THE USE OF ORAL PROMPTS AS AN EFFECTIVE TEACHING STRATEGY IN ORAL READING ACTIVITIES BY PATRICIA BUTCKA SEELY A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IIVERSITY OF FLORIDA 1981 Copyright 1981 by Patricia Butcka Seely ACKNOWLEDGEMENTS I wish to express my sincere appreciation to all members of my committee. Dr. William Powell, chairman of the advisory committee, was helpful in many ways. His support and encouragement through the various stages of my doctoral work and dissertation were invaluable. He epitomizes the true professional educator in eyery way. My special thanks to Dr. James Algina for his willingness to help me through various trying moments. I appreciated his support and words of encouragement. Dr. Lawrence Smith was a helpful and supportive member of my committee. I appreciated his help and advice. My thanks go to Dr. Timothy Blair and Dr. Doyle Casteel who offered their assistance and advice during the various stages of my dissertation. Also appreciated was the help of Dianne Downing who helped with the typing of dissertation drafts. I owe a special note of gratitude to Linda Davidson, my best friend, who was always supportive with a kind work and thoughtful deed. I owe a great deal to my stepfather, Bill Lastinger, and to my sister and brother, Debbie McCullough and Walter Butcka, who helped me in their own special ways. m To my mother, Anne Lastinger, I owe an exceptional debt of thanks for providing an atmosphere filled with love and encouragement to become the best I could be. Her continuous love and support have filled my life in a way that only she and I will understand. IV TABLE OF CONTENTS PAGES ACKNOWLEDGEMENTS iii ABSTRACT vii CHAPTER I INTRODUCTION 1 Statement of the Problem 2 Objectives and Research Questions 3 Hypotheses 4 Basic Assumptions 5 Definitions of Terms 6 Limitations of the Study 9 CHAPTER II REVIEW OF RELATED LITERATURE 10 Rationale for Oral Reading Instruction 10 Feedback 12 Teacher Conceptualizations 14 Teacher-Student Interaction 17 Use of Prompts 21 Summary 24 CHAPTER III PROCEDURES AND METHODOLOGY 26 Pilot Studies 26 Research Study 29 Sample 30 Instruments 31 Procedure 34 Method of Analysis 37 CHAPTER IV RESULTS OF THE STUDY 39 Results 39 Summary and Discussions 63 CHAPTER V SUMMARY AND CONCLUSIONS 67 PAGES APPENDIX A SAN DIEGO QUICK ASSESSMENT WORD LISTS .... 73 APPENDIX B INFORMAL READING INVENTORY 75 APPENDIX C POWELL CRITERIA 82 APPENDIX D CODING SHEET FOR RECORDING TEACHERS' ADOPTABILITY OF ASSIGNED PROMPTS 87 BIBLIOGRAPHY 88 BIOGRAPHICAL SKETCH 93 VI Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE USE OF ORAL PROMPTS AS AN EFFECTIVE TEACHING STRATEGY IN ORAL READING ACTIVITIES By Patricia Butcka Seely August 1981 Chairman: William R. Powell Major Department: Curriculum and Instruction A primary purpose of this study was to examine the extent to which teachers adopted a method of teacher prompts when working with average fifth-grade readers in oral reading acitvities. A second purpose of this study was to examine the use of teacher prompts as an effective teaching strategy with average readers in fifth grade in oral reading activities. Three main objectives were included in the investigation. 1. Teacher ability to adopt a prompting condition when correcting student oral reading errors. 2. Student achievement gains when teacher prompts were used in oral reading activities. 3. Effectiveness of four designated prompting conditions among one another. The six schools selected for the three-and-one-half month study were located in Volusia County, Florida, and represented the three different socioeconomic groups. Twelve teacher subjects were randomly selected. Three teachers were randomly assigned to one of four prompting conditions: 1) un- corrected, 2) graphophonic, 3) semantic, or 4) repeat. They were in- structed to use the assigned prompts as often as possible when correcting the average fifth-grade readers' oral reading errors during the daily 30-minute instructional period. Each teacher was observed 14 times during the study. The 72 average fifth-grade readers were selected from a larger population of fifth-graders. In the final analysis, 67 students were considered. The results of the study suggested that teachers could adopt as- signed prompts when working with average fifth-grade readers. The uncorrected and semantic prompts recorded significant adoptability. The repeat prompt showed no difference in adoptability in relation to the other prompts. The graphophonic prompt recorded no significance in adoptability. The results of the student achievement gains indicated that the uncorrected prompt produced significant student gains whereas the graphophonic, semantic, and repeat prompts produced no significant dif- ferences during the three-and-one-half month study. vi n CHAPTER I INTRODUCTION During an oral reading task the reader is likely to give an in- correct response. The teacher must then decide how to react to the error and whether to provide feedback to the reader. The first decision is concerned with whether to signify to the reader that an error had been made or whether to withhold reaction. Secondly, the teacher must decide at what point she will provide feedback to the student that a specific error had been make. The final decision for the teacher is concerned with the type of prompt she will use to provide feedback to the oral reader. Many educators believe that immediate correction of student errors is most facilitative to oral reading instruction while others support different methods including delayed feedback, cuing, and repeating. Although a variety of instructional techniques are available, reading programs of the past and recent years have favored the immediate feedback approach whereby the teacher corrects a student's errors at the time they occur. Teachers adopting this approach believe that failure to correct errors in oral reading promotes learning of incorrect responses (Niles, Graham, and Winstead, 1977). However, Pehrsson (1974) found that teacher interruptions of student oral reading caused recall to decline. Similarly, Buschke (1974) found that recall with young oral readers would also decline through teacher interruptions during the oral reading stage unless the students were probed at the conclusion of the task. Gattegno (Mitchell, 1979) suggested that not only must teachers be concerned with the errors made by students during oral reading activities but with effective methods of dealing with them. Inter- rupting a student's oral reading performance should not be the only strategy used by a teacher. Goodman (1970) stated that teachers must go beyond the errors. Therefore, educators cannot merely correct errors but must examine effective strategies to use with students. Brophy and Good (1978) argued that teacher education programs do not provide teachers with the necessary skills to work effectively in this domain. They believed that teachers are not trained to adopt different strategies for classroom use nor given practice in refining such strategies. This suggests that teacher responses to errors may be guided to provide valuable and meaningful instruction. This study was concerned with the examination of the different methods of providing feedback to student oral reading errors. The methods under investigation included the following: 1) uncorrected, 2) graphophonic, 3) semantic, and 4) repeat. Specifically, each method was examined with average readers in the fifth grade who were taught by teachers who employed one of the methods during oral reading instruction. This study served as a facilitative examination of the use of prompts as an effective teaching strategy and an impetus for future investigations and research. Statement of the Problem The primary purpose of this study was to examine the extent to which teachers adopted a method of teacher prompts when working with average fifth-grade readers in oral reading activities using basal readers as the primary source of instructional reading material. Since four different types of prompts were studied, each type was observed to gauge its adoptability as a teaching strategy. A second purpose of this study was to examine the use of teacher prompts as an effective teaching strategy with average readers in fifth grade in oral reading activities. Student achievement gains were studied after students had been exposed to one of four designated teacher prompts for a three-and-one-half month period. The four identified prompts were 1) uncorrected, 2) graphophonic, 3) semantic, and 4) repeat. Objectives and Research Questions Listed below were the objectives of this study. Immediately following each objective were the specific questions that were studied. Objective I To investigate the use of teacher prompts as an effective teaching strategy with average readers in the fifth grade. Question IA Are teacher prompts an effective teaching strategy with average readers in the fifth grade? Objective II To investigate teacher adoptability of particular prompting procedures. Question HA Will there be differences in teacher adoptability among the different prompting procedures? Question I IB Will one prompting procedure produce more significant adopt- ability than the others? Objective III To investigate student achievement gains between prompting procedures. Question II IA Will there be differences in student achievement gains between the different prompting procedures? Question IIIB Will one prompting procedure produce more significant gains than the others? Hypotheses The research hypotheses were stated in the null form and are listed below. An explanation for the statistical analysis for each hypothesis will be presented later in the study. Hypothesis IA There will be no significant adoptability by teachers among the different prompting conditions. Hypothesis IB There will be no significant adoptability of a particular prompting condition over the others. Hypothesis HA There will be no significant achievement gains by students between the different prompting conditions. Hypothesis I IB There will be no significant achievement gains by students of a particular prompting condition. Basic Assumptions It was assumed in this study that: 1. All fifth-grade students involved with this study received the same basic oral reading skills training as set forth by county and state regulations. 2. All fifth-grade students involved with this study received equal amounts of teacher time, instruction, and materials although the material goods varied. 3. All oral reading instruction was directed by the participating teacher rather than from outside help (aides, interns, volunteers, etc.). 4. Since the subjects were randomly grouped, home background, prior school background, and prior instruction in oral reading were equal in all classes. 5. Teacher ability to adopt a new oral reading teaching strategy was equal. 6. Outside variables were evenly distributed among all classes. Definition of Terms For the purpose of this study certain terms will be defined according to their usage in this study. Basal Readers Basal readers are referred to as foundation readers. They may be of two types: 1) method readers where the material is primarily selected with reference to phonetic difficulties so that the reader will more readily acquire the independent power to pronounce words from the printed page, and 2) non-method readers where phonetic and diacritical factors are quite subordinated to the thought of the material presented. Coding Coding is a process by which a researcher collects units of teacher behavior at specified intervals. Graphophonic A graphophonic prompt is a type of teacher prompt whereby a teacher is concerned with the visual characteristics and sound charac- teristics of a word and encourages the student to consider one or both characteristics. Informal Reading Inventory The informal reading inventory is an individual test to determine oral and silent reading achievement. It consists of graded word lists, graded reading passages, and comprehension questions for each passage. An informal reading inventory is used to determine an individual's independent, instructional, and frustration reading levels. Instructional Level An instructional level is the highest level at which instruction in reading may be inaugurated. Materials for this level should be difficult enough to be challenging but not so difficult as to be frustrating. Powell Criteria The Powell criteria are the differentiated criteria set by William Powell (1969, 1978) for determining reading levels. Prompt A prompt is an external source of information which helps the student to arrive at a correct response to a problem. A prompt may be given to reinforce the reader's correct response or may serve as a cue to the reader that an incorrect response was given. The study included four types of prompts: 1) uncorrected, 2) graphophonic, 3) semantic, and 4) repeat. Random Sampling Random sampling permits each member of the population to have the same chance for selection in the sample. Repeat A repeat prompt is a type of teacher prompt whereby the teacher provides feedback to students' oral reading performance after the reading passage or complex thought was completed. The prompt is used to cue the reader that the oral reading was different from the printed text. The repeat prompt does not cue the reader to specific errors but to discrepancies between the oral reading and printed text. Semantic A semantic prompt is a type of teacher prompt whereby the teacher provides feedback to students' oral reading performance after the reading passage or complex thought was completed. The prompt is used to cue the reader that a change in the author's meaning or intent resulted from the error. Although the semantic prompt does not cue the reader to specific errors, the reader is made aware of discrepancies in what he orally read and the printed text. Standardized Tests Standardized tests provide a sample of achievement or ability. They may be norm-referenced or criterion-referenced with individual scores reported in derived scores (grade equivalents, percentiles or stanines). Transfer Transfer is regarded as a shift in responsibility for correcting a student's oral reading errors from teacher to student. Transfer may be regarded as a terminal process which would eventually cause the student to be self-sufficient when correcting oral reading errors. Uncorrected An uncorrected teacher prompt is a type of teacher prompt whereby the teacher provides no feedback to students when an oral reading error occurs. Limitations of the Study The findings and conclusions reached in this study were limited to a random sample of fifth-grade students and teachers in Volusia County, Florida, during the 1980-1981 school year and do not necessarily reflect the use of prompts as an effective teaching strategy for all fifth-grade students and teachers in other settings. Each teacher subject received a one-hour training session con- cerning the use of the assigned prompt. No additional training was given nor was feedback given during the study. This insured that all teacher subjects received the same amount of teacher training throughout the study. The study's limited focus was on fifth-grade students who were receiving oral reading instruction on the fifth-grade level. The fifth grade was selected for the study because there were no additional remedial or enrichment reading programs offered at this level which could have affected the amount of oral reading instruction each student was given. The study was concerned with the students' ability to make significant achievement gains in oral reading activities. Student achievement gains for each type of prompt were also studied. The study did not take into account the different grouping processes among the different classrooms. That is, differences between departmentalized, self-contained, and open-space groupings were not studied. CHAPTER II REVIEW OF LITERATURE Rationale for Oral Reading Instruction Oral reading instruction must be a prime consideration of the total reading program. Harris (1961), Sipay (1969) and Durkin (1970) have deemed oral reading as a necessary component of reading in- struction which must be fostered with great care. It should provide the student with the necessary skills and confidence to function in an everchanging world. Moody (1974) described the oral reading process as ... a necessary social and pedagogical skill, and indeed a valid modification, or extension, of skill in spoken English, which can be neglected only to the detriment of any community in which the language has an important role. (315) Although oral reading serves as an integral means of communication, the goal of oral reading should go beyond this basic need. Efforts should be exerted to guide students from teacher-dependency to self- sufficiency. That is, an ultimate goal of oral reading instruction should make the reader more responsible for his own oral reading in- struction. To acquire the necessary skills for self-sufficiency in oral reading the individual must be taught to automatically confront new ideas, overcome vocabulary, and assimilate new thoughts as a simultaneous process. Teachers must make a conscientious effort to guide students through this developmental process using practices and 10 11 strategies appropriate to the learning task. "Not only must oral reading be regarded as an integral part of the instructional program but . . . teachers should use the most pedagogically sound method of practicing reading which will best develop essential oral reading skills" (Cox and Shrigley, 1980, p. 306). Weinstein (1976) believed that oral reading is the teacher's best method for gauging student growth in reading. The teacher has the opportunity to point out accurate and inaccurate responses to print and to engage the student in meaningful dialogue. Such dialogue between teacher and student provides the essence of good oral reading instruction (Hoffman and Eaker, 1981). The dialogue typically involves the teacher giving feedback or instruction to the student's oral reading performance and allows the student to accept or reject the teacher feedback and move on. Pearson (1976) maintained that a goal of good oral reading in- struction should allow the student to regulate his own oral language and to detect oral reading errors. To do this Pearson suggests a three- step system: 1) student must see the relationship between oral speech and print, 2) student's anxiety level must be minimized to promote fluent reading rather than mere guessing, and 3) student must see that both oral speech and print involve the student's own experiences. If the goal of oral reading instruction is to cause students to become more independent, self-sufficient readers, instruction must be designed to guide students through a developmental process which allows them to accept the responsibility of regulating oral reading. Various researchers have maintained the need for teacher awareness and involve- ment in the guidance process while using appropriate teaching practices 12 to aid students (Cox and Shrigley, 1980; Weinstein, 1976; Pearson, 1976). Oral reading instruction cannot be regarded as a continuous dialogue between teacher and student in and of itself; rather, the dialogue between teacher and student evolves into a monologue by the student with himself in control of the total oral reading process. Feedback According to Bourne (1966), feedback is an external source of information which helps the student to arrive at a correct response to a problem. Feedback may be given to reinforce the reader's correct response or may serve as a cue to the reader that an incorrect response was given. Niles and Tech (1980) believe that teachers give feedback for two main reasons: 1) for the purpose of assisting comprehension, or 2) for the acquisition or recognition purpose. Although it would be difficult to approach the two purposes as separate skills that are segregated from one another, the teacher must decide her purpose for using feedback and adjust her strategies to meet her goals. If the primary concern of the teacher is to build comprehension skills, her feedback strategies to students might be quite different than those strategies used to build word recognition skills. Therefore, a teacher must identify her purpose for the oral reading task and adjust her feedback strategies to meet the purpose. When the student makes an oral reading error the teacher may choose from three strategies. The reader can be informed 1) of both right and wrong answers, 2) of right answers only, or 3) of wrong answers only. At that time the teacher makes a decision as to whom should accept the responsibility for correcting the error. If the 13 teacher decides to supply the correct word or cue to the student for both correct and incorrect responses, she accepts the responsibility for correction of the error. If, on the other hand, the teacher decides to cue the student that a wrong response was given, she is transferring the responsibility for correction to the student. During the transfer of responsibility for correction from teacher to student the reader must adopt a set of behaviors to meet the oral reading feedback situations. The process involves three main steps: 1) The reader recognizes that a deviation has occurred and must decide whether to accept the response or to correct the response; 2) The reader decides whether to correct immediately or to delay correction; and 3) The reader decides what strategies to use in the correction. The process is concerned with both the timing factor of correction (im- mediate or delayed) and the type of correction used. Thus, the student must assume responsibility for two decisions when he accepts self- correction. Niles and Tech (1980) stated that the reader must process feedback at the two separate levels simultaneously. While the reader contends with the major task of deriving meaning from print, he must also judge his own oral responses. A reader who receives feedback from the teacher and who accepts responsibility for correction of the oral reading error must process information on two higher levels of thought than a reader who does not receive some type of feedback (Niles and Tech, 1980). Therefore, a reader who receives feedback and accepts responsibility for oral reading corrections may be confronted with a more complicated reading task which encourages the reader to be more self-sufficient. 14 Teacher use of feedback when instructing students in oral reading activities may be very beneficial if the teacher has identified her purpose for providing feedback and the specific strategies to use to achieve the goals. Feedback may also be advantageous in involving students in the correction process through transfer of responsibility for oral reading corrections. If a teacher's ultimate goal in oral reading instruction is to make students self-sufficient in oral reading, feedback may be regarded as an important and necessary factor for students. Teacher Conceptualizations Although research has been conducted on oral reading errors during the early decades of the century (Madden and Pratt, 1941; McCullough, 1946; Monroe, 1928, 1932), investigations were most often limited to remedial situations where teachers were removed from the normal classroom environment. Additionally, teacher practices and strategies for dealing with student oral reading errors were overlooked in favor of analyzing specific student miscues during the oral reading task. This resulted in little practical advice to teachers for ef- fective methods for dealing with student errors. Various researchers of the 1960s recognized the need for in- vestigations into teacher involvement in the oral reading process (Conrad, 1964; Corder, 1967; Wickelgren, 1965). They contended that the teacher must be regarded as an important factor in oral reading programs when assessing specific student miscues. Although lacking in theoretical consideration, they did progress the idea that a teacher should be able to aid students with text discrepancies. 15 Spiegel and Rogers (1980) stated that teacher feedback may be a viable part of the reading instruction program. The teacher's feed- back may aid the student in word identification in actual reading situations. Also, the way by which a teacher offers feedback could provide valuable insight into the teacher's theoretical perception of reading and her expectations of student oral reading performance. For example, teachers who rely on the interrupted method may be primarily concerned with the student's ability to accurately pronounce the printed words on a page. Teachers who rely most often on delayed feedback may be more concerned with the overall meaning or intent of the author. Investigations in the 1970s were conducted by researchers who focused on the behavioral aspects of the teacher when dealing with student errors. Anderson and Brophy (1976), Brophy and Evertson (1974) and Terry and Cohen (1977) believed that a teacher's behavior in providing feedback to student miscues was an integral part of the student correction process. They further believed that such teacher behavior could provide an understanding of particular theoretical frameworks held by individual teachers (Mitchell, 1979). Carroll and Chall (1975) suggested that teacher beliefs about reading heavily influence their instructional practices. Harste and Burke (1977) agreed and stated further that teachers do indeed have identifiable models of reading which are reflected in their teaching strategies. Bawden, Burke and Duffy (1979) identified five conceptualizations of reading which serve as models of reading: 1) linear skills, 16 2) basal text, 3) natural language, 4) interest, and 5) integrated curriculum. Their study of 23 teachers showed that 20 of them had at least two overlapping conceptualizations. The researchers con- cluded that teachers were more easily identifiable by content centered (linear skills, basal text) and child centered (natural language, interest, integrated curriculum). Additionally, the years of teacher experience affected conceptualizations of reading with more experienced teachers being more child centered. However, the results showed that although teacher conceptualizations were a major force in oral reading feedback, other factors as reading ability, grade level and other con- text variables were of more importance when the two sets of conceptuali- zations were compared. This would indicate that teacher conceptuali- zations are very important in guiding students to correct oral reading responses but should be considered in light of other factors. The investigations of Conrad (1964), Corder (1967), Wickelgren (1965) and Spiegel and Rogers (1980) emphasized the importance of teacher involvement in oral reading programs. Not only is the teacher facilitative in teaching word recognition skills to students during oral reading instruction but integral in guiding the students to be- come self-sufficient readers. Through the use of feedback a teacher may provide students with the necessary training and skills to become more independent readers. The ability to guide students towards independence in reading relies heavily on teacher conceptualizations of reading. Carroll and Chall (1975), Bawden, Burke and Duffy (1979) and Harste and Burke (1977) agreed that teachers do hold specific beliefs concerning reading which affect their behavior when instructing students in oral reading. 17 Although research is scarce concerning the actual possibility of teachers being able to adopt new teaching strategies in light of preconceived strategies, indications are that teachers could adopt new practices if sufficient training and experience were provided. Even with preconceived teacher conceptualizations of reading, teachers should be able to effectively provide meaningful feedback to students which would aid the readers in assuming control of the correction process in oral reading activities. Teacher-Student Interaction Feedback in oral reading situations is viewed as an effective way to improve word recognition (Biemiller, 1970; Brady and Lynch, 1976; Jenkins and Larson, 1978). However, the researchers regarded teacher feedback as a terminal practice which would transfer the re- sponsibility of feedback from teacher to student. In this way the student would adopt an independent feedback process comprised of three different prompts: 1) graphophonic, 2) syntactic, and 3) semantic. Biemiller (1970), Goodman (1970) and Weber (1970) believed that the student would progress from teacher responsibility for providing feedback to accepting responsibility for providing his own feedback. The researchers regarded this as a developmental process shared equally by the student and teacher. Investigations have shown that teachers do use different strategies for providing feedback to students that are based on teacher conceptuali- zations. Although most teachers favor a transfer of feedback responsi- bility from teacher to student, in practice they most often fail to transfer the responsibility. Additionally, teacher perceptions of a student's reading ability could be a major factor in a teacher's adoption and transfer of feedback. Anderson, Brophy, and Everston (1977) and Anderson and Brophy (1976) also recognized the need for transfer of feedback from teacher to student but found that teachers would most often supply the word rather than lead the student through the necessary steps to allow for transfer of feedback. The studies used first-grade students as sub- jects in situations where a faster reading pace was of more importance than interrupting a student to employ a particular prompt. Different results might occur if older students were studied. Investigations by Brady and Lynch (1976) found that teachers did not practice a systematic guidance process which would progress from teacher responsibility for providing feedback to student responsibility for providing his own feedback. Instead, most teachers studied were grouped into one of three categories: 1) teachers who used all types of prompts with little consideration of their appropriateness, 2) teachers who used no prompts, and 3) teachers who used a variety of prompts but with little effort to use particular prompts at appropriate times or with little attention given to the guidance of feedback responsibility from teacher to student. Although the transfer feedback process was well accepted in con- cept, in actual practice it was not adopted by most teachers. This suggests two possibilities for the apparent lack of adoptability: 1) the teachers were not trained to transfer the responsibility of providing feedback to students or, 2) the teachers could not do so. 19 Pehrsson (1974) studied the process of providing feedback from both a teacher's and student's viewpoint. He was interested in establishing whose responsibility it was to provide feedback when fifth- graders read orally from 200-word sixth-grade passages. The students read under one of three conditions: 1) uncorrected in which the students received no help, 2) corrected in which the students were asked to pay attention to words, and 3) unaided in which the students were asked to pay close attention to words. Pehrsson found that readers could indeed provide their own feedback if the teacher provided some type of meaning orientation. Therefore, Pehrsson sug- gested that the transfer of responsibility for providing feedback may vary between types of errors, particular prompts, and the meaning orientation established between teacher and student. The transfer of responsibility for providing feedback could also be affected by the teacher's perceptions of student reading ability. Gumperz and Hernandez-Chavez (1972) found that teachers did indeed vary the amount of feedback between good and poor readers with poor readers receiving more interrupted feedback than the good readers. Allington (1980) investigated the same issue with 20 teachers and 147 good and 120 poor readers on the primary level. The tapes of the oral reading selections were analyzed for the selection (no response), timing (point of feedback), and the content of the prompt (graphemes, phonemes, semantic/syntactic, teacher pronounce and other). The pro- portion scores showed that poor readers received more feedback than good readers, 74% to 31% respectively. For semantically unacceptable responses, the poor readers again received more feedback, 76% to 54%. 20 For teacher-pronounced words the poor readers also received more feedback, 50% to 38% for good readers. The results of the Allington study suggested that teachers felt a greater need to help the poorer readers who showed difficulties in fluent reading and comprehension whereas teachers thought the good readers could figure out the correct responses by themselves. This would indicate that teachers could adjust their conceptualizations according to their perceptions of a student's reading ability. The studies by Biemiller (1970), Brady and Lynch (1976) and Jenkins and Larson (1978) regarded feedback in oral reading activities as a terminal process which must progress in a developmental manner from teacher-directed to student-directed. Anderson, Brophy, and Everston (1977) and Anderson and Brophy (1976) found that teachers do recognize the need to guide students through the transfer process although in actual practice teachers often disregard the developmental steps necessary in the transfer process. This would indicate that teachers vary in the concepts they hold concerning the transfer of feedback and actual practice when working with students in oral reading activities. The studies cited suggest the need for further research into the teacher-student transfer process of feedback in oral reading. Research is necessary to determine the extent to which teachers can instigate their concepts concerning the transfer of feedback into actual practice when instructing students in oral reading activities and their effective- ness in guiding students to accept responsibility of the feedback process. 21 Use of Prompts Niles and Tech (1980) suggested that not only must we be con- cerned with the necessity of feedback in oral reading and the ability of the teacher to transfer the responsibility for providing feedback to the student but with the amount and type of feedback. To be more explicit, what type of prompts should be used and how much prompting should occur by the teacher and by the student? Pearson (1976) stated that teachers have long regarded oral reading errors as things that should be corrected. Although teachers were unsure as to the type of prompt to use in particular situations or how often to prompt, they did believe that failure to correct would promote learning incorrect responses among the students. Pearson re- garded the popularity of such practices by teachers as having ". . .a seductive rationale behind them." He gave the example often used by teachers in correcting oral reading errors as having the student read a word off a flashcard or making lists of words missed to be read and reread by the student. Such practices, Pearson believed, placed total responsibility for prompting on the teacher who would more than likely use the same prompting condition for all miscues. Little attention or responsibility was given to the student to decide his own prompting strategies. Instead, Pearson believed that the ultimate goal of prompting should place the responsibility and choice of prompting on the student with care given that the student be trained to use prompts and prompting strategies effectively. Pearson's ideas for student involvement may be traced to the Goodman line of thought. Goodman (1970) maintained that the student 22 is the main provider of feedback and little teacher feedback is needed. Goodman stated further that errors were ever-present when a student was learning new material and should be taught how to provide his own prompts for guiding instruction. From this view- point the teacher plays a minor role in prompting and the type and amount of prompting becomes the major issue. Jenkins and Larson (1978), however, viewed the issue in a re- versed way from Goodman. If one accepts the Jenkins and Larson (1978) belief then one accepts the belief that the teacher is the main pro- vider of prompts with minimal student responsibility. Using a case study/experimental approach with five junior high students, the re- searchers studied the effects of five different prompting conditions with remedial readers. Of the prompting conditions studied (no cor- rection, sentence repeat, end of page review with teacher-pronounced words, word meaning with teacher providing the word and teacher or student providing the definition, and drill with the word correctly taught to mastery), results indicated that drill was superior with the teacher guiding the prompting process. The least effective prompting condition was the no correction condition where the teacher did not guide the student in the prompting process. The results suggested that teacher involvement and guidance in providing prompts would affect student learning. Niles, Graham and Winstead (1977), working with fourth-grade students, studied two types of prompts to gauge the effects of student- directed feedback where the teacher did not provide a correction con- dition and teacher-directed feedback where the teacher provided a 23 graphophonic prompt. The results of the study showed that students could provide their own feedback effectively when the goal of the oral reading task was to express the meaning of the passage. The results also indicated that teacher use of the graphophonic prompt to aid student corrections could be as effective. However, since only one teacher prompt was studied, it would be impossible to predict the effectiveness of other prompts on the basis of this study. Niles (1979), using third-graders, expanded the original study to include two additional prompting conditions: 1) a semantic prompting condition in which the teacher asked if a meaning change deviation made sense, and 2) a repeat condition in which the teacher asked the reader to repeat a sentence which contained a meaning change but was not explicitly told there was a change. The results indicated that the graphophonic prompt produced less semantically acceptable responses and more responses which changed the author's meaning. The semantic and repeat conditions produced results similar to the uncor- rected condition with no significant differences recorded across any of the four conditions. This would indicate that the graphophonic prompt relied mainly on graphic level information already in the text while the other prompts relied on the semantic level information. The results of the series of studies (Jenkins and Larson, 1978; Niles, 1979; Niles, Graham and Winstead, 1977; Pearson, 1976) investi- gated prompting conditions in a limited scope. The researchers attempted to study the effects of particular prompts on teacher-student inter- action and the transfer of responsibility for prompting. They did not, however, sufficiently explore the effectiveness of particular prompts 24 on student achievement, the ability of teachers to adopt particular prompts, nor the amount of prompting necessary for student achieve- ment. Such concerns should be given increased attention to progress the understanding of prompts and feedback in oral reading instruction. Summary If the goal of oral reading instruction is to guide students toward independency in oral reading tasks, teachers must practice appropriate strategies to aid students during this developmental process. Teachers must be aware of the purpose of oral reading in- struction and adjust their instructional practices to meet these purposes. The use of feedback to students during oral reading activities has been viewed as an advantageous instructional practice when used correctly by teachers (Bourne, 1966; Niles and Tech, 1980). It is imperative, however, that teachers understand the different types of feedback and use appropriate types when guiding students in the cor- rection of oral reading errors. Teachers must be aware of their own conceptualizations of reading and be willing and able to adjust their beliefs to meet the needs of individual readers. As Spiegel and Rogers (1980) suggested, feedback may be a viable part of the instruction program if teachers could adjust their use of feedback strategies with their perceptions of reading. Indeed, teachers must strive to provide meaningful feedback to students which would promote reader self- sufficiency in the oral correction process. The goal of oral reading instruction should not begin and end as a teacher-student dialogue. Instead, efforts should be exerted to 25 establish effective teacher use of feedback which ultimately transfers the responsibility of oral reading instruction from the teacher to the student. It should be regarded as a sequential process with the end result being reader independence. CHAPTER III PROCEDURES AND METHODOLOGY Pilot During the summer and fall of 1980, two pilot studies were con- ducted by the researcher. The primary purpose of each study was to compare the extent to which teachers adopted four methods of prompting when working with average readers in oral reading activities using basal readers as the primary instructional reading material. A second purpose of each study was to compare the use of the four teacher prompts as an effective teaching strategy in relationship to student achievement. Each study was conducted in public schools in Volusia County, Florida, with teachers who were willing to participate. The schools represented the three socioeconomic classes as determined from avail- able data on free lunches and reduced lunches to students. The studies were exploratory in nature rather than quantitative. Emphasis was on studying oral prompting methods in broad terms from which specific questions for indepth research would evolve. A discussion of each pilot study is presented below. Pilot Study I Twenty-four students of varying ages and reading levels were selected from a summer school program in Volusia County, Florida. All participants were attending a remedial math program and were not 26 27 designated as deficient readers. Test results from the Comprehensive Tests of Basic Skills (reading subtests) and from the Nelson-Denny Reading Test (grade 10), administered during the spring of 1980 on a countywide basis, indicated that all participants were average readers on their appropriate grade levels. The student groups yielded the following information: 1) Group I consisted on eight third-grade students from a school of middle socioeconomic status; 2) Group II con- sisted of eight sixth-grade students from a school of low socioeconomic status; and 3) Group III consisted of eight tenth-grade students from a school of high socioeconomic status. Within each group two students were assigned to each of the following prompting conditions: 1) uncorrected, 2) graphophonic, 3) semantic, and 4) repeat. Each student orally read an assigned passage of approximately 200 words to the researcher on two occasions with the researcher supplying the appropriate prompting condition when the student made an oral reading error. The students were asked to recall all they remembered. Only the second readings were considered for evaluation. The results showed that students in grade three could recall more information through use of the repeat prompt. Students in grade six could also recall more information through use of the repeat prompt. Students in grade ten could recall more information through use of the uncorrected prompt. The results of Pilot Study I formulated several questions for further research into the use of oral prompts as an effective teaching strategy. 28 Pilot Study II The primary purpose of the study was to examine the extent to which teachers were able to adopt an assigned prompting condition when working with average readers on the fourth-grade level. The student subjects included 48 fourth-grade students selected from five public schools in Volusia County, Florida, during the fall of 1980. The students were identified as average readers from scores on the Comprehensive Tests of Basic Skills (reading subtests). Six students were randomly chosen from eight classrooms and remained unidentified to the teachers. Eight teachers were selected from the five schools and were randomly assigned to one of four prompting conditions. They were instructed to use the assigned prompt as often as possible when in- structing the average readers in oral reading activities for approxi- mately 30 minutes per day for a one-month period. The researcher observed both teacher and student subjects one day per week during the study and used a coding process to gauge the extent to which each participant adopted the assigned prompt. The four prompting conditions included the following: 1) uncorrected, 2) graphophonic, 3) semantic, and 4) repeat. At the conclusion of the one-month study it was found that teachers could adopt one of the assigned prompting conditions with little difficulty. Although the teachers seemed to adopt the uncor- rected and repeat prompts more readily, little variation in adopta- bility was recorded among the four conditions. 29 Pilot Study II held several implications for further research: 1. The study should be replicated over an extended period. 2. The study should provide qualitative data for analysis. 3. The effects of student achievement associated with the different prompting conditions should be studied. The pilot studies generated several questions for further indepth study. First, could teachers adopt a particular prompting condition over an extended period of time when working with students in oral reading activities? Although the pilot studies indicated that teachers could adopt a particular prompting condition over a short period of time, the adoptability of prompting conditions over a longer period was not established. Second, would student achievement in oral reading increase through teacher adoptability of particular prompting conditions? The pilot studies did not adequately assess student achievement. An extended study would be recommended to measure student achievement gains. Pilot Studies I and II provided the researcher with background information for indepth study into the effectiveness of oral prompts as a teaching strategy. The researcher instigated a three-and-one- half-month study to gauge the extent to which teachers adopted a method of prompting and the subsequent effect on student achievement. A description of the study follows. Research Study The researcher conducted a three-and-one-half-month study to gauge the effectiveness of oral prompts as a teaching strategy. 30 Specifically, the study was concerned with the extent to which teachers could adopt a particular prompting condition when working with average readers in the fifth grade and the effect of the prompting conditions on student achievement. The four prompting conditions under investigation included the following: 1) uncorrected, 2) graphophonic, 3) semantic, and 4) repeat. Sample The subjects were 72 fifth-grade students who were selected from a larger population from six public schools in Volusia County, Florida. The students were randomly selected from a pool of average fifth-grade readers with six students selected from each of 12 class- rooms. The students involved in the study were unidentified to the teachers. All subjects were enrolled in one of the participating schools on September 8, 1980. The six schools selected for the study included the following: Bonner Elementary School, Minerva Bond Long Lake Helen Elementary School, Ormond Elementary School, Pierson Elementary School, Port Orange Elementary School, and Spruce Creek Elementary School. The schools were selected because they contained kindergarten through grade six and had principals and faculties who were willing to partici- pate in the study. The schools also represented the three different socioeconomic levels as determined by available data on free and re- duced lunch counts for each school. Since free and reduced lunches to students are based on family income, classification of the schools into socioeconomic groups may be indicated. 31 Bonner Elementary School and Minerva Bond Long Lake Helen Elementary School were considered the low socioeconomic schools with over two-thirds of the students in each school qualifying for free or reduced lunch. Pierson Elementary School and Port Orange Elementary School were considered middle socioeconomic schools with between one- third and two-thirds of the students qualifying for free or reduced lunch. Ormond Elementary School and Spruce Creek Elementary School were considered high socioeconomic schools with less than one-third of the students qualifying for free or reduced lunch. Twelve fifth-grade teachers were selected from the six schools and were randomly assigned to one of four oral prompting conditions (uncorrected, graphophonic, semantic, or repeat). The teacher sub- jects were observed prior to the study as favoring immediate correction of student oral reading errors. None of the teachers were involved in the pilot studies nor given advanced knowledge of the study. It was assumed that all teacher subjects had equal opportunity for adopting the assigned prompting condition. Instruments The student subjects were designated as average readers from scores obtained from a standardized test, the Comprehensive Tests of Basic Skills (reading subtests) and from scores on an informal reading inventory (using Powell criteria). The Comprehensive Tests of Basic Skills (CTBS) was selected for use in the study since it was administered on a countywide basis. The 32 results of the reading subtests for each student served as a locater for the teacher to identify a group of average readers. All designated average readers were then administered an informal reading inventory (IRI) by the researcher. The passages administered ranged in readability from grade five to grade seven. Powell criteria were used to determine the reading instructional level of each student. Only students who were determined to be on the fifth-grade instructional level were given an opportunity to be included in the study. Comprehensive Tests of Basic Skills (CTBS) As validation for the use of the reading subtests scores of the CTBS, the following information was provided. The standardization of the CTBS (Form Q) at all levels was based on a probability sample of the entire national school population. Included in the sample were approximately 212,000 students from grades 2 through 10 from both public and Catholic schools from all fifty states. Geographically, the sample also represented different types of communities and various socioeconomic levels. The Kuder-Richardson formula #20 was used to determine internal consistency for Form Q. Reliability statistics were calculated on the reading subtests and include the following information: Mean S^L KR#20 Total Reading 48.2 17.74 .95 The validity of the CTBS (Form 0) is based on correlation co- efficients with the California Achievement Tests at appropriate levels 33 and the California Short-Form Test of Mental Maturity. Although the data produce an acceptable measure of validity, caution must be taken when generalizing to other situations. Informal Reading Inventory ( I RI ) An IRI is a practical technique for determining a student's skills or lack of skills in comprehension and word recognition. An IRI consists of graded word lists, graded reading passages, and com- prehension questions for each passage. Results from an IRI can produce an independent, instructional, and frustration reading level for each reader. An IRI was administered to each subject in the study to determine the instructional level for each subject. The San Diego Quick Assess- ment (Graded Word Lists) (La Pray and Ross, 1969) was used for the graded word list section (see Appendix A). Each list consisted of ten words which were read orally by each subject to determine the level at which the reading passages should begin. The reading passages were read by each subject to determine reading instructional level. Each passage contained approximately 180 words on levels five through seven and were arbitrarily assigned a reading level of 5.0, 6.0, or 7.0 to equal the readability levels of the grades they were designed to measure (see Appendix B). Each IRI was scored using Powell criteria (see Appendix C). The criteria was used to score the oral readings and to determine the reading instructional levels. 34 Reliability The reliability of the IRI may be more consistent at the lower grade levels than at the upper grade levels. This may be attributed to increased vocabulary, language structure and content. The IRI used in the study may be regarded as a reliable test instrument using Powell criteria for scoring. Validity An IRI is considered to be valid if the readability of the reading passage is equal to the grade level it is designed for. Second, the administration of the IRI should approximate techniques used in normal classroom instruction (Powell, 1969). The IRI used in the study adhere to the two aspects of validity and is regarded as being a valid test instrument. Procedure Testing The CTBS was administered on a countywide basis prior to the study. The results of the reading subtests for each student served as a locater for the teacher to identify a group of average readers for possible inclusion in the study. Other than locating possible average readers for the study, the CTBS results were not used for specific analysis. An IRI was administered by the researcher to each student who was designated an average fifth-grade reader from the CTBS scores. The IRI consisted of graded word lists, oral reading passages, and 35 oral comprehension questions. Students whose instructional level was fifth grade according to the IRI results were considered average readers. At the conclusion of the study the researcher again administered an IRI to each subject to determine possible changes in reading in- structional levels. The same IRI form was used as in the initial testing of the three-and-one-half-month study. The final IRI testing was conducted with all student subjects during the same week under similar situations. It was the intent of the researcher to determine reading in- structional levels only. Independent and frustration reading levels were not considered for inclusion in the study. Assignment of Teachers The 12 teacher subjects were randomly assigned to one of four identified prompting conditions for correcting student oral reading errors. Three teachers were assigned to each of the following oral prompt conditions: 1) uncorrected, 2) graphophonic, 3) semantic, or 4) repeat. Six average fifth-grade readers, who remained unidentified to the teachers, were observed during oral reading activities. During the three-and-one-half-month study each teacher was in- structed to use the assigned prompting condition with all average readers during oral reading activities. Each teacher was expected to instruct the average readers in oral reading activities for approxi- mately 30 minutes per day using the reading materials already in use in the classroom. 36 Teacher Training All teacher subjects were observed prior to the study as being immediate prompters when instructing average readers in oral reading activities. When a student makes an oral reading error an immediate prompter corrects the error when it occurs. The teachers were en- couraged to become delayed prompters when working with their assigned prompting condition. Each teacher received a one-hour training session on an individual basis concerning the use and instructions of the particular prompt. The teachers were instructed to use the assigned prompt whenever possible when instructing the average readers in oral reading activities. The following terms were selected for each prompt: 1. uncorrected a. (Teachers did not attempt to correct student errors but would provide feedback within a five-ten second interval for unknown words.) 2. graphophonic a. "Look closely at the word and see if the word part looks like . " b. "Look closely at the word and see if the word part sounds like . " 3. semantic a. "Does the sentence/passage make sense?" b. "Do you think that is what the author meant?" 4. repeat a. "Repeat the sentence/passage." b. "Try that sentence/passage once again." 37 No additional training was given to the teachers nor was feed- back offered during the study concerning their use of the prompt. This insured that all teachers received an equal amount of teacher training with assigned prompts throughout the study. Observations The teacher and student subjects were observed one time per week for a 30-minute period during the study. Although observation visits were limited to specific times when the average readers met for reading instruction, the visits were made on different days of the week to in- sure an equal observation schedule among the classrooms. Each classroom was visited 14 times during the study for ob- servation purposes. To control for researcher bias a trained coder was used on two visits to each classroom. A tape recorder was used on two additional visits to each classroom after which the researcher coded the responses. The researcher visited the classrooms on the other ten visits. On each visit a code sheet was used (see Appendix D) to record the possibility for teacher use of the prompt, whether the prompt was adopted, and the result of the prompt on student correction of the oral reading error. Method of Analysis All data cards were punched onto IBM cards for analysis. The SAS Introductory Guide for the Social Sciences (Helwig, 1978) was used to aid computation. The computation of data was done by computer analysis at the University of Florida Computing Center, Gainesville, Florida, using the SAS programs for scientific data. 38 The research hypotheses are stated in null form. An explanation for the statistical analysis is given for the hypotheses. Hypothesis I A There will be no significant adoptability by teachers among the different prompting conditions. Hypothesis IB There will be no significant adoptability of a particular prompting condition over the others. An analysis of variance model was used to gauge the significance of the teachers' ability to adopt the particular assigned prompting condition and to determine if one prompting condition showed more significant gains than the others. These hypotheses were tested at a significance level of a = .03. Hypothesis HA There will be no significant achievement gains by students among the different prompting conditions. Hypothesis IIB There will be no significant achievement gains by students of a particular prompting condition. A hierarchical design was used to gauge treatment effects between methods of prompts and student groups. The hierarchical design was used to determine the influence of the social unit of the subjects. Therefore, each score was subjected to a treatment effect, a group effect, and a residual component reflecting error of measurement. CHAPTER IV RESULTS OF THE STUDY Results A primary purpose of this study was to examine the extent to which teachers adopted a method of teacher prompts when working with average fifth-grade readers in oral reading activities. A second purpose of this study was to examine the use of teacher prompts as an effective teaching strategy with average readers in fifth grade in oral reading activities. Student achievement gains were studied after students had been exposed to one of four designated teacher prompts for a three-and-one-half month period. The four identified prompts were 1) uncorrected, 2) graphophonic, 3) semantic, and 4) repeat. The factors involved in the analysis were teacher observations and prompting conditions. Each of the 12 teacher subjects were observed 14 times to gauge the extent to which each adopted the assigned prompt. Three teachers were assigned to each of the four prompting conditions. Table I reported the identification process for teacher assignment to prompting conditions and the prompt IDs. The results for the testing of each hypothesis were presented along with a discussion of those results. Hypothesis IA--There will be no significant adoptability by teachers among the different prompting conditions. Hypothesis IB— There will be no significant adoptability of a par- ticular prompting condition over the others. 39 40 Table I. Identification Process for Teacher Assignment to Prompting Conditions and Prompt Identification Prompt Prompt ID Teacher ID Uncorrected A 1, 2, 3 Graphophonic B 4, 5, 6 Semantic C 7, 8, 9 Repeat D 10, 11, 12 41 Percentages were determined between the possibility for teacher adoptability of the prompt and actual adoptability by the teacher for each of the 14 observations for each teacher. Results were recorded in Table II. The percentages suggested a variance in adoptability among the four prompting conditions. An analysis of variance was used to test for differences in adoptability among the four prompting conditions. Table III reported the results. The value of the F statistic (16.34) indicated that differences did occur among the adoptability of the four prompting conditions at the .05 level. An analysis of variance was used to test for interaction between the observations and prompting conditions. The value of the F statistics showed no interactions at the .05 level. Table IV reported the results in an abbreviated table. A Modified Bonferroni test was used to make Pairwise Comparisons among the prompting conditions at the .01 level. The results indicated that the uncorrected and semantic prompting conditions were not significantly different in adoptability between one another. The graphophonic prompting condition was significantly different in adopt- ability from the uncorrected and semantic prompting conditions. The repeat prompting condition was not significantly different in adopt- ability from the uncorrected, semantic, nor graphophonic prompting conditions. Table V reported the results. This indicated that the uncorrected and semantic prompting conditions were more readily adopted in the study than were the repeat or graphophonic prompting conditions. The graphophonic prompting condition was reported as one prompting condition in the initial reported results. The condition, however, 42 cor^Lr)OLr>or^.oor~-«u")omo oooLnooooooLnoocoi^. coooMflinmodNoionio i — cmooooi — i — or->.ooooo"> NonninowNrNOLnoninn oooi — or^ooo*3-or~or^ lDOLfHDCOlDC0 03 0tDtLnLno(M Lor-^oooocoooooor^-nro looooooor^-r^oi^oocoto oootoooroLnouoooor^Ln ooooocooiDNstoso'tinN cooooi — r^oor^i — ouii — r^. o LOLnooro>^-oncoOLnoon NNOioujiootocoiDNOLnN Nro«tirnflisoooiOi — cm cn O Q- E C O o s- •i- o_ +j i- o O -C D. CL O re S- S_ Q. CD oor^LnOLOOr^oor^Lnoroo r- oooi-oooooMLnooror^ tDOONinwinonNOUJono O r-r- i- r- i— ojooooi — i — ONOnon NoiuinomsNoinoninfi r- r- r- ooor^or--ooo^i-or^or^. IDOLniOOlDCOWOifl^-WOlO CM 1^ r- r- r- coocooo'd-or^.Lnooooroo tvcoraooiocowrsivfrjcono LO '-'- •- ooouolooc — MnMoono OlDONNON«3N^COOCOCa CO r- r- r- ^> oi^Lnoooior^oooooo-i Lnmr-~>3-Lr>c\jmuDO«d-LOLnoc\i lo r- r- ^■i — ooooffltoraoosnn co-=d-OLO«d-r^rooQLO*d-tooo>3- LO LO co LO >— iootoooor^r>or^oorouD r^Oi — lOLni.ntoi-niouD'd-ousi-n C\J ■- ■- oooirsooooLDOLOonr^-LO ocococoouDNrvcorNO'tmN i — i i i rooroi — r^oor^r— ouni — r~o coocoNiooiot^NLnpsNiom CO ■— •— LnLnooro^i-orocoOLnooco N[NO)OUDlOOCOCOVDNOi^rx CO S- "3 CD +-> a. E O S- ^Mn^-irnDis.cocnor-cMn<* CO fO CD s: >> CQ c CD 48 An analysis of variance was used to determine if significant differences in adoptability by the teachers did occur with the grapho prompt when compared with the three remaining prompting conditions. Table VII depicted an analysis of variance abbreviated table for the grapho prompt. The value of the F statistic (16.34) indicated that a significant difference did occur. Pairwise Comparisons among the prompting conditions were made to determine the difference between the grapho prompt and the other three prompting conditions. A Modified Bonferroni procedure was used with a .01 significance level. Table VIII recorded the results which showed that the grapho prompt mean was significantly different from the means of the other three prompting conditions at the .01 level. The lower mean of the grapho prompt indicated that the prompting condition was not as easily adopted by the teachers in the study as were the other three prompting conditions. Table IX depicted teacher adoptability of the phonic prompt in relation to the remaining three prompts (uncorrected, semantic, and repeat). The mean of the phonic prompt (.00) suggested low adoptability of the particular prompt by the teachers in the study. An analysis of variance was used to determine if significant differences in adoptability by the teachers did occur with the phonic prompt in relation to the remaining three prompting conditions. Table X reported the analysis of variance abbreviated table with the value of the F statistic recorded as 25.28 at the .05 level. This indicated that the phonic prompt was significantly different in adoptability from the other three prompting conditions. 49 Table VII. Analysis of Variance Abbreviated Table for Grapho Prompt Source df SS MS F Grapho Prompt 8 .917 .306 16.34* .0009 Error 8 .150 .019 Observations 13 .767 .060 1.58 .1026 Observations 39 1.466 .038 1.01 .4743 and Prompts Error 104 3.883 .037 *p_ < . 05 50 Table VIII. Pairwise Comparisons Among Prompting Condi- tions Using a Modified Bonferroni Procedure: Grapho Prompt Prompt Mean Grouping* Uncorrected .76 A Semantic .76 A Repeat .67 A B Grapho .58 B *£ <-.01 *Means with the same letter are not significantly different. 51 4- 4-> O O- B c o o s- •i- D- +-> s- a o i- Q- C o o S- -C D. Q- cOMnouiosooMnono ooiorNiDrxc3iDinouDNO- r- r- oooooooooooooo OOOOOOOOOOOOOO o o OOOOOOOOCMOOOOO oooooooo.— ooooo o o o o oor-00000000000 OOr— ooooooooooo o ooouDOOrOLnoLnoror^LD OK)COCOOU3NNCONO>*lflS «- r- i- ooocoi — r~. o o i^- i— o tr> i — r-» o 00 "- "- minooto>jonoDomooci 00 o rd 1— +-> Q. E o <- i — Nn>tirnDNcoaiOi — c\iro<^- >1 DO re CQ r0 O) 52 Table X. Analysis of Variance Abbreviated Table for Phonic Prompt Source df SS MS F Phonic Prompt 1 .655 .655 25.38* Error 4 .103 .026 Observations 13 .598 .046 1.15 Observations 13 .300 .023 .58 and Prompts Error 52 2.078 .040 *£ < .05 53 Table XI recorded the results of the Pairwise Comparisons among the phonic prompt and the remaining three prompting conditions. Using a Modified Bonferroni procedure at the .01 level, the results showed that the phonic prompt was different in adoptability than the other three prompts. The results indicated that the phonic prompt was not as readily adopted by the teachers in the study in relation to the other prompts. Hypothesis IA was rejected. Significant adoptability did exist among the different prompting conditions. The results indicated that the uncorrected and semantic prompting conditions were not different between one another whereas the graphophonic prompt was significantly different from both the uncorrected and semantic prompts. The repeat prompt did not show significant differences in relation to the other three prompting conditions. Hypothesis IB was rejected. Significant adoptability by teachers of one prompt from another did exist. The results indicated that the uncorrected and semantic prompts were more readily adopted by the teachers than the repeat or graphophonic prompts. Several factors could have influenced the order of adoptability of the four prompting conditions. The teacher correction statement for particular prompting conditions could have varied in difficulty. The uncorrected teacher prompt required no verbal statement whereas the graphophonic prompting conditions required specific graphic and phonic statements. The semantic and repeat prompting conditions each used a one-statement correction. The use of instructional materials in oral reading activities could have influenced the order of prompt adoptability. Although each 54 Table XI. Pairwise Comparisons Among Prompting Condi- tions Using a Modified Bonferroni Procedure: Phonic Prompt Prompt Mean Grouping* Uncorrected .76 A Semantic .76 A Repeat .67 A B Phonic .00 B £ < .01 *Means with the same letter are not significantly different. 55 teacher used a basal text and materials designated for fifth-grade readers, the basal texts were not from the same series or publisher. Readability, content, and vocabulary approach could have differed among the materials used in each classroom. This could have affected the teachers' use of the assigned prompts and the students' ability to handle the material . A third factor of concern in the order of teacher adoptability of the prompting conditions was the actual use of the assigned prompt when the researcher was not visiting the individual classrooms. Although each teacher subject was instructed to use the assigned prompt as often as possible when correcting the average students' oral reading errors, the researcher could not be certain that the instructions were carried out during the instructional reading periods when the researcher was not present. This could have affected the amount of practice each teacher received in the use of the assigned prompt and the student achievement gains. A fourth consideration of the results of the order of teacher adoptability of the prompting conditions was the grouping of students among the six schools. Certain teacher subjects taught in a depart- mentalized situation and taught only reading to the entire fifth-grade population. This particular grouping situation could have provided the teachers with more opportunity to practice the assigned prompt than teachers in a self-contained classroom situation who taught reading on a lesser time plan. A fifth factor could have been the interest of individual teachers in participating in the study. Although the teacher subjects were willing to participate at the beginning of the study, certain teachers 56 showed a lack of interest in using the assigned prompt towards the conclusion of the study. This could have affected the teachers' adoptability of the assigned prompt. A final consideration could have been the control of the students by individual teachers to maintain order in certain classrooms. Two particular teachers showed difficulty in managing student behavior which affected the amount of time spent on actual instruction. Thus, teacher adoptability of particular prompting conditions could have been affected. Hypothesis IIA stated that there would be no significant achieve- ment gains by students between the different prompting conditions. Hypothesis I IB stated that there would be no significant achieve- ment gains by students of a particular prompting condition. Table XII depicted the means and standard deviations for student scores from an informal reading inventory. As discussed in Chapter III, the IRI score values were arbitrarily assigned as 5.0, 6.0, and 7.0 based upon the grade levels the passages represented. The pretest and posttest means for each prompting condition were recorded. Table XIII recorded the number of student subjects in the final analysis, the mean scores, and standard deviations for student posttest scores by each teacher. Table XIV recorded the frequency and percentage of students on instructional reading levels 5, 6, and 7 at the conclusion of the study. The results indicated that of the 67 student subjects included in the final analysis, 22 were on the fifth-grade instructional reading level, 42 were on the sixth-grade instructional reading level, and 3 were on the seventh-grade instructional reading level. 57 Table XII. Means and Standard Deviations for IRI Scores Pretest Posttest Prompt X S X S Uncorrected 5.00 0 6.06 .57 Graphophonic 5.00 0 5.71 .47 Semantic 5.00 0 5.66 .49 Repeat 5.00 0 5.44 .51 58 Table XIII. Mean and Standard Deviations of Student IRI Scores by Teacher Prompt Teacher ID N Mean Standard Deviations Uncorrected 1 6 6.14 .41 2 4 5.75 .50 3 6 6.17 .75 Graphophonic 4 6 5.67 .52 5 6 5.67 .52 6 5 5.80 .45 Semantic 7 6 6.00 .45 8 6 5.33 .00 9 6 5.67 .52 Repeat 10 5 5.40 .55 11 5 5.60 .55 12 6 5.33 .52 59 Table XIV. Frequency and Percentage of Students: Prompt- ing Condition by Instructional Reading Levels Prompt Instructiona 5 1 Reading 6 Levels 7 TOTAL Uncorrected 2 2.99 11 16.42 3 4.48 16 23.88 Graphophonic 5 7.46 12 17.91 0 0.00 17 25.37 Semantic 6 8.96 12 17.91 0 0.00 18 26.87 Repeat 9 13.43 7 10.45 0 0.00 16 23.88 TOTAL 22 32.84 42 62.69 3 4.48 67 100.00 60 This indicated that 33% of the student subjects scored on the fifth-grade instructional reading level, 63% scored on the sixth- grade instructional reading level, and 4% scored on the seventh-grade instructional reading level. The results suggested that more students were on a sixth-grade instructional reading level at the conclusion of the study than on a fifth-grade or seventh-grade instructional reading level. Since average readers at the end of fifth grade would be expected to be on a 5.9 instructional reading level, the results indicated that achievement gains for the majority of the student sub- jects were average. An analysis of variance was used to test for significant differences in student achievement gains within prompting conditions. Table XV re- ported the results. The value of the F statistic (3.48) indicated that significant differences in student achievement gains were recorded among the different prompting conditions. Duncan's Multiple Range Test was used to determine which prompting condition realized significant student achievement gains in relation to the others at the .05 level. Table XVI recorded the results which indicated that student achievement gains were significantly higher for the uncorrected prompting condition. The graphophonic, semantic, and repeat prompting conditions recorded no significant differences in stu- dent achievement gains in relation to one another. Pairwise Comparisons using the Bonferroni T-test at the .01 level were conducted to determine differences in student achievement gains among the different prompting conditions. The results showed a signifi- cant difference between the uncorrected prompting condition and the 61 Table XV. Analysis of Variance Abbreviated Table Within Prompting Conditions. Source df SS MS * Prompt 3 2.711 .904 3.48 Teachers Within 8 2.121 .265 1.02** a Prompt Error 55 14.283 .259 *.05