"J'-‘-3¢~r~" v ~ I 7}.- ”If: f i} v’ LID1K15 x\L y '5 kitty“ (3., 4,... H. '|1.-\un . 'r‘v I? - -. (j: J LiIlVQI :h)’ 6““ 'vmn-n'm‘mn'W' v".“' 'V“ 1' P‘" "M rRESI: 33“: .O" " 'l M172: NJ This is to certify that the thesis entitled SOME EFFECTS OF VISUAL IMAGERY ON THE READING COMPREHENSION OF THIRD GRADE CHILDREN presented by Beth Bossenbroek Schipper has been accepted towards fulfillment of the requirements for Mega in (fact czg’v“ o .1 Major professor Date filly, “frififzfi 0-7 639 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: ..________.____.__._____.__. Place in book return to remove charge from circuletton record SOME EFFECTS OF VISUAL IMAGERY ON THE READING COMPREHENSION OF THIRD GRADE CHILDREN By Beth Bossenbroek Schipper A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Elementary Education 1979 © Copyright by BETH BOSSENBROEK SCHIPPER 1979 ii ABSTRACT SOME EFFECTS OF VISUAL IMAGERY ON THE READING COMPREHENSION or THIRD GRADE CHILDREN By Beth Bossenbroek Schipper This study investigated the application of visual imagery in teaching reading to third grade children and reports the observed effects on their reading comprehen— sion. This study was designed to be administered to nine classrooms of third grade students from a private school system in the Midwest. One group, the experimental group, received instructions in the use of visual imagery as an aid in reading comprehension. The other group, the control group, received no instructions or suggestions regarding visual imagery. Both the experimental and the control group read the same ten-paragraph prose selection and responded to the same question on that selection. Imme- diately after reading the selection, half of the experi- mental group and half of the control group were tested on their verbatim memory and reading comprehension. These groups were designated as E-l and 0-1. The other half of each group was not immediately tested. These groups were Beth Bossenbroek Schipper designated as E-2 and C-2. One week later all subjects of both groups received the same test to measure comprehen- sion and recall. The results were analyzed with a univariate analy- sis of repeated measures designed to determine whether the visual imagery instructions given to the treatment groups resulted in significantly higher comprehension scores. It was hypothesized that the group receiving instructions in visual imagery would score significantly higher on semantic comprehension and verbatim recall tests than the group receiving no instructions in the use of visual imagery. Five major hypotheses were tested to determine the rela- tionship of visual imagery on verbatim memory and seman- tic comprehension, on both immediate and delayed tests. Conclusions 1. Instructions in the use of visual imagery and encouragement to use it while reading a selection does not significantly affect verbatim memory or semantic comprehen- sion when students are tested immediately after reading the selection or when they are tested one week later without having taken an immediate test. 2. Instructions in the use of visual imagery in conjunction with the reinforcement of an immediate test does significantly affect long-term verbatim memory; how- ever, it does not significantly affect long-term semantic comprehension. Beth Bossenbroek Schipper 3. The use of an immediate test is not signifi- cantly more effective than the use of visual imagery when either long-term verbatim memory or long-term semantic com- prehension was measured. 4. The use of an immediate test significantly improves the scores on both a delayed verbatim memory test and a semantic comprehension test. This study has shown, for the population and methods used, that the use of a combination of visual imagery and immediate testing improves long-term verbatim memory when compared with an immediate test and no visual imagery. Although the differences were not statistically significant, the difference in means also suggests that a combination of visual imagery and immediate testing may also be beneficial for long-term semantic memory when com- pared with an immediate test and no visual imagery. Although this study did not focus on the effects of an immediate test or delayed verbatim memory or seman— tic comprehension, it did show a significant effect on the scores of both tests for the population and methods used. Dedicated to David Fisher for his love, support, encouragement, and, above all, his patience. iii ACKNOWLEDGMENTS To the chairman of my doctoral committee, Dr. William Durr, I wish to express my appreciation for his guidance, advice, and support throughout my doctoral program. I also wish to express my thanks to Dr. George Sherman, who was a source of inspiration throughout my doctoral program, for his knowledge of reading and for keeping things in perspective with his sense of humor. For their kind help, I express my gratitude to Dr. Louise Sause and Dr. Edwin Keller, who graciously served on my committee and gave support and encouragement. A special thanks to Michael York, for his help with the statistics, and to Janelle Bremer for her help in collecting the data; to Jim and Ted, who made me a believer; to Jack DeWaard, Ernie Miesen, and Dr. Juanita Russell for giving me time from school responsibilities in Arabia; and to Jay, Chip, Fran, and Robert, for reasons they will under- stand. Last, but most important, is the gratitude I owe my parents, Rev. Ed and Bette Bossenbroek, who were always there. iv TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . vii Chapter I. THE PROBLEM . . . . . . . . . . . . . . . . 1 Background to the Problem . . . . . . . . 4 Importance of the Study . . . . . . . . . 12 Definition of Terms . . . . . . . . . . . 13 Purpose of the Study . . . . . . . . . . . 14 Hypotheses . . . . . . . . . . . . . . 14 Analysis of the Data . . . . . . . . . . . 15 Educational Implications . . . . . . . . . 16 Organization of the Study . . . . . . . . 16 II. REVIEW OF THE LITERATURE . . . . . . . . . . 18 Theoretical Viewpoints . . . . . . . . . . 18 Visual Imagery in Sentences . . . 22 Visual Imagery and Comprehension of Prose 24 Summary and Conclusions . . . . . . . . . 29 III. METHODOLOGY AND DESIGN . . . . . . . . . . . 32 Overview . . . . . . . . . . . 32 Development of the Study . . . . . . . 33 Selecting the Reading Material . . . . . 34 Formulating the Test . . . . 36 Selecting and Screening the Student Sample . . . . . . . 38 Formulating Instructions in Visual Imagery . . . . 39 Administering the Reading Selection and the Test . . . . . . . . . . . . . 40 Analysis of Data . . . . . . . . . . . . 41 Summary . . . . . . . . . . . . . . . . . 44 Chapter Page IV. RESEARCH FINDINGS . . . . . . . . . . . . . . 45 Introduction . . . . . . . . . . . . . . . . 45 Hypotheses . . . . . . . . . . . . . . . 45 Analysis of Variance . . . . . . . . . . . . 47 Hypothesis 1 . . . . . . . . . . . . . . . 49 Hypothesis 2 . . . . . . . . . . . . . . . 51 Hypothesis 3 . . . . . . . . . . . . . . . 53 Hypothesis 4 . . . . . . . . . . . . . . . 55 Hypothesis 5 . . . . . . . . . . . . . . . 57 Summary of Results . . . . . . . . . . . . . 60 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . . 67 Introduction . . . . . . . . . . . . . . . . 67 Summary . . . . . . . . . . . . . . . . . . 67 Conclusions . . . . . . . . . . . . . . . . 68 Implications . . . . . . . . . . . . . . . . 69 Limitations . . . . . . . . . . . 71 Instructional Significance . . . . . . . . 72 Recommendations for Future Studies . . . . . 74 APPENDICES . . . . . . . . . . . . . . . . . . . . . . 76 A. READING SELECTION . . . . . . . . . . . . . . 77 B. VERBATIM MEMORY AND SEMANTIC COMPREHENSION TEST . . . . . . . . . . . . . . . . . . . . 81 C. PLACEBO TEST . . . . . . . . . . . . . . . . . 83 D. VISUAL IMAGERY INSTRUCTIONS . . . . . . . . . 85 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 88 vi Table 4.1a LIST OF TABLES Comparison of Group Receiving Visual Imagery Instructions With Group Receiving No Visual Imagery Instructions: Immediate Verbatim Test . . . . . . . . . . . . . Comparison of Group Receiving Visual Imagery Instructions With Group Receiving No Visual Imagery Instructions: Immediate Semantic Test . . . . . . . . . . . . . . Comparison of Group Receiving Visual Imagery Instructions and No Immediate Test With Group Receiving No Visual Imagery Instruc- tions and No Immediate Test: Delayed Verbatim Test . . . . . . . . . . Comparison of Group Receiving Visual Imagery Instructions and No Immediate Test With Group Receiving No Visual Imagery Instruc- tions and No Immediate Test: Delayed Semantic Test . . . . . . . . . . Comparison of Group Receiving Visual Imagery Instructions and an Immediate Test With Group Receiving No Visual Imagery Instruc- tions and an Immediate Test: Delayed Verbatim Test . . Comparison of Group Receiving Visual Imagery Instructions and an Immediate Test With Group Receiving No Visual Imagery Instruc— tions and an Immediate Test: Delayed Semantic Test . . . . . . . . . . . Comparison of Group Receiving Visual Imagery Instructions and No Immediate Test With Group Receiving No Visual Imagery Instruc- tions and an Immediate Test: Delayed Verbatim Test . . . . vii Page 49 50 51 52 54 55 56 Table 4.4b Comparison of Group Receiving Visual Imagery Instructions and No Immediate Test With Group Receiving No Visual Imagery Instruc- tions and an Immediate Test: Delayed Semantic Test . . . . . . . . . . . . Comparison of Group Receiving Visual Imagery Instructions and an Immediate Test With Group Receiving Visual Imagery Instruc- tions and No Immediate Test: Delayed Verbatim Test . . . . . . . . . Comparison of Group Receiving Visual Imagery Instructions and an Immediate Test With Group Receiving Visual Imagery Instruc- tions and No Immediate Test: Delayed Semantic Test . . . . . . . Analysis of Variance for Raw Scores on Immediate Verbatim Tests Comparing 1 Treatment Group E1 to Control Group C Analysis of Variance for Raw Scores on Immediate Semantig Tests Comparing 2 Treatment Group E to Control Group C Analysis of Variance for Raw Scores on Delayed Verbatim Test Comparing 2 Treatment Group E2 to Control Group C Analysis of Variance for Raw Scores on Delayed Semantic Tests Comparing 2 Treatment Group E2 to Control Group C Analysis of Variance for Raw Scores on Delayed Verbatim Tests Comparing 1 Treatment Group E1 and Control Group C Analysis of Variance for Raw Scores on Delayed Semantic Test Comparing 1 Treatment Group E1 and Control Group C Analysis of Variance for Raw Scores on Delayed Verbatim Tests Comparing Treatment Group E2 and Control Group C viii Page 57 58 59 62 62 63 63 64 64 65 CHAPTER I THE PROBLEM Our world of the twentieth century has been built and held together in great part by the written word. Con- sider the contracts, advertising, treaties and treatises, love letters, newspapers, notes to the milkman, specifi- cations and reports, instructions and directions, great literature and bad literature, correspondence, examina- tions, medicine bottle labels, dissertations, job applica- tions, and inspiration for lonely people. Any list demon- strating the importance of the written word, especially in today's environment of advanced technologies, heightened sensitivities, and omnipresent media, would indeed be long, if not endless. The corollary is, of course, that in order for all these written words to be of any value, someone must read them and understand them. Most of us do that, fortunately, because we've been taught the skills of reading and have experienced the rewards of the entire process from a fairly early age. The ability to read is now something most of us take for granted in our various daily pursuits. Yet, a 1969 U.S. census report indicated that 31 million adults 25 years and older were at or below a mar- ginal survival threshold in literacy.1 A 1971 Harris survey for the U.S. Department of Health, Education and Welfare showed that over 15 percent of our people beyond the age of 16 were illiterate, functionally illiterate, or suffered from deficiencies in their reading ability.2 The need to upgrade basic standards of literacy in the United States has become a recognized national issue and millions of taxpayer dollars are being spent to improve reading achievement standards in the United States by 1980.3 These funds are not only reinforcing the priority of reading instructions in elementary school curriculums; they are also being used to supplement secondary and college cur- riculums where inadequate reading skills have become appar- ent as well. It has become clear that while the great part of our citizenry can read well enough to help themselves and our society, there is still a significant number of people among us that cannot read or, if they can, they can- not read well enough. Those that cannot are at an obvious 1Wanda D. Cook, Adult Literacy Education in the United States (Newark, Delaware: International Reading Association, 1977), p. 105. 2Roger DeDrow, Adult Reading Development (Washing— ton, D.C.: National Reading Center Foundation, 1972), p. 4. 31bid., p. 108. disadvantage when it comes to giving and receiving their fair share within our system. The need to remedy their situation is apparent. Unfortunately, the remedy is not one that is immediately at hand. Research by educators and psycholo— gists has been going on in the field of reading since the turn of the century. But only within the last decade has really significant attention been devoted to reading com— prehension, the process a person goes through in order to extract meaning from written text. Current research efforts are designed to find procedures for improving that process. But the relatively young field is still forma— tive and hampered by a complexity of ideas, theories, and processes difficult to verify by direct observation or measurement. Research experiments in reading comprehen— sion are necessarily matters of practical application where various techniques are tried out on readers in controlled laboratory and classroom situations and attempts made to determine which techniques might really be helpful in teaching people to read with comprehension. This paper outlines one such study. It deals with the application of visual imagery in teaching reading to eight-year-old elementary students and reports some of the observed effects on some aspects of their reading compre- hension. Background to the Problem Reading is a complex cognitive process. Its com— plexity makes reading research difficult. Exploration is further complicated because the cognitive process involved in reading cannot be directly observed or empirically measured. Educators and psychologists continue to explore and define their processes and their components but their conclusions thus far about the reading process are diverse. Definitions cover a wide spectrum with authorities in the field of reading offering different views in the attempt to clarify what reading is. But they do share one common central conclusion: comprehension is the ultimate goal. As Durr points out: Reading is comprehending; unless the reader under- stands what he reads, he is not, in the truest sense of the word, reading. Although others may define reading as nothing more than pronouncing words on the printed page, the teacher cannot afford the lux- ury of such a loose definition. If our goal is to teach boys and girls to read, we must clearly under- stand that we have not yet achieved the goal until we have taught them to understand the printed page. But there is yet no consensus among researchers as to the elements of reading comprehension. Ekwall states, "Research to date has generally shown that we are not able to accurately differentiate more than about two 4William K. Durr, ed., Reading Instruction: Dimen- sions and Issues (Boston: Houghton-Mifflin Co., 1967), p. 126. or three broad factors (in comprehension).”5 George and Evelyn Spache stress that Reading is not a group of separate skills to be prac- ticed in isolation and later blended together into the total act. It is rather a total act from begin- ning, limited, of course, by the reader's reasoning capacities, his experiences and his reading abili— ties. For Spache and Spache, word identification is not enough to describe the reading process. Most important are the factors involved in word meanings, the relationships between ideas, and the mental processes used in reasoning. Dechant7 views reading as two processes: the mechanical process of perception and the interpretation of the symbols by the reader. Dechant emphasizes that the reading process is not complete until the reader grasps the significance of the meaning. Dechant, like Spache, agrees that reading is more than one process and comprehension is the critical element. Gray8 also views reading as a multi-faceted process 5Eldon E. Ekwall, Diagnosis and Remediation of the Disabled Reader (Boston: Allyn and Bacon, Inc., 1976), p. 53. 6George SpacheznulEvelyn Spache, Reading in Ele- mentary School (Boston: Allyn and Bacon, Inc., 1969), p. 8. 7Emerald V. Dechant, Improving the Teaching of Reading (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1970), p. 400. 8William S. Gray, ”The Major Aspects of Reading," in Sequential Develgment of Reading Abilities, ed. HelenM. Robinson, Supplementary Educational Monographs, 90 (Chicago: University of Chicago Press, 1966). consisting basically of word perception and comprehension. Gray believes that the reader directs his attention to the printed page with his mind on meaning. The meaning may go beyond the literal level to critical levels of thinking-— thus a reaction and fusion of new ideas with former experi- ences and ideas take place. Although comprehension is the goal of the reading process, "we know far less about factors that influence comprehension of sentences and longer passages of dis- course in reading than we do about factors that influence the recognition of individual words."9 And reviews of the literature reveal few research studies in the area of read- ing comprehension until the last decade. Neither is there agreement as to a single theo- retical model of the comprehension process. Gibson and Levin state, "If there is no single reading process--but instead many reading processes, there can be no single model for reading."10 Although no consensus has been reached regarding a theoretical model of comprehension, researchers have moved on to investigate the effectiveness of specific strategies for improving comprehension. Their investigations regarding instructional strategies and other 9Eleanor J. Gibson and Harry Levin, The Psychology of Reading (Cambridge, Mass.: MIT Press, 1976), p. 392. 10Ibid., p. 438. factors affecting comprehension fall into two categories. The first category of research focuses on the manipulation or organization of the text itself. The second category focuses on the effects of instructional or pre-learning strategies on the characteristic behaviors of the student. Cromer11 experimented by changing the structure of reading materials to see if it would benefit those who had poor comprehension but had adequate decoding and vocabulary skills. He organized a story in two ways. In the first, the sentences appeared in regular form. In the second, the sentences were organized according to agreed upon phrase boundaries as prescribed by LeFevre.12 Cromer found that sentences in the regular form produced significant differences in the comprehension of good and poor readers. However, when the phrasing format was imposed on the sen- tences, the poor readers' comprehension was equal to that of the good readers. Thus, changing the organization of printed material proved effective for poor readers. Ruddell13 showed that more comprehension takes place when the material read resembles the sentence llWard Cromer, "The Difference Model: A New Expla- nation for Some Reading Difficulties,” Journal of Educa- tional Psychology 61 (1970): 471-83. 12Carl A. LeFevre, Linguistics and the Teaching_of Reading (New York: McGraw—Hill, 1964). 13Robert B. Ruddell, ”The Effect of the Similarity of Oral and Written Patterns on Language Structure on Reading Comprehension," Elementary English 42 (1964): 403-10. structure of the reader's own oral language pattern than when sentence patterns are strange. Rohwer and Matz14 experimented with children from lower and middle class groups. Both groups differed in IQ and in reading achieve- ment. Stories were read to the subjects with accompanying printed sentences or sketched pictures. The comprehension of both groups was greater when the stories were accom- panied by pictures rather than printed sentences. This increase in comprehension was most pronounced and signifi- cant within the group of poor readers. Since the pic- tures seemed to aid comprehension of both good and poor readers, it seems plausible to ask what significance imposed pictures (as in Rohwer and Matz's study) or induced pictures play in reading comprehension. All of these studies suggest that organization of material may affect comprehension. Farley and Eischens15 studied the effect of adjunct questions on the reading comprehension of students in grades three to six. Adjunct questions inserted in prose passages proved to enhance comprehension both immediately 14William D. Rohwer, Jr., and Robert D. Matz, "Improving Aural Comprehension in White and in Black Chil- dren: Pictures versus Print," Journal of Experimental Child Psychology 19 (1975): 23—36. 1”Frank H. Farley and R. R. Eischens, "Children's Processing of Prose: The Effects of Question Arousal, Text Complexity and Learner Strata on Short- and Long-Term Retention," Technical Report No. 201, Wisconsin Research and Development for Cognitive Learning (1971). and one week later for children in grades four to six. The results from children in grade three are unclear. These studies suggest that adjunct questions call atten— tion to important facts or principles, thereby enhancing comprehension. The second category of research focuses on the effects of instructional or prelearning strategies on the characteristic behavior of the reader. The emphasis of this research is upon the cognitive strategies the reader employs as a result of instructions given by an experi- menter. One such instructional strategy which has received some experimentation is the effect of visual imagery on comprehension. Many studies have shown that visual imagery can be put to conscious use and enhance the learning of paired associates.16 These studies test the learning of lists of unrelated words to which the subjects must apply their own structure. For instance, to learn the paired 16Richard C. Anderson, "Encoding Processes in the Storage and Retrieval of Sentences," Journal of Experimen- tal Psychology 91 (1971): 338-40; Gordon H. Bower, "Imagery as a Relational Organizer in Associate Learning," Journal of Verbal Learnigg and Verbal Behavior 9 (1970): 529-33; Bergen R. Bugelski, "Images a Mediator in a One-Trial Paired—Associate Learning II: Self-Timing in Successive Lists," Journal of Experimental Psychology 77 (1968): 328- 34; J. C. Yuille and Allan Paivio, "Imagery and Verbal Mediation Instructions in Paired-Associate Learning," Journal of Experimental Psychology 78 (1968): 436-41. 10 associate vase-bird, the subject might visualize a vase with a bird sitting on it. The use of visual imagery seems to help form these relationships between discrete items and aid in their recall. It is hypothesized that one item is recalled and acts as a retrieval cue for the asso- ciated item, thus generating the whole. The research here suggests that visual imagery may provide the organizational structure for remembering discrete items. Yet, even though the use of visual imagery has been proven to be a helpful memory strategy in learning unrelated words, little has been done with visual imagery as a strategy for the compre- hension of prose. The comprehension of prose involves cognitive pro- cesses about which little is actually known. Research to date has not proven that visual imagery is as effective a tool for comprehending prose as it is for memory of dis- crete items. There are two major distinctions between the recall of discrete items and the comprehension of prose which require further investigation. The most obvious dis- tinction is the difference between isolated words or phrases and prose. Prose is characterized by its inherent organization. It is not a list of unrelated words or phrases. The words of prose are arranged in units accord- ing to certain rules of grammar and syntax in order to create relationships between words which in turn convey meaning. These meaningful or "semantic" units express the 11 ideas or concepts to be comprehended. The semantic unit to be comprehended is already embedded in the organiza- tional structure provided by the text. The use of visual imagery as an additional device for organizing the whole to aid in comprehension and recall is still in question and the subject for additional research. The second distinction is less clear. It involves the difference, if any, between comprehension and recall. If one recalls something, does it also mean that one has comprehended it? Are "recall" and "comprehension" inter- changeable terms for the purpose of measuring the ability to extract meaning from written text? The literature shows differing views. In this study and paper the term "recall" will not be taken to mean comprehension, as "recall" can also mean the result of rote memory or verbatim recall, which does not necessarily involve true comprehension or understanding. The research discussed above provides some insight into the factors that may influence reading comprehension. But further study of these factors is required in order to validate the new strategies and techniques which are to be included in actual instructional programs. This paper is a further investigation into the possibilities the applica- tion of visual imagery may have in enhancing reading com- prehension. 12 Importance of the Study Although there is an increased interest in research on reading comprehension, there remains a need for studies which are applicable to the classroom setting. The major— ity of studies regarding the factors which influence read- ing comprehension have been done in a laboratory setting. Furthermore, most of the research has been done with adults and is applicable only to adults. In order for the effec- tiveness of instructional strategies to be verified for the classroom teacher, research must be conducted with chil— dren in the classroom setting. Venezky states, If reading research is to influence instruction, then more experimental psychologists will have to be per- suaded to interact professionally with educational planners and developers to concern themselves with the practical side of reading.17 Visual imagery has been studied by psychologists since the beginning of the century, but the relevance visual imagery may have for reading comprehension has been recognized only recently. Scientific proof of the effec— tiveness of visual imagery as a strategy to facilitate reading comprehension in children must still be obtained. 17Richard L. Venezky, "A History on Research on Reading," Theoretical Paper No. 56, Wisconsin Research and Development Center for Cognitive Learning (1975). 13 Definition of Terms Visual imagegy--a mental picture response that occurs as a result of an extrinsic stimulus found in the context of written discourse. Semantic comprehension--The definition of semantic comprehension for this study includes the processes involved in arriving at a literal understanding of text. Comprehension in this study does not include rote memory of words or sentences. (See verbatim memory.) It does include a complex cognitive process used in extracting the meaning from a text and its relation to an established system of knowledge. Verbatim memory--recall of words or phrases as stated in the text. The author of this study does not define verbatim memory as comprehension. Direction--The definition in this study refers to a command to carry out a set of actions; i.e., "use visual imagery." Instructions—-In this study instructions include the training or lessons given prior to performance. In this study the training of visual imagery was done through modeling. Long-term--In this study long—term refers to a one-week period of time. 14 Purpose of the Study The purpose of this study is to investigate the effects of instructions in visual imagery on the reading comprehension of third grade children. Hypotheses Hypothesis 1A: Pupils with visual imagery instruc- tions will have significantly greater verbatim memory on an immediate test than pupils without visual imagery instruc- tions. Hypothesis 1B: Pupils with visual imagery instruc- tions will have significantly greater semantic comprehension on an immediate test than pupils without visual imagery instructions. Hypothesis 2A: When there is no immediate test for either group, pupils with visual imagery instructions will have significantly greater verbatim memory on a delayed test than pupils without visual imagery instructions. Hypothesis 2B: When there is no immediate test for either group, pupils with visual imagery instructions will have significantly greater semantic comprehension on a delayed test than pupils without visual imagery instruc- tions. Hypothesis 3A: When there is an immediate test for both groups, pupils with visual imagery instructions will have significantly greater verbatim memory on a delayed test than pupils without visual imagery instructions. 15 Hypothesis 3B: When there is an immediate test for both groups, pupils with visual imagery instructions will have significantly greater semantic comprehension on a delayed test than pupils without visual imagery instruc- tions. Hypothesis 4A: Pupils with visual imagery instruc- tions and no immediate test will have significantly greater verbatim memory on a delayed test than pupils without vis— ual imagery instructions and an immediate test. Hypothesis 4B: Pupils with visual imagery instruc- tions and no immediate test will have significantly greater semantic comprehension on a delayed test than pupils with- out visual imagery instructions and an immediate test. Hypothesis 5A: When both groups receive visual imagery instructions, pupils who have had an immediate test will have significantly greater verbatim memory uoande popzqeoo asoz MO 830 H 0 mo mooasom .No 050nm Hoganoo 0» mm Qsonw paoausogp mawnsaeoo mumop capamEom opsfluosafl no monoom ssh now ooadwns> mo mflmaasamll.nw.v wands $000.00 00 Hapoe 000¢.H wwuw.mm 30 masouo canvas 000.3 0mv>.H 0000.m 0000.m H mgsoau noospom A00.v m m onsadm wonssvm . . nofipsflmd> moanse 06950800 0862 Ho 83m m 0 Ho mooasom .Ho 030nm Homuqoo Op m Qsonm unmeadohp mafiysmaoo mpmop Sfiusnuo> opswnoasfl so mouoom was How oossfihs> mo mfimzasamll.sm.v oases 63 on¢.®NH hm H0009 mmeH.m moom.omH mm masouo aHnHHs ooo.e ommw.m Hemm.o Hemm.m H masons ammapom Am0.v m m oussvm mohdsvm . . soapsflus> poanwe Hop30800 ado: Ho 83m m 0 no common .mo macaw Honpaoo 09 mm Quouw pacepmonp waHHmQEoo mpwop oHpcsEom 0o>3H90 so moaoom Buy How ooasfiad> mo mHm>HSG0 Hmpoe b0b0.H 0mmb.¢0H 00 masono :anHz ooo.e ammo. «66H. mmmH. H masons ammapmm A00.v m m maddvm momdscm . . aoHpsHHs> UoHQse 00050800 new: Ho 85m H 0 Ho moouzom .No anonm Honaaoo on mm asonm pnospsohp wcfiusgsoo pmop SHpano> comsHou so oaoom 36H you oocsflms> Ho MHmaasc.umH 00 H0509 0000.0 0000.00H 00 005050 555553 000.0 HO0H.0 0000.0 mmmm.> H 005050 5003500 A00.v 0 0 050500 0055500 . . 50550550> 00Hnae 00550800 5002 50 850 5 0 50 0005500 .H0 05050 H055500 055 Hm 05050 550850055 055550800 5005 05550800 0000H00 50 005000 305 505 005055d> 50 0500H05 00Hnmh 00550800 5502 50 850 5 0 50 0005500 .H0 05050 H0555o0 050 Hm 05050 550850055 055500800 05005 8555550> 0005H00 50 005000 385 505 005055a> 50 0500H555II.00.0 0H509 65 0000.00H H0 H5509 omHv.m 5st.¢vH om 053650 555555 000.0 0005.H ~0H0.0 NmHm.v H 005050 5003500 H00.0 0 . 0 055500 0055500 . . 50555555> 00H55B 00550800 5502 50 850 5 0 50 0005500 .H0 05050 H055500 055 am 05050 550855055 055550800 05005 8555550> 0005H00 50 005000 355 505 0055555> 50 0505H55<|I.50.0 0Hn59 66 The next and final chapter will summarize the results and give implications and recommendations for future research. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Introduction The effect of visual imagery instructions on the reading comprehension of eight-year—olds was the focus of the general hypotheses of this study. The results and implications of the findings are presented in this final chapter. Summary This study was designed to test the effectiveness of instructions in visual imagery on reading comprehension. The sample was composed of nine classrooms from a private school system in the Midwest. The pupils of nine classrooms in each school were randomly assigned to one of the four groups: visual imagery instructions with an immediate and delayed test (E1), visual imagery with only a delayed test (E2), no visual imagery instructions with an immediate and a delayed test (C1), and no visual imagery instructions and only a delayed test (C2). The treatment groups (E1 and E2) received 20 minutes of instructions on how to form visual images before reading the selection and were encouraged to 67 68 make mental images as they read. The control groups (C1 and C2) received no visual imagery instructions. All of the groups received the same delayed test, consisting of five verbatim memory questions and five seman- tic comprehension questions. One imagery group and one control group received the same test immediately after read- ing the selection. Analysis of variance with one repeated measure was used to test significant differences between treatment effects on verbatim memory and semantic comprehension. Conclusions 1. Instructions in the use of visual imagery and encouragement to use them while reading a selection do not significantly affect verbatim memory or semantic com- prehension when students are tested immediately after read— ing the selection or when they are tested one week later without having taken an immediate test. 2. Instructions in the use of visual imagery in conjunction with the reinforcement of an immediate test do significantly affect long-term verbatim memory; however, they do not significantly affect long-term semantic compre— hension. 3. The use of an immediate test is not signifi- cantly more effective than the use of visual imagery when either long-term verbatim memory or long-term semantic com- prehension was measured. 69 4. The use of an immediate test significantly improves the scores on both a delayed verbatim memory test and a semantic comprehension test when students have received visual imagery instructions. Implications During the last decade the area of reading compre- hension has received increasing emphasis in research. Almost without exception, reading programs incorporate reading comprehension as a major part of the instructional activities. Many types of activities are suggested by pub- lishers and educators to increase comprehension of mate- rials. However, many of the activities suggested are not supported by research. The majority of activities used are those which follow the reading of a selection, such as those found in workbooks or those used during discussions with the teacher. Few instructional strategies which may estab- lish an organized cognitive pattern of behavior applicable by the reader to any selection in order to improve compre- hension are used prior to the reading of a selection. Further, much of the research done in comprehension has been completed in laboratory settings. There remains the need for proven instructional strategies which are useful in the classroom by the classroom teacher. The use of visual imagery is one such strategy that has been regarded as a relevant aid to learning by some 7O psychologists; however, only recently has there been an interest by researchers in the use of visual imagery to increase the comprehension of prose. This study attempted to explore the possibility that the use of visual imagery while reading would improve the immediate and delayed verbatim memory and semantic compre- hension of students. This study has shown, for the population and methods used, that the use of a combination of visual imagery and immediate testing improves long-term verbatim memory when compared with an immediate test and no visual imagery. Although the differences were not statistically significant, the difference in means also suggests that a combination of visual imagery and immediate testing may also be beneficial for long-term semantic memory when compared with an imme- diate test and no visual imagery. Although this study did not focus on the effects of an immediate test on delayed verbatim memory or semantic comprehension, it did show a significant effect on the scores of both tests for the population and methods used. This study has not, however, provided a definitive answer regarding the application of visual imagery to the comprehension of prose. There are several limitations con- nected with this study which may explain why there were few significant differences between the compared test groups in this study. 71 Limitations The most obvious limitation in this study may have been the small size of the sample. If the number of chil- dren used in this study had been greater than 124, more results might have been significant. Another limitation is the inability to directly observe the use of visual imagery. It is possible that many of the pupils in the control groups were already naturally incorporating the use of imagery while reading. Thus, the differences in scores may not be reflective