C|uality of handwriting, and linguistic fluency. It is interesting 
to note in this connection that in the survey of mental-test 
results • the women excel in tweh'c out of fourteen tests which de- 
pend chiefly upon linguistic fluency. Thus the females excel in 
speed of reading, both oral and silent, in amount of information 
given in describing an object or in making a rept)rt, in the genus- 
species test, in the number of words thought of and written per 
minute, in the ])art-whole test, in the opposites test, in memory 
span for words, in memory for K)gical-verbal material, in the word- 

' (iivin in the various chapters of Wliipplc's Manual of Mental and Physical Trsls. 



SEX DIFFERENCES 69 

building test and in the Ebbinghaus completion test; while the 
males excel in the rate of association and in the sentence building 
test. Apparently the popular belief in the greater linguistic fluency 
of women is not without foundation. Men and boys are superior 
in motor capacities, such as tapping, quickness of reaction, in 
arithmetical reasoning, and in resistance to suggestions as indi- 
cated by the size-weight illusion and the use of suggestive ques- 
tions in testimony. The two sexes seem to be approximately 
equal in associative processes and in most school subjects. The 
amounts of difference, however, are very small. This is particu- 
larly true of all the traits that have been measured in a sufficiently 



1.20 
1.00 

.80 

.60 

.40 - 

.20 



.00 



Girls 



9 10 
Years 



12 



13 



14 



Fig. 29. — Comparison of general intelligence of boys and girls as measured 
by the Stanford revision of the Binet-Simon tests. After Terman ('16, p. 72). 
The numbers along the vertical axis are intelligence quotients as explained in 
Chapter VII. 

large number of persons to make the comparisons safe. Any 
differences lying between 40% and 60% of the number of either 
sex reaching or exceeding the median of the other are practically 
negligible. If 60% of one sex reach or exceed the median of the 
other, it means that 10 persons in a hundred of the one sex, are by 
a small amount superior to the other. Differences larger than this 
have been established with a fair degree of certainty practically 
only in the case of one large field of capacities, namely, that of 
motor abilities. Differences in nearly all other respects in which 
comparisons have been made on large numbers of persons are 
almost entirely within the limits of 40% to 60%. Terman found 
in measuring the general intelligence of nearly 1,000 boys and girls 



70 



EDUCATIOXAL PSYCHOLOGY 



by means of his revision of the Binet-Simon tests that for the ages 
of five to fourteen j,'irls tend to be very slightly sui)crior to boys 
and that after fourticn they are practically equal. His results are 
set forth in Figure 2S. 

It seems a likely interpretation that motor superiority has been 
carried over to include intellectual sui>eriority as well. For centu- 
ries women have been considered intellectually inferior to men. 
They were thought to be incapable of acquiring anything more 
than an elementary education. It has been only since the middle 
of the 19th centur>' that co-education and women's colleges have 
been generally established. Intellectual inferiority has probably 




Fig. 30. Range of ability of men a.nd women in color discrimination. After 
Henmon ('10). 

been inferred chiefly from niotor and mu.scular inferiorit}- and from 
the conditions of a narrower environment and dependency due 
to the bearing and rearing of children. The inference and belief 
of intellectual inferiority is apparently unfounded. This conclu- 
sion may be fairly dra\\ii both from the sjjeoific psychological 
tests that have been cited and also from the recent successes of 
women in the acquisition of higher education. 

Diflference in the Range of Variations in Abilities. Besides 
C()m|);iring the average amounts of any gi\en ability in the two 
se.xes, we may compare also the range of abilities from the lowest 
to the highest in the two se.xes. Such comj)arisons have been made 
in a few traits and the general inference has been that the range of 
abilities is wider among men than among women. The distribution 
of the abilities in the geometrical perception test made upon 193 



SEX DIFFERENCES 7 1 

men and 200 women mentioned in a preceding paragraph, was 
as follows: 

Scores: , 2-3 4-5 6-7 8-9 lo-ii 12-13 14-1S 

193 Men 4.5% 15.4% 33-9% 21.9% 15.4% 5.6% 2.8% 

200 Women 3.2% 20.8% 38.9% 21.6% 10.7% 3.9% 0.8% 

Thus in the extremely high ability of canceling 14 to 15 geo- 
metrical figm"es in one minute, there were 2% more men than 
women, and in the lowest ability of canceling only two to three 
geometrical figures, there were 1.3% more men than women. 
Comparisons of this sort can be made safely only on large numbers 
of individuals, and consequently there is as yet little reliable 
material available. 

The ratio of female to male variability has been computed by 
Thorndike ('14, III, p. 194) on the basis of tests of memory, re- 
action-time, discrimination of length, opposites, and cancellation 
made by Gilbert ('94) upon 100 boys and 100 girls of each age 
from 6 to 17. The average ratio in all tests for the ages of 9 to 12 
was found to be .92, for the ages 13 to 14 1.025, and for the age 
of 15, .97. Girls were slightly less variable at all ages except 13 
and 14. In a test of color discrimination Henmon ('10) also found 
a slightly larger variability among men than among women as 
shown in Figure 30. 

The author made a comparison of the range of abilities in history 
and geography as measured by his tests in these subjects, and 
found the following distributions: 

History, 8th Grade 
Percentages of boys and girls attaining the various scores 

Scores: o-io 11-20 21-30 31-40 41-50 51-60 61-70 

Boys 4-2% 9-3% i5-3% 170% 13.2% 12.2% 11.5% 

Girls 6.2% 22.7% 22.7% 16.4% 12.8% 9.4% 6.0% 

Scores (continued) : 71-80 81-90 91-100 loo-iio Total 

Boys (continued) 9.3% 6.6% 2.1% 0.4% 288 

Girls (continued) 2.3% 2.6% 0.9% 0.0% 352 

Geography, 7 th Grade 

Scores: o-io 11-20 21-30 31-40 41-5° 51-60 61-70 71-80 

Boys 06% 2.8% 4-7% 8.4% 5-3% 12.2% 14.7% 14-7% 

Girls 03% 16% 56% 10.0% 10.0% 12.5% 18.7% 18.7% 

Scores (continued) : 81-90 gi-ioo loi-iio 111-120 1 21-130 Total 

Boys (continued) 10.3% 8.1% 9-i% 6.9% 2.8% 320 

Girls (continued) 10.0% 8.7% 7-2% 5-9% 3-i% 322 



72 EDUCATIONAL PSYCHOLOGY 

The varial)ility of boys in the case of history is somewhat larger 
than that of the girls, ^\hereas in the case of geography it is sub- 
stantially the same. 

ThomJike ('14, III, p. 195) has given the range of ages of boys 
and girls in the third year of high schools in Chicago, Philackli)hia, 
New \'ork, Detroit, Fall River, Los Angeles, Lowell, and Worcester 
as follows: 

Afjc 13 14 15 16 17 i8 19 20 and over Total 

Boys 7 02 504 1246 1203 57-' 1 93 67 3974 

Girls 4 73 562 1351 1289 554 120 34 3987 

There are about twice as many boys as girls at either 13 or 20 
or over. 

In support of the general iH-Iief that the range of general abilities 
is wider in men than in women may also be cited the fact that in 
the history of the world most of the great geniuses have been men, 
and also the statistical fact that male idiots and criminals at the 
other extreme of the distribution curve consideraljly outnum- 
ber the female. The fact that the great geniuses of the world 
have been men rather than women would accordingly be ex-jjlained, 
not on the basis of lack of opportunity, but mainly on the basis 
of greater exceptional ability. The theory seems plausible but 
has been ]iroposed rather in ad\'ance of a con\'incingly wide range 
of e.\])erimental data. If it is true, it would mean that according 
to the perception test the one or two per cent most gifted individu- 
als are men and the i or 2% least gifted individuals arc also men, 
that of the next 10 or 12% of most gifted indi\'iduals aj^proximately 
two-thirds would be men and one-third women, and likewise of 
the next 10 or 12'^ i least gifted individuals at the other extreme, 
about two-thirds would be men and one-third women. For the 
remainder of the distribution the number would be practically 
identical. The facts should not be interi)reted as im])lying that 
men as a rule are superior to women, but would mean simply that 
only the one or two exceptional jiersons in a hundred would be 
superior to the most gifted women. The remaining 96 or 98% 
would be largely identical. 



CHAPTER VI 
THE INHERITANCE OF MENTAL TRAITS 

Problem. In a certain obvious sense, the entire native equip- 
ment of any human being is inherited. The various capacities and 
the relative amounts of them with which a person starts in Hfe 
are derived from the cells from which the individual originates. 
The dififerences among these original cells, even when derived 
from the same parent, are assumed to vary with regard to any 
potentiality according to the normal distribution curve about 
the central tendency of that particular parent. Stalks of corn 
growm from seed taken from the same ear will vary considerably 
from one another because the seeds themselves, even from the 
same ear, are different, but yet they will vary around the general 
type of the parent stalk. It is therefore obvious that children of 
the same parents will not be absolutely alike but that they will 
vary about the central tendency of their ancestors. The specific 
problem is not: Are mental traits inherited? but rather: How much 
do children of the same parents or ancestors resemble one another 
in the amounts of different traits possessed, and in the manner 
in which the various traits combine? To what extent are abilities 
in school work inherited? To what extent are the wide ranges of 
abilities, noticed in Chapter III, due to native equipment or to 
opportunity and environment? To what extent does a person 
make of himself what he does by virtue of his opportunities or by 
virtue of his inherent make-up? What part of the future adult 
individual is really determined by the school as an agency of 
his environment and what part is beyond the control of the 
school? 

Methods of Studying Heredity. Any individual is the resultant 
of the interplay between his inherited equipment and the stimuli 
from his environment. Hence, theoretically, there are two general 
methods of studying the problem: First, by keeping the environ- 
ment constant and varying the ancestry, so to speak; or second, 
by keeping the ancestry constant and varying the environment. 
That is, according to the former plan we would place children of 
entirely different ancestry into the same environment from birth 

73 



74 KDUCATIONAL PSYCHOLOGY 

up to a given puint in life, and then measure the amount of simi- 
larity or dilTerence; or according to tlie latter plan, we would place 
children of the same ancestry into entire!}' dilTerent environments 
from birth to a given point in life, and then measure the amount 
of similarity or dilTerence. Such ideally scientific conditions are 
I)ractically impossible to obtain. The best we can do is to measure 
the resemblances or difTerences of children of the same ancestry 
and compare them with the resemblances or difTerences of children 
of dilTerent ancestry, both groups living in appro.vimately the 
same enxironnicnt. 

General Views Concerning Mental Heredity. Two extreme 
views concerning heredity are possible according to our conception 
of the relative roles ])layed by heredity and environment in the 
production of adult individuals. We may assume on the one hand 
that v.hat a person l:)ecomes is absolutely and entirely determined 
by heredity, and that environment makes no dilTerence whatever; 
or we may assume on the other hand that what a person becomes is 
Completely and entirely determined by his en\ironmcnt, and that 
heredity jjlays no ])art. Neither \iew has been held by any serious 
student of heredity in recent times. Views very closely ajjproach- 
ing these e.xtremes, have, howe\'cr, been held by prominent writers 
and thinkers in times jxist; whereas various views between these 
extremes arc generally being held at the jiresent time, de])ending 
ui)on the conception as to whether the larger, smaller, or ecjual 
share is contributed by heredity or by environment. The ^•iew 
held by most scientific students of the i)roblem to-day gives 
weight to both elements with perhaps the major emphasis upon 
heredity. 

The Similarity of Abilities among Related Eminent Persons. 
This particular method of attacking the probleni was historically 
the first means of approaching the study of the inheritance of 
mental traits. Two extensi\e investigations on this aspect of the 
subject ha VI' been made. The first was carried out by Sir Francis 
Galton and i)ul)iished in iSoc;. Galton made a study of 977 emi- 
nent men, each of whom was the most eminent among 4,000 ]>ersons. 
He i)roceeded to determine how many relati\'es of ec|ual eminence 
and of var\ ing degrees of relationship each ]urson possessi'd. 
In this manner he found that the.se 077 men had the following 
relatives of a like degree of en\inence: 89 fathers, 114 brothers, 129 
sons, 52 grandfathirs, 37 grandsons, 5.^ uncles, and 61 nephews, or 
a total of 5;;5. (lailon further jjointed out that 977 ordinary men 



THE INHERITANCE OF MENTAL TRAITS 75 

selected by chance from the population at large would have only 
four such eminent relatives. He concluded as follows: 

"i. That men who are gifted with high abilities — even men of class 
I E — easily rise through all the obstacles caused by inferiority of social 
rank. 

" 2. Countries where there are fewer hindrances than in England, to a 
poor man rising in life, produce a much larger proportion of persons of 
culture, but not of what I call eminent men. (England and America are 
taken as illustration.) 

"3. Men who are largely aided by social advantages are unable to 
achieve eminence, unless they are endowed with high natural gifts." 

More recently an extensive study was made by Woods ('06) 
on mental and moral heredity in royalty. Woods made a com- 
parison of 671 members of royal families in Europe by giving each 
person a rating on a scale of i to 10 in which 10 signified excep- 
tionally high ability or genius, and i represented exceedingly low 
ability or imbecility. These ratings were made by the judgment 
of Woods himself according to the reports of these persons in 
histories and biographies. On the basis of these estimates, a 
tabulation was then made of the relationship of persons of various 
degrees of ability. He found that most of the eminent persons 
were grouped about four stocks or families out of fifteen, namely, 
the families of Frederick the Great, Queen Isabella of Spain, 
William the Silent, and Gustavus Adolphus. Likewise, he found 
that most of the persons of lowest ability were grouped around 
certain families in Spain and Russia, and the persons of mediocre 
ratings, four to seven, centered about some half dozen royal families 
including the houses of Hanover, Saxe-Coburg-Gotha, Reuss, 
Mecklenburg, Hapsburg in Austria, Holstein, Denmark, Saxony, 
Savoy, Orleans and modern Portugal. The ratings, of course, 
were not absolutely correct measurements of their abilities, but 
they, no doubt, represented greater validity than general impres- 
sions would. He further computed coefl&cients of resemblance in 
intellect and morals as follows: 

I. In intellect : 

Offspring and father 30 

" " grandfather .16 

" " great-grandfather 15 

II. In morals: 

Offspring and father 30 

" " grandfather 175 



76 KDUCATIONAL I'SYCHOLOCiV 

I >r. Woods then altcniplcd to determine whether (^r not thtfact of 
accession to the throne by virtue of birth gave an individual greater 
opix)rtunity for eminence. This he states in the following nianner: 

"There is one peculiar way in which a lilllc more than half of all 
males have had a consi(lcral)le advantage over the others in gaining 
distinction as inifxjrtanl historical characters. The eldest sons, or if not 
the eldest, those sons to whom the succession has devolved, have un- 
doubtedly had greater op[M)rlunilies to become illustrious than those to 
whom the succession did not fall by right to jjrimogeniture. I think 
every one must feel that perhaps much of the greatness of Irederick II 
of Prussia, Clustavus Adolphus, and William the Silent, was due to their 
oflicial positions; but an actual mathematical count is entirely opposed to 
this view. The inheritors of the succession are no more plentiful in the 
higher grades than in the lower. The figures show the number in each 
grade who came into power by inheriting the throne." 

Grades i 23456789 lo 

Total No. in each j; rude 7 Ji 41 49 71 70 68 43 18 7 

Succession inheritors 5 14 26 31 49 38 45 23 12 4 

Percent 71 07 O3 64 69 54 07 54 67 57 

"It is thus seen that from 54 to 71*^ inherited the succession in the 
different grades. The upper grades are in no way composed of men 
whose opportunities were enhanced by virtue of this high position. Thus 
we see that a certain very decided difference in outward circumstances — 
namely, the right of succession — can be proved to have no etTect on 
intellectual distinction, or at least so small as to be unmeasurable without 
much greater data. The younger sons have made neither a poorer nor a 
better showing. ('06, pp. 285-2S6.)" 

"The ujjshot of it all is, that as regards intellectual life, environment is 
a totally inadequate explanation. If it explains certain characters in 
certain instances, it always fails to explain as many more; while heredity 
not only e.xplains all (or at least Qo^t ) of the intellectual side of character 
in practically every instance, but does so best when questions of en- 
vironment arc left out of the discussion. Therefore, it would seem that 
we arc forced lo the conclusion that all these rough differences in in- 
tellectual activity which are susceptible of grading on a scale of ten are 
due to predetermined differences in the primary germ-cells." ('06, p. 286.) 

WTiile heredity no doubt plays an important part in the prmluc- 
tion of intellect and character the |)art attributed to it by VV'chkIs 
that it explains " at least QO% of the intellectual side of character 
in every case " is hardly warranted either by the findings of other 
investigators or by the results of Woods himself. His corrrelation 
between father and offspring is only .30. 



THE INHERITANCE OF MENTAL TRAITS 77 

Similarities of Abilities Among Related Defective and Low 
Grade Persons. Quite a number of studies have been made in 
recent years concerning the frequency with which defective persons 
are either distantly or closely related. One of the first studies 
w^as that of the Jukes reported by R. G. Dugdale in 1877. Max 
Juke, born in 1720, was a shiftless truant, who married an equally 
worthless woman. Up to 1S77 there had been five generations with 
approximately 1,200 descendants among whom have been traced 
the following types of persons: 310 paupers, 7 murderers, 60 habit- 
ual thieves, 50 prostitutes, 130 convicted of crime, 300 died in 
infancy, 440 physical wrecks from debauchery, only 20 learned a 
trade, and 10 of these learned it in prison. The estimated cost to 
the State of New York has been put at approximately $1,000 a 
person. In contrast with this lineage, a comparison has been 
suggested with the Jonathan Edwards family, which had approxi- 
mately 1,400 descendants in the same period of time. Among 
them there have been 120 graduates of Yale alone, 14 college 
presidents, over 100 professors, 135 books of merit have been 
written by various members of the family, and 118 journals have 
been edited by them. Aaron Burr was the only black sheep among 
them and he can certainly not be classed as an intellectually de- 
fective person. (Winship '00). 

Poellman of Bonn (Guyer '16^ p. 271) made a study of a family 
called the Zeros in which 800 descendants were traced through six 
generations back to a female drunkard. Among them were found 
102 professional beggars, 107 illegitimate offspring, 181 prostitutes, 
54 inmates of almshouses, 76 convicted of crime, and 7 murderers. 
The cost to the state was placed at $1,206,000. 

More recently a very interesting study was conducted by Dr. 
Goddard of the Training School at Vineland, New Jersey. Dr. 
Goddard ('12) traced the ancestry of a young girl who had been 
brought to his institution. It was found that the lineage went 
back to a man, Martin Kallikak, a soldier in the Revolutionary 
War, who was the progenitor of two lines of descendants. (See 
Figure 31.) He had an illegitimate son whose mother was feeble- 
minded. This was the establishment of line — A — which had, down 
to the time of the study, 480 direct descendants among whom 
were found the following: 143 feeble-minded, 292 unknown, 36 
illegitimates, 33 prostitutes, 24 alcoholics, 3 epileptics, 82 died in 
infancy, 3 criminals, 8 keepers of disreputable houses, and only 46 
normal individuals. Apparently human nature does not gather 



78 



EDUCATIONAL PSVCIK )L( K i V 



grapes of thorns or figs of thistles. After his return from the war, 
Martin married a woman of normal intelligence and from this 
lineage — B — there had come during the same period of time, 4g6 
direct descendants, of whom all were normal individuals with the 



THE 
LAWFUL WIFE 



©- 



MAATII KALLIXAK SR. 

I THE MMELESS 

FEEBLEUmOEO 6IRL 

liil 



[§ (N) (N) (N) (N) [N] ® 



-® 



RHODA 2ABETH 



(Sw-lli-r-O ^ ^^^' ' '4 (S) (N) d) 



$S<h^^6666 



TAKEN IH D. D, 

GOOD 9 VRS. 4 mt. 
HOME 

D.K TRS. 




Fir,. 31. — Dcsrcndants of tin: Kallikak ramily. Squares = males, drrles = 
females, hiac k sfjiiares or cinlcs = feeMeminded, open scjuarcs or cinles = 
normal persons. Ihe lineage was traced back from Deborah. After Goddanl. 

exception of five, one of whom was reported as mentally defective, 
two as alcoholics, one as sexually immoral, and one as a case of 
religious mania. There were no epili-ptics or criminals, and only 
15 died in infancy. The remainder were good citizens, including 
doctors, lawyers, educators, judges, and business men. 

One thing seems to stand out very conspicuously from the 



THE INHERITANCE OF MENTAL TRAITS 79 

numerous facts of family histories that have been unravelled in 
recent years, namely, that much defective mentality, degeneracy, 
and crime is a m.atter of ancestry. General opinion among persons 
in charge of institutions for defectives is that two-thirds of all 
cases are due to heredity and one-third to environmental or un- 
known causes. Thus Dr. Alfred Wilmarth, Superintendent of the 
Wisconsin Home for Feeble-minded, says: 

"My own observations, and those of others in this country and Europe, 
would indicate that at least two-thirds of the feeble-minded have defec- 
tive relatives. This is significant. Mental accident may occur in any 
family, but it is rarely a second case occurs unless there is a tendency to 
nerve degeneracy. (Quoted by Guyer,'i6, p. 245.) 

"I present to you the results of compiling the histories on 1,000 appli- 
cations, where our information is most thorough; but I am confident 
that these do not tell the whole story. In 3 1 1 of these any neurotic taint 
in the family history is absolutely denied. In 365 cases at least one near 
relative suffers from one of the graver forms of nervous or mental trouble; 
in 170, two relatives were found; in 73 cases, three relatives, and in 81 
cases four or more. These figures agree very accurately with the results 
of other observers in this country and abroad. It is safe to say that less 
than one-third of the defective classes are the results of disease or trau- 
matism in families capable of transmitting a healthy, well developed 
nervous system." 

Dr. Goddard of the Training School, Vineland, New Jersey, 
states in connection with his tests of 2,000 children: 

"But we now know that 65% of these children have inherited the 
condition, and that if they grow up and marry they will transmit the 
same condition to their oft'spring. Indeed, we know that this class of 
people is increasing at an enormous rate in every community and unless 
we do something to stop this great stream of bad protoplasm we shall 
some day be swamped in a sea of degeneracy." 

Likewise Dr. A. C. Rogers of the Minnesota School for Feeble- 
minded, at Faribault, says: 

"We have no survey of mentality in this country except in very small 
areas, but probably about 65% of the feeble-minded children that we 
know of are feeble-minded from heredity; that is, they come from families 
in which there is much feeble-mindedness, usually associated with various 
neuroses or psychoses. There are about 35% approximately that are 
acquired cases. These cases develop from various things. Full develop- 
ment may be prevented during gestation, or early childhood, or early 
adolescence, but these acquired cases are entirely distinct from the 
hereditary ones." (Guyer '16, p. 246.) 



8o EDUCATIONAL PSYCHOIX)GY 

Likewise, Dr. Martin W. Barr of the Pennsylvania Training 
School for Feeble-minded Children states: 

"In my individual study of 4,050 cuscs of imbecility, I tind 2,631 or 
65.34%, caused l)y malign heredities; and of these 1,030, or 25.43%, are 
due to direct inheritance of idiocy; and 280, or 6.91%, to insanity." 

To one who wishes to argue in favor of environment as the chief 
determining element in ability and character, such data as have 
been presented from family histories and relationships are not en- 
tirely convincing. It might be argued that a given family has 
so many individuals of high or low intelligence and achievement 
because its members were born in circumstances which did or did 
not afford opportunities for development and training and for 
achieving higher success. It might be said that the descendants of 
the Edwards family were born and reared among favorable cir- 
cumstances of educational and financial advantages and conse- 
quently were fitted for greater tasks and lived in an environment 
in which larger opportunities offered themselves, whereas the 
members of such a lineage as the Jukes family would have just the 
opposite environment of birth, education, and opportunity in life. 
In answer to all this, we must remember, however, that abihty 
very largely determines the sort of environment in which a person 
is satisfied to live, that a really capable person is quite likely to 
push forward and to find a way out of the en\ironment in which 
he may hapi)en to have been born, or to improve it if he cannot 
leave it, and finally, we must remember that the persons of low 
ability were born in circumstances of a correspondingly low nature 
because of the hereditary stock of the families from which they 
came. Their jxircnts were content to live under the circumstances 
under which they did live because their abilities and desires sought 
for nothing better. 

Similarities between Brothers and Sisters in Special Mental 
Traits. Pearson has shown that the resemhlaiue in physical 
characteristics among brothers and sisters is approximately .50. 
He gives the following coefficients of correlaiitui for various 
physical traits. 

Brother aiui sister 

Hair color 55 

Cephalic index 4Q 

IleiRht SO 

Eye color . . .52 



THE INHERITANCE OF MENTAL TRAITS 



8l 



What, however, is the degree of resemblance in mental traits? 
The general argument is that mental traits are dependent upon 
anatomical and neurological structures, and hence, if these are 
inherited, mental traits must also be inherited. What is the evi- 
dence from experimental and statistical facts? 

In a study made by the writer ('17) a series of tests of capacities 
directly affected by school work and another series of tests of capac- 
ities not directly affected by school work were applied to 18 pairs 
of brothers and sisters in the University of Wisconsin. Each test 
was given twice on two different occasions in order to obtain a 
fairly accurate measurement of the capacities concerned. The pur- 
pose of giving the two types of tests was to ascertain whether 
brothers and sisters were more alike in the traits affected by train- 
ing than in the traits not directly affected by school training. The 
following were the tests and the correlations obtained between 
pairs of children of the same family : 



TABLE 18 
Correlations between abilities of brothers and sisters. After Starch ('17) 



Reading — speed 

Reading — comprehension . . 

Writing — speed 

Writing — quaHty 

Size of reading vocabulary . 

Spelling 

Arithmetical reasoning. . . . 

Addition attempts 

Addition — rights 

Subtraction — attempts. . . . 

Subtraction — rights 

Multiplication — attempts . 
Multiplication — rights. . . . 

Division — attempts 

Division — rights 



Average . 



Memory 

A-test 

Geometrical form test . 
Tapping 



Average 

Coefl&cients based on ranks in all tests combined 



82 EDUCATIONAL PSYCHOLOGY 

In order to grasp the full import of these figures, it is necessary 
to remember that the coefTicient of correlation between mental 
abilities of pairs of unrelated individuals selected by chance would 
be zero, and that any coefficient above zero between pairs of 
brolJiers and sisters means a corresponding amount of resemblance. 

"From the above (able several interesting results appear, (i) The 
resemblance of siblings is apparently no greater in those mental traits 
which are directly affected by school work than in those which arc not so 
affected. The average correlation in the former group of tests is .42 and 
in the latter .38. This seems to indicate that the mental similarities of 
children of the s;ime parents arc due primarily to heredity rather than to 
similarity of environment since the resemblance is no greater in those 
traits which are more directly affected by environment. 

"(2) The resenil)lance of siblings is approximately as great in menial 
traits as in physical traits. Pearson found the correlation between 
brother and t)rother in height to be .50 and in cephalic index (ratio of 
length to width of head) .40. These correlations for physical traits are a. 
Uttle larger than the ones found here for mental traits taken separately. 
The correlation, however, calculated on the basis of a combined rank for 
each person in all mental tests together was found to be .73. This 
greater correlation for all tests combined as compared with the correlation 
for single trails is due partly to the variation of the correlations among 
the single traits and partly due to the imperfections in the separate tests, 
which arc counterbalanced to some extent in a combined ranking." 
(Starch '17, p. 237.) 

Pearson ('04) made a comparison of 2,000 brothers and sisters 
who were rated by their teachers in such traits as vivacity, self- 
assertion, introspection, popularity, conscientiousness, temper, 
ability, and handwriting. On the basis of these ratings he found 
coefficients of correlation ranging from .4 :; to ,64 with an average 
of .52. These results are interesting and indicative of the resem- 
blance of more general traits of character, but they are probably 
rendered more or less uncertain l)y the unreliability of one person's 
ratings of such elements of character. The likelihood is that the 
teachers would be more inclined to estimate alike the children from 
the same families, rather than to estimate them more different 
than they really were. 

Similarities of Brothers and Sisters in Abilities in School Sub- 
jects. In a study made s<.\eral years ago, Earle (03) found a 
correlation of .50 between the spelling abilities of 180 pairs of 
brothers and sisters. The writer ('15) made a study of the scho- 



THE INHERITANCE OF MENTAL TRAITS 83 

lastic records of children from 63 families. The average grade in all 
school subjects was obtained for each pupil and used as the measure 
of his academic ability. The correlations based upon these averages 
were as follows: 

First and second child in a family, 63 pairs 58 

Second and third child in a family, 24 pairs 64 

First and third child in a family, 24 pairs 34 

Average 52 

Further comparisons were made for abilities in specific school 
subjects which yielded the following correlations: 

Spelling, 57 pairs of children from the same parents 21 

Reading, 57 pairs of children from the same parents 49 

Writing (speed) 24 pairs of children from the same parents 18 

Writing (quality) 24 pairs of children from the same parents 06 

Another study was made of 38 children from 11 families. All the 
marks that each pupil had received in each study in grades three to 
eight were averaged. From these averages the following coefficients 
of correlation were obtained. 

Arithmetic, 54 pairs 32 

Spelling, 54 pairs 21 

Reading, 54 pairs 31 

Language, 54 pairs 24 

"No importance, I believe, can be attached to the differences in cor- 
relation between the various studies. The correlation for individual 
studies is lower than that for scholarship in general based on the average 
performance in all studies combined. This is probably due chiefly to the 
fact that the inaccuracies of teachers' marks in individual subjects are 
partly ehminated in the averages derived from all studies. 

"Abilities in special subjects are inherited, apparently, to no greater 
extent in one subject than in another. What is probably inherited is 
either general scholarship or else more specialized traits than ability in 
arithmetic, or ability in language. Each study involves many mental 
faculties and nearly all studies involve the same faculties with varying 
emphasis. 

"In corroboration of this point we may notice the following table of 
average marks for each of nine families in each study. 

"In this table, we must examine the ranks, rather than the marks, of 
the different families in each subject, so as to eliminate the variation in 
-Standards of marking. These famiUes rank very nearly the same in the 



84 



EDUCATIONAL PSNTTTOI^OCV 



various studies. For example, family C is first in every study except 
arithmetic and there it is third. Family (1 is second in every subject 
except arithmetic and there it is fourth. Family I is either third or fourth 
in every subject but one, and family H is last in every subject except one. 

TABLE u) 
.Averages for each f;unily in each subject. 



FVMII.Y 


No. OP 




Children 


A 


3 


B 


5 


C 


4 


D 


6 

4 


E 


1* 


G 


3 
3 


H 


I 


3 



Akitp- 

MtTlC 



Grade 



Rank 



72.3 
75-1 
80.2 
80.4 

77-4 
73.6 8 
77.6 4 
76.1 4 
Si. 8 I 



Spellinc. 



Grade 



Rank 



85.1 
76. I 
89.1 
85 9 
78.5 
80.8 
86.6 2 
78.4 8 
8SS 4 



Reading 



Grade 



R.ink 



73-2 
86. 7 
81. 1 

74-4 

76.1 

84.4 
So. 5 
82 I 



Lanou.m-.e 


Grade 






R.ank 


80 


7 


5 


74 


6 


9 


83 


6 


I 


81 


6 


4 


76 


I 


7 


76 





8 


S3 


2 


2 


So 





6 


82 


5 


3 



Geog- 

RAPBY 



Grade 



Rank 



76.7 8 

75-3 9 
84.6 I 
78.6 6 

79 3 4 
78.6 6 

83.8 2 
78.6 6 
81.6 3 



History 



Grade 



Rank 



77.9 8 

76.1 9 
84.0 I 

78.5 6 
78 7 5 

78.2 7 

83.8 2 

79.6 4 

81.9 3 



"There i.s no evidence, at least from these figures, for the notion that 
special abilities in certain studies run in families. ^Mental trails running 
in families are very likely more specialized than abilities in school studies 
which involve large groups of mental functions. The children of any 
given family are on the average equally good or equally poor in all studies. 
Ability in school work is apparently inherited to the s;ime extent as 
jihysical features since the coeflicients of correlation for children of the 
same parents are ai)proximalely the s;ime for both physical and mental 
traits." (Starch, '15, pp. 609-610.) 

Schuster and IMdcrtoii (07) calculated the resemblance in 
scholarship between brother and brother and between father and 
son among the Oxford honor men and found a coclTicient of correla- 
tion of .40 for the former and .^^i ior the latter. Miss Eldcrton 
further determined the correlations between cousins from records of 
about 300 families and found a coelTicient of .27. 

Miss Emily S. Dexter ' made a study of the scholarship records (^f 
185 pairs of brotJiers, sisters, brothers anrl sisters, graduates of the 
University of Wi.sconsin, and of 69 similar pairs who were gradu- 
ates of the high school at .Ashland, Wisconsin. She reports the 
following coelTicienls: 

' The study was r.nrrird ntif under the <lircction of Profes.s«)r Henirmn and reported 
ill a thc»ii in the liljrury uf the UiiivcrMty of Wiscuiuuii, 1915. 



THE INHERITANCE OF MENTAL TRAITS 



85 



TABLE 20 





Number 
OF Pairs 


General 
Scholar- 
ship IN 
all Sub- 
jects 


English 


Lan- 
guage 


Mathe- 
matics 


History 


Science 


University: 

All pairs 

Bro. and bro . . . 

Sis. and sis. . . . 

Bro. and sis. . . . 
High School: 

All pairs 

Bro. and bro . . . 

Sis. and sis ... . 

Bro. and sis ... . 


185 
44 
71 
66 

69 
10 
26 
23 


.69 

■47 
53 
.62 

.64 
■38 
39 
36 


.64 
.58 


.63 


■55 


.62 

■63 


.60 
.61 



Miss Dexter concludes "that inheritance, to a much greater extent 
than training is responsible for the degree of resemblance found." 

"If it were largely training, we would expect to find the resemblance 
greater between brother and brother, and sister and sister, than between 
brother and sister, but such is not the case. In the high school the cor- 
relations for the three groups are much the same, but, as has been pointed 
out, that may be due to a great extent to chance, for the groups are small. 
However, in the case of the university, where the groups average nearly 
three times as large as in the other school, we find the resemblance be- 
tween brother and sister to be greater than between brother and brother, 
or sister and sister. 

"Again, there is the question as to specialized abilities, and also that of 
general mental ability rather than specialized abilities as a basis of ex- 
planation for the close resemblance found. Thorndike, as has been said, 
finds that heredity is highly specialized. This study, however, seems to 
show a stronger tendency toward general mental ability, if by that we 
mean approximately equal ability in all subjects. It seems, also, to give 
almost no evidence of alternate inheritance; that is, of one individual's 
inheriting ability in one line while his brother inherits ability in another. 
In other words, a student who is above the average, either of his family 
or of the school, in one subject, is usually also above in most, and in many 
cases all, other subjects." 



Similarities of Twins in Special Mental Traits. The two prin- 
cipal investigations on this phase of mental heredity were made by 
Galton and Thorndike. Galton made a general comparison of two 
groups of twins, one group of 35 pairs, which were reported as being 
very similar, in fact so similar that they were frequently reported 



86 EDUCATIONAL PSYCHOLOGY 

as indistinguishable, and another group of twenty pairs of dis- 
tinctly dissimilar twins. The conclusion formulated was to the 
effect that the former twins remained very similar all through 
life in spite of ditTerent environments, while the latter twins re- 
mained dilTerent all tlirough life in spite of similar environments. 
Concerning certain of the twins Gal ton reports: 

"i. One parent says: 'They have had exactly the same nurture from 
their birth up to the present time; they are both perfectly healthy and 
strong, yet they are otherwise as dissimilar as two boys could be, phys- 
ically, mentally, and in their emotional nature.' 

''2. T can answer most decidedly that the twins have been perfectly 
dissimilar in character, habits, and likeness from the moment of their 
birth to the present time, though they were nursed by the same woman, 
went to school together, and were never separated till the age of fifteen.' 

"3. 'They have never been separated, never the least differently 
treated in food, clothing, or education; both teethed at the same time, 
both had measles, whooi)ing-cough. and scarlatina at the same time, and 
neither had any other serious illness. Both are and have been exceed- 
ingly healthy and have good abilities, yet they differ as much from each 
other in mental cast as any of my family differ from another.' 

"5. 'They were never alike either in body or mind and their dissim- 
ilarity increases daily. The e.xteriial influences have been identical; they 
have never been separated.' 

"9. 'The honie-traininp; and influence were precisely the same, and 
therefore I consider the dissimilarity to be accounted for almost entirely 
by innate disposition and by causes over which we ha\e no control.'" 
('83, p. 170, Everyman's Library Edition.) 

Gallon's general impression of his results is as follows: 

"We may, therefore, broadly conclude that the only circumstance, 
within the range of those by which persons of similar conditions of life 
are affected, that is capable of j>roducing a marked effect on the char- 
acter of adults, is illness or some accident that causes physical infirm- 
ity. . . . The impression that all this leaves on the mind is one of some 
wonder whether nurture can do anything at all, beyond giving instruc- 
tion and professional training. There is no escape from the conclusion 
that nature prevails enormously ovx-r nurture when the differences of 
nurture do not exceed what is commonly to be found among persons of 
the same rank of society and in the same country." ('S3, pp. 168 and 1 7 J.) 

Thorndike's investigation (05) was made by more accurate 
methods. He applied the tests, mentioned in the following table, 
to 50 pairs of twins and found the following correlations: 



THE INHERITANCE OF MENTAL TRAITS 



87 



In the A-test R — 

In the a-t and r-e test R — 

In the misspelled word test R — 

In addition R — 

In multiplication R — 

In the opposites test R — 



"If now these resemblances are due to the fact that the two members 
of any twin pair are treated alike at home, have the same parental 
models, attend the same school and are subject in general to closely 
similar environments, then (i) twins should, to the age of leaving home, 
grow more and more alike, and in our measurements the twins 13 and 
14 years old should be much more alike than those 9 and 10 years old. 
Again (2), if similarity in training is the cause of similarity in mental 
traits, ordinary fraternal pairs not over four or five years in age should 
show a resemblance somewhat nearly as great as twin pairs, for the 
home and school conditions of the former will not be much less similar 
than those of a pair of the latter. Again (3) if training is the cause, 
twins should show a greater resemblance in the case of traits much sub- 
ject to training, such as ability in addition or in multiplication, than in 
traits less subject to training, such as quickness in marking off the A's on a 
sheet of printed capitals, or in writing the opposites of words. 

"On the other hand (i) the nearer the resemblance of young twins 
comes to equalling that of old, (2) the greater the superiority of twin 
resemblance to ordinary fraternal resemblance is, and (3) the nearer twin 
resemblance in relatively untrained capacities comes to equalling that in 
capacities at which the home and school direct their attention, the more 
must the resemblances found be attributed to inborn traits. 

"The older twins show no closer resemblance than the younger twins, 
and the chances are surely four to one that with an infinite number of 
twins tested, the 12-14 year-olds would not show a resemblance ,15 
greater than the 9-1 1 year-olds. The facts are: (Thorndike '14, III, 
pp. 248-249). 

TABLE 21 
The resemblances of young and old twins compared. 



Twins, 12-14 



1 . A-test 

2. a-t and r-e tests. . . . 

3. Misspelled word test 

4. Addition 

5k Multiplication 

6. Opposites 

Averages 




8S 



KDUCATIOXAI. rsVCHOLOr.V 



The Influence of Uniform Environment Upon Different Original 
Abilities. All studies cited thus far have attempted to measure 
the amount of similarity in related persons as compared with un- 
related individuals on the assumption that the environment was 
roughly constant for all, that whatever resemblances existed be- 
tween pairs of brothers and sisters or between other types of rel- 
atives greater than that between any pairs of persons selected by 
chance from the population at large, is considered to represent 
the actual amount of similarity in the original inherited natures of 
the individuals springing from the same ancestry. The problem 
may, however, be pursued further from a dilTerent angle, namely, by 
specific control of the environment or some portion of it. Thus it is 
possible to keep some particular part of the environment uniform 
for a group of persons of widely ditTerent abilities and to measure 
to what extent the original ditTerences remain constant, increase, 
or decrease. If the differences remain constant, or increase, the in- 
ference would be that the ultimate dilTcrences of achievement would 
be primarily due to the inherited differences of capacities. If the 
differences decrease materially and finally disappear, the original 
differences would be mainly due to the elTect of environment and 
opportunity. 

A number of such experiments have been carried out. An in- 
vestigation made by the writer ('ii) in which S persons multiplied 
mentally 50 3-place numbers by a i-place number each day for 
14 successive days showed the following amounts of improvement: 

T.MJLE 22 



NUMBKR OF 
lOXAUI'I.KS IN 
1st 10 .Mis. 



NCMBER OF 
EXAMPLKS IN 

Last 10 .Miv. 



Gain in No. of 
Examples 



Pkr Cent 
Gain 



Three best persons . . 
Three poorest persons 



39 



51 



45 
26 



104 



Hence, both the greatest absolute and the greatest relative gain 
was made by the group with the highest initial records. 

Similar results have been found in the practice experiments of 
sul)stiluting numbers for letters as described in the author's Ex- 
pert turn Is in luliualiottal Psyc/iolof^y, Chapter X. The follow- 
ing table gives the highest five and the lowest five records from 
aniong twenty persons. Each person practiced uo minutes. 



THE INHERITANCE OF MENTAL TRAITS 



89 



TABLE 23 
Average number of letters transcribed. 



First 5 Min. Last 5 Min 



Initial highest five persons . 
Initial lowest five persons . 



139 
100 



310 
239 



171 
139 



Again the largest gain was made by the group having the greatest 
initial ability. 

Results pointing in the same direction have been obtained by 
Thorndike, Whitley, and others. For example, Thorndike ('lo) 
found in the case of practice of nineteen persons in adding, the fol- 
lowing results: 

TABLE 24 

The effect of equal amounts of practice upon individual differences in column 
addition of one-place numbers. After Thorndike ('10). 



• 


Average Number of Additions per 5 Minutes 
Corrected for Errors 


Average Time 
Spent in Practice 

FROM Mro-POINT OF 

First Test to Mid- 
point OF Last Test 
(in Minutes) 




First Test 


Last Test 


Gain 


Initially highest 6 in- 
dividuals 

Initially next highest 
6 individuals 

Initially lowest 7 in- 
dividuals 


297 
234 
167- 


437 
345 
220 -|- 


140 
III 

54 


40 

49 
46 







The statistical studies of scholastic histories of pupils through 
various periods of school life, which were discussed in a preceding 
chapter under the heading of correlations of abilities at different 
times of life in the same individual, all tend to corroborate the ex- 
perimental facts here presented. The scholastic records show to a 
remarkable extent the uniformity with which each individual main- 
tains his position throughout his educational career. A very in- 
teresting tabulation was made by L. J. Coubal and E. VanLande- 
gend ^ to show the progress made by pupils in grades 4, 5, and 6 
in one school in the four fundamental operations in arithmetic. 
Progress, was measured by the Courtis tests, Series B, month by 

1 Under the direction of Professor Henmon, and reported in two theses in the library 
of the University of Wisconsin, 1917. 



9© 



EDUCATIONAL PSYCHOLOGY 



month through an tntirc year. The records in the four opera- 
tions for each pupil were combined into a single score. Figures 32, 
eo 



40 



ao 



20 



10 















/ 
















' .' 













//y 





=!^ 










\^' 


/ 




^ 













/ 





,Bept. Oct. Nov. Dec. Jan. Feb. March April May, 
Fig. 2,2. — Progress in the four fundamental operations in arithmetic as 
measured by the Courtis tests, Series li, gi\en at monthly intervals. The 
heavy continuous line represents all the pupils of the 4th grade. The four broken 
lines represent these pupils divided into quartiles. 
TO 



60 



60 



40 



80 



ao 





























> 




\ 












• 
• 


/y^' 


-N 










/J 


yy 








J 


i^''' 


'"^. 


/^ 






\ 

\ 
\ 
\ 




—•— ■ 


yy 


— " ^^ 





























10 ~^^ 



S«pt. Oct. Nov. Dec. Jan. 



Feb. 



March April May 

Fir.. 33. — Same as Fig. 32, for the 5th grade. Pupils divided into three groups 
instead of four. 

3,^, ancl ;^4 give the curves of progress for the respective grades. 
The pupils in each grade were divided into groups according to 
their final performanc e. Thus the pupils in grade 4 were divided 



THE INHERITANCE OF MENTAL TRAITS 



91 



into four groups while those in the other grades were divided into 
three groups. The results reveal the significant fact that the best 
groups in each grade made the greatest progress, the poorest groups 
made the least progress and the intermediate groups made average 
progress. The graphs for the various groups in any grade gradu- 
ally spread apart during the course of the year, indicating that the 
differences increase rather than decrease or remain constant. The 
more gifted pupils profit more by their school work than the less 
gifted. 

All experimental results point in the direction that practice does 

not equalize abilities; in fact, equal practice tends to increase differ- 

70 



60 



60 



40 



30 



20 



10 



Sept. Oct. Nov. Dec. Jan. Feb. March April May 
FiG. 34. — Same as Figs. 32 and S5, for 6th grade. 

ences in achievement and skill rather than to decrease them. The 
more gifted individuals profit more, both relatively and absolutely, 
than the less gifted. This experimental fact is one of the most pro- 
found bits of evidence regarding the whole problem of heredity and 
environment. The talented men not only start with greater initial 
capacities but seem also to be capable of more intense application 
and more zealous desire to improve. "To him that hath shall be 
given" is psychologically true in the sphere of intellectual training 
as well as in the sphere of morality and religion. The man with 
ten intellectual talents will acquire far more than the man with one 
talent. If we may generalize for life as a whole, equal opportunities 

















^^.•' 


























,,-- 


' / 


,y^ 

^ 


'^ 










— -- 


V\ 


— 




■irrzTl 




^ 


y 




/ 






___^- 


,^^^' 


^ 


y 



























92 EDUCATIONAL PSYCIIOUXiY 

for all do not produce equal abilities in all. Men may be born free 
politically, but they are not born equal mentally; they may be 
born equal in opportunities in a democratic society, but they cer- 
tainly are not equal in their ultimate achievements in life. 

Influence of Different Environments upon Various Original 
Abilities. Iv\lensi\c inquiries into the effects of xarious enxiron- 
mental conditions upon the native ability of human beings have 
been made in other fields besides the experimental one which has 
been surveyed. Such investigation as the study made by Dr. 
Rice ('97 and 02) upon the effects of various factors in school life 
upon the attainments in spelling and in arithmetic, the studies of 
places of birth of American men of science made by Cattell ('06), 
or the study of the places of birth of eminent men of letters made 
by Odin, and similar investigations by I)e Candolle ('73), Jacoby 
('81), and L^ilis, have ])een extensively referred to as bearing upon 
the problem of environmental forces in their inteq)lay with heredi- 
tary capacities. The real signilicance and argumentative weight of 
such data seem to the writer to be uncertain and duplex in their 
meaning. Cattell, for example, has pointed out that the number of 
eminent scientific men born about 1S60 in Massachusetts per one 
million population was very much greater than the number of 
eminent scientific men born in proportion to population in other 
states. To cite a few instances, he has com|)uted that per one 
million population there were bom eminent scientists as follows in 
various states: 

Mass 108.8 

Conn 86.9 

R. 1 25.6 

N. Y 47 o 

Wis 450 

111 24.0 

Ala 2.0 

Miss 1 .0 

Similar figures are given for other states, and the inference made 
by Cattell is that the environment of Massiichusetts and similar 
states has been much more conducive to the development of scien- 
tific men and that the number of such men could be determined 
practically by the control of the i^roper educational stimuli. 

Odin, in his study of 5,2.^3 noted French men t)f letters living 
during the period 1400 to 1S30, fi)un(l the following distribution 
according to places of Ijirth: 



THE INHERITANCE OF MENTAL TRAITS 93 

1,229 born in Paris 

2.264 " " other large cities 

1.265 " " small cities 

93 " " country districts 

From this it has been inferred that if France as a whole had 
been as fertile as Paris in the production of eminent men of letters 
there would have been approximately 54,000 great men of letters 
instead of less than 6,000. The difficulty, however, with both 
studies is that Paris and Massachusetts have been more productive 
of eminent men not necessarily on account of better educational 
and social environment, but possibly also because of the fact that 
eminent men of letters and science have by virtue of the location 
in them of educational institutions, scientific and other intellectual 
centers, necessarily been attracted to these places, and conse- 
quently their children were born in these localities. The facts as 
such may actually be used in the support of heredity as against 
enviromiient as much as they have been used in support of en- 
vironment as against heredity. 

Likewise, the study of Rice with regard to the factors affecting 
efficiency in school subjects is uncertain. Rice, on the basis of 
extensive tests in spelling and arithmetic in various parts of the 
country, arrived at the general opinion that practically all external 
conditions of home and school such as foreign or American parent- 
age, home study, methods of teaching, size of class, and time 
devoted to study, made practically no difference whatever in the 
ultimate achievement of the pupils, and the implication is made 
that the final efficiency depends primarily upon heredity. The 
obvious uncertainty of such data as these is that while the facts 
in toto may imply such a situation, it is also quite certain that such 
a massing of data in this manner obliterates the effect of individual 
factors. Favorable conditions may be offset by unfavorable condi- 
tions and thus obscure the entire situation. To infer that good 
teaching and poor teaching make no difference in the ultimate 
results obtained, or that the amount of time given to study makes 
no difference in results, are conclusions that are quite likely to be 
unsound. The reason for the inference drawn by Rice is probably 
the fact that good teaching in some schools may be accompanied 
by other factors which tend to counteract its effect, whereas poor 
methods of teaching in other schools may be accompanied by 
favorable or unfavorable circumstances in other respects. The 
massing together of returns from many schools is bound to ob- 



c;4 i;i)rr.\TION'AL psvciioL<onv 

literati' the effects of the individual conditions. The only certain 
way to ascertain the effectiveness of one factor or another would 
he to control all conditions, or to be able to allow for theni definitely, 
with the exception of the one factor whose efBcacy is to be deter- 
mined. Thus in order to determine whether or not the different 
methods of teaching a given subject make a difference, it would be 
necessary to take a class of pupils in a given school and divide it up 
into two or more groups on the basis of equal initial capacities and 
to have each class taught, preferably by the same teacher and un- 
der the same general circumstances, by a different method. Then at 
stated intervals the two groups should be compared l)y reliable 
measures. In this manner definite results could be obtained as 
to the effectiveness of the various methods, conditions, or amounts 
of time devoted to a subject. Consequently, Rice's figures as they 
stand are of little worth so far as proving the forcefulness of dif- 
ferent environmental factors in the production of results is con- 
cerned. 

A similar criticism applies to such studies as that made by Spill- 
man with regard to the place of birth of various men prominent 
in ])ublic and business life, such as presidents of the United States, 
governors, and railroad ])residents. Spillman ('09) has pointed 
out, for example, that 23 of the first 25 presidents of the United 
States were born in the country, that 41 out of 45 governors, and 
47 out of 62 cabinet ofBcers were bom in the country. It is unsafe 
to argue that because a large percentage of the presidents of the 
United States or other jirominent persons were bom and reared 
in rural districts, that country life is more productive of ability 
and ambition. To argue with any certainty on the basis of such 
facts, it is necessary to know the ancestral antecedents of the 
])ersons springing from various localities and sections of the popu- 
lation. 

General Interpretation. The general impression from all e.\- 
])erimental, statistical, and historical material thus far accumulated 
on the problems of mental heredity would seem to be somewhat 
as follows: Harring jraupers, invalids, and those suffering from 
want of food and shelter due to conditions beyond their personal 
control, and referring to all others living in the same community 
at the same time, the ultimate achievement of any gi\en individual 
is (\uv to his original ability, probably to the e.xtent of Oo to f)0%, 
and to actual differences in opportunity or e.xtemal circumstances 
only to the extent of 10 to 40%. 



THE INHERITANCES OF MENTAL TRAITS 95 

The facts of heredity bear down so heavily that the impression 
gained of the large part played by it leads one almost to a fatalistic 
philosophy. One is almost inclined to believe that persons become 
what they do largely on account of their hereditary capacities, and 
that they are not in the least responsible for their own outcome; 
that if a person is born with great capacities he will achieve high 
distinction, and if he is bom with mediocre or slender capacities, 
he will not achieve anything beyond his limits no matter what he 
may do. While it is certainly true that no one may achieve a 
position higher than his original capacities will permit, it does not 
follow that a mechanical, fatalistic view needs to be taken. Nature 
predominates enormously over nurture only in the relative and 
not in the absolute sense. This distinction must always be borne 
in mind in studies of heredity. In fact, in the absolute sense, 
nurture predominates enormously over nature. A Newton bom 
among Australian bushmen would no doubt have become a re- 
markable bushman, but never a world-renowned scientist. The 
necessary stimuli of environment must be at hand to train and 
develop original capacities. The difference between relative and 
absolute achievement may be illustrated in any of the experimental 
results concerning the effects of equal practice cited in a preceding 
section. The fact that all individuals improve by practice shows 
absolute gain in performance or skill. The fact that the gifted ones 
maintain their lead or even gain in their lead is relative achieve- 
ment. Before practice, no child can write; after practice, all normal 
children can write with more or less excellence. This is absolute 
gain. Before practice, some children have greater original capaci- 
ties for learning to write ; after practice, these same children main- 
tain the same superiority. This is relative gain. A Newton and 
an ordinary bushman born and reared among bushmen would 
probably be superior and ordinary bushman respectively. A 
Newton and an ordinary bushman born and reared in New York 
City at the beginning of the 20th century would probably become, 
respectively, the one a great scientific, professional or business 
man, and the other an ordinary person, able to get on, earn a 
living, and enjoy life within the ordinary Umits. The original 
abilities of ancient civilized peoples were probably very little 
different from the original abilities of modem civilized peoples. 
The differences are probably due to the transformation of the 
environment which is constantly being brought about through 
the efforts of man. A Newton bom in a modern civilized com- 



96 EDUCATIONAL PSYCHOLOGY 

munity would ha\'c ^cater and difTerent stimuli than one horn 
in an ancient or uncivilized community. His ultimate eminence 
woultl br dctfrniini-d l^y liis environment. 

The jxissimi.siic air may further be dispelled by noting the fact 
that hardly one person in a thousand makes all the absolute gain 
possible for him even in a single ca|)acity. It has been proved over 
and over again in numerous abilities which ha\'e been used daily 
in one's occuj)ation that by a little special practice each day their 
efficiency may be enormously improved. Consequently, while the 
possibilities of each individual are limited by his original inherited 
ec|uipment, each one may develop his capacities far beyond the 
usual degree of attainment. While experimental evidence indicates 
emphatically that under equal opportunities the more gifted surpass 
the less gifted, yet rarely does anyone do his l^est or attain his limit 
even in a single capacity. Life is a matter of competition; let every- 
one compete to the fullest extent of his inherited abihty. 



^7 



CHAPTER VII 

THE MEASUREMENT OF MENTAL CAPACITIES 

Problem. Strictly speaking, it is impossible to measure directly 
the original equipment of a human being unmodified by environ- 
mental causes. The nearest approach would be the preparation of 
a complete inventory, and an exact measurement, of all the capac- 
ities that an individual possesses at birth. Even then, pre-natal 
conditions have entered into the growth of the organism. The next 
nearest approach would be a measurement of all capacities which 
are not directly or specifically trained by school, occupation, or 
special circumstances. In fact, no one can live and possess capac- 
ities without any modification of them from outside causes; hence 
the best that we can do is to measure as many capacities and abil- 
ities as possible which have been modified least by special exercise 
or training, and then to consider them as approximately represent- 
ing a person's original abilities, or, to make such allowances as we 
can for the influence of external causes. No human being up tc 
the present time has been measured in all respects at any given 
point in his growth by thoroughly accurate methods. A great 
many persons, however, have been partially measured in a great 
many capacities by more or less accurate or inaccurate methods at 
various stages of their growth. 

General Value. John Stuart Mill has said the "greatest thing 
in the world is man, and the greatest thing in man is mind." To 
this statement we might possibly add that the greatest achieve- 
ment of science would be the measurement of the mind. The im- 
port and value of definite means for measuring the capacities of 
human beings would touch all phases of human life in which in- 
telligence is involved. If we had accurate means of describing a 
given person's capacities in all directions in terms that could be pre- 
cisely defined and understood, we would have an instrument for 
evaluating human beings far beyond our present possibilities. We 
would then be able to obtain a precise notion of the capacities of 
an individual. Consider for a moment what the advantages would 
be! In all sorts of human relations, men are called upon con- 
stantly to pass judgment upon their fellows concerning their fit- 

97 



98 EDUCATIONAL PSYCHOLOGY 

ness, capacity, and promise of success for this or that particular 
line of work. The one tiling which is probably most important of 
all, aside from the special training in a given field, is the intelligence 
and native ability which a person possesses. What aptitudes does 
a person have for this or that type of work? Enormous waste in 
the energies of men are due to mal-adaptation of individual to 
task. The business world is rapidly turning toward psychology for 
help, and if psychology is to give the help it will have to be in the 
form of adequate measurements of the capacities of human beings. 
Sound vocational guidance, in which much interest has recently 
sprung uj), will have to be founded upon a sound vocational psychol- 
ogy whose development lies largely in the future. Courts are 
recognizing that responsibility for conduct rests upon mental 
maturity and intelligence, and that these must be determined first 
before proper adjudication may be made of an individual's behavior. 
Psychological laboratories have therefore been established in recent 
years in connection with juvenile courts. The immigration office 
finds it necessary to make intelligence examinations, even if crude, 
in order to exclude those distinctly unfit. In normal times a con- 
siderable number, So to 100 per month, are returned to the coun- 
tries whence they came on account of mental deficiency. 

One of the large problems of the school is the proper adjustment 
of work and progress to the natural ability of the pupils and, in 
particular, the discovery of the morons and borderline individuals 
so that they may be taken care of in special classes or otherwise to 
the best advantage to themselves and to the other pupils in the 
school. Intelligence examinations would be useful not only in 
connection with the relatively small percentage of backward and 
defective pupils, but also in connection with the normal and supe- 
rior individuals. Such tests would be valuable in conjunction 
with the measurement of attainment in school subjects specifically, 
so that a child's progress and rate of advancement could be deter- 
mined on the basis of both types of measurements. Precise methods 
of evaluating the actual capacities of pupils would be of decided 
value in making possible a more accurate promotion or retardation 
of pupils according to their abilities, and a more accurate prescrij>- 
tion of work to be done and of the progress that can most profitably 
be made. The school has paid relatively more attention to the 
backward pupils by putting them into special classes than to the 
superior ones. And yet the latter will be the ones who will con- 
tribute most to the advancement of society as a whole. Why 



THE MEASUREMENT OF MENTAL CAPACITIES 99 

should there not be special classes for the gifted pupils so that 
they might be led to reach their fullest intellectual growth and 
thus return to society the most that they are capable of? 

Methods of Measuring Original Capacities. In general two 
types of methods have been developed, at least in part, and used 
for determining the native ability of human beings. The term 
"native" of course, must be understood to signify not pure, native 
ability unmodified by experience, but native or original only in the 
sense of not being directly affected by specific training. The one 
method consists of a considerable variety of reactions to questions 
and situations which a child would be able to make as a result of 
normal growth in a normal environment. The tests developed on 
this principle are the Binet-Simon tests and the various modifica- 
tions of them. 

The second general method has proceeded on the basis of meas- 
uring, by fairly precise methods, certain special mental functions 
from year to year, and of determining thereby the mental status 
and growth of the individual. Thus, for example, many capacities 
might be measured by a definite psychological technique from year 
to year, and certain norms might be established for each year so 
that we could say that a given individual's memory has been de- 
veloped to the norm or average of a child of ten. Similar tests and 
norms could be developed in as many different mental capacities as 
would seem to be necessary in order to obtain a fairly complete 
evaluation of an individual's natural abilities. This second general 
method has not as yet been developed to the same degree of com- 
pleteness as the Binet-Simon type ('05) with respect to either the 
selection of the particular capacities that should be tested, or the 
types of tests that ought to be used, or the technique by which 
they should be given. Brief consideration will be given to both 
plans of measurement. 

The Binet-Simon Scale. This series of tests is arranged in 
groups according to years. Thus there is a series of tests for every 
year from age three up to twelve or fifteen, and in some of the 
revisions even to adult life. These tests were first prepared by the 
French psychologist, Binet, and the French physician, Simon, 
who collaborated for a period of twelve or fifteen years in the 
selection of tests and in assigning them to the proper years accord- 
ing to the growth and development of the child. These tests were 
first published in 1905 and since then were revised by the original 
authors in 1908 and in 191 1. A number of investigators have 



lOO EDUCATIONAL PSYCHOLOGY 

attempted to rt\ isc them and to adapt tliem to tlie conditions in 
their respective countries, and to improve them so that they would 
be more reUal)le and more accurately graded according to the 
growth of children from year to year. In this country, the chief 
revisions and imi)rovements have been made by Goddard, Kuhl- 
mann, Verkes, Tennan, and others. Prol)ably the most Siitis- 
factory and careful revision of the original Binet tests is the one 
recently j)repared by Terman and known as the Stanford Re\'ision. 
This revision consists in the elimination of some of the original 
tests, in the addition of a considerable numl)er of new tests, in the 
readjustment of other tests uj) or down the scale of years according 
to their difficulty, and particularly in the develoi)ment of a more 
precise technique for giving and evaluating the tests, so that ex- 
aminers may be guided specifically in the administration of them. 
The following is a complete list of the tests in the Stanford re- 
vision. The detailed directions for giving and scoring the tests 
together with extensive results, are given in Tcrman's The Meas- 
uremcnl of Intelligence. 

The Stanford Revision and Extension 

Ycur III. (6 tests, 2 months each.) 

1. Points to parts of body. (3 of 4.) 

Nose; eyes; mouth; hair. 

2. Names familiar objects. (3 of 5.) 

Key, penny, closed knife, watch, pencil. 

3. Pictures, enumeration or belter. (At least 3 objects enumerated 

in one picture.) 
(a) Dutch Home; (b) River Scene; (c) Posi-OflHe. 

4. Gives sex. 

5. Gives last name. 

6. Repeats 6 to 7 syllables, (i of 3.) 

.\i. Repeats 3 digits, (i success in 3 trials. Order correct.) 

Year I\'. (6 tests, 2 months each.) 

1. Comi)ares lines. (3 trials, no error.) 

2. Dis( rimination of forms. (Kulilmann.) (Not over 3 errors.) 

3. Counts 4 pennies. (No errors.) 

4. Copies square. (Pencil, i of 3.) 

5. Comprehension, ist degree. (2 of 3.) (Stanford addition.) 
"What nnust you do?" "When you are sleepy?" "Cold?" 

"Hungry?" 
(^. Repeats 4 digits, (i of 3. Order correct.) (Slanford addition.) 
A I . Repeats 1 2 to 13 syllables, (i of 3 absolutely correct, or 2 with i 
error eai h.) 



THE MEASUREMENT OF MENTAL CAPACITIES loi 

Year V. (6 tests, 2 months each.) 

1. Comparison of weights. (2 of 3.) 

3-15; 15-3; 3-15- 

2. Colors. (No error.) 

Red; yellow; blue; green. 

3. .^Esthetic comparison. (No error.) 

4. Definitions, use or better. (4 of 6.) 

Chair; horse; fork; doll; pencil; table. 

5. Patience, or divided rectangle. (2 of 3 trials, i minute each.) 

6. Three commissions. (No error. Order correct.) 
Ai. Age. 

Year VI. (6 tests, 2 months each.) 

1. Right and left. (No error.) 

Right hand; left ear; right eye. 

2. Mutilated pictures. (3 of 4 correct.) 

3. Counts 13 pennies, (i of 2 trials, without error.) 

4. Comprehension, 2nd degree. (2 of 3.) "What's the thing for 

you to do?" 

(a) "If it is raining when you start to school?" 

(b) "If you find that your house is on fire?" 

(c) "If you are going some place and miss your car?" 

5. Coins. (3 of 4.) Nickel; penny; quarter; dime. 

6. Repeats 16 to 18 syllables, (i of 3 absolutely correct, or two with 

I error each.) 
Ai. Morning or afternoon. 

Year VII. (6 tests, 2 months each.) 

1. Fingers. (No error.) Right; left; both. 

2. Pictures, description, or better. (Over half of performance de- 

scription.) Dutch Home; River Scene; Post-Oflfice. 

3. Repeats 5 digits, (i of 3. Order correct.) 

4. Ties bowknot. (Model shown, i minute.) (Stanford addition.) 

5. Gives differences. (2 of 3.) 

(Fly and butterfly; stone and egg; wood and glass.) 

6. Copies diamond. (Pen. 2 of 3.) 

Ai. I. Names days of week. (Order correct. 2 of 3 checks cor- 
rect.) 
Ai. 2. Repeats 3 digits backwards, (i of 3.) 

Year VIII. (6 tests, 2 months each.) 

1. Ball and field. (Inferior plan or better.) (Stanford addition.) 

2. Counts 20 to I. (40 seconds, i error allowed.) 

3. Comprehension, 3d degree. (2 of 3.) "What's the thing for 

you to do?" 



I02 EDUCATIONAL l'SVCHOLO(;V 

(a) " When you have broken something which belongs to some- 

one else?" 

(b) "When you are on your way to school and notice that you 

are in danger of being tardy?" 

(c) "If a i)laymate hits you without meaning to do it?" 

4. Gives similarities, two things. (2 of 4.) (Stanford addition.) 

Wood and coal; apple and peach; iron and silver; ship and 
automobile. 

5. Definitions superior to use. (2 of 4.) 

Balloon; tiger; football; soldier. 

6. X'ocabulary, 20 words. (Stanford addition. For list of wunls 

used, see record booklet.) 
Ai. I. First si.\ coins. (No error.) 

.\i. 2. Dictation. ("See the Uttle boy." Easily legible. IVn. i 
minute.) 

Year IX. (6 tests, 2 months each.) 

1. Date. (.Ulow error of 3 days in c. no error in a, b, or d.) 

(a) day of week; (b) month; (c) day of month; (d) year. 

2. Weights. (3, 6, 9, 12, 15. Procedure not illustrated. 2 of 3.) 

3. Makes change. (2 of 3. No coins, paper, or pencil.) 

10-4; 15-12; 25-4. 

4. Repeats 4 digits backwards, (i of 3.) (Stanford addition.) 

5. Three words. (2 of 3. Oral. 1 sentence or not over two co- 

ordinate clauses.) 

Boy, river, ball; work, money, men; desert, rivers, lakes. 

6. Rhymes. (3 rhymes for two of three words, i minute for each 

part.) Day; mill; spring. 
Ai. .1. Months. (15 seconds antl i error in naming. 2 checks of 3 

correct.) 
Ai. 2. Stamps, gives total value. (Second trial if individual values 
are known.) 

"^'ear X. (6 tests, 2 months each.) 

1. Vocabulary, 30 words. (Stanford addition.) 

2. Absurdities. (4 of 5. Warn. S|)onlaneous correction allowed.) 

(Four of Binet's, one Stanford.) 

3. Designs, (i correct, 1 half correct. E.xjxjse 10 seconds.) 

4. Reading and report. (8 memories. 35 seconds and 2 mistakes in 

reading.) (Binet's selection.) 

5. Comprehension, 4th degree. (2 of 3. Question may be repeated.) 

(a) "What ought you to say when some one asks your opinion 

about a person you don't know very well?" 

(b) "What ought you to do before undertaking (beginning) 

something very imjMjrlant?" 



THE MEASUREMEJIT OF MENTAL CAPACITIES 103 

(c) " Why should we judge a person more by his actions than 
by his words?" 
6. Name 60 words. (Illustrate with clouds, dog, chair, happy.) 
Ai. I. Repeats 6 digits, (i of 2. Order correct.) (Stanford addi- 
tion.) 
Ai. 2. Repeats 20 to 22 syllables, (i of 3 correct, or 2 with i. error 

each.) 
Ai. 3. Form board. (Healy-Fernald Puzzle A. 3 times in 5 min.) 

Year XII. (8 tests, 3 months each.) 

1. Vocabulary, 40 words. (Stanford addition.) 

2. Abstract words. (3 of 5.) 

Pity; revenge; charity; envy; justice. 

3. Ball and field. (Superior plan.) (Stanford addition.) 

4. Dissected sentences. (2 of 3.) (i minute each.) 

5. Fables. (Score 4; i. e., two correct or the equivalent in half 

credits.) (Stanford addition.) 
Hercules and Wagoner; Maid and Eggs; Fox and Crow; 
Farmer and Stork; MiUer, Son, and Donkey. 

6. Repeats 5 digits backwards, (i of 3.) (Stanford addition.) 

7. Pictures, interpretation. (3 of 4. "Explain this picture.") 

Dutch Home; River Scene; Post-Office; Colonial Home. 

8. Gives similarities, three things. (3 of 5.) (Stanford addition.) 

Snake, cow, sparrow; book, teacher, newspaper; wool, cotton, 
leather; knife-blade, penny, piece of wire; rose, potato, tree. 

Year XIV. (6 tests, 4 months each.) 

1. Vocabulary, 50 words. (Stanford addition.) 

2. Induction test. (Gets rule by 6th folding.) (Stanford addition.) 

3. President and king. (Power; accession; tenure. 2 of 3.) 

4. Problems of fact. (2 of 3.) (Binet's two and one Stanford addi- 

tion.) 

5. Arithmetical reasoning, (i minute each. 2 of 3.) (Adapted 

from Bonser.) 

6. Clock. (2 of 3. Error must not exceed 3 or 4 minutes.) 

6.22; 8.10; 2.46. 
Ai. Repeats 7 digits, (i of 2. Order correct.) 

Average Adult. (6 tests, 5 months each.) 

1. Vocabulary, 65 words. (Stanford addition.) 

2. Interpretation of fables. (Score 8.) (Stanford addition.) 

3. Difference between abstract words. (3 real contrasts out of 4.) 

Laziness and idleness; evolution and revolution; poverty and 
misery; character and reputation. 

4. Problem of enclosed boxes. (3 of 4.) (Stanford addition.) 



I04 EDUCATIONAL PSYCHOLOGY 

5. Repeats 6 digits backwards. (lof^.) (Stanford addition.) 

6. Code, writes "Come quickly." (2 errors. Omission of dot 

counts half error. Illustrate with "war" and "spy.") (From 
Healy and Fernald.) 
Ai. I. Repeals 28 syllables, (i of 2 absolutely correct.) 
Ai. 2. Comprehension of physical relations. (2 of 3.) (Stanford 
addition.) 
Path of cannon ball; weight of fish in water; hitting distant 
mark. 

"Superior .Adult." (6 tests, 6 months.) 

1. Vocabulary, 75 words. (Stanford addition.) 

2. Binel's paper-cutting test. (Draws, folds, and locates holes.) 

3. Repeats S digits, (i of 3. Order correct.) (Stanford adfiition.) 

4. Repeats thought of passage heard, (i of 2.) (Binel's and Wis- 

sler's selections adapted.) 

5. Rei)eals 7 digits backwards, (i of 3.) (Stanford addition.) 

6. Ingenuity test. (2 of 3. 5 minutes each.) (Stanford addition.) 

The mental maturity or intelligence of a child is expressed in 
terms of a quotient which represents the relation between his 
mental age and his chronological age and is obtained by dividing 
the former by the latter. Thus, a child 10 years old with a mental 
age of 10 would have an intelligence quotient (IQ) of i.oo; a child 
10 years old with a mental age of 11, would have an intelligence 
(|uotient of i.io, or a child 10 years old with a mental age of S 
would have an intelligence quotient of .80. If the Cjuotient is 
under i.oo it means that the child is below the average and if it 
is above i.oo, it means that the child is above the average. Ter- 
rnan has suggested the following cias.sitication of intelligence 
cjuutients with their approximate meanings: 

" IQ Classification 

Above 1 .40 "Near" genius or genius. 

1 .20-1 .40 Very superior intelligence. 

I . lo-i .20 Superior inlL-lligcncc. 

.90-1 .10 Normal, or average, intcllipcnco. 

.80- .90 Dullness, rarely classifial)le as fcehlc-mindcdness. 

.70- .80 Border-line deficiency, sometimes classifiable as dullness, 

ofti-n as feei)le-mindcdncss. 
Helow .70 Definite feel)Ic-mindcdncss. 

"Of the feeble-minded, those between .50 anfl .70 TO include 
most of the morons (high, middle, and low), those between .20 or 



THE MEASUREMENT OF MENTAL CAPACITIES 105 

.25 and .50 are ordinarily to be classed as imbeciles, and those 
below .20 or .25 as idiots. According to this classification the 
adult idiot would range up to about 3-year intelligence as the 
limit, the adult imbecile would have a mental level between 3 and 
7 years, and the adult moron would range from about 7-year to 
ii-year intelligence." 

Terman ('17) has attempted to estimate the boyhood intelligence 
quotient of Sir Francis Galton from such records as are available 
of his youth, and believes it quite certainly to have been 2.00. 
For example, on the day before his fifth birthday he wrote the 
following letter to his sister, the statements of which have been 
corroborated by other general evidence: 

" My dear Adele, 

" I am 4 years old and I can read any English book. I can say all the 
Latin Substantives and Adjectives and active verbs besides 52 lines of 
Latin poetry. I can cast up any sum in addition and can multiply by 
2, 3, 4, 5, 6, 7, 8, (9), 10, (11). 

" I can also say the pence table. I read French a little and I know the 
clock. 

" Francis Galton, 
" February 15, 1827." 

At the age of 10, young Galton wrote the following letter which 
represents maturity of judgment and intellectual interest worthy 
of a high school or college student: 

" December 30, 1832. 
" My Dearest Papa: 

" It is now my pleasure to disclose the most ardent wishes of my heart, 
which are to extract out of my boundless wealth in compound, money 
sufficient to make this addition to my unequaled library. 

The Hebrew Commonwealth by John 9 

A Pastor's Advice 2 

Hornne's Commentaries on the Psalms 4 

Paley's Evidence on Christianity . . . ." 2 

Jones Biblical Cyclopedia 10 

27 

To illustrate concretely the manner of determining the mental 
age of a child, the following record of the examination of a boy 14 



io6 



EDUCATIONAL I'SVCHOUKiV 



years aiul ii months (jld is cited. Thi- test numbers refer to the 
preceding list. 



^\■a^ III. (6 tests, 
I. Passed. 



months each.) 



Year IV. (6 tests, 2 months each.) 

1. Passed. 

2. " 

3. " 

4. ;; 

5- 

6. I'ailed. 

^\•ar \'. (6 tests, 2 months each.) 
1. Passed. 



3. Failed. 




4- 

<;. Passed. 




6. 




t'ear VI. (6 tests. 


2 months each.) 


I. l>assed. 




2. Failed. 




3. " 

4. " 

5. Passed. 




6. Failed. 




Ai. Passed. 





Year VII. (Failed.) 



'Ihis boy passed all the tests of the third year and twelve addi- 
tional tests scattered through the years IV, V, and VI, for each 
of which hi- reccivis two additional months of credit. Hence his 
mental age is three years plus 24 months or five years, and his 
intelligent (|Uolient is .34. He falls into the class of imbeciles. 

.\ dilTerent plan of evaluation has been prepared b}* Yerkes and 



THE MEASUREMENT OF MENTAL CAPACITIES 107 

Bridges ('15) in the construction of their point scale from the 
materials contained in the original Binet tests. This modification 
has proved quite satisfactory in practice. Haines has constructed 
from the same material a similar point scale which is adapted for 
use with the blind. Pintner and Paterson ('17) have assembled 
and standardized a series of fifteen performance tests chiefly of 
the form-board type which are especially valuable for use with the 
deaf and the mute. 

It must not be assumed from the apparent simplicity of the 
Binet tests that any novice can use them. On the contrary it 
requires considerable training and psychological insight to use 
them properly. Some persons by reason of lack of tact and sym- 
pathetic attitude are temperamentally unfitted for mental testing. 
It is desirable that at least a year of training in experimental psy- 
chology should precede practice work in giving the tests; and 
before the examiner may be confident of administering them 



66-65 66-75 76-85 86-95 96-105 106-115 116-125 126-135 133-145 
.33% 2.3% 8.6% 20.1% 33.9% 23.1% 9.0% 2.3% .55^ 

Fig. 35. — Distribution of intelligence quotients of 905 pupils. After Ter- 
man ('16, p. 66). 

satisfactorily he should have practice under supervision with 
thirty to forty cases. The examining room should be plainly fur- 
nished and free from disturbing interruptions or distractions. 
The confidence of the child should be obtained before the examin- 
ation is begun and his efforts should be encouraged so that he may 
react to his best advantage. Above all, the exact formula for giv- 
ing and scoring each individual test must be followed rigorously 
if the tests are to have reliable diagnostic value. 

Measixrements Obtained from the Use of the Binet-Simon Tests. 
Goddard tested some 1,500 pupils by means of the revision of the 
Binet-Simon scale prepared by himself. He found the distribution 
of abilities as shown in the following table in which are specified 



io8 



. EDUCATIONJAL I'S VCHOL( )( ; ^ 



the number of pupils one or more years ahead or behind their 
chronoloj^ical aj^e in mental maturity. The remarkable finding 
is the enormous range of the mental ages of pupils of the same 
chronological age, extending from those who are 4 or 5 years re- 
tarded to those who are 4 or 5 years accelerated. 



T.VBLE 25 



(^hronolog 
ical Age 

4 

5 

6 

7 

8 

9 

10 

II 

12 

'J'otals 



s 


4 


3 


2 


I 


At Age 


I 


2 


3 






I 


2 


2 


3 








2 


4 


8 


40 


40 


lb 


4 






I 





29 


48 


(,9 


9 







I 


2 


8 


15 


114 


50 


4 


3 




2 


2 


I 


87 


86 


18 


12 


3 








^7 


54 


56 


56 


4 


2 






15 


24 


19 


124 


27 


8 


2 




4 


13 


25 


50 


60 


12 


I 




4 


10 


13 


42 


36 


39 


7 






4 


17 


48 


131 


299 


569 


281 


57 


14 



A similar investigation was made by Terman ("16) of 905 pupils 
showing a distribution in terms of the intelligence quotient as 
given in the accompanying diagram, Figure 35. Kuhhnann ('12) 
tested some 1,300 defective individuals in the institution at Fari- 
bault, Minnesota, and found mental ages for various chronological 
ages as follows: 

TABLE 26 



Chronological 
Aci: 


NtMnFR OF 

Cases 


AVKRACK CHRON- 

uLuuicAL Aui:s 


AVKRACF. 

MtNTAL Age 


I- 5 
6-10 


7 
8S 


4.6 

8.7 


2.6 
38 


ii-iS 
16-25 
26-60 


194 

353 
367 


12.9 

20.0 

3(^5 


51 

5 5 
5-5 



Many uses of such measurements as these may be made in 
school. Thus, for example, tlu- number of retarded pupils in our 
schools is usually stated to be very large and the number of ac- 
celerated pupils is usually stated to be very small. The percentage 
of retarded pupils is placed in many schools anywhere from ^5 
to 50*^',. An intelligence examination of 483 pupils in Kansas City 



>JoRMAL 


Advanced 


45% 


6% 


49% 


30% 



THE MEASUREMENT OF MENTAL CAPACITIES 109 

(Dougherty '13), showed that the percentage of pupils as dis- 
tributed both according to intelligence and according to over- and 
under-age plans was as follows : 

Retarded 

Chronological Age 49% 

Mental Age 21% 

It appears thus that the large percentage of so-called over-age 
pupils is not due to lack of capacity but to various other causes. 
The abihties of pupils seem to be distributed approximately in a 
normal, symmetrical manner. The large majority of pupils are 
in the center or at-age with about equal numbers retarded or 
accelerated. Many of the over-age pupils in the numerous age- 
grade statistics that have recently been complied are not mentally 
arrested but entirely normal or average. 

Measurements of Special Capacities in Relation to General 
Intelligence. The second general method of measuring original 
capacities is as yet largely in the experimental stage. However, 
substantial beginnings have been made and a number of investiga- 
tions have been carried out which indicate with considerable 
assurance that it would be possible to select a series of tests of 
specific mental functions which would be correlated closely with 
general ability and could, therefore, be used as symptomatic meas- 
ures of general intelligence. An extensive investigation on this 
order was made by Simpson ('12) in which he attempted to solve 
two problems one of which was to determine what sort of tests 
would be indicative of intelligence, and the other was to apply 
them to two groups of persons widely separated in general in- 
telligence. He proceeded accordingly to apply some 13 or 14 tests 
to two groups of persons, one group of 17 individuals composed of 
professors and graduate students of Columbia University, which 
represented the upper end of the intelligence scale, and the other 
group of 20 persons whom he found at the Salvation Army Indus- 
trial Home and at the mission on the Bowery in New York City, 
which represented the lower end of the intelligence scale. These 
tests were applied from two to four times each and were then com- 
pared with estimated intelligence as ranked by the impressions of 
several persons. The correlations of these various tests with es- 
timated intelligence were as follows: 



no KDUC.VnoNAL I'SNCIK )l.()(iV 

TABLK 27. After Simpson ('12) 
Estimated intelligence and Kbbinghaus Test 89 



" " " Hard Opiwsitcs 

" " " Memory for Words. . 

" " " Kasy Opix>sitL's 

' A "Test 

" " " Memory for Passages. 

" Adding 

" " " Geometric Forms 

" " " Learning Pairs 

" " " Completing Words — 

" " " Drawing Lines — .20 

" " " L.xlcnding Lines — 

From these correlations it appears that some of the tests serve 
\ ery well as indicators of general ability, such as the various mem- 
ory tests, the Ebbinghaus test, and the opposites tests. Certain 
other tests, however, seem not at all to be SNTiiptomatic of general 
capacity since their correlations are approximately zero, such as, 
for example, the drawing of lines and geometric forms. 

A similar study was made somewhat earlier by Burt ('09), in 
which he found the following correlations of various functions 
with estimated intelligence: 

TABLE 2S. After Burt ('09) 

General Intelligcme and Attention (Dotting) .72 

" Apprehension (Pat Iirn) 75 

" " " Adai)tal)ilily (Mirror) 60 

" " " Memory (Words) 67 

" " " Discrimination (AlphalxH) 70 

(Cards) 54 

" " " Reaction (Tapping) 45 

" " " " (Dealing) 36 

" " " Perception (Pitcli) 38 

(Lines) 2s 

" Touch (Two pt.) 03 

" " " Weight Discrimination — . 16 

In an investigation made by the writer a series of eight tests 
was ajijilied to a group of 15 high school pupils. Each test was 
given very carefully to each child indivickially according to a 
uniform techni(|Ue and repealed four times on four dilTerent (Kca- 
sions. The difTiculty with a great deal of testing work has lx*cn 
that the tests have not been applied under sufficiently constant or 
rigon)Us conditions or repeated sulhciently often to yield a fairly 



THE MEASUREMENT OF MENTAL CAPACITIES ill 

accurate measurement of the particular capacity in question. At 
the close of these tests each of the 15 pupils was asked to give a 
rating of the other 14, placing the one that was estimated to have 
highest intelligence as i, and so on down to the 14th. The pupils 
knew one another well and so were able to give fair opinions. From 
these ratings a combined rank of general, estimated intelligence 
was obtained and these ranks were then correlated with the ranks 
in individual tests as well as with the combined ranks in all tests 
put together. The results are shown in the following table: 



TABLE 29 

Estimated Intelligence and Memory of Words 

" " " Memory of Passages 

" " " Opposites 

" " " Mental Addition 

" " " Arithmetical Reasoning . . . 

" All Tests 

" " " " " except Opposites . 



Thus it will appear that the tests individually, with the ex- 
ception of the opposites, as well as collectively, agree very closely 
with the combined estimates of intelligence given by the 15 pupils 
of one another. The estimates of inteUigence of pupils agreed very 
closely among themselves. The pupil who stood highest in the 
estimates of his fellows was estimated first by all pupils but one. 
The ranking of the others, of course, did not agree as closely, but the 
agreement was so close that the combined ranking yielded a rather 
reliable rating. The chief discrepancies between the estimated rank 
and the test rank occurred in the case of two of the fifteen pupils. 
The one was a boy who was estimated considerably lower by his fel- 
lows than was his rank in the tests. His estimated rank was 14 
while his test rank was 5. The other pupil was a girl whose esti- 
mated rank was considerably higher, 5, than her test rank, 11. 
Upon inquiry it was discovered that the boy was not liked well 
by his associates, while the girl was unusually popular. Their true 
mental ability was probably indicated more correctly by the tests. 
The pupil who stood first in all tests combined, and was given 
first place by all his comrades except one, completed the high school 
at fourteen and maintained an excellent record in the university. 

The promise of these and other investigations is sufficiently 
great to make it possible to develop a. series of properly selected 
tests with a definite technique to measure the general ability of 



112 EDUCATIOxNAL PSYCHOLOGY 

human beings with considerable trustworthiness. It would be 
necessary, after the selection of the tests, to develop definite norms 
for each test and for each particular year from infancy up to adult 
life. The advantage of a series of tests of this kind over the Binct 
t\pe, would be that they could be applied and evaluated with 
greater precision; that they would measure more directly certain 
fundamental mental capacities; that they would be less dependent 
upon particular environmental conditions, and that they would 
yield more objective and scientific results. The specific test method 
of measuring intelligence gives greater scientific promise and will 
in the future probably replace the Binet method to a large extent. 



\'0' 



PART II 

THE PSYCHOLOGY OF LEARNING: A. IN GENERAL 



CHAPTER VIII 

ANALYSIS OF PROBLEMS 

Analysis of the Learning Process. What are the mental and 
neural processes involved in various types of learning such as 
writing, reading, spelling, adding, solving problems, operating type- 
writers, grasping laws of nature, retaining facts, playing tennis, 
riding a bicycle, sawing to a straight line, speaking a foreign lan- 
guage, and the like? Probably all forms of learning can be reduced 
to one relatively simple, schematic type: Reception of impressions 
through the senses; assimilation, analysis, and combination of proc- 
esses in the mind ; and redirection of impulses to produce a reaction ; 
or in brief, stimulus, association, response. A child learns to avoid 
a disagreeable stimulus by receiving the sensation, associating it 
with disagreeable consequences, and reacting by avoiding the stim- 
ulus. The chick learns to avoid disagreeable caterpillars by the 
same process. Visual and gustatory stimuli are brought to the ap- 
propriate centers of the brain and there associated so that whenever 
in the future the visual stimuli of the caterpillar are brought in, the 
disagreeable taste associations will also arise which will cause an in- 
hibition of the muscular actions concerned in pecking at the cater- 
pillar. A pupil, on the first day of school, is shown certain black 
marks on a chart or in a book and is told "hat" which is to cause 
him also to say the word " hat." The psychological series of events 
is as follows: First, the visual stimulus from the page transmits im- 
pulses to the visual centers in the brain and simultaneously the audi- 
tory stimulus from the pronunciation of the word by the teacher 
transmits impulses to the auditory center in the brain; second, con- 
nections between these visual and auditory stimuli in the brain and 
arousal by the auditory stimuli of images and meanings of the object 
"hat" which have been established through previous experiences 
before coming to school ; and third, a redirection of impulses to the 
motor centers to attempt to speak the word "hat." A little later 
the pupil is given a pencil and is asked to make these same black 
marks which have the name "hat." The psychological series con- 
sists of, first, visual stimuli from the form of the letters to the visual 
centers in the brain, second, the establishment of connections in the 

115 



Il6 EDUCATIONAL rS\TIIOI.OCV 

brain between the visual centers and motor centers for the hand, 
and third, redirection of impulses from the motor centers to attempt 
to write the word "hat." Then from the muscular movements of 
the hand and arn\, made more or less by trial and error, kinesthetic 
sensations are derixed and associated in the mind with the visual 
stimuli of the word "hat." These two sets of sensations become 
associated and direct the motor resixmses in airryingout the \\Ti ting 
act. Still later the pupil is given (i) the visual or auditory stimulus, 
"If you buy a pencil for three cents and give the clerk five cents, 
how many pennies should you receive back?" which (2) arouses a 
\-ariety of association j)rocesses between various numbers such as 
multijiliaition, division, addition, or subtraction, and out of the 
mass of associations one is selected, namely, five minus three equals 
two, and this in turn (3) directs the im])ulse to the motor centers 
to say "two." All learning, even including reasoning, is jirobably 
of the same fundamental type. The only dilTerence is that there 
are more elements invoh-ed in each of the parts of the three-series 
connections and that, owing to the larger number of elements 
aroused, a selecting or picking out of certain elements rather than 
of others takes place. Learning facts of history, economics, or 
science may be described in the same general schematic manner. 
The facts are either read in a book, heard si)oken by the teacher, or 
observed directly. These sensory impressions are associated, dis- 
sociated, and combined in \arious wa\'s, which in the course of 
time usually lead to some form of reaction either of speech or of 
larger muscular activity. 

The i)receding examples of school learning de]iend for the most 
])art upon simple associations, that is, upon the law that things ex- 
])erienced together or in close succession tend to come back together. 
Thus after a certain number of repetitions the sight of certain black 
marks will set ofT promptly the reaction of speaking the word " hat." 
But in reality the jjrocess of learning is almost never as simple as 
this. While it is true that associatiim bonds must be set up be- 
tween situations and responses, a single situation is almost certain 
to present to the niind of a child of school age a multiplicity of 
aspects. As a consef|uence we find, instead of a single bond joining 
the res])onse to the situation, a num])er of bonds each joining the 
resjMmse to a dilTerent jiart of the situation. Thus the wonl "tri- 
angle" may be associated with an etiuilatend triangle of red card- 
board, Vs iiit'h thick, S inches on each side, showing a dull gray 
edge anil weighing one ounce. Innumerably dilTerent combina- 



ANALYSIS OF PROBLEMS 



IT7 



tions of bonds may accomplish this. There may be, for example, 
a major bond leading to a reaction to the redness, a secondary bond 
connecting the reaction with the size and symmetry of the sides, 
and minor bonds emanating from the thickness and color of the 
edge, while the weight and texture of the paper, the shade of red 
and, most important of all, the number of sides and the angles may 
not emerge from the complex at ail. Clearly such a set of bonds 
would be worse than useless in the presence of a right-angled triangle 
indicated in a book by black lines on a white page with the angles 
labeled by letters and with a base yi inch long. By dint of numer- 
ous experiences with a variety of triangles and with the help of the 
teacher who points out the essential three-straight-sidedness, the 
characteristics peculiar to a triangle finally emerge more or less 
clearly from the complex and become associated with the various 
reactions appropriate to "triangle," The false bonds are either 
destroyed or greatly weakened. Association is still the basis of the 
process, but there is in addition the dissociation from one another 
of the various characteristics which make up the complexes called 
objects. This is conveniently spoken of as learning by analysis and 
abstraction. When complete, the process of analysis and abstrac- 
tion, which makes possible the reaction to parts of situations rather 
than wholes, clearly is an enormous advance over simple associative 
learning. One association thus properly attached to the significant 
part of a situation may function without any further effort in a great 
variety of similar but otherwise entirely novel situations. This 
is probably the essence of reasoning. But again we must note that 
the process is rarely so simple as has been outlined. Rarely are the 
preliminary analysis and abstraction so complete that a reaction 
is transferred without delay to a very novel situation. Besides, 
the attention may be distracted from the often subtle and incon- 
spicuous but significant element in the new situation by the novel 
and striking but irrelevant features. Sometimes some of these ir- 
relevant features touch off a reaction which is entirely inappropriate. 
For example, all but the very brightest pupils in a class, which has 
learned to compute with facility the area of triangles from printed 
problems and diagrams in a book and which knows how to measure 
accurately a straight line, would be completely at a loss to know 
what to do if given a 66-foot tape measure and confronted by an 
area of ground in the shape of a scalene triangle measuring 4 by 7 
by 10 rods, covered with flower beds in a setting of greensward and 
surrounded by an ornamental iron fence three feet high. The 



Il8 r.DUCATION'AL PSVCHOLOOV 

sagacious few, if well taught, will react unerringly to the significant 
element in the situation. Such are the processes involved when a 
child has learned to isolate the significant element from the situa- 
tion: "If you buy a pencil for three cents and give the clerk five 
cents, how many pennies should you receive back?" so that it 
touches off the subtraction reaction rather than that of addition, 
multiplication, or division. 

Common and Special Elements in the Learning I^rocess. If we 
grant that the stimulu.s-assuciat ion-response series is the schematic 
jjattern of learning, it will be convenient to discuss the psychology 
of learning in two j^arts: first, the psycholog^^ of learning in general 
in which the elements common to various tjijes of learning will 
be examined; and second, the psychology of the learning of school 
subjects in particular in which the special elements and processes 
peculiar to each iy])*: of learning will be examined. 

Practically all ty])cs of learning of whate\er sort, have certain 
processes in common. They have in common certain elements of 
sensation and perception which are involved in the reception of the 
stimuli in any sort of learning. Associated bonds are formed in 
certain fundamentally similar ways no matter what the mental 
processes are between which the bonds are formed. Retention, as- 
similation, analysis, abstraction and generalization have certain uni- 
form characteristics. Likewise the redirection and reaction proc- 
esses follow certain general princijjles. But on the other hand, each 
t}pe of learning has its own si)ecial sensor}' material presented and 
perceived in its own particular manner; it has its own special bonds 
which must be formed between its peculiar elements; and it has its 
o\\-n t>'])e of reaction occurring in its own individual way. Thus 
in learning to read, the series is, first, visual-auditor}' stimuli of 
words and groups of words; second, association of \isual-auditory 
impressions and .the memory of the objects which they represent; 
and third, the resjumseof sj)eaking. In learning to spell, the series 
is, first, visual-auditt)ry stimuli of letters in a certain succession; 
second, association of the stimuli in their ])articular orders; and 
third, response in sjieaking or writing. 

The Psychology of Learning in General versus the Psychology 
of Learning of School Subjects. The i)rocedure of learning in 
general lan ])r()ljabl\- not lurni>h the process and technicjue of the 
learning of .school subjects. The pnxcss, the technique, and the 
economy of the learning ol school subjects must be worked out 
cxiKTimenlally in detail for each particular subject. All that the 



ANALYSIS OF PROBLEMS lig 

procedure, technique, and economy common to all types of learning 
can do is to furnish a broad general background. Thus the sub- 
stitution experiment in the author's Experiments ('17), Chap- 
ter X, furnishes certain conclusions with regard to the distribution 
of time according to which the establishment of the bonds in- 
volved in this particular type of learning can be made most 
economically. This experiment shows that it is more profita- 
ble to work at the task 10 minutes twice a day than 20 minutes 
once a day, or than 40 minutes once every other day. It would, 
however, be folly to attempt to generalize from such an experiment 
merely and to say that it is better to teach writing 10 minutes at a 
time twice a day than 20 minutes once a day, or than 40 minutes 
every other day. All that the experiment indicates is the general 
principle that short periods of work distributed at certain intervals 
of time are productive of greater progress in learning material of this 
sort than longer periods distributed at longer intervals. What the 
length of the most economical periods and the intervening intervals 
would be in the case of writing, reading, or any other school sub- 
ject, cannot be inferred on a priori grounds from a general prin- 
ciple, but must be determined in detail for each particular type 
of material and for the particular conditions under which the 
learning must take place. All that the general principle can do is to 
point the way to a more or less probable solution, but the particular 
direction and the course of the path must be determined from fur- 
ther observations. The factors and laws of the mind as set forth 
in general psychology can therefore not be carried over bodily 
into the psychology of a particular type of learning. General 
psychology can furnish its experunental technique and its funda- 
mental laws which will serve as guides in the development of the 
psychology of special t5^es of learning. From this point of view, 
the psychology of school subjects and the pedagogy of these sub- 
jects resulting therefrom, which is likely to be the only sound 
pedagogy worthy of the name, are as amenable to experimental 
attack according to rigorous, scientific procedure, as the problems 
in other fields of psychology have been amenable to the technique 
of experimental methods. 

Program of Problems. According to our analysis, then, the 
following problems result: 

A. The psychology of learning in general. 
I. How are the stimuli received? 



I20 EDUCATIONAL PSVCHOLOC.Y 

a. How do sensory defects interfere? 

b. What arc the factors and conditions of obser\ation 

and perception? 

2. How are they associated, analyzed and combined in the 

mind? 

a. What is the rate and progress in the formation of 

asscKJative bonds? 

b. What factors and conditions promote or hinder the 

most economical formation of the bonds? 
C. W^hat are the effects of one set of associations upon 
other sets of connections? (Transference of 
training.) 

3. How arc they redirected into responses? 
B. The psychology of school subjects in particular. 

1. What are the specific psychological processes involved 

in the learning of each particular subject? 

2. How may the capacities in each subject be measured? 

3. What factors and conditions promote or retard the 

learning in each particular subject? 



n' 



CHAPTER IX 

THE RECEPTION OF STIMULI: A. SENSORY DEFECTS 

Importance of Normal Sense Organs. The first point of con- 
tact with external stimuli is through the sense organs. It is obvious 
that this point of contact should be as perfect as possible so that 
the stimuli which are to furnish the material for learning, may be 
received as accurately and completely as possible. The eye and 
the ear are the most important avenues of information. Defective 
eyes and ears necessarily make school work difficult and disagree- 
able. A child with defective eyes or ears misses a great many 
things which the normal child can see clearly or hear distinctly. 
The sad aspect of the matter is that a great many children have 
sensory defects of which neither they themselves nor their parents 
or teachers are aware. They assume that every one else sees or 
hears just as they do and consequently their attention is not called 
to it. These defects, however, often become serious and remain as 
permanent hindrances throughout life. Many of them, if discovered 
early, may either be kept from becoming aggravating or be reduced 
very considerably. Furthermore, sensory defects, and in particular 
visual defects, have far-reaching consequences upon a child's life 
as a whole through the production of headaches,^ nervousness, and 
dislike for school work. It is even claimed that truancy is in some 
instances indirectly traceable to visual defects on account of the 
dislike for school work produced by them. Dr. J. H. Cliborne has 
given a retrospect of his own school days during which he suffered 
from visual defects, as follows: 

"I now know I have always carried about 1.50 diopters of hyper- 
metropia; in my very early days, possibly more. Books and school were 
to me a nightmare, a source of unutterable disgust. I drove myself to 
my tasks with the scourge of duty; I never took one moment's joy or 
pleasure in the acquisition of knowledge, unless it was the satisfaction 
of a task accomplished or conquest gained. I have no memory of a sense 
of pleasure connected with my studies at school or college. The only 
pleasant memories I have are those connected with outdoor sports, or 
facts gained through observation, or in the lecture-room through my 



122 EDUCATIONAL PSYCHOIX)GY 

tars; and from my boyhood I could never understand why we were 
forced to read from books all that wc learned. 

"Elarly in life 1 pondered over the easiness of the task of those who 
never siit at the feet but who followed the tracks of the peripatetic philos- 
ophers. N'erily, my school and college days woukl have been a joy to me 
had my ears and my distant vision been my means of acquiring knowl- 
edge; and yet I never had a headache in my life at school nor in after 
years until after the commencement of presbyopia. I was nervous to 
the jKjint of madness at times, and the more nervous I was the more 
diligent I l>ecame, anil the nearer I put my nose to my book. I have 
frequently observed that my right eye was crossed after prolonged study, 
or after a long written examination; this was also at times observed in 
my case by a fellow-student. That the difhculty lay in my hj'permetropia 
I have no manner of doubt. I had inherited a love of learning, I felt 
sure, and I had a right to the assurance, and my haired of close applica- 
tion was a mystery to me. I created a frown by my accommodative 
strain, which has ever been a part of me. Prolonged ap[)licalion to 
l)ooks would be followed often by sleeplessness or violence in the field at 
play. I learned for these reasons the art of complete concentration, but 
at what an expense of nervous energy." (Swift '12, pp. 94-95.) 

As concrete illustration of the manner in which undiscovered 
visual defect may interfere with school work, wc may cite the 
following case from the psycholgical laboratory of the University 
of Pennsylvania: 

"On a certain afternoon in ^farch, 1806, Miss Margaret T. AfcGuirc, a 
grade-school teacher in the Philadelphia piublic schools, went to the 
psychological laboratory of the University of Pennsylvania, accompanied 
i)y a lad of fourteen — a well mannered, intelligent lad, industrious in his 
school work; one of the favorite pupils, in fact. Yet this lad was the 
'bog>-' of the teachers who for seven years had had him in their classes: 
lie was a chronic bad speller. This does not moan that he misspelled 
some words sometimes. He misspelled every woril always, and did it in 
the siime careful and serious manner with which he recilcvl the history 
lisson he loved. His reading was as bad as his spelling; he w;xs absolutely 
incapable of getting through a single sentence correctly, </, an, and, ///<•, 
and a few three-letter words being the net result of his seven years* 
schooling. He read saw for was, water for weather; wrote hlat for that, 
soas, for soap, and other picturesque combinations of the sort in endless 
variety. His case seemed hopeless. . . . Dr. Witmer made a long 
examination, the result of which was the discovery that Charles (lilman 
had an ocular defect iie\er, in all these years, so much as suspected by 
either his parents or his teachers: at the distance of about three Ject tft€ boy 



THE RECEPTION OF STIMULI 1 23 

sa"v everything double: 'he lacked the power to direct the two eyes co- 
ordinately upon the same point in space, the left eye looking a Httle 
higher than the right.' A page of ordinary print was thus a blur; when- 
ever he attempted to write, the words doubled under his pen. Curiously 
enough, he had never mentioned this peculiarity — he seemed to think it 
the natural process of vision. And he had repeated three whole years of 
school on this account alone. . . . He was fitted with glasses and later 
operated upon; then for the first time in his Ufe the printed page and the 
words he was tracing with his pen were clear. But his reading and writing 
and spelling were just as bad as ever! The ocuhst had removed the 
defect — he had not removed the effect of the defect : that was in the boy's 
mind. And it was here that the psychologist came to the rescue by show- 
ing just what the effect was and how to remedy it. 

"Now, it is an obvious truism of daily Ufe that in order to recognize a 
thing when we see it again, we must have seen it, at least once, clearly and 
distinctly: a mental image of it must have been left in the mind. Read- 
ing is simply a rapid-fire recognition process by means of the stored 
mental images of words. Charles Oilman had no stores of images of 
words, for he had never seen any — he had seen only blurs of words. He 
was even worse off than the child just groping its way through the primer, 
for he had to unlearn the blurs he had patiently acquired through those 
seven years when nobody knew what his trouble was; then word by word, 
he had to restock his mind with the images of words shown him through 
his glasses. . . . In spite of this handicap, the boy learned to read, write, 
and speU, and was finally graduated from the grammar school only three 
years later than he should have been; which was better than not being 
graduated at all." (Carter '09.) 

T3rpes of Visual Defects. The most common forms of visual 
defects are myopia, hypermetropia, astigmatism, strabismus, and 
color-blindness. 

Myopia, or near-sightedness, is usually due to the fact that the 
eyeball is too long and consequently the rays of light entering the 
eye are focused at a point somewhere in front of the retina. As a 
result the rays of light are again spread out when they reach the 
retina and therefore do not form a distinct image. 

Hypermetropia, or far-sightedness, is usually due to the fact 
that the eyeball is too short and so the rays of light entering the 
eye are not sufficiently refracted in order to form a clear image 
when they impinge upon the retina. The image would be formed 
at a point back of the retina if the rays of hght were extended. 
In some cases myopia and hypermetropia may be due to improper 
refraction of light by the crystalline lense or the cornea. These 
abnormal conditions cause a considerable strain upon the muscles 



124 EDUCATIONAL PSYCHOLOGY 

of the eye which attempt to accommodate the lens in order to 
form as clear an imaj^e as possible. 

Astigmatism is due to the fact that the curvature of the cornea 
or of the lens, usually of the former, is not the same in all meridians. 
The result is that if the lens is accommodated for rays of light from 
some parts of the field of vision, it will not be accommodated for 
rays of light from other parts of the field of vision, and consequently 
a {)orlion of the field of vision will be distinct and another portion 
will be blurred. The curvature of the cornea or of the lens in the 
case of astigmatism may be compared to that of an eggshell. It 
is difTerent in dilTerent directions whereas in the normal eye it is 
similar in all directions like that of a sphere. 

Strabismus refers to the lack of perfect balance in the externa! 
muscles of the eyes, so that the two eyes do not focus upon the same 
point at the same time. This difTerence in point of focus may, of 
course, vary anywhere from perfect coincidence, as it should be in 
the normal eye, to a very large deviation, commonly known as cross- 
eyedness, which can be observed readily by looking at a person's 
eyes. The history of Charles Oilman cited on preceding pages was 
a case of strabismus. 

Color-blindness consists in the confusion of certain colors, nearly 
always red and green. Confusion of the other two fundamental 
colors, yellow and blue, almost never occurs. The cause of it is 
more or less speculative and may be due to the absence, or the im- 
proper functioning, of the color elements in the retina. Color- 
blindness is obviously a drawback in any tjpc of school work in 
which colors are concerned, such as drawing, map work, domestic 
science and manual arts and all scientific studies in which color 
discrimination is involved. 

Causes of Visual Defects. Visual defects are due in general 
to two causes, ii) heredity, and (2) the strained use of the eyes for 
fine distinctions at close range, particularly under poor illumination. 
Color-blindness and strabismus are probably inherited in nearly 
ever)' instance. The other t^i^cs of defects which relate to the 
formation of the image are proably due in part to hereditary- con- 
flitions in the sense organs and in part to overstrained use of the 
eyes. Reading, which has been the great promoter of civilization, 
has also been, in a certain sense, a deteriorator of the eyes through 
the strain put upon them by the fine distinctions that must be 
made ;it close rajige and at a tremendously ra])id rate. Durr fas 
reported by WhippU- Cio, p. 139,)] has attempted to exi)lain the 



THE RECEPTION OF STIMULI 125 

great prevalence of myopia in Germany as compared with other 
countries by the excessive demands made by the German school 
system. He has estimated the number of hours devoted to study 
and to exercise by the typical boy during the years 10 to 19 in 
different countries as follows: 

Hours study Hours exercise 

Germany 20,000 650 

France 19,000 ^,Soo 

England 16,500 4,5°° 

Most hygienists maintain that myopia is an acquired condition 
whereas anatomists are more inclined to regard it as an inherited 
condition. Cohn (Whipple '10, p. 139) reports that myopia oc- 
curred in gymnasia in the following increasing percentages during 
the six years of study: 12.5, 18.2, 23.7, 31.0. 41.3, 55.8. 

Frequency of Visual Defects. Recent years have brought a 
considerable number of investigations as to the percentage of 
children with defective vision. Whipple made an examination of 
the vision of 1,000 white and 100 colored children in Jefferson City, 
Missouri, in which he found the following percentages of visual 
defects: 

TABLE 30. After Whipple 

Visual defects among i 000 white and too colored children in Jefferson City, 

Missouri 

White Colored 

Defective vision (Snellen test) 36.5% 19% 

" " (one eye) 13 . 

" " (both eyes) 22 . 

" " (first 3 grades, 147 pupils) 29. 

" " (high school, 116 pujjils) 40. 

Pain after using eyes in study 29 . 

Probably needing glasses 41.0 

Wearing glasses when examined 3.8 S 

Cross-eyed 3.0 2 

Taussig ('09) has summarized the percentages of visual defects 
in various cities in different countries as follows: 

^ This percentage is probably too high since it was discovered that the colored chil- 
dren took peculiar pride in reporting headaches because they seemed to consider it a 
sign of intellectual keenness. 



8 17 

7 12 

4 

5 

S 34 



1.^6 p:ducatio.\\i, i'sv( ii(H-()(;\ 

TABLE ji. After Taussig Coq) 

I IcitlclhcrR, Germany (1870) 3SO% 

]'.d[n\mT)ih. Scotland (1904) 43-* 

Dunfermline, Scotland (1007) 170 

Cleveland, well-ttvdo district (1907) 32.4 

conjjestcd district (1907) 71.7 

Massachusetts, exccj)! Boston (1907) 19.9 

Boston and environment (1907) 30. 7 

Boston (1908) 23.0 

New York City (190/)) 3^-3 

New York City, Borough of Manhattan 10. 2 

Chicago (1909) •■ 19-4 

JetTcrson City, Mo., either eye (1908) 36.5 

" lx)th eyes (1908) 22.7 

St. Louis County, Mo., cither eye less than 20/20 30.6 

* " " " " both eyes less than 20/30 (1909) 14.3 

" " " " both eyes less than 20/40 (1909) 2.8 

.Xddilionul results from other cities as reported by Gulick and 
A\Tes ('oS, J). 8j) are as follows: 

TABLE 3 J 

Bayonnc, N. J 7-7% 

Camden, N. J. (1906) 27.7 

Milwaukee (1907) i4-7 

Minneajwlis (190S) .13.9 

Pawtucket, R. L (1901) 11 

Utica, N. Y. (1897) 10 

Worcester, Mass 19 

The larRC wiriations in the percenta;,'cs riuoted for different 
cities probal)ly do not re])rescnt actual dilTercnces in the prevalence 
of visual defects. They are probably due mainly to difTcrenccs in 
standards adopted l^y various examiners. Whether an eye is re- 
])()rted as defective or not is, in the milder forms of defect, an 
arbitrary matter. From the purely mechanical stand])oint, an 
absolutely ])erfect eye is undoubtedly \ery rare. Therefore the 
matter resolves itself largely into the question as to whether the 
deviation from perfection is sufTicicnt to interfere apprecial)Iy 
with nomKil distinct vision. As a general statement we may Siiy, 
according to the quoted tables, that approximately 25^^ to 33% 
of the school children have vi.sual defi-cts sulViciently serious to 
demand some attention. Color-blindness fortunately is relatively 
rare. It is estimated that 4% or 5% of men and less than 1% of 
wunien are color-blind. 



THE RECEPTION OF STIMULI 127 

Visual defects seem to increase measurably with successive 
years in school. Thus in Whipple's table the percentage of de- 
fect among high school pupils is 11% higher than that among pupils 
in the first three grades. An extensive comparison was made by 
Gulick and Ayres ('oS) in New York City which showed the follow- 
ing percentages grade by grade: 

TABLE 33 
Grade Per Cent of Visual Defect 



3 21 

4 25 

5 24 

6 24 

7 26 

8 32 



Remedial Measures for Avoiding Visual Defects. The first 
and probably most important suggestion is the examination of 
the eyes of pupils at least once in two years or preferably once a 
year. This would serve as a means of discovering the visual de- 
fects so that measures could then be adopted for the proper care 
of those pupils suffering from them. Excellent results have been 
shown in various cities in which general sensory examinations 
have been introduced. Taussig reports that in Boston the per- 
centage of visual defects dropped from 30.7% to 23% and in 
New York City from 31.3% to 10.2% as a result of the introduc- 
tion of visual examinations. The schools require compulsory 
attendance, but they have not taken sufl&cient steps to make it 
possible for the children to remain in school to their greatest profit. 

In the next place, the proper lighting of schoolrooms is highly 
important. This matter is being taken care of, however, at the 
present time by school architects in a much more thoroughgoing 
manner than was formerly the case. Many of the older buildings 
are poorly arranged and wretchedly lighted. The amount of win- 
dow area to floor area usually recommended as satisfactory, is 
approximately one to five or six. As typical of the inadequacy 
of the lighting in older school buildings we may note that in a 
test of the amount of light at different desks in one of the old school 
buildings in Madison, made by Cohn's light tester, showed that in a 
room of five rows of seats the two rows next to the inside wall 
showed a decided insufl&ciency in illumination. 



128 I.IjLC Al lONAL l'S\ ( IK )L(K;V 

Numerous other ])recautions may readily Ijc exercised by the 
school to avoid the aggravation of existing visual defects. Such 
measures are the breaking up of the school program to spread the 
severe use of the eyes, reduction of close \vork in the early years of 
a child's life, prohibition of work in i)Oor artificial light, adjust- 
ment of the size of the desk to the size of the pupil, the instilling 
of the habit of resting the eyes even for the short interval of a few 
minutes in the midst of eye-straining work. The proi)cr printing 
of books and the use of apiirojiriate pa])er is being looked after 
l)y publishers much more carefully to-day than formerly. 

Frequency of Auditory Defects. The presence of a greater 
or less amount of deafness in one or both ears often interferes very 
considerably with normal school work. \Vhi])ple has reported for 
the 1,000 white and 100 colored children in Jefferson City, Miss<niri, 
auditor}' defects as follows: 

White Colored 

l)cfective hearing (whisper test) 7 7% 70% 

" " (one ear) 6.4 4.0 

" " (both ears) 11.3 17. 

Taussig Cog) has reported the following percentages of de- 
fective hearing in various cities: 



f%. 



TAIiLK 34 

IldinburKli, Scotland (1Q04) 12 

l>unfcrmline, Scotland (1907) 4.0 

Cleseland, well-to-do district (1907) 5.2 

" congested district (1007) 1.8 

Massachusetts, excejjt Boston (1907) 5.8 

Uoston and environment (1907) 7.7 

Boston (1908) 7.6 

iNew ^'ork City (1906) 2.0 

New York City, Borough of Manhattan i .0 

Chicago (1909) 2.7 

St. Ix)uis County, Mo., either ear defective (1909) 7.3 

" " " both cars seriously defective (1909) 2.2 

Additional results reportid by Gulick and .Xyres ('cS) arc as 
follows: 

TABLE 35 

Bayonnc, N. J 2 . 5% 

Camden, N. J. (1906) 4.1 

Minnea|>olis (1908) . 7.7 

Pawtuckct (1901) 4.3 

Utica, N. Y. (1897) 6.6 

Worcester, Mass 6.6 



THE RECEPTION OF STIMULI 



129 



As a general statement we may say that approximately 5% to 
10% of pupils suffer from defective hearing of a sufBciently serious 
character to interfere with their school work and to require medical 
attention. 

Effects of Sensory Defects upon School Work. A number of 
inquiries have been made to determine the amount of hindrance 
which sensory defects have upon the proper performance of school 
work. Dr. Cornell has reported the following results for three 
schools in Philadelphia showing the average school marks of normal 
and defective pupils: 

TABLE 36. After Cornell 

Allison School— 219 children, both sexes, 6 to 12 years old 

Average 

Normal child 75 

Average child 74 

General defectives 72.6 

Adenoids and enlarged tonsils 72 

Deaf 67.2 

Ninth Street Primary School — 84 children, both sexes, 6-10 yrs. old. 

Language Arithmetic Spelling Average 

64 cases normal children 72.9 75.5 75-4 74-6 

84 cases average children 70.5 74 72.8 72.4 

21 cases general defectives 63 .3 70 64.8 66 

8 cases adenoids 60 . o 66 . 7 65 63 . 9 

No cases deaf. 

Claghorn School — 252 children, both sexes, 12 to 15 years old. 

Geography 
and 
Language Arithmetic History Average 

1 79 cases normal children 74.4 72 76.6 74.3 

252 cases average children 72. 7 70 76.5 73. i 

73 cases general defectives 71.4 65 . i 76 . 2 70 . 8 

Whipple in his study in Jefferson City found that among pupils 
of good vision, 26% did unsatisfactory work and among pupils 
with defective vision 38% did unsatisfactory work. Smedley 
reported that in Chicago there were 18% with auditory defects 
among pupils above grade, and 25% among pupils below grade. 
It is difficult to say how much of the scholastic inferiority of those 
having sensory defects is due to these defects and how much is 



I ^o EDUCATIONAL rSYCIIOLOGY 

due to inferior native ability. It has been showTi by Pintncr and 
I'atLTst)n CK)) that deaf children are fully thrcr years behind 
normal children in learninK the di^it-synibol test which, as j^iven 
by them, does not depend upon the use of language. Goddard ('14) 
in an intensive investigation found blindness perceptibly asso- 
ciated with feeble-mindcdness, and deafness with neuropathic 
taint. 

Musical Discrimination, Besides the \arious degrees of deaf- 
ness, the sense of hearing is of direct interest to the school from 
the standpoint of musical instruction. The ability to sing, and 
to some extent, to appreciate music, depends in part upon the 
accuracy of the discrimination of pitch. This al>ility varies widely 
among people, and considerable inaccuracy in musical discrimina- 
tion is a distinct difficulty in learning music. The results of recent 
experiments seem to indicate rather defmitely that accuracy in 
j)itch discrimination can probably not be improved by practice 
or by teaching. The voice in singing and the hand in playing the 
violin, are guided by the accuracy of the ear. If the ear is not 
accurate the individual is unable to guide his voice or his fingers 
with a ])recision necessary for the production of music. 

Seashore ('01), who has studied this ])roblem airefully, has 
suggested that ])upils whose discrimination is two vibrations or 
less ha\'e a sufficiently fine ear to become musicians; pupils who.se 
discrimination lies between 3 and 8 vibrations, which includes the 
large majority of jx'ople, have a sufficiently accurate discrimination 
for ordinary' musical instruction and enjoyment; pupils whose 
discrimination lies between 9 and 17 vibrations should have musi- 
ad instruction only if they have special inclination or desire for it 
and singing in school should be optional for them; and finally, 
pu()ils whose discrimination lies at iS or above, should prob- 
ably not lie required to study music or to attempt to produce 
music. 

Physical Defects. Beside the sensory defects there are several 
t\-])(s of physical defects whose frequency is high and whose inter- 
ference with school work is serious. The |)revalence of these de- 
fects is indicated in the following table, adapted from Gulick and 
Ayres ('13, p. 38): 



THE RECEPTION OF STIMULI 



131 



TABLE 37 



Percentages of children having defects of 
Teeth Throat Nose Glands Others 



Boston (191 2) o 

Chicago (1910) 36 

Cleveland (1910-11) 32 

Newark, N. J. (1910-11) 29 

New York (191 1) 59 

Oakland, Cal. (1910-11) 48 

Pasadena, Cal. (1909-10) 30 

Rochester, N. Y. (1910) 44 

St. Louis (1910-11) 52 



8 


22 


5 


20 


4 


15 


3 


18 





15 


I 


35 


5 


5 


8 


29 





17 



2.6 

3 
9 
o 
8 
9 
4 
7 



8.2 
8.1 
II. 6 
12.3 
II. 9 
18. 1 



12.4 



8.S 



17.7 
8.9 
7-5 

13 -7 
4.6 

30 

6.0 

17.0 

1.2 



Average 41.5 



10.8 



II. 8 



CHAPTER X 

Tin: RECEPTION OF STIMULI: B. PERCEPTIOX AND OBSERVA- 
TION OF THE SENSORY MATERIAL 

Problems. Next to the normal operation of the sense organs 
conic the actual perception and observation of stimuli as pre- 
sented to the sense organs. What material you learn and how 
you leani it <ki)en(ls ui)on what and how you perceive or observe. 
Obviously what you perceive or obser\e depends upon the nomial 
operation of the sense organs. That has long been recognized. 
But it has not been so fully recognized that it depends also ui)on 
the mental a])prehension of the sensory stimuli. Perception and 
observation do not de])end alone on what is presented to the sense 
organ, but also upon how the stimuli are taken into the mind. 
The specific problems in\olved in it are: 

(i) How accurate is the observation of stimuli? 

(2) How large is the range of stimuli observed at a given time? 

(3) How may the accuracy and range of obser\-ation be im- 
l)roved? 

(4) How are the stimuli inteq^reted? 

Accuracy of Observation. Much of the difficulty in learning 
a given material is (hic to inaccuracy and error in the observation 
of the material. The word to be sjK'Ued, the letter to be ^\Titten, 
the plant to be described, the e.xjieriment to be reported, the map 
to be drawn, may all be done and learned incorrectly in part be- 
cause they are juTceived and observed inaccurately and in- 
completely. 

Recent e.xqieriments on the "fidelity of report" have called at- 
tention to the ])revalence and nature of inaccuracies in observation. 
'Jhe observations are not made with sufficient care and attention 
to impress a faithful image of the object upon the mind. Inac- 
curacy in perception and infidelity in rejK)rt have hardly been 
known to e.xist and therefore have not been sufTiciently guarded 
against because, under ordinary conditions of learning in school 
as well as under ordinary' conditions in life, there is seldom an op- 

132 



THE RECEPTION OF STIMULI 133 

portunity for comparing directly the observations as received in 
the mind with the original stimuH as actually presented to the 
sense organs. 

The recent investigations on the fidelity of report have been 
carried out by presenting to a group of observers a series of events 
either in the form of real actions or more commonly by means of a 
picture. The picture is exposed to the observers for a short period, 
say thirty seconds or a minute, after which they are asked to write 
a report of their observations. This is usually supplemented by 
an interrogatory report consisting of answers to questions regarding 
the picture. Experimental inquiry into these matters has been 
stimulated primarily from the practical importance of determining 
the reliability of witnesses in court. Incidentally, the results have 
an exceedingly significant bearing upon the accuracy of observation 
involved in learning. The main results of these experiments have 
been summarized by Whipple ('10, pp. 304 ff.) as follows: 

"The chief single result of the Aussage psychology is that an errorless 
report is not the rule, but the exception, even when the report is made by 
a competent S (subject) under favorable conditions. Thus, in 240 reports, 
Miss Borst found only 2% errorless narratives and 0.5% errorless deposi- 
tions. The average subject, when no suggestive questions are employed, 
exhibits a coefficient of accuracy of approximately 75%. 

"Generally speaking, attestation does not guarantee accuracy: on the 
contrary, though the numbers of errors is nearly twice as great in un- 
sworn as in sworn testimony (according to Stern, 1.82 times, according to 
Borst, 1.89 times as great) there stiU remains as high as 10% error in 
sworn testimony. 

"Reports of children are in every way inferior to those of adults: the 
range is small, the inaccuracy large, and, since the assurance is high, the 
warranted assurance and reliability of assurance are both very low. 
During the ages 7 to 18 years, the range, especially the range of knowl- 
edge, increases as much as 50%, but the accuracy, save in the deposition, 
does not increase as rapidly (20%). This development of capacity to 
report is not continuous, but is characterized by rapid modification at the 
age of puberty. » 

"The one factor that more than others is responsible for the poor 
reports of children is their excessive suggestibility, especially in the 
years before puberty. 

"Not all the features of the original experience are reported with the 
same frequency or with the same accuracy: there is, rather, a process of 
selection, both in the process of observation, and also, probably, in 
memory and in the formulation of the report. In general, we may say 
that persons and their acts, objects, things, and spatial relations are re- 



134 EDUCATIOXAL PSYCHOLOGY 

ported with considerable accuracy (85-90^, ), whereas secondary features, 
especially quantities and colors, arc reported with considerable inaccu- 
racy (reports on color have an error «)f from 40-50' t). 

".Vll authorities agree that the use of interrogatory, whether of the 
complete or incom[)lete form, increases the range and decreases the 
accuracy of the report. 

"The introduction of leading or suggestive questions very noticeably 
decreases the accuracy of re[K)rt for children, an<l, unless the conditions 
of report are quite favorable, even for adults. The greater suggestibility 
of children is shown by Stern's results in which the inaccuracy of bo\s 
and girls, ages 7 to 14 years, was from 32 to 39%, as against 10% in- 
accuracy for young men aged 16 to 19 years." 

The reasons for inaccuracies in reports of obser\'ation must be 
sought in various directions. The chief ones are (i) insufficient 
attention to the material observed so that only a vague impression 
is produced which may easily be disturbed and modified by other 
stimuli, (2) meagemess of ideas and ex})eriences with which the 
obscr\ed material may be connected and inleq^reted, (3) low 
retentiveness of the imi)ressions so that other impressions can 
readily distort them, (4) faint imagery in terms of which to i)icture 
the objects, (5) lack of conscientiousness in keeping apart the 
obser\'ed and the inferred items as a result of which missing or 
doubtful elements are sup])lietl by unconscious inference, (O) the 
effect of suggestion through which the mind is prone to seize upon 
any slight hint and to lit it into the story. 

The Range of Observation. The amount of material observed 
by dilTerentr jiersons within a given limit of time varies over an 
astonishingly wide range. Individual dilTerences in the capacity 
for a])prehending stimuli from the outside world are probably as 
large as those in other mental abilities. One person may report 
several limes as much of a scene or .series of events as another. 
The pu[)il with a wide scope of apprehension and obser\'ation has 
a tremendous advantage over one with a narrow scope of observa- 
tion. This range probably depends fundamentally upon the span 
of attention, quickness of assimilation of items, retentiveness, and 
previous knowledge about the fads to be observ'cd. The span of 
attention as measured by rapid exposure methods, ranges in nonnal 
jKTsons from three or four objects to eight or nine. Such experi- 
ments are carried out by exposing momentarily before the observer 
cards with varying numbers of objects and by recording the num- 
ber of objecl> noticed. 



THE RECEPTION OF STIMULI 135 

Some years ago a number of interesting investigations were 
made by G. Stanley Hall ('83) and others to determine the range of 
observation and knowledge of pupils entering school. The returns 
showed a surprisingly narrow range of information and the great 
extent to which it was determined by the environment in which 
the child lived. The educational value of such inquiries has been 
to emphasize the importance of a wide variety of immediate con- 
tact and experience with real objects by observing, manipulating 
and using them. 

As a concrete illustration, the author made the following simple 
experiment to indicate the range and fidelity of such observations 
as would be made in a class in biology. Some plants in a jardiniere 
were exhibited before a class of thirty-nine students for thirty sec- 
onds with instructions that they were to observe them as carefully as 
possible and that they would be asked inmiediately afterwards to 
record their observations. The shortest report (A) was only fifty- 
seven words in length, and one of the longest and best (B) was 131 
words in length. One of the most erroneous ones is given under (C). 
B is more than twice as complete as A, and fully as free from error. 
C is quite typical with regard to kind and frequency of misstate- 
ments. The erroneous portions are italicized and the corrections 
are given in parentheses. All three persons had had an elementary 
course in botany. 

Report A: 

"There was a high upright geranium plant having no buds but broad 
leaves. This was surrounded by low plants with drooping stems and 
bearing many pointed, small leaves. The leaves had a pinkish center, 
surrounded by a pale green hand, the contour very irregular and the 
general effect bushy. There were no buds on these plants either." 

Report B : 

"From a brown jardiniere arose one stalk of a geranium bearing nine 
big green leaves. The leaves spread out in all directions and are round 
in shape with large scallops. Lower was another plant with much smaller 
leaves and more bushy. It had three large branches, one leaning over 
the pot on each side and one across the front. The leaves of this plant 
were more oblong in shape, rounding at the base and reaching a point. 
The color was a pale pink in the center and back to the base, shading to 
a deep red or wine (pink) color towards the tip end and the whole edge of 
the leaf was green. The stem was much more delicate than the stem 
of the geranium and the leaves were much more numerous." 



136 KDUCATIONAL PSVCIKJLOGV 

Rcix)rl f: 

"The plant was in a brown bowl. There were varieties of plants. The 
one had one large stalk, wilh Jive (nine) branches growing out of it. The 
leaves were large and heart-shaped (rounding). The other plant w:is lower 
and drooped, and had more leaves. The leaves were oblong (heart- 
shaped) and smaller than the leaves of the first plant. They had a center 
of very light green (j)ink) and were outlined by darker green. The leaves 
were smooth and glossy (velvety). The leaves of the tall plant were 
notched." 

It is obvious thai the reporter of B has a great advantage over 
either A or C both in the quantity and in the cjuaiity of his obser\'a- 
tions. In a given period of time B will learn much more material 
and assimilate it in more correct form than either A or C. The 
objection might be raised that a report formulated after the obser- 
vations have been made is bound to be erroneous and that it would 
be fairer to have all three persons stand before the plant and re- 
cord their observations at the time. To this, however, we may 
reply that the range of facts observed and recorded in a given 
period of time would he just as wide and that observed items are 
not really assimilated mentally until they can be adequately 
thought out and expressed. The simple exj)eriment here reported 
probably represents quite fairly the sort of things that occur in 
ordinary observation of material in learning. 

Improvement in Observation. Granting the importance of 
a wide range and a higli degree of accuracy in the perception of the 
material to be learned, what may we do to increase the accuracy, 
scope, and fidelity of observation? Probably the only advice to 
give at the present time is the rather obvious suggestion: Insi.st 
on greater accuracy and attention in observation. This may Imj 
accomplished by defmile discovery of errors and inaccuracies in 
the ol)servalions themselves. E.xix'rimenls have shown that many 
people, especially cliiitlren, do not realize their inaccuracies and 
that calling attention to the discrepancies between objects and 
mental impressions of them results in a material reduction in 
unreliability. Whipple ('10, pp. 309 IT.) has summari/.ed the experi- 
mental results on the possibility of improvement in observation 
by repeated tests witli the same groups of j>ersons as follo>vs: 

"Simple practice in reporting even without special training or conscious 
cfTort to improve, facilitates and betters the reiwrt, as is shown in Ta- 
bic 47, from Miss Horsl. It will be noted that the tendency to oath and 



2 ■ 


3 


4 


5 


39 o 


42.3 


40.3 


42.0 


87.7 


92 


9 


88 


2 


90.0 


96.4 


97 


8 


97 


9 


98.6 


87 





91 





88 





89.0 


89 


4 


92 


6 


89 


8 


90-3 


98 





98 


4 


98 


6 


99.2 


59 


8 


62 


8 


61 


9 


72.1 


53 


2 


58 


5 


57 


5 


66.5 


6 


6 


4 


3 


4 


4 


5-6 


88 


8 


92 


5 


93 





91.7 



THE RECEPTION OF STIMULI ^ I37 

warranted tendency to oath are both particularly improved by practice, 
and that there is also an appreciable improvement in range, accuracy, 
warranted assurance, and reliabiHty of assurance, whereas assurance and 
accuracy of assurance are scarcely affected. Similar practice-effects 
may be discerned in their deposition. From these results, it is clear that 
the several coeflicicnts of report may vary more or less independently." 

Effect of Practice upon Coefficients of Report (Narrative) 

(Borst) 

Number of Report (Test) i 

Range 39° 

Accuracy 86 . 6 

Assurance 96 . 6 

Warranted assurance 84 . o 

Reliability of assurance 87.5 

Accuracy of assurance 970 

Tendency to oath 43 • o 

Warranted tendency to oath 40 . 2 

Unwarranted tendency to oath .... 2.8 

Reliability of oath 93 . o 

NoTi'; : The effect of practice in these tests is somewhat obscured by the fact that the 
first and third tests were made after a 3-day, the others after a g-day interval. 

"The capacity of children to observe and report in a detailed and 
accurate manner may be improved by systematic training. This educa- 
tion may be best secured by appeal to zeal, interest, enthusiasm, or desire 
for improvement on the part of the child; more formal training of an 
intellectual type, e. g., suggestions for systematic observation, specific 
training in sense-perception, instruction designed to augment appropriate 
apperceptive-masses, etc., is much less effective. 

"The inadequacy of the child's report is due, not so much to poor 
memory, as to the fact that he fails to perceive many features in the 
original experience, that he fails to put into words even what he does 
perceive, and especially to the fact that he is absurdly uncritical (his 
assurance, indeed, commonly reaches 100%)." 

Tests such as these, but made with the material of school studies, 
would probably be very useful in bringing about more concen- 
trated attention upon, and greater reliability in, observation. 
Thus a plant or a flower in a course in biology, might be exposed 
for a definite period of time to a class of pupils who would then be 
asked to write as accurate a description of the object as possible. 
This description could then be definitely compared, point by point, 
with the original object and in this manner the errors and inaccu- 
racies would be discovered and noted. Difficulty in acquiring the 



138 EDUCATIOX.\L PSYCHOLOGY 

material in various school subjects is no doubt traceable in larger 
part than we realize to the incompleteness and the unreliability 
of the perception of the material or stimuli to be accjuired. It 
would be an experiment worth making to determine to what extent 
the difficulty of a pupil in learning to spell is due to actual incom- 
pleteness of the observation and perception of the letters in the 
words. 

Interpretation of Stimuli. To a large extent, observation is 
interpreiaiiun. The same identical sense imjjressions are inter- 
preted very difTerently by diflerent observers. This may be dem- 
onstrated perhaps in extreme form in such tests as the one with 
the ink blots outlined in the author's Experiments, Chapter XIII. 
The first ink blot in that series signihed to eight persons the follow- 
ing different things: map, bear, trees, lake, cloud, child, bat, man 
running. The same mental processes occur in a less variable man- 
ner in all kinds of observation. Incoming stimuli are interpreted 
by the association processes aroused in the mind. On this Ixisis 
the traditional doctrine of apperception has been formulated and 
from it have been derived such pedagogical corollaries as, "Link 
the new to the old," or "Proceed from the known to the unknown." 
The theory of apperception has been very clearly expressed by 
James in the following statement: 

"The gist of the matter is this: Every impression that comes in from 
without, be it a sentence which we hear, an object of vision, or an efflu- 
vium which assails our noses, no sooner enters our consciousness than it is 
drafted off in some determinate direction or other, making connections 
with the other materials already there, and finally protlucing what we 
call our reaction. The particular connections it strikes into arc deter- 
mined l)y our past experiences and the 'associations' of the present sort 
of impression with them. If, for instance, you hear me call out .\, B, C, 
it is ten to one that you will react on the impression by inwardly or out- 
wardly articulating I), E, F. The impression arouses its old associates: 
they go out to meet it; it is received by them, recognized by the mind as 
'the beginning of the alphabet.' It is the fate of ever>' impression thus 
to fall into a mind preoccupied with memories, ideas, and interests, and 
by these it is taken in. Educated as wc already are, we never get an 
experience that remains for us completely nondescript : it always reminds 
of something in ciuality, or of some context that might have surrounded 
it before and which it now in some ways suggests. This mental estort 
which the mind supplies is drawn, of course, from the mind's ready-made 
stock. We conceive the impression in some definite way. We dis|)ose 
of it according to our aL(|uired iHjssibilities, be they few or many, in 



THE RECEPTION OF STIMULI 139 

the way of 'ideas.' This way of taking in the objects is the process of 
apperception. The conceptions which meet and assimilate it are called 
by Herbart the ' apperceiving mass.' The apperceivcd impression is 
engulfed in this, and the result is a new field of consciousness, of which 
one part (and often a very small part) comes from the outer world, and 
another part (sometimes by far the largest) comes from the previous con- 
tents of the mind." ('99, p. 157.) 

The doctrine of apperception as here stated by James is simply 
a statement of the psychology of perception as usually accepted. 
The importance of die applications of the doctrine to teaching has 
perhaps been overemphasized in the educational writings of the 
recent past and in the pedagogical methodology that has been 
worked out in accordance with its corollaries. The applications 
of the theory as represented by the injunction, "Link the new to 
the old," is no doubt sound from the psychological side and useful 
from the pedagogical side as a general guiding principle. Illustra- 
tions of the principle would be the teaching of a topic in geography 
by connecting it up with the known facts of geography in the 
immediate environment of the child, or of teaching laws of chem- 
istry by relating them to familiar facts and problems that have 
arisen within the child's own experience, or of teaching the spelling 
of a new word by pointing out its similarities to words already 
known, or of teaching forms of a foreign language by referring 
them to related forms in the language previously acquired. Such 
a procedure is unquestionably valuable whenever it can be em- 
ployed. However, some of the enthusiastic advocates of the 
doctrine have been somewhat blinded to the limitations of it. If 
we regard learning as a process of establishing connections between 
elements of the learning-material, we can conceive of three possible 
kinds of bonds to be formed: (i) between two known elements 
which had previously not been connected, (2) between a known 
and an unknown element, (3) between two unknown elements. 
It is obvious that a great deal of learning consists in the formation 
of the third type of connections. The doctrine of apperception 
can apply only to the first and second type of connections and 
these can very probably be formed more readily according to the 
natural workings of apperception because some of the elements 
had previously been acquired. 

Much of the discussion in favor of the doctrine of apperception is 
really based upon the greater practical value of the known and 
nearer at home, and upon the urgent need of knowing something 



140 '**• J,1)LLA11(J.\AL I'hVLllULUC.V 

about the immediate environment rather than about a distant 
time or territory, whose history or geography may Ik- of httle value 
to tlie child, than uiK)n greater ease in the formation of bonds be- 
tween a known and an unknown element. The main thing in educa- 
tion is not to proceed from the known to the unknown, but rather 
to accjuire the unknown. If this can be done by linking the new 
to the old, well and good, but the chief object is the linking of the 
new. Much of the so-called proceeding from the known to the un- 
known or of the linking of the new to the old, is more or less fruit- 
less, since it neither i)rocceds to the unknown nor links anything 
new. It usually consists of a reawakening of the kno\ni and of the 
old. Learning is fundamentally the acquisition of new sets of 
stimuH-association-response series. When a jiupil first attempts to 
write, he must acquire new neural connections in securing control 
of his hands. When he begins to learn the meaning of printed 
symbols, he is confronted with new stimuli among which new con- 
nections must be established. Much of the so-called teaching ac- 
cording to the theory of apperception consists in setting up prob- 
lems concerning things with which the child is already familiar 
and thus in arousing in him a desire to learn something new. This 
is, no doubt, good teaching, but the important {xirt in learning is the 
new element to be sought and the new associations to be built up. 



CHAPTER XI 

THE RATE AND PROGRESS OF LEARNING 

Problems. The chief problems to be considered are as fol- 
lows: 

(i) How rapidly are new associative bonds formed? 

(2) Does the rate of acquiring new connections and new materials 
continue uniformly per unit of time or per repetition? 

(3) Do variations in rate occur in a uniform manner? 

(4) What causes will bring about variations in rate? 

(5) Does the rapidity of learning occur in a similar manner in 
all types of learning? 

Such problems as these may profitably be raised with regard to 
any sort of learning. If we consider the learning of a language we 
may ask, How rapid is the progress in acquiring the meanings of 
the words, knowledge and use of grammatical forms and ability to 
translate? Is progress uniform or are there times of rapid advance 
occurring in alternation with periods of little or no gain? What 
conditions will promote the learning of the language most ef- 
fectively? If we could answer these and similar questions con- 
cerning any type of learning we would be able to control its 
progress far more economically than we are able to at the pres- 
ent time. 

The Curve of Learning. The rate and progress of learning 
may be expressed in terms of the amount done per unit of time, or 
in terms of time required per unit of work. The relation between 
these two variables is represented by the curve of learning in 
which one function, usually time, is represented along the base 
line, and the other, usually amount accomplished, is represented 
along the vertical axis. Figure 36 represents a typical curve of 
learning in which progress is measured by the amount achieved per 
five minutes of time. It represents the rate of forming associations 
between numbers and letters in transcribing letters into numbers 
as specified in the author's Experiments ('17), Chapter X. 

Characteristics of Learning Curves. Most of the experimental 
work on the course of learning curves has been done chiefly with 
various kinds of skills such as telegraphy, tj^ewriting, tossing 

141 



142 



EDUCATIONAL PSYCHOLOGY 



halls, mirror tracing, substitutions, and tin- like. Little has been 
(lone on the progress of analytical tNT^cs of learning, on the advance 
in the acquisition of facts of a science or of the history of a country, 





































































J 


J 


N 




















J 
















i 


L i 


y 




















y 


V 


















/ 


"^ 


/ 




















/ 


















f' 


/ 


















— 


— 


r 




































































































1 














1 



200 
240 
220 
200 
180 

,ieo 

140 
120 
100 
80 
60 
40 
20 



15 10 15 20 24 

Five Minute Periods 

Fig. 36. — A curve of learning showing the progress of one person in learning 
to substitute numbers for letters in the experiments outlined in Chapter X, 
Experiments in Editcalional Psychology. 

or on the rate at which a child learns to read or to write. Conse- 
quently most of our generaliziUions up to the present time have 
been based upon curves that represent the acfjuisition of skill. 




Fin. 37. — Improvement in telegraphy 
ami Ilarter ('97, p. 49). 



8 12 10 20 24 
Wctks of Practicn 

Individual I 



28 32 80 



L. H. After Hr\'an 



Such curves seem to have in common two general characteristics, 
although it is doubtful whether they are univers;il in all types of 
learning: (i) An initial peritxl of rapid progress, and (2) successive 
periods of no progress, or plateaus, followed by i>eriods of rapid 



THE RATE AND PROGRESS OF LEARNING 



^43 



progress. Theoretically, a curve of learning may have two initial 
directions: (i) rapid progress followed by slower progress, or (2) 
slow progress followed by more rapid progress, that is, a convex or a 
concave form with all possible shapes between these extremes. The 
large majority of curves of learning derived to date are of the former 
sort. All the illustrations reproduced in Figures 37 to 41 have this 
general shape. In the substitution test referred to in Figure 36 
the author found that among twenty curves obtained from as 
many individuals, thirteen were of convex form, six were practically 
straight lines rising from left to right, and one was of concave 
form. Hence, initial rapid gain seems to be a very common feature 




7 10 20 

Weeks of Practice 
Fig. 38. — Improvement in telegraphy analyzed. 
Bryan and Harter ('99, p. 350). 



30 



Individual J. S. After 



in curves of skill. Other types of curves are shown in Figures 42 
to 44. 

The early period of rapid progress may be due (i) to the fact that 
the first elements of a new set of materials or a new set of associa- 
tions may be picked up rather easily and quickly because of their 
simplicity, (2) to the probability that the first stage of practice in a 
new type of learning makes available various elements or activities 
already in the possession of the learner, (3) to the initial zeal in 
beginning a new task, (4) to the large opportunity for progress in 
the beginning, (5) to the physiological limit in many types of skill 
such as typewriting, mirror tracing and writing numbers for letters, 
and (6) to an absolute limit of the number of bonds that the task 
presents to the learner. Thus typewriting has a physiological 
limit in the rapidity with which the fingers can be moved in striking 
the keys. Progress cannot go on indefinitely at the original rate. 
Typewriting also has an absolute limit in the number of strokes 
to be learned. 



144 EDUCATIONAL PSYCHOLOGY 

Book explains the initial period of rapid progress thus: 

"After what hits been Sviid our explanation of the general features of 
our curves can be brief. The first ra[)id and continuous rise is due to 
the fad that the KarniT is niakinfi progress along many different lines at 
once. Rapid strides of improvement are possible and made simulta- 




5 10 15 <!0 25 ao ur> ^U 45 50 55 00 05 70 75 
Days 

Fig. 39. — Progress in karninj^ Rvissian. After Swift ('08, p. 198). 

neously in every department of the work. The learner is not only forming 
and perfecting letter a.ssociations but syllable, word and phrase associa- 
tions as well. lie is simultaneously improving his method of dealing 
with every problem that the writing jiresents; locating the keys, directing 
and controlling his lingers, 'spelling' or initiating the movements, get- 
ting his copy, learning to deal with sju'cial (hiVuulties, learning to keep 
attention more closely and economiially applied to the work, etc. The 
curves will rise rapidly and continuously sy long as many of these j)o.ssibil- 



THE RATE AND PROGRESS OF LEARNING 



145 



ities of improvement exist. As they grow less numerous the rate of gain 
will likewise decline until, as still more skill is acquired, a state is reached 
where most adaptations or short cuts in method have been made; fewer 




5 6 7 8 9 10 
Successive Trials 

Fig. 40. — Improvement in tracing a star outline when seen in a mirror. 
Continuous lines represent reduction in seconds in successive trials. Dotted 
line represents reduction in errors. 





1900 






































1700 





































G 


1500 










Criti 


■1I A 


A 


A_, 


:Sta? 
A 




k 








f^ 


^ 


\ 


1300 






" 




aH 


-A 


-A- 


ff/l 


V' 


t 


r^ 


W^ 


r^f4-. 




i. 


P. 


1100 






vAA 


kh 




\r^ 


^ 


^r\/ 


^ 


!__ 


Criti 


;al £ 


tage 












900 


/ 


^ 
w 


v 


/TV 


^/v' 


























i^ 

M 




7 


/ 

































u 


Ma 




































g 
3 


/300 




































'A 


100 











































































10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 
Amount of Practice in Hours 

Fig. 41. — Improvement in typewriting by the sight method. After Book 
('08 plate, opposite p. 21). 



146 



EDUCATIONAL PSYCHOLOGY 



special habits remain to l)e developed; fewer adaptations arc possible. 
Those possible have become harder and harder to make, l)ecause they 
must be made in the realm of higher habits where the learner has had 
less experience. Every man has had experience with the first stages of 
learning, but little with the later stages because most people touch 
lightly many things and are masters of nothing. There being now fewer 
adaptations to make, and the process of finally perfecting all the special 
associations being so gradual and slow, the learning curve becomes, as 




Fig. 42. — ProRress in hall-tossinR. The horizontal axis represents days. 
The vertical axis represents the number of balls taught. After Swift ('cxS, p. 1 74) . 

the exi)ert stage is approached, almost horizontal. In the later stages of 
learning the sole gain must come from an occasional adaptation and 
from a further perfection of the present habits and methods of work." 
(Book, '08, pp. 90 f.) 



.Swift ancj Rat son have each pvihlished curves based on the in- 
crease in skill in hall-tossing which purport to he of the concave 
iypc. A careful examination of the orij^inal data shows that the 
apparently concave form is in reality due in each case to peculiari- 
ties in the method of plotting. Fortunalely sonu- of the data pul)- 



THE RATE AND PROGRESS OF LEARNING 



147 



lished by Batson permit of being plotted strictly according to the 
principle laid down at the beginning of this chapter, namely, that 
vertical distances represent amounts of performance and horizontal 



500r 



450 



400 



;3oo 



!200 



100 





l^O 




100 




80 




60 / 




40 -l^X- 


-ih Kt 


20 ^3-j±5r 


^^visj^SL u 


r^^^^^^T^^^ 



1 11 21 31, 41 51 61 71 81 
Successive Trials 



20 



4 8 12 16 

Days Practice 

Fig. 43. — Lower graph shows Batson's original curve 
the reconstruction of his curve as stated in the text. 



24 



Upper graph shows 



distances represent equal amounts of time or practice. This method 
of plotting yields the perfectly typical convex learning curve shown 
in Figure 43. 

The types of learning so far investigated have been for the most 
part of a relatively simple sort. Other types of learning may be ex- 



148 



EDUCATIONAL l'.S\ (MIOLOCJY 



pected to bring to light curves of very ilitTerent shape. This is par- 
ticularly true of forms of learning which depend chiefly upon an- 
alysis and selection or in cases where there is no physiological or 
absolute limit within ordinary attainable bounds, such as, for ex- 
ample, learning facts of history. 

Two rather extensive studies on analytical types of learning are 
now available. They reveal a very characteristic type of curve. 
The first investigation, by Ruger ('10), was based upon the number 
of successive solutions of a given mechanical puzzle that could be 



1 2 y 4 .'i 6 7 8 9 10 11 12 

Time Spent in Practice 
Fig. 44. — Curve to .show the progress in solving puzzles, .\fter Rugcr ('10). 
as reconstructed by Thorndike ('14), III, p. 342. 

j)erformed Ijy an indixidual within a certain period of lime. Fig- 
ure 44. It yields a strikingly concave cur\'e in marked contrast to 
those we have previously examined. Unfortunately, practice was 
not continued long enough to reveal the complete curve of this type 
of learning. A second study, by Hull, was based ujion the rate of 
evolution of abstract ideas as shown by their increase in ability 
to function, Figure 45. The material used was an elaborate 
system of Chinese characters combined with nonsense .syllables. 
In this case the work was carried to the point t)f perfection. He 
founcl as a con.secjuence not only the initial concave section shown by 
Ruger but a later {H'riod of diminishing returns. Taken alone, this 



THE RATE AND PROGRESS OF LEARNING 



149 



last section strikingly resembles the learning curves of the simpler 
processes and its course is doubtless determined by the same causes. 
The initial plateau or period of slow progress is probably due to the 
necessity of making a preliminary analysis of the material used be- 
fore proceeding with the remainder of the process. Clearly the 
elements common to many situations, as in Hull's investigation, 
for example, must be perceived as separate elements before they 
can be perceived as common elements. 

It seems also quite likely that in learning facts of history or facts 
of science, in which there is no physiological limit and in which 

12 



10 



o 

Pi 

^6 



O 

O 
2 



















y 


— - 







. 












/ 


/ 


t?^ 




















/ 


/ 






















/"' 




















J 


-^ 






















/ 


/ 

/ 






















/> 
























// 






















/ 


/ 






















// 






















> 


^' 




















; 


y 























120 132 144 



12 24 36 48 60 72 84 96 
Minutes of Work 
Fig. 45. — Progress of two individuals in generalizing abstractions or forming 
concepts. After Hull ('19). 

the number of items that may be learned is practically unlimited, 
the course of the curve, at least for a considerable distance, is con- 
cave. This is hinted at in the results obtained with the author's 
tests in geography and history. Thus in the former test, given at 
the end of the school year to some 1,300 pupils, and in the latter 
test given to some 2,000 pupils, the average scores for the ends of the 
respective years were as follows: 



Grade 5 

Geography 25 

American History 



6 


7 


46 


72 


7 


20 



38 



These scores substantially mean that so many geographical or 
historical items were known to the pupil. Both sets of figures show 
a larger gain from the second year to the third year than from the 
first to the second. These data furnish of course only fragmentary 



I50 EDUCATIONAL PSVCHULOCiV 

portions of the curves of learning in these sul>je( Is. But so far as 
any form may be inferred from them, it is probably of the con- 
cave sort. 

It seems probable that future e.xperimentation will yield similar 
forms of curves in other types of learning. Apparently there are 
types of learning in which continued training brings increasing 
returns. Thorndike states that: 

"Negative acceleration (that is rapid rise or convex form) of any great 
amount is far from being a general rule of learning. On the contrary, 
it may well be that there are some functions, such as amount of knowledge 
of history or geography, or of foreign languages, or of fiscal statistics, 
where, by any jusliliable score for 'amount of knowledge' the rate of 
improvement in hour after hour of practice would rise, giving a pro- 
nounced fx)sitive acceleration. Each item of information may, in such 
cases, make the acquisition of other items easier; learning some one fact 
may involve knowledge of a score of new facts in the shape of its relations 
to the facts previously learned. So knowledge may roll up Uke a snow- 
ball, its sum being, say, as the cube of the amount of time spent. What 
we may call the 'knowledge functions' do, as a rule, show, to say the 
least, very much less of the diminishing returns from increasing practice 
than do the functions of skill in some single line of work which figure so 
often in the experimental studies of practice." ('14, II, p. 257.) 

Whether or not plateaus occur universally in all types of learning, 
and whether they are really unavoidable stages in the course of 
learning, is an oi)en question. They have not been found to occur 
as generally as the initial rise even in curves of skill. Brj'an and 
Harter found periods of slow progress in three-fourths of their 
subjects. Book found them in two of his three persons, and Swift 
reports none. In the twenty curves obtained in the author's sub- 
stitution test, eleven contained plateaus and nine did not. The 
practice periods totaled 120 minutes. It is possible that some of the 
curves might have shown plateaus if the practice had been con- 
tinued longer. 

Batson. who undertook an investigation for the purpose of study- 
ing plateaus, found none in the ball-tossing curves, although the 
training was continued for a long time, but found a pronounced 
plateau in learning to throw shot into a pocket. 

Plateaus may be caused by lagging in energy, by loss of atten- 
tion, interest, and elTort, by fatigue, by periods of mechanization, 
and the like. Rapid progress after a plateau may be due to a re- 



THE RATE AND PROGRESS OF LEARNING 151 

cuperation in physical energy, in attention, interest, and effort, 
to the acquisition of new methods of learning and doing the task 
concerned, and to better use of the bonds which have been made au- 
tomatic by the preceding practice. Bryan and Harter beUeve that 
the plateaus in the learning of telegraphy were due to the establish- 
ment of a hierarchy of habits. During the initial period of progress, 
the simple elements such as the signals for letters, were acquired 
first and when these all had been learned, there came a dead level 
during which the connections became automatized, and then rapid 
progress was again possible by virtue of the acquisition of combi- 
nations of letters into words and words into phrases. Their own 
statement follows: 

"A hierarchy of habits may be described in this way: (i) There is a 
certain number of habits which are elementary constituents of all other 
habits within the hierarchy. (2) There are habits of a higher order which, 
embracing the lower as elements, are themselves in turn elements of 
higher habits, and so on. (3) A habit of any order, when thoroughly ac- 
quired, has physiological and, if conscious, psychological unity. The 
habits of lower order which are its elements tend to lose themselves in 
it, and it tends to lose itself in the habits of higher order when it appears 
as an element therein. 

" 2. The Order of Learning the Habits of the Telegraphic Language. 

" The synchronous curves of Figure 38 and the experience of operators 
agree in showing that from an early period letter, word, and higher habits 
make gains (a) simultaneously, but (b) not equally. 

" (a) The simultaneity in these gains is shown in Figure 38 by the fact 
that from the point where the curves diverge, each continues to rise. 
This is perhaps to be explained by the fact that from an early stage the 
learner practises with sentences, taking them as slowly as necessary. In 
this way there is incidental practice of every language unit and of every 
language unit in its proper setting. 

" (b) The curves of Figure 38 show also, however, that for many 
months the chief gain is in the letter and word habits, that the rate of re- 
ceiving sentences, is in this period, mainly determined by the rate of re- 
ceiving letters and words, and that rapid gain in the higher language 
habits does not begin until letter and word habits are well fixed. This ob- 
jective result is supported by the introspective evidence of operators. In 
the first days one is forced to attend to letters. In the first months one is 
forced to attend to words. If the learner essays a freedom for which he is 
unfit, suddenly a letter or word which is unfamiliar explodes in his ears 
and leaves him wrecked. He has no useful freedom for higher language 
units which he has not earned by making the lower ones automatic. The 
rank and file of operators are slaves to the machinery of the telegraphic 



152 EDUCATIONAL PSYCHOLOGY 

language. They must copy close. They cannot attend much to the sense 
of the message as it comes, but must get its form, and re-read for the 
sense. Only when all the necessary habits, high and low, have become 
automatic, does one rise into the freedom and speed of the expert. 

" 3. The Plateaus. 

" \\'e are now prejiarcd to ofTcr an explanation for the siilient peculi- 
arity of the receiving curve — its plateaus. 

" A plateau in the curve means that the lower-order habits are ap- 
proaching their ma.ximum development, but are not yet sufficiently 
automatic to leave the attention free to attack the higher-order habits. 
The length of the plateau is a measure of the difficulty of making the 
lower-order habits suflkiently automatic." 

The explanation of plateaus probably depends upon the nature 
of the Icarninpj process in which they occur. The theory of the 
hierarchy of habits would probably not apply to such a task as 
mirror tracing. 

Experimenters arc divided in their opinions concerning the in- 
evitableness or the usefulness of plateaus even in those t}pes of 
learning in which they frequently occur. Bryan and Harter, Swift 
and others believe that they serve a beneficial purpose. Swift, 
for example, says: 

"The real advance in the early stages of learning is made during the 
periods of seeming arrest of progress. The manifest advance, that 
which is revealed by the cur\'e or by examination marks, which is the 
same thing, is discouragingly brief. By far the greater part of the learning 
jK'riod is spent on plateaus when both teacher ancl pupil, failing to under- 
stand the situation, feel that they are marking time. Yet it is during 
these days of retardation that the valuable and solid acquisitions arc 
being made. Americans who spend several years in Ciermany pass 
through a long period of discouragement. Though they stuily the lan- 
guage faithfully, and avail themselves of every opportunity to practice 
convers;ilion, they seem to make absolutely no progress. The length of 
this plateau-period varies with different persons, but all experience its 
oppressiveness. Now the mo.st curious feature of this plateau, aside from 
its overpowering monotony, is the suddenness with which it finally dis- 
appears. Several have told the writer that they went to sleep one night 
unable to understand anything, as it seemed to them, and utterly dis- 
couraged, and awoke the following morning to find that they had mas- 
tered the language, that they could understand jiractically everything 
that w:us .said to them. The word associations ami national peculiarities 
of thought sequence had been automatized during the long period when 
no visible progress was being made." ('06, pp. 310 f.) 



THE RATE AND PROGRESS OF LEARNING 1 53 

Other investigators believe that plateaus are not necessary- 
stages in the course of learning, but that they are due to causes 
which may be avoided by introducing new stimuli or new methods 
of attack in learning so that continued progress may be possible. 

Plateaus are apparently not universal in all types of learning, 
nor are they found in all persons in the same type of learning. 
Whether they are useful stages in the learning process is a moot 
question. If they are not necessary, it would be highly important 
for education to prevent their occurrence in the learning of school 
material (i) by removing the conditions which bring them about, 
and (2) by providing stimuli at the points at which they are apt 
to occur so as to continue upward in the course of learning. Further 
experimentation will have to be made to furnish a definite solution 
of the problem. 

Factors Affecting Progress, a. Length and Distribution of the 
Periods of Work. How long at a time, and how often, should the 
learner work at his task in order to make the maximum progress 
for the time devoted to it? Every tj^pe of learning probably has 
an optimum length and frequency of periods of practice. Ebbing- 
haus ('85), in his pioneer study of memory, found that it was better 
in learning nonsense syllables to distribute a given amount of 
time over three days than to spend it all on one day. Sixty-eight 
repetitions made in immediate succession were not as advantageous 
for later relearning as thirty-eight repetitions distributed over 
three days. Practically all investigators who have touched upon 
this phase of learning have found a principle of similar nature. 
Jost ('97), also working with syllables, found, for example, that 
two repetitions a day for twelve days were better than four repeti- 
tions a day for three days. Some of the results of both Ebbinghaus 
and Jost imply that in some instances a decreasing amount of 
time on successive days would be more economical than an equal 
amount on all days; that instead of distributing 24 repetitions 
by having four on six successive days, it would be better to have 
eight on the first day, six on the second, four on the third, three 
on the fourth, two on the fifth, and one on the sixth day. 

Lueba and Hyde ('05), in an experiment on learning to transcribe 
EngHsh words into German script, found that of four plans of 
distributing time, twenty minutes twice a day yielded the slowest 
gain, while twenty minutes every third day yielded better, and 
twenty minutes every day or every other day yielded the best 
results. 



A\erage Time (Seconds) 


First 200 




Letters 


Last 200 


Approximate 


Letters 


41 5 


13 4 


57-5 


17. 1 


47.00 


16.5 


39 5 


18.2 


38 5 


iS 5 


44.00 


21 I 



154 EDUCATIONAL PSYCHOLOGY 

Miss Munn ('09) made an investigation of practice in a sub- 
stitution test consistinj.^ of transcribing 4,000 letters into other 
letters according to a key. Her distributions of time and results 
are given in the following table: 



TABLE 38 

Practice in substituting letters for other letters according to a key. After 

Munn ('09) 



No. of 
Subjects Distribution of Work 

23 200 letter, a day for 20 successive days 
4 Soo letters a day for 5 successive days 

(400 in a. m., 400 in p. m.) 
4 1000 letters a day for 4 successive dajs 
4 2000 letters a day, seven days apart 

4000 letters in one day (1000 at a sitting) 
4 3000 letters a day (at one sitting) 

The highest degree of efTiciency was reached by the 20-day 
group v.ho reduced their time for the last 200 letters to 13.4 seconds. 
A definite com]xirison is a little difficult to make owing to the 
large differences in initial ability among the various groups. 

In the substitution test carried out by the author, Figure 46, 
ten minutes twice a day was productive of the greatest i)rogress, 
twenty minutes once a day was productive of almost as rapid 
progress, forty minutes once a day was productive of considerably 
less progress, while 120 minutes at one time ])roduced saircely 
half as much ])rogress as the ten-minute or twenty-minute periods. 
The toUU time in all four distributions was the same. 

Dearborn ('10), who reported an earlier e.\j)eriment with the 
same substitution test, divided the subjects into two groups work- 
ing ten minutes once a day and ten minutes twice a day respectively. 
lie found a small advantage in favor of the former group. 

Pyle ('13), working with a substitution test, rejjorts that: 

"Generally speaking, daily practice seems to give better returns than 
the same number of periods (listrihuted on alternate days or in twicc-a- 
day periods. However, there is some evidence that in the early stages of 
habituation, the second jiracticc on the s;imc day pives good returns and 
that, later on, alternate days may be the best distribution." 



THE RATE AND PROGRESS OK LEARNING 



155 



Kirby ('13) carried out a practice experiment in addition and 
division with 1,300 pupils in the third and fourth grades. The 
pupils practiced addition in 22.5, 15, 6, and 2-minute periods, and 
division in 20, 10, and 2-minute periods with the following gains: 



260 


- 


y-^^y 


m 250 


- 


y—^jf^^ 


S 240 


- 


/"x^"^^^ 


3 230 


- 


y X 


1 220 


- 


r^^/^ / 


S 210 
gl 200 


- 


€-^^rZ^ h^ 


E 190 


- 


^/ "^^ 1 ^^^/^X/ 


C 180 


- 


/ /^/ y ^""^ 


•- 170 
5 160 


" 


^M/-^^^y^ 


X 150 

g 1« 


: / 


yy r^J^^^'\ /\ 


C 130 


^r 


j^ ^. / ^^^ ^< y ^k J 


2 120 




"/V ^'^^^ ^*^ ^"^ 


^ no 


-IjT 


y\/ 


2 100 




^ 


^ 90 












"S 80 






^ 70 


. 




«4-( 






60 


- 




55 50 


- 




^ 40 






H 30 


- 




^ ?X 


- 




10 


f 1 1 


1 ' 1 1 1 1 ' 1 1 1 1 1 1 ■ ' ' ■' ' ■ 



123456789 1011121314.1516171819 20 2122 23 24 

Successive Five Minute Periods 

Fig. 46. — Practice in writing numbers for letters according to a key. After 
Starch ('12). 

10 min. curve = group working 10 min. twice a day. 
20 " " = " " 20 " once " " . 

40 " " = " " 40 " every other day. 

120 " " = " " 120 " at one time. 



Period 

22.5 

15- 
6. 



TABLE 39 



Addition 

Per Cent Gain over the 

22.5-Minute Period 



Division 



21 % 

I % 

46.5% 



Period 

20 

10 

2 



Per Cent Gain over the 
20-Minute Period 



10. 5% 
77 % 



The superiority of the 2-minute period is probably exaggerated, 
as Thomdike has suggested, by the greater opportunity for out- 
side practice and longer continuation of regular school work, since 
this period was extended over a larger number of days. 

Thomdike ('11) compared the improvement in multiplying 



156 KUUCATIO.NAL I'SVCIIOLOOY 

mentally with three-place numbers continuously, for two to twelve 
hours with Miss Whitley's Cii) sul^jects who did three similar 
problems a day for twenty days. The outcome was in favor of 
the distributed practice, l)ut probably only slightly so when allow- 
ance is made for the etTect of fatigue in the continuous work. 

In general, relatively short periods of work in simple associative 
learning are probably the most economical. It would be unwise 
in the absence of more extensive experimental studies, to generalize 
to all ty])es of learning and particularly to the learning of school 
subjects. What the most producti\e periods for learning reading 
or spelling or Latin or English compo.sition are, will have to be 
determined experimentally in each case. All that we can say at 
present is that each t\q)e of learning probably has its optimum 
length and distribution of jiractice periods. Lyon has stated the 
matter in the following manner as a result of his studies on this 
problem: 

'"With reference to the problem of the most favorable distribution of 
single reading. ... I would siiy that the most general statement that 
can be made, taking all materials and methods of prest^ntation into con- 
sideration, is that the most economical method is to distribute the read- 
ings over a rather lengthy period, the intervals between the readings 
being in arithmetical proportion. P'or example, with one individual, in 
memorizing a poem of twenty stanzas, the highest rctcntiveness was 
obtained by distributing the readings as follows: two hours, eight hours, 
one day, two days, four days, eight days, sixteen days, thirty-two days, 
etc. The practical bearing of the results obtained on education in general 
is that when associations have once been formed, they should he recalled 
before an interval so long has elapsed that the original associations have 
lost their color and cannot'be recalled in the siime shape, time, and order. 
In general it was found that the most economical method for keeping 
material once memorized from dis^ippearing was to review the material 
whenever it started to fade. Here also the interv'als were found to be, 
roughly speaking, in arithmetical proportion. For similar reasons the 
student is advised to review his lecture notes shortly after taking them, 
and, if possible, to review them again the evening of the same day. Then 
the lapse of a week or two does not make nearly so much dilTercncc. 
\\ hen once he has forgotten so much that the various associations orig- 
inally made have vanished, a considerable portion of the material is 
irretrievably lost." ('13, p. 161.) 

b. Fori^rltin^. Learning is, in a certain sense, a fortification 
against forgetting, and from the practical side, it is imi)ortant to 



THE RATE AND PROGRESS OF LEARNING 



157 



know how frequently and in what manner the fortifications should 
be strengthened in order to resist the attacks of forgetting. Only 
a few experimental studies have been made on the rate and factors 
of forgetting. Ebbinghaus ('85) learned nonsense syllables until 
he could give them once correctly, and then measured the rate of 
forgetting by the amount of time required for relearning them at 
diflferent intervals after the original memorizing. Radossawlje- 
witch ('07) used nonsense syllables and poetry, Bean ('12) used 
series of letters, and Magnefif ^ used poetry. The curves of for- 



100 

90 
80 
70 

a 
■'SO 



■■% 

\ \ 



Ebbinghaus ""*'x"-.. 



'''''''''"'■'■ ' "■ 



edCutv" 



5 10 15 20 25 30 35 , 40 

Days 

Fig. 47. — Curves of forgetting. 



getting obtained by these investigators are presented in Figure 47. 
They agree in showing a very rapid rate of loss at first, followed 
by a very gradual decHne afterwards. In retaining syllables, 
Ebbinghaus found that he forgot as much in the first twenty min- 
utes as in the following thirty days; in remembering a poem which 
had been learned to the point of two perfect repetitions, Rados- 
sawljewitch found that his subjects forgot as much in the first 
two days as in the next twenty-five days. 

Inquiry into the rate of deterioration of connections through 
lack of practice have also been made by Book ('08), Rejall,^ Swift 

' As reported by Radossawljewitch. 
^ Reported by Thoriidike, II, p. 309. 



158 • EDUCATIONAL I'SVCIlULOCiV 

and Schuyler ('07) in the case of tj'pcwriting, and by Swift in the 
case of tossinp; l>alls. These seem lo indicate much ^eater per- 
manency in sensori-motor connections than in the memorj' of syl- 
lables or ])oelry. 

The ex])erimental work on forgetting is too limited as yet to 
])ermit of much definite application in the practical ])rocedure of 
learning school material. The one suggestion that may possibly 
be made would be this: Since the rate of forgetting is very rapid 
at first and more gradual later on, it ])robably would be highly 
ad\antageous to have releaniing of a given material come very 
frequently at first and more rarely later on. Thus the facts of a 
lesson in history or the newly accjuired words of a spelling lesson 
should be reviewed the next day or perhaps j^referably on the 
same day, then again two or three days later and then a week or 
ten days later, and so on. 

The effect of long vacations upon the retention of school material 
has been investigated only partially. Measurements of skill in 
arithmetical operations in June and September show heavy losses 
(see page 403) and raise the question as to whether long vaaitions 
are really j^rofitable or detrimental. 

c. Concentration, Effort, and Zeal. "Practice makes perfect," 
is only a half truth. Only ])ractice with zeal and effort is likely to 
bring improvement. A great deal of practice and repetition may 
continue day in and day out Avithout the slightest gain. While 
the factor of zealous attention and interest has long been recog- 
nized as a matter of common-sense observation, its real value, 
however, has never been appreciated until ex^ierimental studies 
])ointed out its actual significance. Bryan and Barter have called 
attention to this point in a very emphatic manner as follows: 

"A fact which seems to he highly signilicanl is that years of daily prac- 
lirc in receiving at oniinary rates will nut brinp a man to his own max- 
imum ability to receive. The proof of this fact is that men whose rc- 
( civing curve has been upon a level for years, frequently rise to a far 
higher rale when forced to do so in order to secure and hold a position 
rrfiuiring the higher skill. That daily practice in receiving will not 
assure improvement is further seen in the fad that in many cases in- 
ferior operators after being tolerated for years are fmally dropped be- 
cause they do not get far enough al)ovc the dead line. One conclusion 
.seems to stand out from all these fads more clearly than anything else, 
namely, that in learning lo interpret the telegra|)hi( language, it is in- 
tense clTort which educates. This seems lo be true ihruughuul ihe whole 



THE RATE AND PROGRESS OF LEARNING 1 59 

length of the curve. Every step in advance seems to cost as much as the 
former. Indeed, each new step seems to cost more than the former. In- 
quiry at the telegraph school and among operators indicates that between 
sixty and seventy-five per cent of those who begin the study of telegraphy 
become discouraged upon the plateau of the curve just below the main- 
line rate. As a rule, ordinary operators will not make the painful effort 
necessary to become experts. Facts of an analogous character will be 
recalled from other fields. 

"The physiological, psychological and pedagogical implications of this 
conclusion are manifestly important. If in our educational methods in 
the past, we have often made the pace that kills, there is possibly the 
danger on the other hand that we shall make school work all play, and so 
eliminate the intense effort which is necessary for progress." ('97, p. 50.) 

A great deal of learning is done without any real zeal or effort 
toward improvement. The usual way in which a great many 
children learn to play the piano illustrates how much practicing 
and learning consists in dawdling with more attention upon the 
clock than upon the music sheet. A great deal of learning of school 
material is done with the same lack of interest and effort. 

d. Specific Practice in the Functions to be Improved. One of the 
striking discoveries of experimental investigations is the very 
rapid progress in specific functions when the practicing is done on 
the particular connections to be established. A surprisingly large 
percentage of pupils make little or no progress in an entire year's 
work in such subjects as reading, writing, and the like, while the 
remaining pupils make only "a part of the progress that they could 
make if their efforts were squarely directed at the material to be 
learned or at the associations to be established. 

The numerous practice experiments that have been conducted 
in psychological laboratories during the last two decades furnish 
overwhelming evidence of the tremendous improvement ob- 
tained under experimental conditions. Only a few examples will 
be cited. 

The writer found that eight persons, practicing mental multi- 
plication of three-place numbers by one place-numbers for about 
15 minutes a day for 14 successive days, made enormously large 
gains as shown in the following table: 



i6o 



KUUCATIOX.M. I'SVCIIOLOGV 



TABLE 40 
Improvement in mental multiplication. After Starch ('11) 



IvnivroiAL 


I'KR to .\1|NUTK.S 

ON Isr Day 


hxAUPLLs Dusk 

I'KK 10 .Ml.S-l'TES 

o.v Uru Day 


Gross Gain 


Perckxtace 
Gai.v 


s 


-5 

37-7 

23-8 

41-7 

14-7 

37 

25 

23 -4 


02. S 
81 

45-4 
71.4 
29 
100 
29.8 
66 


37-5 
43-3 
21.6 
29.7 
14-3 
63 
4.8 
42.6 


150 

"5 

91 

71 

97 

170 

19 

182 


St 

F 


V 

w 

H 


Si 


B 







These subjects gained in a])proximately four hours of practice 
per ])erson over loo^/c, varying, of course, from one jierson who 
made little gain up to two persons who gained nearly 200%. 

Wells found the amounts of gain from 150 minutes of i)ractice 
in addition on the part of ten adults as follows: 



TABLE 41 
Imijrovcmcnt in addition: adults. .Vftcr Wells ('12) 



Individual antj Skx 


NuMllKk l)K AUDiriONS IN I'lVK MlNl'TKS 


Percentage wiiitii 

.\moi:.nt Done on .?Oni 

Day was of .\moist 

Done on 1st Day 




I'lKST Day 


30TH Day 


Gros.s Gain 


if 


150 
180 
200 
220 
225 
22s 

23s 
250 
260 
200 


2S0 
380 

430 
380 
368 
460 
570 
440 
540 
S40 


200 
^30 
160 
143 
23s 
335 
190 
280 
250 


1S7 


2 m 


211 


» m 


215 
173 
164 
204 

243 
176 


4f 


cm 


6 m 


7f 


8f 


of 


208 


10 m 


186 







The gains show ap])ro.\imalily a douiiiing in i-lVicieiuy in the 
course of thirty days. 

Dearborn used vocabularies and poetry in learning e.\periinenli: 
and nportLcl the following results: 



THE RATE AND PROGRESS OF LEARNING i6l 

TABLE 42 
Improvement in ability to memorize After Dearborn ('10) 

Total Prac- Amount Number of Total Time Re- Time Re- 
Subject TicE Time Learned I-ays of Amount quired on quired on 
IN Hours Daily Practice Learned First Day Most Effi- 
cient Day 
Learning the English meanings of French or German words: 

1 61/3 50 21 1050 30 13 

2 6 35 20 700 30 12 

3 6 35 18 630 30 14 

4 81/10 30 22 660 S3 15 

5 72/3 30 20 600 40 15 

Learning poetry: 

7 .-31/3 32 15 480 38 7 

8 32/3 18 16 288 30 8 

10 4^ 17 13 221 30 12 

Similar results have been reported by Bair in tossing shot, by- 
Swift and Batson in tossing balls, by Whitley in drawing lines in a 
maze, by Wells in tapping, by Kline, Wells, and Whitley in can- 
cellation tests, by Thorndike, Wells, and Kirby in adding, by 
Swift, Book, Rejall in typewriting, by Ebert and Meumann, 
Winch, Sleight, Dearborn, and Fracker in memory — in fact in all 
experimental work in which practice enters. Improvement of 
mental functions through practice is well-nigh universal and the 
amount of improvement through specific training under experi- 
mental conditions is almost incredible, particularly when we con- 
trast with it the gains made in school functions in from 50 to 150 
hours devoted to a subject in the course of a year. 

The average gain made by pupils in school activities in the 
course of a year's practice as indicated by the standard scores 
derived from measurements with tests and scales is shown in the 
following percentages of gain at the end of the eighth grade as 
compared with the end of the seventh grade: 

^ Approximate. 



l62 



liUUCATK ).\AI, rSVCHULUGV 



TABLE 43 
IJasi-(l u|Mjn published scores for the various tests 



7TH 


8th 


36 


4.0 


45 


50.0 


750 


83.0 


10.4 


10.9 


6.5 


8.0 


8.5 


10 


6.5 


8,0 


70 


9.0 


II .0 


12. 6 


8.0 


8.3 


41.0 


46.0 



Per Ck.st 
Gain 



Reading: speed, words i)cr second 

ReadinR: comprehension, words written 

Writing: speed, letters per minute 

Writing: quality, Thorndike scale 

Addition: Courtis Series B — rights 

Subtraction: " " " " 

Multiplication: " " " " 

Division: " " " " 

Reasoning: Starch Arith. Scale A 

Language: Starch dram. Scale A 

Composition: Hillegas Scale 



II 
1 1 
1 1 

5 
-\5 
17 
23 
28 

15 
4 



These gains arc surprisingly trivial when compared with the 
gains, often running o\er 100%, reported in connection with 
ex]K'rimental investigations of practice. 

Definite experimental results are not at hand to substantiate 
the following assertion, which may seem doubtful but which is 
not impos.siblc from the present indication of other measurements, 
such as those presented by Dearborn in Table 42 or by the \\Titer 
in Table g (Exj)eriments), namely, that the average high school 
pupil could learn in 20 minutes a day for thirty days, all the Latin 
words (500 words) that he would need in an entire year of Latin. 
He could learn in 30 minutes a day for one-half the school year, 
all the Latin words (approximately 2,000) that he would use in his 
entire study of four years of high school Latin. 

The difficulty with the material of school subjects is that we do 
not, and in some instances we cannot, specify with sufficient def- 
initeness just wherein the improvement is to be made. We can 
point out specifically to a child whether or not he spells a word 
correctly and what part of the word may be incorrect, but we have 
not until recent years made any attempt at determining which 
particular words a child really ought to know how to spell. The 
pupil was given a list of words selected more or less oit the l)asis 
of their unusualness and difiicully rather than upon the basis of 
usefulness or frequency of occurrence. The idea seemed to be that 
if he learns to spell a suffic ii-nt nimibcr <if difficult and useless words, 
he will know how to spell all other wonls in the English language. 
The school has virtually .said, "Learn to spell," but has not said 



THE RATE AND PROGRESS OF LEARNING 163 

what a child should learn to spell. Even in a subject in which the 
associative bonds may be precisely defined so that they can be 
directly attended to, we have not done so. The same situation 
obtains in practically all other subjects with the added difficulty 
that in some subjects the material is of such a nature that specific 
directions and specific material or specific bonds to be formed, 
cannot easily be isolated. This is particularly true of such a sub- 
ject as English composition which represents the opposite extreme 
from spelling and arithmetic. The child is told to improve his 
style, or his language, or his expression, or his originality, or his 
imagination; but he is not told very definitely how he may do this, 
or just what he is to do. The school should, therefore, aim to 
specify exactly what sort of learning is to be done so that a definite 
notion on the part of the learner may be formed of the precise bonds 
and connections to be made. 

e. Definite Knowledge of Success and Error. Much experimental 
work implies that the feeling of satisfaction resulting from success- 
ful trials of a task facilitates the formation of the connections con- 
cerned. It seems obvious therefore as a practical matter that pre- 
cise knowledge of the success or failure on the part of the learner 
is exceedingly important. It will not only serve to arouse the 
feeling of satisfaction but also help to define the particular bonds 
to be estabhshed. Feelings of dislike on the part of the learner 
toward the material to be learned undoubtedly interfere with the 
rapid formation of the connections, and frequently the feeling 
of dislike is accompanied by an attitude of unwillingness or stub- 
bornness indicated by such statements as "I know I can't learn 
languages; I never could." "I never was able to get mathematics." 
"I can't memorize anything." A concrete case that came under 
the writer's observation was that of a man considerably older than 
the average university student, who in the experiment on the trans- 
ference of training (Chapter XI, Experiments in Educational 
Psychology) reported that he was unable to learn vocabulary and 
that the net result of half an hour's work on the first list was ten 
words. The average student is able to learn the entire list of thirty 
words in from twelve to fifteen minutes. He further stated that 
he had always had great difficulty in learning languages. In order 
to ascertain, if possible, the real status of his memory and other 
abilities, he was tested by Terman's revision and some additional 
tests, all of which indicated that he was of average intelligence and 
that his memory was not defective, but approximately average. 



164 F.DUCATFONAI. I'SVCHOUXJV 

He was informed of the results of the tests, that his defective 
memory was largely illusory, and that probably his real trouble 
lay in his contrary attitude toward certain tasks, which was also 
indicated by his own statements concerning his work. The general 
efTect upon his later attitude in learning was a distinctly whole- 
some one. This case is cited because it exemplifies many similar 
instances of persons who feel incapable of learning certain things. 

/. Interest in Improvement. One important element in the re- 
markably large amount of gain through practice in specific func- 
tions, is the fact that the progress is directly observable and 
definitely measureable which in turn produces a real zeal toward 
improvement and toward outstripping the preceding records. 
In an experiment such as the substitution test or the practice ex- 
periments in arithmetical operations, the observation of a definite 
gain is possible so that the learner can see just how much progress 
he is making. The practical value of this suggestion would be the 
creation of circumstances for the learning of school subjects similar 
to the conditions of learning in laboratory experiments by intro- 
ducing forms of measurement through which the progress may be 
determined at frequent intervals, so that the j)U})il may see what 
progress he is actually making. 

In a certain elementary school a series of standard tests ^ was 
applied every month throughout the entire school year. Tests in 
reading, writing, spelling, and arithmetic were given at monthly 
intervals to determine the progress made. Each pupil knew his 
own record in each test and compared it from month to month. 
This condition developed a remarkable interest and zeal in striving 
to surpass the record of the preceding month. The condition 
created thereby was similar to that of a learning experiment under 
laboratory conditions. Each pupil kept his own score and knew 
what gain he was making each month. This condition had a re- 
markable efTect upon the total progress made during the school 
year as shown in the accompanying graphs. These results show 
that the pupils made on the average a gain in some studies 
twice as great as that made ordinarily in the course of a school 
year. This gain cannot be attributed to familiarity with the test 
material since, in the case of reading, dilTerent passages were used 
in each successive test; in writing, a dilTerent sentence was used 

' The ti-sfs were made in The Alice SchfK)!, IlihhitiK. Minnesota, hy Princifvil L. J. 
Couhal, an<l rcporte<l in an unpulilislu-d thesis in the lilir.ir>' of the University of Wis- 
consin. The testa were carried out under tJie direction of Professor V. A. C. Henmon. 



THE RATE AND PROGRESS OF LEARNING 



if>5 



each time; in spelling, the author's six lists were used in rotation, 
one at a time; and in arithmetic the three sets in the Courtis Series 
A were used and rotated so that there was a recurrence of the same 



30 
20 






























^ 


^ 












z::!. 







-_— 


10 



















Fig, 



Sept. Oct. Nov. Dec. Jan. Feb. March April May 
48. — Progress in speed and comprehension of reading combined into 



single scores as measured by monthly tests (Starch reading tests) upon 4th grade 
pupils. The continuous curve represents the progress of the class. The straight, 
broken line is the progress for schools generally based upon the standard scores 
for June of the 3rd grade and June of the 4th grade. 

test every three months, but it is very unlikely that this contributed 
any appreciable share toward the gain shown. It would seem, 
therefore, highly desirable if there could be introduced into the 
schoolroom a similar atmosphere of motivation such as obtains 



40 
30 
20 
10 




























- — 


-^ 


^ ^ 




_ - 


t^" 


--— 


— -^ 

























Sept. Oct. Nov. Dec. Jan. Feb. March April May 

Fig. 49. — Progress in speed and quality of writing (Thorndike scale). Other 
facts same as for Fig. 48. 

in learning experiments in the laboratory. The knowledge of one's 
actual ability and of the actual amount of gain serves as an ex- 
ceedingly powerful spur for the learner to surpass his own previous 
performances. The popular dictum "Nothing succeeds like sue- 



1 66 



EDUCATIONAL I'SVCHOLOCiY 



cess" may \k- partly juslituil hy such experimental results as the 
ones here eited. To see onisilf ^jaininj^ tinds to stimulate greater 
eiTorls toward K'Hin. The educational measurinf^ scales and tests 



ou 
JiO 
•10 

•JO 
































y 












--^ 


^^^-~ 


"'"""' 




















ao 







Sept. Oct. Nov. Dec. Jan. Feb. March April May 

Fig. 50. — I'roj^ress in sjK-Uinj^ (Starch test) of ^nl grade class. Other facts 
the same as for Fig. 48. 

ought to serve a useful purpose at this juncture. They will pro- 
vide means whereby the pujiil may be able to see for himself in 
definite terms the gains he is making. 



ti 6 

5 i 



Addition 




iS 



10 
_ 8 

b£ 6 





I '2 

-g) 



Multiplication 



Divi.sion 



Sept. Oil. 



Dt-c. 



Jan. 



Feb. March April May 



Fig. 51. — Progress in the fundamental operations in arithmetic (Courti.s 
tcstii). Other chita s;imc as for I'ig. .j8. 

.(J. yUnlal Ima^vry. Aflir the early studies on mental imagery 
l)e(ame known, there followed considerable theorizing as to tyjies 
of persons and tyj)es of learners, and with it came the resulting 



THE RATE AND PROGRESS OF LEARNING 167 

endeavors to make applications of these theories to methods of 
teaching. Thus it was said that if a pupil has difficulty in learning 
to spell or in learning a foreign language he may be devoid of, or 
weak in visual imagery; or if he has trouble in learning to write, 
he may be short on motor imagery; or if he finds it hard to learn 
the pronunciation of words, he may be defective in auditory and 
motor imagery; and if he fails in the academic subjects he was sus- 
pected of being devoid in visual and auditory images and strong 
in motor imagery and should therefore turn to manual training. 
The proposed remedy was that the material to be learned should 
be presented to a different sense organ so that the pupil might use 
the imagery natural to him. All these inferences may possibly 
be true; but later additions to our knowledge of mental images 
make us more hesitant regarding the real part played by them in 
learning and concerning the actual differences produced by present- 
ing material to different sense avenues. 

Before we can make changes in practice we must be sure of the 
principles upon which the practice is to be based. It is important, 
therefore, to examine at least the following three considerations: 

In the first place, more careful studies of the sorts of images 
employed by different individuals show that the classification of 
persons into visuals, audiles, motiles, and so on, is fundamentally 
misleading. Studies by Betts ('09) and others have helped greatly 
to clarify the matter by showing that pure types rarely exist. 
During the last six years several hundred students have performed 
the imagery tests outlined in Chapter VII of the author's Experi- 
ments ('17). Among this entire number not more than two or three 
persons were found whose images either were practically all of 
one type, or who had one or another commonly prevalent class 
almost entirely missing. The facts for 95% of all persons are sub- 
stantially as set forth on page 45 (Experiments), namely, that nearly 
all persons have all types of images which are combined in different 
individuals in varying proportions. Mankind as a whole does not 
fall into sharply or even vaguely divided groups of visuals, audiles, 
and so on. They are not found except in rare instances. 

In the second place, recent inquiries indicate that images of the 
class most natural or predominant for a given person may be 
aroused by stimuli coming through another sense. For example, 
auditory stimuli may arouse visual images as well as, or even more 
readily than, auditory images if visual images are more natural to 
the individual. Miss Abbott ('09) found in a detailed investiga- 



!68 EDrC\TroX.\L PSYCHOLOGY 

.-OS) with four subjects, on the memory coDsdooaiess in (Nrthog- 
ni;:^y ikit " imespecm-e oi the method ot |HresaitatioD and the 
maimeT «x karai^sr. the t^-picaJ mode of recall for all ohserMers 
wa5 thrvx:^ the xis^ai irsauary ci the letters" (p. 153). Coase- 
quenth* it dees not loiiow that, even if a person has a strmg lean- 
ing tovard ooe or another type of imageiy, it is necessary or even 
advantageoas to present the material to be learned through the 
sense avenue <tf tk type of im ag ay . 

Cotvin and M~ - .ade an extensive series of tests with 

school diildrer to what es^tent vtsoals retain visual 

matexni best. ^- ^^ ^ :ory m a te rial^ and motiles motor de- 
ments. He condoded that: 

"These seens to be a fudjr definite rdatioa betineen the effectiveness 
ci nKssaocy iit the case of a paitknhr ideal ional type and the memoiy 
imaternl vhich k naost suited to that t3rpe. la partimlar the vsual t3rpe 
letxms best mateeial vith a vEoal contenit and the anrtitoty and motor 
typeSstoalessdegPBe, ^me I M lwith2^:^:2^i:::- : - ~ : rr rontent, as tlse 
case naaybc.'* 

While this oandnsaan is pn^': ~r?s among 

the vanoos types of elements r. ^ 4>s of 

pi^iik are in most instances am. 

In the third place, the mo&: ~' 

pRsentatkn for a given pei^oc 
corresponds to hfe dominant typ 

mental evidence that has been .ted on the problem o: 

modes of presentation of the I -lal is conflicting ir. 

character. It ^ uncertain whet itoiy. motor, audi- 

tory-motor, or visual- jiks are mos: 

advantageous. Henmc z. .^iieiiment which 

has thrown important l^it opoo the problem. He employed four 
cniethods of presentation, ^' ^ditor>*. \isual-auditory, and 

visoal-aoditory-motor: and _ e sorts ci materia], cmcrete 

nouns^ two-pboe numbers, and nonsense s\-iIaUes presented to 
ax subjects with ooe, two. or three repetitions. 

'^ In the visual presentation the subjects read the sdsiul; 1 
the rotating drum and inunediately wrote down as many ' 
ooold be recalled and in the order presented. The subject ~ 
to R|xc95 mtnrments of articidatioa. In the anditocy prcs 
expenmenter read the stimuli horn the drum, the subject ^"T**^ bis 



THE IL\TE AND PROGRESS OF LEAieCTXG 109 

eyes closed and repressing movements of axlicalalioiL In the Tisna]- 
anditor)' preseatation the subject both sair the stimuli and iieaxd them 
read bv' tbe esperimenter. In tiie -visual-aaditoiy-inotar presentation 
the subject Mmself read the lists aloui"^ 

TTJi; results are simimarized thus: 

'" I. Aaditory presentstian is dearlr superior to visaal presenlatian in 
immediate memc - : ;. a result attribnlable to the greater ^Lse and 

freedom of "visni. ^Ith auditory presentatian and the grraiter 

effort of attention required. 

''' 2. This 5ig>eriority of anditarr orer Tisoal presesitatian holds for aD 
miateria3s (nouns, nonsense-syllables- numbers)- for all subjects irrespic- 
Irv-e of ima^e t3'pe, and for one. t^o and three presentations. Tins 
result is not in accord "with tlie cfsnion commonly "held that Tisual pre- 
sentation is superior, especially "srith meaningless material 

^' 3. Combined visual-auditoiy presentation 2= sliErtth- inferioT to the 
auditory- alone and decidedly 5ig>erior to the -^ - "-. The advan- 

tage of a combined method is very much les i_ ;^0"8m in ftarKpr 

investigalions. 

'■'■ £^ \l5ua3-anditory-motor presentatian is sSgfally inferior to the 
andiloiy and the visual-auditory presentations and siqjerior to the visual 
alone. Articiilation or vocalization is of littk value for immediate mem- 
ory-. 

■"'5. The correlations of abilities "snli diaer^- ' ■ ' " ^esemation 

are positive and very Hgh. superiority -Kith . . : radically 

the 'giTn<» degree of siq>eriority "«illi another."" .-J^t^r r-'r^.r.~ rtTi '12.) 

A fair general impresaon of the present status of our knowledge 
of imagery in relation to leammg "would be that distractions among 
types of pupils have been o\-erenipha5ized and that much of Ihe 
endeavor to adjust methods of teaching accordingh* ha^ been mis- 
directed. 

h. Faii^m. As a fmal inoportant factor in the progress of learn- 
ing, we must consider fatigue. EducatiorLal and psychological 
L terature has been replete with discussions res-arding the par: 
~'zl''i, :'i:'.-^je plays in the re" - . . - ■^jjq^-_ W^r 

:'. !;'r-:;al research has ^ t informs tica 

; ; —jng the course of conthraous work and concerning the 
- - r in the efiSdency of the worker as measured by cross- 
A tests at various stages of work, it has not furnished as yet 
.:e knowledge concerning the control of the work of the 
- :hooL Pedagogical literature has been generous in 
- . _ L agns of fatigue and serious consequences of overwork 



170 KDUCATIO.VVL PSYCHOLOGY 

and in suggesting rcmtdits for avoiding exhaustion, yet wc arc 
not sure ^vhcther the so-called symj)lonis are indications of real 
fatigue or whether any sirious or even mild fatigue effects ever 
result from the work as carried out in the great majority of schools. 

In discussions of fatigue it is important to bear in mind two 
distinctions in the meaning of the term, namely, fatigue in the 
sense of decrease in the capacity to do work, and fatigue in the 
sense of decrease in interest in, or willingness to, work. The two 
are plainly different and do not necessarily go together. The one 
is actual loss in efl'iciency; the other is a feeling of ennui or weari- 
ness. Much of our thinking about the proljlem has been confused 
by a failure to distinguish between these two meanings. Fatigue 
in the former sense probably has been greatly exaggerated as an 
educational ])roljlem. Perhaps only in exceptional individuals is 
there injurious overstrain due to menial work. The discussion of 
this topic will, therefore, be abbreviated. 

The experimental methods by which the phenomena of fatigue 
have been investigated will first be mentioned briefly. They may 
be divided into two classes: (i) Indirect methods, and (2) Direct 
methods. 

(i) Indirect methods. The principle, upon which the indirect 
methods have proceeded, has been to measure some physiological 
or psychological functions at different points during the course of 
work in order to compare the efhciency of those functions on the 
assumption that the dilTerence in them would be indicative of 
efficiency in general. One of the first melhiKls was that employed 
by Griesbach, who determined the two-point threshold uj)on various 
parts of the skin at various times of the day on the belief that a 
decrease in sensitiveness or a widening of the threshold indicated 
a reduction of general mental efficiency. He made extensive com- 
jnirisons among school children for the purpose of determining 
the amounts of fatigue produced by various types of .schot)l work, 
and formulated an elaborate series of conclusions with ri-gard to 
them. For examjile, specific fatigue values were assigned by him 
and his followers to the different school subjects. Vannod states 
that mathematics, Latin, and Greek produce most fatigue, and that 
French and grogiajihy produce least. The ditliculty, however, 
with results of this type is that while the two-point discrimination 
ujxjn the skin varies unfler dil'lerent mental and physical conditions, 
it is a rather unsafe basis upon which to make sweeping generaliza- 
tions concerning the general mental ifriciency of a person. In 



THE RATE AND PROGRESS OF LEARNING 171 

fact, the closeness of the agreement of the size of the two-point 
threshold with the actual amount of fatigue is too uncertain to use 
this function as a symptom of general mental or physical fatigue. 
A number of other indirect methods have been employed, such as 
the rate of tapping with a stylus, the variation in blood pressure, 
in pulse, in respiration, the range of visual accommodation, sensi- 
tiveness to pain, and so on. The same criticism applies to these as 
to the two-point discrimination. These functions may have con- 
comitant variations within rough approximations, but they are too 
distant to be precise indications of mental efficiency. 

The use of the ergograph as developed by Mosso and his co- 
workers has probably been the most successful and useful method 
for studying problems of fatigue. As such it is, however, a direct 
method for investigating muscular work and fatigue and only a very 
indirect and doubtful method for investigating mental fatigue. 

Other indirect methods of a more distinct psychological char- 
acter have also been employed. These have consisted of the meas- 
urement of certain mental functions at various intervals in order 
to determine how much variation there may be in these functions 
and to regard them as indications of mental efficiency in general. 
Such tests have been made upon memory, various types of associa- 
tion processes, perception as measured by cancellation tests, and the 
like. These tests have a certain superiority over those mentioned 
in the preceding paragraph since they deal at least with j)sychologi- 
cal functions, but they likewise do not directly measure the course 
of work as it actually occurs. They have, however, been useful 
in comparing efficiency in the same mental capacities at various 
points during the course of a day. 

A considerable num.ber of researches by means of cross-sectional 
test methods have been carried out upon school children as well 
as adults. Thus, for example, Sikorski ('19) tested pupils before 
and after school in writing from dictation, and compared the 
number of errors made. Bolton ('02) measured the memory span 
for digits during the early and the later part of the school day. 
Laser ( '94) made a test with pupils in addition and multiplication 
at hourly intervals, Friedrich ('97) tested 51 pUpils in addition, 
multiplication, and in dictation exercise at hourly periods. Eb- 
binghaus ('97), with the aid of the teachers, gave tests at hourly 
intervals in immediate memory of numbers, in addition and multi- 
plication, and in supplying words and syllables omitted from sen- 
tences. Ritter (1900) used tests in dictations of words, numbers, 



172 



EDUCATIONAL I'SVCHOLOGV 



and si-nlcncc's, and tests in cancelling letters and words. Thorndike 
(1900, '11 and '12) used, early and late in the school day, tests 
in adding, multiplying, cancelling certain words in a printed text, 
and memorizing numbers, letters, and geometrical forms. Heck 
('15) measured the i)erformances of i)upils in adding, subtracting, 
multiplying and dividing at four points during the .school day. 
Miss King ^ used tests at five points during the school day in adding 
and multiplying and in answering questions of a general informa- 
lioiial character. 

i'racticall\' all of the investigations here mentioned that were 
carried out reliably, agree, when interpreted fairly, in showing that 
efficiency in the various functions examined is changed very slightly 
or unappreciably during the course of a school day. Not all of 
the investigators, however, interpret their results in this manner. 
Thorndike has pointed out a very important misconception in the 
interpretation put by some of the experimenters upon their data, 
namely, that of counting simply the number of errors made at 
(litTerent times of the day instead of expressing efficiency in terms 
of both amount and accuracy of work done. This point may be 
illustrated in the case of Friedrich's results presented in Table 44. 



TABLE 44. After Fricdrich 

Tlu- results obtiiincfl hy Friedrich concerning the accuracy of schcxil work at 
difTcrcnt pcriixls of the day 



Time OF Test 



Figures ok Scus a.vu 



Letters, etc.. Wrtt- 
TEN IN Dictations 


PRODurTs IN Com- 
putations 


Pe» Cent 


Pep Cent 


Per Cent 


Per Cent 


RU.IIT 


Wrong 


RiGBT 


\\ rono 


90.8 


. 2 


98.9 


I.I 


9Q.6 


•4 


98 


4 


1.6 


90 3 


■7 


98 





2.0 


OQ.i 


.8 


98 





2.0 


90 4 


.6 


98 


I 


1.9 


90 


I 


97 


8 


2.2 


00 


I .0 


97 


7 


23 


99.8 


2 


gS.i 


1 .9 


99 2 


.8 


07 9 


2. 1 


99 4 


.6 


97 


2.1 


98.9 


1. 1 


97 


h 


2-4 



Morning Session: 

Before ist hour 

.\fter ist hour 

After 2nd hour and S min. rest 

After 2nd hour 

After 3rd hour and two 15 min. rests 

After 3r<i hour and 15 min. rest 

After 3rd hour 

Afternoon Session: 

Before isl hour. . . 

After ist hour 

.After 2nd hour and 15 min. rest 

.\fter 2nd hour 



' .\ii unpublished study rcjxjrlc*! by Thorndike. ('14, III, p. yj). 



THE RATE AND PROGRESS OF LEARNING 



173 



If in this table we compare simply the percentage of errors the 
efficiency of the pupils was over five times as great at the beginning 
of the school day as at the end in the dictation test, and over two 
times as great in the computation tests. If, on the other hand, we 
consider the column giving percentage right we find that the ef- 
ficiency changed but very little. 

To point out further how inconsiderably the performance of 
pupils changes in the course of a day we may note the following re- 
sults from Heck: 





TABLE 45. After Heck ('i 


3) 




Time of Test 


Units of Work Done 


Per Cent Correct 


9:10 a. m. 

11:05 3.. m. 

1:10 p. m. 

2:30 p. m. 


140.37 
142.57 
142.67 
143.68 


87.40 
86.08 
86.17 
85.46 



The amount of work increased slightly while the accuracy de- 
creased slightly from the first test to the last. 

(2) Direct methods. The most fruitful direct methods of measur- 
ing continuous mental work have been the various types of mental 
calculations, particularly addition and multiplication. These 
methods have been used by Krapelin, Thorndike and Aral, Starch 
and Ash, and others. As an illustration of one type of mental 
addition, the writer has used a method consisting in the mental 
addition of 6, 7, 8, and 9 in rotation by beginning with a given 
number and adding each of these numbers in turn to the answer 
last obtained, as described in Chapter XVI, Experiments ('17). 
The advantage of this form of calculation is that it affords suf- 
ficient difi&culty and thus fully taxes the efforts of the individual 
and makes possible a minute record of the amount and accuracy of 
work done during succeeding short intervals of time. Figure 52 
shows a curve obtained by this method, covering a period of con- 
tinuous work of two hours. 

As an illustration of mental multiplication, we may cite the 
experiment carried out by Miss Aral under the direction of Thorn- 
dike. She used the method of multiplying mentally four place 
numbers by four place numbers, as 4,962 times 7,584. She trained 
herself for a considerable period of time in this type of mental mul- 
tiphcation in order to reach an approximate limit of practice. Then 
she did the following experiment: 



174 



EDUCATIONAL reVCHOLOGY 



"On March 3, 4, 5, and 6, that subject did the mental multiplication 
from II A. M. to 11 V. M. without any pauses except the two or three 
seconds between the examples lor recording time. But the subject had 
taken a heavier breakfast than usual at 10 A. M. and a light supper 
after 1 1 P. M. Her health was in good condition and she slept stmndly at 
night. The contents of her consciousness during the experiments were 
ver>' simple, all desires being completely subjected to the one desire to 
get true fatigue curves." (Aral, '12, p. 37.) 

The remarkable result of all experiments with purely mental 
functions has been that mental efficiency is reduced only very 
slightly even after two or more hours of very difBcult, uninterrupted 




Fir,. 52. — Mental work rurvc. Up[>er curve shows number of additions 
made per half minute period. Lower cur\c shows number of errors made. 
Work was continued for two hours. After Starch and Ash (.'17). 



work. Thus in the curs'e, Figure 52, the reduction in the numl^er of 
additions made per thirty seconds, was only irom 14.0 down to 
13.4, or a loss of only 4.3%. Arai found even in the course of 12 
hours of such ditTicult mental multijilication as she carried out, 
that her efllciency was reduced only by about one-lialf. Other 
investigators have shown in general the same facts. 

Seashore and Kent ('05) measured continuously, for as long as 
two hours, the threshold of hearing b}' recording the audibility and 
inaudibility of a sound varied about the limen. The intensity of 
the sound was changed at a uniform rate. As soon as it became too 
faint to be heard the subject gave a signal to the experimenter who 
at once increased the strength of the sound. As soon as it could be 
luard again the subjct I again respondi'd. Then the sound was de- 
creased again, and s<» on without break. A .siimple curve is shown 
in Figure 53. Ten records were obtained which showed that "con- 



THE RATE AND PROGRESS OF LEARNING 



175 



tinuous liminal or moderately faint sounds do not seem to lower the 
efficiency of the ear in a two hour test" (p. 100). 

It would seem, therefore, on the basis of experimental work, that 
fatigue in the sense of decrease in product achieved is practically a 
negligible element in school work. The actual capacity to do work 
with the same degree of accuracy is practically undiminished in the 
course of a school day. Such symptoms of fatigue as have been 
frequently enumerated in pedagogical writings, are apparently only 
superficial signs of monotony, of lack or diminishing of interest, 
or of being bored by school work, and not actual signs of loss of 
capacity to do the work. Such statements as "I simply cannot 
work any longer" made after a half or whole hour's work, are il- 
lusory and probably signify chiefly a weariness with the work 







.10 



20 



30 



80 



90 



100 110 



40 50 60 70 
Minutes 
Fig. 53. — Continuous record of the measurement of the threshold of hearing. 
After Seashore and Kent (.'05). 

which, if it must be kept up by force of conditions, can usually be 
continued without difficulty or harm and usually without being 
seriously boresome. 

The feeling of interest or satisfaction in doing work does decrease 
very materially as the work goes on. Thorndike (17) for example 
found that the satisfyingness of such work as grading compositions 
decreased in the course of two hours to about one-half and in the 
course of four hours to about one-third of the amount of satisfac- 
tion at the beginning of the period. 

The feeling of weariness, from the practical side of school activ- 
ities as well as of mental work generally, is, however, an important 
item. In a certain sense it is a real thing. Even if it is illusory 
it does interfere with the smooth continuation of work. But it is 
very likely a less serious situation than an actual loss of capacity 
to do work would be. Practically it resolves itself into a problem 
of maintaining interest rather than relieving depreciation of ef- 
ficiency. 



CIIAl'lKR XII 
now TU STUDV 

Waste in Studying. Since studying is learning under school 
conditions, it Avould seem worth while to make such suggestions 
as can be made concretely to assist pupils in this inifwrtant phase 
of the psychology of learning. It may seem preposterous to give 
advice about something concerning which each pupil is presumably 
]>roficient after years of jiractice in it, and furthermore to attempt 
to gi\'e suggestions on studying may seem to many to be nothing 
more than an "unprofitable delineation of the obvious." It is, 
however, A'ery certain that there is an uncalculatcd waste of energy 
and a still more prodigal waste of time in so-called studying. If 
we may judge from the possibility of improvement in reading ca- 
j)acity alone, and from the larger accomi)lishments attained under 
favorable conditions of work, we may A-enture to guess that the 
average student could accomplish his work just as efficiently or 
more efllciently, in two-thirds, or less, of the amount of time 
ordinarily consumed, by developing more economical methods 
and habits of studying. Impro\ements in j^ropcr procedure in 
studying have showTi how much more may be accomplished in the 
same length of time or even in a shorter ])criod of tinu*. Vicious 
habits of dawdling in school work are acquired, which may have 
their permanent elTcct throughout the indiNitlual's life. 

Is Studying Worth While? This tjue.stion is Avorth raising in 
view of the belief, prexalent among students, parents, and grad- 
uates, that after all it does not matter much whether a j^ujnl docs 
well in his studies or not, that the boy who does pot)rIy in the 
grammar grades or the high school will outgrow his negligence and 
come into his own when he gets into his college or professional 
course, or that when he gets into the real business of life he will 
outstri]) his more studious mates. To what extent are these beliefs 
true or false? To what extent is early scholastic perlormance 
indicative of similar or different performance later on? To what 
extent is scholastic ])erformance prophetic of performance in life? 

A considerable amount of statistical material has been accumu- 
lated in the atteni])t to answer these questions. Some of this 

176 



HOW TO STUDY 177 

material was presented in the latter part of Chapter IV under the 
heading "Correlations between Early and Later Mental Abilities." 
These correlations were found to be high. Dearborn, for example, 
found that of 472 pupils, whose records were traced through the 
high school and college, only two who were in the lowest quarter in 
the high school rose to the top quarter in the university. It should 
be noted further that these two were just barely poor enough in 
the high school to be classed into the lowest quarter and that 
they rose just barely enough to get into the top quarter in the uni- 
versity. The chances that the pupil who is doing poor work in the 
high school will later come into his own are exceedingly small; 
apparently he has been in his own all along, or, if not, he had better 
have got into his own as soon as possible. 

President Lowell ('10) made a study of the records of the grad- 
uates of. Harvard College for a period of twelve years. He found 
the following situation : 

Men graduating with various Percentage graduating with distinction from 

honors The Law School The Medical School 

A. B.'s with highest honors 60 92 

A. B.'s with great honor 40 87 

A. B.'s with honor 22 76 

A. B.'s without honor 6}4 36 

A. B.'s without honor, of men who 
had entered college with condi- 
tions 3 

The 250 Yale men who graduated from the Harvard Law School 
in 1900-1915 were divided into nine groups according to their 
scholarship at Yale. These nine groups, with the exception of one, ' 
finished the Harvard Law School in the same relative order of 
scholarship that they had held at Yale. 

To many persons a more important problem is the relationship 
between scholastic attainment and success in business or profes- 
sional work. Foster ('16) has summarized in an interesting manner 
much of the evidence pertaining to this problem. He made a 
study of the Harvard College class of 1894. He asked three men, 
the dean of the college, the secretary of the alumni association, 
and a member of the class, to name the most successful men of the 
class. They were free to use their own interpretation of success 
except that they were not to include men whose success appeared 
to be due chiefly to family wealth or position. The three judges 
agreed on twenty-three men. Foster then obtained their records 



17S EDUCATIONAL PSVCIIOLOGV 

in college and comjKirecI them with the records of twenty-three 
other men chosen at random from the same class. The former had 
nearly four times as many highest grades as the latter, namely, 
196 A's as compared with 56 A's. By a similar plan three judges 
selected the most successful men among the graduates of the 
University of Oregon for the period of 1878 to 1901. Of the grad- 
uates designated as successful, 53% had been good students and 
17% had been weak students. Of the graduates designated as 
unsuccessful, 12% had been good students and 52% hiid been weak 
students. 

A study of tlie alumni of Wesleyan University showed that of 
the living graduates for the period of i860 to 1889, 50% of the men 
who hud graduated with honors were listed in Who's Who, and 
only 10% of the men who had graduated without honors were in 
Who's Who. Among the li\ing graduates for the i)eriod of 1890 
to 1S99, 60% of the men graduated with highest honors were listed 
in Who's Who, 30% of the men elected to Phi Beta Kappa were 
listed in Who's Who, while only 11% of the graduates without 
suju-rior scholarship were found in Who's Who. (Nicholson '15). 

E. G. Dexter investigated the records of the living graduates of 
twenty-two colleges and found that 5.9% of the honor scholars 
and only 2% of all graduates were listed in Who's Who. Further- 
more, 56% of the Yale valedictorians were found in Who's Who. 
Their chances were, therefore, more than twent}'-fivc times as 
great as those of other graduates. The records of 13,705 living 
graduates of two New England colleges revealed the fact th:it 
5.4% of those who constituted the highest tenth were listed in 
Who's Who while but i.S^'n of those in the fourth tenth were there 
listed. Who's Who is, of course, not an absolute criterion of suc- 
cess; it is, howe\-er, a rough measure of success. 

A tabulation of the Oxford University men who entered the law 
or the ministry showed the following percentages of men who 
attained distinction in their respective jirofessions: 

Mrn w ith varying,' honors rcrrcntaRCs attuininK distinction 

In the law In the ministry 

Men with 1st class honors 4')' i, 6S% 

" " and " " ^ ^ 37 

" '' 3rd " " J.' 32 

4th " " . 20 JQ 

" " pass dcRrrcs . . 16 21 

" " no degrees. ... 15 9 



HOW TO STUDY . 179 

As a matter of correct interpretation of these extensive statistics, 
it must not be assumed that success or failure is solely attributable 
to the amount of devotion to school studies. The uniform manner 
with which the early scholastic records agree with the later records 
of the same persons, or the pronounced tendency with which scho- 
lastic attainment correlates with business or professional attain- 
ment is probably due to a common cause, namely, original ability -• 
or make-up of the individual. At their face value, these figures 
mean that the person who does well in his school work also tends 
rather strongly to be a person who will do well in his business or 
professional work. However, this array of facts is impressive and 
ought to be brought emphatically to the attention of high school 
and college students. They ought to have a tonic effect upon their 
efforts. While our native make-up deterrnines to a large extent 
what we shall become, yet rarely does any one utilize or develop 
to the fullest extent even the limited measure of ability that he 
possesses. The laggard can find little consolation in the hope of 
somehow redeeming himself later on. 

Types of Studying. For the sake of convenience, we may divide 
studying into three types: 

1. The Reading Type of Studying. In the elementary school 
probably eight-tenths and in the high school and the university 
probably two-thirds of all studying consists essentially in reading. 

2. The Laboratory Type of Studying. This type obviously 
consists of the manipulation of apparatus, the observation of 
material, the recording of observations and experimental data, and 
the interpretation of these data. 

3. The Analytical or Reasoning T>pe of Studying. This type 
is involved in those subjects in which the amount of reading is 
relatively little, but in which the task consists in a thorough mas- 
tery of a relatively small amount of text. Such studying is ob- 
viously involved in mathematics and in a few other types of 
difficult reading, as for example, certain branches of philosophy 
and the speculative and theoretical aspects of the sciences. 

Problems. Every type of studying is different and, in a sense, 
every lesson has its own special material and presents its own 
problems on how to study effectively. It may seem futile to at- 
tempt to give general advice on how to study. Yet upon further 
analysis, it appears that there are several elements common to 
all t}pes of learning. These elements are (i) the control of atten- 
tion in securing the most favorable attitude of work, which would 



I So KDUCATIONAL PSYCHOLOGY 

l)i- in\olvcd in all t}'])is of menial work, (2) common principles in 
the assimilation and retention of the material, and (3) proficiency 
in reatlinj^. Problems involved in all of these elements would be: 
l-'irst, what are the specific processes common to all types of study- 
ing here referred to, and second, how may these various processes 
be facilitated? 

Control of Attention. One of the chief, if not the chief, source 
of waste in studying and in fact in all mental work, is the reluctance 
in beginning, an intellectual task. There seems to 'be in many in- 
dividuals an almost insuperable inertia to overcome before work 
is, or can actually be, begun and continued without constant self- 
])ushing. The common feeling is a dislike to begin work. "I 
don't like to study my history," or "I just hate to write this theme," 
"I don't see why he makes us do this," represent states of mind 
frequently found among the average pupils and to some extent 
even among the better pui)ils who often have severe struggles with 
such a tendency. 

In papers on "Difliculties and Hindrances in Studying and 
How to Overcome Them" collected by the author from about 
one hundred university juniors and seniors, 56 mentioned lack of 
concentration, 26, dislike for or lack of interest in the subject, 
23, getting started, 9, mind-wandering, 5, failure to organize mate- 
rial, and 4, day dreaming. These ma}' all be classed as internal 
psychological difliculties centering around the problem of getting 
the mind to work at the task. Practically every student mentioned 
one or another or several of these four din'iculties. 

Besides attributing this situation to indolence or to stupidity, 
is there anything in the way of concrete suggestions and help that 
can be gi\en to overcome this mountain of dilliculty? I believe 
there are two general procedures which may be followed. One is to 
prit one's teeth and to "go to it"; that is, simply to force oneself 
by shetr \()luntary elTort to begin the task. The other is to put 
oneself into ])hysical surroundings and into a frame of mind in 
which it will recjuire a minimum, or at least a smaller amount, of 
'voluntary effort. Strictly voluntary etTort consumes a large 
amount of mental energy and, if it must be continued for a long 
time, is very wasteful of one's strength. The second is distinctly 
the more advisabli- plan to adopt. With the help of such a control 
of e.xtemal con(litii>ns as is ]H)ssible, the following means of di- 
recting one's energy may, therefore, be suggested: 

(i^ Put yourself into the proj)er physical or bodily attitude of 



HOW TO STUDY iSl 

work. Sit up to your desk or table at which you customarily work: 
This in itself will help to start the mental machinery agoing and 
make it easier for the mental processes to operate. 

(2) Work in surroundings in which there are absolutely no dis- 
tractions as far as possible. Some persons can work under very 
distracting conditions, but these are exceptions, and if one has 
difficulty in beginning work, he should go alone into a separate 
room, shut the door, and sit facing away from the windows, and 
have nothing to look at or to attract his attention. A certain life 
insurance agent of one of the largest companies in America adopted 
the plan of selling to no one except by special appointment in his 
own office from which all possible distractions had been removed. 
There was nothing on the walls and nothing in the room but a desk, 
a telephone, and a couple of chairs. There was nothing on the 
desk except a life insurance policy, which was placed there at a 
certain time of the interview. The purpose was to secure condi- 
tions under which there were absolutely no distractions what- 
ever, and the only thing to think about was the purchase of a life 
insurance policy. For a time there was a calendar of the com- 
pany hanging above the desk. He found that many clients would 
remark, upon leaving the office, about the interesting dates des- 
ignated on the calendar. There was nothing else to distract their 
attention and consequently these stood out in the minds of the 
clients, and, therefore, appeared interesting. He then removed 
the calendar to a rear wall so that even the dates might not dis- 
tract. All these features were a part of his carefully prepared 
sales plan. This man was one of the most successful life insurance 
salesmen among all the agents of that company. In a certain 
month he had the record of selling the largest number of policies 
of all the salesmen of this large company — a record that was 
achieved after only eighteen months of experience in selling life 
insurance policies, immediately after graduation from college. It 
would, of course, be absurd to attribute his remarkable success 
to this one element, but it was nevertheless a very important part 
in a carefully prepared plan of salesmanship. 

The removal of distractions, or what amounts to the same thing, 
the selection of a place for study where there are no distractions, 
is one of the most useful suggestions that anyone can adopt for 
developing concentration in work without a constant and exhaust- 
ing tax upon the worker's voluntary efforts. In the course of time, 
it may be possible to work under even distracting circumstances. 



l82 EDUCATION.\L PSYCHOLOGY 

but probably no one, except the rare, absent-minded genius, 
can work as well among distractions of sights and sounds, and 
in the presence of other peo}jle, as under the complete absence of 
such stimuli. No one is in a position to appreciate the great ef- 
fectiveness in intellectual work under complete absence of dis- 
tractions until he has tried it. The average pupil wastes an in- 
estimable amount of time by having to stud}' in the presence of 
other members of the family who may be conxersing or moving 
about, and every word or action or stimulus of any sort is bound 
to enter the mind and to divert the association ])rocesses to some- 
thing else. E\-en though they are ver}' minor, the}" require a few 
seconds, if not longer, to cause the thought process to return again 
to the subject-matter to be studied. On!}' those j^ersons who have 
compared their own working efliciency under distracting condi- 
tions with their efTKicncy under ideal conditions can appreciate 
the enormous difference in the amount that am be accom- 
plished. 

(3) Begin work. Don't continue to think, "Oh, I just hate to 
do this," but instead go to your desk in your secluded room, sit 
dowTi, take hold of book, paper, pencil, or whatever may be needed, 
and begin to write, or read, or figure. In short, if you have diffi- 
culty in overcoming inertia, just begin to go through the motions 
of work. This will automatically start the mental processes going 
relative to the work to be done, and before you realize it, you will 
be in the midst of the task, reading, thinking, and writing in an 
interested manner concerning the problems at hand. The external 
mechanical movements will act as stimuli for the inauguration of 
associative processes, and arc likely to start mental activities 
without a great deal of voluntary effort. 

A prominent story WTiter relates that he had difficulty in be- 
ginning his wTiting and in working out his plan necessary to finish 
up the details after the plot of the story had been conceivdd. This 
aspect of story writing is work and jjrobably not a matter of in- 
spiration; it involves close application and sometimes drudger}'. 
He found that he was able to get into his writing by simply sitting 
down, taking a pencil and j^aper, and beginning to write whatever 
came to his mind, whether it was very pertinent to his story or 
not. Going through ihc motions started his thought pnKesses 
agoing, and very shortly his associative and imaginative proc- 
esses were almost automatically producing pertinent and excel- 
lent ideas. 



HOW TO STUDY 1 83 

In like manner, begin to study a lesson by taking the book, turn- 
ing to the page, and simply looking at the print. Some voluntary 
effort, of course, must be exercised, if only to take hold of the book, 
but it is more economical to do so than continuously be thinking 
"How I hate to do it." This thought will automatically be driven 
out by the processes started by simply going through the motions 
of beginning work. The more voluntary effort and force one may 
be able to exercise in not thinking about dislike for the task and 
in beginning the motions of the work, probably the better and 
the sooner one is able to start, but this voluntary attention should 
normally pass very quickly into automatic attention and interest. 

In the papers previously mentioned, the students stated that 
they overcame their difficulties of going at their tasks and keeping 
at them, besides "exercising will-power" which was mentioned 
most freciuently, by "setting a certain hour to begin," by "doing 
work in a limited time," by "doing the work under pressure," by 
"dividing number of pages so that they could tell how many would 
have to be read every fifteen minutes," by "copying a sentence 
which helps to get the mind on the subject," by "starting directly 
for if I wait at all a million things would come up which were more 
interesting," by "having a definite schedule of study," by "plan- 
ning the day," by "repeating with lips what is read," by "read- 
ing aloud," and by "studying in one place." 

Common Elements in the Assimilation and Retention of the 
Material. At least five or six specific suggestions applicable to 
any kind of studying may be given. 

(i) Take a problem solving attitude. Know definitely what 
you want to find. Ask questions and then look for the answers. 

(2) Understand what you want to assimilate and retain perma- 
nently. To go through reading matter in a perfunctory manner 
will not cause retention of it except after long, wasteful, and fre- 
quent re-readings. A certain psychologist, in conducting experi- 
ments in memory with words and syllables, had dictated over and 
over a great many times certain series of materials, so that they 
had been completely memorized by several subjects, but he himself 
was unable to repeat the material from memory. The reason was 
that he himself had never paid strict attention to the memorizing, 
and had read them over and over again purely in a passive, in- 
attentive manner. 

Do not try to memorize ideas blindly; memorize understandingly. 
Some material in school must be memorized mechanically, but 



184 EDUCATIONAL I'SVCHULOCJV 

much more (jf it can be learned with a thorough conception of its 
meaning. 

(3) Organize ideas with reference to certain larger ideas and prin- 
ciples. Organize your ideas and think out their relation to general 
principles. Grasp in as large units as possible and note the rela- 
tion of details. 

In the writer's Experiments (17), page 190, is given an ex- 
periment in which two scries of facts of apparently equal kind 
and difTiculty are presented for memorizing. Each list is com- 
posed of five dates of history, five Greek words with their 
English meanings, and five numbers with their sums. In one 
list these ideas are arranged in miscellaneous order; in the 
other list they are grouped by subjects, the five historical dates 
are in one group, the five Greek words in the second group, and 
the five sums in the third group. The time required by ten sul)- 
jects for memorizing the first set was an average of 14 minutes 
and 3 seconds; the time required for memorizing the second series, 
which was arranged in order, was on the average 9 minutes and 1 1 
seconds. The comparison shows a \er>' decided advantage in 
favor of learning the material in organized form. 

(4) Recall at brief intervals the essential ideas of what you have 
read. Stop at the end of each paragraph or two, shut your book 
or your eyes, and recall the essential ideas you have read. Say to 
yourself "What did I rcafl about?" Then try to answer the ques- 
tion. Note here what was said about forgetting in the last chapter. 
The chief value of examinations is the occasion and stimulus which 
they alTord for recalling and organizing the material covered. In 
some respects the most valuable studying done by pupils is done 
in j)rej)arali()n for examinations. The value of the principle of 
recall in learning or memorizing has been thoroughly demonstrates! 
by laboratory experiments. 

Then each day or two, relate the recent material in a given 
subject to the earlier material in that subject. That is, review in 
your mind at short inter\'als, the larger essentials of all the ma- 
terial covered up to date. The princij)le of recall in this form is 
used far too little in studying. These suggestions would be applica- 
ble to every tyi)e of reading which has to be dt)nc rather carefully. 
It would, of course, not be advisable to do so in materials such as a 
novel in which the ideas in detail need not be retained. 

(5) At the earliest possible moment and as frcf|ueiitly as {xissi- 
ble, use the ideas that ha\e been acquired, either by telling them 



HOW TO STUDY 185 

to some one else, or by thinking them over in your mind in con- 
nection with other related materials or situations. This will give 
them meaning in new ways and from new angles, and will help 
to fix them permanently by virtue of the principle of recall. 

(6) In committing material to memory, learn by wholes rather 
than by parts. Poetry or prose can, as a rule, be memorized more 
quickly if the material is read through as a whole from beginning 
to end than if it is memorized in small sections of two or three lines; 
and what is more important, when this method is employed, the 
retention is more permanent. With many persons who are ac- 
customed to memorizing by the part method, there is frequently 
no saving of time in the first learning partly because the whole 
method is new to them and partly because the learner often doubts 
the advisability of using the whole method. 

There are three reasons why the whole method proves in the 
long run to be more economical: (i) Learning by parts establishes 
many useless and interfering connections. Thus in committing 
the first two lines of a poem the association is established between 
the last word of the second line and the first word of the first line. 
But this is not the order in which the lines are to be recalled. 
Rather the connection should be established between the last 
word of the second line and the first word of the third line as is 
done in the whole method. Consequently every portion memorized 
by itself forms at least one detrimental connection and in a long 
selection a very considerable number of such associations are 
formed. These derailing paths probably account for the fact that 
pupils in recit/ing a poem become stalled usually between the 
portions learned piecemeal. (2) Reading the material over as a 
whole gives a view of the entire selection and will serve to give 
meaning and correlation of the parts in the whole. It will help 
to organize the ideas as a whole. (3) Learning by parts is apt to 
produce great unevenness among the various portions of the ma- 
terial in the degree of perfection of the memorizing. Some parts, 
especially the earher ones, will be repeated needlessly a great many 
times and result in much greater over-learning of those parts than 
of other parts. One point of caution in using the whole method 
should, however, be noted. When the learner reads over the entire 
selection to be memorized he does not make much visible progress 
until, after a sufficient number of repetitions, he is able to repro- 
duce most of the material. This situation is likely to be discourag- 
ing, particularly to children. Perhaps the most effective manner 



io6 EDUCAIIOXAL PSYCHOLOGY 

of t'niplo\ iiig the wliolr method is to learn the material in rel;iii\t ly 
large sections instead of as a complete whok-, parlicularly if tin- 
selection is very l«>n<,^ 

Improvement in Reading Ability. Jhe average child, as well 
as the average adult, reads far tt)o slowly, and in fact, far more 
slowly than he is capable of reading. About one- fourth of univer- 
sity students read less rapidly than the average 8lh grade pupil 
does, and about one-fourtli of SUi grade pupils read less rapidly 
than the average 5th grade pupil. Experiments indicate that by a 
moderate amount of defmite i^ractice, with conscious elTort to 
improve, the speed of reading may be increased from 50% to 
100% without loss in the comprehension of the ideas read. The 
moral would be: Force yourself to read more rapidly, which will 
be accompanied by greater concentration of attention and in the 
course of time this more rapid reading will become habitual, so 
that the comjjrehension will be just as complete as at the slower 
rate of reading. Consult the latter part of the chapter on "Read- 
ing" for a more detailed discussion of these points. 

Concrete Rules for Studying. Whipple has presented a series 
of thirty-eight rules which ought to prove valuable for increasing 
effectiveness in studying. Some of these rules involve jxjints that 
have been previously presented in this chapter. Their specific 
character makes them commendable for the student's considera- 
tion and observance. 'I'hey are as follows: 

SUMiMARY OF RULES. After Whipple ('16) 

1. Keep yourself in good physical condition. 

2. Attend to, remove or treat physical defects llial often handicap 
mental activity, such as defective cyesif^ht, defective hearing, det'ectivc 
teeth, adenoids, obstructed nasiil breathing. 

,3. See that external conditions of work (light, temperature, huniidily, 
clothing, chair, desk, etc.) are favorable to stuily. ^ 

4. Korm a place-study hal)it. , 

5. I'orm a time-study habit. " 

6. When possible, prepare the advance assignment in a given subject 
directly after the day's recitation in it. 

7. Begin work promptly. 

8. Take on the atlitudc of attention.. 

0. Work intensely while you work: Concentrate. 

10. But don't let intense application become tlus[cr or worry. 

1 1. Do your work with the intent to learn and to remember. 
\2. Seek a motive or, better, several motives. 



HOW TO STUDY 187 

13. Get rid of the idea that you are working for the teacher. 

14. Don't apply for help until you have to. 

15. Have a clear notion of the aim. 

16. Before beginning the advance work, review rapidly the previous 
lesson. 

17. Make a rapid preliminary survey of the assigned material. 

18. Find out by trial whether you succeed better by beginning with the 
hardest or with the easiest task when you are confronted with several 
tasks of unequal difficulty. 

19. In general, use in your studying the form of activity that will 
later be demanded when the material is used. 

20. Give most time and attention to the weak points in your knowledge 
or technique. 

21. Carry the learning of all important items beyond the point neces- 
sary for immediate recall. 

22. You must daily pass judgment as to the degree of importance of 
items that are brought before you, and lay special stress on the per- 
manent fixing of those items that are vital and fundamental. 

23. When a given bit of information is clearly of subordinate im- 
portance and useful only for the time being, you are warranted in giving 
to it only sufficient attention to hold it over the time in question. 

24. Make the duration of your periods of study long enough to utilize 
"warming-up" but not so long as to suffer weariness or fatigue. 

25. When drill or repetition is necessary, distribute over more than 
one period the time given to a specified learning. 

26. When you interrupt work, not only stop at a natural break, but 
also leave a cue for its quick resumption. 

27. After intensive application, especially to new material, pause for a 
time and let your mind be fallow before taking up anything else. 

28. Use various devices to compel yourself to think over your work. 

29. Form the habit of working out your own concrete examples of all 
general rules and principles. 

30. Form the habit of mentally reviewing every paragraph as soon as 
you have read it. 

31. Don't hesitate to mark up your own books to make the essential 
ideas stand out visibly. 

32. Whenever your desire is to master material that is at all extensive 
and complex, make an outline of it. If you also wish to retain this 
material, commit your outline to memory. 

33. In all your work apply your knowledge as much as possible and as 
soon as possible. 

34. Do not hesitate to commit to memory verbatim such materials as 
definitions of technical terms, formulas, dates and outlines, always pro- 
vided, of course, that you also understand them. 

35. When the material to be learned by heart presents no obvious 



l88 EDUCATIONAL PSYCHOLOCiV 

rational associations, it is perfectly legitimate to invent some artificial 
scheme for IcarninK and recalling it. 

36. In commilling to memory a poem, declamation or oration, do not 
lireak it up into parts but learn it as a whole. 

37. In committing to memory, it is better to read aloud than to read 
silently and better to read rapidly than slowly. 

3S. If your work includes attendance at lectures, take a moderate 
amount of notes during the lectures, using a system of abbreviations, and 
rewrite these notes daily, amplified into a reasonably corapendious out- 
line, organized as suggested in Rule 32. 

Supervised Study. Teachers have come to recognize in recent 
years the waste of time and the blind direction of energ>', or pos- 
sibly lack of energy," in so much of the studying done by pupils 
that a widespread movement has gotten under way for the super- 
vision of studying. The plans for super\'ising studying are carried 
out in so many dilTerent ways that hardly any one plan can be 
designated as typical. The results accruing from the general ef- 
forts in this direction have been in most cases beneficial. Con- 
tinued experimentation during the ne.\t few years with various 
plans of super\'ised study will lead to a more general agreement 
as to the most effective manner of administering it. 

In a recent inquiry of supervised study in schools on the Pacific 
coast, Proctor ('17) found that forty-two high schools employed 
it in one form or another. Of these forty-two schools, thirty-one 
reported the use of a lengthened period distributed as follows: 

(a) 60' ptriod, divided 30-30, No. of cases 3 

(K)' iieriod, divided 35-25, No. of cases i 

60' j)cri(Ki, divided 40-20, No. of ca.ses 15 

60' periix], (livificd 45-15, No. of cases i 

63' period, divided 33-30, No. of cases i 

2 1 

(I)) 70' i>cri(Ki, divided 40-30, No. of cases 4 

70' i)criiKl, divided 35-35, No. of cases 2 

— 6 

(c) 80' period, divided 40-40, No. of cases . i 1 

(d) 85' |K.Ti(Kl, divided 45-40, No. of cases 2 3 

(e) 00' perifxi, divided 45-45, No. of cases i 1 

Tolal. . . 31 

Regarding the cfTects of supcr\ised study. Proctor reports that: 

"Twenty six of the 31 print ipals employing the lengthened period 
Slid that stufly habits had Ixen improved; one could discover no apparent 



HOW TO STUDY 



189 



effect; two said that only the slow students had been helped, the brighter 
ones were not ; and two had no data on which to base their opinions. 

"Wherever the plan had been in use long enough to make possible the 
compiling of statistics as to the effect of supervised study on scholarship, 
there was practically unanimous agreement that the number of failures 
had been reduced and the standards of scholarship had been raised. The 
high school at Snokomish, Washington, reports that the average per- 
centage of failures in elementary algebra for the two years prior to the 
adoption of supervnsed study was 28%. But for the two-year period 
following the adoption of supervised study the failures in the same 
subject were reduced to 17%. Hoquiam, Washington, reports that the 
average marks of the students range 10% higher and that the number of 
honor pupils has been doubled since supervised study was introduced. 
The principal of the Areata high school, California, reports that the 
average mark of the freshman class has been raised from 78% to 82^/2% 
during the first year of supervised study. Santa Cruz, California, com- 
paring the year 1914-15, the last under the old plan, with the year 
1916-17, the second year under supervised study, finds that the increase 
in the total number of high marks has been 157%; the decrease in low 
failures, 188%. Reno, Nevada, reports a decrease of 45% in the number 
of failures, and an increase of 24% in the number of students making 
excellent marks." 

J. Stanley Brown, principal of the high school at Joliet, Illinois, 
reports, as quoted by Hall-Quest ('17), a decided reduction in the 
percentage of failures after the introduction in the high school of 
supervised study, as indicated in the foUowdng table: 



TABLE 46. After Brown and Hall-Quest ('17, p. 386). Supervision of study 
apparently was begun in 191 2 although I have not been able to tind a 
definite statement by Hall-Quest to that effect. 

Table of percentage of failures 



Subject 


igii 


1912 


igi.s 


1914 


Algebra 


24 
26 
29 
21 

22 
10 
12 


22 
20 

19 

20 

19 

9 

10 


15 
12 

17 

13 

16 

8 

8 


12 


Arithmetic 


13 
16 

14 

13 

9 

9 


Geometry 

German. . . 


Latin 


French 


Physiography 



Breslich ('12) made an experiment to determine the effect of 
directed study by dividing an algebra class into two sections, one 



HjO 



LDUCAIIONAI, rs\CH()LO(;V 



of which was comluclal in ihc usual manner of recitation work 
and home study, and the other was conducted by confming all of 
the work to the recitation period of 45 minutes. This time was 
devoted partly to study under the supervision of the teacher and 
partlv to recitation work. The two sections were made uj) of pupils 
of approximately equal ability as indicated by the marks for the 
preceding semester's work, which averaged 81.4 for those who 
constituted the home study group, and 79.4 for those who con- 
stituted the supervised study class. The home study class de- 
Aoted approximately two hours to each lesson including the 45 
minutes for the recitation period. The experiment was conducted 
for a period of fourteen lessons. At the end of that time, the same 
examination over the work that had been covered was given to 




Weeks 

Fig. 54. — The broken line represents the supervised group. The continuous 
Hnc represents the unsupervised groui). After Minnick ('13). 

both sections. The supervised study class made an average of 65.5, 
and the home study group made an average of 62.S. Hence 
the supervised study class obtained as good results as the home 
study class, or slightly better, in spite of the fact that the former 
spent only about two-fifths as much time upon the work. 

Minnick ('i,0 of Bloomington, Indiana, divided a group of thirty- 
six j)Uj)ils in plane geometry into two divisions, j)roviding suptr- 
\ ision of the study to one division and none to the other division 
for a period of fifteen weeks. The results show an advantage in 
scholarshif) for the supervised group as represented in the curves 
of Figure 54. 

Hall-(Juest quotes similar results of impro\ement in scholarship 
as indicated either in the reduction of the percentage of failures 
or in the higher scholastic marks as reported by Loveland at Potts- 
town. Pennsylvania, by Rickard, at Oakland City, Indiana, and 
from tin- high school al Pueblo, Colorado. 



CHAPTER XIII 
TRANSFERENCE OF TRiMNING IN SPECIFIC MENTAL FUNCTIONS 

Problems. To what extent does training in one mental function 
or set of functions modify the operation of other mental functions? 
To what extent will training in mathematical reasoning modify 
reasoning ability about political events or bargains, or vice versa? 
To what extent will training in remembering faces and names 
modify the remembering of the prices of goods or words in a lan- 
guage, or vice versa? To what extent will improvement in pro- 
ficiency in such mental capacities as are involved in school studies 
modify proficiency in any other specific or general activities or 
interests in life? These problems open up all the ramifications of 
the traditional controversy concerning mental discipline, the educa- 
tional value of school subjects, and their related discussions. The 
fundamental problem, however, is not, does training transfer? 
The task is more complex and suggests rather the following three 
fundamental problems: (i) To what extent does training transfer, 
(2) To how closely or how distantly related functions does training 
transfer, and (3) How does the transfer take place? Thus we see 
that transference of training is one of the three or four most im- 
portant perennial problems in the entire field of education. So 
many problems in the administration of schools, in the construction 
of courses of study, in the emphasis upon various aspects of school 
subjects, in fact, the ultimate values of education as a whole, de- 
pend primarily upon our attitude toward the problems of mental 
discipline and transference of training. 

In order to think about these matters clearly, it is' necessary 
to distinguish between two quite different aspects of the discussion, 
namely, (i) The pure disciplinary or training value in the improve- 
ment of a mental function irrespective of the material through 
which it is trained or which is acquired in the training; (2) the 
intrinsic value of the information or material acquired in the proc- 
ess of the training irrespective of the training referred to in (i). 
The one is acquisition of training; the other is acquisition of con- 
tent. The difference between these two aspects may be illustrated 
thus: The learning of shorthand will furnish practice of certain 

191 



1Q2 KDUCATIONAI. I'SS cHOUXiV 

t\']K> of memon' and associative proct'sses; it ^vill also supply the 
individual with certain symbols for recording ideas. The former 
would be the pure training value of the mental functions, the latter 
would be the informational or instrumental value of shorthand. 
If one should ne\er expect to use the syml)ols of shorthand for 
recording ideas, to what extent would the j)ractice in memory 
and associative functions modify memor}- and associative processes 
in other reactions in life? How much value, accordingly, may we 
attach to the practice of these mental processes? In thinking 
about these ])roblems, we must distinguish sharply between the con- 
tent or informational aspect of a given Ix^jc of learning and the 
])ure improvement value in mental functions to be derived from 
the learning. Viewed from the standjwint of the school, the situa- 
tion jjresents two ])roblems: (i) To what extent does training of the 
mental capacities involved in a given school subject airry over and 
])roduce efficiency in other subjects or in other activities in life, 
and (2) are certain school subjects more capable of improving the 
mental functions generally and of carrx'ing the improvement over 
to other responses of behavior? 

The Effect of Improvement in Specific Mental Functions upon 
other Mental Functions. The influence of improvement in t)iie 
function upon others may be one of helj), hindrance, or indifTerence. 
Which it is and how much, can be determined only by recourse to 
facts. Until twenty-five years ago the ])robleni was discussed 
wholly as a matter of opinion. During the last twenty-five years, 
a considcral)le number of researches have been made on many 
asjjects of the jiroblem so that the controversy may be dealt with 
in a more definite and factual manner than was ft)rmerly the case. 

The experimental technique of research in the field of transference 
of training has been practically the same in all investigations, and 
luis consisted (i) of testing the strength of a variety of mental 
capacities, (2) of training one or more capacities for a .s])eciried 
jK-riod of time, and (3) by finally testing again the siime capacities 
tested before the training in order to determine what changes niay 
have been produced in them as a result of the intervening training. 
The tests referred to under (i) and (.0 'ire conveniently called "end 
tests" or the " test series" and the work muler (2) is usually called 
the "training series." Tliis ])lan has been followed in the large 
majority of transfer cx])eriments. A dilTerent plan, however, is 
l)ossiblc and has been em])loyed in a few studies. This consists 
of giving training to a group of jjcrsons in some particular function 



I 



TRANSFERENCE OF TRAINING 



193 



and then giving them practice in another function. Their progress 
in this second function is then compared with that of other individ- 
uals who have not had training in the first function. 

a. James' Experiment on Memory. The first experimental in- 
vestigation was made by James and published in i8go in his Prin- 
ciples of Psychology. This experiment is of interest and impor- 
tance chiefly because of its historical significance in opening the 
problem by an experimental approach. James attempted to de- 
•termine the effect of training in learning one kind of poetry upon 
memorizing other kinds of poetry. He first made the experiment 
upon himself by memorizing in the course of eight days 158 lines 
of Victor Hugo's Satyr. This required a total of 131 Ye minutes. 
He then spent some twenty minutes a day for 38 days in learning 
the first book of Milton's Paradise Lost. At the end of this time he 
again memorized 15S lines from Victor Hugo and found that it 
took i5i>^ minutes. This loss in time was surprising and James 
explained it by saying that he was fagged out by other work at the 
time of the second test on Victor Hugo and that he was not really 
in fit condition for such an experiment. He then repeated the ex- 
periment with four students in a similar manner by using different 
poetry. The results of these early experiments are given in the 
following table: 

TABLE 47. After James ('90, I, p. 667) 



Individual Test Before Training 



158 lines of Victor 
Hugo during 8 days, 
13 1. 8 minutes 

1 28 lines of In 
Memoriam during 
8 days, 14.8 min. 
daily average 

? of Virgil during 
16 days, 13.4 min. 
daily average 

150 lines of ? dur- 
ing 15 days, 3.7 
min. daily average 

? lines of Idylls of 
the King during 
6 days, 14.6 min. 
daily average 



Training 



I St Book Para- 
dise Lost 38 days 

416 lines Schiller's 
translation of the 
iEneid during 
26 days 

? of Scott 



450 lines of ? 



? of Paradise Lost 



Test After Training 



158 lines of Victor 
Hugo during 8 days, 
151.5 minutes 



128 lines of In 
Memoriam during 
8 days, 14.6 daily 
average 

? of Virgil during i 
16 days, 12.3 min. 
daily average * 

150 lines of ? dur- 
15 days, 3^0 



'i 



mg 
min 



daily average 

? lines of Idylls of 
the King during 6 
days, 14.9 min. 
daily average 



194 



mUCATIOXAL PSVCHOUXiY 



This experiment in its essential details, was rei)eated by Peterson 
('12) with two subjects, one of whom showed gain and the other loss. 

/'. Riiulion 'J'inir. The next series of experiments was under- 
taken by GillxTt and Fracker, who attempted to determine the 
amount of transference of training from one tyi)e of reaction to 
other t>T)es of reaction. Three subjects were tested first in simple 
reaction to sound, to electric stimuli, to touch, to \'isual stimuli, 
and likewise in complex reaction to stimuli in\olving discrimination 
and choice. The training series consisted of simple and comi)lex 
reaction to sound only and continued for twelve days. The results 
()])tained in this e.\])eriment are given in the following table which 
shows the percentages of gain made in each of the end tests: 



TABLE 48 

The sfjread of improvement in reacting to \arions sensory stimuli. After 
Gilbert and Fracker ('97) 





The Percentage-s of Tuie Gained bv Practice 


iNDIVIDrAI. 




Simple Reactkjn 


Reaciion wiru Discrimisation 
AND Choice 


To 
Sound 


To 
Electric- 
Shock 


To 
Touch 


To 
color 


SoUhTOS 


EUXTRIC 

Shocks 


Touch 


Blue 

AND 

Red 


J- A. C 

c. c. ]■• 

J. C. P.. 

Averages 


12 

13 
16 


21 
16 

17 


17 

10 

6 

11 


3 
45 
II 

20 


53 
47 
14 

so' 


35 
60 

24 
40 




38 

4 


14 
34 

22 



' ;\vcraKc of J. A. C. and G. C. 1". only. 

J. C. V. was practiced only in reaction time, while the other two were 
practiced in liofh reaction and reaction with discrimination and choice. .Ml 
figures of the al>ove table re[)rcsent i)er < cnt of pain by practice. 

Each of the forms of reaction shdws on the whole a distinct gain 
in the second end tests. How much of this gain is actually due to 
the training series cannot be defmitely determined. Many of the 
earlier investigators did not make control tests, that is, they did not 
repeat the end tests on another group of subjects who did not take 
the practice series but who took only the end tests separated by an 
interval ef|ual to that consumed by the practice series. It is obvious 
that a certain portion of the gain in the end tests is du" to the fact 
that when the second end tests are made, some ad\'antage is derived 
from tin- familiarity or practice in lia\ ing done tin- end tests once 



TRANSFERENCE OF TRAINING 1 95 

before; consequently, the actual amount of improvement in a prac- 
tice experiment can be determined only by subtracting the amount 
of gain made by a control group which has not done the practice 
series in order to obtain the residual amount of improvement ac- 
tually transferred from the training series. 

Another important item frequently omitted in the early investi- 
gations is a statement of the actual amount of progress made in 
the practice series itself. This element is significant because it is 
possible thereby only to determine the amount of gain made in the 
end tests as compared with the improvement in the training series 
itself in order that some definite conception may be formed of the 
amount of gain made in the practice series which is transferred to 
the end tests. Thus in the reaction experiments of Gilbert and 
Fracker, the gain in the practice series is shown in the first and 
fifth columns. It will be noticed that the average gains in the end 
tests in simple reaction to electric shocks, to touch, and to color 
was about as great as in the training series itself, that is, in simple 
reaction to sound. It was 17%, 11%, and 20%, or on the average 
16% in the former, as compared with 16% in the latter. In case 
of the complex reactions, the average gains in the reactions to 
electric shocks, to touch and to color were 40%, 17%, and 22%, 
or on the average 26%, as compared with a gain of 50% in the 
practice series. On the face of it, 100% of the practice effect in 
simple reaction to sound was transferred to the other forms of 
simple reaction, while 52% of the practice effect in complex re- 
action to sound was carried over to the other types of complex 
reactions. Actually the amounts of transfer effects are probably 
considerably less; how much we do not know since Gilbert and 
Fracker made no control tests. 

c, Perception and Discrimination. Thorndike and Woodworth 
('01) made an investigation to determine the transference of prac- 
tice in estimating areas, lengths of lines, and weights to estimating 
areas, lines, and weights of different sizes. They also measured 
the effect of practice in perceiving words containing certain letters 
upon the accuracy and quickness of perceiving other words con- 
taining different letters. The results of this experiment are sum- 
marized in the following manner by Thorndike; 

"Individuals practiced estimating the areas of rectangles from 10 to 
100 sq. cm. in size until a very marked improvement was attained. The 
improvement in accuracy for areas of the same size but of different shape 



196 EDUCATIONAL I'SNCIIOLOCV 

due to this training was only 44% as great as that for areas of the same 
shape and size. For areas of the s;ime shape, but from 140-300 sq. cm. 
in size, the improvement was so% as great. For areas of difTerent shape 
and from 140-400 sq. cm. in size, the improvement was S-'^.c ^'^ great. 

"Training in estimating wciglits of from 40-120 grams resulted in 
only 3g% as much improvement in estimating weights from 120 to 1800 
grams. Training in estimating lines from .5 to 1.5 inches long (resulting 
in a reduction of error to 25% of the initial amount) resulted in no im- 
provement in the estimation of lines 6-12 inches long. 

"Training in perceiving words containing 'c' and 's' gave a certain 
amount of improvement in speed and accuracy in that special ability. 
In the ability to perceive words containing 'i' and 't,' 's' and 'p,' 'c' and 
*a,' 'e' and 'r,' 'a' and 'n,' '1' and 'o', mispelled words and .\'s, there 
was an improvement in speed of only 30^7. as much as in the ability 
specially trained, and in accuracy of only 25% as much. Training in 
perceiving English verbs gave a reduction in time of nearly 21% and in 
omissions of 70%. The ability to perceive other parts of speech showed a 
reduction in time of 3%, but an increase in omissions of over ioo9( ." 

The ex])eriments in markinj]; out words and in estimating \vei<i;hts 
were re})euted Avitli two persons in substantially the same manner 
by Coover. ('16.) 

"Two reagents were trained for 11 days in marking out words con- 
taining e and s in selected columns of the 'Outlook' Magazine. Each 
reagent looked over 12,000 words in each day's practice. 

"Tests were taken before and after training, in marking out 

"(i) Words in 'Outlook' columns containing e-s, i-t, s-p, c-a, c-r. 

" (2) Words on manuscript pages containing a-n, l-o, c-r. 

"(3) Common nouns in 'Outlook' columns. 

"(4) Words in 'Outlook' columns containing e-s." 

Coovcr's results showed a Rain of 44'^(, in tlte training; seri-'s and 
of ;i;i% in the end tests, or 75% as much as in the training series. 
This is a larger transfer effect than that of Thomdike and Wmxl- 
worth whose results, however, were based on five ]H'r.M)ns antl 
showed a gain of 37.7% in the training series and of 17% in the 
end tests, or 48% as much as in the training scries. 

Coover's e.\])eriment in estimating weights was carried out by 
training two ]H-rsons with a set of seventeen blocks ranging from 
40 to 120 grams. I^ach ])erson made 1,700 judgments. The ])er- 
sons were tested, before and after the training, in estimating ten 
conunon objects averaging 67.5 grams in weiglit but falling within 



1 



TRANSFERENCE OF TRyVINING 



197 



the limits of 40 and 120 grams, and in estimating ten common 
objects averaging 552.7 grams but all exceeding 120 grams. 

The experiment yielded a gain in accuracy of estimating weights 
of 23% in the training series and of 29% in the end tests with the 
set of ten smaller objects but a loss of 100% with the larger ob- 
jects. This loss was due to the very large loss of one subject which 
far outweighed the gain of the other subject. The gain in the 
estimation of the smaller weights was greater than in the training 
series itself. Thorndike and Woodworth's experiments showed 
a gain of 45% in the training series and of 38% in the end tests 
with the smaller weights and of 16% with the larger weights. 

Kline had nine persons practice for fourteen days from 30 to 45 
minutes daily in canceling e's and t's on pages of prose. Before 
and after the practice he tested them in canceling nouns, verbs, 
prepositions, pronouns, and adverbs. Eight other persons were 
tested in like manner without doing the practice series. Kline 
found that the practiced group did not gain as much as the un- 
practiced group. This he explains by the introspective statements 
of his subjects that "there was a tendency to cross out words 
containing e's and t's rather than the required part of speech." 
The detailed results follow: 



TABLE 49 
The spread of improvement in marking letters. After Kline ('09, p. 10) 





Nouns 1 Verbs 


Prepositions 


Pronouns 


Adverbs 






§ 






1 






i 






1 






S 






X 








o 






o 




>^ 






^ 


o 






^» 


^ 


Q 


H Q 


'^ 


o 






a 


So 


^ 




-J 

^ n 


^ 








o 


w 


W 9 


n 




a w 


C5 


w 


4 


o 


w 


w '^ 


n 






z 




y. 




^ ^ 


Z 


H 


Z 


H 




Z 






o 

1 


g 

o 


o 
a: 


S 
o 


US 


o 


1 


o 

as 


g 

o 


«2 


o 


O 


Practiced 
























Group 
































After practice 


34.0 


1.6 


12.6 


11.4 


.S.O 


6.0 


28.0 


.S 


7.2 


8.5 


2.3 


6.3 


3,5 


6 


6 .3 


Before practice 


28.6 4.6 


17..? 


9.8 


6.,S 


4.0 


2.S.9 


3.0 


8.2 


6.0 


4.4 


5.0 


6.6 


1.7 


9.3 


Differences. . . 


7.4 


,i.O 


4.7 


1.6 


1.5 


-2.0 1 


2,1 


2.5 


1.0 


2.5 


2.1 1 


-1.3 1 


3.1 


1.1 


3.0 


Unpracticed 


















Group 
































Second period 


.^0.4 


1.4 


10.. 3 


\1.3 


6.0 


7.0 


26.6 


1.7 


9.,? 


5.0 


0.6 


4.0 


5.5 


0.7 


7.0 


First period . . 


2S.S 


.S.I 


17.0 


8.7 


7.0 


.S.O 


16.6 


2.6 


10.5 


4.6 


0.3 


13.7 


4.4 


2.0 


13.0 


Diflferences . . . 


6.9 


i .7 


6.7 


2.6 


1.0 


— 2.0 1 


10.0 


0.9 


1.2 


0.4 


-0.3 1 


9.7 


1.1 


1.3 


6.0 



' — sign indicates loss at second period. 

Bennett ('07) tested a group of sixteen pupils in discriminating 
between shades of red, yellow-green, and orange, and dififerences 



198 KDK AlKiNAL PSYCHOLOGY 

in thu pilch of tones before and after training twice a week for five 
months in discriminating shades of blue. The accuracy in the 
four end tests showed the following gains: 

1234 

Red Yellow-creen' Orange Tones 

Boys 79% 60% 65% 28% 

(Wrls 84% 57% 56% 23% 

Coover and Angell tested four adults in discriminating intensi- 
ties of brightness before and after training in discriminating in- 
tensities of sound consisting of seventeen series of forty judgments 
each. The end tests without the intervening training were also 
given to tliree other subjects. The four trained j^ersons rose from 
56.9',f, of right judgments before the training, to 66.o'/(, after the 
training, while the untrained perscms dropped from 65.5% right 
judgments to 61.7%. 

d. Sensori-motor Association. Bair ('02) attemjjted to measure 
S])read of practice, not by testing certain capacities before and 
after training in some other cai)acity, but In' training the subjects 
in a certain function and then determining the effect of this train- 
ing upon the progress in the subsequent training of other functions. 
His e.\])eriments are described thus: 

"(i) Six keys of a typewriter arc labeled with six symbols (letters or 
figures). Fifty-t'ivc of these letters or figures, in chance order, arc now 
shown one by one, and the subject on seeing one ta|)S the corresponding 
key. The time taken to tap out the series is recorded. Si.\ different 
symbols arc then used with a new series comix)sed of them, and the sub- 
ject's time record is taken as before. This is continued until twenty 
different sets of symbols have been used. Although the symlwls have 
been changed each time, there is a steady improvement, ranging for the 
four subjects in the following decrease in lime: 62 to 52, 05 to 85, 71.5 to 
58, 65 to 56. The major part of this gain could not have been due to 
merely gelling used to the machine or to the general features of the 
experimeiils, for the fourth subject was already used to these and still 
gained about niiie-lenths as much as the other thne. 

"(2) The other experiment consisted in taking daily records for 
twenty days, by means of a stop-watch, of the lime requireii to repeat 
the alphabet from memory. Kach day's experiment was as follows: 
First, the alphabet was repeated as rapidly as jwssible forward; sec- 
ond, the letter n was interpolated between each of the letters; third, 
the alphabet was re[>eate<l backward interpolating n between each two 
of the letters. At the end of twenty praitices in eai h order the subject 



TRANSFERENCE OF TRAINING 1 99 

repeated the alphabet first forward interpolating instead of n the letter 
X and repeating three times; secondly, interpolating r and repeating 
three times; then lastly, repeating backward and in like manner inter- 
polating X and r and repeating three times. There was improvement 
in the test series, the effect of the twenty days' training with the training 
series being to put the abilities in the test series as far ahead as three 
days of the direct training would have done." 

Scholckow and Judd investigated the efifect of knowledge of the 
principle of refraction upon learning to hit a target under water. 

"One group of boys was given a full theoretical explanation of refrac- 
tion. The other group of boys was left to work out experience without 
theoretical training. These two groups began practice with the target 
under twelve inches of water. It is a very striking fact that in the first 
series of trials the boys who knew the theory of refraction and those who 
did not, gave about the same results. That is, theory seemed to be of no 
value in the first tests. All the boys had to learn how to use the dart, and 
theory proved to be no substitute for practice. At this point the condi- 
tions were changed. The twelve inches of water were reduced to four. 
The differences between the two groups of boys now came out very 
strikingly. The boys without theory were very much confused. The 
practice gained with twelve inches of water did not help them with four 
inches. Their errors were large and persistent. On the other hand, the 
boys who had the theory, fitted themselves to four inches very rapidly." 
(Judd, '08, p. 37.) 

Webb ('17) used the plan of determining the effect of acquired 
skill upon the acquisition of other skills. He employed 54 rats and 
21 humans in learning mazes in various orders. He measured the 
results in terms of the number of trials required, the number of 
errors made, and the amount of time needed to learn the mazes. 
The following table gives the savings in learning a second maze 
as compared with the learning of the first one: 

TABLE 50. After Webb 
Average percentage of saving in transfer 





Rats 






Humans 




Mazes 


Trials 


Errors 


Time 


Mazes 


Trials 


Errors 


Time 


A— B . . . 


77.08 


85.81 


83.77 


A— D . . . 


51.98 


94 58 


88.73 


A— D . . . 


69.02 


79.71 


90.42 


A— B . . . 


67.86 


86.64 


67.18 


A— E... 


19.91 


54 63 


63.40 


A— C... 


19.74 


20.20 


29.18 


A— F. .. 


63.01 


42.78 


59-44 










A— C... 


57-85 


46.10 


34 94 











200 KDVCATIONAL PSV(II()I/K;V 

W'thh concluded that "the leaminj? of one maze has a beneficial 
effect in the mastery of a suhsecjuent maze situation" (page 50) 
and that "the dejj;ree of transfer is dependent in part upon the 
dejiH'cc of similarity of two maze patterns" (page 53). Webb 
further attempted to ascertain ^\■hether a new habit has a retro- 
active efTect upon habits ])reviously formed. He had his subjects 
learn one maze, then a second one, and then return to the first one. 
His findings were inconclusive. 

Coover ('16) reports an unpublished investigation by Carrie 
\V. Liddle designed to measure the efTect of practice in discrimi- 
nating and sorting cards bearing colors or geometric signs upon 
discriniinating and sorting cards with dillerent colors or signs, 

" Each set of 102 cards contained six colors, or six designs, was shuffled 
so thai no color or device repeated itself, and was sorted into six com- 
|>artmenls. The tirsl six cards of the pack determined the order of colors 
in the compartments according to which the rest of the pack was to Ix: 
sorted. Nine reagents took part and the experiment continued two 
semesters. There was transference of practicc-edect from one set of 
colors to the other set of colors, and to the geometric forms; and from one 
set of geometric forms to the other and to the colors. Increase*! powers 
of discrimination and attention were thought to be the causes of trans- 
ference." 

Bergstrom ('04) had found previously that training in sorting 
cards by one method interfered with sorting them b}- a ditTerent 
method. The same situation is shown by the card-sorting e.v])eri- 
ment in the author's Experiments, Chapter XV. 

Coover and Angell (07) attem])ted to ascertain the efTect of 
l)ractice in card sorting upon tNTJCwriter-reactions. They trained 
four ])ersons in card sorting on 15 days scattered through a 
period of 40 days. During that time the subjects sorted 4,200, 
3,800, 5,200, and 4,000 cards respectively. Before antl after this 
training they were given j)ractice in t)i)ewriter-reactions. Three 
other persons, as a c<mtrol grouj>, were given practice in tjix^writ- 
ing at two periods separated by an interval of 45 days. The re- 
sults are inter])reted by the authors as indicating transfer, but it 
is doubtful whether there is any transfer and, if there is, how much. 
The practiced grovij) reduced tluir time for the first 100 t\i)e\vriter 
reactions, before the training in card sorting, from S4.4 seconds, 
with an average of 2.1, errors, to 62.3 seconds, with 6.3 errors, for 
the last 100 reactions after the training in aird sorting. The 



TRANSFERENCE OF TRAINING 20I 

unpracticed group reduced their time from 106.3 seconds, with 
$.^ errors, to 80.6 seconds, with 2.3 errors. The trained group 
reduced its time by 26% but increased in errors, while the untrained 
group reduced its time by 25% but decreased in errors. There is 
obviously no appreciable transfer. 

e. Memory. More extensive researches have been made in the 
field of memory than in any other single aspect of the problem of 
transference of practice. One of the most elaborate investigations 
was made by Ebert and Meumann. They measured the amount 
of transfer from memorizing a series of nonsense syllables to vari- 
ous other types of memory, such as immediate memory for num- 
bers, letters, words, permanent memory of prose, poetry, etc. 
The end tests were made at three different times, before the be- 
ginning of practice series, about the middle, and at the close of the 
practice series. The results showed very considerable gains in 
these other types of memorizing. The difficulty in interpreting 
their results, however, is the fact that they did not make the cross 
section tests with a control group according to which a deduction 
could be made for the gain in the end tests themselves. Dearborn 
repeated the end tests on a group of subjects to ascertain the 
amount of allowance to be made. His results together with those 
of Ebert and Meumann are shown in the following table. Dear- 
born found a very considerable amount of gain in these end tests. 
His comments are as follows: 

"The results indicate that a considerable part of the improvement 
found must be attributed to direct practice in the test series, and not 
to any ' spread ' of improvement from the practice series proper. There 
is further, at times, lack of correlation between the amount of improve- 
ment made in the practice and that made in the test series; occasionally 
a larger percentage of gain is made in the latter than in the practice itself. 
This again indicates the presence of direct practice in the test series. 

"Some at least of the remaining general improvement found is to be 
explained simply in terms of orientation, attention, and changes in the 
technique of learning. 

"These results seem to render unnecessary the hypothesis proposed by 
Ebert and Meumann to account for the large extent of the general in- 
fluence of special practice, which their experiments seem to indicate." 

Three subjects of Ebert and Meumann were trained in learning 
64 sets of i2-syllable series; they gained 70%. Three others were 
trained with 48 sets. They gained 50%. 



202 



ri)r(\\Ti()N'AL psvciir)L()(A' 



TABLE SI. After Kbcrt and Meuniann (05) and Dcarlwrn 





Ebert and Mki'mans's Gains 
or 


Dearborn's 3rd 

Cross Section 

Test OVER the 

First' 






2nd Crms 

Section over 

TOE First 


iRD Cross Sec- 
tion OVER the 
KmsT 


Difference 


Mcinon' Span: 

\umbers 


24% 
32 

22 

(>7 
61 
64 
48 
24 
3<> 

— 12 

— 29 

6 
42 


60% 

35 

43 

81 

7(> 
80 
70 

72 
73 

34 
1 1 


12'c, 
29 . 
17 

41 

52 

14 

58 


48% 
6 


Letters 


Syllables 


26 


Memorizing: 

lo-syllable scries 

12- " " 

14- " " 

16- " " 

Geom. Forms (easy) . . . 
" (hard) . . . 
(ierman-Italian Vocab. 
(^0 Dairs) 


40 


Ck-rman-Italian \'ocab. 
(so pairs) 




Poetry 16 lines 


-3 
14 

22% 


Prose 20 linos 





' Quoted !)>■ iKTmission from an un|nil)lishc<! laMf |>rrixiri'<I l)y r)carl)orn. 

Thus the high jJiTccnta^^cs of Kbcrl and Mt-uinann arc reduced to 
an average residual transfer of 22%. 

Fracker has reported a rather extensive series of investigations 
on transfer in memory in \vhich the training series consisted of 
memorizing various combinations of four degrees of lou(hiess in a 
sound. These four loudnesses were i)resentetl in the various possible 
combinations and the responses of the subjects consisted in indi- 
cating the proper order in which the sounds had been received. 
The end tests consisted in determining the memors- aipacities for 
various combinations of four shades of gniy, 9 tones, 8 shades of 
gray, 4 tones, geometrical figures, 9 sets of numbers, arm move- 
ments, and poetry. The results are summarized in the folK)wing 
table which also indicates the an\ount of deduction to be made due 
to the improvement in the end tests made by the control group. 



TRANSFERENCE OF TRAINING 203 

TABLE 52 

Transference of training in memory. After Fracker ('08) 

The improvement made in Training Series by 8 subjects was 21%. 
End Tests: 

Similar to Training Series: 

Four Grays, 8 trained subjects, 

Nine Tones, 8 " " 

Nine Grays, 8 " " 

Four Tones, 8 



36% 4 untrained, 


4% 


32% 


22 4 


ri 


11% 


19 4 


10 


9% 


10 4 


— 2 


12% 



22% 6% 16% 



End Tests: 

Unliiie Training Series: 
Geometrical Figures, 8 trained subjects, 
Nine Numbers, 8 " " 

Movement, 8 " " 

Poetry, 8 " 



13% 4 untrained. 


8 


5% 


4 4 





4% 


04 


— I 


1% 


7 4 


2 


5% 



6% 3% 3% 



An interesting result emphasized by these data is the fact that 
the transfer to the types of memory similar to that involved in the ^ 
training series is considerably greater than the transfer to the 
memory functions unlike the training series. The average residual 
gain in the four similar memory processes is 16%, whereas in the 
four unlike memory processes it was only 3%. 

Sleight made a careful and extensive investigation on transfer- 
ence of training in one sort of memory to other sorts of memory. 
He believed that previous researches had not used enough subjects 
to be statistically reliable. He therefore carried out his first re- 
search with 84 pupils from three girls' schools, averaging 12 years 
and 8 months old. Ten cross sectional tests were made before, 
in the middle, and after the training series, as follows: (i) Re- 
membering and reproducing the location of points in circles, (2) 
two series of six dates each and their corresponding events, (3) 
series of eight syllables, (4) a stanza of from eight to twelve lines 
of poetry, (5) learning a passage of prose, (6) reproducing the con- 
tent of a passage of prose, (7) remembering locations on a map, (8) 
remembering dictated sentences, (9) memory span for letters, (10) 
remembering names. 

The pupils were divided into four groups of approximately equal 
ability as determined by the ten tests before the training series. 
One group was then trained in learning poetry; another in learning 
tables of multiplication, denominations, squares, fractions, etc.; 



204 



KDUCATION'AL PSVCIIOLOr.V 



a third in reproducing the thought content of ])rosc selections of 
scientific, geographical and historical material; and the fourth 
group had no special practice. The training period lasted four 
days a week for six weeks, i)ract icing 30 minutes each day. 

The chief results are ])resented in Table 53. I have com])uted 
the percentage of gain made by each group in Section III, that is, 
the end tests made at the close of the training series, over Section 
I, the tests made before the training series. These percentages are 
givdti in the last column. Sleight has not made such a percentage 
comparison, but has used a dilTerent, and possibly fairer, plan of 
computing the data. I have, however, made this computation in 
terms of percentages as these will be more intelligible to the reader 
unfamiliar with statistical methods. The average percentages at 
the bottom of the table show only slight gains on the part of the 
trained groups, 2, 3, and 4, over the untrained group. The average 
gain of group 2, trained in poetry, over group i, untrained, was 
Z-Z^o'^ of group 3, trained in arithmetical tables, over group i was 
2.6%; and of group 4, trained in prose, over group i was 4.0%. 
The amounts of transfer arc very small. Sleight failed to indicate 
the improvement in the training series themselves so that it is im- 
possible to compare the transferred amount with it. 



T.\BLE 53 

The numbers in the following tahlc arc the avcraRC scores made by each 
RTOup in each test, (irmip i had no siwciai practice, (irouj; 2 was praclicetl in 
learninK poetry, (Iroiq) 3 in learning tables, and (irou[) 4 in learning prose 
substance. 

The column under Section I ^ivcs the scores before the training, untfer 
Section II about the middle of the training, under Section III after the training. 
After Sleight. ('11, p. 413.) 



Section I 
Early Test 



Skxtiov II 
Middle Te-st 



Section III 
Final Tf^t 



pKHrKNTACE 

Cain or III 

OVI.R I 



Points Group i 



Dates 



3 
" 4 
. (iroup I 
" 2 
" 3 
" 4 



73 9 
66.8 
66.5 
58.5 
14.4 
14.7 
18.9 

>7 7 



86.2 
80.2 

77--' 
^<).8 

15 -3 
16.8 
21 .9 
17. 1 



86.5 

845 
00.3 

7^>.S 
18. 1 
20.4 
21 3 



17 
25 
36 

31 

26 

38 
13 
J4 



TRANSFERENCE OF TRAINING 



205 



TABLE 53 — Continued 





Section I 


Section II 


Section III 


Percentage 
Gain of 111 




Early Test 


Middle Test 


liNAL Test 


OVER I 


Nons. Sylls. Group i 


20.7 


20.7 


22.8 


10 


" 2 


19 


8 


24 


9 


27 


3 


33 


" 3 


19 


2 


24 


9 


28 


2 


47 


" 4 


21 


9 


21 





24 


6 


12 


Poetry Group i 


58 


5 


62 


4 


63 


8 


9 


" 2 


56 


5 


59 


4 


57 


9 


3 


" 3 


60 


3 


60 


9 


64 


4 


7 


" 4 


59 


4 


63 


4 


74 


7 


35 


Prose (literal) 
















Group I 


109 


8 


117 


4 


118 


6 


8 


" 2 


lOI 


9 


107 


3 


107 


5 


7 


" 3 


108 


I 


"3 





"5 


6 


7 


" 4 


104 


6 


113 


7 


118 


3 


14 


Prose Subs . Group i 


27 


5 


28 


8 


30 


5 


II 


" 2 


23 


5 


24 


8 


24 


7 


5 


" 3 


23 


5 


27 


I 


27 


I 


15 


" 4 


22 


8 


28 


8 


28 


3 


20 


Map Test . . Group i 


63 


9 


6S 


9 


72 


4 


13 


" 2 


65 


9 


65 


I 


81 


9 


25 


" 3 


65 


9 


64 





74 


5 


13 


" 4 


68 


3 


66 


8 


78 


7 


16 


Dictation . . Group i 


134 


I 


135 


9 


139 





4 


" 2 


129 


6 


130 


9 


130 








" 3 


129 


3 


130 


3 


132 


8 


3 


" 4 


129 


8 


133 


6 


134 


7 


4 


Letters .... Group i 


76 


I 


78 


9 


80 


2 


5 


" 2 


79 


2 


81 


7 


82 


6 


4 


" 3 


76 


5 


78 


4 


80 


8 


6 


" 4 


78 


7 


81 


I 


82 


4 


5 


Names Group i 


32 


7 


41 


5 


41 


4 


27 


" 2 


34 


7 


39 


9 


42 


7 


23 


" 3 


35 


3 


39 


7 


42 


I 


19 


" 4 


35 


5 


41 


5 


45 


9 


29 



Average % of gain of Group i in all tests 130 

" " " " 2 " " 16.3 

" " " " " " 3 " " " 15-6 

" " " " " 4 " " " 17.0 

Sleight, by his method of computation, found only a few in- 
stances of significant amounts of transfer. His conclusion is that 
"There appears to be no general memory improvement as a result 
of practice, nor any evidence for the hypothesis of a general memory 
function " (p. 455). 



2o6 KDUCATIOXAL I'SVCIlULUC.V 

After the conclusion of ihcse cxptrimenls, Slcij^ht rcjicatcd the 
same investigation Avith some modilkations, on a j?roup of young 
women, iS to 19 years old. The results were substuntiully the 
same. 

Coover made a study of the elTect of training in reproducing im- 
ager)' of a simple kind ui)on ability to reproduce imagery aroused by 
materials of various sorts. The tests made before and after the 
training were as follows: (i) recognition or choice of one of two 
letters previously shown, (2) rej^roduction and recognition of letters 
I)resentcd in groups of 12, (,0 discrimination of intensities of sounds, 
(4) niemor)' of visual symbols. The training consisted in i)ractice 
in discriminating intensities of sound, and extended through a 
])eriod of 48 days. These intensities of sound were produced with a 
sound i)endulum (wood) instead of with a fall jihonometer (steel) 
as in end test number (3). The results of the investigation show 
small or doubtful elTects of transfer. 

"The training on (liscrimination of sound did not result in improve- 
ment in cfticicncy with the training material. But, according to intro- 
spective evidence, it cfTcctcd changes in the processes employed. Quan- 
titative analysis showed that the practicc-cflfccl of the evident exercise of 
retention and reproduction of auditory and other imager)' 'sjjread* to 
the tachistoscopic test of Recognition or Choice of One of Two Letters, 
and to the test on the Complete Learning of series of visual sjTnbols, 
both of which iinolvefl releiilion and reproduction of imagery." 

Dearborn made some e.v|>eriments to measure the cfTect of ])rac- 
lice in leaniing vocabulary and poetry upon ability to memorize 
various sorts of material as specified in the following table. He 
did not make the end ti-sts on a control grou]> and hence it is im- 
])()ssible to determine how much of the gain in the end tests was due 
to the practice series. Judging from other ex])erimcnts these gains 
would have to be reduced by one-half or one-third. An interesting 
comparison may be made between the gain in the end tests and the 
training series. The average gain in learning French and Clerman 
voc;d)ulary was 57',. when-as the average gain in the end tests in 
learning l'"reii< li, (iirinan, or l-^nglish verse was only 19% or one- 
third as much. Practice in learning Ptiradisc Lost made 710 im- 
provement in learning chemical formula-. 



TRANSFERENCE OF TRAINING 



207 



TABLE 54 

Transference of practice in memorizing German and French vocabularies and 
ICnglish poetry and prose. After Dearborn (1910), p. 385 



Person 


Practice 
Material 


Percentage 
Gain 


End Test 
Material 


Percentage 
Gain 


I 


French Vocab. 


57% . 


French Verse 


25% 


2 


German " 


60 


German " 


10 


3 


French " 


53 


Enghsh " 


17 


4 
5 


u 


55 
62 


French " 


7 
33 


6 


German " 


S7 


German " 


25 


7 
8 


Victor Hugo 
Horace's Odes 


82 
73 


Browning 
Norse Poem 


52 
17 


9 


Paradise Lost 


68 


Chemical Formulae 





10 


Enoch Arden 


55 


Burke 


2 



Bennett ('07) had one person memorize 16 lines of In Memo- 
Ham a day for 28 consecutive days. This person was tested before 
and after this i)eriod of training by learning a list of 15 names of 
places each day for five days in which he showed a gain of 58 %. 
Another person memorized two stanzas of Faerie Queene a day for 
35 consecutive days. Before and after this period he was tested in 
learning a list of 30 digits each day for five days and in which he 
showed a gain of 22%. No control tests were made on other 
persons without training in learning the poetry. 

Winch tested a group of 34 girls, averaging 13 years of age, by 
having them learn a passage of historical prose. On the basis of 
this test he divided them into two groups of equal ability. Group 
A memorized 18 to 20 lines of poetry each day on four days scat- 
tered through a period of two weeks. Group B meanwhile worked 
sums. At the end of that time both groups were tested with his- 
torical prose. Group A rose from a total score of 1,497 to 2,055, o^ 
37%, while group B rose from 1,497 ^o IjSqo? or 27%, 

Winch ('08 and '10) next tested another class of 34 girls in the 
same general manner, except that the before and after test was 
made with geographical passages and that the poetry for the train- 
ing series was somewhat simpler. He also carried out a similar 
experiment with a third class of girls, using a historical passage for 
the end tests. The results in each case showed a greater gain in the 
practiced group. 

The results from the author's class experiments (Chapter XI, 



2o8 EDUCATIONAL PSYCHOLOGY 

Experiments) indiaito that an im])rovemt'nt of 27% i" learning 
Italian vocabulary is accomijanicd by an improvement of only 
8% in learning French \ocabulary or less than one-third as much. 
Improvement in transcribinj^ letters into numbers was accompanied 
by only 12% as much gain in transcribing numbers into symbols. 
Attention. Coover made a series of tests on transference which 
he lists under the head of attention, but it is doubtful, as he himself 
states, whether they are measures of attention any more than 
many other tests that have been reported under other headings. 
At any rate, attention probably ])layed an im])ortant \ydxi in most 
of the tests that Coover employed. The following were the nineteen 
end tests made before and after training. 

I. Reaction 

1 . Simple sensor>' to sound (50) i 

2. Compound 

a. With discrimination 

(i) Marking out small a's (100) 2 

(2) Marking out o's (100) 3 

b. With discrimination and choice 

(i) Card-sorting (200) 4 

(2) Tyj)e\vTitcr-reaction (200) 5 

(3) Controlled reaction (50) 6 

n. Sensible discrimination of sounds {90) 7 

III. Reproduction 

1. Unequivocal (Rote memory) 

a. Successive presentation 

(i) Memory of s<nmd intensities (50) 8 

(2) Memory of consonants (50) o 

(3) Memory of Arabic numerals (50) 10 

(4) Memory of visual signs (10) 11 

(5) Memory of associated pairs (50) 12 

b. Simultaneous presentation 

(i) Learning 1 2-letter-reclangles 

(a) Free (10) 13 

(b) With distraction (10) 14 

2. Equivocal — Word-completion (10) 15 

3. Free — 2-minutc trains of ideas (3) 16 

rV. Extensive threshold of visual attention 

1. Free (15) '7 

2. With di.straction (10) iS 

V. Maximum voluntary activity — tapping (5 30") 19 

(The figures in parenthesis indicate the number of reactions, memory' units 
or cxi>criments, in the test.) 

"These tests were taken by 10 reagents, 8 of whom took training be- 
tween the first and luKil series which were separated by an interval of 



TRANSFERENCE OF TRAINING 209 

55 days. The first series of tests occupied 12 days during a period of 
36 days; the second, or final, 10 days during a period of 21 days. Each 
pair of tests was separated by an interval of about 66 days. 

"The pairs of tests were also taken by two reagents of a group of 21 
control reagents. There were thus two sets of control reagents: The two 
who took all the tests, and the 21 each of whom took only one pair or a 
few pairs of tests." (Coover, '16.) 

Different ones of the subjects took training in different material : 

"During the 55-day interval between the tests, two reagents (MN., 
Le.) took training 18 days on Test 17; 25 1 2-letter-rectangles were pre- 
sented daily, making in all 450 experiments each. Two reagents (Rt. and 
SI.) took training 18 days on Test 13; 20 1 2-letter-rectangles were pre- 
sented daily, aggregating 360 experiments each. One reagent (Ly.) took 
training in simple reaction to sound for 11 days, 1,100 reactions in all. 
(Le., who took training on test 17, also took training in this simple re- 
action to the extent of about 500 reactions.) Two reagents (He., Cr.) 
took training on memory schemes for about 14 days. And one reagent 
(al.) took training on Test 17 for 8 days, almost consecutive, to the ex- 
tent of 200 experiments." 

The results obtained from the various end tests are rather intri- 
cate and difficult to present in tabular form and somewhat doubtful 
as to their meaning so far as improvement in attention is concerned. 
Coover attempted to interpret their meaning from the standpoint 
of control of attention by comparing the variability in the perform- 
ance of the persons before and after training on the assumption that 
reduction in variability indicated better attention. For a detailed 
consideration, Coover's original report must be consulted. His 
general conclusion was that "as a measure of attention our tests 
are inadequate, and the question of transference of improved con- 
ditions of attention remains open " (page 183). 

g. Analysis and Ingenuity. Ruger ('10), in connection with his 
study of learning to solve puzzles, made observations on transfer 
of practice. His results are difficult to summarize in brief form. 
It will have to suffice, therefore, to say that he enumerates general 
factors of transfer in solving puzzles as follows: (a) The ideal of 
efficiency, that is, "the active search for methods of control;" 
(b) a high level of attention was a precondition of success; (c) 
attitudes — "The change from the self-conscious to the problem- 
attitude occurred sometimes automatically, and sometimes de- 
liberately by means of an ideal. The most powerful stimulus to 



2IO EDUCATIONAL rSVCIIOLOC.V 

change of attitude and so of its transfer ^vas personal success; it 
did not matter much Avhether it Avas accidental or ])lanned;" (d) 
methods of attack. As to special factors he mentions: (a) Related 
ideas — "Geometrical concejjts played an almost ne;^l|gihle jjart in 
the work of solution;" "The greatest transfer in the \va\-^f relatcfl 
ideas was that from similar ])uzzles; " (b) motor habits — "The mere 
presence, in the case of change of conditions, of motor habits aj)- 
propriate to the new conditions did not necessitate positive trans- 
fer," "The degree of positive transfer varied directly with the pre- 
cision of analysis of the similarity of the new case to the old," 
"In some cases a generalized formula developed in connection with 
the first case was essential to effective transfer of motor habits to 
later modifications of the first case," "Transfer was more effective 
in those cases where the formula or general rule was develojied in the 
first few trials, and where the formation of perceptual-motor habits 
had been controlled and inter-ix-'Uftrated by it from the start, 
than when the generalization had been arrived at after those habits 
had been set up." 

//. Cross Education. Cross education refers to the transfer of 
practice from one organ of the body to bilaterally symmetrical 
organs, as for example the spread of training from the right hand 
to the left hand. A number of investigations ha\e been made on 
this problem which show that such transfer takes place to a very 
great extent. .Scri])ture, Smith, and Brown ('04) state that im- 
provement in the strength of grij) with one hand produced ^o' 
as much gain in the other. They also report that V'olkmann found 
that im])rovement in discrimination with the left arm was accom- 
jKinied by a])proximatL-ly So^r as much gain in the other arm and 
that other instances showed similar gains. 

Da\is CqS-'oo) measured the effect of practice in tapping with 
the right great toe upon the rate of tapping with the right hand, 
the left hand and the left great toe. He found that the left toe 
imjiroved iSi'^/f as much as the right toe, with which the ]irac- 
ticing had been done, the right hand 100'^ ,' as much, and the left 
hand 83'^,' as much. He also found that practice in gripping a 
d>'namometcr with the one hand improved the other about 70% 
as much. Practice in hitting a target 100 times with the right hand 
improved the left hand about 75*^,' as mucli. 

Woodworth ('()o) reports that practice in hitting dots with the 
left hand improved the right hand about 50'^, as much and Swift 
found that practice in tossing balls witli the right hand caused the 



TRANSFERENCE OF TRAINING 211 

left hand afterwards to improve in the same exercise more rapidly 
than it would otherwise have done. 

The writer ('lo) measured the amount of transfer of improve- 
ment in tracing with the right hand a star outline as seen in a 
mirror to tracing the same outline with the left hand. The 
test was made by having a tracing made first with the left hand, 
then a series of 25 to 100 tracings with the right hand, and, at 
the close, a tracing again with the left hand. The amount of 
transfer to the left hand is approximately 90% of the total amount 
of gain made by the right hand. The left hand improves nearly 
as much as the right hand although all the practicing had 
been done by the right hand. The results of experiments in cross 
education are somewhat uncertain in- their meaning so far as 
transfer of training is concerned. Improvement in one organ, 
which is uniformly accompanied by a very large improvement 
in bilaterally symmetrical organs, is probably due to the fact that 
many common processes are involved in doing a task with two 
bilateral organs. For example, in practicing with the hands, many 
of the same sensory and neural processes are involved. Thus, the 
same visual processes and the same visual brain centers would be con- 
cerned. Likewise, it is also probable that neural innervations going 
to the right hand in practice also go to the left hand. These would 
tend to improve the control of the left hand without actual practice 
with the left hand. The data on cross education probably have 
only a distant and doubtful bearing upon the problem of transfer. 

Criticism of the Technique of Experiments on Transfer of Train- 
ing. There are three important elements in the technique of 
experimentation in this field which have not been recognized by 
investigators from the beginning and are not recognized by all 
investigators even to-day. (i) The first is the length of the end 
tests. These have been too long in some investigations to give 
as full opportunity as possible to transference. This was one 
of the difficulties in James' original experiment and was recog- 
nized by James himself. (2) In the second place, the end tests 
have not always been repeated on a control group of subjects. 
This is true of nearly all of the early studies. (3) In the third place, 
many investigations do not mention the amount of improvement 
made in the training series itself with which the gain in the end 
tests may be compared. Failure to observe these precautions 
makes impossible an accurate, quantitative interpretation of 
many of the early researches and even of some of the recent ones. 



212 EDUCATIOXAL PSVCIIOIX)GY 

Summary. It may ptrluips he unwise in view of the intricacy 
of the researches and their partial incongruity to attempt to sum- 
marize general conclusions. However, a brief resume will help 
to clarify the reader's thinking about these problems, (i) Practi- 
cally every investigation shows that improvement in one mental 
or neural function is accompanied by a greater or less amount of 
modification in other functions. (2) This modification is in most 
instances a positive transfer, that is, an impro\ement. Negative 
transfer, that is, loss of efficiency in other functions, or interference, 
has been reported princii)ally among sensori-motor habits. (3) 
The amount of improvement in the capacity trained is probably 
never accompanied by an ecjual amount of improvement in other 
capacities, with the possible exception of a few isolated instances 
whose actuality may be questioned. Thus, for example, Thorndike 
and Woodworth found that the gain in \arious types of perception 
or discrimination closely related to the type in which the training 
took ])lace was from o';'^ to about 40^7 as great as that made in 
the particular kind of perception trained. In memory, Fracker's 
results showed that the improvement in different sorts of memoriz- 
ing, so similar to the training series that they were all but identical 
with it, was about 75',r as much as that made in the training 
series; while the improvement made in the forms of memory rallier 
different from the training series was only about 15^^, as much 
as that in the training series. Up to about 1890 when James re- 
ported the first investigation on the j^roblem of transference, it 
was tacitly assumed by many writers that a very large share, if 
not all, of the training derived from one sort of exercise was carried 
over to other sorts of exercise. After the first investigations be- 
came generally known, many writers went to the other extreme 
and assumed that all training is entirely specialized and that 
nothing carries over from one kind of practice to any other kind 
of practice. As a general estimate, on the basis of experimental 
work done thus far, the amount of transference between the ex- 
tremes of 100' ^ and o'( of transfer lies nearer to the zero end and 
is probably in the neighljorhood of 20',') to 30'^^'fi of transfer ti) 
closely allied functions and from that point on down to o'^^ *^f 
transference to more unlike functions. (4) In the fourth place, the 
improvement spread to other functions diminishes very rapidly 
in amount as these other functions become more and more un- 
like the function specifically trained. This diminution occurs at a 
surprisingly rapid rate. 



TRANSFERENCE OF TR^-^INING 213 

■ How Does the Transfer Take Place? If improvement in one 

mental function is accompanied by, or produces improvement in, 
other functions, how may the change in these other mental func- 
tions be explained? How does change in one function carry over 
to others? Two general theories have been proposed: (i) The 
theory of identical elements or special connections, and (2) the 
theory of generalization or common capacities. 

The theory of identical elements has been advocated by Thorn- 
dike and may best be stated in his own words: 

"The answer which I shall try to defend is that a change in one func- 
tion alters any other only in so far as the two functions have as factors 
identical elements. The change in the second function is in amount that 
due to the change in the elements common to it and the first. The 
change is simply the necessary result upon the second function of the 
alteration of those of its factors which were elements of the first function, 
and so were altered by its training. To take a concrete example, im- 
provement in addition will alter one's ability in multiplication because 
addition is absolutely identical with a part of multiplication and because 
certain other processes, — e. g., eye movements and the inhibition of all 
save arithmetical impulses, — are in part common to the two functions. 

"Chief amongst such identical elements of practical importance in 
education are associations including ideas about aims and ideas of method 
and general principles, and associations involving elementary facts of 
experience such as length, color, number, which are repeated again and 
again in differing combinations. 

"By identical elements are meant mental processes which have the 
same cell action in the brain as their physical correlate. It is of course 
often not possible to tell just what features of two mental abilities are 
thus identical. But, as we shall see, there is rarely much trouble in 
reaching an approximate decision in those cases where training is of 
practical importance." (Thorndike,'i4, II, pp. 358-359.) 

The theory of generalization has been advocated by Judd in 
the following manner: 

"The important psychological fact ... is that the extent to which 
a student generalizes his training is itself a measure of the degree 
to which he has secured from any course the highest form of training. 
One of the major characteristics of human intelligence is to be de- 
fined by calling attention, as was pointed out in the chapter on science, 
to the fact that a human being is able to generalize his experience. 
James has discussed this matter by using the example of the animal 
trained to open a particular latch. The animal becomes acquainted with 



214 EDUCATIONAL I'SVCIIOLOGV 

the necessary movements to open one door, but he never has the abiUty 
to generalize this experience. He cannot see that the s:ime method of 
opening doors is applicable to many other latches. The result is that the 
animal goes through life with one particular narrow mode of behavior, 
and exhibits his lack of intelligence by his inability to carry this single 
type of skill over to the other cases which are very familiar to the trained 
human intelligence. 

''James goes on to say that the same distinction appears when we con- 
trast a trained scientific mind with the ordinary mind. The ordinary 
thinker does not see how to deal with a situation in terms of scientific 
principles. James cites the exam{)le of his own experience with a smoking 
student-lamp. He discovered by accident that the lamp would not 
smoke if he put something uniler the chimney so as to increase the air 
current, but he did not realize that what he had done was only one par- 
ticular example of the general princi{)le that combusion is favored by a 
large supply of oxygen. The general principle and its useful application 
belong to a sphere of thinking and experience which the untrained lay- 
man has not yet mastered." (Judd, '15, pp. 413-414.) 

The theory' of identical elements is based on the doctrine that 
' learning or changes in mental capacities consist of the establish- 
ment of specific connections or associations between various 
specific elements. One form of e.xercise has influence upon another 
capacity whenever connections established in the foimer may also 
be used in the latter. In a certain sense the theory of identical 
elements describes or explains transfer of training in a tangible, 
concrete manner. In a certain other sense it does not e.xpiain 
transfer of training at all or else it implies that there is no general 
training in the sense in which formal disciplinarians use the term. 
If special training is general or helps in performing various mental 
activities only to the extent to which the special training has 
elements in it which occur also in tliese other activities then there 
is no spread of training to such activities in which no elements 
are found which also aj^pear in the capacity specifically trained. 
The formal discii)linarian assumes that training of one sort alTects 
capacities of other sorts irrespective of identical elements or simi- 
larity to the activities developed. In tlie last analysis the conlrt)- 
versy comes down to a question of fact, namely, to how dissimilar 
activities does any given form of training spread? The theory of 
identical elements, when the term identical elements is used in a 
liberal manner, has the advantage of describing the situation in con- 
crete, definite concepts and lends itself fairly well to the interpreta- 
tion of experimental results. The discussion of the formal discipli- 



TRANSFERENCE OF TRAINING 215 

A narian is usually not in as tangible terms but is likely not to be very 
' ^ different from the statements of the experimentaUst when the former 
reduces his argument to specific terms. 

I The theory of generalization attempts to explain spread of. 
/improvement in terms of the recognition of application of an ex- 
perience obtained in one connection to other connections and is 
probably more satisfactory to the formal disciplinarian. In the 
author's opinion there is no necessary opposition between the 
theory of identical elements and the theory of generalization. 
The essential difference is in the emphasis upon the conscious rec- 
ognition of identical elements in as many situations as possible. 
Judd has emphasized this in connection with teaching: 

"The first and most striking fact which is to be drawn from school 
experience is that one and the same subject-matter may be employed 
with one and the same student with wholly different effects, according to 
the mode of presentation. If the lesson is presented in one fashion it 
will produce a very large transfer; whereas if it is presented in an entirely 
different fashion it Avill be utterly barren of results for other phases of 
mental life. Itjs quite possible to take one of the objects of nature study, 
for example, and to teach it in such a way that it becomes an isolated and 
utterly formal possession of the student. This has been illustrated time 
and time again by the instruction which has been given in birds and 
plants. A teacher can teach birds and plants in such a way as to arouse a 
minimum of ideas in the student's mind. The training may be as formal 
in these content subjects as it ever was in language instruction. On the 
other hand, the same subject-matter may be taken by a different teacher, 
and under other methods can be made vital for the student's whole 
thinking. Thus the teacher who is dealing with birds as a subject of 
nature study and secures an interest on the part of his students for the 
world in which these birds live, through an examination of the structures 
and habits of the birds, will have in this subject-matter one of the most 
broadly interesting topics that can be taught. In exactly the same way a 
teacher who knows how to make use of the materials given in a Latin 
course may render this subject very broadly productive, as contrasted 
with the teacher who merely gives the formal aspects of the subject. 
Formalism and lack of transfer turn out to be not characteristics of sub- 
jects of instruction, but rather products of the mode of instruction in 
these subjects." (Judd, '15, pp. 412-413.) 

It seems then that the two theories are not necessarily antagonis- 
tic but when sanely interpreted are useful supplements to each 
other. The theory of identical elements has helped to make the 



2l6 EDUCATIU.NAL I'SVLllol.oc.V 

discussion of formal discipline or transfer of training; concrete, and 
the theory of generalization will help to emphasize the conscious 
recognition of the identical elements in as many situations as 
possible. Some writers have assumed that transfer is limited 
to a conscious recognition of elements. This, however, is dis- 
proved by some experiments with human beings and i)articularly 
by the experiments with animals such as those reported by Webb. 



CHAPTER XIV 

TRANSFERENCE OF TRAINING IN ABILITIES IN SCHOOL 
SUBJECTS 

To what extent does the training of the capacities exercised by 
school subjects carry over to capacities concerned in other school 
subjects, and especially to the capacities involved in the usual 
activities of life? This question brings the problem of transfer- 
ence of training directly face to face with the issues of education 
and is the form in which it is usually concerned in discussions of 
mental discipline. It is of more special concern to the liberal 
phases of education in the high school and the college for the reason 
that the subjects taught in the elementary school, in the vocational 
courses in the high school, and in the professional courses in the 
university are directly pertinent to the common needs of life or 
to the various occupations and professions. Most of the discus- 
sion has, therefore, centered about the training value to be de- 
rived from the traditional academic work of the high school and 
the college. 

General Opinions. The beliefs concerning transfer of training of 
the capacities employed in school subjects have been largely matters 
of opinion and not matters of fact. These opinions, held by persons 
prominent in educational affairs, have been for the most part 
rather uniformly confident in the faith that the exercise of the 
mind upon the materials of the school subjects produces a very 
profound improvement in mental powers as a whole. Typical of 
such opinions are the following: 

"But my opinions of the supreme educational value of the great dis- 
ciplinary studies have not changed, and will never change. 

"As a result of my long experience in watching their effects on our 
students I am absolutely and irrevocably sure that certain subjects train 
in thinking straight and reasoning clearly. 

"I am absolutely sure that Latin and Greek, higher mathematics, 
philosophy, the critical study of the literatures of different nations (and 
the better the literature, the better the training it gives, Greek, Latin, 
English, and French literatures leading all others in this respect, and in 
the order named), economics and politics, especially on their theoretical 

217 



2iS EDUCATIONAL PSYCHOLOGY 

side, and Englisli composition are thinking subjects of very high educa- 
tional value." (Thomas. M. ("., "Old Fashioned Disciplines," Journal 
oj Uie Association oj Collegiate Aluvincc, May, 191 7, p. 5S8.) 

In connection with Uie Classical Conference at Ann Arbor, 
Michigan, kjoo, Dr. Harvey Wiley sent out questions to 100 
scientific men asking what they considered to be the value of Latin 
and Greek as preparation for scientific jjursuits. He received 35 
replies distributed as follows: 

Favorable to the study of Latin and Greek 14 

Unfavoriible to the study of l^atin and Greek 17 

Favorable to Latin, but not to Greek . .' 4 

Among the oj)inions expressed in this connection, Profe.s.sor 
R. P. Biglow made this statement: 

"To summarize my opinions in the matter of a scientific education, it 
seems to me that the essentials are of two classes: First, a thorough 
training in the use of the tools required by a scientific man, namely, the 
modern languages and mathematics; second, a training in the scientific 
method, es{)ecially as ajiplied [o the branch of science in which he desires 
to specialize. If to the curriculum, the studies of classics can be addeil 
without interfering with these essentials, then it seems to me that in some 
cases it would be desirable as a means of culture." 

Professor NefT of the University of Chicago regretted the time 
he spent on Latin and Greek: 

" I think everyone realizes as he grows older that he has his limitations. 
I, for one, regret very keeidy that I took a great deal of Latin and Cirtx>k 
and did not s[)end far more time on advanced mathematics antl physics. 
I am, however, not now wasting any time in vain or useless regrets on 
this account, but simply doing the best I can with the knowledge that 1 
have acquired." 

The opinions of prominent business men were rejwrted at the 
Siime ctmference: 

Mr. William Sloane, a New \'ork business man: 

"I believe that the slow processes of translation of the classics make 
goo<l training for the boy \.ho has chosen a business career." 

The Hon. J. W. Foster, of Washington: 

"The mere routine l.dwrs of the translation of Greek and Latin authors 
into one's vernacular, the elTorl to ascertain their exact meaning and the 



TRANSFERENCE OF TRAINING 219 

choice of the words which correctly express that meaning, constitute a 
mental training which will be invaluable to the future lawyer or public 



Probably the most notable assemblage of opinions ever brought 
together was that presented at the conference on classical studies 
held at Princeton University in June, 1917.^ These opinions were 
quoted from some 300 persons prominent in pubUc hfe, business, 
universities and colleges, schools, the ministry, law, medicine, 
engineering, science, journalism, modern literature, history and 
related branches, fine arts, and oriental studies. These statements 
were substantially unanimous in bearing testimony to the value 
of the classical languages. 

But perhaps as many opinions on the other side from men of 
equal intellect could be gathered. The point is that mere opinions 
cannot yield a final appraisal of the training value of school studies. 
Both favorable and unfavorable opinions are bound to be very 
nearly worthless because at best they are apt to be prejudiced by 
personal likes or dislikes and by exceptional instances of benefit 
or lack of benefit from the pursuit of this or that particular subject, 
and most of all because no general observer has at hand stifficient, 
rehable or complete evidence concerning the problem. Experi- 
mental and statistical data are hard enough to interpret because 
of the complication of factors in the production of any type of 
training to say nothing of the settlement of the controversy by 
general impressions. 

We carmot determine by ballot the shape of the earth, or the 
value of a patent medicine, no matter how many testimonials may 
be presented on the one side or the other. Men prominent in 
hfe have testified to the benefits of patented remedies which 
science has shown to be not only valueless but harmful. 

Specific Estimates of the Value of School Studies. The writer 
attempted to obtain specific estimates of the value of school sub- 
jects according to the best judgment that could be exercised by 
persons who are concerned with the work of public education. 
These were obtained not with a view to contributing anything 
toward the solution of the problem but for the purpose of examining 
more precisely the drift of the consensus of opinions held by persons 
immediately in charge of school work. 

Any branch of learning may have three possible values — a 

^ Reported in Value of the Classics, Princeton University Press. 



■220 EDLL.Vl lONAl, l's\ (, l l()l^()( .\ 

disciplinan* value, a utility value, and a cultural value. Thus 
1 the study of English has a certain amount of discij)linary value in 
training the mental capacities involved in the learning and under- 
standing of the material studied; it has an obvious utility value in 
acquiring the ability to use English correctly and efTectively; and 
it has a certain cultural value in acquainting the student with the 
thought and life of mankind. 

Estimates of the three values of each of the subjects listed in 
the table were made as carefully as possilile by fifty-eight super- 
intendents, principals and teachers. In making these estimates 
it was assumed that the pure, disciplinary value of the first year 
of high school English as taught in the average way be equal to lo 
and that all other values be estimated in terms of this assumption. 
If the disciplinary value of algebra was considered to be twice as 
great as that of English, hour for hour devoted to each, then it 
should be estimated as 20. Or if the utility or cultural value of 
English was considered greater or less than its disciplinary value, 
the rating should be indicated accordingly. It was further assumed 
that these were to be the values for the average boy or girl in the 
high school. 

At first glance it would seem that such judgments would be 
rather uncertain and variable, and, as a matter of fact, they were 
quite variable. Xevcrlheless, viewed from the statistical stand- 
point, the judgments present a nonnal distribution spreading over 
a wide range but clinging in large numbers about a central point. 
For exami)le, the judgments of the disciplinary value of American 
histor>' ranged as extremes from 3 to 30, with the largest number 
of estimates on 10 and a gradual decrease in the number of estimates 
on values farther and farther removed from 10. The median 
estimate was 10 and the i)robable error was 3.5, Uiat is, one-half 
of the estimates were between 8 and 15. So that the judgments 
about the various values, even though quite variable and usually 
accompanied by a feeling of uncertainty so far as the indi\i<lual 
judge was concerned, were as reliable and as normal as judgments 
about most matters arc. 

The following table gives the median judgment for each of the 
values listed: 



TI^VXSFEKENCE OF TRAINING 



221 



TABLE 55 
Estimated values of school subjects. After Starch ('17) 



Disciplinary 
Value 



Utility 

Value 



Cultural 
Value 



Totals 



Geometry 

Algebra 

Latin 

Physics 

Gymnastics 

German 

French 

Chemistry 

Manual Training 

Football 

Shorthand 

English (assumed) 

U. S. History 

Physical Geography 

Music 

Cooking 

Bookkeeping 

Civics 

Botany 

Zoology 

Drawing 

Sewing 

Typewriting 

Work of teacher or business 

man 

Earning one's way through 

school 



19 
17 
17 
17 
15 
13 
13 



19 

6 

21 

30 
18 



30 



30 

22 

36 



8 
8 

14 
12 

7 
12 

13 

II 

10 

5 

5 

22 

17 

ID 

25 

9 

7 



15 
10 

5 



38 
36 
41 
47 
36 
42 

37 
43 
45 
23 
2,1 
62 

45 
35 
46 

49 
42 
43 
id, 
32 
36 
49 
36 

73 
78 



An examination of the table reveals some interesting compari- 
sons. The highest disciplinary value is assigned to geometry with 
a rating of 20, or twice as high as that of English. Algebra is next 
with a value of 19. It is rather surprising to find gymnastics and 
football rated as high as they are. It is also interesting to note 
that the disciplinary value of a pupil's earning his way through 
school is rated higher than that of any of his studies. The lowest 
disciplinary value is assigned to sewing and typewriting. 

In the case of utility values, the highest rating is given to English, 
cooking and sewing (30), a value approximately three times as 
great as their disciplinary value. The lowest utility value is as- 
signed to football (6) and the next to algebra (g). Again the utility 



222 EDUCATIONAL PSYCIK )L()(.V 

value of a pupil's earning his way through school (36) is placed 
above that of any of his studies. 

In the case of cultural values, the highest rating is given to 
music (25) and the next to English (22). The lowest value is as- 
signee} to shorthand, typewriting, and football (5). The cultural 
value of a pupil's earning his way through school (18) is placed 
below only music and English. 

A similar study with reference only to the disciplinary aspect of 
college studies was previously reported by Thorndike ('15). Esti- 
mates obtained from 100 teachers are summarized thus: 

"Philosophy (for freshmen) 8; English composition, to; German, 
Chemistry and Logic, 11; Physics, 13; Latin, Greek, Mathematics, 16. 
Wuiting on table is rated at 3; athletics is rated at about 7; work for the 
college paper at 8 or 8; 2; tutoring at 13 or 13J2; and regular productive 
work in the world as teacher, business man or skilled laborer at 14K." 
(Page 281.) 

Such tabulations of opinions are valuable only in showing in 
more accurate terms what teachers and educators tliink about 
the question and not in really answering it. However, so long 
as schools are operated by opinions, a combination of opinions 
may be better than individual ones as guides of educational policies. 
'J'o what extent individual opinions are consciously or unconsciously 
prejudiced by personal interests is shown by the fact that the 
teachers overestimated by nearly one-half the value of their own 
specialties as compared with the average values assigned by 
teachers as a whole. 

Experimental and Statistical Inquiries. <7. Arilhmciic. The 
writer made an investigation to measure the effect of improvement 
in mental multiplication of three-place numbers by a one-place 
number, doing 50 problems a day for 14 days, upon other types of 
arithmetical processes. This experiment was carried out with 
eight subjects who constituted the training group, and seven 
subjects who constituted the control group. The results are given in 
the following table, which shows the percentages of gain of the 
second end tests over the first: 



TRANSFERENCE OF TRAINING 



223 



TABLE 56. After Starch Cii) 






Trained 
Persons 


Untrained 
Persons 


Differences 


Adding fractions 


40 
49 
-3 
58 
53 
3 
47 
45 

49 


12 
10 

— 2 

35 
29 

-5 
10 

25 
20 


28 


Adding thrpc-place numbers 


39 
— I 


Memory span for numbers 


Subtracting numbers 


23 

24 

8 


Multiplying four-place numbers 


Memory span for words 


Multiplying two-place numbers 


37 
20 


Dividing three-place numbers 


Averages, exclusive of memory span 


29 



%9 Ci 

The residual gain on the average was 29%. The average gain 
made by the trained group in the practice series, comparing the 
first day with the 14th day, was 112%. Hence the gain trans- 
ferred to the alHed arithmetical operations was only 26% of the 
gain in the practice series itself. From one point of view, this 
seems to be a very considerable amount of transfer, but when we 
note that some of the end tests were as similar to the training 
series as they could be without being identical with it the transfer 
is small. We might expect almost a complete carrying over to 
the closely similar operations but the largest amount of residual 
gain took place in the multiplying of two-place numbers by a 
one-place number and in the adding of three-place numbers; but 
even there it was only slightly larger than the transfer to the other 
operations. 

Winch conducted a series of experiments to determine the amount 
of transfer from improvement in numerical computation to arith- 
metical reasoning. In each experiment the class was divided into 
two groups of approximately equal ability as shown by a previous 
test in arithmetical reasoning. Then one-half of the class was 
trained in "rule" sums after which a final test in arithmetical 
reasoning was given alike to both groups. 

The first class, composed of 13-year-old girls from a poor neigh- 
borhood, showed improvement in numerical accuracy but no 
transfer to arithmetical reasoning. The second class, composed of 
lo-year-old girls from a poor neighborhood, showed considerable 
improvement in accuracy and a doubtful transfer to reasoning. 
The third class, composed of lo-year-old girls from a good neigh- 



224 r.Dl'CATIONAl. I'SVCIIOLOdV 

borhood l)ut ])Oor in arithmetic, showed transfer in three sections 
but the opi)osite in a fourth section. In the fourth class, composed 
of lo-year-old boys, both practiced and unj^racticed p:roups showed 
about equal gains in arithmetiad reasoning. Winch concludes 
I from these ex])eriments that improvement may take place in 
numerical computation \vithout any certainty of improvement in 
I arithmetical reasoning. 

In a later experiment conducted on the same general plan, he 
used 72 ten-year-old boys. Here again he found no evidence of 
transfer from impro\ement in numerical computation to arith- 
metical reasoning. 

Carrie R. Squire made an experiment regarding the transfer of 
neatness: "At the Montana State Normal College careful experi- 
ments were undertaken to determine whether the habit of jiroduc- 
ing neat ])apers in arithmetic will function in reference to neat 
written work in other studies; the tests were confined to the 
intermediate grades. The results are almost startling in their 
failure to show the slightest improvement in language and 
spelling jxipers although the iriiprovement in arithmetic ])apers 
was noticeable from the first." (Bagley, The Educative Process, 
p. 208.) 

This ex]^eriment was repeated under the direction of Ruediger 
with the dilTerence that along with the s])ecific training in neatness 
in one particular study a general ])ractice of neatness in daily life 
was held uj) before the class as an ideal to be striven for. Care was 
taken not to discuss neatness in the other classes. Sample papers 
were taken in the one subject concemed and in two other subjects 
before and after the training. The seventh grades of three schools, 
located at widely dilTerent jilaces and comi)rising 39 ])upils in all, 
were used in the experiment. The two schools which shoA\ed 
an appreciable improvement in neatness in the study where s])ecial 
training was given -also showed considerable, though less, impro\e- 
ment in the other studies. Thus the two schools showed an average 
improvement of 4.75 ])oints in the study where training was given- 
and 3.1 points, or Os^/'o ^s much, in studies where nothing was said 
about neatness. 

b. Grammar. Grarrunar has been regarded as a highly cfTica- 
cious instrument for training the functions of the mind. Thus 
Commenius stated: " I j^rcsume that no one can raise any objection 
to my ])lacing (Latin) grammar first, since it is the key of all knowl- 
edge." l.ocke saiil on the other hand: "I would fain have anyone 



TRANSFERENCE OF TRAINING 225 

name to me that Tongue, that anyone can learn, or speak as he 
should do, by the rules of Grammar. Languages were made not 
by Rules, or Art, but by Accident, and the Common Use of the 
People." 

Claims for Grammar have been that it 

1. Disciplines the mind. 

2. Prepares for the study of other languages. 

3. Gives command of an indispensable terminology. 

4. Enables one to use better English. 

5. Aids in the interpretation of literatures. 

The Committee of Ten (1893) said: "The study of formal gram- 
mar is valuable as training in thought, but has only an indirect 
bearing on the art of writing and speaking." 

What are the actual facts so far as any are available at the 
present time? Briggs ('13) attempted to determine the extent to 
which the various claims made for grammar are substantiated. 
He outlined the following claims and devised an elaborate set of 
tests to measure the effects of training in grammar. 

"It is held that grammar trains children; 

A. With rules and definitions: 

1. To see likenesses and diflferences. 

2. To critically test a definition. 

3. To thoroughly apply a definition. 

4. To make a rule or definition. 

B. With reasoning: 

5. To test reasons. 

6. a. To take from a mass of data all that are necessary 

and to use them in reaching a judgment, 
b. To demand all necessary data before drawing a 
conclusion. 

7. To reason in other fields, e, g., arithmetic. 

8. To reason syllogistically. 

9. To detect "catches." 

As illustrations of the nature of the tests we may cite the follow- 
ing instances. For measuring the observation of likenesses and 
differences, Briggs used such a test as this: 

"Oue-half of the following 16 words are alike in one respect and in that 



226 EDUCATIONAL PSYCHOLOGY 

respect unlike all the others in the list, rind these eight words and mark 
them with a check (V)." 



biscuit 


pirate 


mountains 


men 


oxen 


geese 


fathers-in-law 


factory 


scholars 


knives 


vessel 


table 


pole 


frame 


children 


mice 


(8 are plurals) 









To determine ability to judge definitions and to amend them, 
he used as tests such statements as the following for shoe: 

1. A portion of clothing. 

2. Something hhick made of leather. 

3. Something to wear on the feet. 

4. A necessary article costing from $1.00 to S5.00 or $6.00. 

These tests were given to 25 or 30 pupils in each grade from two 
to seven in the Horace Mann school. Each class was divided into 
two divisions. Then for three months, three times a week, the 
children of Division I were taught formal grammar. During the 
same three months, the children of Di\ision II had work in com- 
I)Osition and language. They were then given the second set of 
tests similar to the preliminary tests, after which the conditions 
were reversed. Division II then had formal grammar and Divi- 
sion I had language and composition work. At the conclusion of 
this period, the first set of tests was again given to all of the chil- 
dren. The upshot of the whole investigation is simimarized by 
Briggs in the following manner: 

".•\s a result of this experiment it may s;ifcly be asserted that these 
particular children at'ler the amount of formal grammar that they had, 
do not, as measured by the means emi)l()vcd, show in any of the abilities 
tested, improvement that may be attributed to their training in formal 
grammar." 

Hoyt ('06) made a study to determine the relation between the 
knowledge of grammar, ability to interpret English, and ability 
to write English. He cm[)loyed three tests: One for grammar 
consisting of ten questions on four stanzas of Gray's IHt'^y; the 
second for testing ability to inteqiret English, consisting of a 
statement of thought in four other stanzas of Gray's Elegy; and 
the third for ascertaining ability in com])osition, consisting of 
writing a composition in forty minutes. These tests were made 
with 200 pupils in a high school in IndianajKilis. All papers were 



TRANSFERENCE OF TRAINING 227 

marked by two examiners according to the percentage method. 
Correlations were then computed among the different tests, which 
were as follows: 

Grammar and composition 23 

Grammar and interpretation 28 

Interpretation and composition 32 

These coefficients are very low and indicate that a greater or 
less amount of knowledge of grammar is accompanied to only a 
slight extent by greater or less ability to write a composition. The 
same statement holds for the relations between the other compari- 
sons of abilities. The fact that the pupil who knows more or less 
grammar writes respectively a slightly better or worse composition 
is quite likely due to the fact that he is a better or poorer pupil 
rather than to any aid which knowledge of grammar may render 
him in writing a composition. 

Hoyt concludes that "... the teaching of grammar is of 
little avail in strengthening one's ability to use language." 

The writer ('15) made a series of tests in formal grammar and 
in correctness of English usage. The test in formal grammar 
consisted of three parts: First, a passage in which the parts of 
speech of as many successive words as possible were to be indicated 
in three minutes; second, a passage in which the cases of nouns 
and pronouns were to be indicated in three minutes; and third, a 
passage in which the tenses and modes were to be indicated in 
three minutes. The test for usage consisted of 100 sentences each 
of which was stated in two ways. Both might be correct, both 
might be incorrect, or one might be correct and the other incorrect. 
Pupils were allowed fifteen minutes in which to indicate the cor- 
rect expressions. The results are summarized in the following 
tables. 

The tests were made upon 54 university Juniors and Seniors and 
146 high-school pupils. They gave the results shown in Table 57, 
in which the scores for knowledge of grammar are the numbers of 
the parts of speech, tenses, cases, and modes iidicated correctly 
in the specified period of time, and the scores for correctness of 
usage are the i:^imbers of sentences designated correctly in the 
specified period of time. 



)/ 



228 



EDUCATION'AI. I'SVCIIOI.OfiV 



TABLi; 



After Slarcli ('15) 



Years op Foreign 


NlIUIIKR OK 


AvKK\r.E S<llRKS 


FOR 


AVERAOK fVoRrS FOk 


L.\NCUAGES 


SlU DENTS 


KNOMUiUOE OF GK/XMMAK 


CoRKLCINEbS OF UsiACl- 




Umvkrsity 


Juniors 


AND Sk 


MORS 







2 




48.0 




81.5 V , 


2— S 


12 




47.8 




7I.I ^ 


6-9 


25 




58.6 




75-5 


10— IS 


15 




63 -4 




7S-7 




IIu;u 


School 


Pupils 









12 




14-7 




32.2 


8 weeks 


SO 




20. S 




430 


I year 


i8 




25-5 




43-4 


2 years 


39 




24. S 




45-9 


3 years 


27 




28.6 




47-7 




University 


Juniors 


AM) Sk 


MORS 




Years of Latin 















15 




45-8 




70.9 


1—3 


II 




56.1 




75-7 


4 


14 




57-5 




74-3 


5 or more 


9 




51-8 




76.1 



Another test for correctness of usa^e, consisting of sentences 
like the set of one hundred l)ut ammj^ed in the order of increas- 
in^dy difficult stej^s, Avas made on another srouj) of 146 university 
students and 92 hi^h-school jjupils. This test yielded the results 
^dven in the folloAvin^ table. The scores are the numbers of the 
hij^hest stejjs jjassed. The hi<,dier the score is, the greater is the 
ability of using English correctly. 



Years of Latin 



16 



o 
1-4 



TABLE 58 




Ni.MBER OF Pupils 


AVERAOE SrORES 


Umvkrsity Stuui:nts 




47 


10. I 


99 


10.3 


IIic.ji .School I'riMi^ 




-s 


90 


14 


9 3 



These tables agree in sliowing one very significai^.l nsult, namely, 
that the study of foreign languages materially increases a ])U|)irs 
knowledge of English grammar but only slightly increases his 
ability in the correct usage of the English language. Notice, for 



TRANSFERENCE OF TRAINING 229 

example, the upper part of Table 57. The students who had 
10 to 15 years of foreign languages made a score in grammatical 
knowledge of 63, as compared with a score of 47.8 made by the 
students who had 2 to 5 years of foreign languages, a difference of 
32.6% in favor of the former group. For correctness of usage, the 
corresponding difference is only 6.4%. The two students with no 
foreign languages made high scores because they were exception- 
ally good students, but they are too few in number to be con- 
sidered. The high-school pupils show a gain in grammatical 
knowledge of 37.5% from the 8-week group to the 3-year group 
and a gain in usage of only 10.9%. The twelve pupils with no for- 
eign language made low scores because they were exceptionally poor 
pupils. This is indicated by their low scholarship records, by the 
fact that many were over-age, by the fact that they avoided the 
foreign languages, and also by the large difference between their 
scores and those of the 50 pupils who were just beginning foreign 
languages. Eight weeks of foreign languages could hardly have 
produced such a big gain. Their higher scores must be due largely 
to a difference in original nature. 

c. Foreign Languages. Extensive and confident claims have been 
made for the value of general mental training to be derived from the 
study of languages. Thus Lodge states the value of the study of 
Latin as follows: 

"Far above every other subject it trains (i) the process of observation, 
(2) the function of correct record, (3) the reasoning power and general 
intelligence in correct inference from recorded observation. To this 
should be added its great value in developing the power of voluntary 
attention. 

"The value of Latin as a practical subject has to do particularly with 
the effect of the language in the cultivation of English style. In the 
English vocabulary a very large proportion of words in everyday use are 
of Latin origin, and it has been estimated that two-thirds of the Latin 
vocabulary of the classical period has in some form or other come over 
into English speech. For the correct use of synonyms in English and 
the habit of expressing one's thoughts clearly, concisely, and cogently, a 
discriminating knowledge of Latin is indispensable, and while not every 
pupil in the school may be expected to develop a good style, nevertheless 
he should be given the necessary foundation for it. 

"When we turn to literature, we find that Latin is influential every- 
where — particularly in our classical authors— by allusions, by quota- 
tions, by actual domestication. Many of our great English writers are 
permeated with Latin. We cannot expect that all will desire to feed 



2;o 



KDICATKJ.NAL l'SVCH(JL()(;V 



ihcir minds on the works of our greatest authors, however much we 
might prefer it; but certainly we should not deprive them of one of the 
most inifKjrtant elements in their enjoyment should they Ije so minded." 
(Lodge, J). 3S8, in Principles of Secondary Education, Edited by Paul 
Monroe.) 

Swift measured the progress in learning a new language made by 
pupils with ditTerenl amounts of jjrevious language study to de- 
tennine, if possible, the advantage in beginning a new language to 

100% 



90 



80 



txiTD 



'60 



60 



40 





1 






























y 


s^ 


















A 


/ 







/ 


\ 


s* 




/ 


\ 










\ 


V 


7 


N 




V, 


\> 


/ 


\ 


/N 


. I 






r 1 


^ 


/ 




S 


^.^ 


/ 




V/i\ 




/ 














\ 


/ 




\ 


/A\ 




n 






/ 


k 




/^ 


'c" 


\ 






/ 


\ 






\ 


^ 




\ 


/ 






\ 


y' 


'■-V 


f 






/ 




























/ 




























/ 






A -One Year of Latin and German. 
B-Onc Year of Latin. 
C- Spanish Only. 


















































1 

1 



















i 4 



U 7 8 9 10 11 L2 13 U 16 
Weeks 



Fig. 55. — Progress of pui)ils in learning Spanish, .\ftcr Swift ('06). 

be derived from the {)revious study of another language. The ex- 
periment was carried out with two classes, composed of 24 boys 
and 24 girls in a St. Louis high school, who were beginning the study 
of Spanish. Weekly tests were made to measure their progress and 
individual abilities in learning Spanish. The classes were taught in 
the usual manner and the pupils knew nothing of the puq^ose of the 
tests. A record of progress was kept for the first 15 weeks. The 
results are shown in the following grai)lis (Figure 55), which in- 
dicate the relative progress of the iJiree groups, namely, those who 
had previously studied one year of Latin and one year of German, 



TRANSFERENCE OF TRAINING 23 1 

those who had only one year of Latm, and those who had never 
studied a foreign language before. The "Latin" and the "Latin 
and German " groups stood considerably higher at the beginning 
than the "Spanish only" group, but the "Spanish only" group 
gained gradually so that at the end of the 15 weeks it had made up 
about two-thirds of the difference. Swift concludes: 

"The number included in these tests was too small to serve as a basis 
for anything more than tentative conclusions, but the results certainly 
open the question whether the advantage to beginners of a new language, 
so generally thought to accrue from the study of Latin, may not be due, 
chiefly if not solely, to grammatical information that would be carried 
over from one language to another, and which would naturally help 
enormously at the start. In acquiring faciUty in the use of the Spanish 
gender, to cite one example, Latin would aid materially, since the ma- 
jority of Latin feminines are feminine in Spanish, and a large part of 
Latin masculines and neuters become masculine in Spanish. The de- 
clension of Spanish adjectives for gender and number, and their agree- 
ment, in these respects, with their nouns, would give Latin students a 
further advantage. The teacher of the Spanish classes noted that more 
frequent and detailed explanations of case were needed by those who 
had not studied Latin. The order of words, also, was more readily 
mastered by those famihar with the Latin arrangement. Finally, in 
learning the conjugations and in understanding the significance of tenses, 
the assistance of the information acquired under these topics in Latin 
was found to be especially great. The indications, however, are that the 
higher records made by the Latin and German pupils were the result of 
the substance of language information obtained from these studies rather 
than of any so-called 'language' or 'mental discipline.'" (Swift, '06, 
pp. 250 £E.) 

The writer ('15) made a comparison of the scholastic records 
of university students who had entered the university with two 
to four years of Latin with the records of those who had entered 
with two to four years of German. The average grade for the 
four years of college work of each of the graduates of the Col- 
lege of Letters and Science of the year 1910 was computed. The 
median mark of the 104 students who had entered the university 
with Latin was 85.7 and the median mark of the 45 students who 
had entered with German was 84.0. Hence the difference between 
the two groups was only 1.7 points. 

The explanation for this small advantage of Latin over German 
may be sought in three directions : First, the disciplinary difference 



232 EDUCATION'AI, I'SYfllOLOGY 

between Latin and German is cither zero or very small. Second, 
whatever difTerence they may have produced originally may ha\e 
tended to disappear in the four years of college work, owing to the 
freedom of electives, pursuit of dilTcrcnt courses, disciplinary cfTect 
of other studies, etc. Third, the small dilTcrencc in scholastic 
records may be due to an original dilTerence in the students them- 
selves, owing to the possibility that one language may attract a 
better class of pupils than another. 

To determine what part, if any, tlic first two factors played, the 
average grade of each of the 7.^8 P'reshmen of the year 1909-1910 
was computed. The median grade of the 416 Freshmen who had 
entered with Latin was 82.4 and that of the 322 Freshmen who had 
entered with German was 81.0. Hence the dilTerence between the 
two groups was only 1.4 points, or approximately the simie as that 
for the graduates. 

The next problem was to compare the grades of these two groups 
in specific subjects as follows: 

TAHLK 5(). After SUirc'.i ('15) 

Median f^ruflc in modrrn l;inKiiaf;cs of 362 I-'rcshmcn who hiid entered 

with Latin <S4 . 5 

Median ^nifle in modern ianj^iiagcs of 293 l-'reshnicn wlio had entered 

with Cicrman 82 .3 

DifTerence in favor of tlie Latin Kroup. . 2.2 

Median j^rarle in I-reshman J'in>;iish of 54 stu(leni?> wiio had entered witli 

Latin only 83 . 9 

Median j,'rade in Kreshman English of 97 students who had entered with 

(icrman only 82.7 

Difference in favor of the Latin Rroup. . 1.2 

Median i;radc in first-year l-'renc h of 27 I-resiinun wlio had intered with 

Latin only Si .5 

Median >,'radc in first-year Lrem h of 34 Lreshmen who had entered with 

(ierman only 82.0 

DilTerence in favor of the (ierman j^roup 5 

The differences again are very small. The claim of language 
teachers, so commonly made, that beginners in French who have 
had Latin are much superior to those who have not had Latin, or 
that studiiits in I'.nglish with jirevious training in Latin are suiK'rior 
to those wilhoiil such training i'^ ill founded. 

.\nollier tabulation was made to show the scholarshij) records of 



TRANSFERENCE OF TRAINING 233 

Freshmen in relation to the amount of foreign language studied, 
irrespective of what the languages were. 

TABLE 60. After Starch ('15) 

Years of Foreign Number of Median Grade in all 

Languages Students Freshman Studies 

o 25 81.8 

1-2 224 81.9 

3-4 195 83.05 

5 ''> 155 84-0 

The next problem was to measure the extent to which a pupil's 
English vocabulary is increased through the study of Latin. The 
method employed measured the percentage of words of the entire 
English vocabulary, as well as the approximate absolute number of 
words, whose meaning a person knows. The test was made with 
189 university students and with 46 Juniors in the Madison High 
School. 

TABLE 61. After Starch ('15) 

'. . . . "^^ 

Size of EngUsh vocabulary of 139 university students who had studied 

Latin 60.9 

Size of Enghsh vocabulary of 50 university students who had not studied 

Latin 58.2 

Size of Enghsh vocabulary of 14 high-school Juniors who had studied 

Latin 54 . 7 

Size of English vocabulary of 32 high-school Juniors who had not studied 

Latin 50.2 

The differences between the Latin and the no-Latin groups are 
surprisingly small. Nevertheless, the study of Latin does produce 
an appreciably larger English vocabulary. This advantage becomes 
less in university students with whom it is partly counterbalanced 
by the increase in vocabulary due to wider experience. 

Partridge compared the standing in the regents' third year Eng- 
lish examination of 783 pupils by dividing them into groups ac- 
cording to the number of years of Latin they had studied. His 
tabulation is as follows: 

TABLE 62. After Partridge ('15) 

The entire 783 papers divided on a basis of the number of years Latin was 

studied 

Number years studied o 

Number papers written 181 

Average standing (percentage) .... 65 



I 


2 


3 


123 


220 


2.S9 


65 


69 


76 



234 



EDUCATION'AL PSYCITOLOf^V 



The pupils with no Latin may haw liad oni' or more years of 
other languages and conseciuently I'artridge presents the following 
table for students who had studied Latin onI\': 



TABLE 63. After Partridge ('15) 

Includes papers of ptipils having Latin only and no other lanpiiage to their 
credit. Tulal number of papers, 167 

No. of years studied o 1 2 3 

No. of papers written 28 25 42 72 

Average standing (per cent) . . 63 61 69 78 

TABLE 64. After Partridge ('15) 

Inchides papers of pii[)ils having German or Frcm h only and no other lan- 
guage to their credit. Total number of papers, 17O 

No. of years studied o ' i 2 3 

No. of i)apcrs written 28 41 57 50 

Average standing (percentage) 63 61 65 68 

Partridge believes that the "superiority of the classical over the 
non-classical pupils is due not solely to initial natural ability, but 
to the training received in Latin." He has, however, failed to show 
the difTcrences in initial ability and consequently any inference of 
this sort is doubtful. 

Harris made a study of the effect of knowledge of Latin upon 
ability to spell English words by submitting a list of 50 words of 
Latin origin to 324 freshmen in the University of Illinois. He gives 
the following table: 





TABLE 65. After HaFris (' 


IS) 






\kaks of 


.ATIN 





I 


2 


.( 


1 


No. of students 

Average 


00 
82.1 


41 

Sj 4 


05 
So . 2 


54 
S I . 5 


44 
90. 1 



He furthcf submitted to the same group of students 10 words 
of Latin origin which were to be delincd. This test gave the follow- 
ing result: 

TABLE 66. After Harris ('is) 



\ KAkS 111 


1 . \ 1 1 N 





1 


- 


.< 


1 


No. of students. . 
Average 




90 
30-5 


41 
44.2 


95 
45 9 


54 
S3 


44 
853 







' It is ol)vic)iis that the o cohimns in those two lalili-s will ix)ntain record of exactly 
the same puijils. 



TRANSFERENCE OF TRAINING 



235 



He also compared the grades in Rhetoric of students who had had 
various amounts of Latin as follows: 

TABLE 67. After Harris ('15) 



Years of Latin 





1 


2 


3 


4 


No of students 


53 

77.2 


41 
79.2 


66 
795 


28 
80.6 


26 


Grade 


81.8 







Harris concludes: "From these various results the conclusions 
in so far as these students are concerned, are obvious. In all fields 
the four-year Latin students showed a marked lead, and in all but 
the spelling — which I have considered above — there is a steady re- 
trogression although for the practical purposes the one-and-two- 
year Latin students might be classed together." 

The interpretation of these figures is by no means so obvious. 
Harris has made no allowance for the native superiority of the stu- 
dents with more years of Latin study. In fact, the probability is 
that, if we may infer from other studies in which such a deduction 
has been made, a large part of the superiority is due to original 
nature. Harris's results as they stand prove little or nothing con- 
cerning the effect of training in Latin. 

F. M. Foster performed a similar experiment at the University of 
Iowa; 503 freshmen, about equally divided between the sexes, were 
given a spelling test of forty words of Latin derivation. The results 
are given in the following table: 



TABLE 68. After Foster ('17) 

Number of years of Latin o i 2 3 4 

Average % of errors (girls) 23 28 25 24 17 

Average % of errors (boys) 39 37 29 28 27 

As in the study inade by Harris there appears to be a decided re- 
lationship between ability to spell words of Latin derivation and the 
number of years devoted to Latin. In this case, however, it hap- 
pened that Professor Irving King had previously given intelligence 
tests to these same students by means of which it is possible to se- 
cure a more accurate notion of the forces really producing the 
better scores of the Latin students. The following table shows the 
relation between the index of intelligence (that is, the percentage 
above or below the average adult intelligence), the number of years 



25-8, 


spelling av. 5% error 


-9 5, 


" 33% " 


13-4, 


" " 7% " 


-12.4 


" 44% " 


ig.i, 


" " 11% " 


— 14.2, 


" " 34% " 


50, 


" " 23% " 


-27.2, 


" " 63% " 



236 KDUC'.vnoNAi, l•s^(■^()l.()(;\■ 

spent in the study of Latin and the ability to spell words of Latin 
derivation for the two extreme Latin groups. 

TABLE 69. After Foster ('17) 

Best 14 of 4-ycar Latin Rirls, mental al)ility av 

Poorest 13 " " " " " ' 

Best 12 " " boys " " ' 
Poorest 12 " 

Best 10 of o-year girls " " ' 

P(K)rest 10 " " " " ' 

Best 10 " boys " " ' 

Poorest 10 " " " " 

This table shows clearly that students who chose to study Latin 
had on the average a distinctly better nali\e intelligence than the 
non-Latin students and that the ability to spell words of Latin der- 
ivation was to a considerable extent due to this superior intelligence 
rather than to the study of Latin. 

The secretary of the College Entrance Examination Board made 
an extensive tabulation of the records of the classical and the non- 
classical students who took the examinations in 1914, 191 5, and 
1916. The classical students arc the ones who ofTered Latin or 
Greek, or both; non-classical students are those who ofTered neither 
Latin nor Greek. A total of 21,103 candidates arc concerned in the 
following table which is based on the marks in all subjects except 
Latin and Greek (reported in V^dut 0/ Classics, 191 7, p. 366): 

Combined Ratings in All the Non-Classical Subjects 

Candidates who obtained a rating of 90 to 100: 

2.05% of all the < lassical ramiidates. 
2.05/0 of all the non-tlassicul candidates. 

The classical students show a superiority of 44%. 
Candidates who ol)lained a rating of 75 to 89: 

17 31% of 'i" til*" classical candidates 
12.31% of all the non-classical candidates. 

The classical students show a superiority of about 40%. 
Candidates who obtaineil a rating of (k> to 100: 

51 .06% of all the classical candidates 

40 97% of ^" till-" non-clahsical cantiidates. 

The classical students show a suiwriority of about 27%. 



TRANSFERENCE OF TRAINING 237 

A committee in connection with the Princeton Conference made a 
comparison of the honors received by classical and non-classical 
students upon graduation from high schools, academies and colleges. 
The table is based upon 2,799 classical and 5,606 non-classical 
students from 19 high schools and academies, and upon 4,092 
classical and 2,003 non-classical students from 17 colleges and 
universities: 

"The combined data from the nineteen high schools and academies 
reporting yield the following results: 

"Students receiving High Honors at Graduation were 18% of all the 
classical students, but only 7.2% of all the non-classical students. 
"That is: the classical students show a superiority of 150%. 

"Students receiving Honors at Graduation were 32.1% of all the 
classical students, but only 30.8% of all the non-classical students. 
"That is: the classical students show a superiority of 36.7%. 

"Students receiving Honors or Prizes for Debating, Speaking or 
Essay- writing were 8.8% of all the classical students, but only 3.5% of 
all the non-classical students. 
"That is: the classical students show a superiority of 150%." 

"The combined data from the seventeen colleges and universities 
reporting yield the following results: 

"Students receiving High Honors at Graduation were 17.3% of all 
the classical students, but only 6.6% of all the non-classical students. 
"That is: the classical students show a superiority of 162%. 

"Students receiving Honors at Graduation were 46.5% of all the 
classical students, but only 38.5% of all the non-classical students. 
"That is: the classical students show a superiority of 20.7%. 

"Students elected to Phi Beta Kappa were 16.8% of all the classical 
students, but only 8.9% of all the non-classical students. 
"That is: the classical students show a superiority of 88.8%. 

"Students winning Prizes or Honors for Scholarship in Other than 
Classical Subjects were 13.5% of all the classical students, but only 
q.3% of all the non-classical students. 
"That is: the classical students show a superiority of 45.2%. 

" Students serving on the Editorial Boards of Student Newspapers and 
Magazines were 15.1% of all the classical students, but only 9.2% of aU 
the non-classical students. 
"That is: the classical students show a superiority of 64.1%. 

"Students acting as Members of Intercollegiate Debating Teams were 
5.1% of all the classical students, but only 2,--% of all the non-classical 
students. 
"That is: 
in Valiie of Classics, pp. ,381-383.) 



238 



EDUCATrON.NL I'SVCIIOLfK^.V 



These statistics arc interesting; enou/rh, l)ut they represent a 
mingling of training and native ability in the superiority shown and 
as such do not afford conclusive proof for the efficacy of classical 
training. 

Wilcox ('17) made an inquiry with the endeavor to ascertain the 
amount of difference in original capacities. If the superiority of the 
Latin students is due to their study of Latin, we ought to find that 
they were not superior, or at least not as much superior, to the 
other students before they undertook Latin. If, however, we should 
find that the Latin students were as superior before they studied 
Latin as aftenvards, we may infer that Latin had nothing to do 
with their superiority. 

Wilcox tabulated the records of pupils in the Io^\•a City High 
School graduating during a period of ten years. He tabulated 
separately the grades made ])y all the students with Latin or Ger- 
man, but not with both. These results are shown in the following 
table in which the numbers were obtained by transposing the 
symbols E, G, M, P, and F into numerical values of 4, 3, 2, i, and 
o respectively. 

TABLE 70. After Wilcox (*i 7) 

Median grades in English of Towa City High School students who studied 

Latin or (icrman 



SopnoMoRj; 



JlNIiiK 



Sknkik 



AU Latin (184). . . 
All (lerman (120) , 
Diflcrence 



6.21 

4-93 
1.28 



6. 29 
5 29 



5-8o 
1.08 



6. II 
4.92 
I 10 



The comj^arison is graphically shown in Figure 56. It will be 
noticed that the superiority of the classical group is found in the 
freshman year and continues throughout the course. 

A comparison was also made of the Knglish grades of students 
ha\ing four, three, and two years of Latin. This is sho"s\ii in 
the following table: 

TABLi: 71. After Wilcox ('17) 

Median grades in Ilnglish of Iowa City High School students who had 4, 3, 

and 2 years of Latin 





I'RKSHUAN 


.Sol'llOUOHK 


Jl NIOR 


Senior 


4 yrs. Latin (31) 

3 yrs. Latin (27) 

2 yrs. Latin (126) 


7 14 
6.33 
5 9'' 


7-37 
6.56 
5 93 


6.70 
6.30 

5 39 


7-33 
6.4s 
5 SO 



TRANSFERENCE OF TRAINING 



239 



A graphical comparison is shown in Figure 57, It is evident 
that those who were destined to take four years of Latin were 
already in their freshman year clearly superior to those taking 
less Latin. 

Wilcox made a similar investigation of the graduates of the 
high school of Cedar Rapids, Iowa. 









t^atin. 


^. 




i^ 













12 3 4 

Years of High School 

Fig. 56. — Median grades in Eng- 
lish in the four successive high-school 
years of students taking Latin or Ger- 
man. After Wilcox ('17). 




1234 
Years of High School.> 

Fig. 57. — Median grade in Eng- 
lish in the four successive high-school 
years of students who had 4, 3, or 2 
years of Latin. After Wilcox ('17). 



Comparisons were made of the grades of 150 graduates having 
Latin, German or no foreign language. This is shown in Table 
72. 

TABLE 72. After Wilcox ('17) 

Median grades in English of Cedar Rapids High School students who studied 
Latin, German or no foreign language 





Freshman 


Sophomore 


Junior 


Senior 


All Latin (70) 


91 
86 
86 


89 
86 
83 


89 
8S 
84 


86 


All German (60) 

No foreign language (30) . 


87 



It will be observed that in the freshman year the classical group 
was superior to the other two, but that by the senior year there was 
very little difference in the three groups. 

In Table 73, comparison is made of the English grades of stu- 
dents having four, three or two years of Latin. 



240 



EDUCATIONAL PS^'^rIOLO(;V 



TABLE 73. After Wilcox ('17) 

Median prades in English of Cedar Rapids IliKh School students with 4, 3, 
or 2 years of Latin 





FkKSIIM AN 


Soi-noMoiiK 


Junior 


SeVIi R 


4 years Latin (ig) 

3 years Latin (g) 

2 years Latin (38) 


94 

87 

go 


94 

88 
89 


93 
87 
88 


92 

85 

. 84 



Here apjain the ])eo])k' with four years of Latin maintain tlu ir 
su])(.riorily throughout the course. 

" Conclusions: It seems evident, so far as the Iowa City and Cedar 
Rapids high schools are concerned, that the frequently demon- 
strated su])eriority of students who ha\-e had Latin is not due to the 
special discipline or training secured in the study of Latin. It is 
probably due to the fact that, as a whole, the students who elect 
Latin are somewhat superior to those who refuse to take it." 

Perkins ('14) made an investigation to determine the effect of 
emphasizing the derivation of English words from Latin words in 
the instruction of Latin in the commercial course in the Dorchester, 
Massachusetts, high school. This investigation was designed to 
eliminate as far as possible the dilTerences in original abilities 
between pupils with Latin and without Latin. His report follows: 

"Obviously, the first step was to select two sets of pupils of equal 
ability, one set in the second year of Latin, and the other in the second 
year of a modern language. .Vccordingly. we chose pupils such that each 
groui) had virtually the same average mark in Latin, on the one hand, 
an<l modern language, on the other, and also in English, with the result, 
in actual figures, that the non-Latin group in the two studies averaged 
0.5 of iSo tli<^' higher. To make doubly sure that the Latin pupils were 
not favored, the non-Latin group were taken from the section of Mr. Mur- 
dock, a classical scholar, who in his English leaching emphasizes the 
Latin element in the language. There were twenty-one pupils in each 
set, all in the second year chiss of the school. - 

"Five measurements were made, one in spelling, one of the use of 
words in sentences, the third in definitions and parts of speech, the 
f(nirlh in the meaning of words anil siK-lling, and the liflh in excellence 
in vot abulary. 

".Mis» Humphrey selected the words in \os. 1-4, and the subject in 
No. 5. In Nos. I and : the words were taken from the 600 or Soo deriva- 
tives in the nolelKH)ks of a fourth-year pupil of the class, who w;us ex- 
cluded from the measurements. Moreover, to be fair to the non-Latin 



TRANSFERENCE OF TRAINING 24 1 

group, care was taken not to select words too difficult. In No. 3 the 
words were taken from the 'Tale of Two Cities' which the pupils of both 
groups were reading at the time in connection with their work in Eng- 
lish II. Of the twenty words in No. 4, ten were taken from the 'Tale of 
Two Cities' and ten from other sources. The subject in No. 5 was, 
'What I like to do best.' The papers were marked by teachers in the 
English department and the results given to me. Altogether, six teachers 
of English assisted in the measurements. 

"To these five measurements is added a sixth — in my opinion most 
impressive of all. This test was made last June, shortly after I had 
received Professor Holmes's letter, by ]\Iiss Gormley, with her pupils in 
English II. As it happened, IMiss Gormley, who was also the 'home- 
room' teacher of all the pupils and consequently had access to their 
marks, in making up the two groups to be composed of pupils of equal 
ability, took into account not only foreign language and English II, as 
was the case in measurements 1-5, but also the studies the pupils had 
taken during the year. Hence we have even more reason in this case 
than in the others to assume that the pupils were of equal ability. In 
each set there were seventeen second-year students. The words were 
taken entirely from Franklin's Autobiography and Silas Marner which 
all were reading at the time. The Latin pupils were selected from the 
first class I had had in the subject, just as they were completing the 
course at the end of the second year. 

"The result of the six measurements were as follows: 

Averages 
Latin Non-Latin 

Per Cent Per Cent 

January and February, 1914: 

1. Spelling 82.5 72.6 

2. Use of words in sentences 57-5 40.6 

3. Definitions and parts of speech 69.5 HZ 

4. Meaning of words and spelling 57 -o 27.5 

5. Excellence in vocabulary 36.0 6.8 

June, 1913: 

6. Meaning of words and spelling 63.3 12.3 



367-8 193 I 

Averages 61.3 32.18 

32.18 
Difference 29.1 2% 

"In No. I, the spelling measurement, the words were not difficult, 
but such as ordinary pupils of sixteen should know something about, 
whether they had studied Latin or not — as 'valedictory,' 'competition,' 
'occurrence,' 'benevolence,' 'legible.' 

"In No. 2, the pupils composed sentences containing the derivatives, 
some of which, in this measurement also, ought not to be unfamiliar to 



242 KDUfATlOXAI. rSVCII()L(>( .\ 

non-Latin pupils in their second year of English, as 'impediment,' 
'advocate,' 'reference,' 'anticipate,' 'subside.' 

"In the third measurement, the difference in the averages of the two 
groups — 60.5' r, 'I'ifl .3,S-3% — ^^'^s so great that Miss Humphrey thought 
that perhaps too dilVicult words had been selected, or at least words which 
placed the non-Latin students at an unreasonable disiidvantagc. Cu- 
riously enough, in this measurement the words were taken, not from the 
notebooks of a Latin pupil, as in the first two tests, in which the dilTereiu e 
between the two groups was much less, but, as slated above, from The 
Talc of Two Cities. Furthermore, in Xo. t,, the non-Latin pupils were 
so far afield in giving accurate definitions, and so confused in classifying 
the words as to parts of speech, that it was decided to give another test 
in which they should be asked, not to defme words, but to give their 
meanings, with the parts of speech omitted entirely. The results in this 
measurement — 57% and 27.5% — were virtually the same as in No. 3. 

"Since practically every second-year pupil could write at least pass;ibly 
on such a subject as 'What do I like to do best' it was decided to make 
tlie basis of comparison in No. 5, not the average of the two groups, but 
the percentage of rating above the passing mark. Moreover, in this 
vocabulary test, emphasis was laid, not merely ujxjn words of Latin 
origin, but upon any words out of the ordinary, from whatever source. 
The wide difference in the results from the view-point of excellence in 
vocabulary — 36.0% and 6.8% — shows clearly what I have always be- 
lieved and maintained, namely, that the work in commercial Latin 
necessarily gives the pupils the dictionary habit, the results of which 
extend far beyond the Latin derivatives actually studied. 

"Of all the measurements. No. 6, was perha[is the most convincing. 
Li this test, the Latin pui)ils, unlike those in Nos. 1-5. had had during 
the last six months of the two years' course the benefit of drill in a vocab- 
ulary not in the commonest use and yet valuable and even necess;iry to 
educated people. The list of words was taken entirely from Franklin's 
Autobiography and Silas Marner which the jiupils had just read, and 
was not of unusual difficulty, consisting of such words, for example, as 
asperity, promiscuous, mortuary. Yet by referring to the results it will 
be seen that to the non-Latin group of pupils such words were practically 
meaningless. 

"An examination of the m;irks on tliese tests may prove of interest. 
Among the seventeen non-Latin students the highest gra<le was 3o'"r. and 
five zeros were recorded. In the Latin group, on the other hand, the 
lowest mark was3o'"r, while one pui)il received loo'^r, two <)o'"c. two 80%. 
five 70%, and only three had below 50*"^,. The dilTerence in averages of 
the two groups was 53%." (IVrkiiis, '14, pp. 11-14.) 

This invcstipatinn is inlcrfsling and one of the few Avhose results 
were carefull}' worked out to nuikc u j>rccisc comparison after cliuii- 



TRANSFERENCE OF TRAINING 243 

nating differences in original capacity. It is a question, however, to 
what extent the Latin students may have been favored by the 
manner in which the words for the various tests were selected. For 
tests I and 2 they were taken from the lists compiled and studied 
by the Latin pupils. Even when the words are selected from Eng- 
lish sources such as Silas Marucr and Franklin's Autobiography 
there is still the question as to the particular words chosen for the 
test. It is obviously unfair to select words which are relatively rare 
and whose meaning may readily be inferred from their origin. To 
what extent the words were selected fairly cannot be judged since 
Perkins does not give the lists of words used. 

The writer ('17) undertook a study to determine as precisely 
as possible, the relative shares contributed by language training 
and by original ability toward proficiency in English composition. 
A series of tests was carried out with a group of 177 university 
students. These tests together with their findings are given in 
Table 74. 

No. 3 consisted in writing an extemporaneous composition within 
a limited time. These compositions were rated by three judges by 
the Hillegas Scale. 

No. 4 gives the average number of words written by each group 
of students. 

No. 5 gives the average number of different words used in each 
composition. 

No. 6 was a test in speed of reading. The ntmibers refer to the 
words read per second. 

No. 7 gives the number of words written in reproducing the 
thought of the passage read in No. 6. 

No. 8 gives the number of A's canceled in one minute in the well- 
known A-test. 

No. 9 gives the scores made in canceling in one minute a certain 
geometrical form on a page of similar forms. 

No. 10 consisted in reading to the persons a series of ten words 
to see how many they could recall immediately afterwards. 

No. 1 1 was carried out by giving a stimulus word and having the 
persons write as many associated words as they could in thirty 
seconds. 

No. 12 consisted in giving a series of ten words and allowing 
fifteen seconds to each word for writing as many synonyms as 
possible. 

No. 13 was a set of tests in imaging geometrical forms. 



244 



EDUCATIONAL PSYCHOLOGY 



No. 14 consisted in presenting to the subjects seven words, one 
at a time, spelled orally backwards by the e.v])erinienter. The 
pcr>uns wrote down the words which tlK\' could thus recognize. 



TABLE 74. After Starch ('17) 



Years of forei}j;n 
liinj^iiase 

Number of persons . 

Composition (Ilille- 
pas scale) 

Words written 

Different words used 

Readin<^ speed 

Reading comprehen- 
sion 

Perception A-test. . 

Perce[)tion form . . . 

Memory words 

Association free . . . 

Association 
s>Tionyms 

Imagery forms 

Imagery words 

Years of l-'nglish. . . 

tirades, first year of 
high sciiool 













1-2 


3-4 


5-6 


7-8 


9-15 


14 


53 


49 


40 


21 


67.6 


693 


68.7 


71-5 


78.: 


140.7 


158-3 


162. 1 


168.1 


181. 4 


82.3 


86.6 


96 . 6 


96.8 


III .6 


4-5 


5-4 


5-5 


5-2 


6.0 


60.0 


65.0 


70.5 


65.7 


68.2 


66.2 


66.8 


66.2 


67.8 


66.1 


7-5 


7-7 


7-8 


7-5 


8.0 


7-4 


7-2 


7-3 


7.2 


7-2 


23-5 


25-9 


2^.6 


29.9 


28.7 


15-4 


IS-O 


iS-9 


154 


14.2 


7.0 


7.6 


7.0 


71 


7.8 


50 


5-7 


51 


5-7 


6.1 


S-i 


4-9 


5-3 


5-5 


5-5 


83.0 


85.7 


83 -7 


S() . 7 


88. 



Percelntack 




'>-15 liROlP 




t-2 CrOUI- 






No 




Latin 




59 


15 7 


67.9 


28.9 


158.7 


35-4 


89.6 


33-3 


5-4 


13-7 


64.0 


0.0 


66.2 


6.7 


7.8 


-2.7 


7-3 


26.4 


253 


-7-7 


150 


•13 


7-2 


21.4 


5-5 




5° 




84 5 



Lut^n 

112 

72.0 

165.0 

95 2 

5 5 

69.0 

67.1 

7 9 

7 3 

26.3 

15 4 
7 4 
56 
5 3 



85 7 



A general inspection of the tal)le shows that there is a steady in- 
crease in the scores of j^ractically every test from left to right with 
increasing years of language study. Thus the 9-1 5 year group wrote 
considerably better and longer compositions than the 1-2 year 
grouj). The puq)Ose of tests Nos. 6-14 was to ascertain to what 
extent this sujjeriority was due to original sujieriority of a])ility 
or to the effect of language training. Tests S-14 were selected par- 
ticularly because the capacities to do ihem would prol)ably be 
affected very slightly if at all by language training. These show 
on the whole a distinct sujieriority of iiihi-rent ability on the jiart 
of the groups who studied languages for longer periods of years. 

In order to make a crucial comjjarison as to how nuich of the 
greater composition ability was due to tiie greater original ca|)acity 
of the pupils and how mudi was due to their gnater training in 



TRANSFERENCE OF TRAINING 245 

language, the grades received by these students in all the subjects 
carried during the first year of the high school were obtained from 
the entrance records of the University. The amounts of difference 
in original ability of the groups who later pursued varying amounts 
of language work would be definitely indicated by this method, 
since at that time none had had more than one year of foreign 
language. The difference in the scholastic grades at the end of the 
first year of the high school between those who later pursued lan- 
guages for a total of 9-15 years and those who pursued languages 
for a total of 1-2 years could certainly not be due to language train- 
ing. 

Row 16 gives for the different groups the average scholarship 
grades during the first year of the high school. It will be noted that 
there is a steady increase from group to group. The 9-15 year 
group had an average grade of 88.0, or five points higher than the 
1-2 year group. 

The next problem was to compare, in common terms the five 
points of difference in scholarship on the percentage scale with the 
difference of 10.6 in quality of composition as measured by the Hille- 
gas scale. To reduce these two types of measurements to commen- 
surate units, fifty-eight compositions were rated by four persons 
both by the percentage method and by the Hillegas scale. By a 
process of equating values it was found that i.o point on the per- 
centage scale equals 2.1 on the Hillegas scale. The difference of 
five points, percentage scale, in original scholarship between the 
1-2 year group and the 9-15 year group would be 10.5 in terms of 
the Hillegas scale. The surprising result seems to be that the differ- 
ence of 10.6, Hillegas scale, in quality of composition between 
these two groups is approximately equaled by 10.5, the difference 
in original scholarship when expressed in terms of the Hillegas 
scale. The conclusion seems, therefore, unavoidable that the 
difference in abiUty in English composition is due practically en- 
tirely to a difference in original ability and only to a slight or no 
extent to the training in foreign languages. [For the method of this 
computation the reader is referred to the original report of this in- 
vestigation. ('17.)] 

The increase in length of composition and in speed of reading is 
large and very probably in excess of the difference in original ability. 
Training in foreign language seems to have produced a distinct 
effect in greater fluency of words in writing and in more rapid per- 
ception of words in reading. 



>46 



KOUCATK )\AL PSYCHOLOGY 



Miss M. Theresa Dallam ('17), a teacher of English in the 
Western High School of Baltimore, made a series of experiments 
on her pupils to test tlie truth of her conviction " that Latin classes 
do much better work in English tJian the classes that have not 
studied Latin." Out of 114 students she selected 34 fourth-year 
students, 17 Latin and 17 no-Latin or modern language pupils, 
by pairing them on the basis of their general scholarship records, 
so that the two groups would be equal in mental ability. The Latin 
group had an average of 78 and the no-Latin group 79. Kellc)-'s 
Silent Reading test was also made with them. The Latin group 
made an average score of 31.1 and the no-Latin group 33.1. The 
two groups were, therefore, very nearly equal in general ability. 
The Latin group had studied the language for four years. 

Miss Dallam then made seven tests: spelling, reproduction, dic- 
tation, Latin derivations, delinitions, compositions, and English 
grammar. The results were as follows: 



TABLE 75. After Dallam ('17) 
Average percentages attained 



Rki>R( 



Dkriv. 



CoMP. CiRVU 



Mod. Lan. Groui). 
Latin Group 



80.2 
90.7 



63.0 
65 9 



96.0 
95 o 



295 

52.0 



730 

75-2 



71.8 59.7 

75-4 67.6 



Thus the Latin group made a distinctly higher average in deriva- 
tions and grammar, an aj^preciably higher grade in compositions; 
a slightly higher average in sju'lling, reproduction, and definitions; 
and a .slightly lower average in dictation. 

Miss Dallam then comj^uled coeflicients of correlation for the 
Latin group between their grades in Latin and each of the seven 
tests, and for the modern language group between their grades in 
modern languages and each of the seven tests. These correlations 
were as follows: 

TABLE 76. After Dallam ('17) 
Cocfruicnls of correlation 



Sp. 



DiCT. 



Dkrtv. Dkhn. Coiiv. Gram 



Mod. Lan. Group. 
I..2itin Group 



+ .OQ 

+ .OS 



+ .19 



.04 
.16 



— .02 

+ 13 



+ •23 

+ .15 



+ .11 
+ .28 



-f-.28 

+ .46 



TRANSFERENCE OF TRiVINING 247 

These correlations are so low that, with the exception of the ones 
for grammar, no significance can be attached to them except to say 
that there is practically no correlation between the various com- 
parisons made and that the Latin group shows no superiority over 
the modern language group in spelling, reproduction, definitions, 
and a doubtful superiority in dictation, derivations, compositions, 
and grammar. The differences that are shown are non-committal 
and so small that they would have to be substantiated with other 
groups to be conclusive. 

d. Science. Claims of general training to be derived from the 
pursuit of the sciences are practically as extensive and confident 
as those made for the languages. The only difference is that the 
claims for the sciences have not been questioned or investigated 
as much as those made for the languages. Bagley has summarized 
them in the following manner: 

"i. The formation of some useful specific habits, — through training, 
routine, rationalized practice. 

"2. The acquisition of useful information, — through methodical 
study, instruction, and drill. 

"3. The adoption of valuable ideals, or 'emotionalized standards,' — 
inculcated through the inspiration to be gained from the teacher, from 
the lives of great scientists, and from experiences of intimate contact 
with nature. 

"4. The acquisition of facility in the use of facts, ideas, and methodical 
thought processes, for the solution of problems, the overcoming of 
difficulties, and the accomplishment of worthy purposes, — through the 
mental discipline afforded by properly graded practice in the solving of 
scientific problems. 

"5. The development of taste, and power of appreciation, — to be 
gained through a clear apprehension of unity, adaptation, economy, 
order, and system in nature as interpreted by science. 

"6. The development of scientific or philosophic insights, perspectives, 
and attitudes of mind that serve as safeguards to the intelligent inter- 
pretation of contemporary life, — through acquaintance with systems of 
organized knowledge." (Bagley, '11.) 

One of the important values claimed for the sciences is the general 
training of accuracy and fidelity in observation and the transfer- 
ence of this particular type of training to other types of observation. 
Miss Hewins ('16) made an attempt to measure the extent to which 
this improvement is general or carries over to other types of ob- 
servation. She divided each of three classes in botany, composed 
of 34 boys and 50 girls, in the first year of a New York high school, 



248 



EnucAi loxAL I'svriioi.ocv 



into two groups, and gave ihem a series of tests in various kinds of 
observation of a biological and non-biological nature as follows: 



No. 


1)ATE 


I 


April 


1 22 


2 




23 


3 




23 


4 




24 


S 




24 


6 




25 


7 




25 


8 




26 


9 




29 


lO 




29 


II 




30 


12 




30 



13 May 



1 May IS 

2 16 

3 17 

4 20 



5 


21 


6 


22 


7 


23 


8 


24 





27 


10 


28 



I Juno .^ 


2 


•» 


3 


4 


4 


5 


5 


5 


6 


6 


7 


6 


8 


7 





10 


10 


10 


II 


It 


12 


II 


J3 


12 



Scries I 

Test Skries Lxi-obuw: 

Horse chestnut stem 

Picture 30 seconds 
Forsythia flower 

10 syllal)les 30 " 
Lilac leaves 

Nonsense fi>;jure 30 " 

Geometrical figure 30 " 

10 2-pIace figures 30 " 
Scouring rush i minute 

Majiie seedling i minute 

Pea cliart 30 seconds 
Figure in air 

Potato chart 30 seconds 

Practice series 
Matkriai. 
I )escripti()n of lilac Qow cr 

" " box-elder leaf 

" " the stem, leaf, and flower 

of gill-run-over-the-ground 
I)escri])tion of flower stalk and flowers 

of the lily-of-t he-valley 
Descri|)tion of the horse chestnut flower 
" " " buttercup flower 

" " " mustard flower 

" " " dogwiKKl flower 

" " " deutzia flower 

'• " " columbine flower 



Scries 

TtsT Series 
Sassafras stem 
Picture 

Syringa flower 
10 syllables 
Forsythia leaves 
Nonsense figure 
(Jcometri(al figures 
10 2-placc figures 
Moss plant 
Pumpkin see<I chart 
Cr.ipe (hart 
Figure in air 
Wild carrot 



LxPUbUE£ 

30 seconds 

30 " 

30 " 
30 " 

30 " 
I minute 
I minute 

30 seconds 

30 " 



Time puk 
Recokoi.no 

10 minutes 
5 " 
5 " 
I " 
8 " 



RErORDINC 

10 minutes 
for each 
test 



Time for 
Recx>rdim; 

10 minutes 
5 
5 

I " 

8 



TRANSFERENCE OF TRAINING 



249 



The practice series was continued for a period of ten days with 
one section of each of the three classes. The other section of each 
class answered questions from books on the material of the lessons. 
At the end of the period, she gave the tests in Series 2 to both 
sections of each class and obtained the following gains in scores 
over Series i: 



TABLE 77. After Hewins ('16) 
Gross gains in scores 



Biological Tests 



Non-biological Tests 



Practiced groups: 

Boys 

Girls 

Average 

Unpracticed groups: 

Boys 

Girls 

Average 

Residual difference between practiced 
and unpracticed groups 

Percentage gains of residual differ- 
ences ov^er scores in first test series 



8 


06 


6 


41 


7 


23 


3 


03 


— I 


24 




89 


6 


34 


33 


9% 



(Median score in biological tests in Series i before practice 18.7) 
( " " " non- " " " " " " " 39.0) 

Miss Hewins infers a considerable gain in both the biological and 
non-biological types of observation. 

"Table 77 shows that in the biological tests, the average gain of each 
T ractised boy was 8.06 per test for the 5 tests while the unpractised 
showed a gain of 3.03. The practised girls averaged 6.41 gain per test, 
vhile the unpractised lost 1.24 per test. In the non-biological, the prac- 
tised girls gained 6.2 per test for the 6 tests while the unpractised gained 
5.6; the practised boys gained 8.97 per test and the unpractised boys 
gained 5.27 per test." 

"Feeling that the balance of arguments and scientific proofs were 
against formal discipline when this investigation was begun, I am forced 
by the results obtained to admit that in this experiment the proof seems 
to be on the affirmative side. 

"A valuable lesson, T think, can be drawn from one phase of this 
investigation. By consulting the tables and summaries, it will be seen 



250 KDUCATir»\ \|. I'sNcnOLOGV 

that sometimes one division docs not fall in line with the general trend, 
but that a larger number outweighs the negative and shows ix)sitive re- 
sults." (Hewins, pp. 111 and iijj 

In order to make a relative comparison of these gains on the 
basis of the original scores in the first series of end tests, I have 
computed the j)ercenlages of gain in these end tests as shown at the 
bottom of the above table. When such a relative comparison is 
made, the net gain in the non-biological observations is very small, 
being only 5.4% as compared with 33.9% in the biological obser- 
vations, or about one-sixth as much in the non-biological as in 
the biological t}'pes of observation. The improvement in the 
biological end tests can hardly be counted as evidence of spread 
of training since these tests were so similar to the training series 
that they were all but identical with it. The spread of training is 
apparently not as great as Miss Hewins believes it to be. 

c. Geometry. Geometry, particularly in the high school, is urge<l 
to possess a considerable amount of disciplinary value: 

Plato (Republic, Book 7) emphasized this opinion thus: 

" Moreover, the science (Geometry) has indinct clTcrls which are not 
small. 

" ' Of what kind? ' he said. 

"'There are the military advantages of which you spoke,' I s;iid; 
and in all departments of knowledge, as experiem e proves, any one wht) 
has studied geometry is infniitely (juicker of apprehension than one who 
has not. 

Yes, indeed,' he said, ' there is an infinite ditTerence between them." 

Rugg ('16) made a study of the spread of training in the case of 
learning descriptive geometry. He j)erformed three groups of 
tests at the beginning and at the end of the course in descriptive 
geometry, in February and in June resjK'ctively, with 3:26 students 
in the College of Engineering of the University of Illinois. The 
.same tests were made with 7S students in other colleges as a control 
group who did not pursue the C()urse in descriptive geometry. 
The three groups of tests were of a non-geometrical, Tests i and 2, 
(|uasi-geometrical, Test 3, and strictly geometrical nature. Tests 
4 and 5. Partial illustrations of the various tests follow; 

Test I. 

1. Divide ciRhty-onc hy seven. 

2. Divide scventy-ciglil \>y fjuir, itc. 



TRANSFERENCE OF TRAINING 25 1 



Test 2. 

Test 3. 
Test 4. 



Test 5. 



1. Divide eight sixty- two by three. 

2. Divide seven ninety-five by four, etc. 

A test in imaging letters. 

The Painted Cube Test. 

A three-inch cube, painted on all sides, is cut into one-inch cubes. 

1. How many one-inch cubes have paint on three sides? 

2. " " " " " " " " two " ? 

3. " " " " " " " " one side? 

4. " " " " " " " " no side? 

Geometrical Objects' Test. 

The problem: Form a mental picture of each object and count the 

niunber of straight lines which it would take to construct each 

one in space. 

1. A wedge. 

2. Four triangles attached to a square, bases coinciding with the 

sides of the square, etc. 

The main results are represented in the following table: 

TABLE 78. Representing the results for " Rights." Adapted from Rugg's 
table, p. 123 ('16) 

Test 1 Test 2 Test 3 Test 4 Test S 

Train- Con- Train- Con- Train- Cok- Train- Con- Train- Con- 
ing TROL ING TROL ING TROL ING TROL ING IROL 

Group Group Group Group Group Group Group Group Group Group 

Feb. scores 20.0 17.38 20.00 19.3 4.67 4.50 0.59 0.28 2.32 1.92 

June scores 22.81 20.06 23.00 19.15 5.66 4.58 1.50 0.68 3.3i 1.81 

Gross gain 2.81 2.68 3.00 -.15 .99 0.083 .91 0.40 1.01 -0.10 

%gain 14.5 15.6 15.0 -.78 21.2 1.8 157.0 143.0 43.5 -5.0 

Residual 

difference -1.10 15.78 19.4 14.0 48.5 

Averages 7.34 19.4 31.7 

Rugg concludes that the residual gain indicates a considerable 
transfer of improvement to all three types of abilities tested. 

"The study of descriptive geometry (under ordinary class room condi- 
tions throughout a semester of 1 5 weeks) in which such natural and not 
undue consideration is given to practice in geometrical visualization as is 
necessary for the solution of descriptive geometry problems operates: 

" (i) Substantially to increase the students' ability in solving problems 
requiring the mental manipulation of a geometrical nature, the content 
of which is distinctly different from the visual content of descriptive 
geometry itself. 

" (2) Substantially to increase the students' ability in solving problems 



252 EDUCATIONAL PSVCH()I.( »( ;V 

requiring the mental manipulation of spatial elements of a slightly 
geometrical character, i. c., problems utilizing the fundamental elements 
of geometry (the point, line, and plane), but apart from a geometrical 
setting, and in such form as to offer no geometrical aids in solution. 

" (3) Substantially to increase the students' ability in solving problems 
requiring the mental manipulation of spatial elements of a completely 
non-geumetrical nature, i. e.. problems in which the straight line and 
plane do not appear in any way whatever. 

" (4) The training cfTcct of such study in descriptive geometry operates 
more eflkiently in those problems whose visual content more closely 
resembles that of the training course itself, i. e., in those problems whose 
imagery content is composed of combination of points, lines, and planes, 
and in which the continuity of the manipulating movements approaches 
the continuity of those in the training course. (Rugg, pp. 114-115.) 

These results are not \cry clilTerenl from those surveyed in LJie 
preceding chaj)lcr. Rugg states the results perhaps as favorably as 
the data permit, perhaps too favorably. At any rate we ought to 
note that the gain in the non-geometrical tests, Nos. i and 2, is 
only about one-fourth or one-fifth as much as in the strictly geomet- 
rical tests, Nos. 4 and 5. We also should note, as Rugg himself 
jioints out, that only about two-thirds of the persons in the train- 
ing group gained; the remaining one-third did not gain or show 
transfer. 

General Interpretation. The transference of training of the capa- 
cities in\o!\cil in the learning of school material is very small so far 
as present partial data indicate. This seems to be equally true 
of all school subjects, the sciences as well as the languages and 
mathematics. If we represent the possible transfer effect as ranging 
from 0%, or none, to loo^r, or an improvement in other capacities 
cfjual to that in the capacity trained, then the amount of transfer 
is much nearer to the o^/(, end than to the 100% end. It probably is 
o or very nearly o, for all capacities which are not distinctly similar 
or related to the capacities specifically trained. Thus in the author's 
experiment, practice in mental multiplication improved other forms 
of mental calculations about one-fourth as much, but had no effect 
upon immediate memory' of numbers or words. In Winch's ex- 
periment, practice in arithmetical computations harl either no effect 
or a doubtfid clTect upon arithmi-lical reasoning. In Rugg's in- 
vestigation, jiractice in visualization ordinarily done in a course 
in descriptive geoinetry had only a moderate effect upon visuali- 
ziUion (if other sorts. In Miss Hewin's study, improvement in 



TRANSFERENCE OF TRAINING 253 

biological observation improved non-biological observation only 
one-sixth as much. In the author's investigation, the study of 
foreign languages had no effect upon the capacity to write an Eng- 
lish composition, the study of Latin seemed to produce a small 
increase in English vocabulary and a decided increase in the knowl- 
edge of Enghsh grammar, but only a very small increase in dis- 
crimination in correct English. Wilcox likewise found that the 
study of Latin had no appreciable effect upon work in English 
classes as measured by marks. The superiority of the Latin pupils 
was due to their superior native ability rather than to the study of 
the language. Perkins found that Latin as taught by him with 
special emphasis upon word derivations and meanings produced 
a noticeable increase in the abiUty to spell, define, and use English 
words. 

The transfer effects of the training of the abilities in school sub- 
jects is very much less than is commonly assumed. This is prob- 
ably due, in the first place, to the fact that the improvement in the 
capacities exercised specifically by the school subjects is usually 
not as great as is commonly believed by teachers. The modifica- 
tions produced in the minds of the pupils are considerably less than 
teachers usually assume as judged by the modifications produced 
in their own minds after much longer and harder study. To il- 
lustrate, teachers are inclined to believe that a course in mathe- 
matics has produced a much greater improvement in mathematical 
reasoning, or that a course in history has brought about much 
greater facility in handling and interpreting historical material, 
or that a course in psychology has brought about much keener 
insight into the operations of mental process, than these respective 
subjects actually have produced. This is probably the result of the 
teacher's naturally egotistical belief in the effectiveness of his own 
work. 

The small effects of transfer are probably due, in the second 
place, to the fact that the conditions for securing transfer are not 
as favorable on the whole in the case of school subjects as in the 
case of the special laboratory experiments on transference. 

The evidence on spread of training in school material tends to 
support for the most part the theory of identical elements. The 
effects are the largest where there is similarity or identity of material 
as, for example, in the case of the effect of the study of Latin upon 
the study of Spanish, or upon the knowledge of English grammar. 
The fact of identity of material or similarity of procedure makes 



2 54 EDUCATIONAL PSYCHOLOCJV 

possible a greater control of the spread of improvement through 
methods of teaching \vherel)y the identity or the use of identical 
material may be emphasized in as many desirable relations as 
possible. This is illustrated by the spread of the efTects of Latin as 
taught by Mr. Perkins. 

In formulating an opinion concerning general training efTects 
resulting from training of special capacities, we must bear in mind 
that even where the transfer efTect is considerable, as much as one- 
fourth to one-third as much as in the capacity specially trained, 
it is obviously more economical to give practice directly to the 
capacities which we want to train rather than to do it indirectly 
with the hope that the improvement may be transferred to them. 
Concretely, even if the study of Latin under favorable methods of 
teaching does improve the spelling of English words, would it not 
be more economical to study directly the spelling of the words 
which are to be acciuired? Knowledge of the most common Latin 
words from which the largest number of I-^nglish words are derived 
could be obtained in a relatively short period of time, probably a 
year or even less. Learning to play the piano might help in learning 
to play the violin, but no sane person would devote very much time 
to the piano if his sole purpose is to learn to play the violin. 

Even if mathematics may cause some im[)rovement in reasoning 
about bargains, even if the study of Latin may increase English 
vocabulary, or even if a study of animal [jsychology did make a 
man a better teamster, these efTects are relatively very small and 
can be pnxluced much more economically by a direct study of 
bargains, or of the origin and meaning of English words, or of driv- 
ing horses. A course in mathematics or in Latin or in psychology 
will have to stand primarily on its own feet for the content that it 
ofTers or the skill that it develops. These by-products may be useful 
l)ut they cannot be the sole purpose of the etTorts put into a course. 
The value of a meal depends upon the meal itself and not uiK)n the 
crumbs that fall from the table. \\'hene\er a subject loses its con- 
tent value through ( hanged social conditions it seems mysteriously 
to acquire a great deal of discii>linary value. 

An immense amount of confusion in the thinking about the prob- 
lem of mental disc ipline and the value (^f school subjects, even on the 
part of distinguished thinkers, has resulted from a failure to dis- 
criminate between the elTect of a certain kind of education and the 
native capacities of the individuals sul)jected to the education. 
Whenever allowance or detluctions for differences in original ability 



TRANSFERENCE OF TRAINING 255 

have been made, the general discipHnary effect has been found to be 
much less, or, in many mstances, even non-existent. To argue that 
because certain great leaders of men had a certain type of education, 
it must have produced their greatness does not prove the point. 
They probably would have achieved distinction if they had had any 
other sort of education. If the chief argument for pursuing a given 
subject is that it selects the more able pupils, it would be much more 
economical to do so by a shorter and more certain method. Almost 
any fifteen or twenty mental tests that can be applied in a psycho- 
logical laboratory in two hours would separate much more ac- 
curately the gifted from the stupid. 

Finally, the upshot of the experimental and statistical inquiries 
into the transference of training is that effects of training are trans- 
ferred in smaller amounts and within much narrower limits than has 
commonly been assumed. This does not mean that there is no gen- 
eral mental discipline in any form of training, nor that the doctrine 
of formal discipline has been "exploded" but rather that the actual 
limits of general disciphne have been more accurately defined. 
These limits, to be sure, seem to be much narrower than many are 
inclined to believe. So far as the value of school subjects is con- 
cerned, it means that the content value of a subject must be the 
prime reason and the general disciplinary value the secondary 
reason for pursuing it. 

Before leaving this discussion, two further points should be borne 
in mind: The first is that any effect of transfer, even if very slight, 
would probably be worth while if it extended to all or to a large 
number of capacities. If training in botanical observation improved 
all forms of observation in life ever so little, it might still be the 
best form of training in observation. But the implication of the 
evidence thus far at hand is that the spread seems to extend only to 
rather narrow limits. The second point is, that while the trend of 
the arguments here presented would be to reduce the time devoted 
to some subjects, particularly in high school and college, we must 
be sure that we put something better in their places. The advan- 
tage of some of the subjects that would suffer reduction is that they 
are well organized for teaching purposes. Some of the new sub- 
stitutes are not well organized and offer neither form nor content. 
Transitions should be made gradually so that the new branches may 
become organized and extended, and the teachers properly trained. 



r1 



PART III 

THE PSYCHOLOGY OF LEARNING: B. OF SCHOOL 
SUBJECTS 



CHAPTER XV 
THE PSYCHOLOGY OF LEARNING SCHOOL SUBJECTS 

Psychology and Teaching. If education consists in making 
changes in human beings, if psychology is the scientific study of 
the mental processes of human beings, and if teaching consists in 
the facilitation of the changes to be made by the school, then it is 
obvious that knowledge of how to bring about these changes in 
the most economical manner must be based upon an exact knowl- 
edge of the processes involved in these changes. A reliable peda- 
gogy can be based only upon a reliable psychology of the processes 
concerned. Engineering did not become a science until the physi- 
ical and chemical knowledge of the processes involved in a given 
project were thoroughly understood. To build a Brooklyn bridge 
involves a precise knowledge of the laws of gravitation, the strength 
of materials, the means of supporting weights against the force of 
gravity, and the like. To compound an electric cell involves a 
precise knowledge of the chemical action of certain elements. To 
know how to destroy bacteria harmful to plant and animal life, 
it is necessary to understand the biological processes of the par- 
ticular plant and animal life concerned. 

The foundation science for sane and dependable methods in 
education is psychology in its broad sense. The great difficulty 
in establishing a reliable pedagogy is the fact that sure and de- 
tailed knowledge of the psychological processes and laws in learn- 
ing the material of the school subjects is largely unknown. In the 
past the schools have proceeded largely by guess. The future will 
have to map out in detail the psychological steps involved in each 
school subject, and to submit these processes to direct experimental 
investigation. A science of engineering w^as impossible until it 
was discovered how the physical and chemical laws operated in 
the particular conditions under which the bridge had to be built 
or the cell had to be compounded. Knowledge of the law of gravi- 
tation was practically useless until it was discovered how it operated 
under given concrete conditions of materials, distances, and 
circumstances. The psychological laws of learning will be practi- 
cally useless until we shall know how they operate under the 

259 



26o EDUCATIONAL PSVCIIOLOGV 

concrete conditions of the school and with the materials to he 
learned in the school. The reaction times of numerous psychologi- 
cal processes ha\e been studied in great detail for many years, but 
this knowledge is practically useless in giving to the educator 
scientific information by which he may proceed to facilitate 
the reactions of a child in learning to WTite or to read. This 
knowledge is useful in furnishing a general background, in 
jjointing the way, and in suj^plying a general experimental tech- 
ni()ue. But even these must often be materially modified and 
I'.dapted to the solution of a particular problem in a given field. 
The old-time pedagogy has fallen into disrepute because it has 
been almost wholly a matter of personal guesswork. The slate 
must be wiped clean and only those principles and laws whose 
truth has been fully proved can be recorded thereon. 

Problems. If we grant that the method and procedure of the 
school should be based upon a sound psychology' of the processes 
inxolved in learning the special materials of the school subjects, 
it follows that the fundamental tasks to be done are these: 

(i) A thorough and complete analysis of all ])sychological 
])rocesses involved in the learning of a given subject, or in the 
acquisition of skill in it, and of the order and manner in which 
these ])rocesses intermesh. 

(2) The devising of means ])y which these processes may be 
measured and tisted so that the facility in their operation may be 
determined c|uantitively. 

(3) The disco\'ery of the most economical procedures by which 
each ])articular step in the entire i)rocess may be develojxd. 

If we wish to determine how to memorize a poem in the most 
economical and most permanent manner, it is important to know 
the perception, association, and reaction processes involved, the 
means for definitely measuring facility in these operations and 
the means of controlling these processes nrost efficiently. Teach- 
ing is mental engineering; it consists in managing the mental 
processes C(mcemed in learning the materials and in acquiring the 
skill of the school in the most effective and most profitable 
manner. 

In considering the jisychology and jiedagog)' of school subjects 
in the succeeding chapters, this three-fold division of the ])rol)lems 
will be made for each subject and the available information of 
each one surveyed so far as our present knowledge warrants. 



1.-^^ 



CHAPTER XVI 
READING 

Processes or Steps Involved in Reading 

If we trace, for analytical purposes, the successive steps from 
the external presentation of the visual stimuli of the printed 
words on through the complicated elaborations within the mind 
and back to the external expression of reactions in the pronunci- 
ation of the words, we can discern the following order or combin- 
ation of elements: 

(i) Reception upon the retina of the stimuli from the printed 
page. 

(2) The range of the field of distinct vision on the retina. 

(3) The range of attention in apprehending visual stimuli. 

(4) The movements of the eyes. 

(5) The transmission of the visual impressions from the retina 

to the visual centers of the brain. 

(6) The establishment or arousal of association processes where- 

by the incoming impulses are interpreted. 

(7) The transmission of the impulses from the visual centers 

to the motor speech centers. 

(8) The transmission of motor impulses from the speech centers 

to the muscles of the vocal chords, tongue, lips, and re- 
lated parts. 

(9) Execution of the movements of the speech organs in speak- 

ing the words. 
These are the steps as they occur in the developed reading 
process. It is obvious that they do not follow each other in a 
temporal order but that some, as for example (i) to (4), occur 
simultaneously. During the early stage of learning to read there 
occurs, simultaneously with steps (i) to (5), a parallel series of 
steps derived through the ear by which the child learns the pro- 
nunciation of the word, thus: (i) reception in the ear of auditory 
stimuli from the pronunciation of the word by the teacher, (2) 
transmission of the auditory impulses from the ear to the auditory 
center in the brain, (3) transmission of impulses between the 

261 



262 i.Di ( Ai i< >\ Ai. I'svriioT.nr.v 

auditor}' and the visual centers whereby the sound and the sight 
of the word become associated. Silent reading involves only 
the first six steps except in so far as incipient speech mo\-e- 
ments accompany it, in which case the remaining steps enter 
in j)art. 

Such an analysis as this may seem detailed to an unprofitable ex- 
tent; but, in fact, it might be made even more detailed, depending 
upon the extent to which we are able to discern and describe the 
minuteness of the neural and mental functions involved. The 
more complete and accurate our analysis and descrii)tion of the 
steps is, the more sure our knowledge for managing the.se 
processes will be; and ultimately, that is what teaching amounts 
to: The efficient management of the psychophysical processes 
concerned. 

The next i)r()blem is, IIow does each of the elements in the 
reading process operate? 'J'he truth is that concerning many of 
them we know at the present time very little or nothing with cer- 
tainty or completeness. Concerning some of them, however, 
considerable defuiite knowledge has been accuniulaled in recent 
years. What the differences between an efllcient and an inelhcient 
reader are, or what the difficulties in learning to read are at each 
of the steps can be inferred partly, but only partly, from our 
present knowledge about these factors. It is, however, certain 
that the dilTerences and difliculties are to be found in these and 
possibly additional processes. We shall examine each of tliese 
steps in turn and sur\'ey what definite knowledge is available. 

(i) The reception upon the retina of the visual stimuli from 
the printed page dei)en(ls obviously on the one hand, upon the 
size and kind of type, length of lines, indentation of lines, paper, 
and illumination; and, on the other hantl, ujion the inertia of the 
retina in receiving stimuli. We know, for example, tliat for adults 
and for children above 10 or 12 years of age, type smaller than 
8 or 10 points is i)r()bably too small to be perceived easily. Like- 
wise, tyj)e larger than approximately 10 points, spreads out uj)on 
too large an area of the retina to be perceived quickly in as large 
groups as possible. Ivxj)eriments by Dearborn ('06) and others 
have shown that probably the most adwintageous length of line 
is in the neighborluMxl ui 2^2 or ,3^ indies, and that it is better to 
have the lines on a page uniform in length instead of vanning in 
length as is often the casr in reading-texts in which illustnili»)ns 
are set into the margins and the lines made to vary according to 



READING 263 

the space around them. We do not know what the most ad- 
vantageous size of type is for younger children who are beginning 
to learn to read. We feel that it ought to be larger than for older 
children or for adults, but we do not know definitely how much 
larger. 

So far as the inertia of the retina to the reception of the stimuli 
goes, we may infer that it varies in different individuals probably 
according to the normal distribution and that it may be less in 
rapid than in slow readers, but no definite measurements have 
been made to ascertain the truth about it. 

In an investigation, as yet unpublished, by C. L. Hull and W. R. 
Ames, an effort was made to determine the relative effect upon the 
eye of reading from papers of various colors, of various amounts of 
gloss and of various degrees of perfection of the inked impressions. 
Four kinds of paper were compared: a matte white paper, a pink 
newsprint paper, a blue newsprint paper, and a very glossy white 
paper. As measured by the Ingersoll glarimeter, these papers had 
the following glare or gloss values respectively, 18.5%, 41.5%, 42%, 
73.5%. A fair sized book was printed, uniformly upon each kind of 
paper in such a way that the successive pages followed one another 
on a single band of paper. These were placed upon special reels 
in such a way that the pages would be at a uniform distance 
from the eyes, at a uniform angle and would have uniform 
illumination. 

Three measures of the changes in the eyes produced by reading 
were taken: 

1. The number of lines read during a 15-minute period. 

2. The number of spontaneous winks per minute while reading. 
These were recorded by a special device unsuspected by the sub- 
jects. 

3. Extent of failure to recover during a 15-minute reading period 
from artificially produced diminution in the distance at which the 
subjects were able to see faint parallel lines as distinct lines. This 
distance was determined before and after reading by an elaborate 
recording device of considerable precision. Four subjects were 
used. 

The final averages for each kind of paper by each of the tests are 
shown in the following table. In addition, the relative rank of the 
four papers is given for each of the three measures on a scale of 10. 
Lastly these ranks are averaged for a final score of all three meas- 
ures. 



264 



r.nrr.vnoNAL rsvcrioLotiv 



TABLE 79. After Hull utvl Ames. Trom a Thesis in the Library of the Uni- 
\crsily of Wisconsin, 1917 



II 


NAL A\ LK/ 


ua.b 


I"i.v\L Ranks 








Loss OP 
Visual 

ACUITV L\ 
I.SCllKS 


No. or 

\\ INKS 

I'KR 
MiNUTK 


N'o. Li.SKs 
Read per 

15-Mis-. 

Tkriiiu 


Visual 
Acuity 


WiS'KS 


Lines 

Kl All 


A\t.race 

K.\NK 


Matte White. . 
Pink 


1.8 
30 

-'7 
2.0 


7-5 
7 ■ - 
S.I 

«-5 


415 
422 
414 
398 


I 
10 

5-4 
3 4 


3 

I 

7-2 

10. 


3-6 
I. 

4- 
10. 


-• 5 
4- 

.S ^ 
7.6 


BUie 

Gloss White. . . 



Despite ralhcT ."striking differences between the results obtained 
by the thirrl measure and by the other two, the results as a whole, 
as indicated by the fuial average rank, show tliat glare is the de- 
cisive factor in diminishing ocular efficiency. The two white paj)ers 
are respectively the best and the worst of the set. The two colored 
papers which are intermediate in glare are also found intermediate 
in ocular efficiency. These results suggest that color as such has 
little or no inlluence one way or the other. 

A microscopic examination of the texture and perfection of the 
inked impressions of the various papers revealed the glossy white 
papers to be the most perfect, with the matte white, the blue and 
the pink in decreasing order. This order inflicates that within 
ordinary limits the perfection of the printed impression, is of little 
consequence. 

(2) Concerning the size of the distinct field of vision, Ruediger 
('07) made some experiments with the tachistoscopc. Speed of 
reading may depend upon the horizontal area of distinct vision in 
the retina. Ruediger, however, concluded that little correlation 
exists between the areas of color zones on the retina and visual 
acuity and other c|ualities of sight, that the field of distinct vision 
is reduced as much as one-half when the eyes an.' tired from reading 
and that there is only a slight correlation between visual acuity 
and the size of the field of acute vision. 

Judd, McAllister, and Steele ('05) marked the eye near the 
[)upil with a flake of Chinese white and then directly photograplud 
the movements of the eye. Tluy cont huled tliat there is no cor- 
relation between the size nf the horizontal field of acute vision 
and the rate of reading or tlu- nimiber of pauses in reading. The 
correlations found were — .oO and —.10. 



READING 265 

(3) The range of attention in apprehending words and letters. 
Dearborn, in his photographic records of the movements and 
pauses of the eyes, found that the eye takes in, on an average ob- 
tained from five subjects, 1.64 words at one time, or at each fixa- 
tion. He beheves that the word is the unit of reading. This is 
corroborated by the fact that shght misspelhngs in words are often 
not noticed and that such words as "psychology" and "physiology" 
may easily be confused with each other because they are per- 
ceived as wholes. In many instances he found that long words 
take no longer time to be perceived than short ones do. The 
habit of grouping is apparently very important. "It is not the 
short words as such but the words which cannot be easily grouped 
with others, which necessitate separate fixations." He also found 
that the range of attention with slow readers is often only a syllable 
while with the fast reader it is words and phrases. There is thus 
a large and important difference which is probably highly signifi- 
cant in the development of rapid reading as will be pointed out 
later. 

By presenting ordinary printed material for very brief intervals. 
Gray showed that the number of words which could be compre- 
hended at a single exposure was much larger than the number 
actually read at a single fixation — often twice as large. This 
means that the areas of visual apprehension overlap considerably 
at successive fixations. This suggests that the number of fixations 
might be reduced considerably by practice. Gray tried this out 
experimentally with two children, one a good and the other a poor 
reader, and found it to be true. The poor reader decreased the 
number of pauses per line from 15.5 to 6.1 by a 20-minute practice 
period each day for 20 days. 

In this connection the possibility of improving the range of 
visual apprehension becomes important. Some years ago Miss 
Aiken claimed to have increased enormously the range of visual ap- 
prehension of the girls in her school by special training. G. Stanley 
Hall wrote of it, "I would not have thought such rapidity and 
accuracy possible if I had not seen it." Whipple, in an attempt 
to verify these claims under laboratory conditions with adult 
subjects, failed almost completely. Gray, attempting by the same 
methods to enlarge the range of visual impressions of two sixth- 
grade children for printed words, also failed completely. He re- 
peated the experiment, however, with two fourth graders and 
succeeded in approximately doubling it. He concludes that such 



266 EDUCATIONAL I'SV( HOUXiV 

training, to be efTective, should take plate not later than the 
fourth grade. 

(4) The movements of the eyes in reading. Closely related to 
the scope of apprehension is the rapidity and the precision of the 
movement of the eyes. More experimental work has been done 
on these two elements in reading than on any others. It has been 
known for a long lime that the eyes in reading or in examining 
any object do not move along smoothly and continuously but 
that they move by jerks and pauses apparently in a very irregular 
manner. The eyes take rapid glimpses or snap shots of successive 
portions of a line of i)rint and then ])iece them together in obtaining 
the meaning. The reason for it is that the eyes cannot see objects 
distinctly while they are in motion and consequently they per- 
ceive little or nothing during that time. A person cannot see the 
movement of his own eyes in a mirror. As soon as one um see them 
distinctly, they are at rest again. Dearborn found that letters 
fuse when passed in front of the eyes at the rate at which they 
themselves move across a page. Dodge ('00) also found by experi- 
ments that, while the eyes were in motion, no sensiition resulted 
even though the stimulus strikes the eyes. 

A fairly good concejjtion of the nature of eye movements, pauses, 
and fixations in reading may be obtained by taking an o{xni lKX)k, 
placing a mirror on one page while some one reads the opjiosite 
page, and then observing the action of the eyes as rellected in the 
mirror. Javal ('79) counted the pauses of the eyes by means of a 
sound attachment to the eyelids. Laudolt ('91) counted them 
by direct observation. Erdmann and Dodge (08) counted them 
by observing the eyes of their subjects in a mirror. They found 
more pauses with difTicult than with easy reading material, and 
also more pauses in reading a foreign language than in reading 
one's native language. 

The first successful attempt to record eye movements ^\'as made 
by Huey ('9<S-'oo). He attached a jilaster of Paris cup to the 
cornea which was connected with a light aluminun\ ])ointer. This, 
in turn, rested on the smoked drum of a k^^nograph on which the 
movements wire registered. In this manner he attempted to 
study the nature and rapidity of eye movements and the nature 
and length of successive pauses. Dodge then develo])e(l a falling 
plate cami-ra which ])hotographed a beam of light reflected from 
the coniea t)f the e}'e. This method was also used by Dearborn. 
It has the important advantage over Huey's method in that it 



READING 267 

eliminates the attachment of anything to the eyes themselves, 
which quite Hkely interferes with the natural movements of the 
eyes. 

Readings by five subjects of the same newspaper passage 

I 
Sub- Number of Total 

JECTS Fixations Time Average 

J12 152 152 361 

T ST. PE(TE]RSBURlG,No[v.) 2.— Th|e Admiralty... 4 1007 251 

490 110 140 250 

H ST. PE[TERSBURG, [N)ov. 2.— I— TheAd[m)iralty.. 4 1020 255 

416 303 136 138 340 

S [ST. ]PETER|S(B]URG,Nov|. 2.— (Th]e Admiralty. . 5 1334 266 

670 1U6 616 

F ST. P|TERSBUR[G, Nov). 2.— The A[dm)iralty 3 2432 810 

811 148 140 171 351 163 273 

M (S[T. |PE]TER|SBUR[G, )No[v.) 2.— [Th)e Ad|miralty 7 2057 293 • 



2 



T has telegraphe|d to the offi|cers of the |Baltic 3 493 164 

330 220 160 . 240 

H ha[s telegra)phed to th|e o(f6cers] of the [Bal)tic 4 950 237 

314 144 306 23S 233 283 136 

S [) |has telegrap[he) (d] to th[e of)ficers| of th|e Baltic. . 7 1639 234 

291 179 332 587 529 

F [has telegr|aphe|d to the [o)fficers of the B|altic 5 1968 393 

218 195 231 226 171 179 78 

M |ha|s te'egraphed [to) the oflficelrs of the |Ba[lt)ic. ... 7 1301 185 

Figure 58. The vertical lines and brackets indicate the locations of the eye 
fixations. The numbers give the length of the fixations in loooth of a sec- 
ond. After Dearborn ('06). 

Figure 58 shows the locations of successive pauses, their lengths 
in 1 000th of a second, and the distances between them as deter- 
mined by Dearborn. The brackets and vertical lines indicate the 
locations of these fixations. Sometimes there is a slight movement 
after the fixation. The direction and distance of this movement 
is indicated by the parentheses. These regressive or corrective 
movements occur when the eyes have moved too far to the right. 
They also occur when the eyes make the sweep back to the begin- 
ning of the next line but have not gone far enough. These correc- 
tive movements come more often in long lines than in short ones. 
They are probably caused by the fact that the peripheral percep- 
tion of the beginning of the line is not accurate. The exact location 
from the beginning and from the end of the line is gotten as a 
habit after several lines have been read if the lines are of uni- 
form length. For this reason the lines should be of the same 



268 KIHCATIOXAL I'SVCUOLOGY 

length, especially for beginners. The opposite condition, however, 
is often found in reading texts in which each sentence is printed 
as a ])aragraph and in which jiictures are inserted in the margins. 
The latter is still more objectionable when the pictures are colored, 
since the reflex action of colors influences the position of the fixa- 
tions. The numlxTS above the vertical lines, Figure 58, indicate 
in loooth of a second, the length of each pause. It will be noticed 
that these vary considerably from time to time and that on the 
average they are approximately one-fifth of a second in length, 

Schmidt found on the average nearly twice as many refixation 
movements in oral as in silent reading. These movements grow 
l)rogressively fewer and shorter from year to year as children 
.l)rogress through the grades. Gray found frequent refixation 
mo\'ements \'ery characteristic of slow readers. They were also 
much more numerous with all readers in extremely careful ri-ad- 
ing. 

Dearborn found that, with newsjjaper lines, five persons made 
an average of 4.76 fixations per line, that the number of fixations 
tends to decrease and with it the range of apjirehension tends to 
increase as a passage is read over and over again, and that the eyes 
finally develop consideral)le ])recision and accurac}' in the number 
and the position of the fixations. He also found evidence for the 
belief that the eyes form a short-lived motor habit which deter- 
mines the number and jmsition of the fixations for each line in 
reading. The formation of this habit develops during the reading 
of the first three or four lines of print and seems to come much 
sooner for some persons than for others. The fast readers seem 
to form the motor habit more quickly than the slow ones. The 
decrease in the time is due to a shortening of the pause and a 
widening of the scope 01 attention, especially in the latter i)arts 
of the line. 

The fixations may come at any point in the line, between words, 
al the beginning, at the end or in the middle of a word. "The 
first and last fixations generally fall within the edges of the line, 
that is, a little distance from the l)eginning or the end words." 
The first fixation in a line is usually longer than the other fixations. 

There arc fewer pauses in the re-reading of a gi\en material, 
even aftir one month, and the number decreases somewhat as the 
selection is read repeatedly. rrepositi»)nal jjhrases, connectives, 
and substantives make the greatest demand ujKm perception due 
to the fact that a slight change in them affects the whole meaning 



READING 269 

of a sentence. Dashes, punctuation marks, and capitals in the 
middle of a line change the points of fixation, and disturb some 
readers more than others. Reading a foreign language or reading 
aloud requires longer fixation than reading one's native tongue or 
reading silently. Dearborn obtained records of three boys aged 
9, 10, and II years, in the third, fourth, and fifth grades respec- 
tively, and found that the oldest one approached very closely to 
the conditions of fixation of adults, while the younger ones made 
many more fixations or pauses. Fatigue resulting from the use of 
the eyes during an entire day decreases the rate of reading by 
about one-tenth. 

Schmidt ('17) made an investigation of the eye movements, in 
oral and silent reading, of eighty individuals including elementary, 
high school, and university students. He found that they made 
1.6 more pauses per line in oral than in silent reading, that the 
average duration per pause was from 20 to 27% longer, and that 
the perception time was from 44 to 64% longer. 

By an ingenious combination of camera and phonograph, Gray 
determined accurately the relation of the eye to the voice in oral 
reading. He found that the eye always precedes the voice, with 
some subjects as much as four words and with others much less. 
As a general rule a wide eye-voice span was associated with fluent 
reading. 

(5) Transmission of visual nerve impulses to the visual area of 
the brain. Little can be said about this process. All that we know 
about it is that the velocity of nerve impulses varies in different 
persons, and may be slow or rapid in slow or rapid readers, but we 
do not know definitely. Experimental work is necessary to de- 
termine to what extent it may be true. 

(6) The arousal of association processes. A word gradually 
acquires meaning in the life of a child by its recurrence in numerous 
situations and by the connection of the particular significance or 
experience with the word. Thus, the child hears the word "chair" 
in connection with a certain object on which he sits. He hears it 
also after a while in connection with other objects which look 
different from the one he is accustomed to using. But new associ- 
ative bonds are formed and he knows in general what is meant 
when he hears the word "chair." Later on he is shown a certain 
combination of visual characters and is told the word "chair." 
A new bond is then formed between the sound and the visual 
stimulus of the word "chair." An important part of the act of 



270 KDUCATTONAL PSVCII()I,0(;\ 

reading consists, tlunlorr, in the arousal of associative bonds as 
soon as the visual nerve impulses arc brought in from the eye to 
the visual brain areas. The quickness of reading depends, no 
doubt, upon the rapidity Avilh wliich these incoming stimuli arouse 
associated bonds and thereby gi\e meaning to the word. 

That the rate of reading depends upon the rapidity with which 
the visual stimuli are inteq)retcd is shown by such cx]^erimcnts as 
Huey made in which he showed that it takes about twice as long 
to read nonsense as sense material. He also found that words in 
context give fewer associations than words out of conte.xt. When 
the words were shown in context the associations would sometimes 
go ahead of the amount seen and would anticipate what was com- 
ing. 

Hamilton (07) attemi>ted to determine the jiarl j^layed by 
context in reading by ascertaining the time required to perceive 
isolated words and words in various relations, such as in para- 
graphs, in miscellaneous sentences, and in miscellaneous phrases. 
His results, expressed in terms of seconds per word required for 
the reading, are as follows: 

TABLE So. .\flcr Ilamillou 

Pakagraphs Sentences Phrases Words 

Mean. . .429 .456 .466 .660 

M. V. . . .047 .047 .041 .078 
Percent iisinj; |).ir,iKr;i|''' 

as base 100',', 04% 02% 65% 

Isolated words a])parc'ntiy re(|uire a distiiutiy longer tinu- for 
[)erception than words in context. 

Hamilton also found that the first jiart of a word is more im- 
])ortant in giving the meaning of a word than the latter part is, 
that the marginal impression at the right aids in the perception of 
the words seen in the next eye fixation, and that the upj)er 
and lower marginal imj)ressions jirobabiy do not aid in per- 
ception. He also thinks it prol)able that the inter])retation 
takes i)lacc during the rest jjcriod, that is, during the movement 
of the eye. 

The central neural and nuntal activity of ])utling meaning into 
the incoming impressions is probably the most imixirtant step in the 
whole reading process. The important part of reading really is the 
reading of meaning into words. Judd, McAllister, and Steele bc- 
lie\c that tin- essential factors controlling the rate of reading arc 



READING 271 

central rather than peripheral and that it is a matter of assimilation 
rather than a matter of getting material into the brain. Ruediger 
has a similar opinion : 

"In reading, a similar reinstatement of experience takes place as in 
thought or in oral communication. The printed symbol arouses the 
meaning that has through education and experience become connected 
with it. It is to the rapidity with which this meaning is aroused that we 
have to look for the cause of the differences in reading rate. 

"Reading rate may then be taken to depend chiefly upon the rapidity 
with which meaning is aroused in the mind after the symbol is seen. 
This, in turn, is in the main dependent upon the person's native brain 
inertia." 

(7), (8), and (g) Transmission of nerve impulses from the visual 
center to the motor speech centers and thence to the speech organs. 
The motor speech centers concerned in the control of the speech 
organs are highly specialized. In right-handed persons, they are 
located in the left hemisphere of the brain in the region of the fis- 
sure of Rolando. These processes obviously will be involved in 
oral reading and in speech, but they are also active in silent reading 
in the form of incipient speech movements, particularly in the 
tongue and, to a less extent, in other speech parts. They are 
active in the same manner as in speaking, only on a much smaller 
scale. To what extent this inner speech is an important part in 
the reading process is somewhat uncertain. It is believed by 
some investigators to be an important agency for maintaining 
the continuity of the thought aroused by the successive visual 
glimpses of the printed line. 

Disturbances in the motor speech centers are known under the 
names of various types of aphasia which will not be considered here. 
But even in normal persons the rapidity and facility with which 
the neural centers act may have an effect upon the efficiency of 
reading. 

Much valuable information could undoubtedly be obtained 
from a careful laboratory study of individuals with language de- 
fects, such as persons who have difficulty in learning to read or 
difficulty in the speech functions proper. The following case may 
be of interest in this connection. A boy, 17 years of age, in the 
second year of high school, had always had considerable difficulty 
in reading. His reading was extremely slow. He was finally sent 
to a private teacher of public speaking with the hope that his 



272 r.lJlCA'lKAAL I'SVCHOLUGV 

reading ability might be improved. After several months of work 
with this teacher, apparently no progress was noticeable. He was 
then brought to the psychological laboratory for examination to 
see, if possible, wherein his dinicully lay. As careful tests as {X)ssi- 
ble were made to discover in which of the various steps in the read- 
ing process the difficulty might be. These successive processes 
were tested with such means as were available or could be devised, 
beginning with the first, the reception of the stimuli upon the 
retina. Inquiry showed that his eyes had been carefully examined 
and found to be only very slightly defective. Glasses had been 
supplied as his father was a physician. Tests were next made to 
determine the range of distinct vision by obser\ing how large an 
area he could see distinctly at one time. Next, his span of atten- 
tion in apprehending visual stimuli was measured. This was 
done by rapid exposures with a tachistoscope to determine the 
number of words or pictures of objects he could apprehend at one 
time. In the next j)lace, tests were made to determine the con- 
trol of his speech organs. It was thought that possibly his diffi- 
culty in reading might be in an inability to control rapidly his 
speech organs. Tests were made by having him repeat statements 
from memory as ra[)idly as he could and also by having him repeat 
short sentences from dictation. All these tests indicated that 
these various functions were normal. His eyes had only slight 
visual defects, his field of distinct vision was apparently of normal 
size, his span of attention likewise was as large as that of a normal 
individual, and the control of his speech organs was rapid and 
accurate. The inference by a process of elimination was that the 
chief difficulty in his reading ability lay in the central assimilation 
or association [)rocesses. It seemed that visual impressions were 
brought into the brain centers with normal speed and facility, 
but that there was, for some reason, extreme slowiiess in the 
mental interpretation of these stimuli. A test in silent reading 
and in oral reading showed him to be cfiually slow in both cases. 
His rate of reading was approximately that of a child at the end 
of the first grade, namely, 1.5 words per second. His compre- 
hension of what he read was good. His general intelligence, as 
shown by other tests, was normal for his age. His work in other 
school subjects was satisfactory. It seemed, therefore, probable, 
although not absolutely ci-rtain in (he absence of further tests, that 
his difficulty lay in the ( entral interpretation processes. This case, 
similar to others, shows in an interesting niannir tlu' gnat iiitritacy 



READING 



273 



of the reading functions, and tlie difficulty in determining pre- 
cisely what may be the trouble in a child who is backward in a 
given school subject but normal or even superior in all others. 

In order to ascertain more fully the part played by the various 
factors enumerated at the beginning of this chapter, a series of 
tests was carried out by A. D. Mueller under the direction of the 
author,^ with 36 high school pupils. Speed and comprehension of 
reading were carefully measured by three different selections 
according to the writer's method. Then each of the capacities 
mentioned in Table 81 was measured, and the coefficients of cor- 
relation computed. These show that the two elements upon 
which rate and comprehension of reading probably depend most 
are the visual apprehension span for related and unrelated words 
and quickness of association, correlation Nos. 2,3,6 and 12. Ap- 
parently the amount apprehended and the quickness of giving 
meaning to the visual symbols are the most important factors in 
reading. Correlation No. 6 corroborates Nos. 2 and 3 since the 
number of eye fixations per line of print is probably inversely 
I)roportional to the amount apprehended at each fixation. 



TABLE 81 

Correlations between reading ability and various elements entering into reading 

ability 









Speed 




Speed 


Compre- 


PLCS 




hension 


Compre- 








hension 


I. Visual attention span — letters 


.40 


•32 


.41 


2. " " " — unrelated 








words. . . . 


.64 


■73 


.70 


3. " " " — related words 


.70 


•59 


.69 


4. Auditory attention span 


.30 


■i7 


•34 


5. Rapidity of voluntary eye move- 








ment 


.31 


.40 


•38 


6. No. of eye fixations in reading .... 


•54 


■49 


.67 


7. Association with auditory stimuli. . 


•38 


.28 


■35 


8. " " visual " 


■39 


•56 


.42 


9. Quickness of articulation — alphabet 


■34 


.42 


•38 


10. " " " — rhyme. . 


.62 


.76 


•63 


II. " " " — dictation 


■3i 


•43 


•38 


12. Continuous association 


•63 


.46 


.62 






^ Reported in an unpuijlished thesis in Ihe ! 


brary of the L 


niversity of W 


isconsin, iqi8. 



2 74 EDUCATIONAL PSYCHOLOfJV 

These results are supported completely by a \ery similar ex- 
periment of Gray's, although he computed no correlations. As 
in thi' author's experiment there was a close relation between 
reading aljility and the range of visual apprehension for meaningful 
material but, "The ditTerence between the span of attention of 
the good and poor readers disappears in a very large measure when 
nonsense syllables, digits, groups of the same digit or the aussage 
test are given." * 

The Measurement of Efficiency in Reading 

(i) Essential Elements to be Measured. In order to he able 
to determine the elTectiveness of dilTerent methods and procedures 
in learning and teaching a given school subject, it is necessiiry 
to be able to measure with some degree of precision and objectivity 
the general ability in that function. In order to measure such 
abilities it is necessary, furthermore, to determine what the essen- 
tial elements in the process are which ought to be measured. These 
would seem to be in the case of reading as follows: 

(a) In silent reading, speed and comprehension should be 
measured. The prime purpose of reading is the comprehension 
of the thoughts presented upon the printed page. The second 
clement is necessarily the speed at which the thought processes 
may be obtained. 

(b) In oral reading the chief additional element to be measured, 
besides speed and comprehension, is the correctness of the pro- 
nunciation. 

(2) Method of Measurement. Several types of tests have 
been developed for measuring etliciency in the various aspects of 
reading. These methods will simply be mentioned here without a 
detailed, critical discussion of their relative values or of their 
various techniques of administration and evaluation. Tests for 
speed and comprehension of silent reading, have been devised by 
Gray, Kelly, Starch ('15), Courtis, Brown, Fordyce, and others. 
The writer's test consists of a series of eight passages, one for each 
grade. It is arlministercd by having the pupils read for ;^o seconds 
and by having them write out as full an account as possible of the 

' Gray draws the followinR conclusion which is at least worth consideration: — "This 
seems to prove that the iiihcrinl ditTcrenccs iu mental cajvuily which exist between 
memlnTs of this Kfi'I' Jire not tile causes of iliffcrences in the siKin of atlentiou. It 
aj)|K'ars rather that differences in training that is, acc|uirc(l abilities to ileal with mean- 
ings—arc the source uf diilcrcnces in |)crccptual spau." 



READING 275 

portion read. Speed is expressed in terms of the number of words 
read per second ; comprehension is expressed in terms of the number 
of words written after those words have been discarded which 
represent incorrect statements of thought, additions of ideas not 
found in the original passage, or repetitions of ideas previously 
stated. The tests by Gray, Courtis, Brown, and Fordyce measure 
rate and comprehension of reading substantially in the same manner 
except that in some of them comprehension is determined either 
partly or entirely by answers to questions, and in the fact that 
most of them employ fewer than eight test passages. For ex- 
ample, Gray uses three, Fordyce two, and Courtis one passage for 
all grades. 

Thorndike's ('14) tests are designed to measure comprehension 
in reading, either of isolated words or of paragraphs. They are 
constructed on the scale principle. Similar tests have been pre- 
pared by Haggerty ('17). The Kansas silent reading test (Kelly) 
is designed to measure Comprehension primarily and speed second- 
arily. 

Gray ('17) has devised a test for oral reading consisting of a 
series of twelve graded passages which are used to determine the 
rate of oral reading and the number of errors in the pronuncia- 
tion.^ 

(3) Uses and Results of Measurements, (a) The first obvious 
use of exact measurements of reading ability is a determination 
of the actual abilities of the individual pupils, classes, or schools. 
The first logical step in the management of any activity is a diagno- 
sis of the conditions as they exist. Then, upon the basis of this 
diagnosis, it is possible to prescribe more intelligently what should 
be done. In order to be able to make definite comparisons of the 
abilities of the pupils, it is necessary to know the standard norms, 
or averages, of achievement in the various grades. For the writer's 
test these are as follows: 







TABLE 82 


Grades 


I 


2 3 


Speed of reading 

(words per second). . . 
Comprehension 

(words written) 


• 1-5 
15 


1.8 2.1 
20 24 



2.4 2.8 3.2 3.6 4.0 

28 33 38 45 SO 

^ Further discussion of these tests may be found in the original sources in which these 
tests were reported, or in the writer's Edncalioiial Aleasuremenls, or in Monroe's Edu- 
calional Tests and Measurements. 



276 



EDUCATIONAL I'S V( IK )!.()( ;V 



Thtsc measurements are shown ^^Tajjhically in Figure 51) and 
indicate the grade-to-grade progress in sjK'cd and comprehension, 
showing that there is a continuous ini])rovement in l)oth asjjects 
from year to year. liy reference to these norms it is possible to 
exjiress definitely a gi\en pujMl's cajiacities in reading. It enables 
one to say, for example, that pupil A in the 4th grade has a reading 
ability equal to the average ability of pupils in the 6th grade, 
pupil B in the 4th grade has a reading ability e(jual to the average 



i 



Speed 













^^^ 


" 






^^<: 










C'] 


"" 










1 : 


' 


i •! 


5 « 7 8 



Grades 



eo 

c50 
'u 40 

ki 

o 
10 



Comprehension 





j 






/ 














/^ 








^ 




'"' 






^^ 


.^^ 


^^^ 








^ 


'""^ 













I 5 

Grades 



Fir.. 50. — Till' (ontiniious lines represent the shindanl attainments in n-ad- 
inj^. Tlic lirokcii lines represent llie uttainmeiUs in u certain school. 

ability of ])Upils in tin- 4th grade, and ])upil C has a reading ability 
ecjual to the average of pujjils in the second grade, and so on. It 
is possible to describe i)recisely the ability of a ])upil in relation to 
others because the terms in which his abilities are cxprcs.sed are 
accurately defined. Measurements of this sort have disclosed 
enormous ranges of ability in the various school subjects as 
l)ointed out in a ])receding chaj)ter on individual dilTerences. 
They have shown that at the present time ])Uj)ils are not 
promoted according to ability, but rather according to the num- 
ber of years they ha\«' attended school. Thus, for example, the 



READING 



277 



pupils in a fifth grade range all the way in reading ability, from 
the second or third grade on up to the 8th grade. (See Figure 16 
in Chapter III.) 

Another interesting comparison at this juncture is the reading 
ability of the pupils of various ages in each grade. Such a study 
was made of the pupils in one school by the writer, and is shown in 
the following table: 

TABLE 83 

Showing the relation between age and attainment in reading 



Grade 
Age 


3 
Speed Coup. 


4 
Speed Comp. 


5 
Speed Comp. 


6 
Speed Comp. 


7 
Speed Comp. 


8 
Speed Comp. 


7- 




2.2 16.6 












8. 




2.6 29 . 2 


4.6 36.9 










9- 




2.0 29.0 


4-4 4,5-4 


6.4 505 








10. 






4.8 40.0 


5.1 46.0 


4.8 59.9 






II . 






4.6 42.5 


4.6 37.0 


4-4 45-2 


4-5 45-5 




12. 






2.2 47.8 


3-4 -'9-5 


3-4 29.4 


4-7 45-0 


6.5 81.9 


13- 










2.3 28.0 


3-4 39-8 


5.4 64.1 


14- 










2.8 16.0 


4-6 52.5 


45 52.0 


15- 










4-4 21-5 




4.2 60 . 



From the fourth grade on there is a fairly regular decrease in 
reading ability from the youngest to the oldest pupils in the same 
grade. The explanation for it is probably the fact that the stupid 
pupils are on the whole promoted too rapidly and the bright pupils 
too slowly. 

(b) The establishment of accurate norms by means of def- 
inite measiirements has meant, furthermore, a more precise 
estimate of definite aims of attainment. To say that a pupil 
at the end of the eighth grade should be able to read at the 
rate of four words per second and report a correct thought 
content expressed in at least 50 words, means something def- 
inite. The pupil, as well as the teacher, will know what each 
one means. 

(c) The third and probably most important use of measurements 
of reading ability is their employment in the investigation of the 
factors and conditions affecting the learning and teaching of read- 
ing. In the long run the greatest good from tests of reading ability, 
or from tests of any school capacities, will come from their service 
in analyzing and measuring the potency of the numerous elements 
that enter into the acquisition of knowledge and skill in a school 



278 



EDUCATIONAL PSYCIIOLOC.V 



subject. Kcsults on this phase of llic subject will be summarized 
in the following section. 

Speed 



^2 























1 
/ 


\ 

\ 




/ 








/ 

/ 
/ 

1 


'^\..-7- 




_^*''' 


,^^' 




^^ 


^ 


-^ 


y 




^ 


^ 








-^ 















12 3 4 5 6 

Grades 

Comprehension 

60 

c 60 
■C 40 

o 
^ 20 

lU 

1 2 3 4 5 (5 7 « 

Grades 

Fir,. 60. — The rontiniioiis lines refiresent the stand.in] attainments in reading. 

The broken lines represent tlie attainnieiUs in a lertain .selux)!. 











' — -f— 

1 / 
/ 








,^-' 


■-^^ 




>^ 








^^ 












-^ 










^^ 















Economic Proci:i)1'rI': in Lkarninc. to Rf.ao 

(i) Diflferences among Schools and Classes. The difTcrcnccs 
in acliievement in nadinj^ amon;^ classes as a whole are enormous. 
Such differences are shown in Figures 59 and Oo. Thus we sec 
that the best classes in Figure 60 have scores a]>]>roximately twice 
as high as the corres])onding classes in Figure 50. These dilTer- 
enccs are too large to i)e i'.\i)lained by hereditary factors, but must 
be due chiefly to dilTerenccs in learning and in method and sj)irit 
of teaching in these schools. What these dilTerences in procedure 
of learning and teaching are must l)c discovered more fully in the 
future. Hut it is certain that some methods |)ri)duce almost 



READING 279 

double the achievement produced by others. Thus the school 
represented in Figure 60 is, grade for grade, as much as from i to 
3 grades ahead of the average of schools generally, while the school 
presented in Figure 59 is from i to 2 grades below the average of 
schools generally. The former school has attained such proficiency 
in reading, probably chiefly because of the plan inaugurated by 
the principal which did away with nearly all word drills and phonics 
and placed all emphasis upon trying to read and upon reading as 
much as possible. He believes that a child learns to read by reading 
a great deal. He has described the method of teaching as follows: 

"Since first grade reading is in the initial stage materially different 
from that of the succeeding grades, it is treated separately. 

"The reading is begun in the first method reader by means of the word 
and sight method, and the phonics is carried on parallel with the reading, 
but in a separate period. No attempt is made in the early stages to 
correlate the phonics with the reading. During the entire work on the 
first book, the pupils are given words and sentence drills. Each lesson is 
read oraUy by every child after he has read it to himself. The time re- 
quired to cover this book varies from two to four months, depending 
upon the ability of the child. When this book is finished, the child has a 
working vocabulary, and the method is changed to more extensive 
reading. The word drill is now omitted, and the lesson covering from 
five to fifteen pages is read over by the pupils, they or the teacher pro- 
nouncing the difiicult words. Next the pupils read the lesson over 
silently, and the following day they are called upon to read a page each 
without assistance. 

"When the seventh unit reader, at about the close of the sixth month, 
is in the hands of the pupils, the group reading is begun. The pupils are 
seated in groups of two, and each group is provided with an interesting 
reader. One pupil in each group reads aloud; when the first pupil has 
read three or four pages, the second pupil reads an equal number, then 
the first continues. In this way they alternate untU the time is up. A 
fluent reader is often placed with a poor one for the purpose of assistance. 
At the close of a ten or fifteen minute period the place where the reading 
is discontinued is marked by a paper, and the books are laid aside until 
the next group reading period. Group reading is not, as a rule, practised 
oftener than twice a week. 

"Sectional Reading: — This is a phase of group reading. Each group 
of two or more stands before the class. The first pupil reads three or four 
pages orally, the second continues the reading for three or four pages, and 
so on. In this type of reading the teacher may question the child as to 
what he has read, the child may reproduce it, or some listening member 
may tell what he has heard. 



28o r.DrtATIONAL I'SVCIinLOOY 

"Silent Reading: — During the study hour or scat work pcrio<l, silent 
reading is conducted by means of single copies of books containing in- 
teresting material. Every child is given a book, and he reads as many 
pages silently as he can during the jjcriod. When the silent reading jK'ricKi 
is fmished, a mark is placed where the reading ended. At the ne.xt silent 
reading period the pupil continues his reading, and so the work progR'sses 
until he has hnished the book. The silent period may be continued as long 
as the teacher wishes. A record is kept of each child's reading by checking 
ofT a book as soon as it is finished. .Ml through the year, a unit book is 
used during the regular recitation period for the drill that is necess;iry. 

"Books are jilaced in the hands of the children on the first day of 
school, and they are allowed to keep books at their desks to read in 
school or at home as they desire. 

"With this system, each child can go his own gait, reading as many 
books in a year as he can. The best readers will read from thirty-five to 
forty books. The average is about twenty books each." [Reix)rt of Rea<l- 
ing in Dodgeville (Wisconsin) Public Schools, by Supt. II. W. Kircher]. 

The cfTiciency obtained by this process of teaching reading is 

shown definitely l^y means of various tests that have been given 

in this school as shown in Table 84. The results obtained in this 

manner from encouraging pupils to read so e.xtensi\ely are further 

indicated by the large number of books read as shown at the bottom 

of the table: 

TABLE 84 

Attainments in nadinj; in a certain school 

Datk or 

Grades 1 2 J 4 5 6 7 8 Ttsr 

Starch Test: 

Speed in words per minute 96 . 126. . 156 240. . .118 . 2^2 . 264 . VIO 

Comprehension in words written 2M 42.. 46 41.. 46.. 65 Fel). 1917 

Percent above June sLnnilard in S|«ei<l 7. . 17.. 25.. 66.. 90 . J9. . 20 . M 
Percent alxjve June standard in com- 
prehension 16.. SO.. 4U. . 8.. 2.. 30 

Kansas Silent Reading Test: 

Score for each grade 12 19 !(>> 14 1 19 6. 2< J.in. l'M7 

Per cent atwve May 100. . 98. . 22.. 7.. 11.. 11 

Fordyce Reading Test: 

Speed words per minute '. 209.. 272 . 2.W 276 . 2.S0 April 1917 

Quality 76. . 75 . 50 . .50 . 70 

Efficiency 65.. 69.. 22.. 20.. 37 

Courtis reading Te<it : 

Words per minute 150 190.. 220 220. 220 280 May 1917 

Comprehension 75 01 90 92 95 95 

Questions answered . . 34.. 41.. 36.. 40.. 45.. 67 

C.RADF.s 1 2 .1 4 5 6 : 8 

Ma'.imum number o( lxx>ks reail b) 

any pupil .?8 9(> 90 150 10| 120 105 . 100 

Minimum number of txmks rcid by 

any pupil 20 45 41 .'9 . 42.. 20 17 . 18 

Average numlter of books read |>rr 

pupil 31. 65 . 63 . 80 77 47 42 55 

Average age of pui.ils 6. . 7. 1 . .8. J. .9.2 10.? ||.4 12 2 .13.1 

(Dixlgevilie Re|X)rt, page II.) 



READING 



2«I 



(2) The Possibility of Improvement in Reading Ability. Huey 
and other investigators report that they have been able, by special 
effort and practice, to double their speed of reading. Miss Harriet 
O'Shea (under the direction of Professor Henmon) conducted 
an experiment with a group of high school pupils in which she 
attempted to determine to what extent the rate of silent reading 
could be increased by specific practice. Her plan was carried 
out by giving each pupil a book, usually of high literary quahty, 
and asking him to spend 15 minutes each day in reading the book 
until it was finished and to keep account of the number of lines 
read. Her results showed a rather remarkable increase in the rate 
of reading from the beginning to the end of the book. Some pupils 
gained very little and a few others gained very rapidly, as shown 
in the following table: 

TABLE 8, 



Pupils 



Average No. of 
Lines Read i.v 
First Two 15- 
MiN. Periods 


Average No. of 
Lines Read in 
Last Two 15- 
MiN. Periods 


No. of Lines 
of Gain 
OK Loss 


Percentage 

OF G.AIN 

OR Loss 


696 


1445 


749 


107% 


793 


1364 


571 


72 


560 


1038 


478 


85 


553 


964 


411 


74 


461 


648 


187 


40 


430 


758 


328 


76 


420 


715 


295 


70 


416 


655 


239 


57 


403 


569 


166 


40 


367 


732 


365 


99 


365 


528 


163 


45 


364 


467 


103 


28 


355 


390 


35 


10 


322 


667 


345 


107 


303 


362 


59 


19 


275 


480 


205 


74 


263 


685 


422 


160 


260 


415 


155 


60 


253 


371 


118 


46 


245 


240 


-5 


■—2 


241 


484 


243 


100 


241 


377 


136 


56 


235 


343 


loS 


46 


235 


224 


— ir 


—4 


224 


317 


93 


41 


204 


125 


79 


38 


142 


139 


3 


2 



3 
4 
5 
6 

7 
8 

9 
10 
II 
12 
13 
14 
15 
16 

17 
18 

19 
20 
21 
22 

23 
24 

25 

26 

27 



282 KDUCATION'AL PSYCHOLOGY 

Such experiments ought to be repeated again with additional 
attention to the question whether or not comj)rehL'nsion improves 
in a proi)()rtic)nate manner, and also whether tliis gain in the speed 
of reading would airry over to reading in general. At any rate, 
the results are significant in showing such large gains after a rela- 
tively short ])erio(l of practice and elTort to improve the speed of 
reading. 

Peters ('17) undertook an experiment to determine the elTect 
of s])eed drills conducted regularly in connection with the reading 
work during a school year. The pupils in grades three to six in the 
public schools of Royersford, Pennsylvania, were di\ided into 
"drill" and "no-drill" sections. The manner of conducting the 
different sections is described as follows: 

"The groups which were not to have the speed drills, and which were 
to be used as a basis for comparison with those which did have, were 
dealt with after the usual fashion in teaching reading. The other grou{)s, 
in addition to their oral reading, were given daily speed drills, without, 
however, giving a total of any more time to their reading than the other 
group received. So far as feasible both groups were taught reading at 
about the same time of day, or else at equally desirable i)erio<is. They 
used the siune books and the siinie degree of enthusiasm was expected 
to be put into both. The drills were, of course, conducted by the teacher 
in charge of the class, and ran from November jlh to June 2nd. They 
were on relatively easy reading matter, and mostly interesting narrative. 
They occupied ordinarily from five to ten minutes of the reading period. 
The group as a whole was told explicitly where to begin and how far to 
read, and were then all set to silent reading at the s;mie time with the 
exhortation to see who could get it read first. After all, or nearly all, had 
finished someone was asked to tell the substance of what he had read. If, 
in this reproduction, he omitted anything he was questioned on it as a 
guarantee against skimming." 

Tests were gi\en to both groups at four points during the school 
year, comj)aring the drill grou]>s with the no-<lrill groups as a 
base. From the first test to the last, the results showed a g:iin in 
speed of 18.7V0 imd the trifling loss of 1.1% in quality of com- 
prehension. 

Freeman ('16) reports a series of tests made by K. 1). Waldo 
on the i)ossibiIity of increasing sjK'ed. The lower grades ])articu- 
larly made a very large gain in speed which was accompanied by a 
parallel gain in amount reproduced, as indiaited in Table 86. 



READING 



283 



TABLE 86 
Improvement in reading from September to March * 



Grade 3 

September. . . 

March 

Per cent gain. 
Grade 4 

September. . . 

March 

Per cent gain . 
Grade 5 

September. . . 

March 

Per cent gain . 
Grade 6 

September. . . 

March 

Per cent gain . 
Grade 7 

September. . . 

March 

Per cent gain . 
Grade 8 

September. . . 

March 

Per cent gain . 



Rate in Words 
PER Minute 



76. 
149. 

72. 

92. 

163. 

70. 

113- 

129. 

16. 

128. 
130. 



122. 7 

142.8 

21.8 

147.2 

158.9 

II. 7 



Amount 
Reproduced 



13s 
63 

133 
212 

79 

52 
70 
iS 

52 
85 
33 

75 

125 

49 

116 

179 

63 



Percentage of 

Correct Answers 

TO Questions 



44.6 

44- 



56 

60 

6 

16 

25 

9 

27 
35 



(3) The Relation of Speed and Comprehension. This question 
has been one of perennial interest, and a common misconception 
has been held by a great many people regarding the mutual relation- 
ship of these two aspects of reading ability. Many people believe 
that a rapid reader comprehends relatively little of what he reads 
and that a slow reader makes up for his slowness by a more thor- 
ough comprehension of content. In order to obtain some specific 
facts on this question, the writer obtained the results from a careful 
test of reading in an elementary school in Port Townsend, Wash- 
ington, and divided the pupils of each grade into six groups ac- 
cording to their speed of reading, putting the slowest sixth together 
and the next sixth together, and so on to the last sixth, consisting 

^ From an unpublished master's thesis by K. D. Waldo, on file in the library of the 
University of Chicago. 



284 



p:ducatio.\'.\l psvchulu(;v 



of ihc most rapid readers. The results are exhibited in the follow- 
ing table. The first column gives the average number of words read 
by each group, the second column the average number of words 
written representing a correct report of the thought, the third 
column gives the speed of reading in terms of the number of 
words read per second, the fourth column gives the percentage 
of words read in relation to the number of words written. 



TABLE 87 
Relation between speed and comprehension 



, Words Read in 
30 Seco.vds 


Words 
Written 


Speed per 
Second 


I'l-R Cent oi 
Words Written 
OF Words Read 


36 


16 


I . J 


4'>'; 


SI 


-'-' 


1-7 


43 


69 


^4 


2-3 


35 


90 


M 


3.0 


37 


los 


33 


3 5 


31 


147 


54 


4-9 


37 



These results indicate in a striking manner that the rapid reader 
comprehends relatively almost as much out of what he reads as the 
slow reader, and, absolutely, he grasps nearly as many more ideas 
in a given period of time as is proportional to the extra ground 
covered. Specifically, the table .shows that the average speed of 
reading of the slowest group was 1.2 words per second, whereas 
the sj)eed of reading of the fastest group was 4.9 words per second. 
The percentage of comprehension in relation to the amount reacl 
was 4b^/f for the slowest group and .S7% for the fastest group. In 
other words, the percentage of comprehension is almost as large for 
the fast group as for the slow group; or, comparing the first and 
second columns, we note that the fast group read almost exactly four 
times as fast as the slowest group and wrote three and one-third 
times as much as the slowest grouji. In other words, the ratio of 
the s[)ee(l of the reading between the fast and slow group is one to 
four, while the ratio of comjirehension is one to three and one-third. 
The inference is then that the rapid reader derives relatively almost 
as much out of what he reads as the slow reader. Absolutely he 
obtains several times as many ideas. Concretely, the comparison 
may be made in still another way: Of two persons lu-longing re- 
spectively to groups one and si.\, each reading for one hour, the 



READING 285 

fast reader would cover four times as much ground and derive three 
and one-third times as many ideas as the slow reader. The fast 
reader, therefore, has an astounding advantage over the slow reader. 
These results consequently give no corroboration for the common 
belief that an inverse relation exists between speed and compre- 
hension in the fast and the slow reader. 

Similar results have been presented by Judd. These results 
are shown in the following diagram, Figure 61: 

" For the purpose of this study of the relation between rate and quality, 
all of the individual records of Cleveland pupils were divided into classes. 
First the speed records were arranged in order from the most rapid to the 
slowest. The most rapid of these records were designated by the simple 
term 'rapid.' In this class of 'rapid' records were included the most 
rapid 25% of all the records. In like fashion the slowest 25% of all the 
records were set aside and designated as 'slow.' This left half the rec- 
ords, or the middle 50%, which were designated as of 'medium speed.' 
In like manner the 25% of all records which were qualitatively the best 
were designated 'good'; the 25% which were qualitatively the worst were 
designated 'poor,' and the term 'medium' was applied to the middle 50%. 

"It becomes a very simple matter to assign all records in each grade 
to the appropriate class and determine the percentage of the grade which 
falls into this class. Diagram 59 gives the results, the percentages being 
in each case the nearest whole number to the calculated figure, and the 
size of the circle being proportionate to the size of the class indicated. 

"These figures serve to emphasize the fact that good readers are 
usually not slow and poor readers are usually not fast. It is evidently 
not safe to attempt to lay down any absolute rule. There are good read- 
ers who are slow. In some cases such readers may be temperamentally 
slow. But even making allowance for such individual peculiarities, the 
figures show that good reading and slow reading are not incompatible. 
In like manner there are a certain number of children who read rapidly 
but retain little of what they read. With the figures in hand a teacher 
can profitably study her class and determine somewhat more completely 
than it is possible to do for the whole school system what are the special 
explanations of each individual type of ability. 

"For the purpose of this survey the general fact that high rate and 
good quality are commonly related, and that low rate and poor quality 
are commonly related, is of great importance." 

King ('17) tested the reading ability of 94 college students, 
half of whom were asked to read slowly and carefully and half to 
read rapidly and carefully. They read for ten minutes and, by 
observing a clock, the fast group were to read twice as rapidly as 



286 



l.Dll Al lONAL l'SVClI()l.()t.\ 



the slow j^roup. Comprehension of the material was tested by 
answers to questions. The results showed that the accuracy of 
comprehension of the fast group was 44.5% and that of the slow 
group was 5.^,?%. In another test the subjects were divided into 
groups of naturally fast and naturally slow readers as determined 
by a preliminary test. The returns showed that the fastest 25% of 
the group liad a comprehension of 50. 2*^^^, the slowest 25% had a 
comprehension of 48^^^, and the middle 50*^^ had a comprehension 




Rapid Speed and 
Good Quality 




Rapid Spctxl and 
Medium Quality 







Rapid Speed and 
I'cxir Quality 




Medium Speed and 
Good Quality 




Medium Speed and 
Medium Quality 




Medium Spei'd and 
Poor Quality 



o 



Slow Speed and 
Good Quality 




Slow Speed and 
Medium Quality 




Slow Speed and 
Poor Quality 



Fig. 6i. — Relation hetwccn si)ocd and (|ualily oi comprehension. After 
Judd' (16, p. 155). 

of 40.5'/,;- J^i'ig interprets his results in favor of the slow readers. 
As a matter of fact they show, liowi \er, the same relation between 
Sliced and comprehension as fouml by other experiments. The dif- 
ferences in comprehension between the fast and the slow readers 
are very small, being slightly in favor of the latter group; but, when 
one remembers that ttie fast readers in the first experiment read 
twice as much text, the a<l\antages arc distinctly on the side of the 
fast readers. Whip|)le and Curtis ('17) in their study t)f skimming 
in reading also found that the slowest reader was the poorest re- 
j)roducer and the best rei)rii(li.Ker was t)ne of tlie fastest readers. 



READING 287 

(4) Relation between Oral and Silent Reading. Considerable 
attention has recently been given to tlie importance of relative 
stress upon oral as against silent reading or vice versa. The belief 
held by most of the investigators of this problem is that the schools 
have placed too much emphasis upon oral and not enough upon 
silent reading. 

Superintendent Oberholtzer ('14) of Tulsa, Oklahoma, made a 
series of tests to ascertain the relative increase from grade to grade 
in the speed of silent and oral reading. Tests were given to 1,800 
pupils. The following figures give the average speed of oral and 
silent reading in terms of words read per second : 

TABLE 88 
Speed in oral and silent reading. After Oberholtzer ('14) 

Words Read per SEco^fD 
Grade Oral Silent 



3 2.1 2 

4 2.3 2 

5 2.4 3 

6 2.8 3 

7 31 4 

S 3-9 4 



These results indicate that in the third grade the speed of oral and 
silent reading is very nearly identical, but that silent reading in- 
creases thereafter considerably faster from grade to grade, so that 
in the eighth grade the rate of silent reading is approximately one 
word per second, or about 25% faster than oral reading. 

Mead ('15) tested 112 pupils in five classes in the sixth grade in 
both oral and silent reading. He made six tests, each two minutes 
long, and determined the speed by the number of lines read and the 
comprehension by the number of "points" reproduced in writing. 
His results are as follows: 

TABLE 89 
Relative ability in silent and oral reading. After Mead ('15) 

Av. No. Av. No. Points Per Cent Reproduced 
Lines Read Reproduced of Amount Read 

Silent reading 39 . 4 16.4 38.7 

Oral reading 33-6 12. i 329 

Each of the five classes did better in silent read'.ng than in oral 
reading. Mead concludes: "From the results of these five classes 



288 



EDUCATIONAL PSVClltJLUUV 



t 2 3 



u s s 





hCQ 


a: 


^ 


'- — X '■: X X ?. r~ t^ s: -c -c 1- -J c. c: 




< 


— « ;c "^ f J — rj S3 ■* '»< -v •* -c -c 

1 " 






o 


"V-fX t-» — -f« C-*C^CC3 




■M — — — "' "" " ^ ^ oi • ->;""_ ^ 

1 ' 


C 


1- c . 
y. t- r. „ 

1-; H "^ 




X C5 <>« 








..a: 


»^ 


to o 

t-X»05»'0«3«0>.'5 0X»««C^-'-X 






■X. 

f »- r; c X c X '^» s; sr: -v -* r: 'X c r: c 




1- a 
z e. •/ 


C 
< 


'V 


S5'^ — i-r-r;'-<-fr30CC50 — t--c 




oc 


^ 


r; •- r: .-:—.-*•->-: t~ X :s » X '-j cs •c 
'x-f-*c-;->i-*-f-*r:r;«'-;i.-i.--fir5 








'>»Ot^C5»»O'>«5S0C«'.'5t^l^O«X 








y.u, 

a. ui 


^ 
t^ 


Ol^'O'VOO'.O'C-t—i-t^Ot^'Vtt'M 


u 


j~ i~ 11^ lO -f »>< C5 -< »^ r^ — o »>< ••; 1^ — 


•~. 


U X 
u - 


-v 


— — s:-*'V'vr- = r5XO-f^xco 




= — — — r:^ — ri-y^rii-ir: — r^ — -o 








6 

V. 


?« 


ex — — otsccsr^cccc — o 








r; 


-^ '1 'o "^ » o w 1- »- t^ « 00 X 00 X o 



READING 289 

we are more convinced than ever that our schools devote altogether 
too much thne to oral reading and too little to silent." 

Within the following two years, Mead repeated the same tests 
with 340 pupils in grades three to ten, excepting the ninth, and ob- 
tained corroborative results. "Fifteen out of seventeen classes 
did better by the silent method of reading. Seventy per cent of the 
children taken separately did better by this method." The de- 
tailed facts are given in Table 90. 

Pintner made eight tests, two minutes in length, with 23 pupils 
in the fourth grade and found the following results : 

TABLE 91 
Relative ability in silent and oral reading. After Pintner ('13) 

Av. No. Lines Av. No. Points 
Read Reproduced 

Silent reading 28 . 18 . 

Oral reading 20. 15 . 

Thus it appears in all of the tests that silent reading has the lead 
over oral reading in both speed and comprehension. H. A. Brown 
('16) and others believe that there should be no oral reading as 
such after the third grade, that silent reading should be empha- 
sized instead, that above the third grade teaching to read should be 
teaching to study, and that there should be a great deal of sponta- 
neous silent reading. 

Superintendent Llewelyn ('16) of Mt. Vernon, Indiana, at- 
tempted to increase among his pupils the rate of reading and to 
stimulate interest in reading. He adopted the plan of giving oral 
reading three times per week instead of five times, of supplying a 
motive for silent reading by asking questions to test the knowledge 
of silent reading, of using the two extra recitation periods for live 
questions and discussions of what had been read, of stimulating the 
reading of library books and of having frequent book reviews. Each 
book was assigned to two pupils so as to make the discussion more 
interesting and lively. 

No quantitative tests or comparisons were made, but the results 
reported were to the effect that the plan produced a love of reading, 
that teachers became more effective because they had to prepare 
for the giving of suggestions, that oral reading did not deteriorate 
and that reading was much more extensive as indicated by the fact 
that the class read about ten times as many books as before. 



2c>o KDlCAirON'AL PSVCH(>L()(;V 

(5) Phonics. The tendency in recent years has hc^n in the 
direction of less emphasis u|)on phonics and uf)on formal drills in 
general. X'arious inveslif^alors believe that extensive emphasis 
upon phonics and articulation in oral reading tends to establish 
slow habits of pronunciation and interferes with the proper develoi>- 
ment of speed in silent reading. Final experimental evidence on 
this cjuestion as on many other questions is lacking. 

Currier and I)uc|uid attem[)ted to decide by experiment whether 
it was advantageous to teach phonics or not. No definite compara- 
tive tests, however, were made, but their impressions were that the 
I)honic classes concentrated on the word and the sound at thr ex- 
I)en.se of the sense, that their reading was less smooth and slov/er 
and that their ideas were confused. On the other hand, they n-- 
IK)rted that the no-phonics classes enjoyed their reading, that they 
read more swiftly, more expressively, and more for the sense of the 
material but that they did not read quite so accurately. The 
ability to attack new words was about the same. Experiments 
of this sort ought to be carried out more extensively and compari- 
sons of the results should be made by means of more precise, quan- 
titative measures. 

(6) Comparisons of General Methods of Teaching Reading. 
Numerous nK-tiu)(is of teaching reading ha\'e Ik-iii acKocalcd by 
publishers and educators, but no one knows with certainty the com- 
parative merits of these methods nor which ones are most economic- 
ally j)roductive of the best development in reading ability. 

Superintendent Harris ('16) of Dubuque, Iowa, in conjuncticin 
with H. W. Anderson, undertook an experiment to determine the 
relative merits of three systems of teaching reading. The teachers 
had felt for some lime that they were not securing the results in 
read'ng that they might reasonably expect. The e.xperimenl is 
described thus: 

".\s a first move to remedy this unsatisfactory condition, the superin- 
tendent instituted a trial of the Beacon system at the school hereafter 
designated as School A, and of the Horace Mann system at school B. 
The i)rcsfnl year was the second in which these systems have been thus 
used. The Aldinc system wjls continued in use in the four other schools, 
(", I), ]•], F, mentioned in the reiH)rt following. The teachers who worked 
with the Beacon system were enthusiastic in its favor; but their opinions, 
no matter how enthusiast ically decl.ired, were not suiTicient to secure 
general agreement. \\'ith this stale of affairs existing, the problem re- 
solved itself into how to rai^e the cjueslion of the efficiency or worth of 



READING 291 

the various systems of primary reading out of the realm of mere opinion 
and place upon bed rock by scientific evaluation of results actually 
achieved. 

"In order to accomplish this it was decided to test: 

" I. The mechanics of oral reading in the second half of Grade I (lA) 
and in the two divisions of Grade II (2B and 2A). 

" 2. The silent reading in Grade II: for (a) rate; (b) comprehension. 

"These tests were given in the following schools: 

"A Where the Beacon system is used; 
"B " " Horace Mann system is used; 

"C D E F " " Aldine system is used. 

"It was believed that these tests would show the results obtained by 
the Beacon system and the Horace Mann system during the first two 
years of their use, and offer an opportunity for comparison with each 
other and with the Aldine system." 

Oral reading was tested with Gray's oral reading scale, and silent 
reading was tested by Starch's tests measuring the speed and com- 
prehension of reading. The results obtained are given in the fol- 
lowing tables. 

TABLE 92 
Results of the oral reading test 





lA 




IIB 


IIA 




A 
B 


36. 5 
23.6 




52.9 
45- 


55-7 
50.2 


Beacon 
Horace Mann 


C 


10. 




20.3 


42.8 




D 
E 


a 

19 
8 


b 
3 


23.8 
40-3 


43-2 
44. 


Aldine 



"These scores seem to indicate that in each grade, School A, where 
the Beacon system is used, excels all other schools in the simple mechanics 
of oral reading. In fact, the scores of Grade lA at School A are better 
than the Grade IIB scores of both the C and D, and not particularly far 
behind the Grade IIA scores of the Aldine group of schools. The figures 
also indicate that the results at School B, where the Horace IMann system 
is used, are slightly better than at the schools where the Aldine system is 
employed. 

"The results of the Oral Reading Tests seem to show conclusively that 
the pupils trained by the Beacon System are very greatly superior in the 
mechanics of oral reading to those trained under the Aldine or the Horace 
Mann systems of reading." 



292 



EDUCATIOXAL PSVCIIOLOCV 



Results of the silent reading test 
1 AHLK 93. Rate of reading 





A 


A 


n 


(* 


D 


E 


F 


Average 
Median 


2 . 2 


» 3 


' 7 
17 


.s 


g 
9 


1-7 
1-7 


'9 


Average 
Median 


2.9 


2-3 


1.8 


2. 

i.y 


17 
1.6 


1.8 
2. 


-'3 
2 .2 



IIB 
IIA 



"Table 93 shows the average and median rate of silent reading. It 
shows that in Grade IIB the Beacon pupils at School A read at the 
average rate of 2.2 words per second, while the non-Beacon group in the 
same class read at the rate of 1.3 words per second. The Aldine pupils 
read at the following rates: C, .8; D, .9; E, 1.7; F, 2. The Horace Mann 
pupils at School B read at the rate of 1.7 words per second. This shows 
that in this grade the Beacon pu[)ils read .2 of a word faster than the 
nearest competitor (School K) and that they read more than twice as 
rapidly as two of the Aldine schools. The ditTerence between the Beacon 
group of pupils and the best Aldine pupils is not significant, however. 

"In Grade II.\ the Beacon pupils at School A read at the rate of 3 
words per second, while the non-Beacon pui)ils in the same class read at 
the rate of 2.3 words per second. The table shows that the rate of rcailing 
in Grade IIA is clearly faster at School .\ (Beacon pupils) than at any 
other school; the nearest competitors being School V and the non-Beacon 
group at School A, where the pupils read at the rate of 2.3 words per 
second." 



TABLE 94. 


Comprehension of reading 








NON- 
BtACON 


Beacon 


11 M 






.\LniVE 






A 


A 


B 


C 


D 


E 


F 


Average 
Median 


26.2 
30- 


20.5 
18. 1 


18.4 
20. 


10 I 
9-4 


10. 

8.5 


20.6 
19. 


17. 

17-5 


Average 
Median 


31 5 
31 9 


17-5 


151 
19. 


30. 1 
19.9 


1 1. 8 
8. 


24 3 


1S.3 



IIB 
IL\ 



"Thus, the Beacon j)upils in School .\. Grade IIB, on the average 
repro<luced 26.2 words, while the non-Beact)n pupils in the same class 
reproduced 20.5 words. The nearest com[K"titor to the Beacon group of 
pupils is the group at School E, which reproduce*! on the average 20.6 
words— 5.6 words behiml the Beacon group. Schools C and D made the 
remarkably low grades of lo.i and 10. 

"In Grade II.\, the Beacon group of pupils reproduced 31.5 words, 
while the non-Beacon group in the same i lass reprixiuced only iq.q words. 
The nearest comi)etil<)r in I he Aldine group of s« hools was Scluxil E, 



READING 293 

where the average number of words was 24.3. The results obtained 
through the Horace Mann system seem to be below the average, this 
class reproducing only 15.1 words. 

"The results of the Silent Reading Tests seem to show conclusively 
that the pupils trained by the Beacon System are far superior to those 
trained under the Aldine or the Horace Mann systems of reading. 

"It was realized that objections might be raised to the results herein 
shown, on the ground that the teacher rather than the system was the 
strong factor in the results. While it is highly improbable that out of six 
different corps of teachers, the group teaching the Beacon System would 
be uniformly better, in each section tested; yet, as an absolute check upon 
this phase of the matter, the Silent Reading Tests were given in Grade 
HIB, which entered school before either the Beacon System or the Horace 
Mann System was placed on trial in any school and which therefore could 
not have had, in any one of the six schools, its initial training in either of 
the two systems named." (Starch's Reading Test, Series A, No. 3, was 
used.) 

TABLE 95 
Showing median rate and comprehension of silent reading in Grade IIIB 

A B C D E F 

Rate 2-9 1.8 1.9 1.9 2. i.8 

Comprehension 24. 27.5 12.3 29. 27.5 19.5 

"This table shows that the pupils at School A read more rapidly than 
those at any other school, their median rate being 2.9 words per second, 
wliQe the nearest approach to this rate was 2 words per second in School 
E. However, in comprehension three schools excel Grade IHB at 
School A. Pupils at School A reproduced 24 words correctly, while those 
at Schools B, D, and E reproduced 27.5, 29, and 27.5 words respectively. 

"Furthermore, in the tests made in Grade II, while the Beacon group 
at School A excelled all other groups in rate and comprehension of silent 
reading, several of the other schools excelled the non-Beacon group at 
School A in both these points. Thus, in the rate of silent reading, in 
Grade IIB, the non-Beacon group in School A was excelled by Schools B, 
E, and F, and in Grade IIA, the non-Beacon group at School A was 
equalled by School F; and in comprehension, the non-Beacon group at 
School A was excelled by School E in Grade IIB, and by Schools C, E, and 
F in Grade IIA. 

"These facts show rather conclusively that it was not the superiority 
of the teaching which determined the results of the tests, since teachers 
in other schools than School A, both in Grade II and in Grade III, with 
pupils trained under the old systems, secured results as good as or better 
than those secured by the teachers at School A with pupils whose first 
training also had been under the old system." 



294 EDUCATIONAL PSYCHOLOGY 

WTiether or not the results would be generally superior in other 
schools and under other teachers cannot be inferred perhaps from 
these results with complete finality. The experiment, however, 
is interesting and is here cited chiefly for the purpose of showing 
what should be dt)ne in the way of scientific comparisons and tests 
to determine the most proficient methods of learning and teaching 
reading. 

Gray iruule a comparison of the attainments in reading in 44 
schools, 26 of which had used the Aldine method, 17 the Ward 
method, and one a method of its own. The results showed no con- 
sistent or uniform superiority of one method over another. The 
average test scores were approximately the same. 

Waldo ('15) compared the Howe system with the Ward system. 
The latter had been used up to the sbcth grade in all the schools 
except one in which the former had been used. The tests were 
not carried out in a sufficiently careful manner to warrant reliable 
conclusions. 

Hendricks made a comjxirison of schools in which the Rational 
method had been used with schools in which no special methotl 
had been used. He found the former schools superior. This does 
not necessarily j^rove the superiority of the Rational method but 
probably the advantage of well-organized systems o\er less well 
organized s\'stems. The factors making for efficiency are so 
numerous and intricate that much more extensive and far more 
careful c.\])eriments will have to be made to demonstrate compar- 
ative values of the dilTerent systems or methods of teaching reading. 

(7) Suggestions for Improvement in Reading Ability. E.x- 
perinienlal results have brought about a radical shift in emphasis 
uiK)n the aims to be accomi)lished in reading. In the first place, 
there has been a shift from emphasis upon oral reading to emphasis 
upon silent reading because facility in reading, in the sense of 
thought-getting, can be developed to a much higher degree of 
proficiency in silent reading and because nearly all the reading done 
by the average adult is silent reading. In the second place, tliere 
has been a distinct shift from emi^hasis upon slow reading to em- 
[)hasis upon rapid reading because tests have .shown that rapid 
reading does not mean a corresponding loss of thought, as assumed 
by many teachers, but, instead, rajjid reading is accompanicxl on 
the whole by an almost equal ability in comprehension. The rapid 
reader will derive almost as many more ideas in a given period of 
lime as is projjortionate to the greater amount of text covered. 



READESTG 295 

Pupils were formerly told that they must not read fast but that 
they should read slowly because they would then get the thought so 
much better. In the third place, there has come along with these 
two changes a shift from the mechanics of reading to the content 
of reading. Former aims of reading are fairly represented by the 
following answers, given by pupils who had finished the grammar 
school, in response to the question, What is your idea as to what 
the reading lessons were for? Some of the answers were: "To learn 
to pronounce," "To help us in reading before people," "Just to 
pass away the time," and "I thought it was to learn us to use better 
language." (Briggs '13). 

Accordingly then the aim in reading to-day is the development 
of speed in reading and a parallel gain in thought-getting. Recog- 
nizing this change in aim, what definite suggestions can be made 
to facilitate improvement in these aspects of reading ability? 
Experimental results are as yet too few to make many specific 
recommendations with complete confidence. However, several 
important suggestions may be offered. 

(a) As to the speed of reading: Force yourself to read more 
rapidly. Continuous effort and practice in this direction will very 
materially increase the rate of reading as shown by experiments. 
Probably most adults, as well as most children, read far more 
slowly than they are capable of reading. So far as we may judge 
on the basis of experimental investigations of the reading process, 
speed of reading depends chiefly upon the rapidity of the assimila- 
tion and upon the span of attention and less upon the other steps 
in the reading process enumerated at the beginning of this chapter. 
The rapid reader assimilates more swiftly and grasps more words 
at each fixation. 

Forcing oneself to read more rapidly than one's customary rate 
will at first interfere with proper comprehension, but in the course 
of persistent practice the more rapid visual and mental activities 
will become habitual and the comprehension will probably then 
come up to its normal amount. 

(b) As to comprehension of reading: Grasp the thought with 
concentrated attention, (i) Speeding up the rate of reading tends 
also to stimulate greater concentration of attention upon the 
whole thought content. That both speed and comprehension of 
reading may be very greatly improved by practice, by reading a 
great deal with the definite aim of improvement, is shown by such 
results as have been obtained in the Dodgeville schools and else- 



296 EDUCATIONAL I'SVCHOLOGV 

where. The remarkaltle rapidity of reading was accompanied by 
an equally remarkal)le ability in thought-getting. 

(2) Sto]) frequently to recall the essential ideas read. Compre- 
hension will be greatly assisted by stopi)ing at short intcr\'als and 
asking oneself the question, What have I really read? What arc 
the essential ideas? This will not only stimulate reading for 
thought-getting but will also help to fix ideas in mind and to relate 
them to larger units. 

(3) Acquire the habit of looking for the essential ideas. This is 
\ery important in eflicient reading. Skill in so doing will greatly 
facilitate speed of reading by co\'ering more ground and by know- 
ing what may be read very hurriedly or c\-en omitted. 

(4) Tests at frequent intervals. Measurements by means of the 
standard reading tests or by means of ini])rovised tests patterned 
after one or another of these testing plans should be given at fre- 
quent inter^-als for two reasons. In the first place, they will atTord 
the pupil himself a definite basis for discovering his reading abil- 
ity and, by keeping his own record from test to test, they will 
furnish a powerful stimulus to the pupil to suqiass his o\\'n preced- 
ing attainments. This point was elaborated more fully in Chapter 
XI. In the second place, intensive tests with emphasis on both 
rate and comprehension will give the pupil practice in the phases 
of reading in which the school has in the past not furnished ade- 
quate training. The great emj^hasis upon oral reading has tended 
to instill slow habits of reading and placed the primar}- emphasis 
upon the mechanics of reading rather than upon thought-getting. 
Tests of reading ability, for this ])urposc, may be improvised and 
given as often as desirable by ha\ing the pupils in a class turn to 
a si)ecified page, read with their ma.ximum capacity for a limited 
interval, say half a minute, a minute, or several minutes, note the 
])oint of sto])ping, and then write a full account of the thought 
content, or answer (|uestions. Such a procedure nn'ght prolUaljly 
become a regular part of tlie instruction in reading. 



CHAPTER XVII 
HANDWRITING 

Processes or Steps Involved in the Act of Writing 

. An analysis of the various steps involved in writing (or copying) 
similar to that made of the reading process reveals the following 
elements: 

(i) Reception upon the retina of the form of the letters to be 
written. 

(2) Transmission of the visual impressions from the retina to 
the visual centers of the brain, 

(3) Recognition or perception of the letters through the visual 
and other association processes. 

(4) Transmission of nerve impulses from the visual centers to 
the motor centers of the fingers, hand, and arm. 

(5) Transmission of nerve impulses from the motor writing 
centers to the muscles of the fingers, hand and arm. 

(6) Execution of the muscular movements involved in the 
writing act. 

(7) Return kinaesthetic nerve impulses from these muscular 
movements back to the kinaesthetic centers in the brain and 
thence through steps (5) and (6) to help in correcting and con- 
trolling the writing movements. 

(8) Return visual impressions of the letters or marks as actually 
executed back through steps (i), (2), (3), (4), (5), and (6), in 
helping to correct and control the writing movements. 

What do we know concerning the manner of operation and the 
importance of each of these steps in the complete writing process? 
Steps (i), (2) and (3), while important when the child first learns 
to write, practically drop out in the skilled writer in whom the 
ideational and visual processes in the brain and the kinsesthetic 
sensations from the writing act itself serve directly to control steps 
(5) and (6). In the practiced writer they simply serve as a general 
control in securing the proper alinement, size, and spacing of the 
letters. The practiced writer can write about as well with his 
eyes closed as with them open. The chief difference is in such 

297 



298 EDUCATIONAL PSYCHOLOGY 

features as alinement, spacing and heaviness of strokes. The 
letters themselves can be formed practically as gracefully and as 
cjuickly with the eyes closed as with them open. Miss Downey 
(08) found, in her experiment to determine the efTect of dilTercnt 
distractions iij)on the writing habit, that the visual factor has an 
obvious function in ac(|uiring new coordinations but has little 
efTect upon the fully formed habit. The visual perception of the 
form of letters, while important, is probably not as important in 
learning to write, e\en in the beginning, as it is in learning to read, 
because the writing act dejjcnds largely upon the develoi)ment of 
muscular control. The visual perception of the form of the char- 
acters to be written and of those actually written must serve as 
a guide in the attempts at writing, but quickness in visual per- 
ception is not as important in the writing act as in the reading act 
because the writing act is much slower even in the skilled writer 
than the visual jjerception of the forms is, whereas reading depends 
directly upon the rapidity of visual perception. Then, also, the 
child has usually learned to recognize the forms of the letters 
when he begins to learn to write. At any rate its imi)ortance is 
rather secondary in the writing process. 

The chief elements in the writing process are those connected 
with the steps (5) to (8), particularly those connected with the 
steps (5) and (6). l)eveloi)ment of skill in any muscular move- 
ment which is not instinctive or at least not as mechanically pre- 
cise as an instinct, proceeds by trial and error. Attempts are made 
at carrying out the desired moN'ements. At first some of them 
suceed, most of them do not; and through continued trials the 
erroneous attempts decrease and the successful ones increase lentil 
jjerfect control is established. The adult scarcely realizes the 
utter lack of control in the early attempts on the i)art of the child 
in making the writing movements. The nearest approach that 
the adult can make toward realizing the actual dillicullies of the 
child consists in such an e.\]xTiment as the tracing of an outline 
as seen in a mirror. The spatial relations are so com])letely new 
and difterent that a ])erson has little or no conception of the direc- 
tion in which to move. Figure 62 shows a re])roduction of the 
first tracing of a star outline as seen in a mirror. It does very 
little good to reason about it. One may think he is going to move 
in a certain direction i)ut finds upon making the movement that 
he is going in an entirely dilTerent direction. The child learns to 
write very nmch after the Svinie mamur. He proceeds by tri:J and 



HANDWRITING 



299 



error. The successful movements somehow become more deeply 
fixed in the nervous connections and consequently more and more 
numerous. The correct movements finally become associated 
with the visual perception of form and direction so that the move- 
ment can be carried out at will with precision and grace. 

It is a matter of common observation that a child beginning to 
write not only makes the movements very slowly but also with 
much excess pressure. The latter point was investigated by Meu- 
mann. He had children and adults write on a little platform which 
was so supported that the amounts and changes of pressure during 
the writing were transmitted to a light lever which traced them ac- 




FiG. 62. — Record of tracing of a star-outline as seen in a mirror. 

curately on a moving drum. It is natural to find that men tend to 
exert a greater pressure than women. Gross, working in Kraepe- 
lin's laboratory, found that men average nearly twice as great a 
maximum pressure as women. These records also serve to illustrate 
the fact that each letter has a fairly characteristic pressure rhythm. 
PVeeman and others have shown by exact methods that the rate 
at which the writing point moves differs greatly in various parts of 
a given letter. In general the up and down strokes are the most 
rapid while the sharp turns and angles are the slowest. The speed 
curves of different persons writing the same letter consequently 
show striking similarities. 

Sex Differences. The striking differences between the hand- 
writing of different individuals has led many people to believe that 



300 EDUCATIUNAL rs\Cll()U)(;V 

there was a definite relation between the handwriting and the per- 
sonality of the writer. The greater pressure exerted by men as a 
class suggests that the sex of the writer determines at least certain 
characteristics of the writing. This has been investigated indc- 
])endently by Binet, Downey, and Starch who find that untrained 
subjects can determine the sex of the wTiter in from 65 to 75% of 
the specimens (50^0 being pure chance). The author niade a com- 
parison of the writing ability of 2,113 lx)ys and girls in the Madison 
schools and found the differences exhibited in Figures 63 and 64. 
It api)ears from these grai)hs that the median of the girls is above 
that of the boys in both speed and quality, but particularly in 
quality. The difference in speed is very slight. 

Miss Downey ('10) attempted to determine sex difTcrences in 
handwriting by selecting from envelopes by a chance method 100 
samples of men's writing and 100 samples of women's writing and 
by asking thirteen persons to judge whether a given sample had been 
written by a man or by a woman. The thirteen persons made re- 
spectively the following percentages of correct judgments: 60, 60, 
61, 64, 66, 66, 68, 6S.5, 70, 70.5, 71.4, 71.5, 77.5. These show 
that on the average a judgment of sex as revealed in handwriting 
is correct 67% of the times, or two out of three times. Miss Downey 
also reports that the writing of the women showed less variability 
and more conventionality than that of the men. Those samples of 
writing by women which were called masculine were generally from 
persons accustomed to doing a great deal of writing. 

Correlation 0/ handwriting with other traits. A good deal of mis- 
leading character interpretation is based uyum various features of 
handwriting. Experimental work needs to be done in this field, but 
it is probable that there is nothing to the claim of gniphologists 
that handwriting reveals such traits as energy, clearness and sim- 
plicity, vanity, or self-consciousness. 

Gesell ('06) reports that there is a close correlation between 
quality of writing and intellectual ability, but as a matter of fact 
his results show, as pointed out by Thorndike, a correlation of only 
about .30. 

The author ('15) found for children a correlation of .31 be- 
tween writing ability and grneral scholarship. Thorndike ('10) 
reports that for adults thr correlation betwern writing and schol- 
arship is zero. Tin- probability is tliat the small cornlation exist- 
ing in the case of children is due to the fact of receiving instruc- 
tion in writing and to the attention given to it. The better puj)ils 



HANDWRITING 30I 

do somewhat better in writing because they probably pay more 
attention or make more careful efforts. So far as adults are con- 
cerned, poor handwriting is no indication either of high or low 
intelligence, since the correlation is approximately zero. 

Professional graphologists have claimed that a great number of 
specific traits of writing are determined by corresponding traits 
of character on the part of the writer. Thus the hnes of writing 
of ambitious persons are supposed to slope upward from left to 
right. Hull ('19) investigated some of the more persistent claims by 
computing correlations between the traits of character of 1 7 univer- 
sity fraternity men, as judged by their fellows, and exact measures 
of samples of their writing. The correlations in each case were 
approximately zero, showing these claims to be entirely unfounded. 

The Measurement of ErnciENCY m Writing 

(a) Essential elements to be measured. The two important aspects 
of writing that must be measured are speed and quahty, including 
under the latter legibility and form or beauty. 

(b) Methods of measurement. Speed of writing is now generally 
measured in terms of the number of letters written per minute. 
Quality may be measured by either one of several scales, the Thorn- 
dike Scale ('10), the Ayres Scale ('12), the Starch Scale ('19), and 
others. The Thorndike Scale consists of a series of 18 steps or 
qualities of handwriting, each step consisting of one or more spec- 
imens of writing of the appropriate quality. Step zero represents 
an attempt at writing but as such is entirely illegible and devoid 
of beauty. Step 18 is a perfect copper plate specimen. The steps 
from o to 18 represent equal units of increase in quality. The 
Ayres Scale consists of 8 steps designated as 20, 30, up to 90. 
Each step contains three specimens of equal quality, a vertical, 
a medium, and a slant sample. The recent revision of the Ayres 
Scale, the Gettysburg edition, contains only a medium slant 
specimen for each step. The successive steps represent uniform 
increments of legibiUty in writing. The Starch Scale is composed 
of a series of 20 steps arranged in the order of merit or excellence 
of writing as judged by 400 persons. (See the original monographs 
for detailed description of the preparation of these scales.) 

A sample of handwriting is measured by any one of the scales by 
putting it alongside the scale and determining which step it is most 
like in general quality. Speed and quality should ordinarily be 



302 



EDUC'ATIONAI. PSYCIIOIJ )( ; V 



measurt'd sinuillancously in llic same sample l)ecausc these two 
aspects of writing have a functional relationship. A test of writing 
should, therefore, he made by having the i)upils write a short, simple 
sentence repeatedly as many times as they can in, s;iy, two minutes, 
doing it as well as they can. Speed is then measured by the number 
of letters wTitten per minute and quality is rated by one of the scales. 
Freeman ('14) has i)repared a set of five analytical scales for the 
purpose of rating handwriting from (he sland[)oint of uniformity of 
slant, uniformity of aiinemcnl, quality of line, letter formation, and 
spacing. Each scale contains samples of three successive degrees 




1 '-J 3 4 6 6 7 8 

Grades 

Fig. 63. — Se,\ difference in speed of writing;, .\flcr Starch t'lj). 



of merit. These methods of rating ought to be useful in calling 
attention to defects in particular features of writing. 

The A. X. Palmer Company has j)ublished a series of five or si.x 
samples of successive degrees of value for each grade. A percent- 
age value for posture, movement, speed, and formation is given for 
each sample. The value of the sample as a whole is expressed by the 
average of these four estimates. 

The Zaner and Hloscr Com[)any has also issued a set of specimens 
for evaluating haii<lwriting consisting of a series for grades one and 
two, another series for grades three and four, and a third series for 
grammar and high school classes. Each .scries has a number of 
.siimples whose rating is expressed in terms of percentage values. 
Comments concerning the defects or excellencies are ajipended to 
the \ari«)us specimens. 



HANDWRITING 



303 



The manner in which the values of the samples were determined 
is not indicated for either the Palmer or the Zaner scales. This im- 
pairs their scientific value. From the practical standpoint they are 
commendable in that they suggest specific attention to, and evalua- 
tion of, important elements in handwriting. 

(c) Results and uses of measurements. In general the results and 
uses of measurements in handwriting are the same as those pointed 
out for reading. It is possible by means of these measurements 
to fietermine more precisely the actual writing ability of a pupil, 
class, or school and to compare it with standard averages for cor- 




FiG. 64. — Sex differences in quality of writing. After Starch ('13, p. 461). 

responding grades in schools generally. These standards of at- 
tainment for the ends of the respective school years are as follows: 

TABLE 96. After Starch ('16, p. 83) 
Standards of attainment in writing 



Grades 
Speed (letters per minute) . 
Quality (Thorndike scale) . . 

Quality (Ayres scale) 

Quality (Starch scale) 



6.5 



2 


3 i 


5 


6 


7 


8 


31 


38 47 


57 


65 


75 


83 


7-5 


8.2 8.7 


9-3 


9.8 


10.4 


10.9 


27 


2>3 37 


43 


47 


Si 


57 


9 


9.7 10.3 


10.9 


II. 4 


12.0 


12.5 



By reference to these standards of attainment it is possible to 
define quite accurately the speed and quality of writing of a pupil 
or class by saying, for example, that a given pupil in the fifth grade 
is able to write 65 letters per minute at quahty 9, Thorndike scale. 
The value of exact measurements of handwriting, as of any educa- 
tional products, consists in the diagnosis of ability as it actually 
exists in different pupils and schools, in the measurement of the in- 



304 EDUCATIONAL rSYCHOLOCY 

lluence of tiilTtrcnt factors and conditions upon learning to write, 
and in the determination of the mutual relationships of various 
aspects of writing. A sur\'ey of our present knowledge concerning 
these matters will be given in a later section. 

Economic Procedure in Learning to Write 

What influence do the various factors, conditions, and methods 
in learning to write have in promoting or hindering the develop- 
ment of skill in handwriting? This question could he answered 
finally and fully only by tlie careful isolation of each factor under 
experimental conditions and by determining its effect upon the 
progress of learning to write. Substantial beginnings have been 
made in the direction of answering some of these cjuestions, but little 
or nothing is known about most of them. 

(i) Perception of the Forms Written or to Be Written. This 
topic requires consideration of two general questions: (a) WTiat 
are the most advantageous conditions for the visual perception of 
the forms to be written? (b) What sort of form or model should be 
presented? The former question may be answered by the observa- 
tion of certain obvious rules, namely, that the writing surface should 
be placed before the eyes at the proper distance, especially not too 
near, so as to avoid eye strain, and in a position directly in front of 
the eyes so that the points on the i)aper to be successively fixated 
may be at e(|ual distances from both eyes, thus avoiding unecjual 
accommodation in the two eyes. The paper shoukl not be glazed, 
so that it will not produce a glare, and for young children the sur- 
face should be rather rough so that it will easily take pencil marks. 

The second cjuestion is more complicated. The sort of models 
to be presented is obviously highly important since imitation, 
both voluntary and involuntary, jirobaljly jilays a large part 
in the accjuisition of writing skill. The author ('ii) made an 
experiment in which he attempted to measure the unconscious 
effect of different models of writing upon the normal writing of 
adults. Four samples of writing were obtained from each of io6 
university students. In order to avoid any suggestion of imita- 
tion, written rather than oral directions were given stating that 
they were to prtxluce samples of their writing and that tliey should 
proceed at once to write the pa.ssages j^resented without further 
thought or questions. The four passages put before each person 
con.sisted of (i) a tN-pewritten selection, (2) an extreme vertical 



HANDWRITING 305 

model, (3) an extreme slanting model, and (4) a large model with 
many flourishes. The purpose of the typewritten passage was to 
obtain at the outset a sample of the normal writing of each person. 
The other three models were taken from school copy books. 

After the experiment was finished, each person was asked whether 
he had tried purposely to imitate the various models. Three 
persons stated that they had intentionally modified their styles 
of writing. Their records were thrown out. The samples pro- 
duced by the remaining 103 persons were carefully measured to 
ascertain their slant and size. Slant was measured by means of 
a specially prepared, transparent device with ruled lines for de- 
termining the angle of inclination of certain tall letters, such as 1, 
f, and p, with the base line on which the words were written. Size 
was measured by determining the horizontal width of letters by 
measuring the length of words and dividing by the number of 
letters in the word. 

These measurements showed that the average tendency for this 
group of persons was to make the letters distinctly more vertical 
when the vertical model was before them and more slanting when 
the slanting model was before them as compared with their nor- 
mal styles of writing. They also tended to write slightly larger 
when the large model was before them. The amounts of these 
changes were as follows: 

TABLE 97 

Average inclination of 1 in the normal writing 65 . i degrees 

Average inclination of 1 written from vertical copy 68.8 

Average inclination of 1 written from slanting copy 61.5 

Change from normal to vertical 3.7 

Change from normal to slant 3.6 

Total range of change 7-3 

Average width of letters in normal writing 4-33 nini- 

Average width of letters written from large model 4 . 85 mm. 

When we realize that the handwriting of adults is a pretty 
firmly fixed habit, the amount of unconscious imitation is consid- 
erable, being a total of 7.3 degrees in slant and of .52 millemeters 
in width. We may infer that with children whose writing habit 
is in process of formation, the element of unconscious imitation 
plays a much larger part. Furthermore, it seems quite probable, 
although no experimental proof is at hand, that the style and 
quality of writing of the teacher distinctly influences the writing 
of the pupils, especially so because the writing done by the teacher 



3o6 EDUCATIONAL I'iiVCllOLOCJV 

in the presence of the pupils for the purpose of showing them how 
to write, is likely to he more efficacious in securing imitation than 
a static model in a coi)y-book would be. It would seem, therefore, 
highly imperative that every elementary school teacher should be 
a reasonably good writer. 

In connection with the survey of penmanship in the Grand 
Rapids, Michigan, schools, Freeman (Judd, 'lO) reports that 

"Grand Rapids adopted about five years ago a new system of pen- 
manship. Up to that time the writing was not regarded as satisfactory. 
A part of the dilliculty was thought to be due to the inabihty of the 
teachers themselves to write well enough to furnish a good example to 
the pupils. Accordingly, by action of the Board of Education, all teachers 
in the elementary schools were required, as a condition of {jromotion, to se- 
cure a Palmer certificate. This rule has been recently enforced with strict- 
ness and the writing in the schools is reported to be greatly improved." 

Should the model presented to the pupils be vertical or slanting? 
Should it be plain or contain flourishes, decorative curves and 
shading? Should it be angular or rounding? The answers to 
these questions are at present largely matters of opinion and con- 
venience rather than matters of scientific determination. Some 
years ago vertical writing came into general u.se because it was 
thought to be more legible and less productive of spinal curvature. 
But it has largely disappeared for the obvious reason that almost 
everyone naturally falls into the habit of writing a medium slant, 
no matter what style of writing was taught to him previously. 
The average slant for adults, as shown in Table 97 is about 65 
degrees with the base line or about 25 degrees with the vertical 
line. Whether slant writing actually lends somewhat more to 
produce spinal cur\'ature is doubtful. The dilTerence in legibility 
between vertical writing and a medium slant writing is also prob- 
ably very small. The letters should probably be of a medium 
slant and should be relatively plain and free from flourishes since 
these take lime and add nothing to the general value of the writ- 
ing, and finally, they should probably be moderately rounding 
because extreme roundedness is likely to reduce speed and e.x- 
Ireme angularity is likely to reduce legibility. 

Graves ('17) classified 604 .samples of handwriting according to 
slant and then studied the speed and quality len<kiicics of the 
vertical, medium slant, and extreme slant group. The final aver- 
ages of speed and (|uali(y are shown in the following table: 



1 



HANDWRITING 



307 



TABLE 98. After Graves 



Words Written Qdality 
IN 5 Minutes (Ayres scale) 

Vertical 91 . 6 57-98 

Median slant • 96 . i 48 . 22 

Extreme slant lor .7 43-58 

There is revealed clearly a positive connection between slant and 
speed on the one hand and poor quality on the other. That is the 
"extreme slant" writers write more rapidly and more poorly than 
the "vertical" writers. 

(2) Length of Period of Practice. What is the most productive 
practice period in learning to write? Even such a question as 
this, which is capable of definite experimental solution, has been 
answered only in part. The answer given by school programs in 
the time allotted therein for writing, is based largely on opinions 
instead of facts. 

TABLE 99 

Quality of handwriting at roughly the same rate in seven school systems. After 

Thorndike ('10) 

Median results for eighth-grade pupils 



System 
At 20-29 words in 4 min. 
At 30-39 words . 
At 40-49 words . 
At 50-59 words. 
At 60-69 words. 
At 70-79 words. 



.11.5. 

-II-5- 
.11.5. 
.10.3. 
. 10. o. 



B 


C 


D 


E 


F 






•-I4-5- 


..13.0. 


-•IS- 


II. 3. 


..11.6. 


--I2-3- 


..12.3. 


. .14. 


12.0. 


. .12.0. 


..II. 8. 


..12.3. 


..14- 


11.6. 


. . II . I . 


. . II . I . 


..11.6. 


. . 13 . 


U.S. 


..11.5. 


..11.3. 


..11.6. 


..13- 


10.8. 


--II-3 









G 
4. ..14.8 
5. ..14.2 

4---I5-3 
o. . . 11.7 
6 



Median results for seventh-grade pupils 
System A B C D E F 

At 10-19 words 13 . 3 ... 14 . 5 ... 13 . 5 

At 20-29 words 12.3. . .13.3 13.0. . .13.6. . .14.2. . .13.0 

At 30-39 words II. o... II. 8 12. 3. ..13. 3. ..14. 2. ..13. 

At 40-49 words II. o... II. 8. ..II. 3. ..II. 7.. .11.0. ..13. 3. ..118 

At 50-59 words . 
At 60-69 words . 
At 70-79 words. 

Systems A and B devote no time to writing as such in grades 7 and 8. 

System C devotes 50-60 minutes weekly 

" D " 73-100 " " 

Systems E and G devote 60-90 " " 

System F devotes 75 " " " grade ". 

« p « ^Q << « » « g^ 











•13-3- 


.14. 


12.3 


••I3-3 




.13.0. 


.13-6- 


.14. 


II. 


..11.8 




.12.3. 


■^3-5- 


.14. 


II. 


..11.8 


..11.3. 


.11.7. 


.11.0. 


•13- 


IO-3 


..11.4 


. .11 . 1 . 


. II. 0. 


.11.8. 


-13- 


10. 


..11.3 


..10.5. 


. 10. 0; 


.11.4. 


.11. 


9-8. 


.. 9.8. 


.. 9.9. 









3o8 EDUCAllONAL I'SVCHOLOGY 

Thorndike ('lo) compared the writing in seven school systems 
as given in Table 99, and concluded that time was i)ractically 
negligible. He says: 

"What these facts do prove is: First, that at least three systems (C, 
D, and E) get little or no better results at a time cost of about 75 minutes 
a week than two systems (A and B) do xit zero time-cost; second, that 
one system (F) at no greater time-cost than C, D, and E gets results 
about 25% better than they do; and third, that practice for quality may 
secure it only at the cost of speed. The teachers in A and IJ arc better 
paid than those in the other cities, so that the success of these schools at 
no time-cost might not be generally attainable. 

"Leaving F out of account, the ditTerences of these school systems in 
the method of teaching handwriting, in the time devoted to it, and in 
the ideals of the system in respect to it are of inconsiderable intlucnce 
upon efliciency. One makes its pupils write very well at very slow rates, 
the others vary a little in quality with small inverse variations in speed. 
On the whole, in spite of the achievement of system F, efliciency in hand- 
writing seems, like spelling, and unlike arithmetic to be under present 
conditions not very much influenced by the management of the schools." 
(Thorndike, "Handwriting," p. ^],.) 

Freeman ('15) had writing tests made in 47 cities and then 
compared the attainment in these schools with the amount of time 
devoted to the writing-period in each school. His results are set 
forth in Figure 65. Each school is represented in the chart by a 
short vertical line. This line is placed at a position above the base 
line so that it rej)resenls the relative rank of that school in attain- 
ment in penmanship among the 47 schools. 

These results are interesting and valuable, but it is questionable 
whether they prove that lime makes no dilTerence. The difliculty 
with a wholesale set of figures such as these is the impossibility 
of separating the various elements and determining their ef- 
fects individually u[)on the ultimate attainment in writing. The 
schools which devoted 90 to 100 minutes per week to writing and 
obtained no better results than the schools which devoted 40 to 50 
minutes per week, may contain other factors which kept their 
proficiency down, such as, poorer teaching, dilTerent chusses of 
pupils, the quality of writing done in other subjects outside of tlu- 
writing period, whidi probably has ;is much if not more influence 
u|)()n i)roficiency in writing than tin- writing jieriod itself, and so 
on. In fact, we might even imagine that if these same schools had 
flevoted only 40 tu 50 minutes jht week, they might have been 



HANDWRITING 



309 



much worse in writing than they actually were. The real expla- 
nation may perhaps lie in the possibihty that the schools having 
longer writing periods may not use the time to as good advantage 
as those having shorter periods. The latter, by virtue of having 
only a short time to devote to the subject, may work more in- 
tensely and profitably. 

The surest way in which to measure the results obtained in dif- 
ferent periods of practice in writing would be to split up a given 
group of pupils into several sections and to have each section devote 
a different amount of time to the writing period, say 10, 15, 20, and 
25 minutes respectively. All should preferably be taught by the 



40-49 



.« 50-59 



,eo-69 



J 70-79 
§80-89 



90-99 



-1 1 r 



J_L 



J LU I I L 



II III I I LU I I II I 

J 1 1 I I I I I I 



XI L 



_LJ I LL 



10 15 



18 



30 



25.4, 



22.6 2 



20.7 



28 



20 



25 30 
Rank 



35 40 



45 



Fig. 65. — Relation between attainment in writing and time devoted to writ- 
ing. After Freeman ('15). 

same teacher. At any rate, all other elements should be kept as 
constant as possible. Then a comparison by special tests in speed 
and quality made at stated intervals would reveal the effect of 
time upon improvement and would show what period of time 
brought the optimum results. 

The investigations by Thorndike and Freeman have been highly 
valuable in calling attention to this problem and in showing that 
some schools obtain as good results by devoting only half as much 
time as other schools obtain in double the amount of time. The 
general impression is that 15 minutes per day as a maximum is suf- 
ficiently long for the writing period and, under proper methods of 
instruction, can produce as high attainment in writing as the schools 
need to produce for all practical purposes. That there is an opti- 



3IO EDUCATIONAL PSYCHOLOGY 

mum length of the ^\Titing period beyond which the principle of 
diminishing returns operates is Cjuite certain, as indicated by general 
learning ex])eriments such as those cited in Chapter XI. What this 
optimum period for j^ractice in \\Titing is cannot at present be 
S])ecifR(l with ct-rtainty. 

(3) How Great Proficiency Should Be Attained. Tests made 
in a large number of schools show that the average attainment at 
the end of the Sth grade is writing as good as quality 11, Thomdike 
Scale, or quality 60, Ayres Scale, at a speed of about S3 letters ])ct 
minute. The same tests also show that many schools reach much 
higher proficiency than this and that in every school a considerable 
share of pupils far exceed the limits of 11 or 60 in quality and 83 
in speed. Are these averages of attainment in quality and speed 
suflSciently high for practical ])urj)Oses? .^nd is it worth wliile to 
develoj) higher proficiency in writing than these averages rei>- 
resent? 

In answer to the first question, Freeman ('15) made inquiries 
among business firms and found that the majority considered writ- 
ing equal to quality 60, Ayrcs Scale, as sufticiently good for or- 
dinary- business jiurposes. It would seem then that the frequent 
criticism from business men who siiy that pujiils coming to them 
from the public schools cannot write, is ill-founded, and based 
probably on the exceptions rather than on the majority of jiupils, 
since about three-fourths of the ])upils finishing the elementary 
schools can write better than quality 40 or 50, A>Tes Scale, and 
one-half can write better than quality 60. The attainment of ir 
or 60 in quality and of S3 letttrs ])(.r minute in speed reached by the 
axerage pupil upon com])ktii)ii of the elementary school is fully uj) 
to the average requirement of business. The criticism coming from 
business men is probably based upon the 20 or 25% of pupils 
finishing the Sth grade who fall below c|uality 40, and many times 
upon those who have school before completing the Sth grade to 
seek business em])loyment. 

The second question is practically answered by the discussion of 
the first. It probably is not worth while to attempt to reach a 
proficiency in writing much higher than quality 11 to 12, Thomdike 
Scale, or 60 to 70, Ayres Scale. Such higher skill would be gotten 
by too great an expense of time and by too great a s;icrifice of 
speed. Furthermore, the legibility of writing of qualities above 
these limits increases very little. The gain is chiefly in beauty. 
The linif that would be recjuired (o ri-ach these higher degrees of 



HANDWRITING 311 

skill could be devoted to better advantage to other subjects, or to 
the learning of typewriting. Thorndike says: 

" Considering the fact that above quahty 1 1 there is very Httie differ- 
ence in legibiUty, one is tempted to advocate the heresy that children 
are taught to write too well. I personally do advocate it. If school 
boards would furnish, for the use of children electing 'writing' as a study 
in the last two grammar grades, typewriting machines, I should certainly 
advise the transfer to typewriting of a child in these grades whose writing 
at 60 letters a minute consistently reaches quality 13. For, the amount 
of practice required to advance such a pupU to quality 16 at a rate of 
75 letters a minute would much more than suffice to advance him to 
substantially errorless machine writing at that rate. The value now at- 
tached to the high qualities of handwriting is of course largely fictitious. 
Employers who can afford such high qualities of writing, buy machines 
to produce them. For writing cash checks, simple book entries, labels, 
and the hke, a good plain hand or our quality 12 is entirely adequate. 
For attaining the higher qualities (15-18) the machine is a more eco- 
nomical tool than the pen, and in my opinion should be provided by 
those schools which require such qualities. Further, such qualities should, 
in my opinion, be required of children in the elementary schools, only 
when they have elected writing as a vocational subject. For the data 
from the adult women-teachers make it practically certain that ability 
to write above quahty 14 will not be exercised in life except as a part 
of a clerical trade. If very, very few teachers find it worth while to 
maintain qualities above 14, it can hardly be supposed that it will be 
worth while for mechanics, house-keepers, farmers and dressmakers to 
do so." ('10, p. 37.) 

(4) Relation between Speed and Quality. To what extent 
is speed of writing accompanied by good quality? Is there possibly 
an inverse relationship between the two? From general impressions 
we know that if we try to write unusually well we sacrifice speed 
and if we try to write unusually rapidly, we sacrifice quality. Is 
there any balance between these two elements? 

Quite frequently teachers overemphasize either quality or speed, 
usually the former, at the expense of the latter. In Figure 66 
the teachers in the 6th and 7th grades greatly over-emphasized 
quality so that the speed of writing was equal only to that of the 
average 3rd grade pupil. Definite tests and comparisons with 
standards will reveal to the teachers many such aberrations in 
emphasis. 

Sackett had 36 university students write in their normal manner 
and immediately afterwards he had them write the same material 



31 



KDUCATKJNAL rSVCIKJLOGV 



with the knowledge that it was to be used as a writing test. He 
found that on the average the writing was about .5 letters per- 
second slower (original rale about 1.8 letters per second) while 
the qualit}- gained 4 points on the Ayres scale. 




13 

a 
u 
M 11 

C 

^ « 



1 




1 






y 


._— ^ — -^ 


~^^ 


i 








^ 








/ 


.____ 


y^^^ 










>- 


^ 








"''^ 


**'^^*^ 


7 
















Qualit 


J 





















<y 



1 2 3 4 5 ti r 8 

Grades 

Fig. 66. — .AvcraKC speed anrl quality of liaiidwritiiif,' of tin- variou.s graHcs 
in a Riven s< 1iim)I. 'I"1u- broken lines represent the school. The continuous lines 
arc the standard attainments. 

Freeman ('14) made tests to determine what the efficiency would 
be when children were told to write (i) both rapidly and well, 
(2) as well as possible, and [t,) as rapidly as possible. The results 
showed that trying to write well improved quality at the exj^nse 
of speed. Quality iinprovid 6.2% while speed dropped 3. 7*^7- 
Trying to write rapidly ini reascd speed by 27. 2*;^ but decrea.sed 
quality 9.1%. IniproNcnient in both speed and <iuality, however, 



HANDWRITINCx 



313 



can be obtained when instructions are given to stress both aspects. 
Apparently this is the preferable thing to do. 

The author ('15) found in the case of 144 pupils the following 
correlations between various characteristics of handwriting: 

Speed and quality 10 

Speed and legibility li 

Quality and legibility 34 

Freeman also computed the correlation between speed and 
quality on the basis of the writing samples of pupils in Grades 4 to 
8. These he found to be as follows: 



Grade 
Correlation. . 



IV 
.08 (02) 



- . 10 (04) - 



VI 

• 14 (04) 



VII VIII 

•37 -.15(5) 



These correlations are either zero or slightly positive or negative 
and mean that only to a very slight extent is the good writer ex- 
tremely slow or the fast writer extremely poor. 

Judd ('16), in the Cleveland Survey ^ has presented extensive 
data pertaining to this question: 

"After determining the speed and quality of each specimen, it becomes 
possible to work out with great exactness the relation between these two 
characteristics. It is evident from ordinary experience that quahty 
commonly deteriorates when speed is emphasized, and that speed is slow 




40 50 60 70 80 90 

Quality— Ayers Scale 

Fig. 67. — Average speed of handwriting at each quality of writing from 20 
to 90; 10,528 cases from 5th, 6th, 7th and 8th grades. After Judd ('16, p. 72). 

when one tries to write especially well. The school is constantly in the 
position of seeking some reasonable balance between speed and quality. 
"Diagram 67 gives the facts for the 10,528 specimens carefully studied. 



314 EDUCATION.\L PSYCHOLOGY 

In the vertical axis of this diagram are represented the different speeds; 
in the horizontal axis are the various grades of quality. The results from 
each grade are rei)rcscnted separately. Thus, beginning at the extreme 
right end of the bottom line, we see from the diagram that for those 
writers in the fifth grade who show the highest quality (c)o) the rate is 
on the average 51 lelttrs per minute. Advancing along the line toward 
the left, we hnd that those in the fifth grade who show a quality 80 have 
an average speed of 54 letters. 

"The diagram shows that there is a general area between qualities 
60 and 80, and between speeds 60 and 80, where all the grades above the 
fifth may be said to reach a level. Greater speed seems to be purchased 
at an undue sacrifice of quality, and higher quality seems to result in 
much slower speeds. We thus have in our results some indications as 
to the probable area within which teachers will find a desirable biilance 
between speed and cjuality." (Pp. 70-71.) 

(5) Methods of Teaching Penmanship. Experimental efforts 
have thus far not been directed very vigorously toward ascertain- 
ing the specific effect of dilTerent methods of learning to write 
upon rate of improvement in it. It seems, however, very certain 
that different procedures do produce enormously different results. 
This is amply shown by the wide differences in attainment of the 
various grades and schools, even in the same school system. Judd 
found, for example, in the Cleveland Surocy, that the average of 
the best class was twice as proficient, either in speed or in quality, 
as the poorest class. The facts are shown in Figures 68 and 69: 

"Diagram 68 shows the average results for the four upper grades in 
36 schools. The figure is to be interpreted as follows: In the upper 
di.igram, which gives the results for the fifth grades, there are numerous 
small squares, each representing a single fifth grade. In each square is a 
number showing the average number of letters written per minute in a 
grade. Thus in the square at the extreme left of the diagram is the num- 
ber '3Q.' 'I'his means that the average number of letters written per 
minute by that fifth grade was 39. In the next vertical column of squares 
are numbers ranging from 42 to 49. These indicate that there were fifth 
grades showing each of the averages given. 

"One of the most impressive facts which is brought out by this com- 
parison is that the slowest fifth grade is only half as fast as its fastest 
fifth grade. Like statements can be made regarding the other grades. 
These wide diderences cannot be attributed to any native characteristics 
which the children bring to the school. Such di.s[uritics might appear in 
individuals, but the figures report whole grades. .Ml the fifth grades are 
going througli the schools parallel with one another and are oflicially 
ranked as alike. The same statement can be made regarding the other 



HANDWRI'I'ING 



315 



.Speed Records of 
36 Fifth Grades 



Speed Records of 
36 Sixth Grades 



Speed Records of 
36 Seventh Grades 



Speed Records of 
36 Eighth Grades 



56 



64 



61 



79 



76 



47 



55 



55 



52 



50 



60 



60 



60 



75 



70 



87 





78 
77 
77 
77 








76 


89 






76 


89 






75 


86 






75 


86 






73 


85 








73 


85 








68 


73 


84 


97 




58 


64 


72 


83 


96 




57 


62 


71 


82 


91 


Us 


50 


61 


70 


81 


90 



m 



58 



64 



64 



64 



74 



70 



70 



82 



90 



ioil 



Fig. 68. — Distribution of grade averages in speed of writing. After Judd 
('16, p. 64). 



3i6 



EDUCATIONAL PSVCHULUGV 



grades also. Perhaps the most obvious case is that of the eighth grade. 
ChiUlrcn will go out of the various eighth grades into high schools with 
the official assumption that they are equally well fitted for advanced 



Quality Records of 
36 Fifth Grades 





■M 






44 






44 






44 






44 






44 








43 


48 






43 


48 








43 


47 






39 


42 


46 






88 


41 


46 








36 


41 


46 


M 


57 






88 


40 


46 


57 


57 


62 


1^ 


35 


40 


45 


50 


55 


60 



Quality Records of 
36 Sixth Grades 





49 
49 
48 
49 








48 


54 






48 


54 








47 


54 








44 


47 


53 






39 


43 


46 


53 






38 


43 


46 


51 


58 


61 




38 


43 


46 


51 


66 


61 


[^ 


36 


41 


46 


SO 


56 


60 



Quality Records of 
36 Seventh Grades 



30 



48 



46 



46 



46 



bO 



60 



60 



66 



00 



60 



m 



Quality Records of 
36 Eighth Grades 



Fio. f>Q. — Distribution of grade averages in quality of writing as measured 
by the Ayres Scale. After Judd (*i6, p. 64). 





49 




69 
68 






49 


64 


67 






48 


63 


66 






49 


68 


66 






48 


68 


66 






47 


62 


66 


64 


08 




47 


62 


66 


62 


09 




46 


61 


66 


61 


07 


|:b| 


45 


60 


66 


61 


66 



HANDWRITING 317 

work, and yet one eighth grade averages only 46 letters a minute, and 
another averages loi. Is it not evident that there must be a difference 
in emphasis on speed in writing in difTerent schools?" (Pp. 63-64.) 

"Diagram 69 shows in a manner similar to that explained in the 
earher paragraph on speed the results obtained from 36 schools. From 
the figure it will be seen that in quality, as in speed, the most striking 
variation exists between grades which are officially recognized as parallel. 
Furthermore, there is the same overlapping of grades, several of the 
fifth grades ranking higher than the average eighth grade." (P. 68.) 

Here, as in so many other problems, specific experiments with 
conditions rigorously controlled should be carried out in order 
to determine the actual effect of each given element or method in 
teaching handwriting. Such experiments could be carried out 
by teaching parallel sections of a class according to different 
methods, after the general plan outlined under heading (2) of 
this section. 

(6) Factors Affecting the Execution of the Writing Movements. 
The numerous conditions affecting favorably or unfavorably the 
execution of the many complex writing movements such as the 
position of the body, the position of the desk, the position of the 
arm, the position of the paper on the desk, the manner of holding 
the pen or pencil, and the like, are important problems concern- 
ing which likewise we have little scientific information. The pro- 
cedures followed by teachers are based chiefly upon general ob- 
servation and personal judgment. With regard to position of 
body, arm, paper, and desk. Freeman has suggested the relations 
shown in Figure 70. This relationship makes possible a natural 
straight front position before the desk with both arms on the desk, 
and with the paper tilted at an angle of about 30 degrees to the 
left. This position of the paper makes it possible for the hand to 
follow easily along the horizontal line of writing by simply turning 
the forearm on the point on which it rests on the edge of the desk 
as a pivot. Furthermore, the most natural direction of the up and 
down movement of the pen point is directly toward or away from 
the body, and with the paper in the position suggested, the writ- 
ing will have a medium slant of about 25 to 30 degrees from the 
vertical. 

(7) Types of Writing Movements. What sort of writing move- 
ments may be executed most economically in learning to write 
and ultimately in the perfected writing process? Considerable 
controversy has occurred over this question. Theoretically there 



3l8 EDUCATIONAL PSYCHOIX>GY 

arc at least three types of movements possible in the production of 
letters. One would be to do the writing entirely with finger and 
hand movements and to hold the arm absolutely quiet except 
for the turning of the arm from left to right to follow along the 
line of writing. The second would be to do the writing entirely 
with the arm movement and to hold the hand and fingers absolutely 
still. The third would be a combination of these two sets of move- 
ments in varying [)roportions. Advocates of various methods of 
teaching writing favor one or another type of movement. Probably 




Fic. 70. — Position of [)Uj)il in relation to desk and jiaper. .\fter Freeman 
('14). 

the best method is an appropriate combination as suggested in 
the third tyi)e of movement. Freeman states: 

"The arm movement with rest — the so-called muscular movement — 
is an American discovery an<l has l)ccn viRorously exploited in commer- 
cial schools since the last quarter of the last century and more recently 
in certain systems of teaching in the public schools. It seems likely 
that within twenty-five years this form of writing will be practically 
universal in American schools. The chief advantapcs of the movement 
arc two. In the first place, it is made with the fingers relatively rela.xcd, 
thus avoiding crampinR. In the second plat e, the rolling movement of 
the arm u\^n\ the nuis< le pad of the forearm pHxhices a firmness and 
evennessof line, and the fact that the movement is produced from a center 
at a considerable distance from the pen j><)inl results in rcfjularily of slant. 

"The contention that every detail of the letters shall be made by the 



HANDWRITING 319 

movement of the arm while the fingers remain immobile is calculated 
to antagonize reasonable critics. The oscillation of the arm may well 
form the main basis for the upward and downward strokes of letters, but 
to require that every loop and turn and joining be produced by the 
movement of the arm as a whole, instead of the much more flexible hand 
and fingers, is to set up an artificial requirement and one which is not 
made in regard to other types of skilled movement. 

"The form of movement, then, which best meets the requirements 
which may be laid down as the result of experiment and of practical 
experience is somewhat as follows: The hand and arm must be so ad- 
justed that the hand progresses freely along the line during the formation 
of the letters and in the spaces between the words. The hand must rest 
upon some freely sliding point or points of contact such as the finger nails 
or the side of the Httle finger. When, on the contrary, the pen point is 
carried along from one letter to another by means of adjustments of the 
parts of the fingers and the hand, the hand continually gets into a cramped 
position. 

"The movements of the arm and fingers should form a smooth and 
easy co-ordination in which there is a condition of flexibihty in the whole 
member. The rotation of the arm upon the muscle pad of the forearm 
as a center carries the hand along, the upward and downward oscillatory 
movement forms the groundwork of the letter formation, and slight ad- 
justments of the fingers complete the details of the letters. In addition 
to these chief elements of the movement, the wrist may rotate to the side 
to supplement the sideward movement of the arm, and the forearm may 
revolve upon its axis in the movement of pronation as a corrective to the 
increase in slant at the end of the line. There is no good reason for seek- 
ing to eliminate any of these component movements. Each has some 
part to play. Moreover, room must be left for individual differences in 
their relative prominence and manner of combination." ('14, pp. 93-96.) 

(8) Movement Drills. Special drills in movements such as 
ovals, vertical movements progressing to the right, horizontal 
movements from left to right and from right to left, have been 
advocated by various systems of penmanship with the belief that a 
substantial amount of time given to such drills will establish good 
form and speed in writing. To what extent such formal drill or 
how much of it may actually be profitable, is open to question. 
It would be an experiment worth undertaking to teach three 
sections of a class of pupils for a year or more by giving to one 
section a considerable amount of such drill, to the second section 
none, but to devote instead the entire time to drill and practice in 
writing the letters themselves, and to the third section a com- 
bination of the two types of drill. 



320 EDUCATIONAL PSYCHOI/X^Y 

(9) Correct Form in All Writing Done by the Pupils. It is an 
elementary principle of habit formation that an act to be developed 
into a skillful habit, whenever carried out, should be performed 
correctly or at least as correctly as jx)ssible at that stage of learning. 
Othersvise the inaccurate and cureless performance of the act 
tends to counteract the skill already achieved. It would seem, 
therefore, to be a highly desirable plan as an incentive to pupils 
to write at their best at all times, to base their marks in penman- 
ship to the e.xtent of one-half upon their work in the writing-period 
proper and to the extent of the other half upon the quality of 
writing in all other work submitted. One important reason why 
instruction in penmanship, spelling, oral and written composition 
does not carry over into the penmanship, spelling and composition 
generally is that jiujjils are not as careful in their ordinary writing, 
spelling and sjK-aking by observing correct form as they are in 
the respective class periods devoted to these subjects. Telling 
the pupils that their final grades will be made up, half and half, 
as here suggested, will act as a remarkable incentive toward general 
improvement as shown by specific tests in the case of spelling, 
which will be discussed in the next chai)ter. 

(10) Analysis of Imperfections. One of the important by- 
products of the experimental investigation of conditions and fac- 
tors in the learning process is the fact that definite practice in a 
specific function consciously known to the learner greatly imj)roves 
the function. Improvement in any type of skill takes place in 
many instances only when practice is squarely directed towards 
certain specific elements in the process. This is one reason why 
persons in laboratory experiments on learning make such enormous 
progress and why pupils in school make so little i)rogress. The 
function to be trained in the one case is definitely and .sjiecifically 
known to the learner, whereas in the latter case, it is indefinite and 
largely unknown to the learner. It is not enough to say to a i>ui)il, 
"Vou must write better," "write more like the copy," or "watch 
me; write as I do." The specific defects must be pointed out, 
recognized by the learner and then overcome by definite practice. 

Freeman has jjointed out five main tyi)es of defects or character- 
istics of handwriting: uniformity of slant, uniformity of alinement, 
quality of line, letter formation, and spacing. The scales that he 
has devised for rating handwriting from these five points of view 
may be used with advantage in discovering the specific defects in a 
given individual's writing and in centering definite attention and 



HANDWRITING 3 21 

practice upon them. The score card for evaluating handwriting 
prepared by C. T. Gray calls attention to a similar set of elements. 
The methods of judging penmanship suggested by Palmer and by 
Zaner aim likewise to center attention upon defects and excellencies 
in various essential aspects of writing. 



CHAPTER XVII r 
SPELLING 

Processes or Steps Involved in Spelling 

The child Iciinis to syicll by seeing or hearing the letters of a given 
word, and by thinking, s])eaking, or ^vriting them in the order in 
which they are seen or heard. Stated in more minute detail, the 
successive steps arc substantially as follows: 

(i) The reading of the word, that is, the sight, sound and pro- 
nunciation of the word as a whole which involves all the elements 
of the reading process and need not be enumerated here. (See 
Chapter XVL) These are presupposed as the child usually has 
learned to read the word before he learns to spell it. At this point 
the successive steps in the spelling process as such begin. 

(2) Reception upon the retina (or the car) of the visual (or 
auditory) stiniuli of the lirst letter of the word. 

(3) Transmission of the visual (or auditory) impressions from the 
retina (or car) to the visual (or auditor\') centers of the brain. 

(4) Arousal thereby of mental images and other associations 
of meaning. 

(5) Transmission of the impulses from the visual (or auditor^') 
centers to the motor-speech centers or to the motor-writing centers. 

(6) Transmission of motor impulses from the latter to the 
s])eech-organs or to the writing-muscles. This occurs very ]>rob- 
a])ly even in the silent learning of spelling since silent reading is 
accompanied by the so-called inner sjieech. 

(7) E.xecution of the sj)eaking or writing movements in pro- 
nouncing or writing the letters. 

(8) Return kiniesthetic impulses frt)m the speech or writing mus- 
cles to the sensory centers and then to the motor speech or writing 
centers. This series of steps from (2) to (8) is then repeated for 
the second letter, for the third letter, and so on to the end of the 
word. 

The steps here outlined are the ones involved in learning the 
spelling of a word. In the iierfeeted jiroccss, however, stejjs (2) 
and (3) and possibly (4) drop out and step (5) is inaugurated 



SPELLING 



323 



directly either through step (i) or through the idea or image of 
the word to be spelled or written, and from then on the whole 
process of writing or spelling the word consists of a circular series 
of automatic connections between steps (6), (7), and (8) in which 
(8) for the first letter of a word acts as stimulus to step (6) for the 
second letter and so on for the succeeding letters of a given word. 
Step (8) of each letter always acts in turn as the stimulus for the 
series (6), (7), and (8) of the succeeding letter so that in the fin- 
ished habit the mere pronunciation, sight, image or idea of the 
word automatically brings about the succeeding links involved 
in naming or wTiting the letters in correct order. 

Economy in learning to spell consists largely in providing con- 
ditions under which the half dozen links here outlined may be es- 
tablished most easily, most quickly, and most permanently for 
the words whose spelling a child should know. 

Little is known directly concerning the manner of operation of 
each of these factors. The most important step, if any one is more 
important than any other, possibly is number (8). This link 
determines what the next letter shall be in the automatic writing 
of a word. In the original learning of the spelling of a word, steps 
(2), (3), and (4), which together constitute the perception of the 
letters, are highly important since the establishment of the other 
links depends upon the accuracy with which the letters themselves 
are perceived or observed. It seems probable, although not certain 
in the absence of pertinent experimental data, that a considerable 
part of the difiiculty of learning to spell, lies in the inaccurate 
observation of successive letters of a word. The awakening of 
mental images is probably very important, although our informa- 
tion as to the types of imagery concerned, the extent to which 
they are essential, and the methods of arousing them, is relatively 
unreliable. 

The Measurement of Efficiency in Spelling 

(i) Methods of Measurement. On the face of it, it would 
seem to be an easy matter to devise a definite and objective method 
of testing attainment in spelling. All that would seem to be neces- 
sary would be the selection of a series of words and the determina- 
tion of the number or percentage of these words that a pupil or 
class can spell correctly. But a closer study of the possibiHty of 
measuring spelling ability reveals a number of complicated prob- 



324 EDUCATIONAL PSVCH()L()( iV 

knis. What sort of words should be used as a spelling test? How 
should they he presented lo the pupils? How should they he 
scored? Should any wurd be considered equal to any other word, 
or should different \alues be assigned to difTerent words? 

We shall not enter here into any critical discussion of the prin- 
ciples invoked in the construction of si)elling tests, nor into a 
consideration of the technic^ue of administering and scoring them. 
Some of the methods of measuring spelling ability will be men- 
tioned briefly.' 

Up to 191,^, tests of si)elling a])ility were made either by series of 
arbitrarily selected words which were presented either as isolated 
words or as ])arts of dictated sentences, or by determining the 
percentage of misspelled words in spontaneously written composi- 
tions. Since 1913, several more or less scientific methods of 
measuring spelling ability have been devised. 

The writer, ('15) jjrepared one test consisting of 6 lists of 
100 words each by making a selection of words at certain inter- 
vals from the dictionary and then discarding all technical and 
obsolete words. The words in each list were then arranged in the 
order of length. Kach list as a whole was found to be practically 
identical in difficulty with every other list. A\erage standards of 
attainment were then jjrepared for the \'arious grades as sho\\Ti 
in the following table, which gives the ])ercentage of words of any 
one i)f the 6 lists spelled correctly at the ends of the respective 
vears: 



TABLE 99 








I in s|)cIlinK. 


After Starch ('15) 






2 .( 


1 5 (. 


7 


8 


30 40 


SI 61 71 


/S 


85 



Grades I 

Percentage of words correct 10 

The author has more rc-rently prei)are(l a dilTereTit method of 
testing s])eiling abilit\' on the basis of the 2,0j6 most common words 
in the English language. 'J'his list is useful as a study list as well 
as a test list and will be distinctly more valuable beaiuse of this 
double pur])ose. The plan by which these words were selected 
and the method by which they are to be u.sed will be described 
later in this chapter. 

A}Tes prepared a list of words consisting of 10 words for each 

' For .1 (K-tailcil (liNtu-vsion. sti* llic (iriKinal iniiii«)Kr.n()hs or the writer's h'AucaiioHOl 
Mitisurcnu-nli, or Muiiroi-'s JUlutalionjl Tnls and Mcasurrnu-nli. 



SPELLING 325 

grade so selected on the basis of experiments that on the average 
70% of the words for any given grade would be spelled cor- 
rectly by the pupils of that grade. Later, Ayres ('15) prepared 
a very useful test consisting of the 1,000 most common words. 
These words are split up into 26 lists of varying length and so 
arranged that the words in any one list are of approximately equal 
difficulty and that the successive lists from i to 26 become harder 
and harder. The scale gives the average percentage of the words in 
the various lists that pupils can spell correctly in the different 
grades. 

Buckingham ('13) prepared, on the basis of experiments, a list 
of 50 words carefully scaled in difficulty according to the percentage 
of pupils who could spell the words correctly. 

(2) Uses and Results. All these scales have been found useful 
in measuring efficiency in spelling in different schools more ac- 
curately than was formerly possible, in determining individual 
differences in abilities among pupils, in ascertaining progress, and 
in comparing the effects of various factors in the learning and 
teaching of spelling. It is hoped that they will be still more useful 
in the future in discovering the most effective methods for ac- 
quiring proficiency in spelling. 

The facts with regard to the enormous range of individual abili- 
ties in spelling and the consequent overlapping of the abilities of 
the pupils in the various grades are shown in Figure 18, Chapter 
III. The facts, as in other subjects, are astounding. The best 
pupil in the first grade spells as well as the poorest pupil in the 
eighth grade. Certainly some adjustment of the pupils should be 
made according to their capacities. 

Spelling presents one of the more striking examples of mental 
sex differences found in educational psychology. Investigators 
uniformly report girls doing better than boys. Wallin found the 
girls averaging nearly 2% better than the boys in terms of his 
spelling lists. In a study previously quoted in Chapter XIV, Foster 
found that 238 girls and 256 boys, all university freshmen who 
were given a spelling test of 40 difficult words of Latin derivation, 
made respectively 76.6% and 68% successes. Sackett gave 24 
words of Buckingham's spelling list to over 7,000 school children 
and found the girls about a half year of school progress ahead of 
the boys. Sears reports that a test composed of 70 words from 
the Ayres list given to nearly 13,000 children in Oakland, Cali- 
fornia, showed the girls superior to the boys from 2 to 6%. espe- 



326 



KDUfAIlD.NAJ- rSVCIlUUXiV 



cially in the ujiper grades. He suggests that the average girl 
might be graduated from a half to a whole year earHer than the 
average boy. 

The writer found the (HlTerences in one school as shown in Fig- 
ure 71. The difference is in favor of the girls, particularly in the 
upper grades. 

90 

80 

70 

•S50 

to 
I 40 

^80 



w 



























.--^'T^^ 










^ 












J 


^ 










&i»^ 


:/ 










^^ 


"■'Boyr' 










y 














/ 





























4 5 

Grades 



Fig. 71. — Sex differences in spelling as measured by ihe authors test. 



Economic Methods ix the Learning .\ni> Teaching of Spelling 

(i) The Words to be Learned. I'or a number of years the most 
important problem in the economy of learning to spell has been 
the question, What words should a child really learn to spell? The 
words in the spelling-books have for decades been selected mainly 
by the arm-chair method and have consisted to a large extent of 
rare and useless words. Swinton's Spdkr, published in 1S72, states 
in the preface that "It omits the alphabet and the 'abab's' on the 
one hand, and on the other, quite a number of sesciuipedalian 
words common to all old-time sjielling books." It further urges 
as a vantage j)oint, "The j)ractical character of the work which 
aims to set forth, not the lens of thousands of long-tailed words in 
osily and atiou, but the actual vocabulary used in speaking and 
writing." And yet the book contains in one lesson (page 14,5) 
such words as, lelhean, ])harisee, pentagon, pneumatics, theoc- 
racy, anathema, dysentery, etc. In another lesson (page 144) it 
contains oleaginous, farinaceous, argillaceous, lachrymose, sacer- 
dotal, animad\irsion. 



SPELLING 



327 



In view of this situation, recent years have brought forth a 
number of very extensive studies and tabulations to find the words 
which are most commonly used in writing by various classes of 
persons and to discover the frequency with which these words 
occur. The following are the chief tabulations thus made: 

The Eldridge List. Mr. Eldridge ('11), a business man in Buf- 
falo, New York, made a tabulation of 43,989 running words from 
four different newspapers in which he found 6,002 different words. 

2,927 words occurred each, once 



1,079 




' twice 


SI6 




' three times 


294 




' four times 


212 




' five times 


151 




' six times 


lOS 




' seven times 


84 




' eight times 


86 




' nine times 


261 




' ten to nineteen times 


238 




' twenty or more times 



The Ayres List. Ayres ('13) of the Russell Sage Foundation 
tabulated 23,629 words from 2,000 letters, chiefly business letters, 
and found 2,001 different words. 

The Jones List. Professor Jones ('13) of the University of South 
Dakota tabulated 15 million running words from 75,000 themes 
written by 1,050 pupils in grades two to eight and found 4,532 
different words. 

The Cook and O'Shea List. Cook ('14) tabulated 200,000 
running words from the family correspondence of thirteen persons 
and found 5,200 different words. 

These four tabulations represent four distinct iields of writing, 
each being the most extensive in its field, namely, journalistic, 
business, juvenile and private domestic vocabulary. One im- 
portant type of vocabulary has never been tabulated, namely, the 
vocabulary of our best current literary writers. Children ought 
not to be confined to the words which they naturally use (Jones 
List), nor to adult business vocabulary (Ayres List), nor to news- 
paper vocabulary (Eldridge List), nor to the vocabulary of ordi- 
nary family correspondence (Cook List). An important point in 
learning to spell is to learn the meaning of words, especially of 
words whose use will enhance a person's vocabulary. Hence, the 



328 EDUCA'IIOXAL PSVCIIOLO(iV 

writer made a tabulation (Starch List) of the vocabulary of the 
best current literary authors. This taljulation is unpublished and 
on flic in llu- Library of the University of Wisconsin. 

The Starch List. The writer ^ tabulated some 40,000 running 
words, about 1,000 from each of forty authors in eleven current high 
grade magazines. This yielded 5,903 different words as follows: 

3,1 1 1 words occurred each, once 

1,009 " ' " twice 

512 " " *' three times 

280 " " " four times 

189 " " " five limes 

121 " " " six times 

97 " " " seven times 

82 " " " eight times 

53 " " " nine times 

225 " " " ten to nineteen times 

224 " " twenty or more times 

From these five lists, words for spelling and testing purposes 
were selected according to the following plan: Every word occur- 
ring three or more times in the Starch List, every wt)rd occurring 
three or more times in the Eldridge List, every word occurring 
seven or more times in the Cook List, and every word in the Ayres 
1,000 word Hst was selected if it also occurred in one other list 
including the Jones List. A word occurring three or more times in 
the Starch List or in the Eldridge List or seven or more times in 
the Cook List or any word occurring in the Ayres List was not 
included if it occurred only in the one list. To be included it had 
to occur at least once in one other list. This safeguarded against 
the inclusion of words confined to one type of vocabulary only. 
P'or example, the word "cujjfuls" occurred twenty-one times in 
the Starch List but in no other list. Hence, it was excluded. 

The reason for selecting words that occurred three or more 
times in the Eldridge List or in the Starch List was that the words 
found less freciuently are so rare that they constitute a very small 
part of the running words of ordinary writing. This point may be 
shown most emphatically by the accompanying graph, Eigure 72, 
on which the relali\t' number of words of different frequencies is 
indicated. A remarkably close parallel exists between the Eldridge 
and the Starch Li.^ts. The particular point to note in the graphs 

' In «<K')|xT;ifion with I.. C. I)c Bruin. Krimrtcil in a thesis in itic lil)r.'iry of the Uni- 
viTsiiy of \ViM:unMii, lyiO. 



SPELLING 



329 



is the fact that the sharp bend in both curves occurs between 
words whose frequency is between two and three. After three, 
the cui-ve shoots up very rapidly. This same breaking point occurs 
in the Cook List between seven and eight. It is higher in this Hst 
because Cook tabulated a larger amount of writing. Words 
occurring three or more times in the Starch and Eldridge Lists 
constitute over nine-tenths of all running words. 

This process of selection yielded 2,626 words. This number may 

seem small compared with the number of words in former spelling 

3300 

3000 

2700 

2400 

•f 2100- 

^ 1800 
o 

S; 1500 

XI 

s 

^ 1200 
900 
600 
300 

























































































1 






















1 






















/ 






















/ 




















J 










Eldridge 
Starch Li 


List — 






/ 






















.^ 


'i 

















^^ 


'^ 









20 



1 



10 9 8 7 6 5 4 3 
Number of Times Words Occurred 
Fig. 72. — -Number of words occurring with various frequencies in the El- 
dridge and Starch lists. 

books and even in some contemporary spellers. But it is obvious 
that it is not only useless but wasteful of a pupil's time to learn 
words which he will never use in writing in all his years after school 
and at the same time neglect to master thoroughly the words 
which he will actually use. Spelling texts commonly contain from 
ten to fifteen thousand words. In fact, the number of words here 
presented, however, is even larger than that found in the special 
spelling lists prepared by many cities which often do not have 
more than from 1,500 to 2,000 words. 

It seems, however, that a spelling list ought to include all words 
of reasonably common occurrence but not words of extremely 



330 KDLCAllU.NAL PSVCHOLOGV 

uncommon occurrence. This number seems to be approximately 
2,500 or 2,600. Another reason for Umiting the spelling list to tliis 
number is tlie fact that the writing vocabulary of the average 
adult is not over :;,5oo words and probably less for a great many 
people. Cook found that three out of his thirteen persons wlio had 
each furnished 40,000 running words (the largest number obtained 
from any one individual) of family correspondence had used a 
vocabulary of 2,575, i)546, and 2,330 dilTerent words respectively. 
This would mean that a person could write 100 letters each 400 
words in length , making a total of 40,000 running words, and not 
use more than 2.500 dilTerent words. If we may assume that the 
average man or woman, exclusive of persons whose occupation 
involves considerable writing, such as novelists, teachers, and 
journalists, writes one such letter a month, his entire correspondence 
for ten years would not involve more than 2,500 dilTerent words. 

Incidentally we niay also point out that this number furnishes 
enough words to supply two new words for each school day in 
grades two to eight, or 360 a year, counting iSo days to the school 
year, and 106 words for grade one. This is in accord with the 
practice of many school systems of teaching not more than two 
new words per day. 

It is possible that such a list as that described above may need 
to be supi)lemeiited to meet the needs of certain sections of the 
population. Houser tabulated the words used by farmers in 
California in corresjionding with the state department of agricul- 
ture. He found certain radical differences between this list thus 
obtained and that ])ublished by Ayres. 

The Placement of the Words into the Various Grades. Can this 
be done on any scientific basis? Which of these 2,626 words 
should a jiupil leani in each grade? There are three possible wa}-^ 
in which the words might be distributed into various grades: 

(i) We might distribute the words according to their frecjuencies, 
jiutting the most fre(|uent words in the lower grades and the less 
frequent words in the upper grades. 

(2) We might jiut each word into the grade in which ihildren 
first begin to use it rather fre(|uenlly in their wTiting. 

(3) We might ])Ut each word into the grade in which, according 
to the consensus of competent judges, such as teachers, it ought 
to be taught. 

The i)hicement of the words in the present list was made partly 
according to all three principles, but where discrepancies existed 



SPELLING 331 

with reference to any given word final decision was made according 
to the third plan. 

The 2,626 words were first arranged according to their frequencies 
of occurrence and all those words in the list which were also in the 
Jones List were then placed into the grades in which the Jones 
List shows them to be first used by children. A considerable num- 
ber of the 2,626 words were not found in the Jones List. These 
words were placed into higher and higher grades according as they 
occurred less and less frequently. 

After this task was completed the entire list of words was re- 
checked according to the third plan. Fifteen different lists of 
words which had been prepared by various cities or school systems 
for their own uses such as the Boston Minimum List, the Stockton 
List, the Santa Cruz List, the Chicago Speller, etc., were used. 
After each of the 2,626 words occurring in one or more of the lists, 
was written the number of the grade into which the word was 
placed by each list. An average of these placements was then 
obtained and accordingly the word was finally placed into its 
grade. For example, the word "flower" was placed by seven 
lists into different grades as follows: 4, 4, 2, 3, 2, 2, 4. This gives 
an average grade placement of 3.00. " Cough " was assigned by six 
lists to grades 5, 5, 3, 3, 2, and 4, with an average grade placement 
of 3.66. All of the 2,626 words were thus assigned with the excep- 
tion of 126 words which did not occur in any of the lists employed 
and 178 words which were found in only one list. In order to make 
the grade placement of these 304 words with equal confidence, a 
group of seven experienced elementary teachers or supervisors 
were asked to assign each of these words to some one grade accord- 
ing to their best judgment. An average of these judgments was 
then obtained and the words were placed accordingly. 

All of the 2,626 words were then assigned to the various grades 
according to the average grade placement as follows: 

All Words Whose 
Average Grade Placements were 



In Grade One F 

" " Two 

" " Three 

" " Four 

" " Five 

" " Six 

" " Seven 

" " Eight 



om 1 . 00 to 2 . 00 

2.01 " 2.75 

2.76 " 3.28 

3.29 " 3.66 

3.67 " 4.71 

4.72 " 5.66 

5.67 " 6.70 

6.71 " 8.00 



332 EDUCATIONAL rSYCHOLOGY 

This process gave 360 words in each fjrade or two for each school 
clay with the exception of grade one into which the remaining 106 
easiest words fell. 

Measuring the AttainmoU of i/ie Pupils by Means of These Words. 
The problem that was next attacked was this: How many of the 
words of a given grade may we reasonably expect the pupils at 
the end of the year to be able to spell? In pursuit of an answer 
to this question, six lists of 60 words were selected from each 
grade list by taking, for a given list, every sixth word through 
the entire 3O0 words of a given grade. The words of the first grade 
were split up into two lists of 53 words each. These lists were then 
given as a special test at the end of the school year to apjiroximately 
7,000 pupils in 28 schools in 15 cities ranging in size from very small 
towns to a city as large as Seattle. The percentage of words of each 
grade spelled correctly by the pui)ils of that grade was as follows: 

C,K\j^r. 12 3 4 5 6 7 8 

Av. percentage of words 

spelled correctly 56.2 63.6 77.0 So. 4 S3. 9 S5.0 S2.9 So. 7 

Thus a very imj)ortant advantage of these spelling lists ^ is that a 
school or teacher can at any time test the pupils and determine 
their efiiciency by comparing them with the above standard a\er- 
ages. This can be done by selecting a list of 60 words from the 
360 words for a given grade and giving them as a test. For ex- 
ami)le, if at any time the teacher of the fourth grade desires to 
compare the achievement of her pupils with the standard averages 
of other fourth grades in schools generally, all she needs to do is to 
turn to the words for the fourth grade and begin with any one of 
the first six words and then ])ick out every sixth word through 
the list. This will give a total of 60 words. At another time a 
different list of 60 words may be chosen in like manner. The 
same procedure may be followed in any other grade by using 
the words for that respective grade. 

The im])ortant advantage of this ]ilan is the fact that the s;imc 
words which have been used as a study list may be used at any 
moment as a test list and comi)arisons may thus be made with 
the standard averages. 

The averages here jircsented indirati- that the words for all 
grades above the second are of a])])roxiinately equal (lilTiculty for 
the resj)ective grades since tluy all are within a few jKiints of 80%. 

' I'ulili.ihttI in tlic authur's Sitcllin^ Book, iv>V> '>>' Silver, Burdctt &: Co. 



SPELLING 



333 



That the words in the second grade are not too difficult is shown 
by the fact that in several schools the percentages in this grade 
averaged as high as 90. The chief reason for the lower percentage 
in the first and second grades is the variation in amount of actual 
instruction in speUing given in different schools in these grades. 
The best school among the 22 averaged around 90% in every grade 
and the poorest school averaged around 60% in its various grades. 

Aside from this list, numerous special lists have been prepared 
partly on the basis of vocabulary studies or partly on the basis of 
words commonly misspelled by pupils as reported by teachers. 
Illustrations of such lists are the Boston Minimum Spelling List 
(1915) consisting of 762 words, the Nicholson List, consisting of 
3,070 words, prepared for the State of California, and the Chico 
(California) List, consisting of 3,470 words, prepared by Studley 
and Ware. 

(2) The Influence of Rules in Spelling. Cook ('12) made a 
test with 50 words on 70 university freshmen and on 39 high school 
seniors and 30 high school freshmen. These 50 words were ex- 
amples of seven rules with their exceptions. The university fresh- 
men had had drill on spelling rules about seven months before 
the test and the high school classes had finished the study of rules 
six weeks before the test. The results are shown in the following 
table: 

TABLE 100 

Observance of the rules. After Cook 





Conscious of 

Rule While 

Writing 


Unconscious of 

Rule Whilf- 

Writing 


Combination 

of All CrriNG a 

Rule 


Unable to 

Cite Any 

Rule 


RULE 


High 
School 


Univer- 
sity 


High 
School 


Univer- 
sity 


High 
School 


Univer- 
sity 


High 

School 


Univer- 
sity 




No. 
Stu- 
dents 


Av. 

% 


No. 
Stu- 
dents 


Av. 

% 


No. 
Stu- 
dents 


Av. 

% 


No. 
Stu- 
dents 


Ay. 
/o 


No. 
Stu- 
dents 


Av. 

% 


No. 
Stu- 
dents 


Av. 

% 


No. 
Stu- 
dents 


Av. 

% 


No. 
Stu- 
dents 


Av. 

% 


ie-ei 

Final e . . 
Final y . 
Final con 
Final ie. 


16 
31 
11 

15 


79 
81 
74 
78 


25 
20 
18 
32 


87 
87 
94 
88 


15 
21 
18 

27 


71 
78 
67 

72 


5 
9 
13 
2 


87 
94 
95 
87 


31 
52 
29 
42 

5 


75 
80 
70 
74 
80 


30 
29 
31 
34 
18 


87 
89 
94 
88 
95 


38 
17 
40 
27 
64 


73 
82 
73 
75 
61 


40 
41 
39 
36 
52 


86 
88 
91 
84 
69 



"In summary, it may be said that no one rule was quoted by as many 
as 50% of the university students, though more than half of them had 
memorized all these rules, and others besides, only the winter before; and 
many of the students had been over all the rules in the public school. A 
Uttle less than half the high-school students had the courage to try to 



:;34 EDUCAriONAL PSYCHOLOGY 

give the rules they hud learned only six weeks previously. In the univer- 
sity group, those who gave some sort of rule to cover any part of the list 
of words, averaged 4% higher in general spelling efhciency than those who 
could not give any rule. So it is fair to assume that their better observ- 
ance of the rules as shown by Table 100 is the result of their better spelling 
ability in general, and not to any conscious application of the rules as 
such. Not a single rule tested proved to be of real value, except the one 
for the last two words of the list — that relatiiag to final ie." (.Viler 
Cook.) 

These conclusions are interesting and probably correct in their 
chief implications. One further point, however, ought to be con- 
sidered. Inability to cite a rule or unawareness of its ai)plication 
does not necessarily prove the impotency of the rule. It might be 
possible that a rule played a part in the learning of Avords at the 
time the learning took place and then had been forgotten. A 
further investigation is necessary in which a comparison would be 
made between two groups, one of which had learned and ajiplied 
rules while the other one had never had any contact with rules. 

(3) Length of Class Period. Dr. J, M. Rice (97) tested the 
spelling ability of about 33,000 pupils to ascertain the effects of 
different factors ujjon efl'iciency in the subject, such as methods of 
teaching, foreign i)arentage, home environment, amount of time 
devoted to spelling in the school program, and the like. His re- 
sults with reference to the factor of time are presented in the 
table. 

The results as they stand would seem to indicate that length of 
class period makes no dilTerence in the ultimate achievement in 
spelling since schools devoting 10 or 15 minutes daily do as well 
as those devoting 50 minutes daily. Thus, City 15, School B, 
grade IVA, de\oting only 15 minutes daily to spelling, made a 
record of 70.8 in the sentence test, whereas Cit)' i, School B, devot- 
ing 50 minutes daily to s])elling, made a record of only O1.8. Many 
other similar comparisons may be cited. 



SPELLING 



335 





700H3S 


<:pq U< M O < pq CJ < 


pq <: 


AJJ3 


MM Mt^t^t^ONOOO 


O M 

W M 


ISaX 30N3INI3S 


M ■ o • "N c^ • o 1^ ■ cn 


\0 • *N 


< 
w 

00 


AlIVQ 


• -OOOio-Oio 


o • • • o 

CM ■ • ■ rj- 


isax 

NOUISOdWOJ 


• -O •^':^^0■* "N 


M . . . . 


• 1^ C^ • 1^ O^ ^ ^ 0^0\' 


isax 
aONaxNag 


MOCO-INt^- • ■ lO 


lO ■ ■ • •* 


■00>Oior|--t^O rO<^ • • -O 

; ooooco -OOOO 0000 • • -00 


as 
< 
a 
>| 


AlIVQ 

saxnNij\[ 


■ o-OO -OO'OO O -O -O ■ 


• • • o 

r-l . . . ro 


xsax 

N0IXIS0dK03 


■ ■ -l^-Tj-. • IN . .-^<N • -(N 


.o • • ■ • 


• • -00 ■ o ■ ■ tn o ■ -OO- • OOO • • • ■ 
• 0\ • O^ ■ -roON* ■0^0\- ■0^0^■ 


xsax 
aoNaxNag 


t^- • M '"^^M • --o • .wo-r^ 


LO . . . 


O • -cOro-r^M . ■ in ■ -O -^O 

t^ . .'t^ I^ . f^oO • • oo • -00 • GO CO 


0\ . ■ ■ vO 


w 


AlIVQ 

sax.iNij^ 




• • o ■ -o 


xsax 

N0IXIS0dK03 


• • 00 !^i 00 ^ • • «3 M . . 1^ rt ■ ■ . 


. oi . . . 


•oooot^oo ■ -i^o- .OOOO ■ • • 


.<>••• 
.<>..• 


xsax 

aoNaxNag 


IN . t^ lO "N Tt M ■ • LOCO tS M M tJ- . 


• t^ • Tj-cc 


r^-t^M^COMO • --OClro LOO Ov • 


. (N . ro LO 


< 

w 

H 


Aiiva 

saxaNipM 


lo-OioOloOO lO- ■ ■ in ■ ■ 


■ ■ o • • 

• . (N • • 


xsax 
Noixiso<mo3 


• •'^TfcOrO- -VOO ■ -OO- • • 


■ KO o ■ ■ 


• • t^oo t^oo • • t^oo ■ ■ t^ r^ ■ • • 
' O^ 0\ 0\ O • -OvCN- .c^o^■ 


• OO 00 • 
■ O ON • 


isax 

aDNaxNas 


■O • ^ CO O -t M-.ro 0000 <N- 


• O ^ -^vO 


C^•<^^M^oo^»OM . -moo rj-o O rO • 
i-^ • i>.00 t^ J^ t^co • • t^ i^ r^ t-^ t^oo • 


• O t^ o o 

• t-^ t^ t^ t^ 


< 

w 

>< 

w 

H 


AlIVQ 

saxaNiH 


• -OioOO -O lo-o ■ ■ -O • 

• -lO^rOTt-O) .ro--1-- • ••+• 


. . LO • LO 

. ■ CO • CI 


isax 

N0IXIS0cIK03 


■ -OOr^OviO OiO^- -MOO 


• -O CO • • 


'0\0\0^<:X^O^t^O^C^ • •0^0^ ■ 


• O 00 • • 


xsax 
33NaxNas 


mD ■ 00 >0 LOGO • • cyD mD Tl- ^ ^ • 


• ^ 0\ "N VO 


r^ ■MMr^oO'+Lo • -OOmOCOOO • 
O • ^O t^ J^vO \0 i^ • • t^'O o o o o 


• M LO M CO 

<< pq < m < 


lOOHOg 


<eqU<:MU-<pqu<: 


pq <: 




AXI3 


MM Mi>. i>. r^ c^ O^ O O 

M 


O M 

M M 



336 



EDUCATIONAL i'S^'CHOI.Or.\' 





ioon.)s 


53 •< 


a » a 


w ^ -^. ^ ■< 


- 






.U.IJ 


r, 


■-1 »0 1/-, 


■ o i/-- O o o. 


o 














'AV lOOIUv; 


I- 


a> o "t 00 


CO O- O i^ 


00 CO 




1^ 


-O On fO 00 

I^ l~ t^ 1^ 


t>. t>. t^ I^ 


CO CO 




ai 

>< 
E 

00 


ATJva 


• o • 

• -T ■ 


. . r, . . p 


• O ■ lO ■ O • lO • 






XS3X 
NOIXISOdJIOJ 


• »-• 


■ • • -r • • 


• M • l-l •Jt 0> <^ 








. . . o, o> . . 
. . . c> o • • 


■ 0> ■ • O- aoo OS • 

• ^i • • ^v ^ ^ ^t 








• o o 

• 00 aa 


<N coo o> <N M o 

6 O "t CO t^vO CO 

O O- O>00 00 CO CO 


• ■* Cs •* o, <-< c, ,o 

• OsOsOs'to^- o -r 

•OOOOCOCOOCCOCOCC 


CO 

■ >r:S 

■ CO 00 




a: 

>• 
a 

H 


■VIIV(I 

sax.iNii^ 


■ o ■ 


. . M . . . O 

• M ■ . . r» 


• O • uo ■ O ■ O • 


• ■ o 

. . CO 




iS3X 
N0IXIS0dH03 


• 00 ■ 


• ro r>» • 


• (^ • . CN oo lO CO • 




° 


■ 00 • 

■ o • 


■ . • 00 oo • • 

. . . c^ OS . . 


• CO • ■ CO CO 00 o> • 

• o> • • o o- o\ o ■ 




~z 


XSMX 
aO.M3XN3S 


Jcg :' 


CO ■ »^ ro t^ 


• O 1- • 1 00 CO CO 


LO 

t <> t 

00 O CO 


C 


00 ■ o o CO -r -r 

1- ■ O 1-V3 l-O) 


. O I . n c< « looo 
■ CO CO • I-. r^ !>. r^ 1^ 




H 


A1!V{I 

sax.iNijv 


• o ■ 


. • o- • • • o 


■ O • lO ■ O • lo • 


lo ■ O 

CO ■ CO 


< 


NOixiso<mo3 






to w . -o «>• "I t^ • 






lO 


i^ "^ 


f 


rY\ 


r-rt 


00 00 • ■ t^OO 00 »>. • 

OS o> ; • OS o> c- OS • 








r>^» 


•^ 




t/> V 






-t ■ <o 


i^ w •<f iri ir> r^ 


00 O 00 <N Os 1^ • - oo 


<o lO 




lo coco t^ 11 00 "H 

1- 1^ J^ lO t^ O 1^ 


c< Os OsO 1 CO • -00 
t^O O r^O O • ■ O 


t-sO o 


d 


0( 

■<: 


AIIV(I 
S3X.1NIIV 




■ GO O • • • O 

• « « . . ■ n 


• O • lo • lo ■ O 


o • • 
t ■ • 


I 


xsu 

N0IXIS0d«03 


0> 'O • 


. ■ • CO t • • 


00 CO • ■ loo >o <^ 




~ 


O- 0^ ■ 


. . . o i^ • • 

. • . o- Ov • • 


r^oO • • oo oo 00 00 
O- Os • • Os Os o^ o> • 




_t 


ajNaxNas 


O 00 -"t 


t to o 


■^t'T'N nO to toOO <N M 




f*5 •t OS »0 CO - CO 
1^ 1^ 1^ l~ 1^00 OO 


t O^O O^ c^O O to vO cooo 
t^ t^ 1^ i^ c^ i^ t^ t^O «^ 1^ 1^ 




•< 

>" 

a 

M 


A1IV(I 

saxaNijv 


• 1/-) . 


• ■ CO • • p 


• lo O • O ■ O • 


o o 


a 


NOU.IS<jaK<).) 


rrt o • 




o ■ • -r o« • o- • 














Os o> ■ • o- o> • Os • 














aosaxNas 


tea f^ 


M CO '^ 'CO 


MOOMi-iO«t oo 


• t t 




O O fO f^ fO 1^ • • • "t-O O 1^ o> f* O-OO 'n « 


■ t^OO 
• u^O 

< n < 




icx)ius 


sa < 


n PQ Q 


PQ S < H < 


U 






AXQ 


•- "H 


I-. lo ><-, 


lo >n ■^ ^ O 


a 





SPELLING 



337 



Cornman made a similar study and reports substantially the 
same situation. A sample of his results follows: 



TABLE I02. After Cornman ('02) 
Records of 1898 





4th a Gilvde 


5th Grade 


School 


Time Given 

TO Spelling; 

Minutes per 

Week 


Ability of the 
Class in Spelling 


Time Given 

TO Spelling; 

Minutes per 

Week 


Ability of the 
Class in Spelling 


A 


50 
75 
100 
100 
100 
120 
150 
200 


67.0 

75-5 
76.4 
65.0 
66.0 
65.0 
70.3 
76.3 


50 
60 
80 
100 
100 
100 
100 
160 


68.0 


C 


72.5 
835 
57-4 
76.2 


E 


F 


G 


D 


66.0 


I 


67.1 
82.0 


H 







The investigations of both Rice and Cornman are highly im- 
portant but they do not afford conclusive evidence that time 
makes no difference. The fact that some schools devote two or 
three times as much time to spelling as other schools do and obtain 
thereby no better results, does not prove that time plays no part, 
for the reason that many other complicated factors enter, such as 
differences in teachers, method, spirit, and the like, to make the 
effect of any one factor unanalyzable. Indeed, we might infer 
that the lower schools, assuming all conditions the same, might 
have obtained even poorer results than they did if they had de- 
voted to spelling only as much time as the better schools did. 

In order to ascertain definitely the effect of different lengths of 
class periods upon spelling efiSciency, it would be necessary to 
proceed in a more rigorously scientific manner rather than to 
make inferences on the basis of complicated, wholesale statistics. 
It would be distinctly worth while to undertake an experiment by 
teaching several equally able sections of a given class under as 
nearly identical conditions as possible, such as having the same 
teacher, text, method, and environment, and by varying only 
the time element so as to have, for example, a period of 15 minutes 
for one class, of 30 minutes for another, and of 45 minutes for still 
another. Comparisons by adequate tests at different times would 
yield conclusive evidence concerning the effect of time upon ulti- 



338 EDUCATIONAL I'SVCHOLOC.V 

mate achievement. It would seem reasonable to anticipate that 
longer peritxls of equally good class instruction \vould i)roduce 
greater efficiency. The fundamental j)roblem, of course, would 
be the determination of the o[)timum length of the class period in 
relation to the desired efficiency in spelling. 

(4) Methods of Teaching Spelling. Along with the question 
of time, Rice and Cornniun were interested also in determining the 
influence of different methods of teaching. Rice does not present 
definite figures, but, on the basis of extensive inquiries among the 
teachers in the schools which he tested, he tried to ascertain the 
facts coneming methods of instruction and environmental con- 
ditions. His conclusion was as follows: 

"In brief, there is no direct relation between method and results. . . . 
The results varied as much under the same as they did under difTcrent 
methods of instruction. The facts here presented, in my opinion, will 
admit of only one conclusion, viz., that the results arc not determined 
by the methods employed, but by the ability of those who use them. In 
other words, the first place must be given to the personal equation of the 
teacher, while methods and devices play a subordinate part." (.\fter 
Rice.) 

Practically the same criticism, made in the preceding section 
concerning the factor of time, is pertinent here. The situation is 
too complex and the cooperating factors too numerous to make 
such inferences without a careful isolation of the individual ele- 
ments and their separate efTects. Each elenient should be subjected 
to an experimental procedure similar to the one outlined in the 
preceding sections. 

Comman went further than Rice by undertaking an experiment 
in which the spelling i)eriod was entirely eliminated from two 
schools in Philadelphia for a period of three years. He states: 

"It was decided to abandon the use of the spelling book and home 
lessons in the subject, to omit also the period from the school programme 
which had been devoted to its study and recitation and to investigate the 
effect that the abstraction of these influences might produce upon the 
spelling of the pupils of the several school grades. Several methods of 
measuring results were devised which will l>c herein described and statis- 
tically reported upon." (.\fter Comman.) 

On the basis of these tests, made at different intervals, Comman 
found that the spelling ability of the two schools was almost as 



SPELLING 



339 



good as that of the other schools and that the pupils improved 
steadily in spelling, even though special instruction in them had 
been omitted. Some of his figures follow : 



TABLE 103. After Cornman 

Spelling ability measured by uniform examinations for all schools, 
given by the city superintendent. 



50 Schools Giving Specific 
Instruction in Spelling 

7th grade 73 o. 

6th grade 70 . 4 . 

5th grade 71.6. 

Average 7i-7- 



2 Schools in Which for 3 Years no Specific 
Instruction in Spelling had been Given 

69.9 

65.1 

72.7 

69 . 2 



The spelling ability of classes in the Northwest School who had for 
three years been without specific instruction in spelling compared with 
that of corresponding grades of previous years, who had had the full 
amount of drill in speUing. 



Tests of June, 1897 

Sth grade 99 • 4 ■ . 

7th grade 99. i . . 

6th grade 96 . 75 . 

Sth grade 9695- 



Tests of June, 1900 

99-8 

98.6 

99-0 

97-6 



The spelling ability in a test in writing words in sentences of classes 
which had been without specific instruction in spelling for a year and of 
classes which had regular drill. 



Classes With Regular Drill 



Tests of 


June, 1897 


Northv?est 


Sth grade 




7th " 




6th A " 




6th B " 




Sth A " 




Sth B " 




4th A " 




4th B " 




3rd A " 




3rd B " 


Agnew 


4th A grade 




4th B 




3rd A " 




3rd B " 



86.1. 

83.5. 

72.7. 
80.6. 
78.8. 

75- 1- 
85. 8. 
86. s. 
57-8. 

76.8. 
82.5. 

72.3- 
66.1. 



Classes Without Regular Drill 
Tests of June, 1898 

) 90 . 6 ... . 



Difference 



0.8 
-2.6 
-6.0 
— 1 .0 

-3-7 
0.4 
S-6 

-o.S 
-16. 1 

—0.1 

5-2 
1.2 

1-4 
1.6 



340 EDUCATIONAL PSYCHOLOGY 

That is, half of the classes without sjK'cific instruction did better and 
half of them did worse than the corres{X)nding classes with specific in- 
struction. 

Cornman concludes: 

" (6) The amount of time devoted to the specific spelling drill bears no 
discoverable relation to the result, the latter remaining i)ractically con- 
stant after the elimination of the spelling drill from the school i)rogrammc. 

"(9) It is therefore advisable, in view of the economy of time, to rely 
upon the incidental teaching of spelling to produce a sufliciently high 
average result. 

" (10) The average result is what can be and is attained, as shown by 
statistical evidence, by average pupils under teachers of average pro- 
fessional efhciency in classes of average size, i. e., in the elementary 
schools of this country as now organized. To remain strictly within the 
evidence gathered by this investigation, it must be admitted that there 
may be teachers of surpassing ability, who can obtain more than average 
results by the method of the specific spelling drill, and other teachers of 
meaner ability who need the drill to bring their pu{)ils up to the level of 
this average result. It is claimed, however, that there is no evidence 
(whatever may be the wealth of opinion) to prove that such teachers 
exist or to show where they may be found. Moreover, the evidence 
which has been presented in this paper makes their existence at least 
improbable." 

The extensive work of both Rice and Comman has been very 
valuable in attacking a large educational ])r()blen\ by more exact 
methods and in showing that there is undoubtedly an enormous 
waste in the teaching of spelling, as there is ])robably in all sch(X)l 
subjects; but the results are not final ])roof that length of time or 
methods of learning arc negligible elements. Indeed l)oth labora- 
tory exi)eriments in the learning j)roress and the more recent and 
more carefully controlled tests in spelling itself indicate cjuitc 
certainly that length of lime and manner of learning do make 
important dilTerences. 

The author found that there are large dilTerences in the average 
spelling attainment of the various classes of any given grade in 
the same school system. The results for the 10 schools of a Wis- 
consin city, Table 104, City J, were as follows: 



SPELLING 



341 



TABLE 104 

Differences in attainment in spelling in 18 Wisconsin cities as measured by the 

author s test 



Grades 
Standard Scores 

City A 

C, School 



F 

I, School 



J, School 



30 



3 
40 



4 
51 



I 










.36 


. . . .40 

5 • ■ 53 


2 










■34 


0. .47 


3 










.38 


0. .43 




6 


9- 


.29 


3- 


■38 


. . 49 ■ 


I 










■37 


5^^45^ 


2 










•39 


5^^4S 


3 














I 






•30 


8. 


■43 


6. .52. 


2 






•35 


0. 


■34 


0. .47. 


3 


10 


3- 


•31 


4- 


■41 


9. .52. 


4 


6 


I . 


.10 


0. 


■32 


0..47. 


5 


ID 


0. 


•23 


2. 


■39 


2. .51. 


6 


8 


I. 


■25 


I . 


■36 


I 45^ 


7 


23 


I. 


.48 


2. 


■44 


0. .58. 


8 


7 


4- 


•25 


8. 


•37 


7^^44. 


9 


II 


3- 


.28 


9^ 


■44 


2. .56. 





ID 


2. 


■35 


2. 


■43 


2^^54^ 



5 

61 

■ 52. o 

.61.7 

.60.7 

■57^1 
.63.2 



■54- 
.61, 
•52. 
■59^ 
.62. 

■55^ 
.58. 
.62. 

■53^ 
.62. 

■55^ 
.60. 

■55^ 



6 
71 



7 
78 



. .68.8. 












. .68.8. 


76.2. 


75^2 


. .69.0. 


84.0. 


88.0 


. .66.0. 






..82.3. 


70. 5^ 


77^5 


. . 64 . 1 . 






. . 60 . 4 . 


lS-9- 


81.3 


. .79^o^ 


80.0. 


83.0 


. .60.2. 


82.2. 


85.0 


. .6S.0. 


75^o. 


77.0 


■■67^5^ 


69.7. 


78.8 


. .62.7. 


74^o. 


79 7 


■ ■749^ 


76.1. 


79 ■o 


■■67. 3^ 


75^o. 


77-9 


. .67.0. 


73.0. 


81. 5 


.64.8.. 


7i^3-^ 


82.2 



Thus the lowest sixth grade made an average of 60.2 which is 
barely up to the standard of the fifth grade and the highest made 
an average of 79.0 which is above the standard of the eighth grade. 
These differences are actually quite large when we remember that 
they are grade averages and not scores of individual pupils. These 
differences must be due in a large measure to differences in methods 
of teaching. 

Wallin attempted to make a comparison of the careful drill 
method, devised for the Cleveland schools by Assistant Superin- 
tendent W. E. Hicks, with the incidental method employed by 
Cornman in the two schools in Philadelphia from which the spell- 
ing period had been eliminated. Wallin reported results from the 
Cleveland schools distinctly superior to those in the Philadelphia 
schools. 



TABLE IDS. After Wallin ('11 ) 
Combined averages for the composition and column tests, all schools 
Grades 4 5 6 7 8 All Grades 

Percent 98.40 96.31 96^95 97 •03 96.28 97.00 



342 EDUCATIONAL PSVCHULU(JY 

He states: 

"First, the general spelling efficiency for all schools shown (q7^) is 
striking. It is 12.6% higher than Kratz's results (84.49^, for the fourth 
to eighth grades, inclusive). 25'^(, higher than Chancellor's (72*^^, for 
10,000 pupils from the fourth to eighth grade), 25.48% higher than the 
results in Rice's column test (71.52^1), which consisted of a list of dic- 
tated words, and 22.42% higher than the results from his sentence test, 
which consisted of dictated sentences containing 50 test words for the 
fourth and fifth grades, and 75 words for the sixth, seventh and eighth 
grades. It eclii)ses by 25.7' ^ Cornman's average in three term examina- 
tions during three years for eighty Philadelphia schools (7 1.3^6 )> and 
is 27% higher than the results of these examinations in his two experi- 
mental schools (7o'^(), in which the spelling instruction was incidental. 
In four column tests given to these two schools from September to June 
and consisting of lists of fifty words differing from grade to grade, the 
averages were 33%, 49%, 50% and 50% respectively for one school, and 
49%» 57%. 60% and 6S% for the other; while the repetition of Rice's 
column and sentence tests gave an average efliciency in 1897 of 78.9% in 
one school, and 67.1% in the other, and in 1898, 73.i'^o 'I'l^l 65.4% re- 
spectively, for the column lest. The corresjX)nding averages for the 
sentence test were: 82.3'^'(, and 74.6^(, in 1897, and 76.5' o '-^'^^ 77-9% iii 
1898. It will ije observed that there was a loss of efficiency in 189S except 
in the case of the sentence test for one school." (.\fter Wallin.) 

Miss Fulton reports an experiment in which an attemj^t was 
made to ascertain the effect of a specific drill in learning to spell. 
One hundred words, ten each day for two weeks, were taught by 
the following i)lan: i. "Write word upon the hoard." 2. "E.\])lain 
meaning." 3. "Children use the word in a sentence." 4. "Write 
word ten times while saying letters aloud at same time." 5. "Em- 
phasize by intonation of voice or by colored chalk, on blackboard 
the difficult part of words." At the end of the two weeks a test 
with the 100 words was given. 

During the ne.xt two weeks another list of 100 words of similar 
difficulty was used with no directions except to "study tlie lesson." 
The children were "heard" but no special cfTort was made to 
teach the words. A lest with these words also was then given. 
The results of the two i)lans as indicated by the tests given im- 
mediately and three weeks later were as follows: 

TABLE 106. After fullon ('14) 

Wrrn Timrr. Weeks WrnioiT Thrfk Weeks 

Drill I.atek Urill I.xtkr 

AvcraBC % 08 96 73 68 

Daily results y8 78 



SPELLING 343 

(5) Laws of Association. Skill in spelling is primarily a matter 
of forming associative connections between certain arbitrary 33^1- 
bols arranged for the most part in arbitrary order. Economy in 
the learning of spelling reduces itself to this question: Under what 
conditions can these associations be made most quickly, most 
effectively, and most permanently? Of the four laws of associ- 
ation, frequency, vividness, primacy, and recency, the first two 
are most directly applicable. Obviously, frequent repetition is 
necessary to establish the connections. Frequent reviews, monthly, 
weekly, and possibly daily, are indispensable. 

The law of vividness operates in numerous ways. This law 
states that other things being equal, the most vivid or intense 
association is most apt to be recalled. It may be made of service 
in two general ways: (i) By any device that will stimulate the 
clearest possible attention and interest on the part of the learner 
upon the spelling of words, or (2) by any device that will make 
conspicuous the particular part of any word that is most likely to 
be misspelled. These points will be considered more fully in the 
next three paragraphs. 

(6) Careful Attention upon the Successive Letters of a Word. 
Pryor ('15) reports "an experiment to determine the value of 
'spelling the word through' as an aid to learning. Two divisions 
of the fifth grade studied the same list. Conditions as to time, 
length of period, and the like were the same for both divisions. 
For one, emphasis was placed on observing carefully the order of 
letters while studying. Preliminary and final tests given to both 
divisions showed an advance from 50.55% to 83.39%, or an average 
gain of 32.84% for the division working under the usual condi- 
tions. The other division advanced from 48.58% to 89.14%, an 
average gain of 40.56%." 

(7) Personal Incentive to Interest and Effort on the Part of the 
Learner. Aside from competitive contests in their various forms, 
there are at least two ways in which personal effort may be stimu- 
lated: (a) By having at regular intervals definite tests, preferably 
by means of some one of the standard spelling scales or tests so 
that each pupil may know his attainment from time to time and 
measure his progress. This plan will usually arouse very keen 
interest in the individual to surpass his own previous record. See 
Chapter XL This is a form of incentive whose actual effective- 
ness has never been fully appreciated, (b) A second form of in- 
centive is to arrange with the pupils that the grade in spelling will 



344 



EDUCATIONAL rSVCHOLUC^V 



be determined half l)y the work in the spelling-class and half by 
their spelling efTiciency in all written work. This i)lan was em- 
ployed in a school in Potsdam, Xcw York, with the result that the 
pupils developed a remarkable efficiency in spelling as shown in 
the accompanying graph, Figure 73. Each grade is approximately 
one entire grade ahead of the general average. This is one of the 
highest records found in any school measured by the writer's 
spelling test. 

(8) Calling Special Attention to Difficult Parts of Words. One 
of the much needed investigations of spelling is a careful collection 
and classification of errors in the words used as spelling material. 

100 



t 80 



60 



40 



y 20 





1 '^ 3 4 5 6 7 8 

Grades 
Frr,. 73. — AvcraRcs in spelling in grades 5 to 8 in a certain school as measured 
by the author's test. The continuous Hnc represents standard attainments. 















,^- 










.^ 










^ 


^ 








y 


^ 












/ 















P>rors in spelling usually consist in certain letters or parts of 
words only. For example, the error in receive is usually in the two 
letters c-i, or in separate, in the a after the />. (Such a study is 
under way for the 2,626 most common words previously referred 
to.) By means of such a study the teacher would be able to antici- 
pate the probable misspellings and to call especial attention to the 
letters likely to be missed. Such particular emphasis may he 
secured by asking the pupils to focalize the correct order of the 
letters which are usually confused, by writing those letters extra 
large, or by drawing a line around them, or by having them printed 
in larger or heavier type, and so on. It is quite imperative to 
anticipate the troublesome jxirts of words .so as to forestall tlu- 
formation of wrong connections. Instruction in spelling should 



SPELLING 345 

consist in the teaching of correct forms rather than in the un- 
teaching of incorrect forms. 

(9) Writing the Words. We need to know how to spell words 
only in writing. It is a general law of association that connected 
bonds should be established in the manner and order in which 
they are to be used. It would seem, therefore, on a priori grounds 
that the associative bonds between the successive letters should 
be formed by exercise in the writing of the words so that the spell- 
ing may become automatic during the act of writing. Current 
emphasis upon writing the words in sentences is in the right di- 
rection. It would be an interesting experiment to teach two 
sections of a spelling-class by having in one section a great deal of 
writing of the words of the spelling lesson and by having little or 
no writing in the other section. The spelling ability of the two 
sections would have to be compared at various times by appro- 
priate tests. 

(10) Context versus Column Spelling. On the basis of the 
discussion in the preceding paragraph, it would seem obviously 
advantageous to have a great deal of writing of the words to be 
learned, especially in sentences. The belief is held by many that 
the ability to spell words in isolation does not insure ability to 
spell them in context. As a matter of fact, however, there is very 
little dilTercnce in spelling efficiency between these two situations. 
Wallin ('11) used the same test words both in a column test and in 
a dictated composition test and found an average spelling efl&- 
ciency of 97.72%, in the column test, and 96.28% in the composi- 
tion test, giving an advantage of only 1.44% to the former. Cook 
('14) used 60 words in a column and in a composition test and 
likewise found only a very slight advantage in favor of the former 
test. Spelling efficiency in composition is only very slightly lower 
than it is in columns. This slight loss is probably due to the dis- 
traction of attention by the other factors in writing, such as punc- 
tuation, grammatical form, and thought content. The argument 
that spelling should be taught by using words only in sentences 
is not very weighty. 

(n) Teaching Homonyms Together or Separately. Suzzallo and 
Pearson ('13) report an experiment in which they attempted to 
determine the relative effectiveness of teaching homonyms to- 
gether or apart. They used five pairs of homonyms in each of 
grades 3 to 7 in the Horace Mann school. Each grade was taught 
in two sections of about equal ability. One section was taught by 



346 KDrtAIIOXAL PSVt'HOLO(;V 

llic togi'ilur-iiK'lhod and the other by the separate-method. All 
other conditions were kept as nearly alike as possible. All 
classes were tested at the beginning and at the end of the experi- 
ment. The outcome showed a decrease of 2.29 errors by the 
together-method, and of i.Oi errors by the separate-method. 
There is thus a slight advantage in favor of the former methcxl. 
Suzzallo's experiment was repeated with the siime material by 
Knight at Montclair, New Jersey, in grades 3 to 7, inclusive. The 
together-method reduced the errors on the average 2.63% while 
the separate-method reduced them 2.24%, thus supporting Suz- 
zallo's findings. On the other hand, \V. F. Jones ('15) reports 
that 

"Experiments in teaching homonyms have been made by the depart- 
ment of education at the University of South Dakota, which show that 
homonyms should not be brought together until the second one of the 
pair api)cars in the child's vocabulary. This often gives time to fix the 
meaning and the spelling of the I'lrsl member of the pair before the second 
one appears. In such cases the homonyms give relatively little trouble." 

(12) Class versus Independent Study, Suzzallo and Pearson 
further undertook a comjxirison of progress in spelling when the 
pupils studied indeiKudently, with j^rogress wheii the ])upils 
studied under sui)cr\ ision. For nearly a year, one class in each of 
grades 4, 5, and 6 was taught by the supervised-study jilan and the 
other by the independent-study plan. In the latter case, the recita- 
tion period consisted chiefly in testing or lesson hearing. Their 
conclusion is stated thus: 

"The evidence of this experiment, therefore, from whatever angle we 
study it, shows that teaching of the class-study type is far more effective, 
than the independent-study type." 



SPELLING 



347 



TABLE 107. After Suzzallo and Pearson ('13) 
Total decrease in errors 





Independent Study 


Class Study 


Gr^dk 


Room 


FrRST 
Test 


FiVAI. 

Test 
228 


Net 
De- 
crease 


Num- 
ber OF 
Pupils 


Average 
Decrease 
per Pupil 


Room 
111 


First 
Test 


FiNAI 

Test 


Net 
De- 
crease 


Num- 
ber OF 
Pupils 


Average 
Decrease 
per Pupil 


IV 


111 


286 


58 


22 


2.63 
maximum 
20 


275 


144 


131 


22 


5 95 
maximum 
20 


V 


201 


291 


208 


83 


20 


4.15 
ma.ximum 
24 


201 


291 


108 


183 


20 


9.15 
maximum 
24 


VI 


206 


349 


221 


128 


23 


5.56 
maximum 
32 


206 


363 


143 


220 


23 


9.56 
maximum 
32 


VII 


209 


351 


218 


133 


20 


6.65 


209 


388 


206 


182 


20 


9.10 

maximum 
32 






127T 








Average 

Decrease 

per 

Grade 

4.74 




1317 








Average 

Decrease 

per 

Grade 

8.44 



(13) Grouping Words of Similar Spelling. Is it an advantage 
to learn words of similar spelling in groups? Wagner ' made an 
experiment to answer this question. He divided a 6th grade class 
into two sections both of which studied the words in the usual 
manner with the exception that the words were presented to one 
section in groups, according to their similarity, such as lineal, 
lineament, linear and lineage while to the other they were presented 
in miscellaneous combinations. The former section, which studied 
the words in groups, raised its average percentage from 68.36 in 
the preliminary test to 97.14 in the final test, or 28.78% while 
the latter section raised its average percentage from 73.25 to 93.6, 
or 20.35%. The former group, therefore, gained 8.42% more 
than the latter, showing a decided advantage in favor of the 
grouping plan. 

(14) Imagery. Individuals differ in mental imagery, but recent 
inquiries point out the probability that pure types of visuals, 
audiles, motiles, etc., are exceedingly rare and that imagery of any 
sort may be aroused irrespective of the sense organ through which 
the stimuli come. Since practically every child possesses images 
of all classes the safest procedure is no doubt to appeal to a variety 

1 Reported by Pryor. ('15.) 



348 EUUCATIUNAL I'SVCIIOLOGV 

of images. Note here again the discussion of imagery in Chapter 
XI, p. i66. 

(15) Spelling To-day and Formerly. Finally it will he of interest 
to note how the spelling ability ol pupils of to-day compares with 
that of our forefathers, particularly in view of the claim often made 
that the schools to-day do not train the pupils as thoroughly in the 
fundamental subjects. One of the com])arisons made in the Spring- 
field test (Riley, 'oS) was that of spelling ability. The same 20 
words that had been given as a spelling test to 9th grade i)upils 
in 1846 in Springfield, Massachusetts, were given again in 1906 
to 246 pupils of corresponding age in the same school. The pupils 
in 1S46 had made an average grade of 40.6'^e> ^^■hile the i)upils 
in 1906 made an average of 51.2%. A similar test, made in 
Cleveland in the years 1S5S and 1909 showed one error less per 
child in the 1909 test. Apparently the "superior" spelling ability 
in the good old days is largely an illusion. 



CHAPTER XIX 
LANGUAGE 

Psychological Processes Involved in Language 

The subjects thus far considered, namely, reading, writing, and 
spelling, together with the one to be considered in this chapter, con- 
stitute psychologically the complete set of language functions since 
they all play a part in the communication of ideas. Reading deals 
with the reception of ideas; while writing, spelling and language in 
the restricted sense, have to do with the expression of ideas. The 
term language as used here, and as used in the school program, 
refers only to that portion of the complete language process which 
deals with the organization and expression of thought in speaking 
and writing. 

A complete analysis of all the psychological processes involved 
in the language functions would require again an enumeration of 
all the elements in reading, writing and spelling. This is unnec- 
essary; and hence our present analysis will be limited to the steps 
immediately concerned in the expression of ideas in oral or written 
form. These elements are: 

(i) The arising of ideas in the mind. 

(2) The simultaneous or successive arousal of symbols or word 
forms corresponding to the ideas. 

(3) The transmission of the nerve impulses, connected with the 
ideas that axise, to the motor speech centers in oral expression, or 
to the motor writing centers in written expression. 

(4) The transmission of nerve impulses from the latter to either 
the muscles of the speech organs or to the muscles of the hand and 
arm. 

(5) The execution of the speaking or writing movements. 
Steps (3), (4) and (5) are identical with the corresponding ones 

previously enumerated in the analysis of the writing and spell- 
ing processes. The important steps for our present purposes are 
numbers (i) and (2). 

From the psychological standpoint, the arising of ideas in the 
mind is practically identical in the normal person with the arising 

349 



35° KDUCATIONAL rsvciioL()(;v 

of ^vords in the niiiul. since in the normal child words and meanings 
arc built uj) together through oft repeated associations of words 
and meanings. Hence we shall use the term "word-idea " to convey 
this union of symbol and meaning. Furthermore, from the psychol- 
ogical standpoint, the occurrence in the mind of ideas or words 
to be expressed, is fundamentally a mattir of the psycholog)- of 
association. Where do the ideas come from? What aiuses them 
to arise in the mind? Why do certain ones arise rather than others? 
To what extent can the occurrence of ideas be controlled? Com- 
position, either oral or written, is simply the outward expression of 
the ideas that do arise. The occurrence of the ideas and their 
precise \'erbal form takes ])lace in the mind as a i)art of step (i). 
Obviously then, a study of steps (i) and (2) and an attempt to 
answer the questions raised there])y, constitute almost entirely 
a study in the jisychology of association or thinking processes. 
Comj)osing fundamentally is thinking. 

Let us take a t}q)ical example of oral or written composition, 
such as a Int of speaking or writing. I low do the ideas in this 
simple composition arise in the mind? The blunt answer is, they 
arise almost entirely in a mechanical manner according to the 
established neural or mental connections. The first idea in a chain 
arises as the result of a stimulus, either through the senses or 
through previous ideas or images in the niind. No idea probably 
ever arises independently in the mind as though out of the blue 
sky, but always in succession to a preceding link, stimulus or 
occasion. Thinking is simjily the flow of ideas, which occurs 
almost wholly automatically, according to the laws of association 
— frequency, vividness, ])rimacy and recency. Conversation also 
is largely the automatic flow of association jiroces.ses in the minds 
of the ])articij)ants. Each statement comes in resi)onse to the re- 
mark of one of the conversers or in succession to the jjreceding re- 
marks of the si)eaker himself. Even more formal comi)osition, 
such as the writing of a theme, a stor}- or an essay, is i)rincipally 
the result of association processes aroused in the mind by the sub- 
ject of the theme. This occurs, even in vigorous and original 
thinking about a subject, largely according to the mechanical laws 
of as.sociation. Voluntary thinking or composing is controlled 
probably only in two ways: (i) by effort and concentration more 
ideas are likely to arise than by a ])urely passive attitude, and (2) 
by making a selection among the ideas that do arise i)reference to 
further chains of ideas are delermined. lUit the arising of ideas 



LANGUAGE 351 

themselves occurs fundamentally according to the mechanical, 
neural and mental connections previously established. 

The objection is likely to be made by the reader at this point, 
that if this is true, how can any new ideas ever arise? or how can 
any original thinking occur? The blunt answer is: Thinking is 
original only in the sense of making certain selections, rather than 
others, among the ideas that do arise and of allowing further 
associations to arise in connection with them, rather than in connec- 
tion with others. In this sense, the possibility of original thinking 
is indefinitely great. 

Correct English is fundamentally a matter of associating certain 
words in certain orders, rather than in others. The reason why 
people say, "Do it good," is because they have been told since 
infancy to "Do it good." Language is an immensely intricate 
net-work of associative links. Wide vocabulary in speaking or 
writing is due to the arising of numerous words in connection with 
given meanings. Elegant diction is due to the more appropriate 
selection of words among the larger varieties that do arise. In the 
skilled speaker or writer the more appropriate words arise auto- 
matically in the course of time. Greater varieties of words come 
up with given word-ideas because more of them have been previ- 
ously experienced and retained. Two persons may read the same 
literary masterpiece, the one may retain a great deal of the actual 
diction and phraseology in connection with the ideas read, while 
the other may retain very little. In the course of time the former 
will acquire a far wider vocabulary and a much nicer diction than 
the latter because he retains much more of the actual words and 
forms of expression than the other, and consequently when any 
topic is brought to his mind, it arouses a far richer wealth of ideas, 
words and phrases, and by virtue of this wealth he is able to make 
a far superior selection.- Ingersoll appropriately said of Shake- 
speare: 

"The moment his attention was called to any subject — comparisons, 
definitions, metaphors, and generalizations filled his mind and begged 
for utterance. His thoughts like bees robbed every blossom in the world, 
and then with 'merry march' brought the rich booty home 'to the tent 
royal of their emperor.'" (P. 661, Modern Eloquence, Volume V.) 

"Some have insisted that Shakespeare must have been a physician, 
for the reason that he shows such knowledge of medicine, of the symptoms 
of disease and death; because he was so famihar with the brain, and with 
insanity in all its forms. 



352 EDUCATIONAL PSYCHOLOGY 

" I do not think he was :i physician. He knew too much ; his generaliza- 
tions were too splendid. He had none of the prejudices of that profc-ssion 
in his time. We might as well say that he was a musician, a com[X)ser, 
because we find in 'The Two Gentlemen of Verona' nearly every musical 
term known in Shakespeare's time. 

"Others maintain that he was a lawyer, perfectly acquainted with 
the forms, with the expressions familiar to that profession. Yet there is 
nothing to show that he was a lawyer, or that he knew more about law 
than any intelligent man should know. He was not a lawyer. His sense 
of justice was never dulled by reading English law. 

"Some think he was a botanist, because he named nearly all known 
plants. Others, that he was an astronomer, a naturalist, because he gave 
hints and suggestions of nearly all discoveries. 

"Some have thought that he must have been a sailor, for the reason 
that the orders given in the opening of 'The Tempest' were the best 
that could, under the circumstances, have been given to s;ive the ship. 

"For my part, I think there is nothing in the plays to show that he 
was a lawyer, doctor, botanist, or scientist. He had the observant eyes 
that really see, the ears that really hear, the brain that retains all {)ictures, 
all thoughts, logic as unerring as light, imagination that supplies defects 
and builds the perfect from a fragment. And these faculties, these apti- 
tudes, working together, account for what he did." (P. 665.) 

Thinking and language are the two sides of the same shield. 
The language used to e.xpress ideas dejK'iids ujxjn the thinking 
tliat goes on in the mind; and the thinking dei)ends upon the 
verbal-ideational connections established in the neural and mental 
network by reading and hearing successions of words, phrases and 
sentences. Language is not words; it is thinking, thinking by 
means of symbols. 

The Measurement of Efficiency is English 

(a) Methods of Mcasurcmnit. Four l\-]ies of measuring devices 
have been prepared, each for a dilTerent phase of language, (i) For 
measuring the grammatical correctness of language, the writer ('15) 
has prepared a series of scales consisting of sets of sentences, each 
stated in two dilTerent ways. These sets of sentences are arrange<i 
in an order of increasingly diflicult steps. Tlie pupil in doing the 
test is ref]uested to indicate wiiich the correct or preferred form is. 
A similar scale has been prepared for testing ability in punctua- 
tion. (3) For measuring technical knowledge of the grammatical 
structure of Englisli the writer has devised several tests for ascer- 
taining ciuickness and accuracy in indicating parts of speech, 



LANGUAGE 353 

cases, tenses and modes. (3) For measuring general merit in 
written composition, two scales have been constructed. The one, 
known as the Hillegas-Thorndike ('12) scale, is composed of a 
series of compositions, arranged according to a large number of 
judgments in the order of steps of increasing merit from zero to 
nearly one hundred. The second, known as the Harvard-Newton 
scale (Ballou, '14), is composed of a series of four scales for nar- 
ration, description, argumentation and exposition respectively. 
Each scale contains six compositions. These were graded by 24 
teachers and approximate in percentage marks the values of 45, 
55) 65, 75, 85 and 95. A composition is rated by any one of the 
scales by comparing its merit with those in the scale, and giving 
to it the value of the step on the scale to which it is judged equal. 
(4) Trabue ('16) has prepared a series of language scales, each 
consisting of some 10 mutilated sentences. These sentences are 
arranged in steps of increasing difhculty as determined by experi- 
mentation with a large number of pupils. In doing the test, the 
pupil is required to supply the most appropriate words in the 
blank spaces. It is difEcult to say just what this scale measures, 
but it probably tests the ability to think of the proper word for 
a given situation. 

(b) Uses and Results. Without repeating here, we need to say 
merely that these tests serve the same general purposes for the 
language functions as the measurements devised for the subjects 
previously discussed serve for their respective subjects. Measure- 
ments of ability in language have shown incredibly large individual 
differences among the pupils in the same class and the resulting 
overlapping of the abilities of pupils in succeeding grades. Note 
the facts in Figure 25 in regard to ability in composition in a 
certain Illinois high school. The overlapping is enormous and the 
median gain from year to year is surprisingly small. In fact there 
is almost no improvement above the first year. The little im- 
provement that is noticeable is probably due chiefly to the drop- 
ping out of the poorer pupils in the lower years. Figure 26 shows 
similar facts with regard to abihty in recognizing correct gram- 
matical forms. 

Brown and Haggerty ('17) measured the progress in composition 
ability in the case of 78 pupils in three classes during a period of 12 
weeks by having them write an extemporaneous composition each 
week. These compositions were rated ];y the Harvard-Newton 
scale. The median progress of the three classes is indicated in 



354 



KDUCATIOXAL PSVCHOLOGY 



Figure 74, being 4.2 points for Classes I and III, and 5.2 points 
for Class IL Tiiese graphs indicate a noticeable improvement 



Weeks of Practice 

5 7 S 



10 11 12 



100 



90 



80 



70 



2 60 



^ 50 



40 



^ 30 



20 



10 



Fir.. 74. — Median scores in composition for three lii^^li-school clafvsos. Con- 
tinuous line rcprt'scnls :i first-soincstcr freshman class; tlu- dotted line repre- 
sents a second-semester freshman class; and the dash-line represents a first- 
semester sophomore class. After Brown and Ilaggerty ( '17, p. 524). 

from the first week to the Iwchlh, l)ii( show no (htTcrcncc in ability 
between the sophomore class and the freshman classes. 

Economic Methohs for AroriRiNo Skill in the English 

Language 

A greal variety of cxpcricncps has been obtained and an equally 
great \ariety of personal opinions is lu'ld 1)\- teachers witli n-gard 























1 




>< 


::::. 


/. 


7^ 


1 1 ^^^ >'] 


>^ 


^^0^ 
'^H 


f^^^ 


■v^ 


^y'^ 


3" 


^ 


2^>: 

















































































































































































LANGUAGE 355 

to the most effective manner of learning and teaching English. 
But there is a corresponding paucity of scientific facts regarding 
the matter so that we cannot speak with certainty concerning very 
many of the factors and conditions that promote or retard develop- 
ment of skill in the use of language. 

(i) The Acquisition of Ideas. According to the psychological 
viewpoint, from which we are here examining the language process, 
the first and most fundamental element is the acquisition of ideas. 
How this may be done our present psychological knowledge gives 
no specific directions other than the common-sense advice of 
gaining as great a variety of ideas as possible through direct per- , 
sonal experiences and contact with environment and through l 
extensive reading. The extent to which such varied experiences 
and wide reading will be profitable depends mainly upon the extent 
to which the experiences and facts are assimilated and retained. 
Next to the experiences and ideas themselves, excellent retentive 
capacity, or memory is a very essential factor. Re-thinking the 
experiences and facts is likely to be a most helpful exercise in 
making new ideas a part of one's mental machinery. 

(2) The Acquisition of Words and Forms of Expression. The 
next important element is the acquisition of extensive vocabulary 
and phraseology. On this point also little more can be said at ^ 
present than what, on the basis of general experience rather than 
on the basis of demonstrated facts, common sense and good judg- 
ment dictate. Wide vocabulary and precise phraseology can prob- 
ably be acquired mainly (a) by reading and hearing language 
which employs extensive vocabulary and proper phraseology, and 
(b) through definite attempts to express one's thought by means 
of these words and phrases. Probably one of the best means of 
extending one's vocabulary is to take special account of new words 
as one reads or hears them, and to make a special point of using 
them on the first, and on every other, appropriate occasion. A 
useful plan is to record these new and unfamiliar words in a special 
notebook. They must, however, not be kept in the notebook only, 
but they must be referred to frequently so as to make them a part 
of the mental fabric of connections to be used in thinking, speaking 
and writing. Pupils often are required to keep such notebooks, 
but usually the words never get farther than the notebook. Special 
effort must be made to use the words so that they may gradually 
be woven into the automatic association processes of the mind. 

Another plan is to make a list of the less familiar or unknown 



3 5 (J EDUCATIONAL PSYCHOLOGY 

words that are obscr\t'(l in tlic language of good writers, speakers 
and conversationalists, and to think of as many synonyms and 
equivalent phrases as jjossil^le so that when occasion for using one 
of the words arises, synonymous terms will come up which might 
jireferably be used instrad. Frecjucnl drills and tests might very 
profitably be given in accordance with this suggestion. The teacher 
could give such a list of words in mimeographed form to the pupils 
and then see how many corresponding synonyms in a given time 
limit, say 5 or 10 minutes, each pupil can write out. The habitual 
use of words is fundamentally a matter of association processes 
and these can be built up only by a repeated use of the words to be 
acquired. 

For such instruction it would be exceedingly useful to have a 
list of the 10,000 most common words employed by the best current 
writers. Special drill exercises in the use and meaning of these 
words could be devised by the teacher.^ 

(c) The influence of knowledge of other languages. It has 
been generally urged that a knowledge of other languages, par- 
ticularly of those from which many English words have been de- 
rived, is of much value in extending the vocabulary', in develoi)ing 
a finer discrimination in the meanings and uses of words, and in 
facilitating the writing of English. The objective data thus far 
available on this i)roblem have been summarized in the chapter on 
transference of training, to which the reader should turn for a recon- 
sideration of that e\idence. In general the evidence indicates 
(i) that the study of Latin seems to increase the size of a pupil's 
English vocabulary only slightly when Latin is taught as it ordina- 
rily has been taught, or quite considera])ly when taught with specific 
reference to the derivation of English words; and (2) that, when the 
dilTerences in the original abilities of pupils are allowed for, the 
study of other languages aids either ver}' little or not at all in the 
writing of English. 

(3) Acquisition of Grammatically Correct English, (a) The 
iniluence of knowledge of grainnuir. Grammar was introduceil into 

* Such a list has liccn prcpare<l by the writer on the l)asis of the vocabulary studies 
described in Chapter XVIII. The Starch List of 5,003 dilTerent words used by 10 
authors was chetkecj up with the Lldri<l;;e, Ayres, Jones, ami C'tmk Lists to find all 
additional wonls in thcin which were not in tiie Starch List. This maile a total of 1/579 
different words found in .jo.ooo running words (Starch List) of .jo different luRh-Krade 
current niUKa/ine writers, in some 4,1,000 runnini: words (Kldriil^'e List) of newspajK-r 
writing, in some 2.?,ooo runninK words (Ayres Li>t) of corresiK)n<lence, in 15,000,000 
running words (Jones List) of children's compositions, and in 200,000 ruuoing words 
(Cook List) of family corres(K)ndenco. 



LANGUAGE 357 

the schools as a scientific study of the structure of the language 
with the belief that this knowledge would aid in the acquisition of 
correct usage of the language. To what extent is this belief jus- 
tified? All definite investigations of this problem tend to show 
that knowledge of technical grammar is of much less service in 
developing use of correct English than has always been supposed. 
This experimental evidence was reviewed in Chapter XIV on the 
transference of training, to which the reader should turn, and 
hence it will not be restated here. Furthermore, the studies by 
Charters ('15) on grammatical errors in oral and written language 
indicate that a relatively small number of grammatical rules is con- 
cerned in ordinary language. In accord with these results, the 
amount of time devoted to the study of grammar has been mate- 
rially reduced in recent years and a great many of the topics have 
been entirely eliminated. The amount of grammatical knowledge 
which functions directly in establishing correct usage is relatively 
small. For example, the Iowa Committee on Minimum Essentials 
recommended in 19 15 the elimination of the following topics: 

"The exclamatory sentence; the interjection; the appositive; the 
nominative of address; the nominative of exclamation; the objective 
complement; the adverbial objective; the indefinite pronouns; the objec- 
tive used as a substantive; the classification of adverbs; the noun clause; 
conjunctive adverbs; the retained objective; the modes (except possibly 
the subjunctive of 'to be'); the infinitive; the objective subject; the 
participle, except the definition and the present and past forms; the 
nominative absolute; the gerund nominative absolute; sentences for 
analysis and parsing that involve subtle points of grammar; formal 
parsing; conjugation; diagramming; person of nouns." (After Betts, '17.) 

(b) The function of imitation. These investigations imply that 
linguistic forms and expressions are acquired very largely through 
imitation, both conscious and unconscious, perhaps especially the 
latter, of the forms and expressions read or heard, particularly 
those heard in one's customary environment. Language forms are 
psychological habits which become deeply ingrained in the human 
psycho-physical system through constant repetition. To hear 
and say, from birth on, "good" for "well" in such expressions as, 
"Do it good," or "I don't feel good," establishes such strong chains 
of associated bonds that, in spite of better knowledge, such expres- 
sions will continue to be used and not be overcome by pages of 
grammatical knowledge. 



358 EDUCATIONAL PSVCHOLOCV 

The factor of conscious, intentional imitation is probably em- 
ployed far too little in the development of ability in composition. 
Students of art learn to paint by copying the great masterpieces. 
Students of composition are shown great masterpieces and asked 
to read them but are told not to imitate them as that would pre- 
vent the development of originality. Is this not incorrect proce- 
<!i:ie, ])sychologically? Why would it not be excellent practice 
in acciuiring vocabulary, diction, style, and even ideas to ha\c 
])upils read and study, for example, a descrij)tion of a landscape 
and then to ask them to write one of a real landscape in their 
environment by making it just as similar to the masterpiece, in 
l)lan, diction and phraseology, as the actual landscape permits. 
Or why should not such a i^lan as the following be really elTeclive 
in acquiring the use of words. Have the teacher take such a descrip- 
tion as the following one of George Washington: 

"When Washington was elected general of the army he was forty- 
three years of age. In stature he a little exceeded six feel ; his limbs were 
sinewy and well proportioned; his chest broad, his figure stately, blending 
dignity of presence wilh case of manner. His robust (onstilution had 
been tried and invigorated by his early life in the wilderness, his habit of 
occupation out of doors, and his rigid temperance, so that few equalled 
him in strength of arm or power of endurance. His comple.\ion w;is 
florid, his liair dark brown, his head in shape perfectly round. His broad 
nostrils seemed formed to give expression and escape to scornful anger. 
His dark blue eyes, which were deeply set. had an expression of resigna- 
tion and an earnestness that was almost sad." 

Let the teacher select all the important descriptive words and 
phrases from it and put them in a list, such as this: 

stature rigid temperance 

sinewy endurance 

well-proportioned complexion 

stately florid 

blending dignity nostrils 

ease of manner scornful 

robust resignation 

invigorated earnestness 
occupation 

The pupils should first study these words and phrases to make 
sure that they thoroughly understand them; then they should 
write a description of some person of their acquaintance who may 



LANGUAGE 359 

also be known to the teacher, using as many of these phrases as 
may be applicable to this person. The pupils in this case should 
not first read or hear the description from which these words were 
taken. After they have written their own composition, they should 
then compare it with the masterpiece so that they could check up 
their own use of words with the possibly better use in the model. 

Such a plan would seem to be worth while experimenting with 
to ascertain its proficiency in developing language ability. How 
can a pupil acquire proper words to be used in describing a human 
being if he docs not know, and has no means of finding out, what 
such words are? Such a plan would seem to have on a priori 
grounds two distinct advantages: First, it would make conscious 
use of imitation which is unquestionably a potent factor in the 
acquisition of language; and second, it would tend to make the 
learning of language definite and concrete. The great difficulty 
in learning composition is that the advice and suggestions given 
by the teacher are too vague and indefinite. The pupil is told 
that he lacks organization, that he must develop a better vocabu- 
lary, or that he lacks imagination. But how is any child specif- 
ically and concretely to know how to improve in these respects? 
How may a child know that such words and phrases as "stature," 
"stately dignity," "florid complexion" might be the most appro- 
priate to use when describing certain persons? When the child 
learns addition or spelling he knows more precisely what he has to 
learn. Instruction in composition can possibly not be particular- 
ized as fully, at least not as easily, but that should not prevent as 
much particularization as possible. 

The chief objection by teachers of English to such a procedure 
is that they believe it would kill originality and make mere thought- 
less, verbal machines out of their pupils. In answer to this point, 
however, we must remember that the most original people in the 
world are also the ones who use to the fullest extent the work, 
methods, and ideas of others. The most original persons are also 
the most imitative persons. Ingersoll said of Shakespeare: 

"Of course Shakespeare made use of the work of others, and we might 
almost say, of all others. Every writer must use the work of others. The 
only question is, how the accomplishments of other minds are used, 
whether as a foundation to build higher, or whether stolen to the end 
that the thief may make a reputation for himself, without adding to the 
great structure of literature. 

"Thousands of people have stolen stones from the Coliseum to make 



360 EDUCATIONAL PSVCIIOLOGV 

huts for themselves. Thousands of writers have taken the thoughts of 
others with which to adorn themselves. These are plagiarists. But 
the man who takes the thought of another, adds to it, gives it intensity 
and poetic form, throb and life, is in the highest sense original. 

"Shakespeare found nearly all of his fads in the writings of others and 
was indel)ted to others for most of the stories of his plays. The cjuestion 
is not: Who furnished the stone, or who owned the quarry, but who 
chiseled the statue?" (P. 645, Modern Eloqicciice.) 

Some of the prominent literary \\Titers have not only pointed 
out the iniportance of conscious imitation in the development of 
their own styles of writing, but have described their own conscious 
attempts to imitate other great ^\Tite^s. For example, Stevenson 
has said this concerning imitation: 

"That, like it or not, is the way to learn to write. It was so Keats 
learned, and there never was finer temperament for literature than 
Keats's; it is so, if we could trace it out. that all men have learned. Per- 
haps I hear some one cry out: 'But that is not the way to be original!' 
It is not, nor is there any way but to be born so. Nor yet, if you are born 
original, is there anything in this training that shall clii> the wings of your 
originality. There can be no one more original than Montaigne, neither 
could any be more unlike than Cicery; yet no craftsman can fail to see 
how much the one in this time tried to imitate the other. Burns is the 
very type of a prime force in letters; he was of all men the most imitative. 
Shakespeare himself, the imperial, proceeds directly from a school. Nor 
is there anything here that could astonish the consitlerate. Before he 
can tell what cadences he truly prefers, the student should have tried all 
that arc possible; before he can choose and preserve a fitting key of words, 
he should long have practiced the literary scales . . . and it is the great 
point of these imitations, that there still shines beyond the student's 
reach his imitable model." 

"Whenever I read a book or passjigc that particularly pleased mc, I 
must sit down at once and set my.self to imitate that ([ualily of projiriely 
or conspicuous force or hap|>y distinction in style. I was uiisuccessfr.l 
and 1 knew it, but I gt)t some practice in these vain bouts in rhythm, in 
harmony, in construction, and in co-onlination of parts. I have thus 
played the setlulous ape to Hazlitt. to Lamb, to Wordsworth, to Browne, 
to Defoe, to I lawthorne, to Montaigne, to Baudelaire, and to Obermann." 
(Stevenson, R. L., Memories and Portraits, p. 55.) 

Franklin likewise described his attempt to improve his own 
style of writing by carefully studying a volume of The ^Spectator 
and de\iloping a style similar to it. 



LANGUAGE 361 

"I read it over and over and was much delighted with it. I thought 
the writing was excellent, and wished, if possible, to imitate it. With 
that view I took some of the papers and making short hints of the senti- 
ments in each sentence, laid them by a few days, and then, without 
looking at the work, tried to complete the papers again, expressing each 
hinted sentiment at length, and as fully as it had been expressed before, 
in any suitable words that should occur to me. Then I compared my 
'Spectator' with the original, discovered some of my faults, and cor- 
rected them." To acquire a stock of words and a readiness in recollection 
and use of them, he ' ' took some of the tales in the ' Spectator ' and turned 
them into verse; and after a time when I had pretty well forgotten the 
prose, turned them back again." (Quoted by Bolton, F. E., Principles of 
Education, p. 421.) 

Brander Matthews in a more general way has pointed out the 
relative shares of imitation and originality. 

"Consciously or imconsciously every artist is a debtor to the past. 
The most original of innovators has made his originality partly out of 
himself, partly out of what he has appropriated and absorbed from those 
who practiced his art before him. Only a few of his separate contrivances 
are his own, and the most he can claim is a patent on the combination. 

"... The young artist is a weakling if he openly robs any single one 
of his predecessors; he is a dolt if he does not borrow from as many of 
them as may have the separate qualities he is striving to combine. 

"The arts are one in reality; and what is true of painting and sculpture 
and architecture is true also of literature, of prose and verse. For exam- 
ple, there are few men of letters of our time whose prose has been more 
praised for its freshness and its individuality than the late Robert Louis 
Stevenson; but his was an originahty compounded of many samples. 
He confessed frankly that he had sat at the feet of the masters, playing 
the 'sedulous ape' to a dozen or more, and at last slowly learning how to 
be himself. Again, the verse of Dante Gabriel Rossetti has a note of its 
own, a note which many younger poets have delighted to echo and re- 
echo; but Rossetti told a friend that the exciting cause of his 'Blessed 
Damozel' was the 'Raven' of Edgar Allan Poe, and Poe's own indebted- 
ness to Coleridge is obvious, even if it had not been expressly avowed." 

(c) Specific attention to frequently recurring grammatical ^ 
errors. Specific attention to, and special drill in, the precise func- 
tions to be developed have been shown over and over again by 
experiment and experience to be most directly efficacious in pro- 
ducing improvement. Several valuable studies have been made 
in recent years of the most common types of errors in oral and 



362 EDUCATIONAL PSYCHOI.OGY 

writttn language and of tin.- frequency of their occurrence. Char- 
ters and Miller ('15J niadt- a stud\' in cooperation with the teachers 
in schools in Kansiis City, by keipiiig a record for one week of all 
errors in oral language hiard by the teachers in and about the 
school, and by collecting all errors in the written work of the 
pupils for one month. A similar study was made by H. D. Fillers 
('17), at Ronham, Texas, based on results from 900 pupils in grades 
three to eight for written language, and in grades two to eight for 
oral language. 

A third study has been reported by Sears and Diebel ('16) from 
Cincinnati based on oral errors of 1,378 children of grades three to 
eight which were heard b}' the teachers during a jieriod of four 
days. 

A fourth investigation was made by Starch (unpublished) in 
which oral errors of pupils in grades one to eight were collected by 
the teachers, and oral errors of university students by a group of 
special students, and written errors were collected irom some 1,700 
themes obtained from eleven high schools. The high schools varied 
from very small ones to one having more than 1,000 students. The 
total number of oral errors collected was 2,9 16 from grade pu{)ils 
and 1,164 from university students. The total number of written 
errors collected from the high school pupils was 2,316, making a 
total of 6,396 errors. The results of all four studies are sum- 
marized in parallel columns in Tal)le loS. 

On the whole, the corresponding ])arts of aJl four studies agree 
surprisingly well, showing consideraljle reliability as well as striking 
similarity in the tjpes of errors found in dillerent parts of the 
country. 



LANGUAGE 



2>^2, 



TABLE loS 
Errors in language 



Written 


Oral 
















Sears 






Charters 


Fillers 


Starch 


Char- 


FlLL- 


& 


Starch 












ters 


ERS 


DiEBEL 




Nature of Error 


%0F 


% OF 


%0F 


%OF 


%OF 


%0F 


%0F 


%OF 


% OF Total 




Total 


First 


Total 


First 


Total 


Total 


Total 


Total 












21 Ru- 




21 Ru- 










Elem. 


Univ- 






brics 




brics 










Sch'ls 


ersity 


1. Subject of verb not 






















in nominative case. 






















Ex. Me and some 






















boys went 


0-f 


1 








0.2 


4 


5 




4.5 


1.1 


2. Predicate nomjna- 






















tive not in nomina- 






















tive case. 






















E.K. It was not him . . 


0-f 


1 








0.1 


2 





2 


0.7 


3.9 


3. Object of verb or 






















preposition not in ob- 






















jective case. 






















Ex. She appointed 






















Lynawood and I. . . . 


+ 


1 








,? 


1 


1 


o-f 


0.3 


5.9 


4. Wrong form of 






















noun or pronoun. 






















Ex. Theirself, hisself. 


5 


16 


7 


11 


.V9 


1 


I 


+ 


1,4 


0.7 


5. First personal pro- 






















noun standing first in 






















a series. 






















Ex. Me and him went 


n + 


1 


1 


1 


0,1 


2 


2 


5 


4.0 


0.0 


6. Failure of pronoun 






















to agree with its noun 






















in number and gender. 






















Ex. Nobody can do 






















what they like 


1 


4 


1 


1 


6.1 


+ 





+ 


0.1 


4.1 


7. Confusion of de- 






















monstrative adjective 






















and personal pronoun. 






















Ex. Them weeds, 






















them things, them 






















girls 


o-f 


+ 








0.0 


.S 


4 


4 


1.7 


02. 


8. Failure of verb to 






















agree with its subject 






















in person and number. 






















Ex. They was brought 


6 


19 


4 


7 


4.6 


14 


9 


6 


10.4 


11.4 


9. Confusion of past 






















tense and past par- 






















ticiple. 






















Ex. I seen him 


2 


5 


2 


4 


1.2 


24 


20 


19 


13.3 


5.9 


10. Confusion of past 






















and present tense. 






















Ex. They come along 






















and took mine 


4 


12 


12 


19 


0.4 


2 


5 


3 


6.0 


1.8 


11. Wrong tense form. 






















Ex. John's dog went 






















home 


2 


5 


2 


4 


0.5 


5 


i 


'•^ 


3.7 


0.9 


12. Wrong verb. 




Ex. Can we stay? . . . 


2 


7 


2 


4 


14.6 


12 


21 


14 


18.6 


15.5 


13. Incorrect use of 






















word. 






















Ex. If you was to be 






















tardy 


0- 


2 








5.1 


0-f 





+ 


0.0 


2.3 


14. Incorrect compar- 




ison of adjectives. 






















Ex. The winds are 






















much more cooler . . . 


0-f 


0-t- 








0.1 


1 





1 


0.6 




15. Confusion of com- 






















parative and super- 






















latives. 






















Ex. JoUiest (of two). . 


2 


6 





n 




+ 





0-f- 







3<^4 



EDUCATION/VL PSYCHOLOGY 
TABLE loS — Continued 



Wkittfs 


(»R\L 














>KARS 






CUAkTERS 


Fillers 


Starch 


Ch.\r- 

rt-Rs 


Fill- 
ers 


& 
DlEBEL 


SrAR<-n 


Natuke of Erbob 






















7o OK 


' I OK 


' , OK 


' [, OK 


ri OF 


' p OF 


'"„ op 


S"'" 


■"o OF Total 




Total 


liKsr 
21 Ri 


r«ir\L 


llRST 

21 ku- 


Total 


I'OTAL 


Total 


Total 






Elem. 


U.viv- 






BRRS 




URICS 










Sch'ls 


ersity 


16 Confusion of ad- 






















irdive and adverb. 
Kx. Will that there 










































il,,? 


+ 


o-t- 


,! 


6 


2-S 


1 


- 


- 


0.9 


4.1 


17. Misplaced modi- 




tier. 






















Ex. I only went once. 


2 


() 


2 


4 


2 N 









0.6 


3.2 


in. Double negative. 






















E\. I don't have no 






















pencil 


i 


1 


1) 


f) 


1 1 


M 


14 


12 


10 .? 


4 1 


l'> Confusion of prep- 




osition and conjunc- 






















tion 






















E\. lie looks like he is 






















rich 


+ 


1 


I 


1 


i . (< 


1 1 




1 






2(1. Syntactical re- 






















dundance. 






















Ex. Where's he at?. . 


4 


n 


^ 


7 


,V ! 


10 


10 


11 


Ifi f> 


<) 6 


2 1 Wrong part of 






















siKfch due to similar- 






















ity of sound. 






















Ex. I would be 






















known 


11 




If) 


25 


?.l 


1 





0-f- 





8 


22 lailure to put 






















Ixrio<l at end of sen- 






















tence 


30 




n 
















2i. Failure to put 


























of (]ueslion 


2 





















24. Failure to put 




apostrophe to denote 






















posse'ssion 


6 




2 
















2.S. Omission of sub- 






















ect 


^ 





















26. Omission of predi- 






















cate 


2 




















27. Confusion of dc- 






















pen<lent and indc- 






















p<ndint clause. ..... 







8 






































sentence with capital 






















Jctirr 






S 
















20. Use of wrong ar- 






















ticle 






















30. Failure to use quo- 






















talion marks 






















3\. Wrong word or 






















phrase. That for TItr 










10. ,1 








2.."^ 


S 8 


32 Words omitted 






















I l>ern to — 










10.1 








; 


1 7 


33. Miscellaneous.. . . 










7 .■; 








.' 2 





These studies show thit most of tlio errors in lanp;uap;e are con- 
fined to a small numl)er of types. Thus in Charter's tabulation, 
71% of ^'1 oral errors fall under five tN-i)es or j^ammatical rules, 
namely, No. 9, Confusion of Fust Tense and Past Participle, 24%; 



LANGUAGE 365 

No. S, Failure of Verb to Agree with its Subject in Number and 
Person, 14%; No. 12, Wrong verb, 12%; No. 18, Double Negative, 
11%; and No. 20, Syntactical Redundance, 10%. The same situa- 
tion obtains among written errors; 91% of all errors fall under ten 
classes or rules. 

There are several rather striking differences between the errors 
of written and oral composition. Written composition is much 
more apt to get the wrong form of noun or pronoun (16%, 11%) 
than oral (2%, 1%, o - %) and the same is true of confusion of the 
tenses (12%, 19%, vs. 2%, 5%, 3%). On the other hand, oral lan- 
guage is much more apt to confuse the past tense and past parti- 
ciple (24%, 20%, 19%, vs. 5%, 4%), to use the wrong verb (12%, 
21%, 14%, vs. 7%, 4%), and to use the double negative (11%, 

14%, I2%VS. I%,0%). 

Sears and Diebel also tabulated their material to show the 
relative proportions of different types of errors for the various 
grades. (Table 109.) 

TABLE 109 

Classification of Errors Grades 

3 4 5 6 7 8 Total 



Verbs 44.2 60.0 55.4 54. q 43.. 



49 



2. Pronouns 15.9 14.0 6.7 7.7 12.3 18.8 13 

3. Negatives 11. 5 7.1 20.2 7.2, 15.2 14.0 11 

4. Syntactical redundance ... . 8.0 6.6 11. 2 12.6 16.5 9.6 9 

5. Mispronunciations 14.7 7.8 2.2 4.9 1.7 8 

6. Prepositions 3.4 3.2 1.8 5.6 4.1 2.6 3 

7. Adjectives and adverbs ... . 2.0 0.6 2.2 6.6 8.2 4.8 3 

8. Ambiguous expressions .... 0.2 .2 

Percentages of errors in each grade due to each class of mistakes. The 
forty-one specific errors v^^hich Sears and Diebel found most frequent 
follow with their respective frequencies: — 

1 . haven't no for haven't any 233 

2. seen — had saw 180 

3. ain't for am not, isn't, aren't 124 

4. done 113 

5. got, ain't got, haven't got 112 

6. 1 and my brother 96 

7. kin, jist, git, kitch 91 

8. ain't for haven't, hasn't 89 

9. Frank and me for Frank and 1 80 

10. is for are 76 

11. them for those 75 

12. learn for teach 71 



366 EDUCATIONAL l'SMH()l,( >r,V 

13. can for may 60 

1 4. my mother, she 58 

15. got for receive, bciomc, grow, is 53 

16. that there 38 

1 7. don't for doesn't 36 

I S. It was me 36 

1 0. leave for let 34 

20. went for gone 32 

2 1 . come for came 31 

22. never gave 30 

23. by my aunt's 28 

24. drawod, throwed, growed, knowed 27 

25. somepin, for something 25 

2(1. broke for broken 22 

27. lay for lie 21 

28. make dinner for prepare, get 21 

2g. says for s^iid 20 

30. all two, all both iQ 

31. readin, nothin 18 

32. by us for near us 16 

33. he does it like she docs 15 

34. why, and, so, at the beginning of sentence or in middle of sentence. ... 15 

35. that, which, for who and whose 14 

36. onct 12 

37. in back of 12 

38. funny, lots, etc., for queer, many 11 

30. ct for ate 11 

40. run for ran 11 

4 1 . set for sit 10 

R. T. Johnson obtained samples of written e.x]")Osition, narration, 
and descrijition from 132 hip;h school freshmen and 66 college 
freshmen from the Kansas City Ilit^h Sch(Mil and Junior College in 
an effort to determine the persistence of errors in Avritten English. 
The high-school freshmen made 2,160 errors in 50,371 words, while 
the college freshmen made 7S7 errors in 32,603 words. Roughly 
this was 23 errors ])er thousand words for the former and 42 errors 
per thousiind words for the latter. The college freshmen are there- 
fore distinctly superior. The following table show.^ the relative 
improvement for the various t^qx-s of errors. Naturally the raw 
results of such an experiment somewhat exaggerated the improve- 
ment due to four years of training because of the eMmination of 
the poorer high school students before reaching college. In the 
last colimm of the table this factor has been compens;ited for. 



LANGUAGE 



367 



TABLE no 





Total 
NO. Er- 


Errors 


Per 






Per Cent 




of 66 


Cent 






Decrease 




College 


De- 


Rank 


Rank 


IN Errors of 




rors OF 

132 

High 

School 
Fresh- 
Men 


Fresh- 


crease 


IN 


IN 


College 


Types of Errors 


men Im- 


OF Er- 


Pre- 


Per- 


Freshmen 




CREASEU 


rors OF 


va- 


sis- 


after Elim- 




to Pro- 


College 


lence 


tency 


ination HAS 




portion 


Fresh- 






been Al- 




of 132 


Men 






lowed FOR 


1. Mistake in case of pronoun .... 


11 


3.0 


72.7 


14 


13 


71 


2. Other errors with pronoun 


102 


61.6 


40.0 


6 


7 


36 


3. Use of verb 


93 

52 


49.0 
43.0 


47.0 
17.3 


7 
9 


8 

1 


37 


4. Adjective and adverb 


- 1 1 (increase) 


5. Prepositions and conjunctions . 


SO 


37.0 


26.0 


10 


5 


21 


6. Ungrammatical sentence struc 














ture 


220 

46 

232 


49.0 
26.0 
154.8 


77.7 
43.5 
33.0 


8 
12 
3 


11 
4 
6 


74 


7. Unclear meaning 


17 


8. Mistakes in punctuation 


28.3 


9. Use of apostrophe 


150 


115.5 


23.0 


5 


3 


16.5 


10. Capitalization 


196 


184.8 


5.7 


2 


2 


-6.5 














(increase) 


11. Careless omission or repetition. . 


223 


118.6 


47.0 


4 


9 


43. 


12. Mistakes in spelling 


676 


320.3 


52.5 


1 


U 


47.4 


13. Quotation marks 


25 


13.8 


44.8 


13 


10 


46.3 


14. Miscellaneous errors 


85 


35.4 


58.3 


11 


12 


54. 



I have computed the correlation of the two series of ranks in 
the above table by the Spearman rank method which yields .26, 
Thjs means that there is slight connection between the prevalence 
and the persistency of errors, that is, if there is any tendency at all 
it is for the more common errors to be also more persistent. 

Such studies should prove exceedingly useful in helping the 
teacher to devise special exercises and drills designed to eliminate 
these errors and to substitute for them correct forms of expression. 

(4) Oral versus Written Practice in Composition. The processes 
of oral and written composition differ psychologically from each 
other in certain important respects; and the same is even more true 
of formal oral composition and everyday speech. Ordinary conver- 
sation consists for the most part of short verbal responses each last- 
ing but a few seconds, initiated by continuously recurring stimula- 
tions from the other individuals engaged. Formal oral composition 
on the other hand is a reaction many times as long with but a 
single formal impulse at the beginning. With this enormously 
increased length of reaction arise all the complex problems of 
structure and the accompanying strain on the attention, to say 
nothing of the powerful emotional inhibitions of thought and action 
resulting from self-consciousness and fear of failure, which prac- 
tically do not exist at all in ordinary spontaneous speech. 

Again, written composition dififers psychologically very much 



^68 EDUCATIONAL PSYCHOLOGY 

irom oral composition in jK-miitting much more deliberate action, 
if a satisfactory thought floes not present itself, there is no imper- 
ative necessity for immediate action at all costs. There are no 
distractions of the attention from a staring audience. Perhaps 
most difl'erent of all, written composition presents an op])ortunity 
for revision of structure, choice of material and words which oral 
comjwsition cannot permit. Even the vocabulan,- and sentence 
structure is different to a certain extent. 

It has already been pointed out in the chapter on transfer of train- 
ing that i)sychological reactions tend strongly to be limited closely 
to the conditions under which training takes place. The moral as 
to composition is obvious. Oral composition no less than written 
composition requires specific jjractice. 

Probably over nine-tenths of the composition of the average 
adult is oral and only a very small fraction is written. And yet 
the schools until recently have directed about nine-tenths of all 
sjjecific training in composition toward MTitten expression, either 
with the belief that training in writing would carry o\er directly 
to oral composition or with no realization at all of the in\portance 
of oral comjjosition. The whole ])rocess has disregarded the psy- 
chological law of association that associated bonds should be 
formed in the order and manner in which they are to be used. If 
language is to be spoken, the neural links via the motor sjx'ech 
centers should be specifically and correctly exercised. It is needless 
to ])oint out here the inestimable value of speaking and conversing 
in correct, convincing manner. Why should the schools not direct 
their efforts far more fully to specific training in oral com])osition? 
The reply might be made that all speaking in general and all re- 
citing in classes affords practice in oral composition. But the 
trouble is that most of it is done with little or no attention to cor- 
rect, well-organi/.ed ex])ression. Gross errors, to be sure, are 
pointed out to the child; but why should the school not assign a 
theme to each pu[)il in the English class on which he is to give a 
3 or 4 minute oral composition in correct, well-planned and thought- 
out form? A])ro])os of this point, a committee of The Illinois As- 
sociation of Teachers of English arranged 

"A course for the second semester of the ninth grcdc, which was to l>e 
taught in two ways; one class would have only written excrcisi\s; the 
other, a ronihinalion of two-thirds oral and one-lhird written. .\11 
classes taking either course were to be given the same written tests at 



LANGUAGE 36^ 

the beginning, at the middle, and at the end of the semester. All the 
papers written by each class, including these tests, were to be forwarded 
to the committee in charge of the experiment, accompanied by a report 
from the teacher, stating as accurately as possible how much time he 
spent in preparation, in conference, and in correcting papers, and also 
his opinion as to the results of the experiment. 

"The outcome was decidedly favorable to the use of oral composition. 
The sections taking the combined course were better at the end of the 
semester in thought-vigor, freedom and interest than the others; they were 
no worse in spelling and punctuation and better in handwriting — indeed, 
the writing sections showed marked degeneration in all matters of me- 
chanics. Over half of the 22 schools which carried out the experiment in 
full reported greater improvement in the combination sections, while only 
two reported less improvement." (Reported by Hosic, '15, pp. 9S-99.) 

(5) How Much Written Composition is Profitable? Written 
composition, even aside from consuming the time that might more 
profitably be used for oral composition, is probably overdone in 
many English classes. So much is required that it becomes a bore 
with little profit to be derived for the amount of time consumed in 
putting down upon paper a few trivial ideas about fictitious and 
worthless subjects. Professor Lounsbury and other teachers of 
English have begun to doubt the value of such extensive writing. 
He says: 

"I am by no means disposed to go so far as the historian of New Eng- 
land, John Gorham Palfrey, who as I have been told, was wont to ex- 
press the desire that an act of Congress should be passed forbidding on 
pain of death anyone under twenty-one years of age to write a sentence. 
Excess in one direction can not be remedied by excess in the opposite. 
Still, none the less am I thoroughly convinced that altogether undue 
importance is attached to exercises in English composition, especially 
compulsory exercises; that the benefits to be derived from the general 
practice in schools is vastly overrated; that the criticism of themes, 
even when it is fully competent, is in the majority of cases of little value 
to the recipient; that in a large number of instances the criticism is and 
must ever be more or less incompetent; and that when the corrections 
which are made inefiicienlly and unintelligently, as is too often the case, 
the results are distinctly more harmful than helpful." 

William Lyon Phelps, Professor of English Literature at Yale, 
has reached a similar conclusion: 

"On the subject of required English composition, I am a stout, un- 
abashed and thorough sceptic. And although the majority is still against 



370 EDUCATIONAL PSYCHOLOGY 

mc, I am in good company. Professor Child read and corrected themes 
at Har\'ard for about forty years: at the end of the lime it was his fervent 
bchef that not only was the work unprofitable to the student, but that 
in many cases it was injurious. That it is always injurious to the in- 
structor, when it is inlemperately indulged, is certain, \\hen I was an 
instructor at Harvard, 1 one day met Professor Child in the yard. He 
stO{>ped a moment and asketl me what kind of work 1 was doing. I xiid, 
'Reading themes.' He put his hand affectionately on my shoulder, and 
remarked with that wonderful smile of his, in which kindness was mingled 
with the regret of forty years, 'Don't sjxjil your youth.' Professor 
Wendell, wlio inherited the bondage under which his predecessor groaned, 
has never really believed in the etlicacy of the work. Professor Louns- 
bury of Vale has given valuable and powerful testimony against it. 
Professor Cook and Professor Beers — two quite diflerent types of men — 
are in this point in absolute agreement." (P. 117.) 

By way of concrete evidence for his opinion, Professor Phelps 
tells that after recjuiring only a moderate amount of theme ^\Titing 
during the freshman and sophomore years he submitted a batch 
of compositions written by his juniors to one of the Harvard pro- 
fessors. He read them carefully and testified that they were 
exactly as good technically as those done by Harvard juniors. He 
further says: 

"I know of nothing in the world that illustrates more beautifully the 
law of diminishing returns than required courses in conifjosition. .X 
class of sludLMls will never under any circumstances write live times as 
well by writing five themes as they will by writing one; but the reading 
and correcting of five themes require five times the effort on the part of 
the body of teachers." (P. 123.) 

It is evident, from the diversity of opinion and practice regarding 
the question, that a careful and extensive investigation should be 
made to determine if possible the optimum amount of writing to 
be required in I-Jiglish courses. The report of the Committee of 
the Modern Language Association of America ami the National 
Council of Teachers of English on the Cost and Labor of English 
Teaching (1Q13), based on returns from 552 English teachers in 
93 high schools and 345 English teachers in 96 colleges, states that 

"The amount actually written under present conditions averages for 
high schools 3S0 words a week tliroughout the year, and for colleges 6jo 
words a week. Ideal conditions would slightly increase these averages 
to alx)Ut 430 for hi^h schools and 6S0 for colleges, and would make ik)S- 
sible equal attention to oral and to written training." 



LANGUAGE 37 I 

These facts are interesting as indicative of the present practice; 
they are significant as indicative of the consensus of opinions con- 
cerning the ideal amount of writing to be done. No one of course 
knows what the ideal amount is nor where the region of diminish- 
ing returns is located. Opinions do not settle the question. Ex- 
perimental and statistical inquiries must be made to determine 
the actual results produced by various amounts of writing under 
varying conditions and to locate, if possible, the region of dimin- 
ishing returns. 

(6) Good English in All Oral and Written Work. One of the 
great counteracting forces to the influence of the teaching of com- 
position in English classes, particularly in high schools and univer- 
sities, is the utter disregard for proper use of language in work 
outside of the classes in English. Pupils wear their Sunday clothes 
in only one class and go shabby the rest of the time. Instruction 
in composition is effective only when it succeeds in making correct, 
elegant language a part of the automatic association processes of 
thinking. During nine-tenths of the time, the language habits 
are being mechanized in a slipshod manner. How can the instruc- 
tion in English classes prevail against the iron chains of habit 
formed during the rest of the day? 

In his classic chapter on habit, James says: "The second maxim 
is: Never suffer an exception to occur till the new habit is securely 
rooted in your life. Each lapse is like the fall of a ball of string 
which one is carefully winding up; a single slip undoes more than 
a great many turns will wind again." 

Pupils should be required to be as careful in all their oral and 
WTitten work as they are in their classes in English. The organiza- 
tion of schools should be so modified that the necessary cooperation 
between teachers of English and teachers of other subjects could be 
made possible. Why should not all written work in all other sub- 
jects have to pass muster before the teacher of English? While 
all teachers, besides teachers of English, are in theory supposed to 
watch carefully over the English used by their pupils, they do as a 
matter of fact pay very little attention to it, partly because they 
feel they are too busy teaching the content of their own subjects 
to be able to spare the time, and partly because many of them are 
not as competent to correct the English of their pupils as they 
should be. 

A powerful incentive to the pupil to use at all times the best 
language that he is capable of could be given by basing his English 



372 i:i)LCA'll().\AL PSYCHOLOGY 

mark to the extent of say one-third or one-fourth, upon his work in 
the English class, and to the extent of two-thirds or three-fourths 
upon the qualil}' and correctness of his English in other classes. 

(7) Practice in Expressing Really Important and Personally 
Vital Ideas. On the assunij)tion that thought and language are 
intimately related, that in fact for practical puq)oses idea and 
word or thought and language are substantially identical. It would 
seem to be a fair inference that the exercise in oral and written 
composition should be carried on in conjunction with topics con- 
cerning which the pupil really has ideas or concerning which, if he 
has none, it would be really worth while to acquire ideas. From 
a psychological and humanitarian standpoint, it seems almost 
a crime to ask a pupil to write several pages of words on a topic 
concerning which he has no ideas or in which he has not the slightest 
interest, or which has no vital imj)ortance to him or to anvbody 
else. Yet how often is such the easel It is not likely to be very 
stimulating to a ])upil to be asked to write 150 words with this 
phrase as the beginning of the first sentence: "Life to me 
seems .... "; to write a thffme on that veteran of topics "The 
Autobiography"; or on such subjects as "Why Go to Church?"' 
"Why go to College?" "What Picture Impressed me Most," 
"What I got out of Virgil's ^Eneid." 

Brown and Haggerty incidentally point out in their study that 
the pupils wrote distinctly better compositions on certain topics 
than on others. The most striking case was the topic of the sixth 
week, "How I Earned Some Money" which prwluced a rise in the 
curves to as high a point as was reached even at the end of the 
twelve weeks. The topic for the fifth week, "The Pleasures of 
Skating (or some other sport)" and for the seventh week, "The 
Right Kind of a Chum" produced distinct drojxs in the curves. 

Too much of the practice in theme writing has to do with fictitious 
situations and not with vital ideas. The average adult, unless he 
enters a profession of which literar>' work is a necessary part, 
rarely has occasion to write about iictitious or far-fetched subjects. 
His composition, both oral and written, has to do with problems 
which are vital to his life and concerning which he actually has 
ideas or concerning which, if he has none, it is very important for 
him to acfjuire ideas. His welfare, his business and professional 
success may depend u[)on the forcefiilness of his letter or upon the 
convincingness of his interview. Many adults testify to the effect 
that they obtained most of thiir training in organizing and ex- 



LANGUAGE 373 

pressing their thoughts in connection with their business, profes- 
sional, and vocational problems. 

Why should not the composition in school be centered around 
problems that are real to the pupil or that will be real to him? 
Why should not the written work in other school subjects be 
submitted as the work in English composition, and why should not 
the correction and development of language be centered around 
the expression of thoughts actually necessary in his school work? 
The engineering school in a western university severed its instruc- 
tion in English from the academic English department because of 
the purely formal drill that was provided for its students, and 
instead secured the services of a teacher, who was trained both in 
engineering and in English, to give instruction in English by using 
their writing, their laboratory notes and reports, their language work 
in the various courses as the chief basis of instruction in English. 

(8) Effect of Differences in Teaching Ability. The differences 
of the ability of teachers are probably as striking, if not more so, 
in the teaching of English as in any other branch. The differences 
in the average abilities of entire classes are very great. The fol- 
lowing table gives the median value of the compositions written 
on the same topic and under similar circumstances in the high 
schools in a certain county in Illinois. Each composition was rated 
by three or more judges according to the Hillegas-Thorndike scale. 





TABLE III 






High 


Number of 


Median Score in Composition 


School 


Pupils 


(HiLLEGAS Scale) 


I 


6 




40.5 


2 


77 




47-4 


3 


169 




47-7 


4 


15 




52.8 


5 


41 




52-9 


6 


113 




58-5 


7 






62.3 


8 


55 




63.0 


9 


261 




63.1 


10 


22 




77.2 



From this table it appears that the pupils in the best school 
wrote compositions of approximately twice as good quality as the 
pupils in the worst school. Furthermore there is, so far as these 
limited data go, no indication of connection between size of school 
and quality of composition. The probability is that the most potent 
factor in the situation was the teaching ability of the instructor. 



CHAPTER XX 

ARITHMETIC 

The Psychological Processes Involved in Arithmetical 
Operations 

An analysis of the psychological processes involved in arith- 
metical operations may be undertaken in two ways: cither we may 
analyze the mental activities concerned in a topical arithmetical 
operation as performed by an adult or by a practiced pupil, or we 
may trace the genetic development and combination of arithmet- 
ical concepts and processes in the child. The one would be a 
dissection of the linished product as carried out by the trained 
individual; the other would be a synthesis of the arithmetical 
elements as they arise in the mental growth of the child. Both 
methods of attack will lead essentially to the same set of elements. 
Let us i)ursue for the present the first method of apj^roach and let 
us take as an illustration a common everyday type of arithmetical 
computation. Suppose you purchase at a store, fruit for loc., a 
loaf of bread for loc, a pound of butter at 45c., and a bunch of 
celery for 7c. , and give the clerk a dollar bill, how much change 
should you receive? What mental processes are either involve<l 
or presupposed in arriving at the answer, 28c., or rather in the 
clerk's making the change up from 72 until he reaches one dollar? 
An analysis will .show at least the following steps: (i) The conce[)ts 
of numbers and their meaning, (:;) the ability (a) to hear, interpret 
and speak (if only by inner speech, which accompanies even the 
thinking of the numbers) the sounds for the numbers when the 
calculation is carried out mentally, or (b) to write and read the 
symbols for the numbers when the calculation is done with paper 
and pencil; (,0 ihe previously established mechanical assoiialions 
among numbers known as addition, subtraction, multiplication 
and division; (4) The fact that the various articles are put to- 
gether probably suggests the putting together of the numbers 
representing their value. That is, it acts as a stimulus to arouse 
the association process of a(lditii)n rather than subtraction «)r 
multiplication or ilivision: '•■",' the occurrence of the successive 

374 



ARITHMETIC 375 

links of the mechanical associations of the numbers leading to the 
sum and the continued association of counting by units (pennies) 
until ICO is reached. 

In brief the steps in the solution of a representative arithmetical 
problem would be as follows: 

(i) Number concepts previously acquired. 

(2) Ability to read, write or speak symbols for the numbers or 
objects to which they refer. The steps involved in these particular 
processes have been analyzed in the preceding chapters and hence 
need not be repeated here. 

(3) Numerous connections previously established between the 
various numbers, which are known as the fundamental operations 
with numbers, fractions, and so on. 

(4) A clue as to the particular connection to be made at succes- 
sive points. 

(5) The execution of the successive acts with the particular 
numbers involved. 

(i) The number concept. This is partly acquired by the child 
before he enters school. It arises out of, and develops through, 
the endless experiences of the child in contact with objects and 
repeated occurrences of events. Some investigators of the genesis 
of the number idea believe it to arise by means of counting; others 
believe it to arise by means of grouping. The probability is that 
both factors contribute. What happens is substantially this: The 
child deals with various objects and learns through handling them 
that they are separate things and that there are several of them. 
As he continues to handle them, he established definite meanings 
and associations with each one. He furthermore develops the 
idea of number or quantity by finding possibly that one or more 
may sometimes be missing. He thus acquires a group perception 
and gradually widens it to the abiUty to count, which essentially 
consists of associating certain sounds and certain motor speech 
processes in speaking the sounds, with successive objects or acts, 
probably both, as the successive objects are touched and handled 
or at least looked at in turn. These elementary number ideas are 
then enlarged rapidly upon entering school by providing more 
extensive materials and by associating larger numbers with 
them. 

(2) This step involves all the detailed and complicated elements 
enumerated in the reading and writing processes and as such need 
not be considered again as they contain no new elements. All the 



376 j;i)l ( ATIONAL PSYCHOLOGY 

other steps are s[)ccil'ically arithmetical processes and require de- 
tailed consideration. 

(3) The establishment of associations ajjiong the numbers. 
The processes involved in the four fundamental operations of 
adding, subtracting, multiplying and dividing are pure association 
processes supplemented by experiences with illustrative material 
to show the meaning of the various operations. Ullimately and 
fundamentally they are established in the neural and mental 
machinery as associative links which, in the trained individual, 
become purely automatic. Six plus 2 ecjuals 8, 6 minus 2 equals 4, 
6 times 2 etjuals 12, and divided by 2 ecjuals 3, are pure associa- 
tion processes so that when the two numbers with a certain symbol 
between them appear and are spoken in succession, the appro- 
I)riate last link is brought up thereby. 

(4) The clue as to which jirocess is to take i)lacc. The clement 
in a given situation or problem that suggests which of the four op- 
erations at a given i)oint shall take place is the clue which acts as a 
stimulus to arouse the appropriate associative reaction. In the 
illustration used, the fact of putting the articles together probably 
acts as the clue which starts the association of 10 plus 10 plus 45 
plus 7 equals 72. Step (4) in this problem is what we ordinarily 
call reasoning. But fundamentally, arithmetical reasoning, as well 
as so-called reasoning in general, is essentially a matter of the 
perception of ditTerenccs and similarities and a matter of .'^elective 
association processes which are largely automatic in the perfected 
stages of arithmetical skill and largely trial and error in the early 
stages. The child associates the putting together of objects with 
addition; the taking away of objects from a group with subtraction; 
the putting together of groups of ccjual size with multijilication; 
and the taking away of groups of e^iual size with division. This at 
first probably develops through the multiplication of objects in 
these various ways. Something in any given situation or problem 
suggests one or the other of the four operations. Through in- 
definitely repeated situations, the correct operation is probably 
suggested as a matter of association. In a more comjilex situation 
in which the clue does not oj^erate as automatically, the mind prtv 
ceeds largely by trial and error by bringing up in turn dilTerent 
associations until the loirett one arises, or until one arises that 
satisfies the circumstanns. 

In ordinary terminology, we call this process reasoning. It may 
seem as though we were reducing reasoning to mechanical associa- 



ARITHMETIC 377 

tion which naively we do not consider to be reasoning. Perhaps 
in a sense that is really what it amounts to. There probably is no 
such thing as reasoning in the sense of forcing one's thought proc- 
esses into a given desired direction in a straight and direct line by 
sheer force of will. Reasoning, even that of the most original 
and inventive type, probably consists fundamentally in starting 
with a certain idea, desire, or problem, in short, with a stimulus, 
and in waiting for associations to arise and then in following out 
in turn by trial and error, one link after another, and in waiting for 
each one to bring up its links until a chain of successful links arises 
which satisfies the desire or which meets no opposition and which 
is then selected. Probably all that voluntary effort will do is to 
stimulate possibly a more rapid arousal of associative links, and to 
stir up by virtue of stimulating greater neural activity, such addi- 
tional associations as ordinarily do not come up quite so easily. 
This is apparently what happens even in the inventive and most 
original type of thinking or reasoning. The statements of scientists 
and inventors bear witness to the error of the usual belief that 
original thinking goes straight to the goal with unerring step. Thus 
Jevons remarked: 

"In all probability the errors of the great mind exceed in number those 
of the less vigorous one. Fertility of imagination and abundance of 
guesses at truth are among the first requisites of discovery." 

And Faraday said: 

"The world little knows how many of the thoughts and theories which 
have passed through the mind of a scientific investigator have been 
crushed in silence and secrecy by his own severe criticism and adverse 
examination; that in the most successful instances not a tenth of the 
suggestions, the hopes, the wishes, the preliminary conclusions have 
been realized." (Quoted from Lindley.) 

Edison has described the invention of the electric lamp as follows : 

"During all those years of experimentation and research, I never 
once made a discovery. All my work was deductive, and the results I 
achieved were those of invention, pure and simple. I would construct 
a theory and work on its lines until I found it was untenable. Then 
it would be discarded at once and another theory evolved. This was 
the only possible way for me to work out the problem. ... I speak 
without exaggeration when I say that I have constructed 3,000 different 
theories in connection with the electric light, each one of them reasonable 



378 EDUCATIONAL PSYCHOLOGY 

and apparently likely to be true. Yet only in two cases did my experi- 
ments prove the truth of my theory. My chief dilVicully was in con- 
structing the carbon rilameiU. . . . Every cjuarter of the globe was 
rans^icked by my agents, and all starts of the queerest materials used, 
until finally the shred of bamboo, now utilized by us, was settled ujKjn." 
(G. C. Lathrop, '" Talks with Edison," Harpers, \'ol. So, p. 4-'5.) 

However, it must not be supposed that our analysis has more 
than sketched in its broadest outlines one of the most complex 
and difllcult mental activities which himian beings ordinarily per- 
form. Indeed arithmetic is psychologically not a single activity but 
a group of fairly distinct activities, some of which are as unrelated 
to each other as arithmetic as a whole is to English. Thorndike 
found that the average arithmetic grades of the children in a cer- 
tain school for a period of two and one-half years gave a correlation 
(Pearson) of .39 with the grades in English, and of .36 witii those 
in geography. On the other hand, Stone tested 500 children on the 
four fundamental processes and arithmetical reasoning and ob- 
tained the following correlations: 

Addition with subtraction 50 

Addition with multiplication 65 

Addition with division 56 

Subtraction with multi[)lication 89 

Subtraction with division 95 

Multi[)lication with cli\ision 95 

ArithnKticai rcasoninj; with addition 28 

Arithmetical reasoning with subtraction 32 

Arithmetical reasoning with multiplication 34 

Arithmetical reasoning with division 36 

These results are especially striking in their low correlation of 
the four fundamental processes with arithmetical reasoning. It 
is significant also that subtraction, which might be thought to be 
( losely related to addition, turns out to be much more closely re- 
lated to division, the correlations being .50 and .95 respectively. 
These results have added significance when it is remembered that 
all elements of subtraction or multiplication, for example, which 
are normally |)resent in division fundamentals were carefully 
eliminated from the scores. 

Our numerous previous observations of the specil'ic nature of 
psychological activities, however, so far from making us surprised 
at Stone's results would rather h-ad us to expect still greater dif- 
ferentiation of tlic arithmetical processes. Howell found with re- 



ARITHMETIC 379 

gard to the combination 12—8=4 3-^^ 12 — 4=8 that knowing 
the one by no means assured knowing the other. Table 112 shows 
the total number of errors of 300 children in grades three to eight 
on eight pairs of complementary subtraction combinations: 

TABLE 112 

Total Errors Total Errors 

12-4 14 12-7 17 

12-8 o 12-5 o 

13-5 8 iS-8 17 

13-8 IS 15-7 o 

11-6 10 1 1-9 15 

11-5 o 11-2 o 

II-7 22 » 17-9 12 

11-4 5 17-8 I 

Here there is evidently a strong tendency for the smaller of two 
amounts subtracted to yield a surprisingly greater degree of accu- 
racy than its complement. 

Now each of these numerous specialized arithmetical activities 
has a psychology more or less of its own. Most of these processes 
are as yet very imperfectly explored. The psychology of comple- 
mentary processes just considered offers an excellent example. 
Other important ones are concerned with borrowing, carrying, 
and decimal points. Perhaps one of the most significant of all is a 
peculiar type of attention, or memory span, such as is involved in 
accurately keeping in mind and adding one to the tens at appro- 
priate intervals while at the same time adding the units of a long 
column of digits; or, in subtracting, in which it is necessary to re- 
member when borrowing has taken place so as to make the appro- 
priate compensation. Among the specialized aspects of the psy- 
chology of arithmetic is the phenomenon of number preferences. 
The United States census reports show that when people are not 
certain of their ages they tend strongly to give even numbers and 
to avoid odd numbers. Phillips obtained introspections showing 
that people like the even numbers but feel uncomfortable at the 
thought of odd numbers. Jastrow in an unpublished investigation 
had university students estimate the number of steps of two short 
flights of stairs which all climbed nearly every day. The avoidance 
of the odd numbers and particularly of the prime numbers is very 



38o 



EDUCATIONAL PSYCHOLOGY 



strikingly shown by Figure 75 where the results of both sets of judg- 
ments are combined. The avoidance of a given odd number is 
shown by comj)aring its frequency with the frequency of the num- 
bers at each side of it. ICight is an especial fa\orite. 



28r 



eg 



3 4 5 6 7 8 



^>^ n 



y 10 11 12 i:i 11 15 16 17 18 I'J 20 
Estimates 



Fig. 75. — The number of persons whose estimates of the number of steps 
fall on various numbers showing a tendency to avoid odd and prime numbers. 




1 «' :j •• 5 ti 7 8 II 10 n 12 n n 15 10 17 

Totals in Addition 
Fir,. 76. — ShowinR the relative numbers of addition combinations amounting 
to the various totals, .\dapted from laMc uS. 



ARITHMETIC 38 1 

These number preferences are chiefly significant to education in 
their genesis. Some indications as to their origin are incidentally- 
furnished by two studies to be described presently, the results of 
which are represented graphically by Figures 76 and 80. Figure 
76 shows very clearly that there is a strong tendency for additions 
of digits which result in totals of odd numbers and especially of 
prime numbers, to be more difficult and thus to cause the learner 
more trouble than even-numbered totals. From quite a different 
source, namely, the difficulty of mastering number pictures (Fig- 
ure 80), there comes evidence tending toward the same conclusion. 
Indeed it is possible that early experiences of a disagreeable nature 
involving these odd and prime numbers may have caused the 
unpleasant feelings reported by Phillips' subjects to be associated 
with them, and that this unpleasant feeling-tone may have in- 
hibited to a certain extent the choice of odd numbers in Jastrow's 
experiment according to the well-known tendency for the un- 
pleasant to produce avoiding reactions. Moreover, the special 
difficulty encountered with prime numbers suggests that the diflfi- 
culty of the various totals, apart from the factor of size, is closely 
related in its turn to indirect assistance derived from multiplication. 

The Measurement of Efficiency in Arithmetical Operations 

According to our analysis, the measurement of efficiency in arith- 
metical operations consists fundamentally in a measurement of 
the quickness and correctness with which the various operations 
are suggested and carried out under different conditions. Two 
types of measuring devices have been worked out. The first type 
consists of the well-known Courtis ('10) tests. Courtis prepared at 
the outset a series of eight tests now widely used, known as Series A, 
to measure the following operations: Addition, subtraction, mul- 
tiplication, division, copying figures, reasoning without performance 
of the operations, fundamentals consisting of various combinations 
of the four fundamental operations, and reasoning, including the 
performance of the operations.' The general principle of the tests 
consists in the selection of units of the material in each test of 
approximately the same type and difficulty throughout the test 
and of measuring efficiency by the number of operations attempted 
or done correctly in a given period of time. For example, the 
addition test is composed of a large number of combinations of 
single digits, such as 6 plus 3, 7 plus i, etc. The measure of 



382 EDUCATIONAL PSVCHOLOCiY 

efficiency is the number of such additions made in one minute. 
Each of the other tests is constructed on the same general princi- 
ple. The reasoning tests consist of separate problems of approxi- 
mately equal length and difficulty, and the measure of efficiency 
is the number of problems attempted or solved correctly in a 
given number of minutes. 

More recently Courtis ])repared another set, Series B, which is 
confined to the four fundamental operations and contains conse- 
quently one test for addition, subtraction, mulli[)lication and 
division. The chief ditTerence between the tests in Series B and 
the corresponding ones in Series A is that the problems in the 
former contain larger numbers arranged in larger columns so as to 
introduce more complex elements. 

The second general type of tests is constructed on the princii)le 
of a scale of problems of increasing difficulty. The author ('15) 
prepared a test for measuring ability in solving concrete prob- 
lems, designated as Arithmetical Scale A. It is composed of a 
series of problems increasing in difficulty by determined steps of 
difficulty ranging from o to 15. Ability is measured not in terms 
of speed, but in terms of the highest j)oint on tlie scale at which a 
pupil can solve problems correctly. The difficulty of the problem 
and the distances between the steps were determined by extensive 
experiments with pupils. 

Stone ('oS) prepared a set of problems whose difficulty was deter- 
mined experimentally. The problems are, however, not arranged 
in scale form. 

Woody ('16) ])rei)ared a series of tests on the scale plan for 
measuring ability chielly in the fundamental ofXTalions. Each 
scale is composed of a series of problems arranged in the order of 
increasing difficulty. 

Judd and Counts ('16) prepared for the survey of the Cleveland 
schools, fifteen sets of arithmetic tests, four for addition, two for 
subtraction, three for multiplication, four for division, and two 
for operations with fractions. The various sets for any one t>pe 
of operation represented successively more and more complex 
stages of the operation. 

Standards of attainment have bein prejiared by the authors of 
the various tests so that measurements of the abilities of pupils 
and schools can be compared directly with the respective norms for 
the various grades. These tests have been very useful in getting at 
some real facts concerning progress and attainment in arithmetic. 



ARITHMETIC 



383 



Figure 19, Chapter III, shows the range of individual abiUties 
and the overlapping of abiUties in the various grades in a certain 
school as measured by the author's arithmetical scale. The facts 
shown therein, although indicating enormous ranges of differences 
and overlappings, are substantially the same as those found in 
other subjects and hence are no longer surprising. Figure 20 ex- 
hibits the same facts for addition as measured by the Courtis tests. 




5 6 7 8 

Grade 

Fig. 77. — Difference between boys and girls in solving arithmetical problems 
as measured by the author's scale ( '15 ). 

Figure 77 shows the difference between the sexes and indicates 
that in arithmetical reasoning the boys surpass the girls. This is 
one of the few school abilities in which the boys are in the lead. 



Economical Methods of Learning Arithmetical Operations 

(i) The Acquisition of the Number Concept. The fundamental 
psychological materials which the child must acquire are ideas of 
numbers and quantities. He must learn what i, 2, 3, 4, etc., mean. 
The practical question is, How may the pupil acquire these number 
concepts most economically? A considerable amount of experi- 
mental work has been done which bears either directly or indirectly 
on the processes by which the child develops definite notions of 
the significance of number and quantity. The investigations on 
the span of attention or apprehension as measured in terms of the 
number of objects that can be grasped simultaneously have chiefly 



384 EDUCATIONAL I'SVCHOLOCV 

an indirect bearing. Such researches have been made by Cattell 
('S6), Messenger ('03), liurnelt C'oO), Dietze ('85), Warren ('97), 
Nanu ('04), Freeman, and others. The general result of these 
researches has been fo show that the average person has a s])an of 
four, five or si.\' objects or stimuli presented either simultaneously 
or in rapid succession to the sense of vision, hearing or touch. 

Investigations bearing more directly upon the development of 
number concepts have been carried out by Lay ('98 and '07), 
Walsemann, Knilling, and others. 

Pestaloz/.i was [probably the first to attack the problem by 
devising for the purpose of school exercises his Slrichlabellen (stroke- 

' I II III nil mil mill iiiini iiiiiiii niiiiin muilir''' 



^ o 00 000 0000 00000 000000 0000000 00000000 000000000 oooooooooo 
!1 ° ° °o 05 000 000 0000 0000 00000 ooo-~ 

000 



° O 00 000000 00000000 onn^,, IBOmJ 



4 o a 0000 000 coo 0000 0000 Qoooo oaooo, 

" ° 00 00 000 000 0000 0000 0000 o^^*^ 

500 0000 000 000 000 000 000 000 

o " 00 00 000 000 000 000 000 



(JO 00 00 oooo 00 o o o o e o o o 

o eo 00 00 

o oooooooooo iBeotz 1 

o oooo oooooooooo 

Fig. 7S. — Number pictures or arrangements. After Howell ip. 154). 

tables) which were the forerunners of more recent number pic- 
tures. These have been devised in two general forms: (a) those 
in which the dots are arranged in two parallel horizontal rows 
(j)repared by Born, Bussc, and Behmc) and (b) those in which the 
dots are arranged vertically (prepared by Hcntschel, Beetz, So- 
belewsky, and Kaselitz). See l-igure 78. 

Extensive experiments ha\e been made u|X)n school children by 
Lay to determine the relative merits of the various devices ami 
arrangements of the materials used for teaching apprehension of 
the number of objects. Lay investigated such problems as whether 
it is better for teaching numbers to present groujjs of stimuli or 
objects simultaneously or successively; whether it is better to 
present them in single rows or in double rows, in continuous rows 
or in quadrats (groujjs of four); how many objects children can 



ARITHMETIC 



385 



apprehend at one time; what the effect is of distance between the 
objects; and so on. 

The chief results have been summarized by Howell in the follow- 
ing Table 113, and graph, Figure 79. 





71 














































































































— 






{ 






















































\ 


























X - Grade 

7/ = % Mistakes 
















— 






\ 








































u 




\ 


















































— 








: 






















































\ 
















































— 








\ 
















































S 






\ 














































— 


— 








\ 


s 












































~ 








-N. 


\ 










































— 


— 














^ 


-— 






























































S. 


S 




























— 





































■^ 


-— 


— - 


— 


— 


■ — 












1 


X 


IB 

1 


2 


a" 


2B 

1 


3 


1 


3B 

i 


4 


1 


4B 


5\ 


6}. 


'f^ 


8^ 


X 





Fig. 79. — Curve of error for the number pictures 5 to 12 as a whole. After 
Howell ('14, p. 213*. 

TABLE 113. After Howell 

Comparison of the Born, Beetz, and Lay pictures with one another and with 

the Russian Machine in the apprehension of the numbers 5 to 10. 

P= chances for mistakes. M = number of mistakes. 



Numbers 


Lay 

Quadratic 

Pictures 

First-year 

Pupn-s 


Born 

First-year 

Pdpils 


Wendlino 

App. (Born) 

First-year 

Pupils 


Beetz 

First-year 

Pupils 


Russian 

Machine 

First-year 

Pupils 


c 


P 

87 
87 
87 
87 
87 
87 


M 

22 
16 
35 
23 
58 
50 


r/ 

/c 

25 
18 
40 
26 
67 
57 


P 

106 
106 
95 
95 
95 
82 


M 

25 
18 
38 
23 
58 
39 


% 
24 
17 
40 

24 
61 
48 


P 

106 

53 

53 

53 

106 

53 


M 

12 

5 
II 

13 
37 

25 


% 
II 
10 
20 
25 
7,i 
49 


P 

58 
58 
58 
58 
58 
58 


M 

29 
12 

42 
49 
46 

50 


/o 
50 
20 
72 
84 
80 
86 


P 

106 

Si 

53 

106 

53 


M 

26 
^5 
24 
3^ 
28 

17 


/o 
25 


6 


^0 


7 


45 


8 


40 





36 


10 


32 






Total 


522 
12,422 


204 
173 


40 
7 


570 
1,669 


201 

135 


35 
8 


424 
1,011 


103 
59 


24 
6 


348 


228 




424 


141 




Total Training 
School 




Coml)ine(l 


2,944 


377 


13 


2,248 


336 


15 



















386 



EDUCATIONAL rSYCIIOLOGY 



The outcome of these experiments is that the quadratic pictures 
are superior to other known pictures, particularly to the rows, and 
that the Lay anvl Born pictures are practically ecjual in merit. 

Howell continued exi)erinientation with quadratic number 
pictures similar to Lay's. The groups of dots used on the pictures 
ranged in size from three to twelve. Each card was exposed for 
5 seconds. The investigation was carried out with pupils in an 
elementary school. Howell's results are shown in the curve of 
Figure 79. This curve shows a rapid drop in errors from the first 
grade to the second, and that pui)ils reach a high degree of cer- 
tainty between the third and fourth grade in apprehending the 
numbers. Howell also found that certain numbers are not apprc- 



■ - 


y 










LI. 1.1 1 1 






















\ 










X - Number Pictures 
y - Mistakes 






































/ 


V 












^ 






























/ 


\ 






































i 


f 


> 


\ 

























/ 


\ 












/ 






\ 




















/ 


* 


N 


\ 




/ 


■^ 


^ 


1 






\ 






















/ 






\ 


/ 











































































{ 


1 


i 




' 




{ 




{ 


) 


1 





1 


1 


1 


2 


X 












1 


1 










1 











Fig. So. — Xumhcr of mistakes in iiercciving niiml)crs l)y pujiils in various 
grades. After IIowcU ( '14, p. 217). 

ht'iidcd as accurately as others, the most difficult ones being seven 
and elevrn, as sliown in LiL^urc So. 

(2) The Operations to be Learned. One of the lirst important 
problems in the economy of learning arithmetic, as in every school 
subject, is the determination of the operations that pupils should 
really master. Considerable attentit)n has lately been given to 
this Cjuestion with the result that the recent texts and courses of 
study have eliminated much material and have decidedly shifted 
the emphasis from certain topics to others. 

Two methods of attack may be employed in determining the 
topics and the material to be learned in arithmetic. According to 
the one method, we might obtain the consensus of many expert 
opinions as to the topics that .should be included and the relative 
amount of emphasis on each. According to the other method, we 
might procft'd to gatluT a large number of the arithmetical prob- 
lems and operations actually involved in the occupations and 



ARITHMETIC 



387 



professions of all classes of people, and to determine on the basis 
of such a collection, the sort of material that should be taught. 
Some results have been obtained by both methods. 




aO 40 50 tJO 7U 
Per cent 



90 100 



Fig. 81. — Percentage of superintendents in 830 cities who favor elimination 
of the various topics represented by checked surface and those who favor "less 
attention" are represented by shaded surface. After Jessup and Cofifman ('15). 



3SS 



EDUCATIONAL PSVCHOLOGV 



Jessup ('15) and CofTman obtained an expression of opinion 
from superintendents of 830 cities with a population of 4,000 or 
over as to which topics should be eliminated, which should receive 
less emphasis, and which should receive more emi)hasis than they 
do at present. Their results arc shown in Figure Si and Figure S2. 




lU 



L>o ^ 40 Jo 00 70 »o yu 100 

Per cent 



Fin. 9,2. — IVrcLnlaRe of superintendents in 830 cities who favor giving more 
attention to each of these tui)irs. After Jessup and Coffman ('15). 

As a result of these returns Jessup and (\i(Tman recommend the 
elimination of the following topics from the elementary course 
of study: 

"Apothecaries' weight, alligation, aliquot parts, annual interest, 
cube root, cases in percentage, compound and complex fractions of more 
than two digits, comi>ound proportion, dram, foreign money, folding 



ARITHMETIC 



)8g 



paper, the long method of greatest common divisor, longitude and time, 
least common multiple, metric system, progression, quarter in avoirdu- 
pois table, reduction of more than two steps, troy weight, true discount, 
unreal fractions." 

Furthermore, they recommend greater attention to such topics 
as: 

"Time saved through the omission of the material mentioned in the 
foregoing may be wisely devoted to the study of social, economic and 
arithmetical issues involved in such facts as saving and loaning money, 
taxation, public expenditure, banking, borrowing, building and loan 
associations, investments, bonds and stocks, tax levies, insurance, profits, 
pubhc utihties, and the like." 

G. M. Wilson ('17) collected 5,036 problems from 1,457 persons, 
representing practically all varieties of occupations and profes- 
sions. He then classified these problems according to the type of 
operation involved, and the number of problems of each type as 
shown in Table 1 14. 



TABLE 
Addition i-Pl 

2-PI 

3-Pl 

4-Pl 

Over 4-PI 

Total 

Multiplication i-Pl 

2-Pl 

3-Pl 

4-Pl 

Over 4-PI 

Total 

Subtraction i-Pl 

2-Pl 

3-Pl 

4-PI 

Over 4-PI 

Total 



114. After Wilson ('17) 

30 Accounts 251 

706 Addition of Fractions 3 

748 Amount 11 

193 Area i 

65 Average Weight 14 

1742 Banking 18 

Board Measure 12 

Cancellation 26 

1660 Capacity 10 

904 Circular Measure i 

195 Cubic Measure 56 

1 7 Debts 56 

3 Decimals 4 

2779 Discount 5 

Division of Fractions i 

40 Dry Measure 5 

407 Exchange 5 

406 Insurance 10 

167 Interest 66 

65 Liquid Measure 14 

1085 



39° 



EDUCAIK ).\ AI. I'SVCHOLOGY 



TABLE 114 — Continued 



Division. 



Over. 
Total . 

Fractions. 



i-Pl 
2-ri 
3-Pl 
4-Pl 

.4-ri 



10 plus 
I plus 

i-S 
I plus 

5 plus 



■ Total 

United States Money i-Pl 

2-Pl 

3-PI 

4-Pl 

Over 4-PI 

Total 



3.^4 

3iy 

I 21 

48 

5 

839 



534 



86 



60 

1035 

23 

2982 

1714 

550 

247 

5516 



MakinR Change 3 

Miasurin}; 21 

PcrccntaKC 217 

Plastering 2 

Practical Measurement 70 

Profit and Lt>s.s 16 

ProjKjrtion 5 

Receipts i 

Square Measure :■} 

Taxes fj 

Time Measure 13 

Buying 3128 

Selling 646 



"The problems solved in actual usapc are brief and simple. They 
chiefly require the more fundameiUal and more easily maslered proc- 
esses. 

"In actual usage, few problems of an abstract nature are encountered. 
The problems arc concrete and relate to business situations. They 
require simple reasoning and a decision as to the processes to be em- 
ployed. 

"The study justifies careful consideration of the following question: 
After the development of reasonable sf)eed and accuracy in the funda- 
mentals an«l the mastery of the simple and more useful arithmetical 
processes, should the arithmetic work not be centered largely around 
those i)roblems which furnish the basis for much business information?" 
(Wilson.) 



ARITHMETIC 391 

W. S. Monroe ('17) has compiled the problems in four text- 
books and classified them according to the types of operation 
involved and then compared the frequency of these types of prob- 
lems with the number of workers in the different occupations. 
His preliminary report states: 

"In the first place, out of a total of 1,023 types of practical problems 
found in four text-books, 720, or 71%, occur in occupational activities. 

"A study of the frequency with which type problems occur reveals 
a significant fact; viz., the frequency ranges from one to 434. 

"This wide variation in frequency shows that the authors of our text- 
books are far from being in agreement on the type problems of arithmetic. 
Only one author out of four has recognized 511 out of 1,023 type problems 
and 140 type problems have received the recognition of only two authors 
out of four." (Monroe.) 

(3) Length of the Class Period. J. M. Rice made an investiga- 
tion of efficiency in arithmetic after the general plan of his investi- 
gation of spelling. He tested some 6,000 pupils in eighteen dif- 
ferent schools in seven cities. His results are exhibited in Table 
115, which 

"Gives two averages for each grade as well as for each school as a 
whole. Thus, the school at the top shows averages 80.0 and 83.1, and 
the one at the bottom, 25.3 and 31.5. The first represents the percentage 
of answers which were absolutely correct; the second shows what per 
cent of the problems were correct in principle, i. e., the average that 
would have been received if no mechanical errors had been made. The 
difference represents the percentage of mechanical errors, which, I be- 
lieve, in most instances, makes a surprisingly small appearance." 



392 



KDUCATION'AL rSVflKJLOtiV 



Pi 



•^ o -t - - 



"^ in -t t^ r^ t>. 



■^•t-Ov'SO ^< •^©■oo 0>'^'-« '^5'-' lo'^ <N>o 



"-,-/:> •- -o "j C O f^oo >o r^ O •"< -• zo ►" -t 



'o~imO " 'OO*^'^ 



O fO'^O^'toO t^'N Por^O^iN ro*"' O 



r^ i^ . lO o 'O "I lo O>Ci0 in <o ►" O 00 f^ 't' i-i 



« « « rO ■* ■-' 



»-tr4r^^MMr4rv)M 



-'t -*» -^ -^ 



;juuuuuuuuuuuuuuu(ju 



o « „- 



O (J c 
C t- § 
O 4J 3 



ARITHMETIC 



393 



1 


z z 




w u 




« :^ 




^hi 




. >■ 




Sri 








•^o 




I^ ir, 








fa ij 




o < 








^" 




H 




i-l 




S p 


■a 


t s 


3 


w S 


C 


> Oh 




< 


c 




o 

u 










yi U 


u 


« 










o 




s 


iJ 


43 


fc 






C 


S Ai 




>< 


tn 


n H 




H J 


o 


OO p 


_G 




i^ 


PS 






ni 








-o 


u 

h) 


> 


& 






-O 


g 


.h 




o 






>l 


en 


m a 










rt 


M 


<u 


P<i 


> 




< 






•J 




O 




o 




R 




Ol 



< < <i < C-," Ph' Ph < &h' < A^' dl < Cl^ Oh < fC C^ 



+ 



r<^ ■^ r^O oo t~» 1^ uooo O •-< O OO lo m O C^ 



M ro w C< POOO O^ '^00 vO ^» ^0 OO >0 0\0 lO 



O v3 cooO ro t^ lO M O^ looo O^ O lo O vo i/i ro 






r^ 0^ l^ >'^ »O00 M\0 voiot-^w OvN ro 



rO'M<7>'^o0t~»f^'-' •vO<N uno tN m lo ^ »-< 



cs lo vovO vO O^ ^ 



r< M tNioioiow O O^ 






MMNC«^«^MPIMMrO'^HMCSC<0HMrO 



!>>>>, 



>>>>>>> 
1 G u u u "0 u G G G u u u u u G u u 



. 00 



ir> ro '*■ 



'3 S 



2 a 



i-ii-i "^ 



rt c JJ 



394 EDUCATIONAL PSVCIloLUGY 

With reference to the factor of lime in relation to efficiency in 
arithmetic, Rice concludes thus: 

"A glance at the figures will tell us at once that there is no direct 
relation between lime antl result; that special pressure tloes not ncccs- 
s;irily lead to success, and, conversely, that lack of i)ressure does not 
necessarily mean failure. 

"In the first place, it is interesting to note that the amount of time 
devoted to arithmetic in the school that obtained the lowest average — 
25% — was practically the same as it was in the one where the highest 
average — So% — was obtained. In the former the regular time for arith- 
metic in all the grailes was forty-five minutes a day, but some additional 
time was given. In the latter the time varied in the dilTerent classes, 
but averaged fifty-three minutes daily. This shows an extreme variation 
in results under the s;mie appropriation of time. 

"Looking again toward the bottom of the list, we find three schools 
with an average of 36%. In one of these, insuflicient pressure might be 
suggested as a reason for the unsatisfactory results, only thirty minutes 
daily having been devoted to arithmetic. The second school, however, 
gave forty-eight, while the third gave seventy-five. This certaiidy seems 
lo indicate that a radical defect in the quality of instruction can not be 
offset by an increase in quantity. 

"If we now turn our attention from the three schools just mentioned 
and direct it to three near the lop — Schools 2, 3 and 4, City I — we find 
the conditions reversed; for while the two schools that gave forty-five 
minutes made averages of 64% and 67'^f., res[)ectively, the school that 
gave only twenty-five minutes succeeded in obtaining an average of 60%. 
This would appear to indicate that while, on the one hand, nothing is 
gained by an increase of time where the instruction in arithmetic is 
faulty, on the other hand, nothing is lost by a decrease of time, to a 
certain jKjinl, where the schools are on the right path in teaching the 
subject. I'erhaps the most interesting feature of the table is the fact that 
tlie schoul giving twenty-five minutes a day came out within two of the 
toj), while the school giving seventy-five minutes daily came out prac- 
tically within one of the bottom." 

Stone (08) made a similar investigation, testing some 6,oco 
pupils in the 6th grade in twenty-six school systems. lie reports 
results practically identical with those of Rice, namely, that while 
the amount of time devoted to arithmetic in dilTerent schools 
varied from 7^^, to 22^/( of t'li' \ol:i\ school time, yet a comixirison 
of time expenditure with the efTiciency attained .showed, according 
to his interpretation, that time plays a negligible jiart. 

These results and inferences are interesting and valuable but 



ARITHMETIC 395 

they cannot be interpreted with absolute assurance. The various 
factors cooperating or counteracting are so intricate that a more 
careful isolation of the effect of the time element is necessary. In 
general, the same criticism made in connection with Rice's and 
Cornman's investigations of spelling applies here. Dependable 
conclusions could be reached only by an experimental procedure 
similar to the one there suggested. 

The findings of Rice and Stone probably represent correctly 
the situations in the schools examined. A possible explanation of 
the fact that the schools giving more time to arithmetic did not 
obtain on the whole any higher efficiency than those devoting 
less time to it may, perhaps, be sought in the likelihood that the 
schools giving longer periods of time may not have worked as 
intensively and used their time to as good advantage as the schools 
devoting less time. 

(4) The Effect of Various Environmental Factors. Both Rice 
and Stone massed their results with reference to ascertaining the 
effect of such factors as the home environment of the pupils, size 
of classes, age of pupils, the time of day of the test, amount of 
home-work required of the pupils, method of teaching, teaching 
abihty, the course of study, the superintendent's training of the 
teachers, etc. Rice reports that none of the factors had any influ- 
ential part in producing efficiency in arithmetic. The results are 
open to the same criticism of complication of factors as were pointed 
out previously. It seems quite improbable that these elements 
played no part. It is rather a question of more rigorous isolation of 
the effect of different factors. Stone, for example, found that the 
correlation of excellence in the course of study, as rated by judges, 
with efficiency in arithmetical reasoning was .56, and with effi- 
ciency in fundamentals .13. 

That environmental factors, and perhaps particularly method 
and spirit of teaching, do make important differences in the attain- 
ments of pupils is shown clearly in such results as those exhibited 
in Table 116 which gives the distribution of class averages of the 
grades in sixteen different schools as measured by the author's 
Arithmetical Scale A. 



396 KDUCATIONAL PSYCHOLOGY 

TAI51J-: 116 

Average scores attained in various schools as measured by Arithmetical Scale A 

(Starch) 

1 ■ — 

Grades 3 4 5 6 7 8 

City A 9.7 

B School I 13. 1 

2 7.2 10.4 10.6 II . 2 

C School 1 5.1 5.9 y. 2 9.2 

2 3-9 5-6 6.9 

3 3-9 S-3 5-6 7-5 9-2 126 

G School 1 90 10.9 II. 6 14.5 

2 8.9 12.0 13.0 13.7 

3 7-5 IO-2 9.2 10.9 IIS 

4 lo.o 10.6 II .3 

I School I 5.1 6.0 

2 6.2 

3 10. o 10.2 I I o 

L School I 4.6 5.8 S.5 9.8 II. 9 140 

2 46 7.4 

3 6.0 8.5 II. 3 

4 6.6 8.8 

Thus \vc note that the best eighth grade attained an average of 
14.5 as compared with the poorest one which attained an average 
of only 9.7. Such dilTerences would not be surprising if they 
were the scores of individual pupils. They are, however, the mean 
scores of whole classes. It is quite unlikely that the hereditary 
dilTerences of the groups as wholes dilTer so much from one another. 
It seems quite probable that the environmental circumstances, 
and chief among them the teacher and the attitude of the learner, 
were mainly responsible for the ultimate dififerences in achieve- 
ment. 

Similar results have been reported by Judd for the fifth and 
eighth grades in ninety schools in Cleveland, as measured by his 
Test A in simple addition. Figure 83. The best fifth grade made 
an average score nearly three times as high as the poorest fifth 
grade, and the best eighth grade made a score nearly twice as 
high as the poorest eighth grade. 

(5) Drill in Fundamental Operations. Various methods of 
drill in the fundanuiital operations have Inrn devised. 

Studebakcr, Assistant Su|)erintendent of Schools at Des Moines, 
has preparer! a scries of drill cards. The various combinations of 
numbers in fundamental operations are given on one side of the 



ARITHMETIC 



397 



card. Below each example there is an opening through the card 
in which the pupil may write his answer on the sheet of paper 
placed underneath the card. The pupil works as rapidly as he can 
within a certain limit of time. Then he turns the card over and 
places it again over the sheet of paper containing his answers so 





21 

21 
21 
21 




23 
23 
23 




Fifth Grades 






21 


22 


23 






21 


22 


23 






21 


22 


23 








21 


22 


23 






20 


21 


22 


23 








20 


21 


22 


23 








19 


20 


21 


22 


23 




26 










18 


19 


20 


21 


22 


23 


26 


27 








18 


19 


20 


21 


22 


23 






26 


27 








18 


19 


20 


21 


22 


23 


24 




26 


27 


28 






16 


17 


18 


19 


20 


21 


22 


23 


24 


26 


27 


28 


40 


|14| 


16 


17 


18 


19 


20 


21 


22 


23 


24 


25 


26 


27 


28 


1301 1 1 lail 1 1 1 


38| 40 1 



22 



22 



23 



25 



25 



25 



26 



26 



26 



27 



27 



27 



28 



28 



28 



Eighth Grades 



29 



30 



30 



33 



33 



34 



35|36|37|38|39|40|41| 



Fig. 83.— Median scores of the sth grades and of the 8th grades in 90 schools 
in simple addition. After Judd ('16, p. 112J. 

that they can be seen in the openings of the card and compared 
with the correct answers printed on that side of the card. 

This plan of drill work has a number of advantages, such as an 
incentive to rapid and accurate work, immediate self-checking of 
the answers, and so on. 



398 KDUCATIONAL PSYCHULOCN 

A considerable number of careful experimental studies on the 
influence of drill are now available and wiihoul exception they 
show drill to be distinctly valuable. Thorndike had nineteen 
university students practice adding 48 lo-digit columns of figures 
daily for seven days. While the work required on the average less 
than an hour in all, there was an improvement of 29% over the 
original rate. 

J. C. Brown performed two elaborate comparative experiments 
to determine whether children under controlled school conditions 
profit more by giving a small part of each class period to drill or 
by spending the entire period in ordinary routine work in arith- 
metic. In each experiment the children were first tested with the 
Stone Arithmetic tests and then divided into two groups of equal 
ability on the basis of their performance in the tests. One group 
was given the special drill as a part of the regular class work while 
the other did the class work as usual. At the conclusion of the 
drill, both were tested again by the Stone tests to see wliich had 
made the greater gain. In the first experiment 51 children from 
the sixth, seventh, and eighth grades were used. They averaged 
thirteen and one-half years of age. Drill on the four fundamental 
operations was given to one-half of the group for the first five 
minutes of each class period of twenty-five minutes. About half 
the drill was oral and half was written. The drill lasted thirty 
periods. In the second experiment 222 children were used and the 
drill was given for twenty periods. The results of the two experi- 
ments are given in parallel columns in Table 117. In each case 
section I received the drill and section II received the regular class 
work. The pupils did not know that any experiment was in 
progress. 



ARITHMETIC 



399 



TABLE 117 



Section 


Per Cent of Improvement of Second 
Test over First in 


Per Cent 
First Experi- 
ment 
(51 Children) 


Per Cent 
Second Experi- 
ment 
(222 Children) 




Number of problems worked 
Fundamentals, Addition 
Fundamentals, Subtraction 
Fundamentals, Multiplication 

Fundamentals, Division 

Total number of points made 

Number of points made on the 
first six problems (averaged) 

Number of points made on the 
first six problems (averaged) 


21.2 
9.8 

33-4 
II. 8 

36-9 
I3-I 
30.0 

13-7 
28.0 . 

193 
32.0 

14.7 
5-8 
2.4 


16.9 




6.4 
18.5 




6.8 
32.0 




II. 9 

24.1 




10.9 
34-2 




15-4 
24.2 




9-4 


II 


II. 7 

-1.8 



In both the experiments there was a decided advantage in using 
a part of the recitation period for drill. In the first experiment, 
the drilled group gained about twice as much as the undrilled 
group; while in the second experiment the drilled group improved 
about two and one-half times as much as the undrilled group. The 
sixth grade gained the most (35%) and the eighth the least (13.8%). 
In order to determine whether group I had gained on fundamentals 
at the expense of reasoning, both groups were tested in arithmet- 
ical reasoning before and after the drill. Here again the drilled 
group did better, making a gain of 6.3%, while the undrilled group 
gained only 3.0%. This last factor is interesting in the light of 
the small amount of connection between fundamentals and arith- 
metical reasoning pointed out above as well as the small amount 
of transfer of one arithmetical process to another (Chapter XIV). 
Since the improvement in reasoning, which had not been drilled 
at all was almost exactly the same proportionately as the processes 
which were drilled, it suggests that the drill had a tonic effect 
upon the remainder of each recitation period following the drill, to 
which much of its value was due. 

In order to discover the permanency of the effects of drill. Brown 
tested both groups once more after a twelve-weeks vacation and 



400 



EDUCATIONAL PSYCHOLOGY 



found that the drilled group was also superior in retention, having 
lost .2% while the undrillcil group had lost 2.2(/'/i. 

The same experiment was repeated by Y. M. Phillip.s. He had 
6() children for subjects and gave to one group drill in fundamentals 
and in reasoning, both oral and written, for eight weeks. Neither 
teachers nor students knew the purpose of the tests. He found 
that, "The improvement in fundamentals of the combined drill 
groups was 15% greater than that of the non-drfll groups. In 
reasoning, the drill groups improxed 50'"^ more than the non-drill 
groups. . . . The greatest gains were made in the si.xth grade and 
the least in the eighth." Almost all the gain on fundamentals 
was in multiplication. 

Mary A. Kerr under the direction of Haggerty reported an 
experiment carried on for six weeks at liloomington, Indiana, on 
the effects of five minutes of drill in addition at the beginning of 
each class period. The drill was begun by adding five three-place 
numbers per column, which were gradually increased to nine three- 
place numbers ])er column. Four hundred and twenty-three chil- 
dren took the drill. Table 118 shows the average performance on 
the Courtis tests, .Series B, before the drill began and at its con- 
clusion in June and, for comparison, the May scores of the best 
twenty Indiana cities for the previous year. 

TABLE 118 





Attkmits 

liLlK)UIN<;TON 

I-KU. Jim: 


IlK;nF.sT 
.Mkdian 

S( (IKI'.S (IK 

20 Ind. 

C"|T1KS 

(May) 


Ri(;iir.s 
Fkb. Junk 


IIlGIIESr 

Media.m 

Scores of 

20 1 vn. 

Cities 

(.May) 


Ter Cent Accirai y 


Gradi: 


ni.OOUINCTON 

Feb. June 


HlCOE-ST 

Median 
SroRE.s or 20 
l.ND. Cities 

(.May) 


6B ... 


8.7 9-7 


8.0 


ss 


6.9 


5-6 


64 72 


65 


6A ... 


9.0 10.5 




5-3 


7-5 




59 71 




7B... 


9.7 10.8 


9.4 


5-6 


8.1 


6.4 


60 75 


68 


7A... 


9.8 II. 8 




6.0 


8.4 




62 71 




8B ... 


11.4 12.0 


IO-3 


6.9 


9 3 


7-2 


61 78 


(>9 


8A ... 


"S 13 7 




6.3 


10 4 




55 76 





The decided advance in each grade and the great suix>riority to 
the best twenty Indiana cities in each test bears eloquent testimony 
to the value of drill in addition fundamentals. 

Su|)t. Herman Wimmer of Rochelle, Illinois, conducted a series 
of comparative e.xptriments on the etTects of drill in arithmetic 



ARITHMETIC 40I 

under various conditions. Each experiment lasted six weeks. The 
time spent in drill was in all cases subtracted from the regular 
class time, in arithmetic. The Courtis tests, Series A, were given 
before and after each experiment. In Experiment I, the drill group 
was rather miscellaneous, grade 5 being drilled five minutes per 
day (probably in fundamentals, though it is impossible to tell 
from Wimmer's account), grade 6E five minutes per day, three- 
fifths on reasoning and two-fifths on fundamentals, grade 6W 
fifteen minutes in a single period per week on reasoning and funda- 
mentals in the same ratio as 6E. In Experiment II, two sections 
of the sixth grade, which had equal ability as shown by tests, were 
drilled five minutes daily, one for speed and the other for accuracy. 
In Experiment III, the seventh grade was drilled five minutes 
daily on reasoning, while the eighth grade was drilled five minutes 
daily on fundamentals. The results of all three experiments are 
shown by percentages of gain in Table 119. As in the previous 
experiments, drill as such is shown to have a decided value though 
much more for reasoning than for fundamentals, the advantage of 
the drilled groups being 32.7% and 4.4%, respectively. In Ex- 
periment II, drill for speed is seen to have a distinct advantage 
over drill for accuracy. Here the gain is considerably more in the 
fundamentals than in reasoning. In Experiment III, we find that 
the class drilled on reasoning gained very largely in reasoning 
alone, while the other class trained in fundamentals gained almost 
exclusively in fundamentals. This presents a contrast to Brown's 
experiment where, it will be remembered, drill in fundamentals 
showed as much gain in reasoning as in fundamentals themselves. 



402 



EDUCATIONAL PSYCHOLOGY 



TABLE 119 
Wimmcr's results 



Probixu 



Grades Used as Sub- 
jects 
Number or Subjects 

Type op AcrrviTV 



ExPFRIirF«JT I 

I)Rii.L VS. No Drill 



5 th 


TTn 


t)TU 


«ru 


79 


70 


Drill 


No 




Drill 


3i i 


16.2 


15-6 


"3 


58 0.. 


251 


68.1 


35-4 



Differ- 

KNtE IS 

Kavor 

OF 

Driix 



experimtvt ii 
Drill for Speed vs. 
Drill for Aix-irac-v 



6E 


6W 


22 
Drill 

SptED 


22 
Drill 
Ac-citt- 

ALY 


1 1 . 1 


8.8 


12.9 


6.9 


II .0 


0.6 



Differ- 

tNCE I.> 
I-AVOR 

ut Drill 

FOR 

Speed 



ExPtRlME.VT III 

Drill in Rea- 

SOVINC \'S. 

Drill in Fu.vua- 

UENTALS 



/TU 

35 
Drill 
Reason- 
ing 



6th 

35 
Drill 

Fl'NDA- 
IIENTALS 



All tests, attempts 
and riRhts 

Fundamentals .... 

Reasoning, at- 
tempts and 
rights 

Reasoning, rights 
only 



17. 1 
4 4 



32-9 

3-'. 7 



23 
6.0 



13 



9.1 
4-7 



18.6 
19. 1 



8.6 
13 3 



(6) The Optimum Distribution of Drill. A number of experi- 
ments have been conductL'd to determine the most economical 
distribution of time, that is, the most economical duration of drill 
periods in arithmetic. T. J. Kirby carried out such experiments 
on a large scale for both addition and division. He used special 
blanks for the practice in both experiments. In addition funda- 
mentals there was a beginning; and a linal test each of t'lfteen 
minutes. Between these periods forty-five minutes of drill were 
variously distributed. The following table shows the distribution 
of the periods, the median initial ability of each group in examples 
correct, and the gross gains as measured by three different methods 
of calculation. Sevan hundred and thirty-two fourth-grade children 
were used as subjects, 

TABLE 120. After Kirbv 





No, 


IsirivL 


Group 


SlB- 


Ttsr 




jrxrs 


Period 


I 


194 


15 min. 


II 


KM 


15 •• 


III 


205 


IS ■• 


IV 


229 


15 • 



DlSTRlnUTION OP 
ISTERVrSINC 45 

Minutes 



2 periods 15 min. 

3 15 •• 
7 " 6 " 

and one 3 " 
21 periods 2 
and one .1 



Final 
Test 
Period 



72} min 

IS •• 

IS " 



Mfjjian 
Initial 

.\llILITY 

(KXAII- 

PLILS 

Correct' 



22 9 
2S.4 
21,0 

25.1 



Gain I)i e to Drill 



Av. 
Gross 
Gain of 
Individ- 
uals 



11 
n 6 
10 7 



Mfd. 
Gross 

Gmn 

Individ- 
uals 



or 



9 S 
II 
9.6 



OP 



Av 
.Mfd. 
Gross 
Gains 



10 2 
9.6 
9.4 

13 9 



ARITHMETIC 



403 



When accurate correction had been made for dififerences in 
initial ability, the gains were then in proportion 100, 121, loi, and 
146 >^ respectively. There was no very distinct tendency observ- 
able here except that the short periods of two minutes yielded a 
distinct advantage over the rest. Unfortunately for the consist- 
ency of these results the next shortest period (6 min.) yielded 
nearly the least gain of all. 

The same general plan was followed by Kirby in the experiment 
on division. Six hundred and six children from the second half 
of grade three and the first half of grade four were used. The 
following table shows in detail the various distributions of the 
drill together with the results by each method. 



TABLE 121. After Kirby 





No. 
Sub- 
jects 


Initial 
Test 
Period 


Distribution of 
Intervening 40 
Min. of Drill 


Final 
Test 
Period 


Median 
Initial 
Ability 
(Exam- 
ples) 
Correct) 


Gain Due to Drill 


Group 


Av. 
Gross 
Gain of 
Individ- 
uals 


Med. 
Gross 
Gain of 
Individ- 
uals 


Av. op 
Median 

Gross 
Gains of 
Classes 


I 

II.... 

III. . . 


204 
209 
193 


10 min. 
10 " 
10 " 


2 20 min. periods 

4 10 " 
20 2 " 


10 min. 
10 " 
10 " 


38.4 
33.4 

41.4 


25. 1 
25.5 
42.6 


22.6 
23.5 
40.4 


20.6 
25.1 

44.7 



When inequalities of initial ability had been removed, the gains 
were found to be in the proportion of loo, iio>^, and 177, Thus 
we find here very consistent and decided advantage in favor of the 
shorter drill periods. Unfortunately in each of these experiments 
it is impossible to tell how much of the gain in the shorter practice 
periods was due to spontaneous practice outside the class. It is 
needless to say, that the children were not permitted to take any 
of the practice cards away from the class. 

Kirby's experiments as a whole both in addition and division 
showed great improvement. Addition with a practice period of 
60 minutes yielded an improvement of 48%, while division with 
a practice period of 50 minutes yielded an improvement of 75%. 
Accuracy was not disturbed in addition, but in division it improved 
2.6%. 

Kirby also investigated the permanence of the improvement 
resulting from the drill. He found that from June to September 
fourth-grade children lost 17% of the ability possessed in June and 
required 58% as much time to regain the efficiency which they pos- 
sessed the preceding year. In division there was a loss of 21%. 



404 



EDUCATIONAL PSVC^OLOf;^• 



It required 60% as mucli time to recover the preceding year's 
efficiency. 

Hahn and Thorndike repeated the addition part of Kirby's 
experiment. Each grade was divided into two sections, section B 
receiving periods exactly half as long as section A. All received 
a total of go minutes of drill, which was preceded and followed by 
a 15-minute test as in Kirby's investigation. Table 122 shows 
the distribution of drill for the various groups, the initial ability, 
the gain, and the advantage of group A according to two diflerent 
methods of scoring the results. 



T.VBLE 122. After Ilahn and Thorndike 





Number 
OF Sub- 
jects 


Length of 
Practice 
Period 


Score Hasfu o.v Rights 

ONLY 


1 Score Based upon the 
Rights + O.ve-Half thk 
Wroso AsSWfRS 


C-tADE 


Average 
Initiai, 
Score 


Avrrage 
Gain 


Advan- 
tage IN 
Kavor ok 
Group A 


Average 

I.S'ITIAL 

.Score 


Average 
Gai.v 


Advan- 
tage IN 
l-AVOR or 
Group A 


7 A . . . 
7 B 

6 A . 
6 B . 

5 A . 
5 B 


10 

16 

1.5 
12 



12 


22K min. 
U'A •• 

20 
10 

( '-J " 


25.9 
26.0 

16..? 
17.4 

i.v."; 

11.8 


2.V7 
17.5 

10 7 
10.7 

11.4 
10.0 


8.2 

.0 

1 4 


.12 2 

,Vi 1 

22 
22 2 

17 2 
19 5 


23.8 
17.6 

11.3 
14.7 

15.3 
17.6 


6.2 

-3.4 
-2.3 



There is no clearly defined advantage for cither the long or the 
short period as was the case with Kirby's experiment on drill in 
adding. What little tendency there is, however, is in favor of the 
longer periods. 

Superintendent Wimmer, in connection with his drill experiments 
previously reported, also investigated the problem of economy of 
long and short periods of jiractice. The drill, three-fifths on rea- 
soning and two-lifths on fundamentals, was given to one group 
for five minutes at the beginning of each class period. The other 
group received one 15-minute period per week. Drill lasted for 
six weeks. The results are shown in the following table. They 
are distinctly in favor of the longer drill period despite the fact that 
only three-fifths as much time was spent by this method. 



ARITHMETIC 405 

TABLE 123 

After Wimmer 



Problem Favorable Distribution of Time for Drill 

Grades Used 6E 6W Difference in 

Number of Subjects 22 22 Favor of Drill 

Type of Activity 5 Min. 1 5 Min. once per Week 

Drill Daily Drill Weekly 

All tests, attempts and rights. . . 34.3 45.1 10.8 

Fundamentals i4-7 181 3-4 

Reasoning, attempts and rights. . 67 . 2 83 . 5 16.3 

Reasoning, rights only 75.9 102. i 26.2 

This is particularly the case with reasoning. The rather slight 
advantage of the long periods on fundamentals in connection with 
the uncertain indications of Hahn and Thorndike's results and 
the opposite finding of Kirby suggests that there is little or no 
advantage in the distribution of time in arithmetical drill on fun- 
damentals but that the longer periods are more favorable for drill 
in reasoning. 

(7) Special or Economical Methods of Drill. A number of 
special methods of giving drill in arithmetic have been advocated 
and used. That of Studebaker has already been noticed. Courtis 
has also published practice pads for drill purposes. Flora Wilbur 
undertook to determine experimentally the value of this kind of 
drill at the Fort Wayne, Indiana, training school. Two classes 
of 14 children each were divided each into two groups of equal 
ability on the basis of the Courtis tests, Series B. One section 
of the fifth grade received four and one-half minutes of drill with 
the pads at the beginning of each class period, and one section 
of the sixth grade received similar drill for four minutes. The 
remaining sections received the regular class work. The experi- 
ment lasted from September to May. The results are shown in 
percentages of gain in the following table. The drill was clearly 
of value in both grades and in all four processes. 



4o6 



KDUCATK )\AI. PS^'C•HOLOG V 









TABLE 


124. 


After Wilbur 












Grade Five 






Grad 


■-Six 






Speed 


y^ 


tCURACY 


Si'EED 


Aix-uiLArv 














u u 






u w 














(^ 



























f.t 






r:h 






FF 






h (^ 




u 




< < 


H 












u 




;J3 




u 




X X 






« at 






Q£ tt 


u 






b 


u 




< 


U 






!i 


lb C 


c 


U 


K 




X 

(1. 


t 


z 




t 


z 

X u 


E 




at 


n 


c z 







< 


<> 





ec 


< 't 





■< 


<i: 





;S 


5!^ 




y. 


cu 


Oo 


'A 


C 


Oo 


z 


0. 


Oo 


y. 


£ 


Oo 


Addition . 


30 


43 


13 


30 


37 


7 


43 


60 


17 


12 


20 


8 


Subtrac- 


























tion . . . 


27 


54 


27 


29 


49 


20 


28 


4Q 


21 


2 


21 


19 


Multipli- 


























cation. . 


12 


148 


136 


26 


31 


S 


SI 


88 


37 


18 


20 


2 


Division. . 





144 


144 


43 


4^ 


— 2 


76 


129 


53 


28 


34 


6 



Division and multiplication profited most and addition least, 
as is usual in such experiments. The real question remains, how- 
ever: Is drill with the Courtis practice pad more or less efiicient 
than drill as ordinarily given? 

Kirkpatrick performed two comparative experiments to de- 
termine the relative economy of various metho<ls of memorizing 
multiphcation tables. His subjects were twenty nomial school 
men, divided into groups of equal ability. As these subjects knew 
the ordinary tables, he had them learn the products of 7 multiplied 
by all the prime numbers between 17 and 53. One group simply 
memorized the table by rote for the first five days, then spent the 
periods of the next live days in writing down the answers on a 
blank with a card containing the table before them for reference. 
The other section spent the periods of all ten days working on 
the blank with the table for reference. The time consumed during 
the five days of memorizing was about an hour. A test at the end 
of the experiment showed that, in a period of two minutes, the 
practiced group put down 46.2 answers wliile the memorizing 
group put down 40.9, thus showing a distinct advantage for the 
practiced grouj). 

A second experiment was performed with two groups of normal 
school students of ef|ual ability, twenty-five in each group. One 
group practiced witli tlic keys and blanks as in the first experiment 
while the other group spent the same amount of time multijjlying 
out the products as needed. The experiment extended over eight 



ARITHMETIC 407 

days. When tested at the end of the experiment, those using the 
key put down 25.4 answers in two minutes, while the computers 
wrote down 44.3 examples, showing a decided advantage for the 
computation method. He concludes that tables should be learned 
by use rather than by memorizing. 

Conrad and Arps ('16) investigated the effect of suppressing artic- 
ulatory movements upon the effect of drill in rapid adding. They 
divided sixty-four high school students into equal groups of equal 
ability. The students were then given eight periods of drill in 
rapid addition of columns. The pupils in one group were per- 
mitted to add in their ordinary way which involved a great deal 
of articulation or inner speech. The other group was cautioned 
repeatedly and emphatically to "think results only." The former 
was called the traditional method and the latter the economical 
method. The percentages of gain by the two methods were as 
follows : 

The traditional method gained in attempts 8.5% and in rights 
2.5%. The economical method gained in attempts 34.4% and in 
rights 30.9%. This gave an advantage in favor of the economical 
method of 25% in attempts and of 33.4% in rights. These results 
came out almost startlingly in favor of "thinking results only." 
The evil effects of articulation and lip movements have been no- 
ticed in connection with reading (page 287). It is probable that 
the cause is the same in both cases. 

P. B. Ballard investigated the comparative efl&ciency of the 
"equal addition" method and the "decomposition" method in 
subtraction. 

"In the equal addition method the compensation is made — accounts 
are squared — at the very first number dealt with after the minuend has 
been disturbed. In subtracting 37 from 85, after taking 7 from 15 the 
disturbed relationship of difference between minuend and subtrahend is 
immediately restored by increasing the 3 tens to 4 tens. In the method 
of decomposition, however, it is the 8, the second figure dealt with, that 
has to be changed to restore the balance. If the minuend figure is zero, 
the balancing of accounts is still longer deferred." 

Ballard gave tests in fundamentals to 71 English schools of 
which 23 had been taught subtraction by the method of equal 
addition while the rest had been taught it by the method of de- 
composition. While there was little difference in the average ability 
of the two groups in the other three fundamental operations, 



4o8 



EDUCATIONAL PSYCHOLOGY 



there was a very striking superiority in the score for subtraction 
of the equal addition group. At 13 years of age it amounted 
to over loVo and at earher ages it amounted to over 40%. (See 
Figure 84.) Inasmuch as the decomposition, or less efficient 
method, is the one in general use in this country, it is evident 
that this matter deserves careful attention. 



OB 


- 


y^ 


60 


_ 


/ ^ 




■" 


/ ^^ 




- 


^ 




" 


y^ / 




™ 


f / 


60 


- 






- 


# s^ 




~ 


T/ ^-^V 


40 


: 






- 






/ / 




~ 


/ / 




- 


/ / 


:« 


_ 


/ / 




:; 


/ / 






/ 




_ 


r~J 




_ 


/ 


20 


_/ 
/" 


/ 


10 


- 


J — 1 -. 1 1 1 1 1 1 I 1 1 



8i 8i JH 



OJ lOi 103 IIJ llj \2\ 12J \i\ 13| 
Ages 



Fig. 84. — ShovviriK superiority of teaching suljlriulion liy the "c(iu;i! addi- 
tions" method. After Ballard ('15). 



Mead and Sears performi-d two experiments in comparative 
economy of methods in aritlimelic. The first was to compare the 
efficiency of the ordinary "take away" method of subtraction 
wliich involves the learning of an entire subtraction tabic, as 
compared with "addition" subtraction which permits the use of 
the addition table, thus saving the learning of an entire table. 
In the first the formula is "8 minus 2 equals what?", in the second, 
"2 plus what t'(|uals <S?" Two second-grade classes of approxi- 
mately equal median ability as indicated by the Courtis tests were 



ARITHMETIC 409 

taught subtraction by the respective methods thirty minutes per 
day for four months, all other factors being equalized as fully as pos- 
sible. Tests given periodically throughout the experiment showed 
that at the outset the addition method was superior but the "take 
away" method gradually overtook it until at the end of the four 
months the "take away" method was superior by 4.5 points which 
was nearly one-third of the final median score made by the addition 
group. This difference disappeared, however, when both groups 
were tested on longer examples. 

The second experiment was to compare the multiplicative 
method of division. The formula of the first is, " Five into twenty 
how many times?", that of the latter, "Five times what equals 
twenty?" Two third-grade classes were used as subjects in this 
experiment. Other conditions were similar to the experiment on 
addition. In this experiment, the final test on combinations 
revealed the fact that the multiphcative class stood 4.3 points 
above the "into" class, which was about one-fifth of the final 
score of the "into" class. This difference disappeared, however, 
when the class was tested on longer examples just as that in ad- 
dition noted above. 

J. A. Drushel investigated the relative efficiency of two methods 
of determining the position of the decimal point of the quotient in 
the division of decimals. The rule of method A, the older one, is: 
"There are as many ])laces in the quotient as those in the dividend 
exceed those in the divisor." The rule of method B, the newer of 
the two, is: "First render the divisor an integer by multiplying 
both dividend and divisor by 10 or some power of ten. Then 
proceed as with integral divisors." A short test in division of 
decimals was given to 576 freshmen at Harris Teachers' College. 
Of these, 507 had studied division of decimals by method A, while 
69 had studied it by method B. The results show that the stu- 
dents taught by method A had the very low accuracy of 66% in 
placing the points, while those taught by method B had the very 
excellent accuracy of 99%. If future investigations confirm these 
results, method B should be generally adopted. 

(8) Speed vs. Accuracy. Thorndike investigated the relation 
between speed and accuracy in simple addition. Six hundred and 
seventy-one students were tested apparently on two different occa- 
sions in a class experiment in adding columns of nine digits. The 
subjects were then grouped according to speed as shown by the 
following table: 



4IO 



EDUCATIONAL PSVCHULOGY 



TABLE 125. After Thomdike f 'is) 



NrunEK or 
Inuiviui'als in 

IJROUP 


NuuiiKK OP Additions i-kk UH) 

Skc-osus (Counting; thk Imt or 

W'kiiim: tmk Answlk Fxjlal to 

Onk Addition's Iimk) 

1 


Ai'I'Koxiuate NiuuER or KkkokSPtit 

KXKJ Additions, i. f.., \\ moni; Answers 

ptK 100 It-s-DioiT .Additions 




Eably Test 


Late Test 


Eakly Test 


Late Tkst 


65 

108 

86 
"5 

109 
103 

65 

20 


150 

108 

88 

75 
04 

55 
46 

37 


162 
120 

99 

87 

75 

66 

58 

46 1 


7.0 
9.1 

10 3 
12.0 
12.7 
12.6 
14.4 
17 5 


3-8 
6.5 
6.7 
8-3 
9.0 

9 3 

10 5 
14.4 



With a (lecreasL- in tlic rate of additions there is a stcarh- increase 
in the number of errors j)er 1,000 additions. Tliorndike concludes 
" that the sort of individual who is quick in adding is more accurate 
also than the one who is slow." 

(9) Limits of Attainment. Since there is general agreement 
that the fundamental number combinations should become auto- 
matic association processes, it is pertinent to ask, how high a 
degree of skill should be developed in pupils? This question is 
similar to the one discussed in connection with quality of hand- 
writing. It is reasonable to maintain that it is probably uneco- 
nomical to attempt to develop a degree of speed beyond a certain 
point. 

How great speed is practically necessary or worth while? Obvi- 
ously the school may devote relatively too much attention to the 
development of speed in the four fundament. d operations at too 
great a cost of time. It would, therefore, be important to know 
what degree of proficiency is needed for the practical affairs of life. 
An investigation of the problem is needed. 

(10) Errors. 'I'hc detection and classification of errors and the 
discovery of the frequency with whii h they occur are highly useful 
facts in any school subject because such infonnation will lielp to 
make instruction specific. It will indicate the particular points 
at which drill should be directed. 

Howell made an analysis of the mistakes in division (Hcurring 
in the Courtis tests as applied to the pui)ils in his school. He 
found the following rubrics of errors: 



ARITHMETIC 41I 

"i. Making the quotient figure the same as the divisor, 

(a) When a difference of only one exists between the divisor and 

quotient ; 

(b) When the quotient is commonly used as the divisor of the 

given dividend. 

" 2. Making some factor (other than the divisor), commonly used as 
the divisor of a given dividend, the quotient figure. 

"3. When dividing a digit by itself, making the quotient figure the 
same. 

"4. When dividing a digit by itself, making the quotient figure zero. 

" 5. When dividing by one, making the quotient one. 

"6. When the dividend is zero, making the quotient the same as tlie 
divisor. 

"7. Pupils whose associations are as yet feeble or become so through 
fatigue or distraction are commonly observed to resort to running up the 
table. 

"They frequently miss count and get a quotient figure one removed 
(say) from the right one. 

"8. Making one of the quotient figures the quotient. 

"9. Substituting multipHcation for division. 

" 10. Unclassified." 

The frequency with which the different types of errors occurred 
is shown in Table 126. 

TABLE 126. After Howell 
Showing the number of mistakes in the division tables falling into each class 



lDE 








Kinds of 


Mistakes 






I 


2 


3 


4 


5 


6 


7 


8 




a&b 
















3 


16 


12 


2 


14 


I 


2 


26 


6 


4 


13 


7 


2 


20 


4 


4 


23 


3 


5 


S 


7 


3 


23 


3 


I 


14 





6 


4 


8 





32 


16 


4 


16 


2 


7 


10 


6 


I 


27 


4 


2 


7 


2 


8 


9 


9 


4 


65 


11 


II 


5 


3 


tal 


57 


49 


12 


181 


39 


24 


91 


16 



Total Mistakes 


10 




43 


124 


49 


133 


4 


62 


17 


99 


19 


78 


3 


120 


35 


616 



From this table it appears that certain types of error occur 
much more frequently than others. For example, Error No. 4, 
"when dividing a digit by itself, making the quotient figure zero," 
occurs 181 times; whereas Error Nos. 3 and 9 occur only 12 times 
each. More extensive studies of this sort are much needed. Learn- 
ing in any school subject is apt to be more economical the more 



412 EDUCATIONAL PSVCHOUKJY 

specifically the learners' allcntion may be flirccted to the particular 
processes to be exercised. 

A. S. Gist tabulated the errors in subtraction, multiplication, 
and division of 812 arithmetic test papers from six schools in 
Seattle. Table 127 shows the percentage of different types of errors 
separately for each of the three processes and for each of the grades 
from 4 to 8. It is noticeable that the greatest difficulty was pre- 
sented in subtraction by borrowing, in multiplication by the 
tables and in division by the remainder. For the most part the 
proportions of the various errors are fairly constant from year to 
year. 

Besides the prevalence of the general tj'pe of errors noted above, 
the relative difficulty of the various elementary combinations is 
a matter of much practical importance. C. L. Phelps attempted 
to determine this for the addition combinations by finding the 
percentage of errors made on each of the 55 addition combinations 
of the Courtis tests in 5,950 papers made by repeatedly testing 
238 eighth-grade children. These results are shown in detail in 
Table 128. There is a clear relation between the per cent of errors 
and the size of the total resulting from the combination. The 
writer computed the correlation between the two by Spearman's 
rank method and found a coefficient of .57. The relation is shown 
in detail by the curve of Figure 76. For the most part the o<ld 
and especially the prime numbers are more difficult than the even 
numbers adjacent to them. While there is seen to be a fairly 
steady rise up to 10, there is an abrupt increase at the beginning 
of the teens which continues as far as the investigation extends. 



TABLE 127 

The per cent of each type of error in the txamples of subtraction, multiplica- 
tion, and division respectively from the 812 papers from six schools in 
Seattle. (After Gist ('17).) 

Grade 4Tn ^rn 6th 7th 8th 

Subtraction: 

HorrowinK 54 5^ 5^ $1 55 

Combinations 36 38 45 44 41 

( )missions 2 i 2 3 i 

Reversions i 2 }i o o 

7 — o, o, etc 5 3 ' / o o 

Left-hand digit o o o o 2 



ARITHMETIC 413 

TABLE 127— Continued 

4TH 5TH 6th 7TH 8th 
Multiplication: 

Tables 79 

Addition 18 

Cipher in multiplier 1.5 

Division : 

Remainder too large 34 

Multiplication 22 

Subtraction 11 

Last remainder o, and o in dividend ... 7 

Multiplicand larger than dividend 7 

Failure to bring down all of dividend . . 7 

Failure to bring down correct digit .... 2 

Failure to place all of quotient in quo . . 7 

Cipher in quotient, as 908—98 3 

A somewhat similar investigation was carried out by Holloway. 
He tabulated the number of errors made on each of the addition 
combinations by 1,065 third-grade children. His results are given 
in Table 128 in the order of the number of errors, parallel to those 
of Phelps. While there is considerable agreement between the two 
studies as to the relative difficulty of the various combinations, 
there are also rather striking disagreements. How much these 
differences are due to the stress which had been put upon the 
various combinations in the teaching of the children cannot be 
determined. The latter defect is much less likely in Holloway's 
results because of his much wider range of subjects. 

Holloway also tabulated the errors in the multiplication com- 
binations in the test papers of 1,215 third-grade children. They 
are given in detail in Table 129. 



73 


73 


77 


75 


20 


22 


19 


20 


6 


5 


4 


5 


39 


27 


19 


10 


15 


19 


37 


55 


14 


18 


25 


23 


15 


19 


7 


II 


4 


I 


I 


I 


4 


3 





6 


I 


4 


4 


6 


I 


I 


3 


3 


7 


8 


4 


7 



414 KDUCATIONAI. I'SN C1I<)L(K.\ 



TABLE 128 



Table showiriR the relative difficulty of the various addition combinations for 
the third and eighth grades resjiectively 





NlMBER OP Er- 


Per Cent op Errors 










Com- 


ROKS BY 1.065 


IN 5.950 Papers 


( 


LoM- 






binations 


3rd CIradk Sub- 


PROM 2.?8 8tu Gr.\de 


binations 


HoLLOWAV 


Phelps 




jects.— Hollo- 


CUILDREN. — 












way 


Phelps 










9+8 


95 


2.44 


8 + I 


19 


.62 


9 +7 


90 


332 


3 


+ 1 


19 


■45 


9+6 


82 


2.60 


4 


+ 3 


18 


.67 


8 + 7 


69 


2.30 


3 


+ 2 


17 


54 


8 +S 


68 


3.10 


6 


+ I 


17 


2.22 


8+6 


66 


1.04 


I 


+ I 


17 


• 27 


7 +S 


56 


2.25 


4 


+ 2 


16 


•50 


9+4 


51 




9 


+ 2 


15 


1. 16 


7 +6 


5° 


i'.S6 


5 


+ I 


15 


•72 


9 +S 


49 


2.50 


4 


+ I 


15 


I 30 


7 +4 


48 


1-95 


5 


+ 2 


13 


.86 


9+3 


43 


2-55 


9 


+ I 


13 


I 09 


8+3 


41 


1.94 


8 


+ 2 


13 


.88 


8+8 


37 


1.98 


5 


+ 5 


9 


.07 


8+4 


37 


2.38 


2 


+ 2 


9 


•37 


7 +3 


37 


2.02 


4 


+ 4 


8 


. 12 


6+4 


34 


.69 


3 


+ 3 


8 


1.46 


6 +s 


32 


2.27 





+ 8 




.62 


9+9 


29 


•30 





+ 5 




■ 52 


5+3 


26 


1.31 





+ 




•39 


7 + 2 


24 


1.32 





+ 3 




.28 


2 + I 


21 


.64 





+ I 




• 25 


7 +7 


20 


•14 





+ 4 




.24 


6+6 


20 


•35 





+ 7 




• 24 


5 +4 


20 


• 72 





+ 9 




•J5 


6+3 


20 


1. 19 





+ 2 




• 14 


7 +1 


20 


2.42 





+ 6 




•OS 


6 + 2 


19 


•71 











ARITHMETIC 



415 



TABLE 129 



Table showing the order of difficulty 
of 1,215 third grade children 

II X 

7 X 
9 X 
9 X 

II X 

8 X 
II X 

6 X 

II X 

8 X 

6 X 



as determined by the number of 
at end of year. After Phelps. 



II X 11 . 


•735 


12 X II.. 


•65s 


II X 10.. 


..638 


12 X 10. 


■ .542 


12 X 8.. 


. .460 


9X7. 


•455 


12 X 7- 


.438 


8X7 


•435 


12 X 12. 


•425 


9X8.. 


. .422 


12 X 9 ■ 


.417 


9X6.. 


•390 


8X8.. 


.361 


12 X 6.. 


• 361 


8 X 63 . . 


■ 42 


9X4 


. .292 


7X6.. 


..28s 


12 X S-- 


. .271 


7 X 7•■ 


. .268 


9 X 9 • 


• 263 


12 X 4 • 


• 250 


10 X 10. . 


. .241 


8X4- 


• 235 


7 X 4-- 


. .192 


12 X 3- 


.•183 



6 X 
II X 

6 X 

11 X 
10 X 
10 X 
10 X 

12 X 
10 X 

4 X 
4 X 

7 X 
10 X 

8 X 



181 
181 
169 
168 
167 

151 

144 

138 
137 
137 
133 
129 

113 
102 

99 
94 
86 

85 
81 

79 
78 
76 

71 
58 
58 



5X4 


■55 


6X2.. 


• 50 


5X3 


.46 


II X 2.. 


.46 


I X I.. 


■ 41 


9X2.. 


• 39 


10 X 3 


•38 


7X2.. 


.38 


5X5 


■ 34 


4X2.. 


■ 32 


10 X 4 


31 


10 X 2.. 


■ 31 


II X I.. 


• 31 


4X1.. 


• 31 


3X1.. 


. .28 


5X2.. 


. .26 


3X3 


• 25 


9X1.. 


. .22 


3X2.. 


. .21 


9X1.. 


. .21 


6X1.. 


. .21 


12 X I.. 


. .20 


5 Xi.. 


. .20 


2X1.. 


. .20 


2X2.. 


. .18 


8X1.. 


. .18 


10 X I. . 


. .12 



CHAPTER XXI 
HISTORY 

Psychological Processes in Learnlng History 

It is more difficult, at least more uncertain, to make an analysis 
of the psych()]op;ical processes concerned in the learning of history 
than it is in case of most of the subjects treated thus far, for the 
reason that teachers are not as fully agreed as to what is to be 
learned in history. In general there are two extreme views: One 
would hold that the learning of history means the learning of the 
main facts — names, dates and events — of the human race; while 
the other would hold that history means the acquisition of ability 
to interpret the significance of human events. In their most ex- 
treme forms the two views would be a memorizing of isolated 
facts versus an interpretation of facts with little emphasis on 
facts. Probably no one holds either of these extreme positions and 
the distinction is useful only for analytical purposes since the 
psychological processes involved in learning history would be 
quite difTerent in these two modes of approaching the subject. 
Those who stress the interpretational aspect of history would 
stress the conception that the purpose of history in the ])ublic 
schools is training for citizenship or the development of patriotism. 

Let us assume for our present analysis that the obvious aim of 
what is to be learned or accjuired in history is facts of the events 
of human beings and the connections among these facts — a view 
to which probably every one could agree. What are the psycho- 
logical functions concerned in acquiring and connecting historical 
facts? Let us take as a concrete instance a given historical event 
and see what mental operations are necessary to grasp, interpret 
and remember it. Let us take the statement that Columbus dis- 
covered .Anurica in 1492. The psychological jirocesses involved 
or assumed in learning and grasping this statement would be 
substantially as follows: To begin with, it would involve all the 
steps enumerated in the analysis of the reading process, or of the 
process involved in understanding spoken language when the facts 
are heard instead of read, since practically all history is learncxl 

4.6 



HISTORY 417 

through reading. The chief difference would be in the emphasis 
and elaboration of some of the steps. Taking the factors as enu- 
merated in Chapter XVI, the main difference between ordinary 
reading and learning history would be in step (6): ''The establish- 
ing or arousal of association processes whereby the incoming im- 
pulses are interpreted." These specific processes of association 
and interpretation over and above ordinary reading, necessary for 
grasping a historical statement are as follows: 

(i) A mental picturing or conceiving of the persons, actions, 
locaHties and objects, concerned in the event. 

(2) A mental picturing or conceiving of points and locations in 
time. 

(3) Processes (i) and (2) applied to events preceding the one in 
question. 

(4) Processes (i) and (2) applied to events succeeding the one 
in question. 

(5) A judgment concerning the internal motives of the persons 
and the external conditions that led to the event. 

(6) A judgment concerning the effect of the event upon the 
motives of the persons and environmental conditions involved in 
succeeding events. 

(7) A remernbering of the mental processes in steps (i) to (6), 
in so far as a permanent memory of them is considered important. 

Steps (i) to (4) use primarily the imagination, steps (5) and (6) 
judgment and reasoning, and step (7) memory. 

Thus the grasping, interpreting, and remembering of the state- 
ment that Columbus discovered America in 1492 would mean, (i) 
An imagining or conceiving of Columbus as a person, his asso- 
ciates, ships, water, the voyage and landing on new soil; (2) an 
imagining or conceiving of the time so as to give a notion of how 
long ago 1492 was; (3') and (4) a similar procedure with events 
coming before and after this particular one, such as Columbus's 
interview at the court of Queen Isabella of Spain, the need for 
another route to India, succeeding voyages, settlement of the new 
country, and the like; (5) and (6) judgments concerning the effect 
of the preceding events in bringing about the particular event 
under consideration, and judgments concerning the effect of the 
latter in bringing about later events; and (7) a repetition of the 
learning of the fact with its connections to fix it in mind. 



41 8 EDUCATIONAL PSYCHOLOGY 

Measurement of Attainment in History 

The cliftu ultics that beset any endeavor to devise an objective 
and generally acceptable methcKi of measuring attainment in 
history are very great for the reason that historians ditTcr very 
widely in the selection, emphasis and interpretation of facts and 
in the manner of stating the facts; and also for the reason that 
teachers as well as texts dilTer in the relative emphasis on the learn- 
ing and remembering of facts as compared with their interpreta- 
tion. The author has prepared a plan with the aim of meeting 
these difficulties as far as possible so as to secure a test that could 
be used fairly wherever American history is taught; it is necessary 
to exercise much care in selecting the right material for testing 
purposes. The scheme finally carried out was as follows: Five 
widely used text-books in American History were carefully com- 
pared and all facts and interpretations given in all five were selected 
and formulated into statements or sentences. This gave a total 
of 278 statements — a remarkably small body of facts common to 
five texts. These statements were then made into a mutilated 
text or completion test. Certain important words or phrases were 
omitted which are to be supplied by the pupils doing the test. The 
entire 278 statements would be too long as a single test. Hence 
they were spUt up into four parallel sets, each containing 69 or 70 
statements, by taking for the first set, statements numbers (i), 
(5), etc.; for the second set, numbers (2), (6), etc.; tliese four tests 
may be used interchangeably at different times in testing a class. 
Direct comparisons and measurements of progress can thereby 
be made. The score of a pupil is the number of omitted parts 
correctly supplied. The following statements serve to illustrate 
the nature of the resulting test. They are the first ten statements 
of the first test. 

1. discovered .Vmcricu in 14Q2. 

2. John Cabot exploring for the in 1407 landed 

on the coast and claimed the country for 



3. Sillied around the globe in 1510-1521. 

4. discovered the Mississippi River in 1541. 

5. Two expeditions sent out by to settle Vir- 
ginia in i5<S5 and i 5S7 respectively, failed. 

6. — was governor of \'irginia after Delaware 

left. 



HISTORY 419 

7. in service of the Dutch East India Com- 
pany, explored the —river in 1609. 

8. was Governor of the Dominion of New Eng- 
land, which was composed of (i) — (2) 

and (3) . 

9. John Winthrop came to America in 1630 and settled 



10. New Hampshire was founded in . 

The following are the average scores for the ends of the different 
grades obtained from approximately 2,000 pupils: 

Grade 6 7 8 H. S. 

Scores 7 20 38 38 

Individual differences and overlapping of successive grades as 
shown in Figure 22 are enormously wide. In a certain eighth grade 
composed of thirty-six pupils, the best pupil made a score of 102 
and the poorest a score of 4. The differences among various schools 
are indicated in Table 130. Very wide differences exist, which are 
probably due chiefly to differences in methods of teaching. Thus 
the best eighth grade made a score of 66 and the lowest one a 
score of 19. Even in the same school system the difi'erences among 
schools may be very wide. For example, in city N, the best eighth 
grade averaged 52 and the poorest 19. Another striking observa- 
tion is the fact that the average attainment of the high school 
pupils is no better than that of the eighth grade. Apparently the 
pupils relearn in the high-school course in American history about 
as much as they forgot during the intervening two or three years 
since they left the eighth grade. This does not necessarily mean 
that American history in the high school is useless since the re- 
learning will help to guard against further loss. More extensive 
tests are needed on this point. 

Striking sex differences in knowledge of history have been re- 
vealed by such a test as the one here described. The median scores 
for boys and girls in the writer's test were as follows: 

Boys Girls 

Number Median Score Number Median Score 

High School 47 41 73 36 

8th grade 288 45 352 31 

7th grade 94 24 loi 17 



420 EDUCATIONAL PSVClIOLUCiV 

Ik'll and McCollum applitd a history test to 1,500 pupils and found 
similar dilTerences. 'I'lii- hoys in the elementary schools did 28% 
heller and in the hi.L'h schools 31'^'^ better than the girls. This 
superiority on the pari of the hoys may he due to their greater 
interest in battles which in turn may be due to the greater strength 
of the lighting instinct. 

TABLE 130 
Scores in the writer's .Vmcrican History Test, Series A 

Grade 7 8 H. S. 

City A — 66 35 

" B — 30 — 

" C - 32 48 

" D, School I — 22 — 

"2 — 59 — 

"3 — — 42 

" E 17 33 ~ 

" F — 40 — 

" G, School 1 22 — — 

"2 17 65 — 

" H " 1 17 40 — 

"2 19 48 — 

"I " 1 29 — — 

"2 — 33 — 

"3 — — 39 

"J - - 41 

" K 15 36 - 

" L - 32 — 

" M — 29 — 

" N School I — ■ 33 — 

"2 — 19 — 

"3 - 52 - 

"4 17 50 — 

"5 — 29 — 

Economic Methods in the Learning and Teachin(; of History 

Kxjx'rimental work that has been done up to the present time in 
the jisychology and pedagogy of school subjects has been contined 
almost entirely to the subjects thus far considered. Yet the prob- 
lems and factors entering into such a subject as history are exceed- 
ingly intricate and as much worth while and for the most j)art as 
caj)ahle of experimental determination as most of the problems in 
the other suljjects. The discussion of factors and conditions 
alTecting the most economic i)rocedure in history will, therefore, 



HISTORY 421 

have to be limited to a few suggestions and beginnings in experi- 
mental work. The problem for the future will consist in determin- 
ing the factors which promote or retard the elements enumerated 
in the first section of this chapter. Economic procedure in learning 
history resolves itself into discovering the most favorable means 
of, and a measurement of their actual effects in, grasping, imagin- 
ing, judging and remembering the important events of the human 
race. 

(i) By what means may the imagining or conceiving of a given 
event be brought about most effectively? Numerous devices are 
employed to assist the imagination, such as pictures, dramatiza- 
tions, pageants, etc. These are probably used with profit but no 
one has ever determined to what extent they actually contribute, 
or whether they contribute at all to the better understanding of 
the event, or how much time devoted to dramatization, for ex- 
ample, is worth the returns it may bring. Experimental work 
should obviously be undertaken. Dramas and pageants may be 
easily overdone and may often deal with unimportant phases of 
the persons or events concerned. 

(2) By what method may the imagining or conceiving of a given 
event in time with respect to other events be accomplished? Ori- 
entation in time is a very complex psychological process and prob- 
ably develops rather gradually through the years of a child's 
experience. It probably develops from the immediate perceptions 
of changes in the child's environment to the gradual extension to 
longer historical periods v/hich are not directly perceived but are 
thought of in symboUc form. Thus the writer pictures different 
periods and points in history in spatial terms by imagining a hori- 
zontal line about three feet in length extending from a point, which 
represents the present, toward the left, that is back to the past. 
The discovery of America is located about four inches to the left 
from the beginning point, the birth of Christ about a foot and a 
half to the left, and so forth for other approximate locations in 
time. As a mere opinion, the writer believes that it would be 
advantageous to introduce the pupil to the study of history by 
giving a bird's-eye view over long stretches of time, as this would 
probably aid the imagination in conceiving time. Historians 
usually object rather strenuously to this method of introducing or 
teaching history. Psychologically, it would seem easier to imagine 
long periods of time and the relative location of events in them by 
viewing all history pretty much at a glance, and by giving then 



422 EDUCATIONAL I'SVCUOLOtiV 

more and more detailed considtniliun to each period. But this 
again is a problem requiring experimental determination. 

(3) How may judgments about the personal motives of histori- 
cal figures and causal relations among events be best developed? 
Nobody knows. All we can say is to encourage the making of 
such judgments and interpretations according to the best insight 
of the pupil and then to check them up with those of comjietenL 
historians. 

(4.) What are the most effective methods of remembering his- 
torical events? In spite of the extensive experimental work in the 
tield of memory, there is very little in the way of concrete advice 
that can be given to a pupil to assist him at this point. The fol- 
lowing, partly general and partly specific, suggestions may be 
given. These have been corroborated by experimental data and 
have been stated in Chapter XII on How to Study, and in Chapter 
X\T on Reading, and will, therefore, be only mentioned here. 

a. Thoroughly understand the facts you wish to remember. 

b. Systematize the facts to be remembered. 

c. Look for the essentials. 

d. Recall, after every paragraph or two, the essential ideas read. 

e. At longer intervals, re-think or review the essential ideas 
again. 

f. Develop your own special means, associative links, or schemes 
ft)r remembering certain facts. Systems of memory such as that 
developed by Loiselte for remembering, for exam[)le, the names of 
the Presidents of the United States, consist in establishing certain 
associations of similarity in sound between certain parts of the 
successive names, as shown in the following illustration: 

(K<ir;,'f \V;ishingTON In. "Ton" and "John" make a fairly good 

JOHN' .\ilams In. \>y .sound. 

JOII.N .\dams In. "John" and "Thorn" (the "h" is silent 

'I'lIO.Mas JclTerson in both names) make an IN. by sound, 

imijcrfect but adequate if noticed. 

Thomas JeffcrSON In. Both names terminating with the same 

James MadiSON syllable, "son," makes a clear case of In. 

by sound and spelling. 

J.\MKS Madison In. This pair of names furnishes an example 

J.AMKS Monnx: of |)crfect In. by soimd and sjHjlling in 

the Christian names. 



HISTORY 423 



James MONroe In. "Mon" and "John" give us a good In. 

JOHN Q. Adams by sound. 

JOHN Q. Adams In. "Jack" is a nickname for John — a r^se 

Andrew JACKson of Synonymous In. 

Etc. (Loisette, p. 26.) 

The main objection to such a plan is its artificialty. The chief 
advantage is that it does draw specific attention to the facts to be 
remembered; but everyone can develop his own special links or 
clues for retaining facts with which he has difficulty. These are 
likely to be more natural, more serviceable and more permanent 
than artificial ones forced upon the learner from the outside. The 
main point is that each one should attempt to establish such clues 
which will usually result in discovering useful associative links and 
at the same time force attention upon the facts to be retained. 

(5) Another very important problem in the economy of learning 
history is the question of essential material. What should the 
child really be expected to master? What facts, names and dates 
shoiild he actually learn? What interpretations should he be led 
to make and acquire? In general there has been a distinct shift 
from regarding history as a chronicle of wars to regarding history 
as a tracing of the development of political, industrial and social 
institutions. In recent years, various committees have been at 
work to decide upon a body of minimum essentials. Thus the 
committees of Iowa and Minnesota have made the following sug- 
gestions as summarized by Betts ('17): 

"Wars. Limit the study of wars to their remote and immediate 
causes; their general geography; resources and problems of nations in- 
volved; general plan of military operations; a few critical battles; im- 
portant leaders; what the war settled, and the after effects; cost in men 
and treasure. This plan will reduce the war phase of history study by 
more than half. 

"Eliminate the detailed study of battles except: Battle of Quebec; 
Lexington and Concord; Bunker Hill; Saratoga; Yorktown; Lake Erie; 
Merrimac and Monitor; Gettysburg; Vicksburg; Manila. 

"Dates. Limit the memorizing of dates to events of central impor- 
tance like the following: 1492, the discovery of America; 1607, settlement 
of Jamestown; 1619, slavery introduced; 1620, Pilgrims land at Plymouth; 
1643, confederation of colonies; 1775, Lexington, Concord and Bunker 
Hill; 1776, Declaration of Independence; 1781, Cornwallis surrenders; 
1789, First Congress; 1793, Whitney's cotton gin; 1803, Louisiana Pur- 



424 EDUCATIONAL PSYCHOLOGY 

chase; 1807, Fulton's steamboat; 1812, war with England; 1820, Missouri 
Compromise; 1823, Monroe Doctrine; 1826, first railroad; 1844, first 
telegraph; 1S46, sewing-machine invented; 1845, first reaper; 1S46- 
1848, Mexican War; 1S61, secession and Civil War; 1863, Emancipation 
Proclamation; Gettysburg. \'icksburg; 1866, Atlantic cable; 1S76, first 
telephone, 1S7S, electric light invented; iSg8. war with Spain; 1903, 
first wireless across Atlantic; i()i4, world war in Europe. 

"Other omissions. Detailed i)rovisions of various tarifT acts (but the 
meaning of tariff should be understoo<l); details of political campaigns 
except JefTerson's, Jackson's, Lincoln's and any current campaign in 
progress; critical study of political party principles (but give broad 
distinctions between chief rival parties); financial panics except those of 
1837, 1873, 1893." (Pp. 271 and 272.) 

liagley ('15) has been engaged on working out a possible scien- 
tific plan for determining the relative emphasis upon various 
portions of history by discovering the frequency of reference to 
persons and events made in current magazines and newspapers. 
He concludes that such a study is suggestive but doubts whether 
it may ser\'e as a final criterion for determining the amount of 
time and emphasis to be given to various phases of histor}'. 

Bagley and Rugg ('16) made a study of twenty-three text-books, 
published between 1865 and 191 2, by comparing the amount of 
space given to various topics in each book, by determining the 
shift in emphasis in the course of this period of time as measured 
by the space given to ditlerent topics, and by listing the topics and 
names included in all, or in a certain fraction of these texts. 

"(i) In so far as can be determined from the materials presentcti in 
the text-books, elementary .American history as taught in the 7th and Slh 
grades has been antl still is predominantly political and military history. 

"(2) Within the past fifty years, the emjihasis upon military' affairs 
as measured l)y the proportion of space devoted to wars has declined. In 
general, battles and campaigns are treated less in detail than was formerly 
the rule, while projxjrtionately more space is devoted to the causes and 
the results of the wars. The lessening emphasis upon details of the wars 
is first noticed in some of the text-books published between iSSi and 
1888, and the tendency has been general and decided since that time. 

"(3) The later books give a perceptibly hca\ner emphasis to the facts 
of economic and industrial development than do the earlier books, al- 
though iK)litical development still constitutes the essential core of ele- 
mentary historical instruction. 

"(4) .\s regards the treatment of six-cific eras or ejMx hs, the jirincipal 
increases in emj)hasis are to be noted in connection with: (a) the period 



HISTORY 425 

1783-1812 (especially in the treatment of the so-called 'critical period' 
between the close of the Revolution and the adoption of the Constitu- 
tion); (b) the non-military affairs of the period 1812-1861; and (c) Euro- 
pean events preceding and during the periods of discovery, exploration, 
and settlement. 

"(8) Numerous changes have taken place in the construction of ele- 
mentar}^ text-books in history during the past fifty years. The more 
important of these are: (a) a movement toward a simpler 'style' with 
larger emphasis upon clear statements of causal relationships: (b) the 
introduction and development of the 'problem' as a method of teaching 
history, and a consequent encouragement of 'judgment' as contrasted 
with rote memory, — of rational as contrasted with verbatim mastery; 
(c) a marked decline in the employment of imaginative pictures as illus- 
trations and an increase in the use of pictures that represent sincere at- 
tempts to portray actual conditions; (d) a marked decline in the use of 
anecdotal materials; (e) a larger and wider use of maps." (Pp. 56 and 
57.) 

Horn ('17) conducted an investigation after the manner of 
Bagley's magazine-newspaper method by checking through 
twenty-seven recent books on current industrial, political and 
social problems, in order to ascertain the facts, persons and dates 
referred to and the frequency of reference. His general impression 
of this inquiry is stated thus: 

"This investigation has not attempted to answer the question as to 
the complete content of the course of study in history. Neither does it 
assert that the purpose of history is to throw hght on modern social 
problems, or that this is even one of the chief purposes of studying history. 
Without regard to what the aims of teaching history are, this investiga- 
tion has been carried on to examine into the implications of one par- 
ticular assertion: namely, that history should render pupils more in- 
telligent with regard to modern conditions, problems and activities. If 
one assumes (i) that this is the function of history, (2) that the method of 
research here followed is satisfactory, and (3) that sufficient data have 
been collected, then there seems to be no escape from the conclusion that 
the present elementary and high-school courses of study in history are 
in very serious need of reconstruction." (Horn, '17, p. 171.) 



rHAPTFR XXTI 
MARKS AS MKASl ki;s (J I- SCHOOL WORK 

Importance of Marks. In order to determine the fruitfulness 
or wastefulness of methods of learning and teaching school sub- 
jects, it is necessary to evaluate the achievements of pupils as 
accurately as possible. Furthermore, the successful operation of 
a school demands an accounting of the work of its pupils. 

Marks have been the universal measures of school work. So 
many problems in the management of a school — credit, failure, 
promotion, retardation, elimination, graduation, honors, recom- 
mendations for positions, indeed the entire scholastic machinery 
of a school — hinge upon the assignment of marks that it is highly 
imperative to e.xamine in detail the value, accuracy and reliability 
of marks as well as to ascertain the possibility of some sort of 
standardization of marks. 

Variations among Teachers and Schools in the Distribution of 
Marks. The manner in which marks are distributeil to pupils 
varies enormously from teacher to teacher and from school to 
school. No one realized the seriousness of the situation until 
specific tabulations and comparisons were made. 

Meyer j)ublished the distribution of the marks assigned by 40 
different professors at the University of Missouri to their students 
during a period of five years as exhibited in Table 131. 



436 



MARKS AS MEASURES OF SCHOOL WORK 427 



TABLE 131. After Meyer ('08) 

Distribution of the marks of 40 teachers in the University of Missouri for a 
period of five years. The numbers are the percentages receiving the 
various grades. 

Teachers A 

Philosophy 55 

Latin 1 52 

Sociology 52 

Mathematics 1 40 

Economics 39 

Greek 39 

Latin II 36 

French 36 

Pohtical Science 34 

Mathematics II 32 

German 1 30 

Psychology 1 30 

German II 26 

Elocution 20 

Geology 22 

History 1 14 

Zoology 1 21 

Psychology II 19 

History of Art 25 

Bacteriology 20 

Freehand Drawing ... 18 

Chemistry 1 23 

English 1 21 

Astronomy 13 

History II 11 

Zoology II 24 

German III 22 

Chemistry II 9 

Education 18 

Mathematics HI 19 

Mathematics IV 25 

Physiology 20 

Anatomy 19 

Mathematics V 16 

Engineering 1 13 

Mechanical Drawing. . 18 

Mechanics 18 

Engineering II 16 

Chemistry III i 

English II 9 









Total No. 


B 


C 


F 


or Students 


33 


10 


2 


623 


42 


6 





130 


30 


13 


5 


958 


31 


16 


13 


208 


37 


19 


5 


461 


26 


24 


11 


287 


40 


19 


5 


577 


29 


25 


10 


295 


30 


27 


9 


592 


29 


23 


15 


145 


29 


20 


II 


586 


36 


24 


ID 


907 


3« 


25 


II 


941 


61 


19 





917 


48 


22 


8 


293 


53 


27 


6 


779 


45 


28 


6 


479 


47 


29 


5 


238 


40 


30 


5 


68s 


45 


31 


4 


263 


47 


25 


10 


506 


40 


31 


6 


205 


41 


30 


8 


964 


49 


33 


5 


225 


51 


33 


S 


806 


37 


31 


8 


250 


37 


28 


13 


441 


48 


43 





21 


38 


35 


9 


266 


36 


26 


19 


182 


29 


36 


10 


380 


33 


40 


7 


426 


34 


36 


II 


544 


34 


35 


15 


209 


36 


42 


9 


813 


29 


41 


12 


538 


26 


42 


14 


495 


26 


46 


12 


826 


II 


60 


28 


1.903 


28 


35 


28 


1,098 



428 EDUCATIONAL PSVCHOLOGY 

Similar tabulations have been published for Harvard University 
by Foster, Table i^^^; for the University of Wisconsin by Dearborn, 
Table 134; and for Cornell University by Finkelslein. Finkelstein 
('13) has shown the etTect of the personal c(iuation in the marks 
assigned to the same students in a year course which was in charge 
of one teacher in the lirst semester and of another in the second 
semester. 

TABLE 132. After Finkelstein 

No. OP 
Students Per Ce.vt Receivino the Various Marks 

0-30 40-44 45-49 50-54 55-59 60-64 65-69 
I St semester .. . 2(\\ .4 .4 j.,^ 4.5 1 i.^ 7 11 S 

and semester . . 257 — — 1.2 .8 1.5 12.1 12. 8 

No. OF 

Students 

70-74 75-79 80-84 85-89 90-94 95-100 E.\cm|)t 
I St semester, con. . . 263 16.7 15.6 20.2 6.4 5.7 .4 12.5 

2nd semester, con. . . 257 10.5 13.6 9.8 33.9 38 — 37-7 

The instructor in the lirst semester exempted from the tinal 
examination 12.5% of the class, while the instructor in the second 
semester exempted ,57.7^r- The latter obviously graded very 
much higher than the former. 



MARKS AS MEASURES OF SCHOOL WORK 429 



TABLE 133. After Foster ('11). 
Harvard College. Distribution of 8,969 grades. Elementary Courses 

Group I A% B% C% D% E% Abs.% Total 

Astronomy 16 13 45 19 6 i 69 

10 17 48 17 7 2 130 

Botany 11 28 38 14 2 7 183 

4 32 44 13 K 6 219 
Chemistry 6 26 45 12 9 2 334 

8 19 45 17 II o 319 
Economics 10 18 37 25 7 3 531 

7 19 43 21 7 3 436 

Engineering 11 15 31 28 12 3 114 

Engineering 10 13 27 21 21 9 121 

139 
129 

English I 13 52 28 3 3 603 

I ir 51 32 3 3 564 

Fine Arts 2 33 45 10 2 9 58 

6 27 67 o o o 49 
French 11 25 35 21 4 4 156 

12 19 36 19 10 4 145 

Geology 5 26 45 20 3 2 489 

5 25 33 28 2 7 8s 
Geology 2 28 48 10 7 5 122 

4 20 43 24 7 2 108 

German ■. . . . 7 21 31 26 11 4 259 

6 14 32 27 17 2 293 
Government 6 16 39 28 8 3 356 

9 23 37 21 7 2 419 
Greek 35 28 21 13 i 3 72 

IS 36 34 7 5 3 61 

History 7 20 44 21 5 2 347 

7 24 42 20 6 2 380 
Hygiene 18 29 $$ 18 i o 87 

8 23 48 14 4 3 139 
Latin 17 25 41 10 7 o 143 

15 27 41 5 10 2 128 

Mathematics 18 24 18 31 11 o 85 

14 22 31 23 II o 95 

Philosophy 7 31 -41 15 2 5 229 

7 23 61 8 o K 215 
Spanish 10 24 43 16 4 3 106 

7 13 38 33 8 2 119 

Zoology 2 13 48 30 5 I 149 

184 

Averages 7 20 42 21 7 3 213 



43° 



KDULATIONAL I•S^■C1IUL()(JY 



TABLE 134 

Percentages of prades assipnefl by 45 individual instructors in the University of 
Wisconsin lo freshmen and sophomores. After Dearborn ("lo; 

Ex. 
History: 

1 4-9 

2 9.8 

3 3 4 

4 7-4 

5 16.7 

6 9.1 

7 6.3 

Hnglish : 

8 19.3 

9 12.5 

10 6.4 

II 1.9 

12 7-4 

13 6.3 

14 40 

15 16.7 

16 7-3 

17 16.9 

18 II. 4 

19 8.9 

20 6.1 

Mathematics: 

21 16.6 

22 24. 1 

23 12.1 

French : 

24 21.2 

25 17-5 

26 22.3 

27 155 

28 14s 

29 23.9 

Physics and Chemistry: 

30 27.9 

31 21. 1 

Latin : 

32 I f ) . I 

33 n 7 

34 2<> . I 



G 


F 


P 




X 


No. OF 










Cases 


26.2 


32.8 


26.2 


9.8 


183 


52.9 


31-6 


3" 


2-5 


193 


22 


38.9 


26.7 


8.9 


558 


25.2 


37-4 


234 


6.5 


107 


52.4 


28.6 


2-3 





42 


39 4 


27 3 


18.2 


6.0 


33 


274 


30.8 


23.6 


II. 8 


237 


22.6 


19-3 


38.7 





31 


30 


37 5 


12 


5 


7-5 


40 


45 


312 


14 


7 


2.7 


109 


33-2 


45-5 


12 


9 


6.4 


202 


30 


6 


38.0 


19 





4 9 


121 


33 


7 


40 


14 


7 


5-2 


95 


38 


8 


32.6 


14 


3 


10.2 


49 


22 


8 


36.0 


17 


5 


7.0 


114 


39 


6 


375 


13 


5 


2.0 


96 


50 


6 


19 5 


II 


7 


1-3 


77 


42 


9 


27.2 


14 


3 


42 


70 


53 


9 


21.8 


7 


6 


7.6 


78 


25 


5 


235 


37 


8 


71 


98 


27.8 


28. 5 


14.2 


12.9 


302 


25 


15-7 


10. 4 


15-7 


108 


17.9 


193 


27-4 


233 


223 


43-8 


22.5 


7-5 


5 


80 


43-8 


275 


7 


5 


3-7 


80 


35-7 


21.4 


12 


5 


8.0 


I I z 


32 


27.2 


19 


4 


5-8 


103 


24.2 


24.2 


27 


4 


9.6 


62 


35-4 


24.6 


13 


8 


2.3 


130 


37.8 


21.1 


10.8 


2.4 


204 


35 


24.6 


156 


3-8 


289 


46 


JO. 7 


9.2 


8.0 


87 


4f). 2 


26 


ifi 





119 


47 


.6 


IS 9 


7 


•4 


2.8 


107 



MARKS AS MEASURES OF SCHOOL WORK 



431 



German: 

35 26.3 

36 12.0 

Zl 34-3 

38 II. 4 

39 17-3 

40 179 

41 147 

42 12.3 

43 29 . o 

44 21 .6 

45 22.4 



.80 



n 



.80 



J 



10 







34-2 
49.1 
40-3 
34-4 

35 

29-5 

27.4 

30.1 

42.0 

39-6 



21.9 
24.1 
19.4 
27.9 

29-3 
26 

33-5 
27.4 
22.6 
21 .6 
20.7 



12.3 



14 

12.5 

15s 



5-2 
3-7 
2.9 
3-2 
9-3 
4.0 

31 
10.9 

4-3 
2.27 

1-7 



114 

108 
67 
61 

75 
123 

95 
73 
93 
88 
S8 




70 75 80 85 90 95 100 

Fig. 85. — Distribution of all grades in a high school. After Gray ( '13, p. 66). 

These tables agree in showing extremely wide differences among 
teachers in the manner of giving marks. In the tabulation for the 
University of Missouri, one professor assigned the grade A to 55% 
of his students, and the grade F to only 2 %, while another prof es- 



432 



KDFCATIDXAL P.SVCIIOr.Of^.V 



sor assigni'cl the grade A to only i% of his students and the grade 
F to 28' <'. At Harvard, one professor gave the grarle A to 35% 
and the grade E to 1%, while another professor gave A to 1% and 
E to 32 ^/V' ^f his students. 

The situation in high schools is substantially the same. Gray 
tabulated the marks assigned by all the teachers in eight high 



«* 



J 



m 



JZ 



2 



70 75 UU 85 90 l<6 lOu 

Fig. 86.— Distribution of all grades in iinollicr high school, .\fter Gray ('13,11.67^. 

schools. The distributions of two of these schools are shown in 
Figures 85 and 86. The one grades high and the other grades low. 
In one school the great mass of the pupils receive 85 to 100; in the 
other they receive 85 to 70. 

Table 135 shows the distribution of the grades by the different 
teachers in a high school of about 150 pupils. 



MARKS AS MEASURES OF SCHOOL WORK 433 



TABLE 135 

Distribution of 8,490 grades by the different teachers in a high school during 
four years. (From a private report by Superintendent J. F. Waddell, 
Evansville, Wisconsin.) 

— 74 75-80 81-86 87-92 93-100 

English (1913-1914, 1914-1915) •• 33% 39% 13% 12% 3% 

English (1915-1916) 15 27 27 25 6 

Latin and German 5 15 13 $^ 34 

Mathematics (1913-1914) 22 31 18 21 8 

Mathematics (1914-1915) 23 27 24 13 13 

Mathematics (1915-1916) 5 12 15 28 40 

History (1913-1914) 11 25 22 3s 9 

History (1914-1915; 1915-1916). . 10 18 25 30 17 

Science (1913-1914) 11 36 25 26 2 

Science (1914-1915; 1915-1916) . . 10 33 24 19 14 

Domestic Science o 12 27 51 10 

Distribution of grades for the year after the above tabulation was made known 
to the teachers. Extreme variations are considerably reduced. 

English (1916-1917) 6% 32% 27% 27% 6% 

Latin and German (1916-191 7) ..9 25 22 29 15 

Mathematics (1916-1917) 5 20 28 27 20 

History (1916-1917) 2 17 37 34 10 

Science (1916-1917) 5 26 33 22 14 

Domestic Science (1916-1917) . . . o 10 40 41 9 

Variation among Teachers in the Evaluation of the Same School 
Products. A more direct and crucial method of examining the 
variations of teachers' marks than the tabulation of the grades as 
distributed by different teachers is to measure experimentally the 
differences in the values assigned by different teachers to the same 
pieces of work. 

Starch and ElUott ('12 and '13) made a series of investigations 
in which two final examination papers in first-year-high-school 
English were graded by 142 EngUsh teachers in as many high 
schools, one final examination paper in geometry was graded by 
118 teachers of mathematics, and one final examination paper in 
American history was graded by 70 teachers of history. The 
variations in these marks are shown in Figures 87, 88, 89 and 90. 
The differences are astounding; the marks for any given paper run 
practically over the entire range of the percentage scale ordinarily 
used. The marks of the first English paper run all the way from 



434 j.dllahonai. I'svciiolckjv 

64 to qS, of the second Kiiglish paper from 50 to 98. of the geometry 
paper from 28 to 92, and of the history paper from 43 to yo. 



• • • 

• • • 



• ••••••••• 

•••••••••••a 



66 70 75 80 85 90 d5 

Fig. 87. — Distribution of the marks assigned by 142 English teachers to a 
final examination paper in high-school freshman English. After Starch and 
EUiott ('12). 



I • • • • • • • 



6060 65V0 7ob0boil0 9o 

Fig. 88. — Distribution of the works assigned by 142 engiish teachers to an- 
other final examination paper in high-school freshman English. After Starch 
and Elliott ('12). 



•••»•■•••••• ••••••••• 



28 5a 55 60 65 70 75 80 85 90 

Fig. 8q. — Distribution of marks assigned by 114 mathematics teachers to a 
final examination paper in geometry. After Starch and Elliott ('13). 



40 6U ik) To NJ 00 

I'lC. 00. — Distribution of marks assigned by 70 history Ic.k hers to a tmal 
examination i)aiK.r in American history. After Starch and Elliott ('13). 



MARKS AS MEASURES OF SCHOOL WORK 435 

This investigation has established two conclusions: first, that 
teachers dififer enormously in evaluating the same pieces of work 
in terms of the ordinary percentage scale; and second, that they 
differ as much in one subject as in another. They disagree as 
much in evaluating a paper in mathematics as in Enghsh or his- 
tory. Apparently mathematical papers are not marked with 
mathematical precision any more than any other papers are. 

The author made a further investigation by having ten final 
papers in freshman Enghsh in the University of Wisconsin graded 
by ten instructors of freshman Enghsh. The marks are shown in 
the following table: 

TABLE 136. After Starch ('13) 
Marks assigned by ten instructors to ten final examination papers in English. 

COEFFI- 

Instructors CIENT 

Pa- Aver- Mean ofVaria- 

PERS 123456789 10 age Var. bilitv 

1 85 86 88 85 75 80 88 87 85 87 84.6 2.8 .034 

2 77 80 87 80 62 82 82 87 85 87 80.0 4.6 .057 

3 74 78 78 75 69 84 91 83 79 80 79.1 4.4 .056 

4 65 65 62 20 26 60 55 68 55 50 52.6 12.3 .233 

5 68 82 78 82 64 88 85 86 78 80 79.1 5.7 .070 

6 94 87 93 87 83 77 89 88 88 89 87.5 3.2 .036 

7 88 90 95 87 79 85 96 91 87 89 88.7 2.6 .029 

8 80 84 73 79 72 83 85 91 77 76 80.0 4.6 .058 

9 70 70 68 50 44 65 75 81 79 79 68.1 9.1 .118 
10 93 92 85 92 81 83 92 89 84 85 87.6 4.0 .045 

Av. 79.4 81.4 79.8 73.7 65.5 78.7 83.8 85.1 79.7 80.2 78.7 5.3 .074 

The variations shown in this table are practically as large as 
those found in the previous inquiry. It was thought that the wide 
range of marks shown in the first study might be due to the fact 
that the teachers were in different schools. However, Table 136 
shows that teachers in the same department differ almost as much. 
Less extensive results obtained by having various members of a 
department grade the same paper show that as much variation 
exists in other subjects as in English. 

Causes of Variation. Why do teachers differ so much in estimat- 
ing the worth of a given product and in the distribution of marks 
to groups of pupils? Four possible factors may be mentioned: 
(i) Differences in the standard of severity or leniency in different 
schools; (2) differences in the standard of severity or leniency of 
different teachers; (3) differences in credit or penalty assigned by 
different teachers to any given fact or error in a piece of work; and 
(d) minuteness of the discrimination between successive steps of 



436 EDUCATIONAL PSYCHOLOGY 

merit or quality in a given scale of qualities. How potent is each 
factor in producing the total variation in evaluating a given paper? 

The first thought that occurs in regard to the wide range of 
marks for the same ])apers as shown in P'igures S7 to 00, was that 
it must be due to the fact that these teachers were situated in 
different schools with different standards and ideals. It turns out, 
however, that factor one is relatively insignificant. If we compare 
the mean variation of the marks of the ten English papt-rs assigned 
by ten instructors in the same department with the mean variation 
of the marks of the two English papers assigned by teachers in 
different schools we can determine the part played by factor one. 
The mean variation of the former set of marks is 5.3 and of the 
latter is 5.4. Hence the mean variation of the marks assigned by 
teachers in the same department is only o.i less than the mean 
variation of marks assigned by teachers in different schools. 

The potency of factor two may be ascertained from the data in 
Table 136. The general average of all the grades assigned by the 
ten instructors to the ten pa])ers is 78. 7. If we comj)are the aver- 
age of all the marks given by any one instructor with the general 
average we obtain a measure of his particular standard of severity 
or leniency. Thus instructor 5 graded on the average 13.2 points 
lower and instructor 8 graded 6.4 points higher than the general 
average of all the teachers. If now we raise or lower each instruct- 
or's grades by as many points as the average of his grades is below 
or above the general average, we find that the mean variation of 
these weighted marks is 4.3. This mean variation is only i.o 
point smaller than the mean variation of the original unweighted 
marks. Hence factor two accounts for a relatively small share of 
the total variation. Factors three and four must then account for 
the remaining mean variation of 4.3. The strength of factor four 
can be determined e.x-ijcrimentally by having the same teachers 
re-grade their own i)apers without knowledge of their former 
marks. The author carried out such an e.\i)eriment and obtained 
the results e.\hibited in Table 137. 



MARKS AS MEASURES OF SCHOOL WORK 



437 



> 
o 
o 

sis 

s 






4 


Q 


00 ooo 00 00 CO C^ r^ C^cO 


lO 

00 


H 


O ro M lo lOOO oo <-o <^ IT) 

!>. O\00 t^ t^ I>.00 OO OCO 


o 

CO 


s > &: 
a "^ S 
S u o 


Q 




lO 


Q 
Z 


Ot^roi-<OwoOOOio 
t^ t^ i^oo CO O iJOOO O vo 


t^ 


H 


LO I7-J lO 

O O • looo O ■ M • o 


t^ 


1-3 U2 

w < n 

« > H 




lOMD o <^< M >-< r-. 


OO 

N 


Q 
Z 


loO CO "O O O lo iJO 
r^oo CO t^ t^oo O r^ 


00 


H 


O O 00 'i- r^ lo looo 

t^OOGO t^I^OO'OMS 




■o 


a Hi M 

H < a 

<l > H 

a o( g 

M M O 
H Z'^ 




';trO<^l<^)0'N(NrOO<~0 


M 


Q 
Z 


O rovoO t^ t^OO lOTj- 

MD i>- t^ C^o OO 00 lo t^vo 




H 


vo O r^co oi Ov^ co^or^ 
lot^r^oOOOOoO loi^vO 








M vO O O loco CO o t^ 


00 


Q 
Z 


H t^ t^ OO CO O t^ o 
loO ot^ooor^ooON 


CO 


H 


vOM MCO'^'~'*~'vO^O 


d 


o 

S 
w 




ro 


Q 
Z 


Ov rO '^0 IM 00 0\ OOO O O 
t^OO O OOO t^OO 00 I^OO 


00 


H 


lO t^ O O roOO rOOO OO fO 
00 00 O OOO t^ OOO t^oO 


lO 

LO 

CO 


h) 

, < 
>< > 

m OS 

> ^ 

o 


p 


w -^coi-H '-' Looio tJ-o 


rf 


Q 
Z 


I^ O O O <~0CO looo lOO 
00 00 00 OOO 00 t^OO CO OO 


00 


H 


00 t^oo 00 CO OOO OOO O 


00 

vO 
00 








< 



438 KDUCATioN'AL ps^'c^OLo^,^• 

The mean variation of these marks, comparing the first with the 
second for each jjaper, is 2.2 points. A part of this variation, 
however, is due to the slight shift in standard on the part of each 
teacher from one grading to the next. By appl)-ing the same 
process of weighting e.\])lained in connection with factor two, the 
mean variation droj)s from 2.2 to 1.75 points. Hence 1.75 repre- 
sents the amount of variation contributed by factor four to the 
total mean variation and by subtraction we find that factor three 
contributes 2.55 points. The four factors therefore contribute 
the following amounts: 

Factor one 10 p>omts 

Factor two i .00 " 

Factor three 2.55 " 

Factor four i , 75 " 

Total mean variation 5 40 " 

It is obvious then that factors three and four are the most im- 
portant ones in producing the large dilTerences of values assigned 
by teachers to a gi\en piece of school work. 

How Large Should the Units of a Marking Scale Be? The 
answer to this question depends jirimaril)- upon the fineness of the 
discriminations of successive degrees of quality in terms of the 
scale used, and, secondarily, ujjon the convenience of using a given 
scale of marks. The smallness of distinguishable shades of quality 
of anything can be determined by ascertaining the amount of 
dilTerence in terms of a gi\en scale that can be discriminated in 
the long run by the judges. 

A general principle that has been followed commonly in psycho- 
logical measurements, in which the units depend upon the discrimi- 
nation of judges, has been to regard a dilTerence in amount of the 
thing in cjuestion which can be distinguished correctly by 75'Jf of 
the judges or judgments concerning it as the snKillest psychological 
unit that can be used with reasonable certainty. For e.\amj)le, if 
we take ten shades of blue and ask 100 judges to arrange them in 
the order of blueness from left to right, we would regard that dif- 
ference in blueness between any two successive shades which 
75 of the judges agree in perceixing the one bluer than the other as 
the least that can be distinguished with a fair degree of confidence. 
The 75% l)oint is chosen because it is midway between ])ure 
chance and absolute certainty. If only one of two possible judg- 
ments ntay be made, that is, if a given shade is either more or less 



MARKS AS MEASURES OF SCHOOL WORK 439 

blue than another, then 50% of the judgments would be correct 
by pure chance guesses. If 100% of the judgments are correct, 
it means that the difference is so large that it can be recognized 
correctly every time, and the amount of difference may range 
anywhere from just large enough to be always recognized up to 
an infinite difference. 

According to this principle, how large would the steps be on 
the marking scale? For example, if we took two papers, a and b, 
in a given subject which differed in quality just enough so that 
three-fourths of the examiners or teachers would consider h better 
than a, how large would this difference be, say on the usual 100 
percentage scale? Data for answering this question with approx- 
imate accuracy are found in Figures 87 to 90. The probable 
error or median deviation of the marks given by the teachers to 
the papers represented in these four figures are 4.0, 4.8, 7.8 and 7.1 
respectively, with an average of 6.4. By definition, two times the 
probable error includes the middle half of the measures or marks. 
For example, in Figure 88 the median is 80.2 and the probable 
error is 4.8, that is, the middle 71 of the 142 marks lie between 75 
and 85. Obviously one-fourth or 35 of the marks lie above 85. 
Consequently so far as this particular paper is concerned the next 
better paper would have to be 4.8 points better so that three- 
fourths of the examiners would consider it better. 

Now the average probable error of the four sets of marks is 6.4. 
Hence the difference between two papers in general must be approx- 
imately 6.4 points so that three-fourths of the examiners would 
consider one better than another. On this principle then the step 
on the 100 percentage scale, with 70 as the usual passing grade, 
turns out to be approximately 7 points. This would produce a 
scale of steps as follows: 70-76, 77-84, 85-92, and 93-100. That 
is, the marking scale would have five steps, failure and four passing 
steps above 70. which may be designated as excellent, good, fair, 
poor, and failure, or perhaps preferably by the symbols A, B, C, D, 
and E. Such would be the size of the steps so that three- fourths 
of the examiners of a given set of papers would agree in distinguish- 
ing between the qualities of the papers. 

However, any individual teacher agrees with himself more 
closely in re-grading a set of papers than he agrees with other 
teachers, as indicated in Table 137. This table shows that the 
probable error or median deviation of a given teacher's marks in 
re-grading his own papers is approximately 2 points. By the same 



440 KDrCATIOXAL rSVCIIOLOGY 

reasoning the amount of difference in quality between two papers 
would have to be 2 points in order that an individual teacher would 
consider one paper bcttt-r than another in three out of four in- 
dependent niarkinj^s. Hence the marking scale for an individual 
teacher, who grades papers from his own \'iewpoint and compares 
them only with his own judgments, could have each step in a live- 
step scale subdivided into three smaller steps of about 2 jioints each 
by using the plus and minus. That is 70 to 76 would become 
70-72 or D - , 73-74 I), and 75-76 D-I-, and so on. 

Whether a fine marking scale such as the 100 percentage scale or 
a coarser five-step scale should be used is largely a matter of con- 
venience and personal habit. The advantage of a coarse scale is 
perhaps that it a\oids gi\ing the pupil the impression that the 
evaluation of a piece of work is more accurate than it actually is. 
The advantage of a fine scale is that it probably encourages the 
examiner in making as fine distinctions as possible. In practice 
a fine scale can probably be used as readily and as quickly as a 
coarse one if the teacher is accustomed to using a fine scale. .\ 
person may use as fine a scale as he wishes provided he recognizes 
the amount of the probable error in terms of the units of that par- 
ticular scale. In terms of a 100 percentage scale the probable error 
is about 6 or 7 points; in terms of the five-step scale it is about 
one step which is 6 or 7 times as large as a point on the percentage 
scale. The absolute amount of variation is substantially the same 
on the two scales. 

How Should Marks be Distributed to Groups of Pupils? If a 
five-step scale is used, what percentages of pupils should in the 
long run receive each of the five marks? The an.swer to this ques- 
tion that I advoaite is that the marks of large numbers of unse- 
lected pupils should be distributed approximately in conformity 
with the normal distribution or probability curve. Three lines of 
evidence for this position may be jiresented, the last two of which 
are fundamentally based upon the first: 

First, mental and i>hysical traits, when measured in large num- 
bers of indi\iduals. are distributed in a manner which yields a 
distribution surfaci- very near!}- identical with that of the proba- 
bility curve. Concrete evidence for this statement has been pre- 
stntifl in Chapter III, Figures 7 to 10, to which the reader should 
turn. It seems reasonable to infer that abilities in school subjects 
an- very probal)!}' distributed in the same manner as other mental 
traits. 



MARKS AS MEASURES OF SCHOOL WORK 44 1 

In the second place, when abilities in school subjects are meas- 
m^ed by objective methods, they are found to be distributed in 
very close conformity to the probability curve. Concrete evidence 
for this is presented in Figures 16 to 27, Chapter III. 

Thus, for example, the scores of 662 seventh grade pupils in the 
author's geography test shows the following distribution when the 
total range of the base line is divided into five equal sections: 

Scores 0-27 28-54 5S-8i 82-108 109-135 

% of pupils 6 24 37 24 9 

This is a remarkably close conformity to the theoretical dis- 
tribution proposed on the following pages. 

In the third place, the distribution of marks assigned by many 
teachers to large numbers of students conforms fairly closely to 
the normal distribution curve. When the marks of many teachers 
are combined, the idiosyncrasies of individual teachers tend to be 
counterbalanced. Tables 138 to 142, and Figures 91, 92, 93, 
and 94, show the distributions of marks in various institutions and 
the extent to which they differ from the theoretical probability 
curve. 

TABLE 138 

Distribution of grades in the College of Letters and Science, University of 
Wisconsin, for the years 1907, 1910 to 1915. From the reports of Presi- 
dent E. A. Birge. 

CoNDrriONED Ex- No. of 

Incomplete & Failed Poor Fair Good cellenx Grades 

Elementary Course . . 3.6 9.3 15.3 33.2 29.4 9.2 42,557 

Advanced Courses. . . 3.2 3.5 7.9 30.9 41.8 12.7 39,302 

TABLE 139 

Distribution of grades at Cornell University for the years 1902, 1903 and 19 11. 
Adapted from Finkelstein ('13, p. 22), to give the distributions for a five- 
point scale, 60 being the passing grade. 

Number of Grades 
0-59 60-69 70-79 80-89 90-100 
9.2 22.5 30.0 27.2 11. 1 20,348 

TABLE 140 

Distribution of all grades for two academic years at Harvard College. After 
Foster ('11, p. 262) 

Number 
Totals E% D% C% B% A% of Grades 

Elementary Courses 7 21 42 20 7 8969 

Intermediate " 4 13 37 28 12 2456 

Advanced " 2 2 13 38 36 476 



442 EDUCATIU.NAL l'.s\ CJiOl.UUV 

TABLE 141 
Distribution of grades at the University of Missouri. After Foster, p. 289 

Nl'MBER 

Ufiwn) E I) C H A or Grades 

Aug. 190S 3.5 15.6 8.7 41.2 23.3 7.7 

Feb. 1909 5.0 8.5 13.7 47.5 20.7 4.6 

June 1909 3.8 8.0 138 48.8 21.0 4.6 

Feb. 1910 3.5 6.5 14.4 4(> 6 21 3 47 24,979 

Averages 3.7 9.5 12.7 46.8 21 6 5.4 

First year after 

new system 

went into effect 9.0 14.5 50 21.7 4.9 11,342 

TABLE 142 

Average percentages for Cornell, Missouri, and the clcmcntar>- courses for 
Har\'ard and Wisconsin. These percentages do not total 100 because the 
incomplete grades for Wisconsin and Missouri arc not included. 

E D C B .\ N'cMBER UK Gkases 

8.7 17.9 38.0 24.5 8.2 Of',853 

If we grant that marks in the lon<^ run slioultl be assigned ac- 
cording to the normal distribution curve, wl^at jierccntage of 
])U})ils should receive each of the five stejis of the marking scale? 
If tlie base line of the probability curve in Figure 15, Chapter III, 
is di\ided into five ecjual di\isions, then the area above the various 
ilivisions would comprise the following i)ercentages of the total 
area: ' 

A, E.xcellent, or 93-100 = 7% 

B, Good, or 85- 92 = 24% 

C, Fair, or 77- 84 = 38% 

D, Poor, or 70- 76 — 24% 

E, Failure, or 69 = 7% 

Figures qi to 04 indicate how closely distributions of the marks 
at Wisconsin, Cornill, IIar\ard, and Missouri run i)arallel to the 
theori'tical cur\e. The only difference is a slight skewing to the 
ri|,'ht. Not (|uile as many D's are assigned and ver}' slif^hll}- more 
JC's and A's are assigned than the tlieorelical distribution would 
demand. Thus the marks as actually assigned by hundreds of 

' The ends of the prot)al)ility curve wiiuld readi the \>^m.- line only .it infinity. Hcnrc 
nn arliilrary point of termination must l)e selcctcti. This has heen placc<l at a iwint 
3.f>5 I*. K. values from the median. This [Kjlnt has l)een seleite<l hinausc it yields 7% 
for the K and .\ surfaces which is uppro.\imatcIy the |>erccntagc of pupils receiving thc-sc 
grades in many insliluUuns. 



MARKS AS MEASURES OF SCHOOL WORK 



443 



teachers to thousands of students furnish impressive support for 
the theory of the probability distribution of grades. 

Certain objections, however, both of a theoretical and a practi- 
cal kind, must be considered. In the first place, the soundness of 




Con. & Failed Poor Fair Good Excellent 

Fig. 91. — Distribution of 42,557 grades, broken line, in elementary courses 
in the College of Letters and Science of the University of Wisconsin. The con- 
tinuous Kne is the theoretical distribution. After a report by President E. A. 
Bilge. 

the theory rests on the supposition that the pupils are unselected, 
chance specimens of mankind as a whole. This supposition, of 
course, never obtains absolutely for any group of hmnan beings 
brought together anywhere. The very reason that brings any 
group together at the same time selects them. Pupils in school are 




0-59 60-69 70-79 80-89 90-100 

Fig. 92. — Distribution of 20,348 grades at Cornell University. After Fin- 
kelstein ('13). 

not random samplings of human beings of their respective ages — 
the less so as one goes up the educational ladder. The tendency 
is that every rung of the ladder selects on the whole slightly better 
and better specimens. The fact, however, seems to be that the 
selection which does take place is not of the sort that materially 
modifies the form of the distribution curve but rather tends to 
contract its base. The selection that does take place is not an 



444 



EDUCATIONAL PSYCHOLOGY 



abrupt cutting off, but a gradual slicing off along a large share of 
the distribution surface. 




E D C B A 

Fig. q3. — Distribution of 8,969 grades in elementary courses at Harvard Uni- 
versity. After Foster ( '1 1). 

The writer undertook to ascertain the actual elimination of 
university students as it really takes place on the basis of the records 
of 476 freshmen tabulated by Dearborn. It was found that the 




E D C B A 

Fig. q4. — Distribution of 24,979 grades at the University of Missouri. After 
Meyer ('08). 

following percentages of students dropiK'd out of (he University 
in the various grades of scholarship at the end of tlie freshmen and 
sophomore years: 

TAHLK I4J 

rnVDITIOMF.D 

& I'AiLEO Poor Fair Good Excellent 

PcrccntaKc of students of e.uh 
Rrade dropped durinR fresh- 
man year 100 52 19 1 1 o 

JVrcentaRe of those remaining 
in each Kr.uie, dropped during 
s<iphomorc year 45 16 80 



MARKS AS MEASURES OF SCHOOL WORK 445 

This table reads that all students whose average grade was 
"conditioned" or "failed" dropped out during the freshman year; 
52% of those whose grade was "poor" dropped out during the 
freshman year and 45% of those remaining whose grade was 
"poor" dropped out during the sophomore year, etc. It is obvious, 
therefore, that there is elimination from all classes of scholarship 
with the exception of the highest from which there is very little or 
no loss. The general effect of the actual elimination upon the 
distribution cxurve is to shift the left end of the curve tow^ard the 
right and to change the general form of it only slightly as indicated 
in Figure 95. 

The outcome of this evidence is that the distribution of the 
grades for the freshman year of the college as well as of the high 




65-68 69-72 73-76 77-80 81-84 85-88 89-92 93-96 97-100 



Fig. 95. — The continuous line shows the theoretical distribution of the marks 
of students. The upper broken Hne represents the change in this curve due to 
the dropping out of students during the freshman year. The lower broken Hne 
represents the change in the curve due to the elimination during the sophomore 
year. After Starch ('13). 

school should conform quite closely to the theoretical distribution 
curve and that slight shifts to the right should be made for the 
successive four years. It may seem curious to recommend that 
after the elimination of the successive years of the high school, the 
distribution to be followed in the freshman year of the college 
should be approximately normal again. The explanation is that 
the standards of the college are somewhat higher than those of the 
high school; so that, even if the high school should eliminate all of 
the poorest 7% of its pupils, the next poorest 7%, who are able to 
complete the high school, are likely to be unable to meet the de- 
mands of the college. 

The second objection urged by teachers against the adoption of 
the theoretical distribution of grades here recommended is that it 
would be unfair to lay down a rule that 7% of the pupils should 
be failed. How do we know; possibly by good teaching all pupils 



446 EDLCATIONAL PSVCHOLOGY 

Tiiav reach a sufTicienlly high attainment to be passed in the course. 
The answer to this statement is that the effects even of the best 
teaching will so rarely raise the attainments of pupils sufficiently 
hif^h so that none of the ])upils would fall below the jiassing grade; 
and furthermore, in the interests of reasonably high standards of 
scholarship the attainments of approximately -j^c of large num- 
bers of j)upils will very probably not merit a passing grade. There 
should be doubly good reasons for passing all students or for failing 
considerably less than 7%. Many of the cases of "good teaching' 
or "unusual classes" prove to be spurious when it is possible to 
check them up by outside means. 

A third point is not so much an objection as a question of practi- 
cal use of the principle of the distribution of grades; namely, in 
how large classes or groups of pupils should we expect fairly close 
conformity, and how close confonnity should be expected? The 
answer to the question which I shall give, on the basis of e\-|)erience 
in attempting to observe the jirinciple in the assignment of grades, 
is that for groups of students of 100 or more quite close conforniity 
should be ex])ected. By quite close conformity I mean a de\iation 
of not more than about 25'^'f, above or below the number of grades 
that theoretically should fall on a given step of the five-division 
scale. For example, the theoretical distribution demands that 
7% of the pupils should receive the grade of A or Excellent. For 
grou]is of 100 or more pupils, this i>ercentage should ordinarily 
not be lower than 5 nor higher than 9; the percentage of B's should 
ordinarily not run lower than iS nor higher than 30; the percentage 
of C's should not run higher than 48, nor lower than 2S, etc. The 
larger the number of pupils concerned, the closer the conformity 
should be. For groups smaller than 100 a wider latitude should be 
])ermjssible whenever there is genuine reason for wider deviation. I 
advocate conformity to the theoretical distribution within the limits 
of common sense with as much deviation as may seem permissible 
for good cause. However, really genuine reasons for large devia- 
tions, even with classes as small as 25 i)upils, unless obviou.sly 
selected by special cause, is much rarer than teachers ordinarily 
believe. 

In su])i)ort of this contention, the author ('15) reported an 
»x])eriment in which twenty-four comjiositions WTitten by sixth 
and seventh grade ]iu]iils were graded by 23 teachers according to 
the usual jK-rcentage method with 70 as a passing grade. After 
the papers had thus been graded, the teachers were requested to 



MARKS AS MEASURES OF SCHOOL WORK 447 

grade them according to a five point scale and give the grade of E 
to two papers, D to from four to six papers, C to from eight to ten 
papers, B to from four to six papers, and A to two papers. Even 
if the teachers felt, for example, that there were no papers good 
enough to receive the grade of A, they were to select the two best 
ones and call them A. The outcome was that those teachers who 
in their original grades differed most from the combined judgment 
of all the teachers were forced to comply more closely to the actual 
average marks as given in the first grading. One teacher marked 
the highest paper 85 in the original grading, and objected to giving 
it a grade of A in the forced distribution on the ground that no 
paper in the lot was good enough to receive so high a grade, and 
yet the average of the marks given by all the teachers to this 
paper was 92.9, the best paper in the entire group. 

The theory of the probability distribution of marks should be 
observed with sense and reason and not in a purely mechanical 
manner. A blind, unintelligent observance of the principle is 
bound to lead to injustice, particularly with small classes. In one 
such case which came to the author's attention it led to the giving 
of a mediocre grade to a pupil of very high ability. 

A fourth point frequently raised by teachers to justify unusually 
high or low marks is that the particular class in question is an 
unusually good one or poor one. Such a claim ought to be allowed 
only if it can be justified by good evidence. There are, of course, 
differences in classes, but these are almost never as great as we are 
inclined to believe. Large differences between successive classes 
in the same subject are for the most part illusory for the reason 
that the judgment of an individual teacher is more likely to deviate 
from a correct estimate than the average ability of a group devi- 
ates from the average of other groups. The teacher who says to 
each succeeding class that this is the best class he has ever had in 
this subject would possess, if this judgment were correct, a magic 
power for elevating the intellectual level of human beings. 

The feeling on the part of teachers that a given class is an un- 
usually good or poor class is quite often due to one or two unusu- 
ally good or poor individuals whose impression upon the mind of 
the teacher is outstanding, rather than to a higher or lower level 
of the class as a whole. 

As concrete evidence of the extent to which a teacher may err 
in such opinions and of the manner in which the opinion of the 
teacher may be checked up, the curves in Figure q6 are Dresented. 



44 



R I.DrCATKAAL rsvnioLOGV 



The contiiuiuus curve shows the distrihution of the pradcs of a 
teacher in Latin and German. When her attention was called to 
the predominance of hi<i;h marks, she claimed that her jiupils werj 
exceptionally K<>*«1- 1 1^*-" l^roken curve shows the extent to which 
her claim was unfounded since it shows that, according to their 
al)ilitics in other subjects, they Avere an average group. 

How May Variation in the Assignment of Grades be Reduced. 

I. 1})- a common sense compliance in the distribution of marks 
with the normal distribution or probability cur\'e. 

(a) To this end the administrator of a school should tabulate 
at stated intervals the marks assigned by each teacher and exhiljit 
the tabulation to the teachers. This in itself will usually lead 
without request or compulsion to a very considerable correction 




Fig. q6. — The continuous line shows the distril)Ution of the marks of a teacher 
of Latin and (Icrman in a high school. The broken line shows the distribution 
nf the murks of the s;ime pupils in their other subjects, .\ftcr an xuipublished 
report of Sui)l. J. I". Waddell, Evansville, Wisconsin. 

of aberrations on the part of those teachers who deviate most 
widely. At the University of Missouri the adojjtion of a i)lan of 
distriljution in conformity with the normal curve reduced the 
irregularity in grading in the ratio of five to two. 

(b) The teacher himself will find it useful to tabulate at frequent 
intervals the distribution of his grades. In making out the marks 
of a set of i)apers and particularly in making out the final grades 
for a course, the author has followed for se\-eral years Llie practice 
of plotting a distribution of the grades as tentatively made out. 
If the assignmint is decidedly abnormal in having considerably 
too many or too few of the dilTerent grades, a shift is made of 
borderline cases, unless there is an obvious reason to the contrary'', 
to obtain a reasonably normal distribution. Every teacher feels 
that there is a considerable number of cases concerning which he 
is in doubt as to whetlier they should have the one or the other 



MARKS AS MEASURES OF SCHOOL WORK 449 

grade. For example, if the tentative list of grades contains too 
many or too few A's, the lowest A's may be shifted to B's or the 
highest B's may be shifted to A's. 

2. Variability and uncertainty in grades may be reduced by 
adopting particularly in departments containing several teachers, 
a plan of giving certain weights or penalties for certain types of 
errors or defects. This should be done by departmental conference 
so as to secure a consensus of judgments on the various types of 
errors and amounts of penalties. Much could be done in this 
direction toward greater uniformity in methods of grading. If 
organizations of teachers would take this matter up, much of the 
chaos which is now present in methods of grading could be re- 
duced to order. 

At the present time A's or B's obtained from different teachers 
often mean quite different things. By observing the points here 
suggested they w^ould mean more nearly the same thing. Evalua- 
tion of achievement in terms of judgment depends obviously upon 
the judge. Marks as such will at best depend upon the examiner. 
They will probably always have to be used. More impersonal and 
objective methods for determining achievement in school work 
are being developed at the present time. To what extent these 
educational measuring devices will be able to replace the usual 
examinations and grades will depend upon their future develop- 
ment. 



BIBLIOGRAPHY 

Abbott, E. E.: 1909. On the Analysis of the Memory Consciousness in 
Orthography. Psychological Review Monograph, 11: No. i, 127- 

158. 
Arai, T.: 191 2. Mental Fatigue. Columbia University Contributions 

to Education, No. 54. 
Ash, I. E.: 1914. Fatigue and Its Effects upon Control. Archives of 

Psychology, No. 31. 
Ayres, L. P.: 191 2. A Scale for Measuring the Quality of Handwriting 

of School Children. Russell Sage Foundation, New York City. 
Ayres, L. P.: 1913. The Spelling Vocabularies of Personal and Business 

Letters. Russell Sage Foundation, New York City. 
Ayres, L. P.: 1915. A Measuring Scale for Ability in Spelling. Russell 

Sage Foundation, New York City. 

Bagley, W. C: 191 1. Educational Values. 

Bagley, W. C: 1915. The Determination of Minimum Essentials in 
Elementary Geography and History. Fourteenth Yearbook of the 
National Society for the Study of Education, 131-146. 

Bagley, W. C, and Rugg, H. O.: 1916. The Content of American 
History as taught in the Seventh and Eighth Grades: An Analysis of 
Typical School Text-books. University of Illinois Bulletin. 

Bair, J. H.: 1902. The Practice Curve. Psychological Review Mono- 
graph, No. 19. 

Baldwin, B. T.: 1914. United States Bureau of Education, Bulletin 
No. 581. 

Ballard, P. B.: 1915. Norms of Performance in Fundamental Processes 
of Arithmetic, Journal of Pedagogy, 3:9-20. 

Ballou, F. W.: 1914. Scales for the Measurement of English Composi- 
tion. Harvard University. 

Batson, W. H.: 1916. Acquisition of Skill. Psychological Review 
Monographs, 21, No. 3. 

Bean, C. H.: 191 2. The Curve of Forgetting. Archives of Psychology, 
No. 21. 

Bennett, C. J. C: 1907. Formal Discipline. Columbia University. 

Bennett, N. F.: 1916. The Correlations Between Different Memories. 
Journal of Experimental Psychology, 1:404-418. 

Bergstrom, J. A. : 1894. The Relation of the Interference to the Practice 
Effect of an Association. American Journal of Psychology, 6:433- 
442. 

4SI 



452 BIHLIUGRAPHV 

Ikils, (1. II.: iQoq. The Dislrilnilion ami Functions of Mental Imagery. 
Columbia University Contributions to Education, No. 26. 

Ik-Its, G. li.: 10 1 7. Class-Room Method and Management. 

Binet, II., and Simon, T.: i()05. (a) Sur la neccssite d'etabler un diagnos- 
tic scientific des etats inferieurs de rintelligence. .Vnnee Psycho- 
logique, 1 1 :i63-i()o. (b) Methodes nouvelles jwur le diagnostic du 
niveau intellectuel des anormau.x. Ibid., iyi-244. (c) Application 
des methodes nouvelles an diagnostic du niveau intellectuel chez 
cnfants normaiLX et anormaux d'hospice et d'ecole primaire. Ibiil., 
245-336. 

Boas, F. : 1896-1807. Growth of American Cliildren. Re{X)rt of United 
States Commissioner of Education, 1 1 :iss5- 

Bolton, T. L.: ic)02. Ueber die Beziehungen zwischen Ermiitlung, 
Roumsinn der Haut and Muskelleistung. Psychologische Arljeiten, 

4:i75--'34- 

Book, W. F.: iqoS. The Psychology of Skill: with Special Reference to 
its Acquisition in Typewriting. University of Montana, Publica- 
tions in Psychology; Bulletin No. 53, Psychological Series Xo. 1. 

Borst, M.: 1004. Recherchcs exp'llles sur I'educabilite et la fidelite du 
lemoignoge. .\rchives de Psychologic, 3:233-314. 

Hreslich, E. R.: 191 2. Teaching High School Pui)ils to Study. School 
Review, 20:505-515. 

Briggs, T. H.: 1913. Formal English Grammar as a Discipline. Teach- 
ers' College Record. 14, No. 41. 

Hriggs, T. II.: '13. The Right Way to Read. F'ducalion, 33:559-562. 

Brown, II. A.: 1014. Measurement of Efhcicncy of Instruction in Read- 
ing. Elementary School Teacher, 14:477-490. 

Brown, H. A.: 1916. The Measurement of .Mjility to Read. Department 
of Public Instruction, New Hampshire, Bulletin No. i. 

Brown, J. C.: 191 3. An Investigation on the \'alue of Drill Work in ll.e 
Fundamental Operations in Arithmetic. Journal of Educ. Psych., 
2:8i-S8; 3:561-570. 

IJrown, M. D., and Haggerty. M. E.: 191 7. The Measurement of Im- 
provement in English ComiK)sition. The ICnglish Journal. 6:51 5-527. 

JSrownc, C. E.: 1906. The Psychology of the Simple Arithmetical 
Proccs.scs. .\merican Journal of Psychology, 17:1-37. 

Bryan, W. L.. and Harter. N.: iS()7-iS99. Studies in the Physiolog>' and 
Psychology of the Telegrajihic Language. Psychological Review, 
4:27-53; 6:348-375. 

Buckingham, B. R.: 1913. Spelling .Ability: Its Measurement and 
Distribution. Teachers' College Record. 

liurk, C. F.: i<>oo. The Collecting Instinct. Pedagogical Scminan.-, 
7:179-207. 

iiurnett, C. J.: i(k>0. The Esliniation of Number. Harvard Psycholog- 
ical Studies, 2:349-404. 



BIBLIOGRAPHY 453 

Burt, C: 1909. Experimental Tests of General Intelligence. British 
Journal of Psychology, 3:94-177. 

Carter, M. H.: 1909. The Conservation of the Defective Child, Mc- 
Clure's Magazine, 33:160. 

Cattell, J. M.: 1886. Inertia of the Eye and Brain. Brain 8:294-309. 

CatteU, J. M.: 1886. Ueber die Zeit des Erkennens, u. s. w. Philos. 
Studien, 3:94-127. 

CatteU, J. M.: 1905. Examinations, Grades and Credits. Popular 
Science Monthly, 66:367-378. 

Cattell, J. M.: 1906. A Statistical Study of American Men of Science, 
III. The Distribution of American Men of Science. Science, New 
Series, 24:723-742. 

Charters, W. W., and Miller, Edith: 19x5. A Course of Study in Gram- 
mar. University of Missouri, Education Bulletin, No. 9. 

Colvin, S. S., and Myers, E. J.: 1909. The Development of Imagination 
in School Children and the Relation between Ideational Types and 
the Retentivity of Material Appealing to Various Sense Depart- 
ments. Psychological Review Monographs, 11: No. i, 85-126. 

Conrad, H. E., and Arps, G. P.: 1916. An Experimental Study in Eco- 
nomical Learning. American Journal of Psychology, 27:507-529. 

Cook, W. A. : 191 2. Shall we Teach Spelling by Rule? Journal of Educa- 
tional Psychology, 3:316:325. 

Cook, W. A., and O'Shea, M. V.: 1914. The Child and His Spelling. 

Coover, J. E.: 1916. Formal Discipline from the Standpoint of Experi- 
mental Psychology. Psychological Review Monographs, 20: No. 3. 

Coover, J. E., and Angell, F.: 1907. General Practice Effect of Special 
Exercise. American Journal of Psychology, 18:327-340. 

Cornell, W. S.: 1908. The Relation of Physical to Mental Defect in 
School Children. Psychological Clinic, 1:233. 

Cornman, O. P.: 1902. Spelling in the Elementary School. 

Counts, G. S.: 191 7. Arithmetic Tests and Studies in the Psychology 
of Arithmetic. Suppl, Educ. Monographs, i : No. 4. University of 
Chicago. 

Courtis, S. A.: 1910. Manual of Instruction. 82 Ehot Street, Detroit, 
Michigan. 

Dallam, M. T.: 191 7. Is the Study of Latin Advantageous to the Study 
of EngHsh? Educational Review, 54:500-503. 

Davidson, P. E.: 1914. The Recapitulation Theory and Human In- 
fancy. (Columbia University.) 

Davis, W. W.: 1898-1900. Researches in Cross-Education. Studies 
from the Yale Psychological Laboratory, 6:5, 6, and 8, 54-108. 

Dearborn, W. F.: 1906. The Psychology of Reading. Archives of Psy- 
chology, No. 4. 



454 liIllLl(H,l<.\l'HV 

Dearborn, W. F.: iqoq. The Cieneral Effects of Special Practice in 
Memory. Psychological Bulletin, 6:44. 

Dearborn, W. ¥.: lyog. The Relative Standing of Pupils in the High 
School and the University. Bulletin of the University of Wiscon- 
sin, Xo. 312. 

Dearborn, W. ]•'.: igio. E.vperiments in Learning. Journal of Educa- 
tional Psychology, 1:373-388. 

Dearborn, W. F.: igio. School and University Grades. Bulletin of the 
University of Wisconsin, No. 368. 

De Candolle, A.: 1873. (Second Edition, 1885.) Hisloire des sciences 
et des savants depuis deu.x siccles, etc. 

Delabarre, E. B.: iS{j8. .V Method of Recording Eye Movements. 
American Journal of Psychology', q. 

Dietze, G.: 1885. Ueber den Bewusstseinsumfang. 

Dodge, R.: 1900. Visual Perception During Eye-movements. l*sy- 
chological Review, 7:454-465. 

Dougherty, M. L.: 1Q13. Report of the Binct-Simon Tests (iiven to 
483 Children in the Public Schools of Kans;is City, Kansiis. Journal 
of Educational Psychology, 4:338-352. 

Downey, J. E.: igo8. Control Processes in Modified Handwriting: An 
E.xperimental Study. Psychological Review Monographs, g, Xo. i. 

Downey, J. E.: 19 10. Judgments on the Se.x of Handwriting. Psy- 
chological Review, 17:205-216. 

Drushel, J. A.: igi-. .\ Study of the .Vmount of .Arithmetic at the Com- 
mand of High School Graduates who have had no Arithmetic in 
their High School Course. El. School Journal, 17:657-661. 

Earle, E. L.: 1003. The Inheritance of the Ability to Learn to Spell. 
Columbia Contrib. to Phil., Psy., and Ed. 2, Xo. 2:4i-4.j. 

Ebbinghaus, H.: 1885. LU'ber das (iediichtniss. 

EI)biMghaus, H.: i8g7. Ueber cine neue Melhode zur IViifung geistiger 
Fiihigkeiten und ihre anwendung bei Schulkindern. Zeitschril'l fiir 
Psychologic, 1 3 :40 1 -4 5(). 

Ebert, E., and Meumann, E.: 1905. Ueber einige (Irundfragen der Psy- 
chologic des Gediichtnisses. Archiv fiir gesiimle Psychologic, 4:1-32. 

liidridge, R. C: 191 1. Six Thousand Common English Words. Xiagara 
Falls, Xcw York. 

ilrdmann, B., and Dodge, R.: igo8. Psychologischc Untcrsuchungen 
iiber das Lcsen, auf Experimenteller Grundlage. Halle. 

lillers, H. D.: 1917. Oral and Written I>rors in Grammar. Educa- 
tional Review, 54:458-470. 

I inkelstein, I. E.: i()i3. The Marking System in Theory and Practice. 

Foster, F. M.: i()i7. Tiie Results of a Recent Spelling Test at ilu* 
University of Iowa. Sihool and Society, 5:506-508. 



BIBLIOGRAPHY 455 

Foster, W. T.: 191 1. Administration of the College Curriculum. 
Fracker, G. C: 1908. On the Transference of Training in Memory. 

Psychological Review Monographs, No. 38:56-102. 
Freeman, F. N.: 190-. Untersuchungen iiber den Aufmerksamkeitsum- 

fang und die Zahlanffassung. Padagogische-Psychologische Ar- 

beiten, i :88-i68. 
Freeman, F. N.: 19 14. Practical Studies of Handwriting. Elementary 

School Teacher, 14:167-179. 
Freeman, F. N.: 1914. The Teaching of Handwriting. 
Freeman, F. N.: 1915. Handwriting. Fourteenth Yearbook of the 

National Society for the Study of Education. 
Freeman, F. N.: 1916. The Psychology of the Common Branches. 
Friedrich, J.: 1897. Untersuchungen iiber die Einfliissc der Arbeits- 

dauer und der Arbcitspausen auf die geistige Leistungsfahigkeit der 

Schulkinder. Zeitschrift fiir Psychologic, 13:1-53. 
Fulton, M. J.: 1914. An Experiment in Teaching Spelling. Ped. Sem., 



Galton, F. : 1869 and 1892. Hereditary Genius: An Inquiry into its Laws 
and Consequences. 

Galton, F.: 1883. Inquiries into Human Faculty and its Development. 

Gesell, A. L.: 1906. Accuracy in Handwriting, as Related to School 
Intelligence and Sex. American Journal of Psychology, 17:394-405. 

Gilbert, J. A.: 1894. Researches on the Mental and Physical Develop- 
ment of School Children. Studies from the Yale Psychological 
Laboratory, 11:40-100. 

Gilbert, J. A., and Fracker, G. C: 1897. The Effects of Practice in Re- 
action and Discrimination for Sound upon the Time of Reaction and 
Discrimination for other Forms of Stimuli. University of Iowa 
Studies in Psychology, 1:62-76. 

Gist, A. S.: 1917. Errors in the Fundamentals of Arithmetic. School 
and Society, 7:175-177. 

Goddard, H. H.: 191 2. The Kallikak Family. 

Goddard, H. H.: 1914. Feeble-mindedness: Its Causes and Conse- 
quences. 

Graves, S. M.: 191 7. A Study in Handwriting. J. of Educ. Psych., 
7:483-494. 

Gray, C. T.: 1913. Variations in the Grades of High School Pupils. 

Gray, C. T.: 191 5. A Score Card for the Measurement of Handwriting. 
Bulletin of the University of Texas, No. 37. 

Gray, W. S.: 1917. Studies of Elementary School Reading Through 
Standardized Tests. Suppl. Educ. Mo nog., i. No. i. 

Gulick, L. H.,and Ayres, L. P.: 1908. (Second Edition, 1913.) Medical 
Inspection of Schools. 

Guyer, M. F.: 1916. Being Well-Born. 



456 BIHLUXil^M'HV 

laggerty. M. E.: 1017. Scales for Rca(iing Vocabulary of Primary 

C'liildrcn. Elementary School Journal, 17:104-115. 
lahn, \V. H., and Thorndikc, E. L.: 1914. Some Results in Addition 

under School Conditions. Journal of Educ. Psych., 5:65-84. 
laines, T. 11.: 1916. Mental Measurements of the Blind. Psych. 

Rev. Monog., 21, Xo. i. 
lall, G. S.: 1883. Published first in Princeton Review, 2:249-272. 

Then rei)rinte(l in " .Aspect of Cliild Life and E<lucalion," 1907. 
lall, (i. S.: 1904. Adolescence, 
[all, (;. S.: 190.S. Youth. 
iaII-CJuest, A. L.: 191 7. Supervised Study, 
lamilton, V. M.: 1907. The Perceptual Eactors in Reading. Archives 

of Psychology, No.' 9. 
larris, J. II. and Anclfrson, H. W.: 1916. Measuring Primary Reading 

in Dubuque Schools (Iowa), 
[arris. L. H.: 1915. A Study in the Relation of Latin to English C'om- 

Ix)silion. School and Society, 2:251-252. 
[eck, W. IL: 1913. A Study of Mental Eatiguc. 
leck, W. IL: kh.S- A Second Study of Mental Eatigue in Relation to the 

Daily School Program. Psychological Clinic, 7:29-54. 
[enmon, V. A. C: 1910. Sex DilTcrcnces and \'ariabilily in Color Per- 
ception. University of Colorado Studies, 7, Xo. 4. 
[enmon, V. A. C: i9[2. The Relation between Mode of Presentation 

and Retention. Psychological Review, 19:79-96. 
lewins, N. P.: 1916. The Doctrine of Eormal Discipline in the Light of 

E.x'perimental Investigation, 
lillegas, M. B.: 191 2. A Scale for the Measurement of Quality in English 

ComiK)sition by "S'oung People. Teachers' College Record, it,, Xo. 4. 
[oge, Mildred, and Stocking, Ruth, J.: 191 2. .'\ Xote on the Relative 

\'alue of Punishment and Reward as Motives. Journal of .Xnimal 

Behavior, II, 4,3. 
lollingworth, II. L.: 191 2. Correlation of ;\bilities as .VfTected by Prac- 
tice; Journal of ^Educational Psychology, Jan., T912. 
lollingworth, H. L.: 1914. Individual DilTercnccs, During and After 

Practice. Psychological Review, January, 1914- 
lolloway, II. v.: 191 7. The Relative Difficulty of the Tables, Tren- 
ton, N. J. 
lorn, E.: 1917. Possible Defects in the Present Content of ,\nierican 

History as Taught in tlie Schools. Sixteenth VearlxHik of the 

National S<iciety for the Study of Education. 
[osic,J. !•".: 191 5. Ihe Essentials of Comix)sit ion and (iranunar. Eour- 

teenth YearlKxjkof the National Society for the Study of Education. 
Ilouser, J. D.: i(ji7. An Investigation of the Writing \'ocabularies of 

Representatives of an Economic Class. Elcm. School Journal, 

17:708-718. 



BIBLIOGRAPHY 457 

Howell, H. B.: 1914. A Fundamental Study in the Pedagogy of Arith- 
metic. 

Hoyt, F. S. : 1906. Studies in the Teaching of English Grammar. Teach- 
ers' College Record, 7, No. 5. 

Huey, E. B.: 1898 and 1900. Experiments in the Physiology and Psy- 
chology of Reading. American Journal of Psychology, 9:575-586; 
11:283-302; 12:292-312. 

Huey, E. B.: 1910. The Psychology and Pedagogy of Reading. 

Hull, C. L.r 1919. Quantitative Aspects of the Evolution of Concepts. 
Psychological Review Monograph Supplements, 191 9. 

Hull, C. L., and Montgomery, R. B.: 1919. An Experimental Investi- 
gation of Certain Alleged Relations between Character and Hand- 
writing. Psych. Rev., 26:63-74. 

Jacoby, P.: 1881. (Second Edition, 1904.) fitudes sur la selection dous 

ses rapports avec I'heredite chez I'homme. 
James, W.: 1890. Principles of Psychology. 
Javal, E.: 1878 and 1879. Sur la Physiologic de la Lecture. Annoles d' 

Oculistique. 
Javal, E.: 1879. Conditions de la lecture facile. Comptes rendus de la 

Societie de Biologie, page 8. 
Jessup, W. A.: 191 5. Current Practices and Standards in Arithmetic. 

Fourteenth Yearbook of the National Society for the Study of 

Education, 14:116-130. 
Johnson, R. I.: 1917. The Persistency of Errors in English Composition. 

School Review, 25:555-580. 
Jones, W. F.: 1913. Concrete Examination of the Material of English 

Spelling. 
''"Jones, W. F.: 1915. Spelling Book. 
Jost, A.: 1897. Die Assoziationsfestigkeit in ihrer Abhangigkeit von der 

Verteilung der Wiederholungen. Zeitschrift fiir Psycho logie, 14: 

436-472. 
Judd, C. H., and McAllister, C. N., and Steele, W. M.: 1905. Analysis of 

Reaction Movements. Psychological Review Monographs, 7:141-184. 
Judd, C. H.: 1908. The Relation of Special Training to General In- 
telligence. Educational Review, 36:28-42. 
Judd, C. H.: 1915. The Psychology of High School Subjects. 
Judd, C. IL: 19 16. Measuring the Work of the Public Schools. 
Judd, C. H. (Edited by) : 1916. School Survey, Grand Rapids, Michigan. 



4 



Kelley, T. L.: 1914. Educational Guidance. Columbia University. 
Kelly, F. J.: 1915. The Kansas Silent Reading Test. Bulletin of State 

Normal School, Emporia, Kansas. 
Kerr, Mary A.: 1916. The Effects of Six Weeks of Daily Drill in Addi 

tion. Indiana University Studies. 



45^ BIBLIOGR^\PHY 

King, I.: 1017. Comparison of Rapid and Slow Readers in Comprehen- 
sion. School and Society, 4:330-334. 
Kirby, T. J.: 1013. The Results of Practice under School Conditions. 

Columl)ia University. 
Kirkpatrick, E. A.: 1014. An Experiment in Memorizing vs. Incidental 

Learning. Journal of Educ. Psych., 5:405-413. 
Kline, L. \V.: iqoq. Some E.xperimcntal Evidence on the Doctrine of 

Formal Discipline. Hulletinof State Normal School. Duluth, Minn. 
Kraepelin. K., and I loch. A.: iSt>5. Uebcr die Wirkung der Theelx'stand- 

thcile auf korj)erliche und geistige Arbeit. Psychologische .\rbeiten, 

i:37S-488. 
Kraepelin, E., and Hylan, J. P.: 1Q02. Ueber die Wirkung kurzer 

.\rbeitszeiten. Psychologische Arbeiten, 4:454-494. 
Kraepelin, E., and Rivers, \V. H. R. : 1S96. Ueber Ermiiilung und Er- 

holung. Psychologische Arbeiten. 1:378-488. 
Kraepelin, E.: 1902. Die .\rbeitscurve. Philsophische Studien, 19:450- 

507; 
Kraepelin, E.: 1903. Ueber Krniuilungsmessungen. Archi\e fiir die 

gesamte Psychologic, i :9-3o. 
Kuhlmann, F. : 191 2. Revision of the Binet-Simon System for Measuring 

the Intelligence of Chiklren. Journal of l'sycho-.\sthenics, Mono- 

grai)h Supplements, i. No. i. 

Laser, IL: 1S94. Uebcr geistige Ermiidung beim Schulunterricht. 
Zeitschrift fiir Schulgesundheitspflege, 7:2-22. 

Lay, \V. A.: 1S9S. Fiihrer (lurch den ersten Rechenunterricht. Wies- 
baden. 

Lay, W. .\.: 1907. Fiihrer durch den Rechenunterricht der Unterstufe. 

Laudolt: 1891. Nouvelles recherches sur la physiologic des mouvements 
des yeux. Archives d'Ophthalmologie, 2:385-395. 

Lindley, E. II. : 1897. A Study of Puzzles, .\merican Journal of Psy- 
chok)gy, 8:431-493. 

Llewelyn, E. J.: 1916. Reading in the Mt. \ernon Schools. Elementar>' 
School Journal, 17:123-127. 

Lodge, ().: 1915- Chapter X. Principles of .Second;ir)' Education. 
Edited by Paul Monroe. 

1 xiisct te, A. : 1 899. .Assimilat ive Memory. 

I^)well, A. L. : 1910. College Stuilies and the Professional School. 
Harvard Graduates' Magazine, Dec, 1910; also Educational Re- 
view, Oct., 191 1. 

Lueba, J. H., and Hyde, W.: 1905. An Experiment in Learning to Makt 
Hand Movements. Psychological Review, 12:351-369. 

Lyon, I). ().: 1913. Relation of Length of Material to Time Taken for 
Learning and the Ojjtinumi Distribution of Time. Journal of Edu- 
cational Psychology, 5:155-163. 



BIBLIOGRAPHY 459 

Matthews, B.: The Study of Fiction in Counsel upon the Reading of 
Books by Henry VanDyke. 

Mead, C. D.: 1915. Silent versus Oral Reading with 100 Sixth Grade 
Children. Journal of Educational Psychology, 6:345-348. 

Mead, C. D.: 191 7. Results of Silent versus Oral Reading. Journal of 
Educational Psychology, 8:367-368. 

Mead, C. D., and Sears, Isabel. 1916: Additive, Subtraction and Mul- 
tiplicative Division Tested. Journal of Educ. Psych., 7:261-270. 

Messenger, J. F.: 1903. The Perception of Number. Psychological 
Review Monographs, 5:123-144. 

Meumann, E. : 1914. (First Edition, 1907.) Vorlesungen zur Einfiirhung 
in die Experimentelle Padagogik und ihre Psychologischen Grund- 
lagen. Vols. I, II, and III. 

Meyer, M.: 1908. The Grading of Students. Science, 28:243-250. 

Minnick, J. H.: 1913. An Experiment in the Supervised Study of 
Mathematics. School Review, 21:670-675. 

Monroe, W. S.: 1917. A Preliminary Report of an Investigation of the 
Economy of Time in Arithmetic. Sixteenth Yearbook of the Na- 
tional Society for the Study of Education, 1917:111-127. 

Mosso, A.: Fatigue. 

Munn, A. F.: 1909. The Curve of Learning. In Studies in Development 
and Learning. Edited by E. A. Kirkpatrick, Archives of Psychology, 
No. 12. 

Nanu, H. A.: 1904. Zur Psychologie der Zahl-Auffassung. Wurzberg. 
Nicholson, F. W.: 191 5. Success in College and in After Life. School 
and Society, 2:229-232. 

Oberholtzer, E. E.: 1914. Testing the Efficiency of Reading in the 
Grades. Elementary School Journal, 15:313-322. 

Partridge, E. A.: 191 5. High-School Latin — A New Phase of an Old 
Subject. Classical Journal, 10:404-412. 

Pearson, K.: 1904. On the Laws of Inheritance in Man. II. On the 
Inheritance of the Mental and Moral Characters in Man, etc. 
Biometrika, 3 :i3i-i90. 

Pearson, K. : 1907. On the Relationship of Intelligence to Size and Shape 
of Head, and to Other Physical and Mental Characters. Biometrika, 
5:105-146. 

Perkins, A. S.: 1914. Latin as a Vocational Study in the Commercial 
Course. Classical Journal, 10:7-16. 

Peters, C. C: 191 7. Influence of Speed Drills on Silent Reading. Jour- 
nal of Educational Psychology, 8:350-366. 

Peterson, H. A.: 191 2. Note on a Retrial of Professor James' Experi- 
ment on Memory Training. Psych. Rev., 19:491-492. 



460 BIBLIOGRAPHY 

I'lielps, C. L.: IQ13. A Study of Errors in Tests of Adding Ability. 

Elem. School Jourii;il, 14:20-39. 
I'1k1[)s, W. L.: Teaching in School and College, 
riiillips, F. M.: 1913. \"alue of Daily Drill in .Xrithmctic. Journal of 

Educ. Psych., 4:150. 
I'inlncr, R.: 1913. Oral and Silent Reading of Fourth Grade Pupils. 

Journal of Educational Psychology, 4:333-337. 
Pintner, R., and Paterson, D. G.: 1916. Psych. Rev. I^fonog., 20: No. 4. 
Pintner, R., and Paterson, D. G.: 1917. A Scale of Performance Tests, 
i'roctor, M.: 191 7. Supervised Study on the Pacific Coast. School 

and Society, 6:326-327. 
i'ryor, II. C: 1915. Spelling. Fourteenth Yearbook of the National 

Society for the Study of Education, 1915:78-89. 
PulTer: 191 2. The Boy and His Gang. 
Pylt-', W. H.: 1913. Economical Learning. Psychological Bulletin, 10:73. 

Radassawljewitsch, P. R.: 1907. Das Behaltcn und Vergesscn bci 
Kindern und Erwachscnen nach e.xperimentellen Unlersuchungen. 

Rice, J. M.: 1897. The Futility of the Spelling Grind. The Forum. 23: 
163-172, 409-419- 

Rice, J. M.: 1902. Educational Research: A Test in /Vrithmetic. The 
Forum, 34:281-297; Causes of Success and Failure in Arithmetic. 
The Forum, 34:437-452. 

Riley, J. L. : iqoS. The Springfield Tests, 1846-1005, 1906. 

Ritter, C: 1900. Ermiidungsmessungen. Zeitschrift fiir Psychologie, 
etc., 24:401-444. 

Ruedigcr, 1907. The Field of Distinct Vision. Archives of Psychology, 
No. 5. 

Rugcr, H. A.: 1910. The Psychology of Efficiency. Archives of Psy- 
chology', No. 15. 

Rugg, II. O. : TO 16. The E.xperimental Determination of Mental Dis- 
cipline in School Studies. 

Sackett, L. W.: Measuring a School System by the Buckingham Spelling 
Scale. School and Society, 2:860-864, 894-898. 

Schmidt, W. A.: 191 7. An E.xperimental Study in the Psychology of 
Reading. Suppl. Educational Monographs (Chicago), i, No. 2. 

Schuster and l'>l(lerton: 1907. Inheritance of Ability. Publications of 
Eugenics LalKiratory, Ix)ndon. 

Schuyler, W., and Swift, E. J.: 1907. The Learning Process. Psy- 
chologic.il Bulletin, 4:307-310. 

Scrii)ture, E. W., Smith, T. L., and Brown, E. AL: 1S94. On the Educa- 
tion of Muscular Control and Power. Studies from the Yale Psy- 
chological Lalxjratory, 2:114-119. 

Sears, J. B.: 1915. Spelling Efllcicncy in the Oakland Schools. School 
and Society, 2:531-537. 569-574- 



BIBLIOGRAPHY 461 

Sears, Isabel, and Diebel, Amelia: 191 6. A Study of Common Mistakes 

in Pupils' Oral English. El. School Journal, 17:44-54. 
V' Seashore, C. E.: 1901. Suggestions for Tests on School Children. Educa- 
tional Review, 22:69-82. 
Seashore, C. E., and Kent, G. H.: 1905. Periodicity and Progressive 

Change in Continuous Mental Work. Psychological Review 

Monographs. 
Sikorski, J.: 1879. Sur les effects de la lassitude provoquee par les 

travoux intellectuels chez enfants le I'age scoUaire. Annales d'hy- 

giene publique, 2:458-464. 
Simpson, B. R.: 191 2. Correlations of Mental Abilities. Columbia 

University. 
Sleight, W. G.: 191 1. Memory and Formal Training. British Journal 

of Psychology, 4:386-457. 
Smledey, F.: 1900-1901. Rept. Dept. Child Study and Pedagogic In- 
vestigation, No. 3. 
Smith, F. O. : 191 2. A Rational Basis for Determining Fitness for College 

Entrance. University of Iowa Studies in Education, N. S. 51. 
Sneed, C. M., and Whipple, G. M.: 1909. An Examination of the Eyes, 

Ears, and Throats of Children in the PubUc Schools of Jefferson 

City, Missouri. Psychological Clinic, 2:234-238. 
Spillman, W. J.: 1909. The Country Boy. Science, 29:739-741; 30:405- 

407. 
Starch, D.: 1910. A Demonstration of the Trial and Error Method of 

Learning. Psychological Bulletin, 7:20-23. 
Starch, D.: 191 1. Experiments in Educational Psychology, 191 1. 

Revised, 191 7. 
Starch, D.: 1911. Unconscious Imitation in Handwriting. Psychological 

Review, 18:223-228. 
Starch, D.: 191 1. The Transfer of Training in Arithmetical Operations. 

Journal of Educational Psychology, 2:306-310. 
Starch, D.: 191 2. Periods of Work in Learning. Journal of Educational 

Psychology, 3:209-213. 
Starch, D. : 1913. The Measurement of Handwriting. Journal of Educa- 
tional Psychology, 4:445-464. 
Starch, D.: 1913. Reliability and Distribution of Grades. Science 

38:630-636. 
Starch, D.: 1913. Correlation Among Abilities in School Studies. Jour- 

• nal of Educational Psychology, 3:415-418. 
Starch, D.: 191 5. The Measurement of Efficiency in Reading. Journal 

of Educational Psychology, 6:1-24. 
Starch, D.: 1915. The Measurement of Efficiency in Handwriting. 

Journal of Educational Psychology, 6:106-114. 
Starch, D.: 191 5. The Measurement of Efficiency in Spelling. Journal 

of Educational Psychology, 6:127-186. 



462 biblio(;kapiiv 

Starch, D.: 191 5. The Measurement of Ability in English Grammar. 

Journal of Etlucational Psychology, 6. 
Starch, D.: 191 5. Can the Variability of Marks be Reduced? School 

and Society, 2:242-243. 
Starch, D.: 191 5. The Inheritance of Abilities in School Studies. School 

and Society, 2:608-610. 
Starch, D.: 191 5. Some Experimental Data on the Value of Studying 

Eoreign Languages. School Review, 23:697-703. 
Starch, D.: 1916. The Measurement of Ability in Arithmetic. Journal 

of Educational Psychology, April, 1916. 
Starch, D.: 1917. The Estimated Value of School Subjects. School an<l 

Society, 5:59-60. 
Starch, D.: 191 7. Eurther Experimental Data on the Value of Studying 

Foreign Languages. School Review, 25:243-248. 
Starch, D.: 1Q17. Similarity between Brothers and Sisters in Mental 

Traits. Psychological Review, 24:235-238. 
Starch, D., and Ash, I. E.: 191 7. The Mental Work Curve. Psycholog- 
ical Review, 24:391-402. 
Starch, D., and Elliott, E. C: 1912. The ReliabiHty of Grading High 

School Work in English. School Review, 20:442-457. 
Starch, D., and Elliott, E. C: 1913. The ReliabiHty of Grading Work in 

Mathematics. School Review, 21:254-259. 
Starch, D., and Elliott, E. C: 1913. The Reliability of Grading Work in 

History. School Review, 21:676-681. 
Stone, C. W.: 1908. Arithmetical Abilities and Some Factors Deter- 
mining Them. Columbia Contributions to Education. No. 10. 
Studebaker, J. W.: 1916. Economy Practice Exercises in Arithmetic. 

Scott, Foresman & Company. 
Suzzallo, H.: 1910. The Teaching of Spelling. 
Suzzallo, H., and Pearson, H. C: 1913. Comparative E.xperimcnlal 

Teaching in Spoiling. Teachers' College Record, 13: No. i. 
Swift, E. J.: 1908. The Mind in the Making. 
Ssvift, E. J.: 191 2. Vouth and the Race. 

Taussig, A. E.: h)oo. The Prevalence of \'isual and .\ural Defects among 

the Public School Children of St. Louis County, Missouri. The 

Psychological CHnic, 3:149-160. 
Terman, L. M.: 1916. The Measurement of Intelligence. 
Terman, L. M.: 191 7. The Intelligence Quotient of Francis Gallon in 

Cliil(llux)d. American Journal of Psychology. 28:.mo. 
Thompson (Woolley), H. H.: i()03. Psychological .Norms in Men and 

Women. University of Chicago Contributions to Pliilosopliv, 

IV, No. I. 
'Ihorndikc, E. L.: 1900. Mental Fatigue, P.sychological Review. ::4r(- 

4«^ 547-579- 



I 
i 



BIBLIOGRAPHY 463 

Thorndike, E. L.: 1905. Measurements of Twins. Archives of Phil. 

Psychology and Scientific Methods, No. i. 
Thorndike, E. L.: 1910. Handwriting. Teachers' College Record, 11: 

No. 2. 
Thorndike, E. L.: 1910. Practice in the Case of Addition. American 

Journal of Psychology, 21 :483-486. 
Thorndike, E. L.: 191 1. Mental Fatigue. Journal of Educational 

Psychology, 12:61-80. 
Thorndike, E. L.: 191 2. Education. 

Thorndike, E. L.: 191 2. The Permanence of Interests and Their Rela- 
tion to Abilities, Popular Science Monthly, Nov., 191 2. 
Thorndike, E. L. : 191 2. The Curve of Work. Psychological Review, 

19:165-194. 
Thorndike, E. L.: 1914. Educational Psychology, Vol. I. The Original 

Nature of Man. Vol. II. The Psychology of Learning. Vol. III. 

Mental Work and Fatigue and Individual Differences and Their 

Causes. 
Thorndike, E. L.: 1914. The Measurement of Abihty in Reading. 

Teachers' College Record, 15, No. 4. 
Thorndike, E. L.: 191 5. The Disciplinary Values of Studies: A Census 

of Opinions. Education, 35:278-286. 
Thorndike, E. L.: 191 5. The Relation between Speed and Accuracy in 

Addition. Jour, of Educ. Psych., 5:537. 
Thorndike, E. L.: 1917. The Curve of Work and the Curve of Satisfy- 

ingness, Journal of Applied Psychology, 1:265-267. 
Trabue, M. R.: 1916. Completion- test Language Scales. Columbia 

University Contributions to Education, No. 77. 

Uhl, W. T.: 1916. Use of the Results of Reading Tests as Basis for 
Planning Remedial Work. Elementary School Journal, 17:266- 

275- 

Waldo, K. D.: 1914. Tests in Reading in the Sycamore Schools, Ele- 
mentary School Journal, 15:251-268. 

Walhn, J. E. W. : 191 1. Spelling Efficiency in Relation to Age, Grade and 
Sex, and the Question of Transfer. 

Walsemann, H. J.: 1907. Anschauungslehre der Rechenkunst. 

Warren, H. C: 1897. The Reaction Time of Counting. Psychological 
Review, 4:569-591. 

Webb, L. W.: 191 7. Transfer of Training and Retroaction. Psycholog- 
ical Review Monographs, 24: No. 104. 

Wells, F. L. : 1908. Normal Performance in the Tapping Test. American 
Journal of Psychology, 19:437-483. 

Wells, F. L.: 191 2. The Relation of Practice to Individual Differences. 
American Journal of Psychology, 23:75-88. 



464 BIBLIOGRAPHY 

Wfst, A. F. (Editor): 191 7. \'aluc uf the Classics, Princeton Univcrsily 
Press. 

Whipple, G. M.: igio. Maiiiuil of Mental and Physical Tests. 

W hippie, G. M.: 1916. How to Study ElTectively. 

Whitley, M. T.: 191 1. An Empirical Study of Certain Tests fur ! 
dividual Differences. Archives of Psychology, No. ig. 

W illjur, Flora: 1916. E.xperiments with Courtis' Practice Pads. Indiana 
University Studies. 

Wilcox, M. J.: 1917. Does the Study of High School Latin improve 
High School English?" School and Society, 6:58-60. 

Wilson, G. M.: 1917. A Survey of the Social and Business Use of .\rith- 
metic. Sixteenth Yearbook of the National Society for the Study 
of Education, 1917:128-142. 

Wimmer, IF: 191 6. An Experimental Study of the Effects of Drill in 
Arithmetical Processes under \'arying Conditions. Indiana Univer- 
sity Studies. 

Winch, W. IF: 1904. Immediate Memory in School Children. British 
Journal of Psycholog)', 1:127-134. 

Winch, W. H.: 1906. Immediate Memory in School Children: Auditory. 
British Journal of Psychology, 2:52-57. 

Winch, W. H. : 1908. The 'Fransfer of Improvement in Memory in School 
Children: I. British Journal of Psychology, 2:284-293. 

Winch, \V^ H.: 1910. The Transfer of Improvement in Memory in 
School Children: IF British Journal of Psycholog>\ 3:386-405. 

Winch, W. H.: 1910. Accuracy in School Children. Does Improvement 
in Numerical Accuracy Transfer? Journal of Educational Psy- 
chology, I ■.S57-5^9- 

W'inch, W. IF: igii. Further Work in Numerical .\ccuracy. Journal of 
Educational Psychology. 2:262-271. 

Winshij), A. E.: i()00. Jukes-Edwards: A Study in Education and 
Heredity. 

Woodworth, R. S.: 1899. The Accuracy of \'oluntary Movement. Psy- 
chological Review Monographs, No. 13. 

Woodworth, R. S., and Thorndike, E. L.: looi. The Influence of Im- 
provement in One Mental Function ujxin Other P\mctions. Psy- 
chological Review, 8:247-261; 384-305; 553-564. 

Woods, F. A.: 1906. Mental and Moral Heredity in Royalty. 

Woody, C: 1916. Measurements of Some .\chievements in .Vrithmetic, 
Columbia University Contributions to Education, No. 80. 

^'erkcs, R. M., Bridges, J. M., and Hardwick, R. S.: 1015. A Point- 
Scale for Measuring Mental Ability. 



INDEX 



Abbott, E. E., 167. 

Aiken, 265. 

Aldine system of reading, 290, 291, 
293, 294. 

Ames, W. R., 263, 264. 

Anderson, H. W., 290. 

Angell, 198, 200. 

Apperception, doctrine of, 139; eval- 
uation of, 140. 

Arai, T., 173, 174. 

Arithmetic; steps involved in, 374f.; 
ntmiber concept, 375; establish- 
ment of associations, 376; clue as to 
process required, 376f.; number 
preferences, 379f.; methods of meas- 
uring efficiency in, 38if.; economic 
methods in learning, 383f.; acquiring 
number concept, 383f.; number pic- 
tures, 384f.; operations to be learned, 
386f.; length of class period, 391; 
certain environmental factors, 395f.; 
drill, 396f.; optimum distribution of 
drill, 402f.; special methods of drill, 
40Sf.; speed versus accuracy, 409f.; 
limits of attainment, 410; errors, 
4iof.; relative difficulty of combina- 
tions, 414. 

Arithmetic tests: 

Courtis, 20, 35, 37, 89, 165, 166, 381, 

400, 408, 410, 412. 
Starch, 382, 383, 396. 
Stone, 382. 
Woody, 382. 
Judd and Counts, 382. 

Arithmetical reasoning, 20; Fig. 4, 
Fig. s, 21; Fig. 19, 35. 

Arps, G. F., 407. 

Association, see Learning, Spelling, 
Correlation. 

Attention, largely controlled by in- 
stinct, 13; control of in studying, 
180. 



Auditory defects, frequency, 128. 
Aussage tests, 133. 

Ayres, scale in writing, 35; in spelling, 
324- 

Bagley, W. C, 224, 247, 424, 425. 

Bair, J. H., 161, 198. 

Baldwin, B. T., 59, 60. 

Ballard, P. B., 407, 408. 

Ballou, F. W., 353. 

Barr, Martin W., 80. 

Batavia plan, 42. 

Batson, W. H., 146, 147, 150, 161. 

Beacon system of reading, 290, 291, 
292, 293. 

Bean, C. H., 157. 

Beetz, 384, 385. 

Bergstrom, J. A., 200. 

Betts, G. H., 167, 423. 

Biglow, R. P., 218. 

Binet, H., 99, 300. 

Binet-Simon Tests, 69f.; 99f.; Ter- 
man's revised scale, loof., 107; nec- 
essary qualifications for tester; 
measurements resulting from use of, 
io7f. 

Birge, E. A., 441, 443. 

Boas, F., 18. 

Bolton, T. L., 171. 

Book, W. F., 144, 145, 146, ISO, 157, 
i6r. 

Bom, 384, 385, 386. 

Born, Busse, and Belune, 384. 

Borst, M., 133, 136. 

Breslich, E. R., 189. 

Briggs, T. H., 225, 226. 

Brown, H. A., 274, 275, 289. 

Brown, J. C, 398. 

Brown, M. D., 353, 354, 372. 

Brown, J. Stanley, 189. 

Brinckerhofif, Morris arid Thorndike, 
55- 



465 



iC)() 



L\DEX 



Bn-an, W. L., and Harter, N., 142, 

14^?. 150, 151, iS3> 158- 
Buckingham, B. R., 325. 
Burk, C. F., 22. 
Burnett, C. J., 384. 
Burns, 55. 
Burt, C, 53, 54, 59, no. 

Cambridge plan, 44. 

Capacities, definition of, 10; variation 

in, sec Individual Differences. 
Carter, M. H., 123. 
Cattell, J. M., Q2, 384. 
Charters, \V. W., 357, 363. 
Charters, W. W., and Miller, Edith, 

362. 
Classical versus non-classical students, 

236, 237f. 
Cliborne, J. H., 121. 
Coffman, 387, 388. 
Cohn, 125, 127. 
Color-blindness, 124, 126. 
Colvin, S. S., 168. 
Commenius, 224. 
Conrad, H. E., 407. 
Cook, VV. A., 333, 334. 
Coover, J. E., 196, 198, 200, 206, 208, 

209. 
Cornell, W. S., 129. 
Cornman, O. I'., 337, 338, 339, 340, 

.395- 

Correlation: problem stated, 49; co- 
eiTicient of, 49f.; among specific 
mental abilities, jof.; among abil- 
ities in school subjects, 54f.; conclu- 
sion concerning, 54, 57f.; between 
special capacities and general intel- 
ligence, 59; between mental and 
physical traits, 59; between early 
and later mental abilities, 6of. 

Coubal, L. J., 89, 164. 

Counts, G. S., 382. 

Courtis, S. A., 274, 275, 286, 382, 
406. 

Courtis tests, 20, 35, 37, 89, 165, 166, 
381, 400, 408, 410, 412. 

Craig, Helen, 43. 

Cross education, 2iof. 

Culture e[)ochs theory, 25. 

Currier, 290. 



Cur\'e of learning, 141; characteristics 
of, i4if.; initial rise in, i43f.; pla- 
teaus, i5of. 

Dallam, M. There- y, 246. 

Davidson, P. E., 24. 

Davis, W. W., 210. 

Dearborn, W. F., 60, 154, 161, 162, 
177, 201, 202, 206, 207, 262, 265, 
266, 267, 268, 269, 428, 430, 444. 

De Bruin, L. C, 328. 

De CandoUe, A., 92. 

De.xter, E. G., 178. 

Dexter, Emily S., 84, 85. 

Diebel, Amelia, 362, 363, 365. 

Dietze, G., 384. 

Distribution curve, 28, 30, and as- 
signment of marks, 443f.; memory 
ability, 26; A-test, 27; cancellation, 
27; association, 28; chest measure- 
ments, 30; height of women, 31; 
head girth of boys, 31; of tos&ings 
of pennies, 32; and marks, 443-448. 

Distribution of mental abilities, 29. 

Distribution of practice, i53f. 

Dodge, R., 266. 

Dougherty, M. L., 109. 

Downey, J. E., 298, 300. 

Drill in writing, 3i9f.; in spelling, 
339f.; in arithmetic, 396f. 

Drushel, J. A., 409. 

Dugdale, R. G., 77. 

Duquid, 290. 

Durr, 124. 

Earle, E. L., 82. 

Ebbinghaus, H., 69, 153, 157, 171. 

Ebbinghaus test, no. 

Ebert, E., and Meumann, E., 161, 
201, 202. 

Edison, T. A., 377. 

Education: Definition of, 1; funda- 
mental problems of, 2. 

Elliott, E. C, 45, 433, 434. 

Ellis, 92. 

English: methods of measuring effi- 
ciency in, 352f.; results of measure- 
ments in, 353f.; economic methods 
of acquiring skill in, 354f.; acquisi- 
tion of ideas, 355; of words and 



INDEX 



467 



forms, 355; grammar and correct 
English, 3s6f.; imitation in acquir- 
ing, 357; specific attention to errors 
in grammar, 36if.; oral versus writ- 
ten practice, 367f.; good English in 
all classes, 37 if.; types of topics 
which should be given, 372; effects 
of differing teaching ability, 373; 
and Latin, see Latin. 

Environment, see Environmental In- 
fluence. 

Environmental influence on different 
original abihties, 88; experiments 
concerning, 88f.; conclusions con- 
cerning, 91. 

Erdmann, B., 266. 

Extra-work plan, 43. 

Family resemblances, see Mental 

Heredity. 
Faraday, 377. 
Far-sightedness, 123. 
Fatigue, methods of studying, 170; 

experiments on, 17 if.; and school 

work, 175; see also Progress of 

Learning. 
Feeblemindedness, inheritance of, see 

Mental Heredity. 
Fillers, H. D., 362, 363. 
Finkelstein, I. E., 428, 440, 443. 
Fordyce, 274, 275. 
Forgetting, i56f. 
Formal discipline, 217, 252. 
Foster, F. M., 235, 236, 325. 
Foster, J. W., 218. 
Foster, W. T., 177, 428, 429, 441, 442, 

444. 
Fracker, G. C, 161, 194, 195, 202, 

203, 212. 
Franklin, 360. 
Freeman, F. N., 282, 299, 306, 308, 

309,310,312,313,317,318,321,384. 
Friedrich, J., 171, 172. 
Fulton, M. J., 342. 

Gallon, Sir Francis, 74, 85, 86, 105. 

General intelligence and special ca- 
pacities, 109; results of Simpson's 
experiment on, 1 10; Burt's study of, 
iiof.; conclusion concerning, nif. 



Gessell, A. L., 300. 

Gilbert, J. A., 67, 71; and Fracker, 

G. €., 194, 195. 
Gilman, Charles, 122, 123, 124. 
Gist, A. S., 412. 
Goddard, H. H., 77, 78, 79, 100, 107, 

130. 
Gormley, 241. 
Graves, S. M., 306, 307. 
Gray, C. T., 321, 431, 432. 
Gray, W. S., 265, 268, 274, 275, 294. 
Griesbach, 170. 
Gross, 299. 
Gulick, L. H., and Ayres, L. P., 127, 

128, 130. 
Guyer, M. F., 77, 79. 

Haggerty, M. E., 275, 353, 354, 372, 
400. 

Hahn, W. H., 404, 405. 

Haines, T. H., 107. 

Hall, G. Stanley, 17, 19, 23, 135, 265. 

Hall-Quest, A. L., 189, 190. 

Hamilton, F. M., 270. 

Handwriting: steps involved in, 297; 
sex differences in, 299f., 303f.; cor- 
relation of with other traits, 300; 
measurement of efiiciency in, 3oif.; 
economic procedure in learning to 
write, 304f.; perception of forms in, 
304f.; length of practice periods, 
3o7f.; standard of proficiency in, 
3iof.; relation of speed to quality 
in, 3iif.; methodsof teaching, 3 i4f.; 
factors affecting execution of move- 
ments in, 3i7f.; movement drills, 
319; correct form in, 320; analysis 
of imperfections in, 32of. 

Handwriting scales, Thorndike, 301, 

303, 310- 

Ayres, 301, 303, 310, 312, 313, 316. 

Starch, 301, 303. 

Freeman, 302. 

Palmer, 302, 321, 

Zaner, 302, 321. 

Gray, 321. 
Harris, J. H., 290. 
Harris, L. H., 234, 235. 
Heck, W. H., 172, 173. 
Height and age, Fig. i, 18. 



468 



INDEX 



Hendricks, 294. 

Hcnmon, V. A. C, 70, 71, 84, 164, 
168, 169, 281. 

Ilentschcl, But/., SoMewsky and 
Kaselitz, 384. 

Ikrhiirt, 139. 

Ikrtdity, see Mental Heredity. 

Hewins, X. P., 247, 249. 

Hicks, VV. E., 391. 

Hilkgas scale, 35, 39, 40, 41, 245. 

]Iistor>': psychological steps invoked 
in, 4i6f.; measurement of attain- 
ment in, 4i8f.; economic methods 
in learning, 42of.; suggestions for 
method of study, 42 2f.; essential 
material in, 423f. 

History tests: 
Starch, 418. 
Bell and McCoUum, 420. 

Hoge, Mildred and Stocking, Ruth J., 
14. 

II()lling^vorth, H. L., 52. 

Ilolloway, H. V., 413, 414. 

Horace Mann system of reading, 290, 
291, 292, 293. 

Horn, E., 425. 

Hosic, J. E., 369. 

Houser, J. I)., 330. 

Howe system in reading, 294. 

Howell, H. B., 378, 384, 385, 386, 
410, 41 1. 

How to study, 176; waste in studying, 
176; value of study, i77f.; types of 
studying, 170; problems in study- 
ing; see als<j Studying. 

Hoyt, E. S., 226, 227. 

Huey, E. B., 266, 270, 281. 

Hull, C. L., 148, 149, 263, 264, 301. 

Ihmiphrey, 240. 

Hypennctroijia, 123. 

Imagcr>', i66f. 

Iiuli\idual difTerences, 26; quantita- 
tive nature of, 2(>; means of graphic 
representation, 28; range of, 28f.; 
in reading, 33; in writing, 34; in 
spelling, 34; in arithmetical reason- 
ing. 35; '" addition, 37; in English, 
38, 40, 41, 42; in geography, 39; 
in drawing, 40. 



Individual instruction plan, 42, 44. 

Infallibility of instincts, i^i. 

Ingersoll, L. R., 351. 

Instinct, and motivation, 12; collect 
ing instinct and age, Eig. 6, 22. 

Instincts, defined, 9; and refle.xes, 9 
and capacities, 10; classified, lof. 
overemi)hasis of in education, iif. 
role in education, iif; sudden de- 
velopment of, i9f.; unrevivability 
of, 2 if. 

Instincts, theories based ujxjn dy- 
namic theory, 14, isf.; transitori- 
ness, 14, 17; recapitulation, 14, 

22f. 

Intelligence quotient, 104; how de- 
termined, io5f. 

Interest in relation to learning, see 
Progress of Learning. 

Interests, permanency of, 62. 

Jacoby, P., 92. 

James, VV., 21, 23, 138, 139, 193, 211, 

212, 213, 214, 371. 
Jastrow, J., 379, 381. 
Javal, E., 266. 
Jessup, W. .\., 387, 388. 
Jevons, 377. 
Johnson, R. I., 366. 
Jc>nathan Edwards family, 77, So. 
Jones, W. E., 346. 
Jost, A., 153. 
Judd, C. H., 199, 213, 214, 215, .V14, 

28s, 286, 306, 3T3, 314, 315, 316, 

.S96, 397- 
Jukes family, 77, 80. 

Kallikak family, 77f.; diagram of de- 
scendants, 78. 
Kansas silent reading test, 275. 
Kelly, E. J., 274, 275. 
Kelley. T. L., 61, 62. 
Kent, (1. H., 174, 175. 
Kerr, Marj' A., 403. 
King, Miss, 172. 
King. Irving, 235, 285, 286. 
Kirby, T. J.. 155, 402, 403, 404, 405. 
Kirclur, H. \V., 2S0. 
Kirki)atriik, E. .\., 406. 
Kline, L. \V., lOi, 197. 



INDEX 



469 



Knight, 346. 
Knilling, 384. 
Kraepelin, E., 299. 
Kuhlmann, F., 100, 108. 
Language, psychological processes in- 
volved in, 349f. 

Landolt, 266. 

Language scales: 
Starch, 352, 353. 
Hilligas-Thorndike, 353, 373. 
Harvard-Newton, 353. 
Trabue, 353. 

Laser, H., 171. 

Lathrop, G. C, 378. 

Latin: general value of, 229; effect of 
study of modern language, 23 if.; 
and scholarship, 232; effect on Eng- 
hsh, 233f., 24if., 246f., 356; oh rhet- 
oric, 23s; and college honors, 236f.; 
and original capacity, 238f. 

Lay, W. A., 384, 385, 386. 

Learning: problems in rate and prog- 
ress of, 141; curve of, 141; early 
progress in, 143; various curves of, 
142-149; analytic types of, 148; 
distribution of practice in, iS3f.; 
plateaus in, 142, 149, isof.; see also 
Progress in Learning and Curve of 
Learning. 

Learning process: analyzed, nsf.; 
common and special elements in, 
118; general versus special, 119; 
problems concerning, ii9f. 

Length of practice periods, see Prog- 
ress of Learning. 

Liddle, Carrie W., 200. 

Lincoln, 30. 

Lindley, E. H., 377. 

Llewelyn, 289. 

Lodge, G., 229, 230. 

Loisette, A., 422. 

Lounsbury, 369. 

Loveland, 190. 

Lueba, J. H., and Hyde, W., 153. 

Lyon, D. O., 156. 



Marks: importance of, 426; variations 
in distribution of, 426f.; variations 
in evaluation of same school subject, 
433f.; causes of variations, 43sf.; 
evaluation of factors involved, 438.; 
size of units on marking scale, 438f.; 
how distribute marks, 44of.; objec- 
tions to use of distribution curve 
in assigning marks, 443f.; methods 
of reducing variations in grades, 
448f. 

Matthews, Brander, 361. 

JMeasurement of mental capacities, 97; 
value of, 97f.; methods of, 99, 109. 

Mead, C. D., 287, 288, 289, 408. 

Memorizing ability, regular increase 
during school life, igf. 

Memory, see Correlation, Transfer of 
Training, Learning. 

Mental heredity; problem of, 73; 
methods of studying, 73; views of, 
74; Galton's study of, 74!; Wood's 
study of in royalty, 75; in various 
low grade families, 77f.; in the 
Jonathan Edwards family, 77; its 
effect on degeneracy and crime, 79; 
on feeblemindedness, 78, 79f.; gen- 
eral interpretation, 95f. 

Messenger, J. F., 384. 

Methods of teaching, their relation to 
psychology, 3. 

Meumann, E., 2oif., 299. 

Meyer, M., 426, 427, 444. 

Mill, John Stuart, 97. 

Minnick, J. H., 190. 

Monitorial group plan, 42, 43. 

Monroe, Paul, 230. 

Monroe, W. S., 275, 391. 

Mosso, A., 171. 

Motivation in learning, 165. 

Mueller, A. D., 273. 

Munn, A. F., 154. 

Musical discrimination, 130. 

Myers, E. J., 168. 

Myopia, 123. 

Myth, Creation of Woman, 63f. 



McAllister, 264. 

McGuire, Margaret F., 122. 

Magneff, 157. 



Nanu, H. A., 384. 

Nature's infallibility, see Infallibility 
of Instincts. 



47° 



INDEX 



Ncar-sit;htedness, i3,^ 
Neff, 218. 
Newton, 95. 

Oberholtzcr, K. E., 287. 

Obsenution, accuracy of, I32f.; in 
children, 133; reasons for inaccu- 
racy in, 134; range of, i34f.; how im- 
prove, 136; effect of practice on, 

137- 

Odin, Q2. 

Original abilities, affected by diflering 
environments, Q2f.; C'attell's inves- 
tigation of scientific men, q2; French 
men of letters, 93; efliciency in 
school subjects, 93; birth places of 
eminent men, 94. 

O'Shea, Harriet, 281. 

O'Shea, M. V., 327. 

Overlapping, 36f.; extent of, 38; im- 
{wrtance of, 39; remedies for, 41 f.; 
percentages of as basis for com- 
parison, 66, 69. 

Parker, F. W., 55, 59. 

Partridge, E. A., 233, 234. 

Paterson, D. G., 107, 130. 

Pearson, K., 61, 80, 82, 345, 346, 347, 
378. 

Perkins, A. S., 240, 242, 243. 

Peters, C. C, 282. 

Permanency of interests, 62. 

Pestalozzi, 384. 

Phelps, C. L., 412, 413, 414. 

Phelps, William Lyon, 369, 370. 

Phillips, F. M., 379, 381, 400. 

Physical defects and school work, 
i3of. 

I'intncr, R., 107, 130, 289. 

Plateaus, sec learning. 

Plato, 250. 

Poellman, 77. 

Point scale tests, 107. 

Probability ciir\e, 33. 

Probability, integral, 32. 

PrtKtor, M., 188. 

Progress of learning, factors in: length 
an»i distribution of work peritnls, 
'5jf| 15^; forgetting, i.s6f., and 
school work, 158; concentration, 



isSf.; specific versus general prac- 
tice, i59f.; relation to sthool work, 
i62f.; practice with knowledge, 1O3; 
interest in improvement, i04f.; 
imagery in, i66f.; conclusion on 
imagery, 169; fatigue, i69f., 75. 

Promotion, by groups, 44, 45; by sub- 
jects, 46; further suggestions for, 
46f. 

Pr>or, H. C, 343, 347. 

Psycholog>' and teaching, 3, 258f. 

Psycholog>' of learning, and methods 
of teaching, 3; need for more ex- 
tended and exact studies in, 3f.; 
analysis of [)roblems of, i isf.; state- 
ment of problems of, i i9f.; in school 
subjects, 257; problems of in school 
subjects, 260. 

Pueblo Plan, 42. 

Puffer, 24. 

Pyle, W. H., 154. 

Kadossawljewitch, P. R., 157. 

Rate of ta[)i>ing, Fig. 2, 19. 

Rational methcKl in reading, 294. 

Reading sailes, Gray, 274, 275, 
291. 

Kelly, 274, 275. 
Starch, 274, 291, 292. 
Courtis, 274, 27s, 280, 
Brown, 274, 275. 
Fordyce, 274, 275, 280. 
Kansas silent reading, 275, 280. 

Reading, steps involved in, 261; re- 
ception of stimuli, 262f.; size of field 
of distinct vision, 264; range of at- 
tention, 265f.; eye movements, 266f.; 
transmission of ner\c impulses to 
visual center, 269; arousal of asso- 
ciation processes, 269f.; transmis- 
sion of ner\e impulse from visual 
center, 27if.; me;isuremenl of efli- 
ciency in reading, 2 74f.; results of 
measurements, 275f.; economic pro- 
cedure in learning to read, 278f.; 
improvement in reading ability, 
2Sif., 204f.; relatitmof spcxd tocom- 
|)rehension, iS^f.; relation between 
oral and silent reading, 2S7f.; phon- 
ics, 390; comimrison of teaching 



INDEX 



471 



methods in reading, 29of.; Beacon 
system, 290-295; Aldine system, 
290-294; Horace Mann system, 290, 
293; Ward method, 294; Howe 
system, 294; Rational method, 
294. 

Reception of stimuh, problems, 132. 

Rejall, 157, 161. 

Report, unreUabihty of, i32f.; sources 
of error in, 134. 

Rice, J. M., 92, 93, 94, 334, 335, 
337, 338, 340, 391, 392, 393, 394, 

395- 

Rickhard, 190. 

Riley, J. L., 348. 

Ritter, C, 171. 

Rogers, A. C, 79. 

Royal families, heredity in, see Men- 
tal Heredity. 

Ruediger, 224, 264, 271. 

Ruger, H. A., 148, 209. 

Rugg, H. O., so, 250, 251, 424 

Sackett, L. W., 311, 325. 

Schmidt, W. A., 268, 269. 

Schocklow, 199. 

Schuster and Elderton, 84. 

Schuyler, W., 158. 

Scripture, E. W., Smith, T. L., and 

Brown, E. M., 210. 
Search, P. W., 42. 
Sears, Isabel, 408. 
Sears, J. B., 325, 362, 365. 
Seashore, C. E., 130, 174, 175. 
Sensory defects: their effects in 

general, 120; on school work, 

I29f. 

Sex differences: educational signifi- 
cance of, 63; popular vs. scientific 
view of, 63f.; quantitative nature of, 
64; methods of comparing, 65, 69, 
summary of, 68f.; in range of abil- 
ities, 70, 72. 

Sikorski, J., 171. 

Similarities between brothers and sis- 
ters, 80; in special mental traits, 
8if.; in abilities in school subjects, 
82f.; in university work, 84. 

Similarities of twins, 85; Galton's in- 
vestigation, 86; Thorndike's experi- 



ments, 87f.; young and old twin? 
compared, 87. 

Simon, T., 99. 

Simpson, B. R., 53, 59, 109. 

Sleight, W. G., 161, 203, 204, 205, 
206. 

Sloane, WilUam, 218. 

Smedley, P., 18, 20, 129. 

Smith, A. G., 55. 

Smith, F. O., 61. 

Spearman, 367, 412. 

Specific topics for educational psy- 
chology, brief outline, 4f. 

Spelling, lists of common words in: 
Eldridge, 327, 328, 329, 356. 
Ayres, 327, 328, 356. 
Jones, 327, 328, 331, 356. 
Cook and O'Shea, 327, 328, 329, 

356. 
Starch, 327, 328, 329, 356. 
Boston minimum, 331, 3$$. 
Stockton, Santa Cruz, Cliicago 

Speller, 331. 
Nicholson, Chico, 333. 

Spelling: steps involved, 322f.; 
methods of measuring efficiency in, 
323f.; results from use of scales, 
325f.; economic methods in learning 
to spell, 326f .; determination of com- 
mon words in, 327f.; proper place- 
ment of words, 33of.; influence of 
rules in, 33f.; length of class period, 
334f.; methods of teaching, 338f.; 
effect of drill in, 339f.; waste in 
teaching of, 340; laws of association 
in, 343 ; centering attention on order 
of letters in, 343; personal incentive 
to effort, 343f.; special attention to 
diflicult parts of words, 344; writing 
the words, 345; context versus 
column spelling, 345; teaching 
homonyms, 345f.; class versus inde- 
pendent study, 346; grouping of 
words, 347; imagery in, 347f.; pres- 
ent day compared with past effi- 
ciency in, 348. 

Spillman, W. J., 94. 

Squire, Carrie R., 224. 

Stanford revision of Binet-Simon 
tests, 69, 70, 100. 



472 



indf:x 



Steele, 264. 

Stevens, W. J., 45. 

Stevenson, J. A., 51, 52. 

Stevenson, K. L., jfK). 

Stimuli, reception of, 132; inteipreta- 
tion of, 138. 

St. Ix)uis |)l;in, 44, 45. 

Stone, C. \V., 378, 394, 395. 

Strabismus, i 24. 

SlucJebaker, J. W., 396, 405. 

Stuilky and Ware, 33. 

Sludyinji, waste in, i7(); valueof, i77f.; 
types of, 179; problems in, 179; 
control of attention in, iSo; ditli- 
cultiesin, 180; suf;gestions for, i83f.; 
Whipple's ndes for, i86f.; whole and 
part method in, i85f. See also How 
to Study. 

Supernormal child, need for differen- 
tiated training for, see Overlap- 
ping. 

Su|)erviscd study, 42, 43, 44, i88f. 

Surface of frequency, 28. 

Suzzalo, H., 345, 346, 347. 

Swift, E. J., 144, 150, 152, 158, 161, 
-\SO. 231. 

Taussig, A. E., 125, 126, 127, 128. 

'relegra[)hy, i42f. 

'J'erman, L. M., 69, 100, 105, 107, 
108. 

Thomas, 64. 

Thomdike, E. L., i, 10, 16, 40, 50, 55, 
56, 62, 66, 67, 68, 71, 72, 85, 86, 87, 
89, 148, 150, 155. 157, 161, 165, 172, 
173, 17s, 195. 197, 212, 213, 222, 
275, 300, 301 T 307, 308, 309, 311, 
378, 404, 405, 409, 410. 

Trabuc, .M. R., 353. 

Transference of training: proiilems, 
191; e.xiKjrinuntal teihnique, 192; 
critici.sm of , 2 1 1 f . ; in memory, James 
cx{)criment, i93f.; other experi- 
ments, 201-208; in reaction time, 
i94f.; in perception and discrimina 
tion, i95f.; in sensori-molor associa 
tion, i98f.; in attention, 2o8f.; in in- 
genuity, 20()f.; summary concerning, 
212; mcllKKls by which transfer 
takes place, 2i3f. 



Transfer of training in school sub- 
jects; opinions concerning, 21 7f.; 
specific estimates of values of sch(X)l 
studies, 2ogf.; table giving estimated 
values of studies, 221 ; in arithmetic, 
222; in grammar, 224f.; in foreign 
languages, 229f.; in Latin, 233f.; in 
science, 247f.; in geometr>', 25of.; 
general interpretation concerning, 

25 2f. 

Twins; similarity of, 85; experimental 

work with, 86f. 
Typewriting, 145^ 

\'an Landegend, E., 89. 

Varying abilities in school subjects; 
range, 3$, 36; general causes, i^. 

Visual defects, 120, 121; tviMJs of, 123; 
causes of, i24f.; frequency of, I25f.; 
relative frequency in colored and 
white children, 125; reasons for 
variations in reports on, 126; in- 
crease in higher grades, 127; 
methods of avoiding, 127!. 

Waddell, J. v., 433. 

Wagner, 347. 

Waldo, K. I)., 282, 283, 294. 

Wallin, J. ]•;. W., 325, 341, 342, 
345- 

Walsemann, H. J., 384. 

Ward method in reading, 294. 

Warren, H. C, 384. 

Washington, 35S. 

Waste in education due to lack of 
exact data on various learning proc- 
esses, 3f. 

Webb, L. W., 199, 200. 

Wells, K. L., 160, 161. 

Whi|)i)le, t). M., 50, 68, 124, 125, 
127, 128, 129, 133, 136, i86, 265, 
286. 

Wl'itliy, M. T., 89, 156, 161. 

Whole and part methods in learning, 
see Studying. 

Wilbur, Eliira, 405, 406. 

Wilcox, M. J., 238, 239, 340. 

Wiley. Harvey. 218. 

Wilmarth, .\lfre<l. 79. 

Wilson, G. M., 389, 390. 



INDEX 



473 



Wimmer, Herman, 400, 401, 402, Woolley, Helen Thompson, 66, 67. 



404. 
Winch, W. H., 161, 207. 
Witmer, 122. 
Woods, F. A., 75, 76. 
Woodworth, R. S., 195, 197, 210, 

212. 
Woody, C, 382. 



Word lists, see Spelling. 
Writing, see Handwriting. 

Yerkes, R. M., 100; and Bridges, 107. 

Zero family, 77. 



Printed in the United States of America 



^ I "'HE following pages contain advertisements of a 
few of the Macmillan books on kindred subjects 



i 



BY THE SAME AUTHOR 

Educational Measurements 

Diagrams, 8vo, 202 pages, $1.25 

"Educational Measurements" is the first book in which are 
assembled together all standard scales and tests available to 
date for measuring ability in school studies. The author has 
made a general survey of the subject of educational measure- 
ments, and has incorporated in this book the best of standard 
scales and texts, with full instructions as to how to use them. 
Not only elementary subjects, but also high school subjects, 
receive attention. The careful reading of this book is equiva- 
lent to a year's graduate study on the subject in a university. 



"This volume is one of the most comprehensive of its kind. 
Himself the foremost investigator and lecturer in this field, 
Starch has collected a considerable amount of experimental 
evidence and has included with it in this text a large number 
of Standard Tests in at least ten school subjects." — School 
and Home Education. 

"The author is to be commended mostly for organizing for 
the casual reader a possible course in educational measurements." 
— The School Review. 

"It is obvious that Mr. Starch's book will meet a great need, 
presenting as it does such valuable material in compact form." 
■ — Educational Administration and Supervision. 



THE MACMILLAN COMPANY 

Publishers 64-66 Fifth Avenue New Tork 



py THE SAME AUTHOR 

£Lxperiments in Educational Psychology 

///., 12o, $I.OO 

This l)Ook is designed to be a laboraton- guide for exiK-ri- 
mental work running parallel with a course in Educational 
Psychology, or other courses in Education in which the psycho- 
logical aspects are emphasized, or in Elementar}- Psychology 
in which the educational aspects are emphasized. The material 
for ex])eriments is supplied in the text or can be obtained at 
ver)- little cost. 

"This book is a pioneer in the new field of educational psychol- 
logy — and as such is a valuable contribution for which every 
worker in the field will be grateful. It will doubtless be widely 
used as a text where courses in educational jisychology are 
given, and will be freely borrowed from when not used as a 
text." — The University of Chicago Press. 

'". . . Boosts the whole scheme of Ex-perimentation in Edu- 
cational Psychology by bringing together in an understandable 
and usable-by-anybody way all the latest jjhases of testing 
Individual Differences, Visual Tests, Auditory Tests, Mental 
Images, Progress in Learning, Transference, Association, Ap- 
perception, Attention, Memory, Work and Fatigue. 

"Almost any teacher can use almost anything in these mul- 
titudinous experiments. 

"Dr. Starch eliminates the vague and visionary, and magni- 
fies the clear and direct in presenting the philosophy and prac- 
tice of the schools." — Boston Journal of lulucation. 



THE MACMILLAN COMPANY 

Publishers 64-66 Fifth Avenue New York 



The Learning Process 



By S. S. COLVIN 

Professor of Educational Psychology in Brown University 

Cloth, i2mo, jj6 pages, $1.25 

The general nature of the learning process is outlined ; first, chiefly 
from the biological standpoint, considering the nature of the learning 
process throughout the animal world, and discussing the basis of 
learning in instinct, and the learning process in the formation of 
habits. The learning process is then analyzed in terms of its con- 
scious factors. 

"It may safely be said that any teacher who becomes familiar with 
the contents of this book will have gained in an agreeable form the 
best that has been accomplished up to date in the experimental study 
of the learning process. ... It ought to prove of distinct service 
in soUdifying American educational theory." — The Dial. 

"One of the best books on those phases of psychology which apply 
to education." — Journal of Philosophy, Psychology and Scientific 
Methods. 

"A decided step in the right direction — away from generality and 
technicality, and towards concrete facts and their specific application." 
— American Journal of Psychology. 

"Should prove stimulating to thoughtful students of educational 
problems." — The School Century. 

"The experienced teacher can well afford to go through this book 
carefully for the purpose of measuring his own habits of work against 
the standard obtained by the scientific study of the instrument upon 
which he seeks to perform — the mind of the child. Those who have 
not kept in touch with recent advances in psychological research will 
find the chapters on the association methods of Jung and Freud, and 
on the transfer of training especially valuable." — American Teacher. 



THE MACMILLAN COMPANY 

Publishers 64-66 Fifth Avenue New York 



f. 



Genetic Psychology 



Jiv KDWIN A. KTRKPATRICK 

Director of the Child Study Department of the Titchburg (Mass.) 
State Normal School 

Cloth, i2mo, $t.zy 

"(lenelic PsychoIoRV," by Professor Kdwin A. Kirkpatrick, will stand 
almost alone in its buildinp up of a system of psychological study by strictly 
biological methods. In his working out of the genesis of intelligence, while 
concerning himself largel.v with the dawning of intelligence among the lower 
forms of animal life, he has gi\en a quantity of facts, carefully groiocd s<j 
as to show not merely the earlier phenomena of mind but its behavior under 
an increasing complexity of structure and environment. Psychology thus 
studied must necessarily be intenvovcn with the facts of anatomy and 
physiolog>', and in the ]>rocess the author has used a wealth of illustration, 
and has marshalled the vast body of information gathered in twenty years 
of investigation and study so as to produce a fascinating work which will 
not fail to stimulate individual study. 

"In origin and plan of treatment pedagogical interest has played a large 
part. After nearly a score of years devoted to the study of child psychology, 
the impulse to formulate the broader truths of genetic i)sychology as a 
distinct subject came from the experience of giving courses in both subjects 
to summer students at Columbia and Chicago Universities. The interest and 
comprehension shown by those students develoi)cd the belief that the facts 
of genetic psychology, incomplete as they arc, could be profitably formu- 
lated for the use of educators. Subsequent experience in giving parts of 
this book in nearly their [jresent form, to a class in a normal school, has 
confirmed this belief. It is probable also that popular interest b great 
enough to make the bt>ok accc|)table to the more serious of those interested 
in animal beha\iour." — From the Preface. 

CONTENTS 
Preface — Literature. 
chapt1';r 

I. Intrwluction. 
II. Structural Basis of Behaviour. 

III. Tyjx-'s of Animal Behaviour. 

IV. Complex Ik'haviour Characteristic of S[)ccie.'^ — Instincts. 

V. Behaviour of Individuals- — .\cquisition of Habits and Ideas. 
\T. Structures Concerned in Complex Behaviour and in Ideation. 
VTl. Consciousness. 
VI II. Specific Conscious Slates. 
IX. Types of .Adaptive .\ctivity or Intelligence. 
X. Tyi)es of Learning .Activity. 
XL Racial and Individual Development. 
Index. 

THE MACMILLAN COMPANY 

Publishers 64-66 Fifth Avenue New York 



Y\ 



i 




University of Toronto 
library 






I 



a 

u. 

CO 



o 

J 



« 
o 

-P 

O 

SI 



DO NOT 

REMOVE 

THE 

CARD 

FROM 

THIS 

POCKET 







Acme Library Card Pocket 

Under Pat "Ref . Index File" 

Made by LIBRARY BUREAU