COMPREHENSION or RATE - ALTERED DISCOURSE 3v PRIMARY SCHOOL CHILDREN WITH IDENTIFIED . AUDITDRY OR VISUAL STRENGTHS Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY DAVID C. BROSKI 197.4 g'LIBRA R y E ‘3 Unrest-sit“ l: ! ».. Jr" ' This is to certify that the thesis entitled COMPREHENSION 0F RATE-ALTERED DISCOURSE BY PRIMARY SCHOOL CHILDREN WITH IDENTIFIED AUDITORY OR VISUAL STRENGTHS presented by DAVID C. BROSKI has been accepted towards fulfillment of the requirements for PhoD 0 degree in InStrUCtional Development & Technology (QLWJO W W Major professor . Date W7 2l1/f77 0.7 639 ABSTRACT COMPREHENSION OF RATE-ALTERED DISCOURSE BY PRIMARY SCHOOL CHILDREN WITH IDENTIFIED AUDITORY OR VISUAL STRENGTHS BY David C. Broski The major purpose of this study was to investigate the applicability of advantaged auditory instruction with learning disabled primary school children. Advantaged auditory instruction, in essence, implies that teaching to the learner's strengths and bypassing his deficits results in effective learning. The key notion underlying such a concept is that instruction be designed to capitalize upon individual strengths of learners. Deficits are not remedia- ted directly. To investigate this concept of advantaged auditory instruction, a sample of thirty learning disabled primary school children with reading problems were divided into two groups, auditory and visual, on the basis of identified communication channel strengths measured by the Illinois Test of Psycholinguistic Abilities. Each group listened to passages of connected discourse at an expanded, normal, and compressed rate of presentation. Each passage was David C. Broski followed by an oral administration of a cloze technique comprehension measure. The cloze test had been modified ”by a standardized algorithmic procedure to provide subjects with two alternatives from which to choose in supplying de- leted words. The hypothesis of major interest was that thosesubjects identified as auditory learners would be more able to comprehend rate-altered instruction than those identified as visual learners. An Analysis of Covariance was used to control for initial differences in order to equate the groups formed through non-random assignment. Reading comprehension was used as the covariate. Academic grade level served as a blocking variable to increase precision. The design was two-way, fully crossed, with a single repeated measure. The two design variables, Type of Learner and Academic Grade Level, were analyzed with an Analysis of Covariance. The repeated measure variable of Word Rate and the interaction of Type of Learner and Word Rate Measure were analyzed by an Analysis of Variance. All hypothesis test- ing was done at the .05 level of confidence set prior to the experiment according to traditional educational re- search conventions. Findings Four major hypotheses were tested in the investiga- tion. None were rejected at the .05 level of confidence. David C. Broski The obtained probability of the F-ratio for Type of Learner main effect and Academic Grade Level main effect was .08. Conclusions Based on the findings, since the null hypotheses could not be rejected, the alternative hypotheses could not be accepted. Implications While the findings of this investigation were not significant at the .05 level of confidence, the potential value of advantaged auditory instruction as an educational tool could not be summarily dismissed. Through retrospec- tive analysis, two potential sources of investigator error were discussed. The discussion led to recommendations for further study as well as implications for the use of ad- vantaged auditory instruction. COMPREHENSION OF RATE-ALTERED DISCOURSE BY PRIMARY SCHOOL CHILDREN WITH IDENTIFIED AUDITORY OR VISUAL STRENGTHS BY .r‘fi David CL Broski A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1974 To Shari and Amy ii ACKNOWLEDGMENTS This study could not have been completed without the cooPeration, expert assistance,and encouragement of many people. My most sincere gratitude goes to those who cared enough to share in this effort. An initial installment of thanks on a long-term debt of love and encouragement goes to my parents, Curtis and Virginia Broski. To Joseph and Sophie Undraitis goes my appreciation for never wavering in their confidence and patience. Dr. James R. Nord, committee chairman and thesis director, supplied me with assistance when it was needed, direction when it was asked for, and warm encouragement at all times. More importantly, perhaps, he provided a model to emulate in future professional endeavors. To my committee members: Dr. J. Dennis Hoban, Dr. James L. Page, and Dr. Randall P. Harrison goes my sincere respect and appreciation for their assistance. My colleagues in the Great Lakes Region Special Education Instructional Materials Center: Nancy A. Carlson, James L. Fleming, 8. Joseph Levine, and Richard iii Burtschi, S.J. provided expert information and technical assistance at critical points in the study. The consultants in the Office of Research Consulta- tion, especially George Sargent and Linda Glendening, pro- vided invaluable help in the methodological design of the study. The staff of the Waverly Learning Center: Donna Seney, Dorothy Ludwig, Margaret Boucha, and Elaine Krueger are truly teachers making a difference in the liVes of children. I owe them much for their cooperation. My fellow graduate students, especially Paul Wood- worth and Randy Gross, have helped to make the doctoral ex- perience at Michigan State the most rewarding time of my life. My wife, Sharon, and my daughter, Amy, have sacri— ficed the most during this time. Yet, they continued to offer constant love, warm encouragement, and patient under— standing. It is not my accomplishment that this study is complete. It is ours. This dissertation is dedicated to them. iv TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . vii LIST OF FIGURES . . . . . . . . . . . . ix LIST OF APPENDICES . . . . . . . . . . . X Chapter I. STATEMENT OF THE PROBLEM . . . . . . 1 Need . . . . . . . . . . . . l Auditory Communication . . . . . . l Advantaged Instruction . . . . . . 5 Purpose . . . . . . . 9 Limitations of the Study . . . . . . 10 Definitions of Key Terms . . . . . . 12 Theory and Rationale . . . . . . . 15 Hypotheses . . . . . . . . . . 17 Overview of Thesis . . . . . . . . 17 II. REVIEW OF RELATED LITERATURE . . . . . 19 Listening and Reading Skills . . . . 19 Development of Listening and Reading Skills . . . . . . . . . . 20 Listening Comprehension Measures . . 27 Individual Differences . . . . . 33 Auditory and Visual Modality Consid—. erations . . . . . . . . . . 37 Rate-Altered Instruction . . . . . . 40 Review of the Illinois Test of Psycho- linguistic Abilities . . . . . . 46 Descriptions of the Tests . . . . . 48 Research with the ITPA . . . . . . 49 Summary . . . . . . . . . . . 52 III. DESIGN OF THE STUDY . . . . . . . . 54 The Population . . . . . . . . . 54 Taking the Sample . . . . . . . 56 Chapter Page Design . . . . . . . . . . . . 62 Design Matrix . . . . . . . . . 65 Hypotheses . . . . . 67 Type of Learner Main Effect . . . . 67 Grade Level Main Effect . . . . . 67 Word Rate Measure Main Effect . . . 68 Interaction . . . . . . . . . 68 Stimulus Material . . . . . . . . 68 Comprehension Measures . . . . . . 70 Experimental Procedures . . . . . . 71 Analysis . . . . . . . . . . . 73 Summary . . . . . . . . . . . 74 IV. ANALYSIS OF RESULTS . . . . . . . . 76 Results . . . . . . 76 Type of Learner Main Effect . . . . 76 Academic Grade Level Main Effect . . 77 Word Rate Measure Main Effect . . . 77 Type of Learner by Word Rate Measure Interaction . . . . . . . . . 77 Summary of Results . . . . . . . . 78 Discussion . . . . . . 79 Reading Comprehension and Listening Comprehension . . . . . . . . 80 Word Rate . . . . . . . . . . 81 Summary . . . . . . . . . . . 83 V. SUMMARY AND IMPLICATIONS . . . . . . 85 Summary . . . . . . . . . . . 85 Findings . . . . . . . . . . . 86 Conclusions . . . . . . . . 87 Discussion and Recommendations . . . . 87 Implications . . . . . . . . . . 90 APPENDICES . . . . . . . . . . . . . . 91 BIBLIOGRAPHY . . . . . . . . . . . . . 117 vi LIST OF TABLES Listening Equivalents to Reading Grade versus Reading Grade . . . . . . . . Aural and Visual Vocabulary Development for Children in Grades One Through Eight . Listening and Reading Word Recognition Rates . . . . . . . . . . . . . Five-Month Stability Coefficients for the ITPA Scores for Eight Year Old Children . . Median Internal Consistency Coefficients for Difference Scores Among Four ITPA Subtests . . . . . . . . . . . . Gates-MacGinite Reading Comprehension Reliabilities . . . . . . . . . . Orders of Presentation . . . . . . . Listening Passage Rates . . . . . . . Comprehension Test Reliabilities . . . . Analysis of Covariance Results for Listen- ing Comprehension Scores . . . . . . . Analysis of Variance Results for Listening Comprehension Scores . . . . . . . . Comprehension Test Mean Scores and Standard Deviations by Type of Learner Groups Accord- ing to Academic Grade Level . . . . . . Means and Standard Deviations of Listening and Reading Comprehension Measures by Type of Learner and Academic Grade Level . . . Vii Page 22 23 24 58 59 64 65 70 72 78 78 79 80 Table Page Mean Scores and Standard Deviations at Ex- panded, Normal, and Compressed Rates for Auditory and Visual Groups by Academic Grade Level . . . . . . . . . . . 82 Summary of Results . . 84 viii LIST OF FIGURES Figure Page 2.1 The Process of Auditory Reception . . . 21 3.1 Procedural Design Over Time . . . . . 62 3.2 Design Matrix Over Variables . . . . . 66 ix Appendix A. B. LIST OF APPENDICES Sample Scores for the ITPA . . . . Sample Scores for Gates-MacGinite Read- ing Comprehension Primary Forms A - C . Fry Readability Formula . . . . . Carver Reading-Input Algorithm . . . Passages and Measures . . . . . . Comprehension Test Scores for the Sample Page 94 96 98 102 107 115 CHAPTER I STATEMENT OF THE PROBLEM Need The need for this investigation is developed under two general topics. These are: (l) The auditory channel of communication; and, (2) Advantaged instruction. Auditory Communication Americans live, today, in a verbal society. A large proportion of their communication time is spent in the reciprocal auditory-vocal interchange. To illustrate, it has been hypothesized that, "A person might write a book a year; read a book a month; and speak a book a week; but hears and listens a book a day."1 This phenomenon extends into the American schools and plays an important role in the delivery of instruction. Stanford Taylor, for example, reports that over 50 percent of a child's time in the elementary classroom is spent in the act of listening.2 Further support for Taylor's con- tention comes from Lundsteen when she states that, "Listening lSara W. Lundsteen, "Thinking Improvement Program," unpublished manuscript, 1966, p. 47. 2Stanford E. Taylor, Listening: What Research Says to the Teacher, National Education Association, 1973, p. 3. 1 accounts for over one-half the time in school activities. It is the most continuously needed Language Arts skill; and yet, traditionally, it has been the most neglected."3 In the primary and intermediate grades, according to Taylor, listening skills are more advanced than reading skills for children of average intelligence and scholastic ability. He feels that children in these grades prefer to listen rather than read, if offered a choice. Listening is a more common act, one that children have had many years of . . . 4 practice 1n perform1ng. As children encounter reading problems, the reliance upon listening becomes even more marked. For example, Taylor concludes: In general, less competent students, those judged to be less intelligent and scholastically below average, show a marked preference for listening over reading in most learning situations and do retain more from listen- ing. The slower student depends on the special attri- butes of listening for much of his understanding. In listening, he is assisted in interpreting content by the phrasing and expression oftflmespeaker, while in reading he must construct his own linguistic units in order to realize meaning. Durrell adds further support by claiming, "In all primary grades, listening vocabulary is much superior to reading vocabulary. Listening is a broader channel for 3Sara W. Lundsteen, op. cit., p. 47. 4Stanford E. Taylor, Op. cit., p. 16. 5113351., p. 17. acquiring information than is reading at this level, since reading skills are immature."6 Although the auditory channel of communication is heavily relied upon as the primary information receptor for many children, research and material development in the area lags far behind efforts in reading. Keller, for example, terms listening research, "embryonic" in comparison to re- search in reading and speaking.7 Anderson, in calling for needed research in listening reports the following: Unfortunately, however, progress in the teaching of listening is hampered by the paucity of research in the field. In comparison with reading, virtually no research has been done in listening.8 He concludes: The more than 3,000 studies which have been made in the field of reading have given us considerable understanding of the physiology, psychology, and sociology of reading. We have no such body of know- ledge about listening. Similarly, reading material development outdistances auditory material development. For example, a survey of 382 members of the Council for Exceptional Children's Divi— sion on Children with Learning Disabilities was recently 6Donald D. Durrell, "Listening Comprehension versus Reading Comprehension," Journal of Reading 12, 6 (March, 1969): 457. 7Paul W. Keller, "Major Findings in Listening in the Past Ten Years," in Listening: Readings, ed: Sam Duker (New York: The Scarecrow Press, Inc., 1966), p. 145. 8Harold A. Anderson, "Needed Research in Listening," Elementary Education English Journal 34 (April, 1952): 216. 9 Ibid., p. 223. conducted to identify materials and techniques used with learning disabled children. By far the greatest number of identified materials and techniques were in the areas of "remedial reading" and "visual perception" (44 percent). Only 5 percent of the materials and techniques identified related to the auditory area and they were defined as "audi- 10 tory perception" materials. Although a proportion of those children identified as learning disabled suffer read— ing problems, it appears that few auditory materials are currently available for their instruction. This lack of attention to auditory channel learning is apparent throughout much of the educational system and is not isolated within those materials employed with the learning disabled. For example, a report of the United States Department of Health, Education, and Welfare expresses this concern: Although educational studies have consistently demonstrated that the major percentage of children's class time and of adult communication is devoted to listening, proficiency in auditory communication skills has received only superficial and insufficient consideration. Lack of attention to children's listen- ing competence is apparent both in the area of instruc- tion and of research. Emphasis in elementary education has been directed to the development and perfection of those visual skills which facilitate learning; educa- tors have been tireless in programming diagnostic, developmental, and remedial reading programs. Pub- lishers have been unceasing in the production of 10Corrine Kass and Rena Lewis, "Favorite Methods and Materials of the DCLD Membership," Division for Children with Learning Disabilities Newsletter III, 1 (Winter, 1973). illustrated books and visual aids. Auditory skills, meantime, have been neglected or taken for granted despite outstanding advances in audio educational equipment. To date, reading research and development efforts have far surpassed such efforts in listening. The need for systematic investigation into auditory channel instruction seems appar- ent. Advantaged Instruction There is a current emphasis upon individualizing in- struction with optimum efficiency for educators at all lev- els. Individualizing instruction, for purposes of this study, is defined in terms of advantaged instruction. The Consortium on Auditory Learning defines advantaged instruc- tion as: Instruction that focuses on those areas that are functioning at the highest levels and that takes ad- vantage of the child's functioning ability to promote further learning; this approach bypasses those func- tioning aspects of the child that are low in an attempt to provide necessary information that can promote normal performance in his environment.12 Advantaged instruction, in essence, implies that teaching to the learner's strengths and bypassing his deficits re- sults in effective learning. The key notion underlying such a concept is that instruction is designed to capitalize upon 11U. S., Department of Health, Education, and Wel- fare, Experimental Listening Curriculum, Project No. 6-8477, 1969, p. 2. 12Glossary of Terms, The Consortium on Auditory Learn- ing, unpublished manuscript, Michigan State University, 1974, p. l. individual strengths of learners. Deficits are not remedia— ted directly. If, for example, a figure-ground problem were diagnosed in a child, an instructional program would be de- signed to bypass this problem rather than to remediate it. The child's strengths would be assessed and the instructional programming would be designed to match those strengths. Ad- vantaged instruction is, then, matching the medium of in- struction to learner strengths-—not to remediating deficits. The advantaged instruction concept has been opera- tionalized, to some degree, at Oakland Community College, Bloomfield Hills, Michigan. Students are administered a battery of tests upon admission. Instructional formats are recommended matched to strengths identified in the testing. The administrators of the institution, Joseph E. Hill and Derek N. Nunney, further this notion of advantaged instruc- tion by claiming: "An aim of the teacher is to diagnose the style of the student, determine his strengths, and begin to instruct him, utilizing media which will capitalize on his strengths. The task, then, is one of matching the cognitive style of the student to the style of the mode of presenta- tion of information."13 One method of individualizing instruction in an ad- vantaged manner may be the diagnosis of communication channel l3Derek N. Nunney and Joseph E. Hill, Personalized Educational Programs, Oakland Community College Press, p. l. strengths prior to instructional programming. Scholars in- dicate that a segment of the general learning population ex- perience difficulty when attempting to process information through the visual channel. For example, Wunderlich states, . . .clinical data suggests that as many as 25% of children may be primary auditory learners."14 Wepman takes this identification of auditory learners one further step by stating, "The need to individualize instruction, at least to the point of grouping visual learners and auditory learn- ers separately at the onset of reading instruction, seems an obvious way to minimize the problem."15 Witkin, in discussing applications of listening re- search for educators, substantiates the grouping by strength idea by making this observation, "Because some children learn better auditorially than visually, auditory educational meth- ods are assuming a larger role in the educational process."16 By diagnosing the generalized attribute of communi- cation channel strength prior to instructional programming, more effective and efficient learning may result if, in fact, those attributes are matched with the instructional medium. 14Ray C. Wunderlich, M.D., Kids, Brains, and Learn- ing (St. Petersburg, Florida: Johnny Reads, Inc., 1970), p. 137. 15Joseph M. Wepman, "Auditory Discrimination, Speech, and Reading," Elementary School Journal 60 (1960): 332. l6Belle Ruth Witkin, "Auditory Perception—-Implica- tions for Language Development," in Listening: Readings, Volume 2, ed: Sam Duker (Metuchen, N.J.: The Scarecrow Press, Inc., 1971), p. 365. Auditory instruction in the past suffered from a fixed rate of presentation set by the speaker and beyond the control of the listener. This is no longer the case. Rate-altered instruction, a relatively new technology, has the ability to change the rate of presentation with little effect upon vocal pitch and quality. When expanded (slowed) and matched with students needing more processing time (typing students, language students, retardates) they can be more effective than normal rates.l7 When compressed and matched with auditory learners, they may produce equiva- lent achievement in less time than normal rates. Thus, by providing input by way of an advantaged modality, the time saved could be used for additional instruction or remedia- ting deficits. This technology has advanced to the point where stu— dents will soon be able to manage word rates for themselves. The Cambridge Research and Development Group recently an- nounced the granting of a patent for a Variable Speech Con- trol. The mechanism will electronically speed up or slow down recorded speech without distortion. Manufacturers will be able to adapt the device to standard audio cassette re- corders with estimated costs of less than $50 per unit.18 l7Emerson Foulke, "Exploiting the Opportunity to Read by Listening," Learning Through Listening, California State Department of Education, 1973, p. 22. 18Center for Rate Controlled Recordings Newsletter, Vol. 8, No. 2, February 15, 1974, p. 2. A financially affordable innovation will be available for classroom use in the near future. If educators are to make profitable use of such available technologies, it seems im- portant that those students most likely to benefit from such use be identified. Mullaly, in 1972, called for such research. He in— dicated the need for studies that match the learning style strengths of children to rate-altered instruction.19 That rate—altered instruction matched to learner strengths may provide effective and efficient learning will be investi- gated in this study. Purpose This research seeks to identify an advantaged in- structional medium for use by learners who receive and process information primarily through the auditory channel. The fundamental question for this investigation might be framed as: "Will matching the medium of instruction to identified individual perceptual strengths result in more effective and efficient learning?" Specifically, this research has a two-fold purpose. First, to identify those learners most able to benefit from auditory instruction through the administration of an 19Lee J. Mullaly, Comprehension of a Narrative Passage by Primary School Children as a Function of Listen- ing Rate and Reading Comprehension Level (Ph.D. disserta- tion, Michigan State University, 1972), p. 63. 10 existing standardized test (Illinois Test of Psycholinguis- tic Abilities).20 And, secondly, to examine the effect of altering the rate of speech on comprehension as a function of the learners' perceptual strengths identified through the aforementioned testing procedure. This study will attempt to investigate the notion that teaching to perceptual strengths-—matching the medium of instruction to communication channel strengths--is an effective and efficient method of advantaged instruction. Limitations of the Study This study intends to investigate the comprehension of rate-altered instruction as a function of modality strength. Two limitations, however, must be stated. First, this investigation does not intend to train or remediate modality weaknesses. Rather than attempting to change learner characteristics, educators may do well to identify learner styles and match instructional media to those styles. Advantaged instruction based upon individual learner strengths may be more profitable than attempting to remedi- ate deficits. Consideration must be given to ethical questions re- garding the remediation of learning styles. A cautionary 20Samuel A. Kirk, James J. McCarthy, and Winifred D. Kird, Illinois Test of Psycholinguistic Abilities, Re- vised Edition (Urbana: University of Ill. Press, 1968). 11 note from Keogh illustrates the underlying philOSOphy of this thesis: On the basis of the evidence, it is not entirely clear whether permanent changes in functional styles can be achieved easily, if at all. That question awaits empirical test. A matter for consideration at another level is whether such changes should be attempted. There is little question that some modes of perceptual and cognitive organization are more compatible than others with the usual educational tasks; however, these same modes may be less adaptive for other areas of behavior or achievement. Value judgements as to relative im- portance of outcome behaviors affect questions of modification. The issue underlying modification of modes or styles may be essentially an ethical one. Modification may be inappropriate unless an individual child's functional style is so extreme as to be patho- logical or to preclude educational progress, since change toward a modal pattern may reduce or limit a number of aspects of individual differences of value. Traditionally, efforts at modification have been fo- cused almost exclusively on bringing about changes in children. It seems reasonable to respond to differ- ences in children's functional styles by modification or individualization of educational programs and goals. The ethical question is whether the child or the edu- cational system will be the major focus of change.21 Second, this research is not an aptitude-treatment interaction investigation. While learning modality strengths will be identified and will provide the basis for discrete grouping, only one measure will be applied--an auditory measure. It is not the purpose of this study to demon- strate an interaction between separate auditory and visual treatments with auditory and visual learners. It is, 21Barbara K. Keogh, "Perceptual and Cognitive Styles: Implications for Special Education," The First Review of Special Education (Philadelphia, Pennsylvania: Buttonwood Farms, Inc.), 1973, p. 98. 12 rather, to investigate the applicability of a particular instructional technique--rate-altered speech--with auditory learners. Visual learners, for purposes of this study, pro- vide a control group. The somewhat unique characteristics of the pOpula- tion under investigation here dictates to some degree these limitations. The sample will be taken from a population of children with learning problems; primarily, reading problems. These children, in effect, may be experiencing difficulty in visually processing symbolic information. Because this population is composed of children with reading difficulties, auditory instruction may provide an alternative to conventional classroom methodologies. Audi- tory instruction may truly be of an advantaged nature in this situation. It will bypass those low functioning areas of the learners in question while providing instruction through a functional channel. Definitions of Key Terms Definitions of terms used in this study are provided in this section. Terms defined are not arranged alphabeti- cally; rather, they are ordered in groups bearing some degree of relationship. 1. Hearing: The process by which speech sounds in the form of sound waves are received and modified by the ear. 22Stanford E. Taylor, Listening: What Research Says to the Teacher, National Education Association, 1973, p. 6. 22 13 2. Listening: The process of becoming aware of sound sequences. In listening to speech, the person first identifies the component sounds and then recognizes sound sequences as known words through the avenues of auditory analysis, mental reorganization, and/or association of mean— ing.23 3. Auding: The process by which the continuous flow of words is translated into meaning. Auding involves one or more avenues of thought——indexing, making compari- sons, noting sequence, forming sensory impressions, and ap- 24 preciating. 4. Compressed Speech: Speech that has been accel- erated by electro-mechanical processes through the use of the Varispeech I for the purposes of this study. The speed of the narrative passage is reproduced in less time than that of the original recording, materially eliminating distor- tions in vocal pitch and quality as may be common with ac- celerated or rapid speech. 5. Expanded Speech: Speech that is slowed or re- duced in rate from the original recording through a similar process described in (4) above. 6. Modified Cloze Technique: This is a test of comprehension that samples from the domain of the information 23Ibid., p. 6. 24Ibid., p. 6. 14 presented. After a listening passage is presented to the subject, it is presented again with every 5th word deleted. Two Options are given the subject to replace the deleted word. The subject chooses a replacement from the options. 7. Auditory Learner: A child whose preference for or measured strengths would indicate a facility for learning through the auditory modality. 25 8. Visual Learner: A child whose preference for or measured strengths would indicate a facility for learn- ing through the visual modality.26 9. Varispeech I: Used in the study, it is an elec- tronic compressor/expander developed by Professor Francis Lee, an electrical engineering faculty member of M.I.T., and manufactured by Lexicon, Inc. This rate changer in- cludes a cassette transport on which the signal to be al- tered is reproduced, and a small, special-purpose computer which obtains from the input signal the samples that are 27 reproduced consecutively in the altered output. 10. Specific Learning Disability: Children with special (specific) learning disabilities exhibit a disorder in one or more of the basic psychological processes involved 25Glossary of Terms, The Consortion on Auditory Learning, Unpublished manuscript, Michigan State Univer- sity, 1974, p. 1. 26Ibid., p. 6. 27Emerson Foulke, ed., Center for Rate-Controlled Recordings Newsletter, Vol. 7, No. 5, May 15, 1973. 15 in understanding or in using spoken or written language. These may be manifested in disorders of listening, thinking, talking, reading, writing, spelling, or arithmetic. They include conditions which have been referred to as perceptual handicaps, brain injury, minimal brain dysfunction, dyslexia, developmental aphasia, etc. They do not include learning problems which are due primarily to visual, hearing, or motor handicaps, to mental retardation, emotional disturbance, or to environmental disadvantage.28 Theory and Rationale The primary purpose of this study is to examine the effects upon learning of matching the medium of instruction to identified communication channel strengths. This research is conceptually grounded in what Salomon terms, a "Prefer— ential Model" of instruction. Salomon emphasizes the poten- tial of capitalizing upon generalized learner aptitudes in educational programming. To be able to capitaliZe on and make use of an existing aptitude it needs to be general enough so that the learner can use it in processing the new information. If it is not general verbal or visual ability, but more task specific, there is little you can capitalize on. Rather, a more stable attribute, 28U. S. Congress, Subcommittee on Education of the Committee on Labor and Public Welfare, "Notes and Working Papers. . .," Education of Handicapped Children (Washington, D. C.: U. S. Government Printing Office, May, 1968), p. 14. 16 perhaps even a "style" may appear to be more helpful in the present case. Underlying this theoretical model are such assump- tions as: 1. Children bring unique sets of inter— and intra-individual learning abilities to an instructional experience. 2. These abilities may be assessed and measured, at least in a gross, generalized manner. 3. Various instructional media and formats de- mand different learning abilities for success- ful interaction. 4. When learning abilities are matched to the medium of instruction, more effective results are produced. It is important, however, to experimentally investi- gate the underlying constructs of such a preferential model prior to its use in the classroom. General research questions such as the following should be answered. First, will matching the medium of in- struction to individual learner's strengths result in more effective and efficient instruction? More specifically, will students identified as auditory learners be better able to comprehend compressed speech than those identified as visual learners? Also, will different types of learners (auditory or visual) differ in their abilities to comprehend expanded speech? 29G. Salomon, "Heuristics for the Generation of Aptitude-Treatment-Interaction Hypotheses," Paper from the Hebrew University of Jerusalem, Israel, 1971, pp. 12-13. 17 Secondly, are children in higher academic grades more able to use rate-altered instruction than those in the lower grades? And finally, can the ITPA (Illinois Test of Psycholinguistic Abilities) successfully identify those stu- dents most able to benefit from a particular instructional technique? Hypotheses Based upon the purposes of this study and framed within a preferential model of advantaged instruction, the following hypotheses have been drawn. In each case, the ability of "learning disabled" primary school students to learn from rate—altered instruction will be tested with a modified cloze technique. 1. The Auditory learners' mean score will exceed the Visual learners' score. 2. The Third grade level mean scores will exceed Second grade level mean scores, which will in turn exceed First grade level mean scores. 3. The Expanded rate mean score will exceed the Normal rate mean score, which in turn will ex- ceed the Compressed rate mean score. 4. There will no learner by rate interaction. Overview of Thesis The remainder of the thesis is organized in the tra- ditional manner. In Chapter II, the pertinent literature is reviewed under five major categories. In Chapter III, the design of the study is presented, including a description 18 of the sample; the measures developed and employed; the design matrix; the testable hypotheses; and the analysis techniques. The results of the study are presented in Chap- ter IV. A summary, discussion, and implications for future research are presented in Chapter V. CHAPTER II REVIEW OF RELATED LITERATURE The literature review is organized under five major headings. These are: (l) The relationship between listen- ing and reading skills in the primary grades; (2) Listening comprehension measures; (3) Individual differences; (4) Rate- altered instruction; and (5) The Illinois Test of Psycho- linguistic Abilities. Listening and Reading Skills Auditory reception is viewed as a three-stage hier— archy by Taylor.1 These stages are ordered from the gross act of hearing a sound to the finer acts of discrimination and attaching meaning. The first stage in the sequence is termed hearing; the gross act of receiving speech sounds. The second stage is listening; the recognition of sound se- quences as words. In the final stage, auding, the flow of recognized words is translated into meaning. A gross visual analogy to the auditory reception stages of hearing, listening, and auding might be looking, seeing, and reading. lStanford E. Taylor, Listening: What Research Says to the Teacher, National Education Association, 1973, p. 6. 19 20 The stages in the hierarchy and the factors which act upon the auditory reception process appear in Figure 2.1 on the following page. Development of Listening and Reading Skills Consideration of auditory strengths is a central focus of this research. Large amounts of classroom time are spent in the act of listening; especially in the elemen- tary schools. Many scholars have examined the comparative develOpment of reading and listening skills by children in these grades. Durrell, for example, developed comparable, stand— ardized measures of listening and reading comprehension for grades one to eight. These language comprehension tests were equated to provide raw score comparisons between listen- ing and reading. They were standardized on populations of three to four thousand children at each grade level. One index of the listening-reading relationship used by Durrell was that of listening grade equivalents. These grade equiva— lents are presented in Table 2.1.2 From the data in Table 2.1, it can be seen that lis- tening comprehension raw scores exceed reading scores until grade eight where they are equal. Furthermore, the last 2Donald D. Durrell, "Listening Comprehension versus Reading Comprehension," Journal of Reading 12, 6 (March, 1969): 458. 21 Speech Sounds 1 cs 2: ; Auditory Acuity -—-) 5 é—- Masking I: . . . (— Auditor Fati ue Binaural Con51derations —-) Y 9 Attention and Concentration c 3 .2 “a m u H "" ‘3 g S..S 9 Experience and 8 3'2 E Background o -H m - . . . . H m o m Auditory Discrimination —) g H 8 z ‘ Context ............... . «4 to U) i L9 :3 5 (-—— Delivery 5 Z 2 0 H p ' z: 2 m 0 F : m I H I _J ! Identified and E Recognized Word 5 : Rate of Input -d> ‘?'—' Unrelated Associations m Experience and c c o 0 Background m o -H .a o m m o H c w c c: m a) a) ~~l -a a. s u u x E O‘ m 04 to o o o c E -a '2 U U) H H 8 H m o~ E E u £9 a c o o a z .,4 .,4 [‘4 U) a. H x u C a; S “’ g «I <2 /; It Meaning . . 3 Figure 2.1.--The Process of Auditory Reception. 3Ibid., p. 5. 22 TABLE 2.l.--Listening Equivalents to Reading Grade Versus Reading Grade (Vocabulary). Listening Reading Listening above Reading (yrs-mo) Equivalent to S Score Reading Grade core Grade Primary Forms Grade 1 75 3.1 37 1.5 1.6 Grade 2 82 3.6 58 2.5 1.1 Intermediate Forms Grade 3 46 4.6 35 3.5 1.1 Grade 4 54 5.4 45 4.5 .9 Grade 5 61 6.2 55 5.5 .7 Grade 6 69 7.1 64 6.5 .6 Advanced Forms: Grade 7 130 7.9 122 7.5 .4 Grade 8 137 8.5 137 8.5 .0 column of the table indicates the grade equivalent superi- ority of listening comprehension for grades one through seven. This normative data may be used to assess the level of reading a child could understand if the information were presented orally. Studies in the vocabulary development of children reveal similar findings. Armstrong conducted a study of aural and visual vocabulary development for two hundred children in grades one through eight. His findings are 4 summarized in Table 2.2. The data from Armstrong's study 4Hubert Coslet Armstrong, "The Relationship of the Auditory and Visual Vocabularies of Children" (Ph.D. disser- tation, Stanford University, 1953), p. 107. 23 TABLE 2.2.-—Aural and Visual Vocabulary Development for Children in Grades One through Eight. Mean Number of Words Mean Number Of Auditorially Known Age Words Visually (Includes Number Known in Column 2) 6-6 848 3,048 7-5 1,184 3,476 8-6 1,900 4,240 9-6 4,040 5,120 10-6 6,040 6,600 11-5 6,080 6,640 12-6 7,240 7,480 illustrates the apparent strength of the auditory vocabulary for children in the primary and elementary grades. Taylor compared listening and reading skills of children in terms of word recognition rate. This efficiency measure indicates the relative advantage of listening in the early grades in terms of recognition time.5 Not until grade six do children recognize words visually as quickly as those presented auditorially. Furthermore, reading rate does not approach listening rate until grades five or six. It appears that listening is a more efficient channel of instruction in the early grades. If an adequate level of comprehension can be maintained at accelerated listening rates, the auditory JStandord E. Taylor, op. cit., p. 16. 24 oom 0mm 0mm nmm vmm «Hm vow mma mma mnH mma mma mHH om UZHommm quszqu t mma..mha v AmuscflE Hod mpnoz Gav oumm omnuo>< NH. mm. em. mm. mm. mm. mm. Om. mm. mm. mm. mg. mm. mm. OzHodmm .E.m.3 mwalmhaunmpmu mcflxmomm Daspm omwuo>o co pommm OszmBqu RI .omm mm. 1 Amocooom cflv puoz pom mEHB ceeuflcmoomm mmmue>< ucwflo .HOO NH Ha OH 0 w h o m w m m H lemmm >uo> Ho>oq opouo .meumm cofluflcmoomm UHOB mcflpmom pen meecoumeqll.m.m mqmae 25 channel becomes even more efficient in terms of learning per unit of time. Many scholars have indicated the relative superior- ity of the auditory channel in the early grades. Reading skills, at this age, are not yet developed to the same ex- tent as listening skills. This comparative advantage tends to be maintained for those children with reading problems at the primary level and extends until adulthood. For ex- ample, Emslie and others presented eight stories to 132 ele- mentary students during four sittings. One story was read aloud to the pupils and one was read silently at each of the sittings. Poor readers did better in listening than in read- ing.6 Similarly, Friedman reported that elementary child- ren had better comprehension of material presented orally than of material they read. Slow readers showed the great- est difference in favor of oral presentation. Fenwick suggested that many slow-learners, while possessing marked reading retardation when measured by standardized test scores, had developed compensatory 6Elizabeth A. Emslie, Margaret E. Kelleher, and Judy D. Leonard, "A Comparison of Achievement in Silent Reading and Listening in Grade four," in Listening Bibliography, Second Edition, ed: Sam Duker (Metuchen, N.J.: The Scare- crow Press, Inc., 1968). 7Robert M. Friedman, "A Comparative Study of the Retention Level of Verbal Material Presented Visually and Orally: Fifth Grade Pupils," in Listening Bibliography, Second Edition, ed: Sam Duker (Metuchen, N.J.: The Scare- crow Press, Inc., 1968). 26 abilities in aural-oral skills which enabled them to func- tion on a cognitive level substantially higher than that I normally indicated by scores from either achievement test batteries or individual measures of intelligence. He ex- plains, ". . . the true learning capacity of the student might be more nearly realized in terms of those situations depending primarily on instructional strategies involving the aural approach."8 Further research relating listening and reading skills for individuals with reading deficits is reported by Sticht: . . . many men who read poorly preferred to learn by listening rather than by reading. This suggests thattjuaprovision of listening materials in addition to reading materials in training programs and on the job might motivate learning where it otherwise would not even be attempted.9 The listening literature indicates the feasibility of using auditory instruction on many levels. Certainly, it has application in the primary grades. As children leave school they will listen much more than they will read. It seems important that auditory instruction be em- phasized in the classroom, especially for those children with reading problems. If a child with an intact auditory channel 8James J. Fenwick, "Slow—Learners and Listening," Listenigg; Readings, Volume 2, ed. Sam Ducker (Metuchen, N. J.: The Scarecrow Press, Inc., 1971), p. 104. 9Thomas G. Sticht, "Learning by Listening," Paper presented for C.O.B.R.E. Research Workshop on Language Com- prehension and The Acquisition of Knowledge, Rougemont, North Carolina, March, 1971, p. 23. 27 is experiencing difficulty in reading, then a viable alter- native may be aural instruction. Listening Comprehension Measures There is little doubt that reading and listening are related language skills. Durrell, for example, terms be- ginning reading as, ". . . essentially a task of searching for speech patterns in print: letters and letter clusters are related to speech sounds; printed words and sentences are translated into speech."lO Many scholars have investigated this relationship. Duker, for example, reviews twenty—three such studies re- porting coefficients of correlation between listening and reading. These ranged from .45 to .70, with a mean of .59, indicating a strong positive relationship between the two skills.ll Much of this research, however, used available standardized listening comprehension measures which have met with recent criticism. Devine, for example, reports:. Investigators regularly note that both kinds of be— havior are related in that (a) both are concerned with the intake half of the communications process, (b) each seems to be a complex of related skills components, (c) the same higher mental processes seem to underlie both, (d) high correlations exist between test scores in read- ing and listening, and (e) the teaching of one seems to affect the other. loDonald Durrell, op. cit., p. 455. 11Sam Duker, Listening: Readings, Volume 2 (Metuchen, N. J.: The Scarecrow Press, Inc., 1971), p. 69. 28 He questions some of the underlying assumptions re- sulting from past research, however: The second basic assumption about reading and listining to be reexamined in the light of recent research is that listening and reading test scores correlate highly. . . . recent studies have raised serious questions about the listening tests used to establish correlations. Anderson and Baldauf analyzed the Sequential Tests of Listening Comprehension Test (Form 4), and concluded that heavy loadings in verbal comprehension suggested that achievement on the test may be a matter of verbal comprehension and not listening as a distinct ability. Kelly studied both the STEP Listening Comprehension Test and the Brown-Carlsen Listening Comprehension Test, and concluded that the construct validity of each was questionable because neither test correlated significantly higher with the other than with read- ing and intelligence tests. It may be that much of the existing statistical evidence commonly accepted in studies of the reading-listening relationship is in- valid. A third published standardized listening comprehen— sion test, the Orr-Graham, was reviewed in The Seventh Mental Measurements Yearbook. Those who believe that the teaching of all the communication skills is a primary function of the schools may be heartened by the appearance of a new listening test. They will, of course, View with dismay the fact that more than 40 years after Paul Rankin's pioneering inquiry into the state of listen- ing there are only three published standardized tests of listening. They will, moreover, despair over the realization that none of the lonely company of listen- ing tests comes anywhere near the level of sophistica- tion that has been attained in the measurement of other skills. The Orr-Graham Listening Test does not, unfortunately, provide an alternative to the Sequential Tests of 12Thomas G. Devine, "Reading and Listening: New Research Findings," Listening Readings, Volume 2 (Metuchen, N. J.: The Scarecrow Press, Inc., 197l),pp. 82, 84. 29 Educational Progress: Listening or the Brown-Carlsen Listening Comprehension Test for its focus is much too narrow.I3 The review is concluded with the statement, "The education profession is still without a first-rate instrument for the measurement of listening skill."l4 Existing standardized instruments for measuring listening comprehension provide limited alternatives as research tools. The result, then, has been for researchers to develop their own measures. One such development tech— nique is the cloze procedure.15’l6 Carrol defines this form of comprehension as, ". . . the ability to supply missing elements in messages."17 The procedure involves taking a passage of text and deleting words in it by some rule, e.g., every 5th word, every other noun, or every other "function" word. A subject is then presented with the passage and asked to guess the missing words. Usually the passage is presented in written form, in which case the missing words are indicated by blanks of a stand— ard size, but techniques are also available for presenting the passage in auditory form. l3Oscar K. Buros, ed., The Seventh Mental Measurements Yearbook, Volume II (Highland Park, New Jersey: Gryphon Press, 1972), p. 636. 14 Ibid., p. 639. 15Milton Dickens and Frederick Williams, "An Experi- mental Application of 'cloze' Procedures and Attitude Meas- ures to Listening Comprehension," Speech Monographs, June, 1964, 31: 103-108. 6 - . . 1 . . 1 Ronald D. Carver, 'On Tne Relationship Between Understanding and Information Stored During the Reading and Auding of Prose Materials," Unpublished manuscript, 1971. 17John Carrol, ed., Language Comprehension and the Acquisitgqn_gf_gpgglgdgg (Washington: V. H. Winton: dis- tributed by the Halsted Press Division of Wiley, New York, 1972), p. 18. 3O . . . The procedure has gained considerable acceptance as a measure of the individual's degree of comprehension of a given text. Suggested by Foulke for use in testing comprehension of auditory materials, the cloze technique bypasses some of the problems usually inherent in other researcherbdeveloped testing instruments.18 David Orr describes: Most typically, the multiple-choice listening ana- log to the standard reading comprehension test is the thing that gets used to determine comprehension. There are several particular problems with this approach. In the first place, it's very difficult to produce a research instrument which is a psychometrically ade- quate test. . . . There is also a problem which can be called the "domain" problem. Ideally, the test ought to be an unbiased and representative sample from the domain of material which was presented to the listener for comprehension. . . A third point with respect to multiple-choice tests has to do with prior knowledge. Questions which the individual can answer on the basis of knowledge that he got elsewhere than listening to the passage are irrelevant with respect to measuring comprehension, at least of that particular presentation.19 The Cloze technique, however, is not without its problems, too. In factor-analyzing the ability to comprehend time—compressed speech, Carver and others concluded, "The cloze test used as a measure of comprehension includes a large component of variance unrelated to comprehension 18From a taped presentation by Emerson Foulke at Michigan State University, November 14, 1973. 19David B. Orr, "The Measurement of Listening Com- prehension," Proceedings of the Second Louisville Confer- ence on Rate and/or Frequency—Controlled Speech (Louisville, Kentucky, 1969), p. 220. 31 and quite specific to the cloze technique itself."20 Carver does report, however, "The normal Cloze variable loaded high- est on the Cloze factor (1.10), yet it also loaded substan- tially on the Comprehension factor (.42). This latter re— sult lends support for the definition of Factor I as a ..21 Comprehension factor. He further states, "The Cloze task is a recall task which is probably quite appropriate for studying the memorization of prose material."22 The standard cloze technique requires subjects to supply deleted words from a prose passage previously pre- sented in its entirety as a stimulus. This task has been found to be inappropriate for use in the primary grades. Deutsch and others consider the standard cloze procedures too difficult for adequate listening comprehension measure- ment with young children.23 20Ronald P. Carver, Raymond L. Johnson, and Herbert L. Friedman, "Factor Analysis of the Ability to Comprehend Time-Compressed Speech," Journal of Reading Behavior, 4, 1 (Winter, 1971-72): 10. 21 Ibid., p. 47. 22Ronald P. Carver, op. cit., p. 10. 23Martin Deutsch, Alma Maliver, Bert Brown, and Estelle Cherry. Communication of Information in the Ele- mentary School Classroom, Cooperative Research Project No. 908 (New York Medical College: Institute for Developmental Studies, 1964), p. 82. 32 A modified cloze technique termed, "Reading-Input," has been developed to replace standard cloze procedures.24 Reading—Input may be considered a second generation Cloze technique, retaining the advantages of a standard cloze, but having two distinct benefits beyond it. First, Reading-Input measures differ from traditional cloze measures in that options are given subjects to choose from in replacing the deleted word. For example, if the fifth word of a passage were deleted and the subject were asked to supply that missing word, a Reading-Input measure would give the subject two or more alternatives to choose from in replacing the deleted word. Traditional cloze meas- ures do not allow the selection from alternatives. This greatly reduces the difficulty of the task, making it more appropriate for younger children. Secondly, Reading-Input measures have a further ad- vantage in that a standardized algorithmic procedure is used to construct the measures. In this manner, experimenter bias can be eliminated in both the construction of the ma- terials and the scoring of the results. Carver points out that the Reading-Input technique bears resemblance to the standard cloze in both format and use . 4Ronald P. Carver, "Revised Procedures for De- veloping Reading-Input Materials and Reading—Storage Tests" (Washington, D.C.: American Institutes for Research, Octo— ber 12, 1973). 33 First of all, reading-input may be used to assess the readability or difficulty of materials as Taylor originally used the cloze technique. Reading-input may also be used to facilitate the amount learned from prose materials in a manner similar to the way that Cloze has been employed. A third possible use of reading-input would be to facilitate improvement in reading skill in a manner similar to the way that Cloze has been employed. Fourth, the reading-input technique could also be used to assess the degree to which an original prose passage had been understood or compre- hended, as the cloze technique has been used.25 Reading-input measures of listening comprehension appear to offer a viable alternative to the three published listening tests available. (STEP, Brown-Carlsen, and Orr- Graham). Additionally, the standardized reading and listen- ing comprehension tests developed by Durrell are available with only single sentence comprehension measures at the primary level. Furthermore, the reading-input teChnique seems to offer advantages to researchers developing their own measures, especially at the primary level. Individual Differences Providing for individual differences in learning situations has received much attention in educational lit- erature. While most educators would agree that accommoda- tions should be made for aptitudes of learners entering instructional systems little, seemingly, is evidenced in 251bid., pp. 10-11. 34 the nation's classrooms. James Beaird expresses it this way: It is true that we have spent, as educators, a great deal of time and effort talking about provisions for individual differences in our instructional pro- grams. Unfortunately, woefully little active adapta- tion to individual differences has been made in instructional programs. This is especially true in the development of instructional media.26 Beaird continues by indicating profitable emphases for in- structional technologists: The most fruitful approach and the one that Cron- bach (l967)28 terms the most "psychologically interest— ing" approach is that of modifying the instructional setting such that it is adaptive to salient and mean- ingful psychological differences in individuals. While the latter course is not an easy one to negotiate, increased attention to such alternatives by instruc- tional technologists can significantly contribute to successful adaptations to individuality.27 Concerns of educators might be viewed, then, as involving the matching of media, materials, and methodolo— gies to the relevant characteristics that individual learn— ers bring with them to the classroom. As Cronbach states, 26 James H. Beaird, "Learner Variables and the In- structional Technologist," The Contribution of Behavioral Science to Instructional Technology, Oregon: Teaching Research, A Division of the Oregon State System of Higher Education, III—3. 27Ibid., III-4. 28U.S., Department of Health, Education, and Wel- fare, Lee J. Cronbach and Richard E. Snow, Final Report: Individual Differences as a Function of Instructional Variables, Contract No. 4-6—061269-1217, p. 175. 35 "The demand is that educators invent new programs to open opportunity to persons who would not succeed in attaining traditional goals in traditional ways."29 The development of new technology should greatly extend the range of methodologies for teachers to deal with learner variations. Hopefully, these innovations will be designed to teach to perceptual strengths of students. The process of remediation is of questionable value. As Beaird reports: The underlying assumption to remediation, of course, is that once the student has eliminated his deficiencies he may rejoin the rest of the group ready to try again. Logically it can be expected that remediation will lead to improvement in many important aspects. All too often, the group, continuing at its pace, has tended to move even farther ahead of the remedial student; thus plac- ing his future success in a tenable position. Keogh cites the importance of individual variations in cognitive style by stating: It is presumed that cognitive styles interact with instructional strategies to facilitate or impede learn- ing. Inclusion of cognitive style as a significant variable in planning of curricula and instructional programs may maximize learning efficiency. Despite Special education emphasis on individualization of program, application of the construct of cognitive style to exceptional children has been limited. The area of study seems promising, especially when applied to children with educational problems. 291bid., p. 175. 30James H. Beaird, op. cit., III-l4. 31Barbara K. Keogh, 0p. cit., p. 84. 36 There should be payoff for children if their per- ceptual strengths are matched to the instructional program. Sabatino and Streissguth divided a sample of children with learning disabilities into two groups--one with auditory perceptual strengths, the other with visual perceptual strengths. An experimental Word Form Configuration Training Program (all visually oriented) significantly modified visual perceptual behavior in the experimental groups of visual children, but not the EHKUJ£3 children. They con- cluded that teaching to perceptual strengths, rather than concentrating on remediating weaknesses appears to be a more effective strategy.32 Some researchers have investigated the efficacy of choice and preference in educational programming rather than a measured assessment of learner aptitude.33 Sticht, how- ever, found that preference for one mode of instruction over another (listening over reading) did not predict suc- 34 cess in that channel. Although student involvement in the 32David A. Sabatino and W. O. Streissguth, "Word Form Configuration Training of Visual Perceptual Strengths with Learning Disabled Children," Journal of Learning Dis- abilities, 5, 7 (August-September, 1972). 33Mark L. Berman, "The Effects of the Magnitude of Response Cost and Reinforcement on Speed and Accuracy on Programming Materials" (Ph.D. dissertation, Arizona State University, 1969). 34Thomas G. Sticht, "Learning by Listening, Paper presented for C.O.B.R.E. Research Workshop on Language Comprehension and the Acquisition of Knowledge, Rougemont, North Carolina, March, 1971. 37 selection of instructional formatting is to be encouraged, preference for a particular format may not serve as well as measured ability in the prediction of success. This may be especially true in the primary grades. Differential educational programming for individual learner characteristics seems to provide an effective and efficient avenue for instructional design. Faced with the practical constraints of schools today, homogeneous group- ing of learners might be a starting point. Auditory and Visual Modality Considera- tions Much research has been conducted on the efficacy of audio, visual, and audio-visual presentations. Day and Beach analyzed thirty-four studies comparing visual and auditory presentations. From these, eleven generalizations were drawn. Those most pertinent to this study are pre- sented: l. Meaningful, familiar material is more efficiently presented aurally; meaningless, unfamiliar material is more efficiently presented visually. 2. The greater the intelligence of the receiver, the greater the relative advantage of a visual presentation. 3. The greater the reading ability of the receiver, the relatively more effective a visual presentation. 38 4. The relative efficiency of a visual presentation increases with age. At the age of six, a visual presentation is less effective than an aural presentation. At the age sixteen, a visual presentation may be more effective than an aural presentation. 5. Material that is organized and related--such as prose or factual information--is better understood with an auditory presentation; material that is comparatively dis— crete and unrelated--such as a code--is more effectively received with a visual presentation. 6. The comprehension of material can be tested either by the ease with which the material is learned or by the amount that is retained after a period of time. As a rule, measures of learning tend to favor a visual presen- tation, while measures of retention are higher after an auditory presentation.35 Edling and Paulson, in reviewing the literature of learning modalities and instructional media report the find- ings of Fairbanks, et. a1.:36 35Willard F. Day and Barbara R. Beach, "A Survey of the Research Literature Comparing the Visual and Auditory Presentation of Information," A. F. Technical Report No. 5921, November, 1950. 36Jack V. Edling and C. F. Paulson, "Understanding Instructional Media," The Contribution of Behavioral Science to Instructional Technology, Oregon: Teaching Research, A Division of the Oregon State System of Higher Education, IV-l:17-18. 39 The audio-visual mode was superior to both (either audio or visual), suggesting that the individual dif- fers in capability to handle one or the other mode of presentation, i.e., some people are better able to comprehend materials by one or the other modality, and the individual's preferred mode of receiving information was used in the audio-visual presentation. It does ap- pear_to be an individual difference factor in the capa- bility to use either auditory or visual modes of pre- sentation. 37 Basing his conclusions upon Miller's theory of finite limits to channel capacity, Travers states: . . . the evidence indicates that multiple sen- sory modality inputs are likely to be of value only when the rate of input of information is very slow. The common practice of filling both the audio and the visual channels with a continuous flow of infor- mation would seem to have little support, except perhaps that it may satisfy some of the compulsions of film producers. 8 The debate over single modality versus multiple modality presentations will probably continue for a time. Each modality has its own special applications for instruc- tion. A summary statement from Edling and Paulson reflects the focus of this research: . the issue is no longer audiovisual versus audio versus visual, but what is the optimum rate for presenting information to learners of various capaci- ties regardless of the modality employed.39 c.“ 37G. A. Miller, "The Magic Number 7, Plus or Minus Two: Some Limits on Our Capacity to Process Information," Psychologigal_Review, 1956, (E3, pp. 81-97. 380.8., Department of Health, Education, and Wel- fare, "Research and Theory Related to Audiovisual Informa- tion Transmission," by R. M. W. Travers, Contract No. 3-20- 003, Revised Edition, 1967, p. 267. 39 IV-1:l4. Jack V. Edling and C. F. Paulson, 9p. cit., 40 Rate—Altered Instruction Rate-altered speech is the accelerating or de- celerating of the rate of presentation of auditory informa- tion with little decrease in vocal pitch and quality. The regulation of the rate of presentation of auditory infor— mation may be used to accommodate the needs and abilities of the listener. To illustrate, Taylor suggested that children recog- nize words auditorially much more quickly than they do visu- ally until about grade six.40 These figures were based on an average adult speaking rate of approximately 175 to 185 words per minute. Technologically, devices exist to in- crease the rate of presentation and further accentuate this gap. Instruction matched to this apparent auditory strength is truly of an advantaged nature. The auditory presentations for purposes of this study were varied over three separate rates. One was pre- sented at an expanded rate (95 w.p.m.); one at normal (125 w.p.m.); and one at a compressed rate (175 w.p.m.) to each of the subjects. The relatively simple regulation of the rate of presentation intended to accommodate individual dif- ference variables among the learners. Research has focused on many personological varia- bles of listeners to determine their relationship with 4OTaylor's findings appear in Table 2.3 of this chapter. 41 comprehension of rate-altered speech. Fairbanks and 41 Sticht,42 43 others, and Woodcock and Clark, investigated mental ability and its relation to comprehension of com- pressed speech. As might be expected, comprehension of com- pressed speech generally increases as does mental ability. Rossiter44 investigated sex differences in the comprehension of compressed speech. He examined both the sex of the lis- tener and the sex of the speaker. No significant differ- ences were found. Gropper investigated the interaction of subject char- acteristics with performance at different levels of speech compression in an effort to isolate a set of predictor vari- ables. He found large individual differences indicating that there is not one most efficient speed for everyone. He states, "In most cases, however, a speed much slower than 416. Fairbanks, N. Guttman, and M. Miron, "Auditory Comprehension of Repeated High-Speed Messages," Journal of Speech and Hearing Disorders 22 (1957): 20-22. 42Thomas G. Sticht, "Some Relationships of Mental Aptitude, Reading Ability, and Listening Ability Using Normal and Time-Compressed Speech," Journal of Communication 18 (1968): 243-258. 43R. M. Woodcock and Charlotte Clark, "Comprehen- sion of a Narrative Passage by Elementary School Children as a Function of Listening Rate, Retention Period, and IQ," Journal of Communication 18 (1968): 259-271. 44Charles M. Rossiter, Jr., "Rate-of-Presentation Effects on Recall of Facts and of Ideas and on Generation of Inferences," Audio-Visual Communication Review 19, 3 (Fall, 1971): 313-324. 42 normal will not add much to comprehension, while speeds about twice as fast as normal will take too much away from comprehension to warrant their use."45 Woodcock summarized a series of interrelated studies pertaining to the application of rate-altered speech in the classroom. The series involved approximately seven hundred subjects over a period of two years. The subjects in- cluded normals, mental retardates, and the culturally dis- advantaged. The Ss were drawn from grades 3 through 6, and from classrooms for adolescent mental retardates. Conclu- sions drawn by Woodcock pertinent to this study are as follows: 1. A pupil will achieve the highest score on a test over a passage at expanded rates of 75 to 125 words- per-minute. 2. A pupil's most efficient learning will take place at compressed rates of approximately 250 to 300 words- per-minute. (It is of interest to note that the normal speaking rate of 150 to 175 me provides neither the most effective rate nor the most efficient rate.) 3. In respect to the relationship of performance to intelligence, mental age is a very significant S variable. 45Robert L. Gropper, "Comprehension of Narrative Passages by Fourth Grade Children as a Function of Listening Rate and Eleven Predictor Variables," roceeiincs of the Second Louisville Conference on Rate and! trolled Speech, Louisville, Kentucky, 196 43 IQ, when mental age is held constant, does not seem to be an important variable.46 A pilot study undertaken in 1972, with emotionally disturbed children, measured the listening and reading com- prehension of fifteen subjects using comparable forms of the STEP Listening Tests. A difference in comprehension in favor of listening was found for eight of the fifteen children in the class. For those who showed a better reading compre- hension than listening comprehension, the average dif- ference score was 13.86. For those who showed better listening comprehension than reading comprehension, the average difference score in favor of listening compre- hension was 23.62. On the strength of these results, an effort was made to provide recorded texts for those whose listening comprehension was better than their‘ reading comprehension. Although no formal assessment of the effects of this experiment has been attempted, their instructor is certain that their rate of educa— tional progress has increased.47 In another study, Callaway, Gleason, and Klaeser measured subjects' listening skills before and after ex- posure to compressed speech by use of the STEP Listening East. Their purpose was to examine the effects of exposure to compressed speech upon improvement of listening skills. They found no significant differences. It is interesting to note, however, that they made no attempt to examine dif- ferences in comprehension of rate-altered speech by those 46R. M. Woodcock, "The Application of Rate-Controlled Recordings in the Classroom," Proceedings of the Second Louisville Conference on Rate and/or Frequency—Controlled Speech, Louisville, Kentucky, 1969, p. 100. 47Emerson Foulke, ed., and Michael Potters, Center for Rate-Controlled Recordings Newsletter, January 15, 1972. 44 having high listening skills and those with low listening skills as measured by the STEP.48 Rate-altered recordings have been used as an in- structional medium in elementary schools on a limited basis. Goldhaber reports, ". . . evidence that perhaps we may be able to teach via compressed speech in the early grades of 49 Furthermore, both elementary and secondary education." Woodcock and Clark demonstrated that listening to compressed speech can be an effective learning medium for elementary school children.50 It appears that altering the rate of the presenta- tion gives the educator some degree of flexibility in accom— modating individual differences in the classroom. Considera- tion of the difficulty of content is an important factor. Although difficulty level of the material was not found to be significant by George51 in his investigation, it never- theless plays a role in comprehension. Taylor, for example, 48Roland Callaway, Gerald Gleason, and Barbara Klaeser, "The Relationship of Listening Skills to the Utili- zation of Compressed Speech," Proceedings of the Second Louisville Conference on Rate and/or Frequency—Controlled Speech, Louisville, Kentucky, 1969, pp. 332-337. 49Gerald M. Goldhaber, "Listener Comprehension of Compressed Speech as a Function of the Academic Grade Level of the Subjects," Journal of Communication XX (June, 1970): 167. 50R. M. Woodcock and Charlotte Clark, op. cit., 1968, pp. 259—271. 51Robert Glen George, "Retention of Prose Material as a Function of Rate of Presentation and Difficulty of Material," AV Communication Review XVIII (Fall, 1970): 157. 45 reports that, "Considering that the average listener is ex- posed daily to speaking rates ranging from 135 to 175 words— per-minute, it appears that the listening mechanism readily adjusts to variation in rate of input, especially when the content is at or below the academic level of the listener. It is also quite probable that for short periods of time, considerably higher rates of presentation can be tolerated without a significant loss of retention, as long as the con- tent is within the usual comprehension range of the lis- tener."52 Rate-altered recordings have the potential for indi- vidualizing instruction especially when matched with learn- ers demonstrating strengths in auditory reception. Foulke reports that equipment will soon be available allowing a 53 This device learner to manage word rates for himself. will take the form of an attachment to a standard cassette tape recorder. A financially affordable, technological in- novation will then be ready for classroom applications. The rate of presentation will not be determined by a fixed out- side source, but by the learner's own listening rate pref- erence . 52Stanford E. Taylor, op. cit., p. 12. 53Emerson Foulke, ed., Proceedings of the Second Louisville Conference on Rate and/or Frequency-Controlled Speech, Louisville, Kentucky, October 22-24, 1969, p. 2. 46 David Orr summarizes the future of this audio tech- nology: The applied dimension lies in the realm of educa— tion. Today's pressures on education, created by the burgeoning knowledge and culture to be transmitted to the next generation, demand an efficient educational process. Each individual student likewise has a greater need for appropriate education to take a useful place in an increasingly complex and technical society. It is clear that auditory educational methods are assum- ing a larger and larger role in our educational process since some children learn better auditorially; since the use of audiovisuals is growing; and since the new educational technologies such as computer-assisted in- struction, dial—access tape lectures, tele-lectures, etc., involve auditory presentations.54 Review of the Illinois Test of Psycholinguistic Abilities The ITPA was conceived as a diagnostic rather than a classificatory tool. Its object is to delineate specific abilities and disabilities in children in order that remedia- tion may be undertaken when needed.55 Based upon the communication model of Osgood,56 the ITPA was first published in 1961 and revised to include twelve sub-tests in 1968. The revised edition was employed, in part, for this study. 54David Orr, "A Perspective," Journal of Communica- tion 18 (September, 1968): 291. 55 Samuel A. Kirk, et. al., pp. cit., p. 5. 56Charles E. Osgood, "A Behavioristic Analysis," Contemporary Approaches to Cognition (Cambridge, Mass.: Harvard University Press, 1957). 47 The present model upon which the ITPA is built in- cludes three levels of cognitive abilities: (1) Channels of communication; (2) Psycholinguistic processes; and (3) Levels of organization. These dimensions are described by Krik, 1. et. al., as follows:57 Channels of communication-—These are the routes through which the content of communication flows. Included here are the modalities through which sense impressions are received. Psycholinguistic processes--Three main processes are considered: (a) the receptive process, that is, that ability necessary to recognize and/or understand what is seen or heard; (b) the ex- pressive process, that is those skills necessary to express ideas or to respond either vocally or by gesture or movement; (c) an organizing proc- ess which involves the internal manipulation of percepts, concepts, and linguistic symbols. Levels of organization--The degree to which habits of communication are organized within the indi-' vidual determines the level of functioning. Two levels are postulated in the clinical model of the ITPA: (a) the representational level, which requires the more complex mediating process of utilizing symbols which carry the meaning of an object; (b) the automatic level, in which the individual's habits of functioning are less voluntary but highly organized and integrated. The model described has been used to generate twelve discrete tests for the purpose of assessing specific abili— ties and disabilities in children from the ages of two to eleven. 57 Kirk, et. al. pp. cit., p. 7. 48 Descriptions of the Tests Four tests of the ITPA were used in this study; they were: 1. Auditory reception-—This is a test to assess the ability of a child to derive meaning from verbally presented material. Since the receptive rather than the expressive process is being sampled, the response throughout is kept at the simple level of a "yes" or "no" or even a nod or shake of the head. The test contains 50 short, direct items, such as "Do dogs eat?", "Do dials yawn?", "Do carpenters kneel?". Visual reception--This test is comparable to the Auditory Reception Test, but uitlizes a different sense modality. It is a measure of the child's ability to gain meaning from visual symbols. In this test there are 40 items, each consisting of a stimulus picture on one page, and four response pictures on a second page. The child is shown the stimulus picture for three seconds with the di- rections, "See this?" Then the page of response pictures is presented with the directions, "Find one here." Auditory Closure--This is basically a test of the organizing process at the automatic level. It assesses the child's ability to fill in missing parts which were deleted in auditory presentation and to produce a complete word. Auditory closure is an automatic function which occurs in everyday life in situations such as understanding foreign accents, speech defects, or poor telephone connec- tions. In this test the the child is asked, "What am I talking about-~bo/1e? tele/one?" There are 30 items ranging in difficulty from easy words such as "airpla/‘to more difficult ones such as "ta/le/oon" and /ype/iter." Visual closure--This test assesses the child's ability to identify a common object from an incom- plete visual presentation. There are four scenes, presented separately, each containing 14 or 15 examples of a specified object. The objects are seen in varying degrees of concealment. The child is asked to see how quickly he can point to all 49 examples of a particular object with the time limit of 30 seconds for each scene.58 Research with the ITPA The ITPA was used in this research as a tool to directly assign subjects to groups based upon strengths and weaknesses diagnosed in communication channels. Children demonstrating auditory strengths were assigend to an audi- tory group and children demonstrating visual strengths were assigned to a visual group. It is important, therefore, to provide evidence that the ITPA is able to differentiate in the communication channel dimension. Burns and Watson recently factor analyzed the re- vised ITPA with underachieving children. Their subjects demonstrated marked academic learning problems not associa- ted with mental retardation or sensory impairment, as did the subjects in this research. They concluded: There is definite evidence for the existence of two channels of communication, auditory-vocal and visual-motor; and such a concept as this would have practical implications for both individual program- ming and remediation as well as for regular classroom instructional procedures.59 A few studies have used the ITPA to identify percep- tual strengths and weakness as the basis for grouping and 58Descriptions of all ITPA subtests taken from Samuel A. Kirk, et. al., op. cit., pp. 9-12. 59Gary W. Burns and Billy L. Watson, "Factor Analy- sis of the Revised ITPA with Underachieving Children," Journal of Learning Disabilities 6, 6 (June/July, 1973): 41. 50 subsequent experimentation. Bateman,6O divided subjects into auditory and visual learners after administering the experimental edition of the ITPA. An auditory method and a visual method were used with half of the students in each modality group. She concluded that the auditory method was significantly better for all groups. Her study, however, was met with some degree of criticism. The basis for group— ing was determined by Psycholinguistic Age (PLA) scores rather than standard scores. The standard-score profile allows for meaningful comparisons across subscales; the PLA profile does not permit an interpretation of compara- tive strengths and weaknesses across subscales. Compari- sons across subtests in a profile are meaningful only when the different subscales have equal variances. The standard score meets this criterion, but the PLA score does not. Waugh tested 166 second graders with the Auditory Reception, Auditory Association, Visual Reception, and Visual Association tests of the ITPA. Children were grouped as auditory or visual learners based upon two levels of dis- crepancies represented by their scores. The more stringent criteria for grouping required that the mean score of the auditory tests be at least 12 standard score points above the mean of the two visual tests, (two standard deviations), 60Barbara D. Bateman, "The Efficacy of an Auditory and a Visual Method of First Grade Reading Instruction with Auditory and Visual Learners," Perception and Reading (Newark, N. J.: International Reading Association, 1968). 51 in order for a child to be classified as an auditory learner. Less stringent criteria required a discrepancy of at least 6 standard score points. Approximately one-third (n = 54) of the population sampled were classifed as either auditory or visual leaners through this process. Visual and auditory instructional treatments were counterbalanced in the search for disordinal interaction. No disordinal interaction was found. There were, however, significant differences between groups at both levels of stringency for the auditory recall tasks. These differences demonstrated that auditory learn- ers, as classifed by Waugh, did perform significantly better than did visual learners on an auditory recall task.61 Citing the usefulness of the ITPA as a tool for diag— nosing perceptual and linguistic assets and deficits, de Hirsch, et. al., state, "We feel that exploration of modal- ity strength and weakness is cmf more than theoretical in- terest and should largely determine teaching methods.62 They conclude, "In our opinion, therefore, one method of teaching cannot be favored over another as a matter or prin- ciple. (Most discussions of the subject seem to miss this point.) Approaches to teaching should depend on the 61 R. P. Waugh, "Relationship Between Modality Pref- erence and Performance," Exceptional Children 39 (March, 1973): 465-469. 62Katrina de Hirsch, Jeannette Jefferson Jansky, and William S. Langford, Predicting Reading Failure: A Preliminary Study (New York: Harper & Row, Publishers, 1966), PP. 82-83. 52 individual child's strengths and weaknesses in the different modalities." Although the ITPA suffers from the criticisms and cautions accompanying most standardized tests, it seems to accomplish its primary objectives. Its underlying theoreti- cal framework has its basis in the diagnostic-prescriptive teaching model. The ITPA is intended for use in identify- ing strengths and deficits extant in children. Summary While children in the primary grades spend a major- ity of their classroom time listening, there seems to be a lack of research on the systematic application of auditory channel instruction in the classroom. It appears that listening and reading skills are strongly correlated. Many authorities, however, do not dis- count the notiontflun:a child may have reading deficits and concurrent listening strengths. Rate-altered recordings have been successfully ap- plied to the primary grades with normal and exceptional children. The simple regulation of the rate of presenta- tion has been used to individualize instruction. These recordings have been slbwed for those needing more process- ing time, and accelerated for those needing less. Reading comprehension seems to be strongly related to the ability to comprehend compressed speech. Future investigations in 53 the area of rate-altered speech seem to have greatest po- tential when applied to classroom settings. Many educators today emphasize advantaged instruc- tional programs; that is, the assessment of a child's par- ticular aptitudes followed by a matching of the mode of instruction to those aptitudes. Few educational technolo- gists have been conducting research matching the properties of various media forms to learner characteristics. Con- currently, scholars indicate the great potential contribu- tions of such forms of research. The ITPA provides a diagnostic tool for identifying strengths and weaknesses in children. While it is not clear that the ten single abilities that the test hopes to identify are in fact separately measured, it does seem to measure communication channel differences. The literature seems to indicate the need for apply- ing existing educational technologies to diagnosed learner strengths. CHAPTER III DESIGN OF THE STUDY The design of the study is presented in this chap- ter. This includes: (a) a description of the population and sampling procedures; (b) a description of the measures used to gather data from the sample; (c) the design matrix; (d) a statement of testable hypotheses; (e) analysis tech- niques used; and (f) a summary of the procedures. The Population The population under study in the research is gener- ally termed "learning disabled." That is to say, a discrep- ancy exists between expected and actual academic achievement for these pupils not due to physical handicaps nor mental retardation. Generally, the population fits this description by Bateman: . . . children who have learning disorders are those who manifest an educationally significant dis- crepancy between their estimated intellectual poten- tial and actual level of performance related to basic disorders in the learning processes, which may or may not be accompanied by demonstrable central nervous system dysfunction, and which are not secondary to generalized mental retardation, educational or cultural deprivation, severe emotional disturbance, or sensory 54 55 loss. Frequently these learning disorders seem to fit into one or more of three broad types--reading problems, visual-motor disturbance, and verbal com- munication disorders . . . .1 It should be emphasized that this population is not retarded, disturbed, or handicapped intflmegeneralized sense. These children exhibit a normal range of intelligence. Dis- crepancies exist, however, between potential and performance. More specifically, one of the three broad types iden- tified by Bateman--children with reading problems--is of prime importance in this investigation. A segment of this reading deficient population at- tends classes in a specialized learning center located in a large suburban school district. This school district is adjacent to a midwestern, urban, metropolitan area and draws from a broad range of socio-economic backgrounds and racial groups. The learning center provides a remedial Language Arts program for approximately 100 elementary school stu- dents, divided evenly in half—day sessions. When not at- tending the center, students are mainstreamed into regular classrooms within the district. The center is staffed by four professionals with training in the field of Learning Disabilities. Children 1Barbara Bateman, "An Educator's View of a Diag-' nostic Approach to Learning Disabilities," Learning Dis- orders, Volume 1 (Seattle, Washington: Speical Child Publications, 1965), p. 220. 56 attend small classes, usually no more than twelve to fifteen per class. room. A para-professional is assigned to each class- Criteria for admission to the learning center in- clude the following: 1. The pupil should be two or more years behind his classmates in reading level according to a recognized standardized test, such as the Stanford Achievement Test or the Gates-Mac- Ginitie Diagnostic Reading Survey. A child who belongs in a special education class is not suitable for the learning center, although children with less serious physical, psychological, or social difficulties are readily accepted. Children who are seeing several other special- ists, such as speech therapists and social workers, or have a disjointed school day which would preclude success at the learning center should not be assigned to the learning center. Taking the Sample The sample for this investigation was taken from that population of elementary children attending classes at the learning center. The center was chosen as the ex- perimental site for two reasons: 1. The center is one of only two intfluastate spe— cializing in remediating reading deficits. Thus, access to a relatively large and relatively homogeneous group was available. This specific research was undertaken as part of a larger, ongoing investigation at the center. Thus, the cooperation of parents, 2 From a directive of the Office of Student Services, Waverly Public Schools, October 8, 1973. 57 pupils, staff, and administratorsmwas,more_I readily gained. The sample for this study was taken from the first, second, and third academic grade levels in the center. Sub- jects ranged in age from seven through ten years. The fourth grade level was omitted for two reasons: 1. Chronological ages of the fourth level students were above the norms of the testing instrument used to de- termine sample groups. 2. The academic grade level of students served primarily as a blocking variable in this study. That is to say, division of the sample by grade level was used to in- crease precision through an accommodation of developmental differences in subjects by grade level. An adequate number of subjects were found in grades one through three. An initial examination of health records for these primary students assured that those subjects who might com- prise the sample possessed auditory and visual faculties within a normal range. This screening procedure left 78 primary pupils for application of the sampling instrument. Subsequently, four subtests of the Illinois Test of Psycholinguistic Abilities (ITPA) were administered in- dividually to determine sample subsets. These were: Audi- tory reception, Auditory Closure, Visual Reception, and Visual Closure. Administrators of the subtests were trained by a qualified, experienced examiner before the testing phase was begun. 58 The ITPA was standardized on a sample of 962 normal children. Test-retest coefficients of reliability for four, six, and eight year old children are provided by the au- thors. Table 3.1, below, provides reliability data for the eight year age group. TABLE 3.l.--Five-Month Stability Coefficients for the ITPA Scores for Eight Year Old Children.a Auditory Reception .63 Auditory Closure .71 Visual Reception .66 Visual Closure .82 aJ. N. ParaskeVOpoulos and S. A. Kirk, The Develop— ments and Psychometric Characteristics of the Revised 11- linois Test of Psycholinguistic Abilities (Urbana, Illinois: University of Illinois Press, 1969), p. 108. The stability coefficients given in Table 3.1 were based on data from eight year old subjects in the standard— ization sample. These age group norms, of the three given by the test's authors, are the most representative of the ages of sample subjects in this research. The four subtests were selected from the composite ITPA (12 subtests) to differentiate intraindividual abil- ities in the auditory and visual communication channels. It was important, then, that the obtained difference scores between subtests for the same individual be reliable. Median 59 reliabilities for difference scores between the subtests used across all age levels are presented in Table 3.2,below. TABLE 3.2.--Median Internal Consistency Coefficients for Difference Scores among Four ITPA Subtests.a Visual Auditory Visual Com osite Reception Closure Closure p Auditory Reception .88 .84 .79 .84 Visual Reception .82 .74 .82 Auditory Closure .73 .82 Visual Closure .68 aJ. N. Paraskevopoulos and S. A. Kirk, The Develop- ments and Psychometric Characteristics of the Revised Il- linois Test of Psycholinguistic Abilities (Urbana, Illinois: University of Illinois Press, 1969), p. 11. The subtests selected were representative of the composite ITPA. Two channels of communication were tapped --auditory and visual. Both the representational and auto- matic levels ofcuganizingcxmmmnication by the individual were measured by these tests. Finally, the means and standard deviations of these four subtests, as with all sub- tests of the composite ITPA, are equal; 36anui6, respectively. A standard (scaled) score profile in the test manual was used to provide meaningful comparisons across subtests. 60 Subjects were directly assigned to either auditory or visual groups on the basis of test results. Either one of two criteria were applied in making this assignment. First, if a subject's mean score on the two auditory sub- tests was at least six standard (scaled) score points above his mean score on the two visual subtests, he was assigned to an auditory group. For example, Student X receives the following scaled scores on the four subtests: Auditory Reception—-43; Auditory Closure--46; Visual Recep- tion--38; and Visual Closure--36. These scaled scores yield an auditory mean of 44.5 and a visual mean score of 37.0, with a difference of 7.5. Thus, according to Cri- terion one, this subject was assigned to an auditory group. Visual groups were built conversely. A similar criterion was used by Waugh, in 1973.3 Secondly, differences between the mean scaled score of all four subtests and the scaled score of any particular subtest constituted a discrepancy if the magnitude of that difference was seven or more points. To illustrate, Stu- dent Y receives the following scores on the four subtests: Auditory Reception--27; Visual Reception-~25; Auditory Clos- ure--27; and visual Closure-~40. The mean of all scaled scores, in this case is 29.75. The difference between this 3R. P. Waugh, "Relationship Between Modality Prefer— ence and Performance," Exceptional Children 39 (March, 1973): 465—469. 61 mean of all scaled scores and the scaled score for the Visual Closure subtest is 10.25, indicating a visual strength. In this instance, Student Y is assigned to a visual group. Auditory groups were built in a conversely similar manner. This criterion is recommended for deter- mining discrepancies in psycholinguistic functions by the test's authors. Some subjects were assigned to groups in an opposite direction to measured deficits. For example, if a subject's mean scaled score were 35, and a visual subtest score of 24 were recorded, the subject would be assigned to an auditory group. Strengths, in this case, are assumed to be those abilities that are not diagnosed as deficits.5 The ITPA subtests were administered to 78 students, one at a time, for approximately 30 minutes each. Those subjects not exhibiting discrepancies according to the afore- mentioned criteria were rejected from sample inclusion. Next, the remaining subjects were assigned to either an auditory or visual group, depending upon strengths meas- ured. Each group was then broken into grade levels; grades one, two, and three, respectively. Finally, an equal num- ber of subjects at each grade level was obtained by dis- carding those subjects with the lowest intraindividual 4S. A. Kirk, et. al., Examiner's Manual, ITPA, 1968, 5ITPA Scores for the sample appear in Appendix A. 62 discrepancies. To illustrate, if according to Criterion one the mean score difference between auditory and visual subtests were to be six scaled score points or more and to ensure equal cell size one subject needed to be discarded, the subject in that cell with the lowest mean difference score was dropped. The sampling process identified fifteen auditory and fifteen visual subjects for a total sample size of thirty. The auditory and visual groups were further di- vided into grade levels (one, two, and three) with five subjects at each level. Design This research was of a quasi-experimental design. Multiple measures were administered. Each was followed by a post-test. This appears graphically in Figure 3.1, below. AUDITORY X101X202X303 VISUAL X101X202X303 Figure 3.1.--Procedural Design Over Time. The three X's represent exposure to the experimental varia- ble, rate of presentation. The 0's are post-tests. The 6Donald T. Campbell and Julian C. Stanley, Experi- mental and Quasi-Experimental Designs for Research (Chicago: Rand McNally & Co., 1963). 63 dashed line indicates that the two experimental groups, auditory and visual, were not randomly assigned. Subjects were directly assigned from ITPA results as previously described. Since subjects were not randomly assigned to groups, their selection was considered as a threat to in— ternal validity. Therefore, an Analysis of Covariance was used to accommodate initial differences between groups. The covariate was reading comprehension, measured by the Gates-MacGinite Reading Tests Primary Forms A through C. Reading comprehension was found to be a significant subject variable in the comprehension of compressed speech by Mullaly,in 1972.7 That is, those subjects with higher reading achievement level scores were significantly more able to comprehend compressed speech. It was controlled for in this research to better equate the groups. The Gates—MacGinite Reading Tests had been admin- istered district-wide as part of a state assessment pro- gram while this study was underway. Thus, scores were readily available from subjects' files. Standard scores of reading comprehension were used as the covariate. The n Comprehension Test, according to the authors, . . . measures the student's ability to read complete prose 7Lee J. Mullaly. "Comprehension of a Narrative Passage by Primary School Children as a Function of Lis- tening Rate and Reading Comprehension Level? (Ph.D. dis- sertation, Michigan State University, 1972). 64 passages with understanding."8 Alternate form reliabilities for each form of the test administered appear in Table 3.3 below.9’lo TABLE 3.3.--Gates-MacGinite Reading Comprehension Reliabili- ties. Primary A .83 Primary B .81 Primary C .87 A concern for external validity arose from what might be termed multiple-treatment interference. This may occur if the effects of an earlier treatment are still present when the subject encounters a subsequent treatment. Often called carry-over effects, they were controlled by systematically ordering the presentation of the recorded listening passa- ges. Table 3.4 illustrates on the following page. In this manner, three listening passages were pre- sented to subjects; and, each was prerecorded at an expanded, a normal, and a compressed rate. When presented according 8Arthur I. Gates and Walter H. MacGinite, Teacher's Manual, Gates—MacGinite Reading Tests (New York: Teacher's College Press, Columbia University, 1965), p. 1. 9 Ibid., Technical Manual, p. 8. lOGates—MacGinite Reading Comprehension Scores for the sample appear in Appendix B. 65 TABLE 3.4.--Orders of Presentation. Expanded Normal Compressed Tape Tape Tape 1 2 3 l 2 3 l 2 3 A B C C A B B C A F E D E D F D F E to the schedule in Table 3.4, no subject in a cell was given the same passage at the same rate. Design Matrix The design matrix took the form ofaatwo-way, fully crossed design having a single repeated measure. Equal num- bers of observations were made in all cells. The design variables were Type of Learner and Academic Grade Level. Academic Grade Level was used primarily as a blocking varia- ble to increase precision. The repeated measure variable was Word Rate. The dependent variable was Listening Com- prehension measured by a modified cloze test following each exposure to the rate—altered listening passages. The design matrix appears in Figure 3.2 on the following page. Two analysis techniques were used to examine the data. An Analysis of Covariance was employed to investi- gate the effects of the design variables--Type of Learner and Academic Grade Level. Reading comprehension served as Type of Learner 66 WORD RATE Grade Expanded . Compressed Level 30% Normal 30% s1 lst . S . l 3 51+1 0 ;fi 2nd o 53 . 521 S2i+1 3rd 531 S3i+l lst S41 S4i+l H E 2nd U) 0 H > 551 S51+1 3rd . S61 Figure 3.2. -—Design Matris over Variables. 67 the covariate to accommodate initial differences between groups. To examine the effects of the repeated measure variable, Word Rate, an Analysis of Variance was employed. This analysis technique is suggested by Winer for examin- ing the effects of a single repeated measure.11 Hypotheses The ability of learning disabled, primary school children to comprehend rate-altered instruction was meas- ured by a modified cloze test. The following was hypothe- sized: Type of Learner Main Effect Null Hypothesis: NO difference will be found in comprehension of rate-altered instruction as measured by mean modified cloze test perform- ance between auditory and visual groups. Alternate Hypothesis: The mean score of the audi- tory group will exceed that of the visual group in comprehension of rate-altered in- struction as measured by a modified cloze test. Grade Level Main Effect Null Hypothesis: No difference will be found in comprehension of rate—altered instruction as measured by mean modified cloze test perform- ance between first, second, and third academic grade levels. 11B. J. Winer, Statistical Principles in Experi- mental Design (New York: McGraw-Hill Book Co., Inc., 1962), p. 614. 68 Alternate Hypothesis: Third grade mean scores will exceed second grade mean scores, which in turn will exceed first grade mean scores in compre- hension of rate-altered instruction as measured by a modified Cloze test. Word Rate Measure Main Effect Null Hypothesis: No difference will be found in comprehension of rate-altered instruction as measured by mean modified cloze test perform- ance between expanded, normal, and compressed word rates. Alternate Hypothesis: The expanded rate mean score will exceed the normal rate mean score, which in turn will exceed the compressed rate mean score in comprehension of rate-altered instruc- tion as measured by a modified Cloze test. Interaction Null Hypothesis: There will be no Type of Learner by Word Rate Measure interaction. Alternate Hypothesis: There will be a Type of Learner by Word Rate Measure Interaction. Stimulus Material Three listening passages of approximately 100 words each were selected for use in the study. Passage One was taken, by permission, from the Reading Progress Scale by Ronald P. Carver.12 Passages Two and Three were excerpted 12Ronald P. Carver, Manual for the Readipg Progress Scale (Silver Spring, Maryland: Revrac Publications, 1971). 69 from a story in basal reader.l3 An experienced announcer recorded the passages and a set of introductory statements for presentation to the subjects. All recording and dupli- cating was done with professional quality equipment in the Great Lakes Region Special Education Instructional Materials Center at Michigan State University. A rate of approximately 125 words per minute was chosen as the normal rate for the passages. A11 expansion and compression was done from this base rate. The nature of the pOpulation influenced the selection of this base rate; that is, learning disabled primary school children. Additionally, Carver and others used a similar normal rate in recent compressed speech research.l4 Each original recording was expanded and compressed approximately 30% with a Lexicon Varispeech I and duplicated in cassette form. This process yielded nine cassette tapes: three different passages at the normal rate; an expanded version of each passage; and a compressed version of each passage. Demographic data for each listening passage appears in Table 3.5 on the following page. Original recorded rates l3Patricia K. Miller and Iran L. Selegman, "Kangar— oos," in Sounds of Laughter by Bill Martin, Jr. (New York: Holt, Rinehart & Winston, Inc., 1966), pp. 107-120. 14Ronald P. Carver, Raymond L. Johnson, and Herbert L. Friedman, "Factor Analysis of the Ability to Comprehend Time-Compressed Speech," Journal of Reading Behavior 4, 1 (Winter, 1971—72): 40-49. are shown, sion and expansion. 70 TABLE 3.5.--Listening Passage Rates. as well as those for an approximate 30% compres- N E . Total Tiizal Normal T:p:nded Expanded gizzr Compr. Words (Sec.) WPM (Sec.) WPM (Sec.) WPM Passage One 102 49.5 123.6 64.0 95.6 35.3 173.4 Passage Two 104 50.5 123.6 65.1 95.8 35.9 173.8 Passage Three 100 48.9 122.7 63.1 95.1 34.7 172.9 Comprehension Measures Prior to its use, each listening passage was scaled for readability by the Fry Readability technique.15 Each passage selected was found to be appropriate for a primary audience--approximately at the third grade level. This procedure was used to provide continuity in the level of difficulty across all three passages. Although developed originally as a measure designed to evaluate reading, reada- bility formulas can be applied to listening passages as well. Sticht reports: With regard to the difficulty level oftfluamaterial, the reading and listening performance of both groups 15Edward B. Fry, Time," The Journal of Reading 11 "A Readability Formula that Saves (April, 1968): 577. 71 declined as the difficulty of the material was in- creased. Thus, the readability formula appears to have been appropriate for scaling "listenability" also. Subsequently, a modified cloze test, termed reading- input, was developed for each passage. These reading-input measures were developed through a standardized algorithmic procedure created by Carver.l7 The purpose of the resulting comprehension instruments was to measure the ability of a subject to recall deleted portions of a listening passage.18 Three reading—input measures, each consisting of twenty items, were produced. These were hand scored using a stand- ard correction for guessing--rights minus wrongs. Reliability coefficients were calculated for the measures developed with the Kuder—Richardson 21 formula. These are given in Table 3.6 on the following page. Experimental Procedures The repeated measure (a rate—altered listening passage followed by a comprehension measure) was adminis- tered on a one-to-one basis, at approximately 30 minutes per subject, in a listening carrel within a conference 16Thomas G. Sticht, "Learning by Listening," op. cit., 1971, p. 5. 17Ronald P. Carver, "Revised Procedures for Develop- ing Reading—Input Materials and Reading—Storage Tests, Wash- ington, D.C.: American Institutes for Research, October 12, 1973 (manuscript). 8Comprehension measures appear in Appendix D. 72 TABLE 3.6.--Comprehension Test Reliabilities. Passage Grade One Two Three One 0.81 0.87 0.87 Two 0.81 0.92 0.92 Three 0.81 0.95 0.92 room in the learning center. Each subject received an ini- tial set of introductory statements followed by the listen- ing experience and an appropriate comprehension measure. The introduction and listening passages were presented by cassette tape through a Wollensak 2550 unit. AKG K-180 stereophonic headphones adapted to monaural were used by each subject. A predetermined schedule of presentations (See Table 3.4) assured that each subject received a different passage at the expanded, normal, and compressed rates. This dictated a total Of nine separate tapes of the origi- nal three passages. A tenth tape, at a normal rate, in- troduced the task to the subjects. This introductory tape explained the task, directed the adjustment of the volume of playback, and cautioned subjects to listen care- fully. Then, the experimenter told the subjects, "We are going to listen to three short stories. After you listen 73 to a story the first time, I'm going to read it to you again, but I'll leave some of the words out. You help me to fill in the missing words. Let's try this one for practice." The experimenter than read the following sentence. Charles asked his mother for a nickel to buy an ice cream cone, but she said he could not have it.19 The experimenter then said, "Now, I'm going to read that sentence to you again, but this time I'll leave some words out. Help me fill in the missing words." "Charles asked his . . . (pause) Was the next word ggg or mother?" Student responds. "For a nickel to . . . ppy_or walk?" Student responds. "An ice cream cone . . . papy or ppp?" Student responds. "She said he could . . . 293 or pf?" Student responds. "Have it." When necessary, the process was repeated until each subject understood the task. The listening passages were then presented with an oral administration of the compre- hension measure in the same manner as the demonstration. Expanded passages were presented first, followed by the normal rate, and then the compressed. Analysis Subjects' scores were hand-coded and keypunched into cards. These scores and that of the covariate of 19Ronald P. Carver, Manual for the Reading Progress Scale (Silver Springs, Maryland: Revrac Publications, 1971). Used by permission. 74 reading comprehension were analyzed via the Michigan State University CDC 6500. An Analysis of Covariance (ANCOVA) pro- gram supported by the STAT. system directed the actual analy- sis for the design variables. An Analysis of Variance (ANOVA) program directed the actual analysis for the repeated meas- ure variable. The hypotheses were tested at the .05 level of significance. Use of the Michigan State University com- puting facilities was made possible through support, in part, from the National Science Foundation. Summary This study took the form of a quasi-experimental de- sign with a single repeated measure. Subjects were selected and assigned to two groups, auditory and visual, using the Illinois Test of Psycholinguistic Abilities to discern com- munication channel strengths. An Analysis of Covariance was used to control for initial differences using reading comprehension as the covariate. Academic grade level served as a blocking variable. The hypothesis of major interest was that those sub- jects identified as auditory learners would be better able to comprehend rate-altered instruction than those identi- fied as visual learners. Each subject was presented a systematically varied sequence of expanded, normal, and compressed listening pas- sages. Comprehension was measured by a modified cloze test 75 administered orally by the experimenter. An Analysis of Covariance was used to determine the effects of the design variables upon the dependent variable, Listening Comprehen- sion. An Analysis of Variance was employed to determine the effect of the repeated measure variable, Word Rate, upon the dependent variable. Hypotheses were tested at the .05 level of confidence. Results are presented in Chapter IV. CHAPTER IV ANALYSIS OF RESULTS The results of the study are presented in this chap— ter. Testable hypotheses are restated individually and findings Obtained from the investigation are provided for each. A summary of results is provided in tabular form. A discussion follows to give a meaningful interpretation of the findings. Results Results of testing each hypothesis are presented below. Each hypothesis is stated in testable form and the findings are presented individually for each. Type of Learner Main Effect Null Hypothesis I: No difference will be found in comprehension of rate-altered instruction as measured by mean modified cloze test per— formance between auditory and visual groups. The Analysis of Covariance produced an F-ratio of 3.2733 for the Type of Learner main effect which was not significant at the .05 level of confidence. The null hy- pothesis, therefore, was not rejected. 76 77 Academic Grade Level Main Effect Null Hypothesis II: No difference will be found in comprehension of rate-altered instruction as measured by mean modified cloze test per- formance between first, second, and third academic grade levels. The Analysis of Covariance produced an F-ratio of 2.7087 for the Academic Grade Level main effect which was not significant at the .05 level of confidence. The null hypothesis, therefore, was not rejected. Word Rate Measure Main Effect Null Hypothesis III: No difference will be found in comprehension of rate-altered instruction as measured by mean modified cloze test per- formance between expanded, normal, and com- pressed word rates. An Analysis of Variance produced an F-ratio of 0.9101 for the Word Rate Measure main effect which was not significant at the .05 level of confidence. The null hy- pothesis, therefore, was not rejected. Type of Learner by Word Rate Measure Interaction Null Hypothesis IV: There will be no Type of Learner by Word Rate Measure interaction. An Analysis of Variance produced an F-ratio of 0.7588 which was not significant at the .05 level of con- fidence. The null hypothesis, therefore, was not rejected. 78 Summary of Results A summary of findings are provided in Tables 4.1 and 4.2. TABLE 4.1.--Analysis of Covariance (ANCOVA) Results for Listening Comprehension Scores. ._-» _—.. “v. ,-, . _, Source of Variation d.f. Mean Squares F—Ratio p Between Types of Learners 1 57.7391 3.2733 0.0836 Between Academic Grade Levels 2 47.7790 2.7087 0.0879 Interaction Between Type of Learner and Academic Grade Level 2 3.8625 0.2190 0.8051 Error Term 23 17.6392 N = 30 TABLE 4.2.--Analysis of Variance (ANOVA) Results for Listening Comprehension Scores. Source of Variation d.f. Mean Squares 'F—Ratio Between Word Rate Measures 2 16.0444 0.9101 Interaction Between Type of Learner and Word Rate Measure 2 13.3778 0.7588 Interaction Between Academic Grade Level and Word Rate Measure 4 17.7111 1.0044 Interaction Between Type of Learner, Word Rate Measure, and Academic Grade Level 4 10.6444 0.6037 Error Term 48 17.6333 79 All hypothesis testing was done at the .05 level of confidence. No null hypothesis was rejected at this level. Discussion The result of the Analysis of Covariance between Type of Learner groups (Auditory and Visual) was not sig- nificant at the .05 level of confidence. The probability of the obtained F-ratio was .08. An inspection of the data in Tabel 4.2, below, reveals differences in the com- prehension test mean scores according to learner type which are generally consistent with the hypothesized superiority of auditory groups to more readily comprehend rate-altered instruction. Thus, while significant differences were not found through an Analysis of Covariance, the data does sug- gest a difference not discernable in the F-test and which is not likely due to chance. TABLE 4.3.--Comprehension Test Mean Scores and Standard Deviations by Type of Learner Groups According to Academic Grade Levels. Grade Aud. Aud. Vis. Vis. Level Mean S.D. N Mean S.D. N 1 6.933 3.69 5 3.200 5.19 5 2 9.867 2.56 5 9.333 5.35 5 3 11.067 4.68 5 8.267 3.42 5 80 The generally ascending mean scores according to grade level further indicate a superiority for academically advanced students to more readily comprehend rate-altered instruction, as might be expected. Reading Comprehension and LEStening Comprehension It is well substantiated that elementary students spend more classroom time listening than reading. That, in fact, children listen mgpa than they read. A major conten- tion of this thesis, however, is that some children may listen better than they read. If that were true, auditory instruction of an advantaged nature might be provided them. An examination of the data from Table 4.4, below, provides some degree of support for the notion that some children listen better than they read. TABLE 4.4.—-Means and Standard Deviations of Listening and Reading Comprehension Measures by Type of Learner and Academic Grade Level. Auditory Group Visual Group Grade List. List. Rdg. Rdg. ' List. List. Rdg. Rdg. Level Mean S.D. Mean S.D. Mean . S.D. Mean S.D. 1 6.93 3.69 40.20 4.09 3.20 5.19 40.20 10.23 2 9-86 2.56 50.08 13.44 9.33 5.36 57.20 11.39 3 11.07 4.68 39.60 6.62 8.27 3.42 44.60 7.77 81 Subjects with identified auditory strengths had greater observed mean scores in listening comprehension of rate-altered instruction than those subjects with identified visual strengths. At the same time, however, the reading comprehension scores measured by the Gates-MacGinite read- ing tests for the auditory group were lower, on the average, than those for the visual group. Thus, while the reading comprehension abilities of the auditory group were lower than or equal to those of the visual group, their listening comprhension abilities were greater in each case. A Pearson product-moment coefficient of correlation between listening comprehension and reading comprehension was computed from the data. This revealed a rather low in- dex of 0.271 for the sample. Although listening skills and reading skills are no doubt strongly related, it appears that being a "poor" reader does not preclude being a "good" listener. Word Rate Altering the rate of presentation within a small range, approximately 90 to 175 words per minute, appeared to have little effect upon comprehension. In Table 4.5 on the following page, mean scores of learner groups at each of the rates are presented. 82 e.m e.m m.m a.os o.e m.m e.m e.m m.m o.m m.m o.ma m m.m 4.4 0.0 ~.m m.e o.m mm.m N.m H.m o.e m.e e.m H . . . . . . . . . . . . . . . . . . Hm>mu o m deoo a m Euoz o m axe o m deoo a m 5602 a m dxm mpmuo Hmsme> >uo use: .Ho>eq opmuw OHEmpmoé >b mmsomo answfl> pen mnouflps< MOw mmumm commopmeoo pan .Hchoz .pwpcmaxm um macaume>oo pumpccum can mmuoom cmozll.m.v mqmma mo. ecu up Omumoe coHuommeIcoz cofluomflwuncoz COHDomflmuIcoz cofluomhmuucoz .cofluomuwucH enamooz mucm pnoz >9 weaned mo meme 0: mb Haez mumne .mopmu OHOB commend 1800 can .HmEuoc .Oopcmmxo coosuob oocmEuomqu umou omoao Omflwepoe some >b Oousmmoa mm COADOSHumce Oououam nouns mo COAmcocondEoo ca pcsow on Haas mucouomwep oz .mam>oa opmum OHEoOmom Ouflzp can .Ocooom .umuew com3umb ooccEnomumm pump mNoHo beamepoe cmoE Sb OoHSmmoE mm :oHuosnumce Omuopam loumu mo concmcoumEoo Ce OCDOM on HHHS mocoummmep oz .mmsoum answe> can wuouflpsm somsuob mocmEHOwumm umop mnoao poflmepoe some an Oousmmoe mo coauusuumce pmumuHMIeunu mo coemcocoumeoo CH OCSOM on Haas museumwwep oz .>H .HHH .HH coHuooflmHIcoz Ho sofluommom mo pcoEmumum mononu0d>m HHDZ .muHSmmm mo wumEEDmll.m.v mqmde CHAPTER V SUMMARY AND IMPLICATIONS Summary The major purpose of this study was to investigate the applicability of advantaged auditory instruction with learning disabled primary school children. Advantaged audi- tory instruction, in essence, implies that teaching to the learner's strengths and bypassing his deficits results in effective learning. The key notion underlying such a con- cept is that instruction be designed to capitalize upon individual strengths of learners. Deficits are not remedia- ted directly. TO investigate this concept of advantaged auditory instruction, a sample of thirty learning disabled primary school children with reading problems were divided into two groups, auditory and visual, on the basis of identified com- munication channel strengths measured by the Illinois Test of Psycholinguistic Abilities. Each group listened to passages of connected discourse at an expanded, normal, and compressed rate of presentation. Each passage was followed by an oral administration of a cloze technique comprehension measure. 85 86 The Cloze test had been modified by a standardized algorith- mic procedure to provide subjects with two alternatives from which to choose in supplying deleted words. The hypothesis of major interest was that those subjects identified as audi- tory learners would be more able to comprehend rate-altered instruction than those identified as visual learners. An Analysis of Covariance was used to control fOr initial differences in order to equate the groups formed through non-random assignment. Reading comprehension was used as the covariate. Academic grade level served as a blocking variable to increase precision. The design was two-way, fully crossed, with a single repeated measure. The two design variables, Type of Learner and Academic Grade Leve1,v«nxaanalyzed with Analysis of Covariance. The repeated measure variable of Word Rate and the interaction of Type of Learner and Word Rate Measure were analyzed by an Analysis of Variance. All hypothesis testing was done at the .05 level of confidence set prior to the experiment according to traditional educational research conventions. Findings Four major hypotheses were tested in the investi- gation. None were rejected at the .05 level of confidence. The obtained probability of the F—ratio for Type of Learner main effect and Academic Grade Level main effect was .08. 87 Conclusions Based on the findings, since the null hypotheses could not be rejected, the alternative hypotheses could not be accepted. Discussion and Recommendations The findings of no significant difference were dif- ficult to accept. The importance of the concept of ad- vantaged auditory instruction as a potential tool was difficult to summarily dismiss. Retrospective analysis indicates two potential sources of error on the part of the investigator which should be pointed out to others who would follow this path. First, there are the experimental factors which could, and should be reviewed if a second study of this nature were to be performed. Second, there is the experimental design itself--particularly the alpha level set to determine accep- tance or rejection of the hypotheses. Some experimental factors such as the small sample size immediately reveal a weakness in the discriminative power of the experiment. If this study were to be repli- cated with a larger sample, perhaps there could be a sig- nificant finding, even at the .05 level of confidence. The listening comprehension test itself was, and still remains, one of the weak features in any effort to investigate ad- vantaged auditory instruction. The modified Cloze, like 88 its predecessor the standard cloze procedure, seems less sensitive to large differences in understanding than other measures. The inability of this investigation to reveal significant differences at the .05 level of confidence may be due, in fact, to this insensitivity. The alternatives to researcher developed listening comprehension measures are limited, however. Future research endeavors with audi- tory instruction would certainly benefit from the refine- ment of existing listening comprehension instruments and the development of new techniques. The Illinois Test of Psycholinguistic Abilities was used to assign subjects to sample groups. Although the composite appears to be able to differentiate communication channel abilities adequately, only four of the twelve tests were used. Furthermore, the use of more stringent criteria for designating a subject as an auditory or visual learner might have resulted in more discrete grouping. Potentially, this procedure could have resulted in revealing significant differences. Finally, the rates of presentation were kept within a narrow range (Approximately 90 to 175 w.p.m.). If an additional com- pressed rate of 225 w.p.m. or higher were to be added, differences between auditory and visual groups might have become more apparent. The selection of an alpha level of .05 in the ini- tial stages of the study is viewed as the second major source of potential investigator error. 89 The convention of setting an alpha level at .05 or lower has grown from the desire to avoid the acceptance of educational research findings without a high degree of con- fidence in their "truth." This is most commendable in cases where decisions need to be made from a set of alternatives; some of which may hold dire consequences for the educational progress of children. A medical analogy is the investiga- tion of a new drug prior to its introduction to the general public. To be sure, the consequences of deleterious side- effects need to be carefully considered and pre-introductory experimentation must be rigorous and stringent, indeed. Alpha levels, in this instance, may not be tolerable un- less they are set at very low levels. A second medical analogy, perhaps somewhat more closely related to the existing status of educational prac- tice, is the case of the terminal cancer patient. In mak- ing decisions for the treatment of terminal patients, the alternatives may be many in number. The consequences of selecting any one of the alternatives for implementation, however, will certainly not have a more deleterious effect than to allow the status quo to exist; that is, to allow the disease to run its course. In this case, the amount of error that is tolerable is most certainly increased. Children with reading problems, in effect, may be plagued with a terminal disease. To be unable to read 90 destines a child to a life of limited potential. Supple- mental and alternative means of instruction are needed in addition to the status quo if, in fact, the terminal dis- ease is to be arrested. In retrospect, the selection of the .05 level of confidence appears to have been inappropriate for this in- vestigation. The consequences of drawing implications from obtained probability levels under .10 seems to be out- weighed by the consequences of accepting current educa- tional practice in its entirety. Implications Given the context of the two sources of potential experimenter error, it is felt that the Preferential Model exemplified in the concept of advantaged auditory instruc- tion cannot be summarily dismissed. Thus, the following series of implications for future research have been drawn. These are not conclusions. They are, rather, a set of ob- servations in need of more conclusive investigation. 1. Learning disabled primary school children with identified auditory strengths generally appear to be more able to comprehend rate-altered instruction than those children with visual strengths. In this investigation, the obtained probability of the F-ratio for this Type of Learner main effect was .08. The area seems worthy of future re- search. 91 2. Past research has indicated that children in higher academic grade levels seem to be more able, generally, to comprehend rate-altered instruction than those children in lower academic grade levels. The obtained probability of the F-ratio in this investigation for the Academic Grade Level main effect was .088. 3. When the rates of auditory presentations are kept within reasonable and narrow limits (approximately 90 - 175 w.p.m.) their alteration tends to have little effect upon comprehension. If comprehension at faster rates is equal to comprehension at slower rates, then it appears that the compressed rate is preferable when learning per unit of time is a consideration. 4. Reading comprehension and listening comprehen- sion ability correlated weakly (0.27) in this investigation. In predicting the success of rate-altered instruction with learning disabled primary children, a diagnosis of communi- cation channel strength may serve as a stronger predictor variable. The area seems to be worthy of future research. 5. The Illinois Test<1fPsycholinguistic Abilities appears to be a viable tool for measuring communication channel strengths and weaknesses. It may be useful, theref fore, in predicing success in advantaged instructional treatments matched to those strengths. 92 6. Auditory instruction in general, and rate- altered instruction in particular, are valuable educational tools for children with low reading skills. Although read- ing and listening seem to demand a somewhat similar set of skills, being a "poor" reader does not appear to preclude being a "good" listener. 7. It seems important that consideration of stu- dents' strengths and preferences be made before they are programmed into a lock-step instructional sequence. If such a philosophy is embraced, it is imperative that optional learning experiences be provided learners to ensure maximum success. If existing technologies and diagnostic techni- ques can successfully match students to programs, it seems far better to adapt such programs to students, rather than students to programs. 8. Learning disabled children in general, and those with reading problems in particular, are in great need of alternative methods of receiving instruction. It is not recommended that the development of reading skills be deemphasized, but, rather that advantaged instructional methodologies be researched and developed to provide in- struction concurrent with remediational activities. APPENDICES 93 APPENDIX A SAMPLE SCORES FOR THE ITPA 94 95 APPENDIX A.--Samp1e Scores for the ITPA. _. __,, _ . ._.-.-.‘~_ ._._._ _.-._ _ ._ __ Student Grade Auditory Visual Auditory Visual Number Reception Reception Closure Closure Auditory 10 1 35 38 38 23 11 1 43 38 47 36 12 l 40 31 36 29 13 l 36 22 28 28 14 l 34 25 36 29 20 2 30 35 44 31 21 2 28 19 31 33 22 2 28 39 42 31 23 2 42 41 36 29 24 2 37 31 46 32 30 3 27 32 46 36 31 3 35 27 44 35 32 3 30 24 44 20 33 3 30 21 34 26 34 3 43 40 48 31 Visual 15 l 32 36 17 31 16 1 27 25 27 40 17 l 21 37 29 32 18 l 27 32 22 35 19 l 25 29 20 30 25 2 22 25 16 28 26 2 31 36 19 34 27 2 20 35 32 33 28 2 29 49 14 36 29 2 20 40 35 31 35 3 22 4O 34 34 36 3 26 38 24 50 37 3 25 38 25 33 38 3 23 35 27 31 39 3 15 32 01 42 APPENDIX B SAMPLE SCORES FOR GATES-MACGINITE READING COMPREHENSION PRIMARY FORMS A - C 96 97 APPENDIX B.--Gates-MacGinite Reading Comprehensidn Scores. Student Number Grade Level Reading ComprehenSion Standard Score Auditory 10 1 37 ll 1 40 12 1 40 13 l 37 14 l 47 20 2 53 21 2 59 22 2 30 23 2 64 24 2 44 30 3 47 31 3 29 32 3 41 33 3 42 34 3 39 Visual 15 l 33 16 l 44 17 l 44 18 l 53 19 l 27 25 2 39 26 2 66 27 2 53 28 2 64 29 2 64 35 3 47 36 3 47 37 3 53 38 3 32 39 3 44 APPENDIX C FRY READABILITY FORMULA 98 99 PROCEDURE FOR USING FRY'S READABILITY TECHNIQUE Choose a passage from a book or basal reader (about 100 words in length). Book, page, and paragraph Count the number of words in the selection. Words (W) Count the number of sentences in the selection. Sentences (S) Count the number of syllables in the selection. Syllables Substitute the above numbers in the following formulas: Syllables per 100 words Sentences per 100 words Plot the two numbers on level. Approximate grade level 100 (Syllables) W 100 (S) W the graph below to find approximate grade is . 100 PROCEDURE FOR USING FRY'S READABILITY TECHNIQUE (Continued). Short words Long words '08 Ill short sentences Average number of sentences per 100 words Long sentences References 1. Fry, Edward B. The Journal of Reading 11 (April, 1968): 513-516. 2. . The Reading Teacher 22 (March, 1969): 534-538. 3. Maginnis, George H. The Reading Teacher 22 (March, 1969): 516-518. READABILITY DATA Passage One Passage Two Passage Three 102 14 139 104 12 137 100 13 128 101 Words Sentences Syllables Words Sentences Syllables Words Sentences Syllables 13.72, 136 Third Level 11.5, 131.7 Third Level 13, 128 Third Level APPENDIX D CARVER READING-INPUT ALGORITHM 102 103 By following certain manual developmental proced- ures, the alternative words for a 30-item reading-input passage can be produced for a lSO-word prose passage in about 15 minutes. These manual steps are as follows: Step 1. Select the parameter, Nr' the size of the repetitive unit. For most research purposes, Nr = 5, the same as regular cloze. Step 2. Select parameter, Xi, the location of the item within each repetitive unit. Step 3. On the prose itself, mark out the correct word for each item. For example, if Nr = 5 and Xi = 2, then the second word in the prose should be crossed out, and then every fifth word thereafter should be crossed out. Step 4. Write or type the remaining words on a page in a column with the Nr - 1 words between items typed on each row. Leave space at the right hand margin of the page for the items (i.e., Na + 1 lines per item), and skip a line to designate a new paragraph. Step 5. A random number table may be consulted to mark on the work sheet which of the word positions will contain the correct word. For example, odd numbers in a sequence of random numbers can be used to designate the upper position as containing the correct answer, and even numbers can be used to designate the lower position as containing the correct answer. 104 Step 6. Copying from the original prose, write in the correct words for each item on the work sheet in the position designated in Step 5. Step 7. Select the parameter, Np’ which is number of words preceding and succeeding an item which are included as possible candidate alternatives. Select the exclusion conditions and the subsequent Nb' Nj’ & NC parameters as needed. The manual implementation of the steps is greatly facilitated when: (a) Np is a multiple of Nr’ (b) N = N b r -l, (c) and Xj = Xil For example, consider the Figure 2 example: (a) Np value, (b) Nb = 4, i.e., Nr - 1, so that the row of four words that the item is on is excluded as well as the follow- 25, and 25 is the multiple of 5, the Nr ing row of four words, (c) Xj = 3, i.e., Xj = Xi’ so that all of the correct item words in the final column of items may be excluded. Step 8. Assuming that the recommended parameter values in Step 7 are adOpted, Step 8 involves three substeps. These substeps involve the use of a series of random numbers and they will be explained in concrete terms using the Fig— ure 2 parameters as examples. Substep 8A. Enter a table of random numbers and select a one digit number. If it is an odd numbered digit, the candidate word will be in the pool of words preceding the item. If the random number is an even numbered digit, the candidate word will be in the pool of words following the item. 105 Substep 88. Inspect the subsequent one digit num— bers in consecutive order until one of the digits 2, 3, 4 or 5 is found. This digit will designate the number of rows following or preceding the item wherein the candidate alterna- tive word will be found. Row number 1 will always be ex- cluded when Nb = Nr - l. The upper limit number, 5 in this case, will always equal NP / Nr' Substep 8C. Continue to inspect the series of dig- its until one of the digits 1, 2, 3, or 4 is found. This digit will designate the order position of the candidate alternative word within the row. The above set of digits will always correspond to the number of words per row, i.e., 1 through Nr - 1. Step 9. Check the candidate word for the item against the correct word (and any previously selected al- ternative word when Na > 1) to see if it is the same word as the correct word. This is the first exclusion condition, E = 1.. If a match results, then the candidate word is re- jected, and another candidate word is chosen according to Step 8. Step 10. Select the parameter, Nc' The manual im- plementation of this step is facilitated when NC = Nr - l, i.e., when the number of words preceding and following an item which cannot include the same word as the alternative word is equal to the number of words in the same row and the following row. 106 Step 11. Assuming that the parameter value recom- mended in Step 10 is adopted, Step 11 involves a compari- son between the candidate word and each word in the same row as the item and the row following the item. If the word matches one of these words, then it is rejected and another candidate word is chosen according to Step 8. If the word does not match any of these words then this word becomes an alternative word, and it is written in on the work sheet in the proper blank. The alternative word is capitalized whenever the correct word is capitalized, and when the correct word is not capitalized then the alterna- tive word is not capitalized. The remaining alternative words are chosen accord- ing to the same procedures, i.e., according to Steps 8 and 9 using the same parameter values as were used for the first alternative word. APPENDIX E PASSAGES AND MEASURES 107 108 INTRODUCTION Hello there! Now that you have your earphones on, I will talk for a while so that you can adjust the volume, if you need to. We are going to listen to three short stories. They are yery short, so listen carefully. When each story is over, you will be asked some questions. Try hard to listen well. My voice is going to sound different for each story. So, even if it sounds strange to you, listen carefully. It's supposed to sound that way. 109 PASSAGE ONE This is our Post Office. It is in our city. Many people work here. There is a Post Office in every city in our country. And Post Offices in every country in the world. A Post Office helper must be honest. He must be a good worker. A Post Office helper handles lots of mail. A Post Office helper handles lots of money. The Post Office sends letters and packages, maga- zines, and newspapers all over the world. It sends small animals and plants, too. It sends money for use. It saves money for us. It puts money to work for us, too. 110 MEAS URE ONE This is Dcountry Post Office. It is Dthere [3 our Din . . Upost . DPost our City. Many people Dwork here. There is a DBe . . . Dis (jgood Office in every City Din our country. And Post DOffices . . Dthe in every country in El city world. A Post Office gri‘ilper must be honest. He D handles D Every . Dmust be a good worker. DA Post Office D lots . [3 animals helper handles [land of mail. A Post Doffice helper handles lots of Dand. Dmoney. The Post Office sends Dworker and packages [3 letters ’ . D helper D It magaZines, and Dnewspapers all over the world. DHelper sends small animals and (Splints, too. It sends money Dfor Usends Doffice us. It saves money Elfor us. It puts money Dto work for us too. D and 111 PASSAGE TWO Do you know where baby kangaroos live? They live in their mother's "pocket." All baby kangaroos are called joeys. The mother kangaroo carries her joey with her, in her pocket. A baby kangaroo is only one inch long when it is born. It has no fur. It cannot see. As soon as it is born, it crawls into the pocket. The baby kangaroo stays in the pocket for four :months. The pocket keeps it safe and warm. The mother xnakes milk in her body to feed the joey. If you could lcxak in the pocket, you would see how much bigger it is growing . 112 MEAS URE TWO D know DO You Eltheir where baby kangaroos live? I] With . . . , El her D They live in their mother s [3 pocket. [3 called E] her All baby kangaroos are Joeys. The mother kangaroo E] in E] fur D carries [1 is her Joey Wlth her ner pocket. [3 with U kangaroo USee C] Joey DIt has no fur. It [jcannot Clit . . Dfur soon as Clthe is born it crawls Dinto E] kangaroo C] no [I] long A baby E] the is only one inch when it is born. see. As the pocket. [lit The baby E] for stays in the pocket D keeps Dbody it safe and warm. four months. The pocket C] The . . El her ClLook mother makes milk in Elmonths Cl Joey C] much' U feed C] see body to feed [3 growing the D in If you could look the pocket, you would how much bigger it is growing. lll‘l'l‘l‘ 113 PAS SAGE THREE Kangaroos live together in small groups. A group of kangaroos is called a mob. The strongest male kangaroo is the leader of the mob. He had to fight the other male kangaroos to become the leader. He had to show that he was the strongest kangaroo. He had to show that he could take care of the mob. Kangaroos sleep during the day. At night they move about, looking for food. Kangaroos eat grass. They nibble leaves from small trees. They eat fruit and vines. A kangaroo has a long, heavy tail. It rests on its tail when it sits. 114 MEASURE THREE Kangaroos live gfiggether in small groups. A group of kangaroos is called Ellive mob. The strong- e [la Dthat . Elsmall est male Dkangaroo is the leader of Dthe mob. He had to Dfight the other male kangaroos [jis become [:1 strongest [j to Uhad Dwas the leader. He Bother to show that he Dthey the C] they [3 sleep strongest kangaroo. He Dhad to show that he Dcould Dmob take care of the E] food' . [3 he . Kangaroos sleep during the [1 day' At night they move D about looking for food Kangaroos D during D leaves ' E] eat . [1 care grass. They nibble leaves D from small trees. They eat DfrUlt and vines. Elsleep A kangaroo g;::ts a long, heavy tail. EEEOd Dthey rests on its tail it sits. Cl when ‘I‘11 (II) APPENDIX F COMPREHENSION TEST SCORES FOR THE SAMPLE 115 116 APPENDIX F.--Comprehension Test Scores for the Sample. Student Number Expanded Normal Compressed Auditory 10 2 2 12 ll l4 14 12 12 2 10 4 l3 8 2 10 14 2 2 8 20 18 10 12 21 8 12 8 22 12 4 4 23 14 10 10 24 8 4 14 3O 12 4 10 31 4 18 18 32 l6 16 14 33 0 4 8 34 10 - 14 18 Visual 15 2 2 4 l6 - 2 8 6 l7 - 2 - 4 - 2 18 14 12 8 l9 - 2 - 2 6 25 2 4 2 26 8 4 2 27 l4 14 12 28 18 16 10 29 4 14 16 35 12 4 10 36 10 10 4 37 2 18 14 38 12 10 10 39 4 0 4 BIBLIOGRAPHY 117 BIBLIOGRAPHY Anderson, Harold A. "Needed Research in Listening." Ele- mentary Education English Journal 29 (April, 1952): 215-224. Armstrong, Hubert Coslet. "The Relationship of the Audi- tory and Visual Vocabularies of Children." Ph.D. dissertation, Stanford Univeristy, 1953. Bateman, Barbara. "An Educator's View of a Diagnostic Ap- proach to Learning Disabilities." LearningfiDis- orders Volume 1. Seattle: Special Child Publica- tions, 1965. Bateman, Barbara. "Implications of a Learning-Disability Approach for Teaching Educable Retardates." Mental Retardation 5 (June, 1967): 23-25. Bateman, Barbara. "The Efficacy of an Auditory and a Vis- ual Method of First Grade Reading Instruction with Auditory and Visual Learners." Perception and Reading. Newark: The International Reading As- sociation, 1968. Beaird, James H. "Learner Variables and the Instructional Technologist." The Contributions of Behavioral Science to Instructional Technology. Oregon: Teaching Research, A Division of the Oregon State System of Higher Education. III-3. Berman, Mark L. "The Effects of the Magnitude of Response Cost and Reinforcement on Speed and Accuracy on Programmed Materials." Dissertation Abstracts, 30/02, 1969, 859. ‘ Bond, Guy L. and Tinker, Miles A. Reading Difficulties: Their Diagnosis and Correction. New York: Apple- ton Century-Crofts, Inc., 1967. Burns, Gary W., and Billy L. Watson. "Factor Analysis of the Revised ITPA with Underachieving Children." Journal of Learning Disabilities 6, 6 (June/July, 1973): 36—41. 118 |.l |.lll|lllll.l|| III I!) 119 Buros, Oscar K., ed. The Seventh Mental Measurements Year- book, Volume II. Highland Park: The Gryphon Press, 1972. California, Department of Education, "Exploiting the Oppor- tunity to Read by Listening." Learning Through Listening by Emerson Foulke. 1973. Callaway, Roland, Gerald Gleason, and Barbara Klaeser. "The Relationship of Listening Skills to the Utilization of Compressed Speech." Proceedings of the Second Louisville Conference on Rate and/or Frquency-Con- trolled Speech. Louisville: 1969. Campbell, Donald T. and Julian C. Stanley. Experimental and Quasi-Experimental Designs for Research. Chicago: Rand McNally & Co., 1963. Carrol, John and Roy 0. Freedle, eds. Language Comprehen- sion and the Agguisition of Knowledge. Washington: V. H. Winton. Distributed by the Halsted Press Division of Wiley, New York, 1972. Carver, Ronald P. Procedures for Constructing a Varietygof Information-Processing Measures Appropriate for Prose Materials. Manual for the Reading Progress Scale. Silver Spring: Revrac Publications, 1971. . Raymond L. Johnson, and Herbert L. Friedman. "Factor Analysis of the Ability to Comprehend Time- Compressed Speech." Journal of Reading Behavior 4, 1 (Winter, 1971/72): 40—49. "Understanding, Information-processing, and Learning from Prose Materials." Journal of Educa- tional Psychology 64 (1973): 76-84. . "Revised Procedures for Developing Reading-In- put Materials and Reading-Storage Tests." Washing- ton: American Institutes for Research, October 12, 1973 (manuscript). Day, Willard F. and Barbara R. Beach. "A Survey of the Re- search Literature Comparing the Visual and Auditory Presentation of Information." Air Force Technical Report No. 5921, November, 1950. 120 De Hirsch, Katrina, Jeannette Jefferson Jansky, and William S. Langford. Predicting Reading Failure: A Pre- liminary Study. New York: Harper & Row, Publish- ers, 1966. Deutsch, Martin and others. Communication of Information intjszlementary School Classroom. Cooperative Re- search Project No. 908. New York Medical College: Institute for Developmental Studies, 1964. Devine, Thomas G. "Reading and Listening: New Research Findings." Listening Readings, Volume 2. Metuchen: The Scarecrow Press, Inc., 1971. Dickens, Milton and Frederick Williams. "Experimental Ap- plication of 'Cloze' Procedures and Attitude Meas- ures to Listening Comprehension." Speech Mono- graphs 31 (June, 1964): 103-108. Duker, Sam. Listening: Readings. New York: The Scarecrow Press, Inc., 1966. . Listening Bibliography, Second Edition. Metuchen: The Scarecrow Press, Inc., 1968. . Listening: Readings, Volume 2. Metuchen: The Scarecrow Press, Inc., 1971. Durrell, Donald D. "Listening Comprehension versus Reading Comprehension." Journal of Reading 12, 6 (March, 1969): 455-460. Edling, Jack V. and C. F. Paulson. "Understanding Instruc- tional Media." The Contribution of Behavioral Science to Instructional Technology. Oregon: Teaching Research, A Division of the Oregon State System of Higher Education, IV-l: 17-18. Emslie, Elizabeth and Others. "A Comparison of Achievement in Silent Reading and Listening in Grade Four." In Listening Bibliography, 2nd ed., Edited by Sam Duker. Metuchen: The Scarecrow Press, Inc., 1968. Fairbanks, G., N. Guttman, and M. Miron. "Auditory Compre- hension of Repeated High-Speed Messages." Journal of Speech and Hearing Disorders 22 (1957): 20-22. Fenwick, James J. "Slow-Learners and Listening." Listening: Rggdiggs, Volume 2, edited by Sam Duker. Metuchen: The Scarecrow Press, Inc., 1971. 121 Foulke, Emerson, ed. Proceedings of the Second Louisville Conference on Rate and/or Frquency-Controlled Speech. Louisville, Kentucky, October, 1969, 22-24. , and Michael Potters. Center for Rate-Controlled Recordings Newsletter (January 15, 1972). . Center for Rate-Controlled Recordings News- letter. Vol. 7, No. 5, May 15, 1973. Foulke, Emerson. Tape-Recorded Presentation at Michigan State University, November 14, 1973. Friedman, Robert M., and others. "A Comparative Study of the Retention Level of Verbal Material Presented Visually and Orally: Fifth Grade Pupils." In Listening Bibliography, 2nd ed., edited by Sam Duker. Metuchen: The Scarecrow Press, Inc., 1968. Fry, Edward B. "A Readability Formula that Saves Time." The Journal of Reading (April, 1968): 11, 513-516, 575-578. Gagne, Robert M. "Learning Theory, Educational Media, and Individualized Instruction." A paper presented at the Faculty Seminar on Educational Media at Buck- nell University, November 16, 1967. Gates, Arthur I., and Walter H. MacGinite. Teacher's Manual, Gates—MacGinite Reading Tests. New York: Teacher's College Press, Columbia University, 1965. . Technical Manual, 1965. Geroge, Robert Glen. "Retention of Prose Material as a Function of Rate of Presentation and Difficulty of Material." AV Communication Review, XVIII (Fall, 1970). Goldhaber, Gerald M. "Listener Comprehension of Compressed Speech as a Function of the Academic Grade Level of the Subjects." Journal of Communication XX (June, 1970): 167-173. GrOpper, Robert L. "Comprehension of Narrative Passages by Fourth Grade Children as a Function of Listen- ing Rate and Eleven Predictor Variables." Proceed~ ings of the Second Louisville Conference on Rate and/or Frequency-Controlled Speech. Louisville: 1969. 122 Hollingsworth, Paul M. "Can Training in Listening Improve Reading?" In Reading Instruction edited by William K. Durr. Boston: Houghton Mifflin Co., 1967. Kass, Corrine, and Rena Lewis. "Favorite Methods and Ma- terials of the DCLD Membership." Division for Children with Learning Disabilities Newsletter Vol. III, No. 1 (Winter, 1973). Keller, Paul W. "Major Findings in Listening in the Past Ten Years." In Listening: Readings, edited by Sam Duker (New York: The Scarecrow Press, Inc., 1966). Keogh, Barbara K. "Perceptual and Cognitive Styles: Im- plications for Special Education." The First Re- view of Special Education. Philadelphia: Button- wood Farms, Inc., 1973. Kirk, Samuel A., James J. McCarthy, and Winifred D. Kirk. Illinois Test of Psycholinguistic Abilities, Re- vised ed. Urbana: University of Illinois Press, 1968. Lundsteen, Sara W. "Thinking Improvement Program." 1966 (manuscript). McDonald, Charles W. "Problems Concerning the Classification and Education of Children with Learning Disabili- ties." In Learning Disorders, Vol. III, edited by Jerome Hellmuth. Seattle: Special Child Publica- tions, 1968. Miller, Patricia K., and Iran L. Selegman. "Kangaroos." In Sounds of Laughter, edited by Bill Martin, Jr.. New York: Holt, Rinehart, & Winston, Inc., 1966. Miller, George A. "The Magic Number 7, Plus or Minus Two: Some Limits on Our Capacity to Process Information. Psychological Review 63 (1956): 81-97. Mullaly, Lee J. "Comprehension of a Narrative Passage by Primary School Children as a Function of Listening Rate and Reading Comprehension Level." Ph.D. dissertation, Michigan State University, 1972. Nunney, Derek N. and Joseph E. Hill. Personalized Educa- tional Programs. Oakland Community College Press. 123 Orr, David B. "Time Compressed Speech--A Perspective." Journal of Communication 18 (September, 1968): .288- 292. . "The Measurement of Listening Comprehension." Proceedings of the Second Louisville Conference on Rate and/or Freguency-Controlled Speech. Louis- ville: 1969. Osgood, Charles E. "A Behavioristic Analysis." Contempor- ary Approaches to Cognition. Cambridge: Harvard University Press, 1957. Paraskevopoulos, J. N. and S. A. Kirk. The Develgpment and Psychometric Characteristics of the Revised Illinois Test of Psycholinguistic Abilities. Urbana: University of Illinois press, 1969. Rossiter, Charles M. "Rate-of—Presentation Effects on Re- call of Facts and of Ideas and on Generation of Inferences." AV Communication Review 19, 3 (Fall, 1971): 313-324. Sabatino, David A. and W. O. Streissguth. "Word Form Configuration Training of Visual Perceptual Strengths with Learning-Disabled Children." Journal of Learn- ing Disabilities 5, 7 (August-September,-1972). . A set of definitions distributed to members of the Consortium on Auditory Learning. Unpublished, 1973. Salomon, G. "Heuristics for the Generation of Aptitude- Treatment-Interaction Hypotheses." A paper from the Hebrew University of Jerusalem, Israel, 1971. Sedlak, Robert A. and Paul Weener. "Review of Research on the Illinois Test of Psycholinguistic Abilities." The First Review of Special Education. Philadel- phia: Buttonwood Farms, Inc., 1973. Sticht, Thomas G. "Some Relationships of Mental Aptitude, Reading Ability, and Listening Ability Using Normal and Time-Compressed Speech." Journal of Communi- cation 18 (1968). . "Learning by Listening." A paper presented for C.O.B.R.E. Research WorkshOp on Language Compre— hension and the Acquisition of Knowledge. Rouge- mont, North Carolina (March, 1971). IIIIIJ' 11(‘I 124 Taylor, Stanford E. Listening: What Research Says to the Teacher. National Education Association, 1973. U.S., Department of Health, Education, and Welfare. Project No. 6-8477, 1969. U.S., Department of Health, Education, and Welfare Contract No. 3-20-003. "Research and Theory Related to Audio- visual Information Transmission," by R. M. W. Travers. Rev. ed., 1967, Welfare Contract No. 3-20- 003. Waugh, R. P. "Relationship Between Modality Preference and Performance." Exceptional Children 39 (March, 1973): 465-469. Waverly School Directive from the Office of Student Services, Lansing, Michigan, October 8, 1973. Wepman, Joseph M. "Auditory Discrimination, Speech, and Reading." Elementary School Journal 60 (1960): 325-333. Winer, B. J. Statistical Principles in Experimental Design. New York: McGraw-Hill Book Co., Inc., 1962. Witkin, Belle Ruth. "Auditory Perception--Implications for Language Development." In Listening: Readings, Vol. 2, edited by Sam Duker. Metuchen: The Scare— crow Press, Inc., 1971. Woodcock, R. M. and Charlotte Clark. "Comprehension of a Narrative Passage by Elementary School Children as a Function of Listening Rate, Retention Period, and 1.0." Journal of Communication 18 (1968): 259- 271. "The Application of Rate—Controlled Recordings in the Classroom." Proceedings of the Second Louis- ville Conference on Rate and/or Frequency-Controlled Speech. Louisville: 1969. Wunderlich, Ray C., M.D. Kids, Brains, and Learning. St. Petersburg: Johnny Reads, Inc., 1970. "I1111111411111115