Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A psycholinguistic investigation of the cloze responses of grade eight students Cram, Ruby Victoria 1980

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1980_A2 C73.pdf [ 8.11MB ]
Metadata
JSON: 831-1.0078359.json
JSON-LD: 831-1.0078359-ld.json
RDF/XML (Pretty): 831-1.0078359-rdf.xml
RDF/JSON: 831-1.0078359-rdf.json
Turtle: 831-1.0078359-turtle.txt
N-Triples: 831-1.0078359-rdf-ntriples.txt
Original Record: 831-1.0078359-source.json
Full Text
831-1.0078359-fulltext.txt
Citation
831-1.0078359.ris

Full Text

c7 A PSYCHOLINGUISTIC INVESTIGATION OF THE CLOZE RESPONSES OF SECONDARY SCHOOL STUDENTS by RUBY VICTORIA CRAM B.A. , The Univ e r s i t y of B r i t i s h Columbia, 1966 M.Ed., The University of B r i t i s h Columbia, 1973 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION i n \ THE FACULTY OF GRADUATE STUDIES Department of Reading Education We accept t h i s t hesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May 1980 @ Ruby V i c t o r i a Cram, 1980 In presenting this thesis in p a r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It i s understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 D E - 6 B P 7 5 - 5 1 1 E ABSTRACT This exploratory study investigated the r o l e of exact and non-exact-replacements of cloze responses i n the assessment of reading com-prehension. Two modes of discourse, narrative f i c t i o n and expository prose, were investigated. Two t h e o r e t i c a l assumptions guided the study: from p s y c h o l i n g u i s t i c s , that reading involves responses to the grapho-phonemic, s y n t a c t i c , and semantic cue systems of language (Goodman, 1976a); and from discourse a n a l y s i s , that a "schema" or cognitive map d i r e c t s the reader i n the search for discourse cues (Winograd, 1977). Subjects were p r o f i c i e n t and l e s s p r o f i c i e n t secondary school students at two l e v e l s of maturity. Attitude to reading was also examined. Operational d e f i n i t i o n s of discourse were: ( i ) na r r a t i v e f i c t i o n or conventions of a story, and ( i i ) expository prose or coherent explan-ati o n of a t o p i c . Subjects were entering grades nine (N = 107) and twelve (N = 100) i n Lord Byng Secondary School, Vancouver, B r i t i s h Columbia. To examine exact replacements (E.R.'s) and a t t i t u d e s , only subjects whose primary language was English were e l i g i b l e . Exact and non-exact-replacements (N.E.R.'s) were examined for a random sample at each grade l e v e l of Good (N = 20) and Poor (N = 30) readers i d e n t i f i e d from scores on the compre-hension subtest of the Iowa S i l e n t Reading Test (1973). To v e r i f y how l i n g u i s t i c cues t r i g g e r responses, s i x subjects were randomly drawn from each p r o f i c i e n c y group for retrospective v e r b a l i z a t i o n interviews, which i i were taped and transcribed. Each subject (N = 207) completed the Estes  Reading Attitude Scale and two cloze tasks: a narrative f i c t i o n and an expository prose, from the B r i t i s h Columbia Reading Assessment 1977, Grades 8 and 12. Responses were tested f o r exact match to the author's word (Bormuth, 1975). To evaluate N.E.R's, the investigator adapted the Cambourne Reading Assessment Procedure (1978), based on the Goodman  Taxonomy of Reading Miscues (1969). Following two p i l o t studies, the c l a s s i f i c a t i o n scheme was made consistent with discourse theory and the coding s i m p l i f i e d . A synonym replacement for the exact response was acceptable i n three categories: syntax, semantics, and discourse. S t a t i s t i c a l procedures included c o r r e l a t i o n , independent t - t e s t s , and two-way analysis of variance. For the. o r a l protocols, categories were induced from the t r a n s c r i p t i o n s . Frequency of response was analyzed using the chi-square s t a t i s t i c , supported by q u a l i t a t i v e d e s c r i p t i o n . A ttitude to reading had a generally weak c o r r e l a t i o n with the selected indices of comprehension. For exact cloze scores, r e l a t i o n -ships with the standardized measure were s i g n i f i c a n t , p a r t i c u l a r l y with expository prose f or poor grade twelve subjects. Narrative f i c t i o n scores exceeded expository prose scores. Good readers were d i f f e r e n -t i a t e d from poor readers. The N.E.R. score discriminated between pro f i c i e n c y l e v e l s . At grade nine, n a r r a t i v e f i c t i o n scores exceeded expository prose, but at grade twelve, differences were not s i g n i f i c a n t . The mean i n t e r - r a t e r agreement, calculated by the Arrington Formula ( F e i f e l & Lorge, 1950), was 91.6 percent. The interviews demonstrated that three cue systems operated most frequently: syntax, semantics, and discourse; and two much less often: i i i grammatical function and l i f e experience. S i g n i f i c a n t differences i n frequency were found between modes of discourse and pr o f i c i e n c y l e v e l s . For combined exact scores plus synonyms, i n grade nine, narrative f i c t i o n scores exceeded prose scores, but i n grade twelve the reverse occurred. Discrimination between p r o f i c i e n c y l e v e l s was noted; how-ever good readers were s i g n i f i c a n t l y superior with expository prose. Data based conclusions were (1) a t t i t u d e i s not correlated with either proficiency^ or comprehension, (2) comprehension scores d i f f e r e d f o r modes of discourse: narrative f i c t i o n and expository prose, (3) exact cloze score discriminated between p r o f i c i e n c y l e v e l s , (4) N.E.R. scores revealed differences i n the use of cue systems by a b i l i t y groups, (5) a l l readers used the same cue systems: syntax, semantics, and discourse, to gain meaning, but control of the set of cue systems, e s p e c i a l l y with expository prose, distinguished the good reader, and (6) the addition of synonym scores to exact cloze scores d i f f e r e n t i a t e d between p r o f i c i e n c y l e v e l s , modes of discourse, and maturity l e v e l s . i v TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES v i i LIST OF FIGURES x ACKNOWLEDGEMENTS x i Chapter I INTRODUCTION AND STATEMENT OF THE PROBLEM 1 Statement of the Problem 2 Background of the Problem 3 Research Problem 21 Research Questions and Hypotheses 22 Limitations of the Study 24 Significance of the Study 24 De f i n i t i o n s 24 Summary 27 Organization of the Study 27 II REVIEW OF THE LITERATURE 28 Evaluations i n the Problem Area 28 Psyc h o l i n g u i s t i c Theory 33 Error Studies 34 Miscue Analysis 35 Cloze Procedure 38 Discourse Analysis . . . . , 46 Theories of Discourse 47 Methodology 51 Attitude 56 Summary 58 III METHODOLOGY 59 Research Design 59 The Population Sample 60 Instrumentation 62 Procedures . . . . . 63 Data Processing 64 Hypotheses 67 Summary 72 v IV RESULTS 73 S t a t i s t i c a l Analyses 73 Post Hoc Analysis 95 R e l i a b i l i t y of Scoring 100 Retrospective Verbalizations 103 S t a t i s t i c a l Analysis of the Frequencies 107 Summary of Findings 115 Summary 117 V SUMMARY, CONCLUSIONS, DISCUSSION AND IMPLICATIONS . . 119 Discussion of the Findings 121 Implications 125 Suggestions f o r Further Research 127 Concluding Statement 128 REFERENCES 130 APPENDIX A 142 APPENDIX B 154 APPENDIX C 171 v i LIST OF TABLES Table 1.1 Taxonomy of Non-Exact Replacements 7 1.2 Results of P i l o t Study 1 — N a r r a t i v e F i c t i o n Grammatical Function of N.E.R.'s 9 1.3 Results of P i l o t Study 1 — N a r r a t i v e F i c t i o n Syntactic A c c e p t a b i l i t y of N.E.R.'s 9 1.4 Results of P i l o t Study 1 — N a r r a t i v e F i c t i o n Semantic A c c e p t a b i l i t y of N.E.R.'s 10 1.5 Results of P i l o t Study 1 — N a r r a t i v e F i c t i o n Loss of Meaning . 10 1.6 Results of P i l o t Study 2 — E x p o s i t o r y Prose Grammatical Function of N.E.R.'s 12 1.7 Results of P i l o t Study 2 — E x p o s i t o r y Prose Syntactic A c c e p t a b i l i t y of N.E.R.'s 12 1.8 Results of P i l o t Study 2 — E x p o s i t o r y Prose Semantic A c c e p t a b i l i t y of N.E.R. 's 13. 1.9 Results of P i l o t Study 2 — E x p o s i t o r y Prose Loss of Meaning 13. 1.10 Grammatical Function of Cloze Responses on F i c t i o n and S o c i a l Studies Materials 15 1.11 Syntactic A c c e p t a b i l i t y of Cloze Responses on F i c t i o n and S o c i a l Studies Materials . . . . . . . . 15 1.12 Semantic A c c e p t a b i l i t y of Cloze Responses on F i c t i o n and So c i a l Studies Materials 16 1.13 Loss of Meaning i n Cloze Responses of F i c t i o n and S o c i a l Studies Materials 16 1.14 Category 4: Cambourne's Taxonomy of N.E.R.'s . . . . 17 1.15 Revised Taxonomy of N.E.R.'s 19 2.1 H i e r a r c h i c a l Components within some Taxonomic D e f i n i t i o n s of Reading 29 2.2 Levels of Comprehension 30 2.3 Levels of Comprehension and the Cognitive Domain . . . 30 2.4 Tasks within Levels of Comprehension 31 3.1 Sources of Variance and Degrees of Freedom f o r a 2 x 2 F a c t o r i a l Randomized Design, Fixed E f f e c t s Model 65 3.2 Weighting System for Four Variables of Non-Exact Cloze Replacements 66 4.1 Intercorrelations among Selected Indexes for Grade Nine and Twelve Readers 75 4.2 Intercorrelations among Selected Indexes for Good and Poor Readers i n Grade Nine 7g 4.3 Intercorrelations among Selected Indexes f o r Good and Poor Readers i n Grade Twelve 78 v i i 4.4 C o r r e l a t i o n C o e f f i c i e n t s of P e r c e n t Complete D i s c o u r s e A c c e p t a b i l i t y S c o r e s and P e r c e n t E x a c t C l o z e Scores f o r N a r r a t i v e and E x p o s i t o r y Passages by Good and Poor Readers i n Grades Nine and Twelve 81 4.5 Comparison Between Exact C l o z e N a r r a t i v e and Ex a c t C l o z e E x p o s i t o r y Means f o r Grades Nine and Twelve . 83 4.6 Comparison between Means of E x a c t C l o z e N a r r a t i v e and Exact C l o z e E x p o s i t o r y f o r Good and Poor Readers i n Grades Nine and Twelve 85-4.7 Comparison between Means of Non-Exact-Replacements of C l o z e Responses on Modes of D i s c o u r s e by P r o f i c i e n c y L e v e l i n Grade Nine 86 4.8 Comparison between Means of Non-Exact-Replacements o f C l o z e Responses on Modes of D i s c o u r s e by P r o f i c i e n c y L e v e l i n Grade Twelve 86 4.9 A n a l y s i s of V a r i a n c e o f Exact C l o z e S c o r e s o f N a r r a t i v e and E x p o s i t o r y Modes by Good and Poor Readers i n Grade Nine 89 4.10 A n a l y s i s o f V a r i a n c e o f E x a c t C l o z e S c o r e s o f N a r r a t i v e and E x p o s i t o r y Modes by Good and Poor Readers i n Grade Twelve 89 4.11 Means of Weighted Scores A c c o r d i n g t o Grammatical F u n c t i o n , S y n t a c t i c A c c e p t a b i l i t y , Semantic A c c e p t a b i l i t y , and D i s c o u r s e A c c e p t a b i l i t y of Non-Exact-Replacement Words i n a N a r r a t i v e and E x p o s i t o r y Passage by 20 Good and 20 Poor Readers i n Grade Nine 91 4.12 Means of Weighted Scores A c c o r d i n g t o Grammatical F u n c t i o n , S y n t a c t i c A c c e p t a b i l i t y , Semantic A c c e p t a b i l i t y , and D i s c o u r s e A c c e p t a b i l i t y o f Non-Exact-Replacement Words i n a N a r r a t i v e and E x p o s i t o r y Passage by 20 Good and 20 Poor Readers i n Grade Twelve 92 4.13 A n a l y s i s of V a r i a n c e of Ex a c t C l o z e P l u s Complete D i s c o u r s e A c c e p t a b i l i t y Scores of N a r r a t i v e and E x p o s i t o r y Modes by Good and Poor Readers i n Grade Nine 96 4.14 A n a l y s i s o f V a r i a n c e o f Ex a c t C l o z e P l u s Complete D i s c o u r s e A c c e p t a b i l i t y S c o r e s o f N a r r a t i v e and E x p o s i t o r y Modes by Good and Poor Readers i n Grade Twelve 96 4.15 A n a l y s i s of V a r i a n c e o f Complete D i s c o u r s e A c c e p t a b i l i t y Scores of N a r r a t i v e and E x p o s i t o r y Modes by Good and Poor Readers i n Grade Nine . . . . 97 4.16 A n a l y s i s of V a r i a n c e o f Complete D i s c o u r s e A c c e p t a b i l i t y Scores of N a r r a t i v e and E x p o s i t o r y Modes by Good and Poor Readers i n Grade Twelve . . . 97> 4.17 P e r c e n t a g e of Agreement between I n v e s t i g a t o r and Independent Judges 1 and 2 i n the S c o r i n g of Non-Exact-Replacements o f C l o z e Responses i n the N a r r a t i v e Mode by Grade Nine Readers 104 v i i i 4.18 Percentage of Agreement between Investigator and Independent Judges 3 and 4 i n the Scoring of Non-Exact- Replacements of Cloze Responses i n the Expository Mode b y Grade Twelve Readers . 104 4.19 Frequencies and Percentages from Narrative and Expository Modes by Good and Poor Readers i n Grade Nine with Chi Squares . 109 4.20 Frequencies and Percentages from Narrative and Expository Modes by Good and Poor Readers i n Grade Twelve with Chi Squares I l l 5.1 Comparison between Cloze C r i t e r i o n Levels and Mean Percentages on Exact Cloze Scores on Modes of Discourse by P r o f i c i e n c y and Grade Levels . . . . 126 5.2 Comparison between Cloze C r i t e r i o n Levels and Mean Percentage Exact Cloze plus Complete Discourse A c c e p t a b i l i t y Scores on Modes of Discourse by Pr o f i c i e n c y and Grade Levels 126 i x LIST OF FIGURES Figure 4.1 Mean Exact Replacement Cloze Scores for Narrative F i c t i o n and for Expository Prose Passages by 20 Poor and 20 Good Readers i n Grade Nine 93 4.2 Mean Exact Replacement Cloze Scores for Narrative F i c t i o n and for Expository Prose Passages by 20 Poor and 20 Good Readers i n Grade Twelve . . . . 94 4.3 Mean Exact Replacement Cloze Plus Complete Discourse A c c e p t a b i l i t y Scores for Narrative F i c t i o n and for Expository Prose Passages by 20 Poor and 20 Good Readers i n Grade Nine 98 4.4 Mean Exact Replacement Cloze Plus Complete Discourse A c c e p t a b i l i t y Scores f o r Narrative F i c t i o n and for Expository Prose Passages by 20 Poor and 20 Good Readers i n Grade Twelve . . . . 99 4.5 Mean Complete Discourse A c c e p t a b i l i t y Scores f o r Narrative F i c t i o n and for Expository Prose Passages by 20 Poor and 20 Good Readers i n Grade Nine 101 4.6 Mean Complete Discourse A c c e p t a b i l i t y Cloze Scores f o r Narrative F i c t i o n and for Expository Prose Passages for 20 Poor and 20 Good Readers i n Grade Twelve 102 x ACKNOWLEDGEMENTS There are many who were e s s e n t i a l to the completion of t h i s study. My thanks go to Mr. Barney O'Brien, P r i n c i p a l of Lord Byng Secondary School, for h i s u n f a i l i n g cooperation. To Dr. E. G. Summers, who thought not only that I could but should, and then demanded my best. To my d i s s e r t a t i o n committee—Drs. M. Dank, G.' M a l l e t t , and K. Reeder— for t h e i r i n t e r e s t and support. Special thanks to Dr. H. R a t z l a f f , who was invaluable i n the design and i n t e r p r e t a t i o n of the s t a t i s t i c s , and to Dr. R. D. Chester, my thesis advisor, for h i s guidance, patience, and f a i t h from the beginning. To fellow students S. Craddock, H. Halpern, G. Walker, and colleague T. Dun f o r t h e i r support and assistance as judges. To Nina Thurston for the meticulous preparation of the t y p e s c r i p t . To my mother, my two chi l d r e n , and above a l l , my husband, Tom, who encouraged me and put my needs before t h e i r own. This d i s s e r t a t i o n i s dedicated to my father, who had a quiet, shining f a i t h i n h i s daughter. x i CHAPTER I INTRODUCTION AND STATEMENT OF THE PROBLEM Comprehension i s a complex process in v o l v i n g p h y s i c a l , psycho-l o g i c a l , i n t e l l e c t u a l and emotional responses to language. It i s the process whereby meaning i s extracted from written language. (Project B u i l d , 1979, p. 7) The i n v e s t i g a t i o n of reading comprehension i n terms of the q u a l i t y of the responses of students to written language has s i g n i f i c a n t implica-tions for the teacher i n the secondary school. Report 4 of Project  Build (Bringing Unity into Language Development), prepared for Vancouver School Board, recommends that a l l subject teachers have " . . . some knowledge of the reading process together with an understanding of the s p e c i a l reading demands of d i f f e r e n t subjects. . . ." (p. 15). Espe-c i a l l y important i s the d i f f i c u l t y inherent i n the forms of language. Various e f f o r t s have been made to examine the r o l e of language i n read-ing from the p s y c h o l i n g u i s t i c viewpoint of the mental processes under-l y i n g i t s use. In p a r t i c u l a r , the analysis of responses to the system-a t i c deletions of words i n a passage, c a l l e d cloze procedure, o f f e r s a promising perspective f o r the assessment of the i n d i v i d u a l needs of students as a supplement to t r a d i t i o n a l t e s t i n g . Simons (1971) emphasized: Process oriented research i s motivated by the assumption that the effectiveness of an i n s t r u c t i o n a l technique i s i n part dependent on the extent to which those techniques c a p i t a l i z e on the actu a l psychological processes which students u t i l i z e i n learning, (p. 340) 1 2 Research directed both at uncovering the basic processes of reading com-prehension and at the diagnosis of the strengths, as well as the weak-nesses of students, i s one important way of improving i n s t r u c t i o n . Statement of the Problem The use of the response has been the basis of extensive research i n reading. Simons (1971) stated: "The comprehension process i s the mental operations which take place i n the reader's head while he i s reading" (p. 340). Since these are not observable, the covert mental process must be i n f e r r e d from e i t h e r the products, as test r e s u l t s a f t e r reading, or more recently, from the descriptions of behaviour, such as responses during reading. For example, i n cloze procedure responses are evaluated f o r the exact match to the author's word (Bormuth, 1975). Many e f f o r t s have been directed to the examination of the cueing systems i n language. The inadequacy of e x i s t i n g schemes f or studying responses i n terms of errors only, documented by Weber (1968), l e d Goodman (1969) to develop a technique f o r the in-depth analysis of o r a l reading responses. From the study of the patterns of "miscues," defined as deviations from the printed text, was formulated a p s y c h o l i n g u i s t i c theory that reading i s an integrated response to three kinds of cue systems: (1) the grapho-phonic cues, which come from the v i s u a l information of the printed symbols: (2) the synta c t i c cues from the structure of the sentence; and (3) the semantic or meaning cues provided by the context of the whole sentence and by the mode of discourse, or the pattern of organization of the en t i r e passage. Recently, Cambourne extended the Goodman method by devising a taxonomy for the analysis of the responses 3 to cloze tasks i n an e f f o r t to understand the s i l e n t reading process, i n narr a t i v e f i c t i o n (1977) and i n the text-book prose of biology (1978). Other methods such as protocols and retrospection-introspection have also been used to investigate responses, but r e l a t i v e l y l i t t l e r e l a t e d s p e c i f i c a l l y to the secondary school. A dearth of studies on the response of students to the cues i n d i f f e r i n g modes of discourse, such as na r r a t i v e f i c t i o n or expository prose was found. None concerned a comparison of the differences i n responses to cloze tasks between the two modes of language use. And although other research suggests that a f f e c t i v e factors such as i n t e r e s t , a t t i t u d e , and self-concept, a f f e c t reading comprehension (Athey, 1970), none combined any of these factors with response. The present exploratory study was designed to investigate the responses i n reading of secondary students to e s t a b l i s h a better perspec-t i v e on the comprehension process. S p e c i f i c a l l y , the primary purpose was to examine whether the a c c e p t a b i l i t y of responses to cloze tasks varies with two modes of discourse: narrative f i c t i o n and expository prose. The secondary purpose was to investigate whether the responses vary with the a t t i t u d e of the student. Background of the Problem The problem of i n v e s t i g a t i n g reading comprehension i n terms of the responses of students resulted from three factors which led to the two p i l o t studies: the demands of reading-to-learn, survey r e s u l t s , and the l i m i t a t i o n s of the s k i l l s model. Gibson and Levin (1975) stated: "Reading to learn i s an e s s e n t i a l and e n t i r e l y d i f f e r e n t a c t i v i t y from reading a poem or novel for 4 pleasure" (p. 10). Not only Is the purpose for reading d i f f e r e n t , but also the structure of the language which i s at the farthest pole from either the students' own spontaneous natural language or the f a m i l i a r patterns of f i c t i o n . Yet, as Rosen (1972) observed, the secondary school o f f e r s : . . . the only chance, c e r t a i n l y the main chance, of acquiring the language and thought of impersonal observation and descrip-t i o n , g e n e r a l i z a t i o n and abstraction, theories, laws, the analysis of events remote i n time and space, argument and speculation. (p. 121) One objective of secondary education i s to teach students how to compre-hend the d i s t i n c t i v e formal language of expository prose, which i s the main form of discourse i n the subject d i s c i p l i n e s . Olson (1977a) stated: "Children's encounters with text or the preparation for those encounters constitute the one absolutely d i s t i n g u i s h i n g feature of schooling" (p. 76). The d i f f i c u l t i e s many students at a l l l e v e l s of a b i l i t y encounter i n reading content area materials are well recognized. For example, the B r i t i s h Columbia Reading Assessment, 1977, Grade 8 and 12, a survey of a l l the students i n the province by the Mi n i s t r y of Education, showed that, at both l e v e l s , scores on narrative f i c t i o n passages exceeded those on the expository s e l e c t i o n s . As a r e s u l t , teachers were urged to emphasize s k i l l development i n "comprehending expository materials" (1977c, p. 44). Instru c t i o n based on s k i l l s has been the most common method of teaching reading, e s p e c i a l l y at the secondary l e v e l (Palmer, 1979, p. 7). Goal 4 of the Secondary G u i d e — E n g l i s h 8-12 (Revised 1978) stat e s : "The secondary school English program w i l l develop i n students a range of reading and study s k i l l s " (1979, p. 10). In the s k i l l s approach, 5 reading i s an a c t i v i t y that i s learned sequentially. The diverse s k i l l s and s u b s k i l l s are arranged i n a h i e r a r c h i c a l structure, with each s k i l l depending on the one below and competency increasing as the reader moves from l e v e l to l e v e l . The reader f i r s t learns to match sounds and l e t t e r s , then combines other phonic and "word attack" s k i l l s to develop "word perception" and f i n a l l y , comprehends the meaning of the sentence and then the e n t i r e discourse. " T y p i c a l l y the a c q u i s i t i o n and use of reading s k i l l s i n h i e r a r c h i c a l order are explained through models of reading known as taxonomies" (Palmer, 1979, p. 3). The emphasis on the s k i l l s model i s evident i n standardized tests (Farr, 1969), reading c l i n i c s and courses (Sawyer, 1974), p r o f e s s i o n a l reading textbooks (Herber, 1970; Shepherd, 1973; Burmeister, 1974; Thomas & Robinson, 1977; Estes & Vaughan, 1978), and i n prescribed i n s t r u c t i o n a l materials; for example, such s e r i e s as Be a Better Reader, Success i n Reading, and.Advanced ' S k i l l s i n Reading. As d e s c r i p t i v e t o o l s , taxonomic models c l a r i f y the range of i n s t r u c t i o n a l tasks for the teacher, but the f a i l u r e to deal with how language i s structured i s a serious l i m i t a t i o n . As the only basis f o r the assessment of i n d i v i d u a l student's a b i l i t i e s and responses, the s k i l l s approach i s equally inadequate. T y p i c a l l y , the poor reader i s assessed i n terms of missing reading s k i l l s . The underlying assumption i s that with correct diagnosis, the s k i l l s can be i s o l a t e d and the s t u -dent taught to improve. Sawyer (1974) observed that often the only d i f f e r e n c e i n reading s k i l l s i n s t r u c t i o n from one grade to the next may be the teacher and the reading materials. Secondary students who have d i f f i c u l t y with reading often change l i t t l e and may even f a l l further behind, as the demands of reading content area materials increase. 6 In a survey of the l i t e r a t u r e to determine possible d i r e c t i o n s for research, MacGinitie (1975-76) stated that the question of what cogni-t i v e operations c h i l d r e n perform i n learning to read i s an important one. It i s necessary to analyze current i n s t r u c t i o n a l materials to assess the range of cognitive demands, to locate the most frequently required oper-ations, and to compare the demands of d i f f e r e n t materials. Research i s needed i n how information i s conveyed by the conventions and l o g i c of sentence i n t e r r e l a t i o n s h i p s and by larger organizational s t r u c -ture which are the devices by which attitudes are communicated. Such st r a t e g i e s should be studied i n r e l a t i o n to reading comprehension. The question of how students read led to two p i l o t studies by t h i s inves-t i g a t o r . P i l o t Study 1 Cambourne (1977, 1978) developed a taxonomy, based on the Goodman procedure (Table 1.1), for evaluating the non-exact-replacements (N.E.R.'s) of cloze d e l e t i o n s . Each response i s examined for i t s grammatical func-t i o n , s y n t a c t i c a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and for loss of meaning. The purpose of the p i l o t study was to r e p l i c a t e the Cambourne Read- ing Assessment Procedure with selected average and low average grade eight readers. Subjects were ten high-average and ten low-average readers from a suburban j u n i o r secondary school i n Delta, B r i t i s h Colum-b i a . The English teachers each selected the students, male and female, from t h e i r subject classes. Cloze deletions were performed on an unfamiliar, complete short story, "Voodoo" by Frederic Brown. The f i r s t sentence was i n t a c t ; the remainder was typed with fifth-word deletions for a t o t a l of 66 blanks. 7 TABLE 1.1 TAXONOMY OF NON-EXACT REPLACEMENTS Aspect of Psy c h o l l n g u l s t i c Sub-categories of Answers to Concern Question Asked Questions 1 Grammatical Does the reader's Y Yes replacement serve the N No Function same grammatical P Cannot t e l l function as the intended word(s)? 2 Syntactic Does the reader's Y Yes, the complete T-unit replacement r e s u l t i n i s acceptable A c c e p t a b i l i t y a s y n t a c t i c a l l y P Yes, but at sub."T-unit leve acceptable construc- N No. S y n t a c t i c a l l y tion? unacceptable 3 Semantic Does the reader's Y Yes, at the whole story replacement r e s u l t i n l e v e l A c c e p t a b i l i t y a semantically T Yes, but only at T-unit acceptable construc- l e v e l tion? ?! Yes, but only at sub T-unit l e v e l with p r i o r portion of the sentence ^2 Yes, but only at sub T-unit l e v e l with following portion of the sentence N No. T o t a l l y unacceptable 4 Loss of Does the reader's N No. There i s no loss i n replacement r e s u l t i n meaning at the whole Meaning a loss of meaning? story l e v e l P Change of unimportant d e t a i l M Change i n major character, incident or sequence Y Yes. T o t a l l y incongruous to the story (Cambourne, 1978) 8 Each N.E.R. was analyzed according to the four categories of the taxon-omy. The r e s u l t s showed that low average readers replaced words of a d i f f e r e n t grammatical function to the author more than two times the rate of the high average readers, 14.1% vs. 34.2% (Table 1.2); had about three times as many t o t a l l y unacceptable syn t a c t i c u n i t s , 9.7% compared to 27.3% (Table 1.3); and maintained the semantic i n t e g r i t y of the sentence 42.8% of the time, compared to 77.2% (Table 1.4). The low group were concerned more with the T-unit l e v e l of meaning (4.1% vs. 11.1%) and with the portion p r i o r to the cloze d e l e t i o n (.4% vs. 5.9%). The trends were substantiated i n the "Loss of Meaning" category. Because the d i s t i n c -tions between "no l o s s " and "change of unimportant d e t a i l " were minimal, i t was f e a s i b l e to collapse the two categories, with the r e s u l t that the high average group maintained meaning or the story l i n e 83.5%, while the low average group comprehended 57.3% (Table 1.5). The r e s u l t s substantiated Cambourne's (1977) study with younger readers and were generally as the Goodman model predicted. The high average readers were s i g n i f i c a n t l y more p r o f i c i e n t i n the replacement of cloze deletions with s y n t a c t i c a l l y and semantically acceptable responses and appeared to maintain the meaning of the e n t i r e passage better than readers judged low average readers. The overwhelming majority of replacements acceptable only with the p r i o r part of the sentence by the low group (5.9%) suggests that the t o t a l context of the sentence was not used e f f i c i e n t l y . The fa c t that high average readers had only .4% of such responses suggests that the con t r o l of the reading process d i f f e r s markedly. In accordance with the p s y c h o l i n g u i s t i c view, such readers are concerned p r i m a r i l y with TABLE 1.2 RESULTS OF PILOT STUDY 1—NARRATIVE FICTION GRAMMATICAL FUNCTION OF N.E.R.'s Pr o f i c i e n c y Y (Yes) Same grammatical function as author N (No) Diff e r e n t to author Can't t e l l High average Low average z value 84.1 63.9 3.26** 14.1 34.2 -3.32** 1.3 2.2 -.48 * p < .05 ** p < .01 Note. A l l scores i n percentages TABLE 1.3 RESULTS OF PILOT STUDY 1—NARRATIVE FICTION SYNTACTIC ACCEPTABILITY OF N.E.R.'s Pr o f i c i e n c y Y (Yes) Complete T-unit acceptable P ( P a r t i a l ) A unit smaller than T-unit i s acceptable N (No) Not acceptable High average Low average z value 88.3 66.1 3.74** 3.0 6.4 •1.14 9.7 27.3 -3.20** **p < .01 Note. A l l scores i n percentages 10 TABLE 1.4 RESULTS OF PILOT STUDY 1—NARRATIVE FICTION SEMANTIC ACCEPTABILITY OF N.E.R.'s Y (Yes) T Pi ?2 N (No) At whole With p r i o r With l a s t Not accept-passage At T-unit portion of portion of able at Prof i c i e n c y l e v e l l e v e l only T-unit only T-unit only any l e v e l High average 77.2 3.6 .4 2.8 14.3 Low average 42.8 11.1 5.9 1.1 39.1 z value 4.96** -2.03* -2.23* •.87 -3.96** * p < .05 ** p < .01 Note. A l l scores i n percentages TABLE 1.5 RESULTS OF PILOT STUDY 1—NARRATIVE FICTION LOSS OF MEANING N (No) Pro f i c i e n c y No loss P Change unimportant d e t a i l M Change of maj or d e t a i l Y (Yes) T o t a l l y incongruous High average 79.6 3.9 Low average 43.7 13.6 z value 5.22** -2..4 3* .4 1.6 -.85' 16.6 41.1 -3.82* * p < .05. ** p < .01 Note. A l l scores i n percentages 11 meaning. It also suggests that p r o f i c i e n t readers use the t o t a l con-text of the passage. Within the l i m i t a t i o n s of the exploratory study, the technique of q u a l i t a t i v e analysis of cloze responses has s i g n i f i c a n t p o t e n t i a l for understanding na r r a t i v e materials. The f a m i l i a r format of the story and simple vocabulary are easy to comprehend, but the less well'-known sy n t a c t i c structures and patterns of organizations of expository prose may not be. This problem was investigated i n the second p i l o t study. P i l o t Study 2 The responses to expository prose were examined with the same sample as p i l o t study 1, but with N = 9 i n the high average group and N = 8 i n the low average group, due to absences. The passage, deleted with 50 blanks, according to standard cloze format (Bormuth, 1976) was a portion of the grade 8 s o c i a l studies s e l e c t i o n "Australian C i t i e s " by A. J . Rose from the B r i t i s h Columbia Reading Assessment, Grades 8  and 12, 1977. It was predicted that scores of high average readers would exceed those of low average readers i n a l l four categories of the Cambourne taxonomy. The r e s u l t s were generally as predicted. High average readers were s i g n i f i c a n t l y more p r o f i c i e n t with grammatical function and syn-t a c t i c a c c e p t a b i l i t y than low average readers (Tables 1.6 and 1.7). The scores were more nearly a l i k e at the l e v e l of the T-unit i n the semantic category (36.6% vs. 30.1%). In the loss of meaning from minor d e t a i l s , scores were 25.3% vs. 22.5% r e s p e c t i v e l y . In reading for meaning i t appears that the responses may be affected by the mode of discourse (Tables 1.8 and 1.9). TABLE 1.6 RESULTS OF PILOT STUDY 2—EXPOSITORY PROSE GRAMMATICAL FUNCTION OF N.E.R.'s Y (Yes) N (No) P (Same grammatical (Different P r o f i c i e n c y function as author) to author) (can't t e l l ) High average 72.4 25.7 0 Low average 65.8 34.2 0 z value 1.01 -1.31 * p < .05 ** p < .01 Note. A l l scores i n percentages TABLE 1.7 RESULTS OF PILOT STUDY 2—EXPOSITORY PROSE SYNTACTIC ACCEPTABILITY OF N.E.R.'s Pro f i c i e n c y Y (Yes) (Complete T-unit acceptable) P ( P a r t i a l ) (A unit smaller than T-unit i s acceptable) N (No) (Not acceptable) High average Low average z value 80.6 68.2 2.01* .5.7 '3.5 '."74 12.9 28.2 -2.68" * p < .05 ** p < .01 Note. A l l scores i n percentages 13 TABLE 1.8 RESULTS OF PILOT STUDY 2—EXPOSITORY PROSE SEMANTIC ACCEPTABILITY OF N.E.R.'s Y (Yes) T Px P 2 N (No) At whole With p r i o r With l a s t Not accept-passage At T-unit portion of portion of able at Pr o f i c i e n c y l e v e l l e v e l only T-unit only T-unit only any l e v e l High average 37.7 36.6 2.6 4.0 16.4 Low average 21.6 30.8 4.6 2.5 40.0 z value 2.49* .87 -.76 -.60 -3.71** * p < .05 ** p < .01 Note. A l l scores i n percentages TABLE 1.9 RESULTS OF PILOT STUDY 2—EXPOSITORY PROSE LOSS OF MEANING N (No) Pr o f i c i e n c y No loss P Change unimportant d e t a i l M Change of major d e t a i l Y (Yes) ' T o t a l l y incongruous High average -41.7 25.3 Low average 24.1 22.5 z value 2.65* .46 16.0 17.9 -•36 16.4 36.0 -3.15** * p < .05 ** p < .01 Note. A l l scores i n percentages 14 Comparison Between P i l o t Studies 1 and 2 The cloze responses on f i c t i o n and s o c i a l studies were then com-pared f o r the high average and low average readers i n the four categories. Scores on grammatical function and s y n t a c t i c a c c e p t a b i l i t y were s i m i l a r . The low average group obtained means of 64.3 vs. 65.8 and 66.5 vs. 68.2 r e s p e c t i v e l y , for f i c t i o n and s o c i a l studies. The scores of the high average group were 84.1 vs. 72.4 on grammatical function and 88.1 vs. 80.6 on s y n t a c t i c a c c e p t a b i l i t y f or the two modes of discourse (Tables 1.10 and 1.11). The differences between the scores of students at the two l e v e l s of p r o f i c i e n c y were much more marked i n the two categories that deter-mine the a c c e p t a b i l i t y of the meaning. The r e s u l t s of a comparison between the two p i l o t studies on the category Semantic A c c e p t a b i l i t y are given i n Table 1.12. The three l e v e l s of a c c e p t a b i l i t y : the T-unit (T), p r i o r to the cloze d e l e t i o n ( P i ) , and a f t e r the d e l e t i o n (P2), were collapsed into the " P a r t i a l " category (Table 1.1). The data of 38.6% for the high average readers and 37.8% f o r the low average readers suggest that both groups have equal d i f f i c u l t y with expository materials, i n contrast to 6.3% and 18.1% re s p e c t i v e l y , on n a r r a t i v e f i c t i o n . In the loss of meaning category, the scores for minor d e t a i l were 3.9 vs. 13.6 f o r f i c t i o n and 25.3 vs. 22.5 for s o c i a l studies for the high average and low average readers. On loss from major d e t a i l , the scores were 16 vs. 17.9, r e s p e c t i v e l y (Table 1.13). The r e s u l t s sug-gested that the mode of discourse a f f e c t s the reading comprehension at both a b i l i t y l e v e l s . In a study of college students reading biology text-book prose, 15 TABLE 1.10 GRAMMATICAL FUNCTION OF CLOZE RESPONSES ON FICTION AND SOCIAL STUDIES MATERIALS P a r t i a l l y Not Acceptable Acceptable Acceptable P r o f i c i e n c y F i c t i o n S.S. F i c t i o n S.S. F i c t i o n S.S. High average 84.1 72.4 1.3 0 14.5 25.7 Low average 64.3 65.8 2.2 0 34.2 33.4 (percentages) z value 3.20** 1.01 -.49 -3.25** -1.19 * p < .05 ** p < .01 TABLE 1.11 SYNTACTIC ACCEPTABILITY OF CLOZE RESPONSES ON FICTION AND SOCIAL STUDIES MATERIALS P a r t i a l l y Not Acceptable Acceptable Acceptable P r o f i c i e n c y F i c t i o n S.S. F i c t i o n S.S. F i c t i o n S.S. High average 88.1 80.6 5.7 5.7 9.7 14.5 Low average 66.5 68.2 9.2 3.5 26.9 32.2 (percentages) z value 3.65** 2.01* .94 .74 -3.14** -2.96** *p < .05 ** p < .01 16 TABLE 1.12 SEMANTIC ACCEPTABILITY OF CLOZE RESPONSES ON FICTION AND SOCIAL STUDIES MATERIALS Pr o f i c i e n c y High average Low average (percentages) z values Acceptable F i c t i o n S.S. 77.2 37.7 42.8 21.6 4.97** 2.49* P a r t i a l l y Acceptable F i c t i o n S.S. 6.3 38.6 18.1 37.8 -2.55* .12 Not Acceptable F i c t i o n S.S. 14.4 16.4 40.3 43.5 -4.11** -4.18** * p < .05 ** p < .01 TABLE 1.13 LOSS OF MEANING IN CLOZE RESPONSES OF FICTION AND SOCIAL STUDIES MATERIALS Change of Change of Unimportant Major T o t a l l y No Loss D e t a i l D e t a i l Incongruous Pr o f i c i e n c y F i c t i o n S.S. F i c t i o n S.S. F i c t i o n S.S. F i c t i o n S.S. High average 79.6 41.7 3.9 25.3 .4 16.0 16.6 16.4 Low average 43.7 24.1 13.6 22.5 1.6 17.9 41.1 36.0 z value 5.22** 2.65* 2.43* .46 -.8S .36 -3.82**-3.15** * p < .05 ** p < .01 17 Cambourne (1978) found that, although responses were acceptable at the f i r s t three l e v e l s , at the l e v e l of the enti r e context, they were " b i o l o g i c a l l y unacceptable." It appeared that ". . . the reading of tec h n i c a l prose requires highly s p e c i f i c scenarios of background knowl-edge and word-use i f meanings are to be extracted and comprehension i s to occur" (p. 10). Discourse Analysis The problem of the e f f e c t of the mode of discourse on responses evolved from the p i l o t studies. The Loss of Meaning category (Table 1.14) i n Cambourne's (1977) taxonomy seemed inadequate f o r i d e n t i f y i n g responses at the l e v e l of the whole passage. TABLE 1.14 CATEGORY 4: CAMBOURNE1S TAXONOMY OF N.E.R.'s Loss of meaning Does the reader's N No. There i s no loss of replacement r e s u l t i n meaning at the whole story loss of meaning? l e v e l P Change of unimportant d e t a i l M Change of major character, incident, or sequence Y Yes. T o t a l l y incongruous to the story In expository materials where the emphasis i s on l o g i c a l order, terms such as "character" are i r r e l e v a n t . A c r i t e r i o n f o r the judgment between "unimportant" and "major" d e t a i l i s lacking; and the re v e r s a l of the coding "N" and "Y", confusing. There was an overlap i n the "whole story" l e v e l , as i t also appeared i n Semantic A c c e p t a b i l i t y . In category 3, Weber's (1977) comment on the a c c e p t a b i l i t y of miscues i n synt a c t i c and semantic context seemed relevant: 18 In s p i t e of t h i s prominence and constant reference to meaning i n the discussion, however, the findings with respect to these categories are not reported f u l l y or methodically enough to be assessed. Nor i s the value of any of the a n a l y t i c categories questioned or t h e i r interrelatedness s u f f i c i e n t l y examined. Nor i s the treatment of semantics as a f i e l d at a l l adequate; i n f a c t , i t misrepresents the current l e v e l of i n t e r e s t and a c t i v i t y now going on. Both prac-t i t i o n e r s and researchers w i l l be disappointed i n the exposition on matters of meaning, because i t i s not cogent enough to be i n s t r u c -t i v e , (p. 419) Kintsch (1977) theorizes that "top-down" processes of cognitive structure or "schema" of the reader i n t e r a c t with the "bottom-up" cues provided by the text. Comprehension depends on the perceptual s i t u a -t i o n s , the subjects' goals and expectations, and the nature of the syntax and mode of discourse. From discourse a n a l y s i s , the study of the organ-i z a t i o n of e n t i r e passages or contexts, has come theories of schema and structure to suggest why n a r r a t i v e f i c t i o n i s more e a s i l y comprehended than expository prose. The f a m i l i a r i t y of the elements of f i c t i o n , such as s e t t i n g , p l o t , and character, permits e f f i c i e n t p redictions; whereas, pe c u l i a r patterns and syntax demand a d i f f e r e n t kind of competence. On the basis of these theories the fourth category of Cambourne's taxonomy became Discourse A c c e p t a b i l i t y ; the other changes appear i n Table 1.15. The taxonomy appears to have considerable p o t e n t i a l for understanding how readers use s t r a t e g i e s , but whether the p r o f i c i e n c y i s the cause or the r e s u l t of the e f f e c t i v e use of context i s s t i l l not c l e a r . Non-verbal factors such as i n t e r e s t s , a t t i t u d e s , and past experience a f f e c t reading s t r a t e g i e s . A d d i t i o n a l i n s i g h t s into the reasons why readers respond may be gained from another technique, r e t r o -spection. Retrospection. Cambourne suggested that "loud-thinkers" are needed to explore how " . . . c e r t a i n l i n g u i s t i c cues t r i g g e r the approp-r i a t e scenarios" (1978, p. 11). Farr (1969) emphasized: 19 TABLE 1.15 REVISED TAXONOMY OF N.E.R.'s Aspect of P s y c h o l i n g u i s t i c Sub-categories of Answers Concern Question Asked to Questions 1 Grammatical Does the reader's Y Yes function replacement serve P Cannot t e l l the same grammat- N No i c a l function as the intended word(s)? 2 Syntactic Does the reader's Y Yes, the complete T-unit i s a c c e p t a b i l i t y replacement acceptable r e s u l t i n a syn- P Yes, but at sub T-unit l e v e l t a c t i c a l l y N No, s y n t a c t i c a l l y unaccept-acceptable able construction? 3 Semantic Does the reader's Y Yes, at the whole sentence a c c e p t a b i l i t y replacement l e v e l r e s u l t i n a T Yes, but only at T-unit l e v e l semantically Pi Yes, but only at sub T-unit acceptable l e v e l with p r i o r portion of construction? the sentence P2 Yes, but only at sub T-unit l e v e l with following portion of the sentence N No. T o t a l l y unacceptable 4 Discourse Does the reader's a c c e p t a b i l i t y replacement r e s u l t i n an acceptable meaning within the whole mode of discourse? Y Yes. The meaning i s i n t a c t at the whole discourse l e v e l P Change of l o g i c a l d e t a i l (Expository Prose) OR Change i n major character, incident, or sequence (Narrative F i c t i o n ) N No. T o t a l l y incongruous to the discourse (Adapted from Cambourne, 1978) 20 By studying students' responses, i t may be possible to determine i f the student goes through a d i f f e r e n t mental procedure i n com-prehending science material than he does i n comprehending s o c i a l studies materials. (1969, p. 121) Retrospection i s a technique adopted from cognitive psychology. In reading research, the subject i s required to describe h i s thoughts a f t e r reading (Pierkarz, 1954; Strang & Rogers, 1965; Smith, 1967; Fareed, 1971). In studies originated by Jenkinson (1957), and continued by Laing: the data were allowed to determine the c r i t e r i a best suited to c l a s s i f y and describe the reading processes used by the sub-j e c t s to complete the meaning of the (cloze) reading passages. (Laing, 1974, p. 56) The taping of one subject's d e s c r i p t i o n of how he read a passage on the exploration and settlement of A u s t r a l i a suggested that u s e f u l i n s i g h t s can be gained by the technique. For example, the boy a s s o c i -ated " f u r - t r a d e r s " with exploration, implying that his background knowl-edge i s l i m i t e d to Canada. The post hoc analysis of the recorded descriptions revealed q u a l i t a t i v e differences i n the responses to nar-r a t i v e f i c t i o n and expository prose. The Role of Attitudes It was apparent from observation of students' comments during p i l o t study 2 that an a f f e c t i v e component was operating. The reading act involves more than cognitive and l i n g u i s t i c systems. Attitudes have been defined i n various ways. Alexander and F i l l e r (1976) con-sider a t t i t u d e s : to consist of a system of fee l i n g s r e l a t e d to reading which causes the learner to approach or avoid a reading s i t u a t i o n . A learner's attitudes may vary with his personal predispositions and may be affected i n unique ways by v a r i a b l e s within the learner and h i s environment. (p. 1) 21 Although research suggests that attitudes tend to be personal, unique and highly unpredictable (Squire, 1964), the consensus i s that a p o s i t i v e a t t i t u d e i s a factor i n maximal success i n reading. Athey pointed out (1976) that a v a i l a b l e evidence supports the view that good readers are "more i n t e l l e c t u a l l y oriented," exhibit more p o s i t i v e a s p i r -ations, show more c u r i o s i t y , and are more p o s i t i v e to school i n general — r e a d i n g i n p a r t i c u l a r , than less p r o f i c i e n t readers (p. 366). It appears, however, that t h i s aspect of the reading process has received l e s s attention than i t deserves (Alexander & F i l l e r , 1976, p. 1). The main d i f f i c u l t y i s to integrate a f f e c t i v e variables into reading theory without also bringing i n the context of a personality theory, which may be "armchair" rather than based on experimental e v i -dence (Athey, 1976, p. 366). Since p s y c h o l i n g u i s t i c models which are the basis of t h i s study have no a f f e c t i v e component, i t was necessary to investigate the r o l e of attitudes i n an a n c i l l a r y manner. Research Problem The main purpose of the study was to inve s t i g a t e whether the com-prehension process as measured by the differences i n responses i s affected by the mode of discourse. The primary objective was to determine whether there were s i g n i f i c a n t differences i n reading comprehension between selected p r o f i c i e n t readers and les s p r o f i c i e n t readers at two le v e l s of maturity, grade nine and grade twelve, as measured by the number of (1) the exact replacements of cloze responses, and (2) the non-exact-replacements of cloze responses, i n two modes of discourse, nar-r a t i v e f i c t i o n and expository prose. The secondary purpose was to determine i f the number of exact and non-exact-replacements varied with 22 the a t t i t u d e to reading as measured by an a t t i t u d e scale. In the main part of the study, the independent variables were: 1. grade l e v e l — n i n e and'twelve 2. mode of d i s c o u r s e — n a r r a t i v e f i c t i o n and expository prose 3. l e v e l of p r o f i c i e n c y — g o o d and poor. The dependent variables to measure the responses to cloze dele-tions i n the reading selections were: 1.. the exact-replacement scores 2. the non-exact-replacement scores i n four categories: a) grammatical function b) s y n t a c t i c a c c e p t a b i l i t y c) semantic a c c e p t a b i l i t y d) discourse a c c e p t a b i l i t y 3. complete discourse a c c e p t a b i l i t y scores i n which responses were acceptable i n three categories: a) s y n t a c t i c a c c e p t a b i l i t y b) semantic a c c e p t a b i l i t y c) discourse a c c e p t a b i l i t y Attitude to reading was measured by scores on the Estes Reading Attitude  Scale (E.R.A.S.) (1975). The o r a l responses to cloze deletions were explored by the technique of retrospection. Research Questions and Hypotheses Research Question 1 Are there c o r r e l a t i o n a l r e l a t i o n s h i p s among selected indices of reading comprehension and a measure of a t t i t u d e for secondary students i n grades nine and twelve? Hypothesis 1.1 There are s i g n i f i c a n t c o r r e l a t i o n a l r e l a t i o n s h i p s among (1) scores on a reading comprehension measure, (2) the number of exact replacements 23 of cloze responses i n narr a t i v e f i c t i o n mode of discourse, (3) the number of exact replacements of cloze responses i n expository prose mode of discourse, and (4) at t i t u d e scores. Research Question 2 Does the subject comprehension vary with the mode of discourse? Hypothesis 2.1 There i s a s i g n i f i c a n t difference i n reading comprehension between the narrative f i c t i o n mode of discourse and the expository prose mode of discourse, as measured by the number of (1) exact replacements of cloze responses and (2) non-exact-replacements of cloze responses, over given l e v e l s of grade and p r o f i c i e n c y . Research Question 3 Do students at two l e v e l s of p r o f i c i e n c y , good and poor, vary i n comprehension? Hypothesis 3.1 There i s a s i g n i f i c a n t d i f f e r e n c e i n reading comprehension between good readers and poor readers as measured by the number of (1) exact replacements of cloze responses and (2) non-exact-replacements of cloze responses, over given reading sel e c t i o n s and grade l e v e l s . Research Question 4 Is there s i g n i f i c a n t i n t e r a c t i o n between mode of discourse and pr o f i c i e n c y level? Hypothesis 4.1 There :is> s i g n i f i c a n t i n t e r a c t i o n across modes of discourse (narrative f i c t i o n and expository prose) and pr o f i c i e n c y l e v e l s (good and poor) i n grade nine and i n grade twelve, i n the number of exact replacements of cloze responses. 24 Limitations of the Study The l i m i t a t i o n s of the study were as follows: 1. The findings are l i m i t e d to the two types of modes of discourse selected: narrative f i c t i o n and expository prose; however, implica-tions may be extended to other s i m i l a r passages. 2. Discourse variables were l i m i t e d to mode; narrative schemata and r h e t o r i c a l patterns were not co n t r o l l e d . 3. Elements from discourse theory were employed to uncover how native speakers use the cues i n comprehension; no attempt at the analysis of passages was made. 4. The subjects from grade nine and grade twelve, randomly selected from the e n t i r e grade on the basis of t h e i r reading p r o f i c i e n c y , were from one secondary school. Conclusions r e l a t e primarily to t h i s population; however, implications may be extended to other s i m i l a r populations. 5. Selection of subjects was further l i m i t e d to those who were native speakers of English. 6. The subjects were directed to complete cloze deletions and to ve r b a l -ize t h e i r responses. Consequently, an a r t i f i c i a l s i t u a t i o n arose which may not be ne c e s s a r i l y s i m i l a r to the process an i n d i v i d u a l may have i n independent reading. Significance of the Study The d e l i n e a t i o n of.responses to cues of language by base-line research i n the comprehension process may lead to an appropriate method for group diagnosis, presently lacking, f o r content area teachers i n secondary schools. D e f i n i t i o n s For the study, the following d e f i n i t i o n s were used: 25 1. Reading ". . . i s a ps y c h o l i n g u i s t i c guessing game. It involves an i n t e r a c t i o n between thought and language" (Goodman, 1976c, p. 498). 2. Comprehension ref e r s to the responses of the reader to written l a n -guage, as measured by the number of exact replacements of cloze dele-tions and by the number of non-exact-replacements of cloze deletions i n four categories: grammatical function, syntactic a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y . 3. A response i s a purposeful, rule-based procedure f o r the comprehension of a written language. 4. Grammatical function r e f e r s to the part of speech of the cloze response; that i s , noun, pronoun, verb, adjective, adverb, preposition, coordin-ate conjunction, subordinate conjunction, a r t i c l e , i n t e r j e c t i o n , as evaluated on a scale of: 1. Y Yes 2. P Cannot t e l l 3. N No 5. The syntactic a c c e p t a b i l i t y of the non-exact-replacements means that the cloze response accords with the l i n g u i s t i c judgement of native speakers, as evaluated on a scale of: 1. Y Yes, the complete T-unit i s acceptable 2. P Yes, but at the sub T-unit l e v e l 3. N No, s y n t a c t i c a l l y unacceptable 6. The semantic a c c e p t a b i l i t y of the non-exact-replacement i s an evalua-t i o n of the congruency of meaning of the cloze response with the con-text of the sentence, as measured on a fi v e - p o i n t scale: 1. Y Yes, at the whole sentence l e v e l 2. T Yes, but at T-unit l e v e l 3. Pi Yes, but at sub T-unit l e v e l with p r i o r portions of the sentence 4. P 2 Yes, but at sub T-unit l e v e l with following portion of the sentence 5. N No, t o t a l l y unacceptable 7. Discourse a c c e p t a b i l i t y of the non-exact-replacement i s an evaluation 26 of the congruency of the meaning of the cloze response with the con-text of the complete passage, according to the mode of discourse, as measured by a scale: Y Yes. The meaning i s i n t a c t at the whole discourse l e v e l P Ghange of l o g i c a l d e t a i l (Expository Prose) OR Ghange i n major character, incident, or sequence (Narrative f i c t i o n ) N No. T o t a l l y incongruous to the discourse 8. A T-unit i s defined as the minimal terminable unit that contains a verb, f o r example, a main clause, plus any subordinate clauses attached to i t . 9. Discourse r e f e r s to a self-contained sequence of connected sentences that constitutes e i t h e r a complete story (narrative f i c t i o n ) or a coherent explanation of a p a r t i c u l a r topic (expository prose). 10. Narrative f i c t i o n refers to the convention of a sequence of connected sentences which t e l l a story. A simple plot may involve a problem facing a main character, a sequence of attempts by the main character to solve the problem, and an eventual r e s o l u t i o n (Kintsch, 1977). 11. Expository prose re f e r s to the convention of a sequence of connected sentences which i s an explanation or d e s c r i p t i o n of a p a r t i c u l a r topic and i s characterized by r h e t o r i c a l devices such as "therefore," that convey patterns of_reasoning. 12. Discourse analysis r e f e r s to a method for e x p l i c a t i o n of a text and the cognitive structures and processes of language users (Winograd, 1977) according to a schema. 13. A schema i s a body of related knowledge to be used i n reasoning. 14. Miscues are deviations from the printed text i n o r a l reading. 15. Exact replacements (E.R.'s) are the exact match to the author's word which has been deleted i n a cloze t e s t . 27 16. Non-exact-replacements (N.E.R.'s) are the replacements which d i f f e r from the exact word deleted. Summary The t h e o r e t i c a l and p r a c t i c a l importance of research on the a c q u i s i -t i o n of information from texts through the process of reading i s generally acknowledged. Equally apparent i s the d i f f i c u l t y of the task: the com-p l e x i t i e s of the comprehension process and of the study of semantics, and, in p a r t i c u l a r , the lack of good tools f o r prose-related research 7(Pearson, 1978). The n a t u r a l i s t i c approach to research which advocates in-depth studies of small numbers of students and materials t y p i c a l of the content areas i n the actual classroom s e t t i n g appeared to be a v i a b l e approach (Goodman, 1976b). This study was, therefore, designed to study reading comprehension from the ps y c h o l i n g u i s t i c view that reading involves the integrated response to the l i n g u i s t i c and semantic cues i n language. Organization of the Study The report of the i n v e s t i g a t i o n i s presented as follows. 1. Chapter II contains the basic t h e o r e t i c a l framework and a review of related l i t e r a t u r e . 2. Chapter I II describes the design and procedures followed i n the c o l -l e c t i o n of the data. 3. Chapter IV reports the quantitative and q u a l i t a t i v e analyses of the findings of the study. 4. Chapter V presents the summary of the study, a discussion of the fin d i n g s , conclusions, and implications f o r teaching and further research. CHAPTER II REVIEW OF THE LITERATURE The purpose of t h i s chapter i s to present the extensive research on the e f f o r t s to understand the comprehension process by the q u a l i t a t i v e examination of the responses to written language, e s p e c i a l l y the cueing systems i n language. The chapter begins with an evaluation of the s k i l l s model and of the task of measurement i n reading. The p s y c h o l i n g u i s t i c framework which forms the r a t i o n a l e for t h i s study follows. Research on responses as errors, as o r a l miscues, and as cloze replacements i s pre-sented next. Recent developments i n discourse analysis which o f f e r the p o s s i b i l i t y of examining responses i n terms of the organization of the ent i r e passage are then summarized. Studies i n retrospection, as a method of understanding responses, and i n attitudes conclude the review of l i t e r a t u r e . Evaluations i n the Problem Area The S k i l l s Model The l i m i t a t i o n s of the s k i l l s model to explain comprehension have been recorded by many educators, i n sp i t e of the pervasive nature of the approach evident i n i n s t r u c t i o n a l materials. The lack of agreement i n the d e f i n i t i o n s of terms has been documented (Shafer, 1978; Emans, 1979). Stern (1973) found four c e n t r a l uses of the term "comprehension": 1. 'comprehension' i s used as degrees or l e v e l s of grasping the meaning of the printed page; 2. 'comprehension' i s used as a set of s k i l l s and a b i l i t i e s which can be measured; 28 29 3. 'comprehension' i s used as grasping the meaning of various l i n g u i s t i c u n i t s ; 4. 'comprehension' i s used as a process equivalent to thinking or understanding. (p. 256) Stern concluded that reading methodologists tend to use the term i n a manner " . . . divorced from any t h e o r e t i c a l formulations about language and l i n g u i s t i c performances . . . they seem more concerned with psycho-l o g i c a l considerations about the teaching of meaning" (p. 257). The lack of knowledge about what meaning i s leads to further ambiguity. Palmer (1979, p. 3) pointed out the differences i n the hierarch-i c a l components of the reading process within some taxonomic d e f i n i t i o n s (Table 2.1). TABLE 2.1 HIERARCHICAL COMPONENTS WITHIN SOME TAXONOMIC DEFINITIONS OF READING William A. S t e r l Olive S. David H. Emerald V. S. Gray A r t l e y N i l e s Russell Dechant (1960) (1966) (1969) (1972) (1974) 1. Word per- 1. Word per- 1. Word 1. "Barking 1. Word Recog-ception, ception, recognition at Words" n i t i o n pronuncia-r. form, t i o n , meaning meaning 2. Compre- 2. Comprehen- 2. Associa- 2. L i t e r a l 2. Understand-hension sion of t i o n of comprehen- ing stated or meaning with sion implied printed meaning symbols 3. Reaction 3. C r i t i c a l 3. L i t e r a l 3. Interpre- 3. Reaction and Evalua- and emotion- comprehen- t a t i o n t i o n a l response sion 4. A s s i m i l a - 4. Applica- 4. Interpre- 4. "Shock of 4. Integration t i o n t i o n of t a t i o n r e a c t i o n " ideas to behaviour 5. Evalua-t i o n 6. A s s i m i l a -t i o n 30 Palmer's chart, rearranged In order of time, also demonstrates that recent theories of language are not incorporated. Report 4 of Project Build categorized four l e v e l s of comprehension: l i t e r a l comprehension, i n t e r p r e t a t i o n , c r i t i c a l reading, creative reading (p. 7).. The Secondary G u i d e — E n g l i s h 8-12, l i s t s three: the l i t e r a l l e v e l , the i n f e r e n t i a l l e v e l , and c r i t i c a l reading s k i l l s (p. 10). The l e v e l s of comprehension from selected textbooks i n reading i n the content areas appear i n Table 2.2. TABLE 2.2 LEVELS OF COMPREHENSION Herber Shepherd Estes & Vaughan (1970) (1973) (1978) 1. L i t e r a l 1. L i t e r a l 1. L i t e r a l 2. Interpretive 2. Interpretive 2. I n f e r e n t i a l 3. Applied 3. C r i t i c a l 3. A p p l i c a t i v e 4. Creative Burmeister (1974) distinguished three l e v e l s of comprehension, which she re l a t e d to the seven l e v e l s of the cognitive domain (Table 2.3). TABLE 2.3 LEVELS OF COMPREHENSION AND THE COGNITIVE DOMAIN 1. L i t e r a l 2. Interpretation 3. C r i t i c a l — c r e a t i v e 1. Memory 2. Transl a t i o n 3. Interpretation 4. A p p l i c a t i o n 5. Analysis 6. Synthesis 7. Evaluation 31 Thomas and Robinson (1977) divided twelve s k i l l s into two categor-ie s : The comprehension processes . . . are i n a sense a hierarchy advancing as they do from tasks that involve reading for l i t e r a l meaning, (tasks 1-4 are often l i t e r a l ) i n which students read what i s a c t u a l l y printed i n the l i n e s , to beyond-the-literal (tasks 5-12) responses, i n which they read "between and beyond the l i n e s . " (p. 171) (Table 2.4) TABLE 2.4 TASKS WITHIN LEVELS OF COMPREHENSION L i t e r a l 1. Grasping d i r e c t l y stated d e t a i l s or f a c t s . 2. Understanding main ideas. 3. Grasping the sequence of time, place, ideas, events, or steps. 4. Understanding and following d i r e c t i o n s . "Beyond-the-literal" 5. Grasping implied meanings and drawing inferences. 6. Understanding character (emotional reactions, motives, personal t r a i t s ) and setting s . 7. Sensing r e l a t i o n s h i p s of time, place, cause and e f f e c t , events, and characters. 8. A n t i c i p a t i n g outcomes. 9. Recognizing the author's tone, mood, and intent. 10. Understanding and drawing comparisons and contrasts. 11. Drawing conclusions or making generalizations. 12. Making evaluations. Johnson and Pearson (1975) contended that the notion of an orderly, l o g i c a l progression of s k i l l s i s at best a "pedagogical convenience," that ignores the r o l e of the purpose for reading and the d i f f i c u l t y of the material i n comprehension. A f f e c t i v e components i n reading: i n t e r e s t s , a t t i t u d e s , f e e l i n g s , and emotions are seldom included (Athey, 1970). In a landmark review of research on the comprehension process, Simons (1971) argued: 32 . . . that the s e t t i n g up of categories of s k i l l s has not aided us greatly i n understanding the reading comprehension process because of a basic confusion over the precise behaviour and cognitive domain of these s k i l l s . This had led to the naming of s k i l l s which are global and vague i n nature and which have f a i l e d to d i s t i n g u i s h between: (1) reading and thinking; (2) the objects and processes of comprehension; (3) the use of comprehension, the procedures f or teaching comprehension, and the psychological processes involved i n comprehension, (p. 346) The i n a b i l i t y of the s k i l l s approach to explain the comprehension process i s due i n part to the lack of a t h e o r e t i c a l framework (p. 340). Language i s the other missing element that extends the understanding of reading as a process which i s centred on meaning. The taxonomic models f a i l ". . . t o deal with how language i s structured within d i f f e r e n t modes of discourse" (Palmer, 1979, p. 9). Measurement. The task of i n v e s t i g a t i n g comprehension i s so com-plex (Farr, 1969; Simons, 1971; Bleakley & Johnson, 1978) that a v a r i e t y of measures i s recommended. Norm-referenced standardized t e s t s are s a t i s f a c t o r y screening devices, but lack content and construct v a l i d i t y of s p e c i f i c s k i l l s (Goodman, 1968; Guzak, 1970). A number of informal measures which assess reading performance by sampling behaviour over a number of d i f f e r e n t occasions provide " . . . more r e l i a b l e and more v a l i d measures than the standardized reading t e s t s " (Farr, 1969, p. 98), e s p e c i a l l y to evaluate i n d i v i d u a l student performance. Three recent methods, the cloze procedure, miscue a n a l y s i s , and the Cambourne  Reading Assessment Procedure (1978), a technique which combines the two procedures, are based i n ps y c h o l i n g u i s t i c theory. 33 Psyc h o l i n g u i s t i c Theory Theories from p s y c h o l i n g u i s t i c s , the science which unites cognitive psychology and l i n g u i s t i c s , the study of language, have been applied to reading to explain the process. Goodman (1976c) noted p a r a l l e l s between his research i n o r a l reading and Chomsky's (1957) model of sentence production, which stimulated p s y c h o l i n g u i s t i c research i n speech compre-hension. On the major c r i t i c a l assumption that: "Written text and o r a l speech are merely alternate forms of the same language process" (Cambourne, 1976, p. 609), Goodman (1968) evolved the model of reading as a "psycho-l i n g u i s t i c guessing game.11 "Primary emphasis i s on the psy c h o l i n g u i s t i c steps of the decoding process" (Geyer, 1972, p. 455), that i s , the innate p r e d i s p o s i t i o n f o r language that enables the c h i l d to acquire the gram-matical rules to proceed from the sound system or "surface"structure to the "deep structure of meaning." Shafer (1978) summarized: Stated simply, i n the ps y c h o l i n g u i s t i c model, the reader i s a continual seeker a f t e r meaning. The brain i s constantly going through a decision-making process to decide what's out there i n terms of incoming information and p r i o r expectations, constantly attempting to reduce uncertainty by applying what i s already known from previous experience to each incoming message. (p. 310) The reader uses v i s u a l information from p r i n t , but to a much greater extent, u t i l i z e s h i s non-visual storehouse of knowledge of language and experience with the conventions of the printed page which are stored i n "cognitive structures" i n the brain. The patterns and consistencies of p r i n t , c a l l e d "redundancies," represent a s i g n i f i c a n t l i n g u i s t i c cueing system (p. 312). The other non-visual source i s the semantic system, co n s i s t i n g of a l l previous experience with meaning and knowledge of the world. The fluent reader goes d i r e c t l y to meaning, and makes only minimal use of the t y p i c a l "phonic" s k i l l s . Research i n the reading 34 process has been extensive, based on the premise that s i g n i f i c a n t f a c -tors, such as r h e t o r i c a l organization, influence the reader's cognitive a b i l i t y to process information. Error Studies The use of o r a l reading errors to study the reading process has a long t r a d i t i o n (Weber, 1968; Gibson & Levin, 1975). Based on the assumption that o r a l reading approximates s i l e n t reading, the focus i s on errors as i n d i c a t o r s of unsuccessful learning. Weber, however, found many inadequacies i n the methods. C l a s s i f i c a t i o n schemes often d i f f e r e d so much that comparisons across studies were almost impossible. A l l e r rors, even hesi t a t i o n s and "poor" enunciation were counted as mis-takes. The schemes focused on words or l e t t e r s ; the r o l e of errors i n sentences or discourse was l a r g e l y ignored. Inaccuracies were handled as i s o l a t e d units or as perceptual inaccuracies. No e f f o r t was made to discriminate between kinds of errors. Goodman (1965) studied the r e p e t i t i o n s of 100 grade one to three c h i l d r e n and found that almost a l l were to correct an error. Many sub-j e c t s made substitutions that maintained the meaning and e a s i l y read words i n context that they missed i n l i s t s . Goodman (1969), and l a t e r Weber (1970), demonstrated that errors showed a s e n s i t i v i t y to grammat-i c a l structure. Y. Goodman (1968), noted that as the reader matures, the a b i l i t y to expl o i t the l i n g u i s t i c constraints to maintain meaning increased. Errors were usually the same part of speech as the o r i g i n a l word. The focus of research s h i f t e d from a concern with errors i n the i n d i v i d u a l word to a focus on the syntactic and semantic constraints within the context of the e n t i r e sentence. 35 Mlscue Analysis The Instruments Since then there have been many o r a l reading studies over a wide range of t o p i c s . The two main instruments, the Goodman Taxonomy of  Reading Miscues and the Reading Miscue Inventory (RMI), were developed concurrently by Kenneth Goodman (1969, 1976a, 1976c), i n c o l l a b o r a t i o n with Yetta Goodman and Carolyn Burke (1972a, 1972b). Cambourne (1976) outlined the " n a t u r a l i s t i c " research method Goodman used to b u i l d h i s taxonomy and h i s model of reading. Goodman (1976b) presented the p s y c h o l i n g u i s t i c p o s i t i o n . In recommending i n -depth studies of small numbers, he emphasized: "Real people using r e a l language i n various r e a l s i t u a t i o n s must be the objects of research i f we are to understand reading as i t r e a l l y i s " (p. 98). The miscue technique d i f f e r s from a l l other commonly used diagnostic and evaluative instruments i n o r a l reading. The deviations from the exact responses of the printed text are evaluated f o r the degree to which meaning i s disrupted. The o r a l reading errors are c a l l e d miscues to i n d i c a t e that they are not random, but cued by the same language and thought processes as correct responses (Goodman & Burke, 1972a, p. 5). The taxonomy and inventory were based on the idea that the q u a l i t y of the miscues was more important than the quantity and that some indicated strengths, not weaknesses. Beebe (1976) emphasized: " A l l miscues are not 'equal', because some r e t a i n grammatical and semantic a c c e p t a b i l i t y and therefore detract l i t t l e from t o t a l comprehension, while others d i s t o r t the mean-ing considerably" (p. 55) (Goodman & Burke, 1969; Rode, 1974). 36 Reading Process Page (1970) observed that readers produced a unique form of the reading process with each encounter with material of d i f f e r i n g grade l e v e l . Menosky (1971) concluded that length was a key f a c t o r , since miscues decreased with larger portions of text. The complete text affected the comprehension s i g n i f i c a n t l y , compared to words either i n i s o l a t i o n or i n sentences ( M i l l e r & Isakson, 1976; Vorhaus, 1976). Miscues change q u a l i t a t i v e l y with maturity, becoming ". . . more complex, more productive, more demonstrative of sophisticated processing" (Burke, 1976, p. 41; Y. Goodman, 1976). Differences i n A b i l i t y Studies c o n s i s t e n t l y h i g h l i g h t differences i n reading p r o f i c i e n c y (Brody, 1973; Poole, 1977). Many of the errors of better readers pre-serve meaning and those that make a d i f f e r e n c e are corrected. Poorer readers, overly concerned with "accuracy," are often unaware of nonsen-s i c a l errors (Smith, 1978, p. 235). In research on the cue-testing behaviour of 98 grade eight and nine students, Otto (1977) reported that the primary t e s t i n g of semantics distinguished good readers from poor readers. Content Area Studies Carlson (1970) analyzed the o r a l reading patterns of three boys and three g i r l s i n grade four reading basal readers, science, and s o c i a l studies materials with the Goodman taxonomy. A s h i f t i n emphasis to a greater concentration on s y n t a c t i c cues with content area materials was detected. A q u a l i t a t i v e and quantitative d e s c r i p t i o n of a sample of 47 eighth grade students reading i n both the n a r r a t i v e and expository modes supported Carlson (Brazee, 1976). 37 With both l i t e r a r y and h i s t o r i c a l n a rratives, S t a n s e l l , Harste and De Santi (1978) conducted in-depth i n v e s t i g a t i o n s with RMI of the cue systems u t i l i z e d by three groups of readers: second grade students, ninth grade students, and mature adults past s i x t y . "The studies yielded a view of s t r i k i n g s i m i l a r i t i e s i n cue system u t i l i z a t i o n across wide age v a r i a t i o n s " (p. 27), regardless of the material. In contrast, Kolczynski's (1973) data from RMI revealed that the reading process of 20 average readers or above entering the s i x t h grade remained stable across passages of science, s o c i a l studies, mathematics, and l i t e r a t u r e . Six ninth graders, the subjects of Stansell's (1977) research with the RMI, tended to be unaware of the basic differences between expository and narrative w r i t i n g . Limitations of Miscue Analysis The advantages of i n d i v i d u a l observations by miscue analysis i n research and i n c l i n i c a l diagnosis are countered by some disadvantages for classroom use. Time and cost necessitate group diagnosis methods, e s p e c i a l l y i n secondary schools. Because an experienced c l i n i c i a n i s r e a l l y required, only poor readers are examined and the average readers' weaknesses are often undetected. The observer e f f e c t , even with the highly trained, influences r e s u l t s . On the basis of research with reading s p e c i a l i s t s , Page (1976) recommended that attention be given to the v a r i a b i l i t y factors a f f e c t i n g observers' responses, such as d i a l e c t differences and perceptual problems. Some a u t h o r i t i e s disagree with Goodman's p o s i t i o n on r e p e t i t i o n that " . . . since the r e p e t i t i o n was only made to correct an error, i t should not be counted as an e r r o r " (Ekwall, 1976, p. 267). Ekwall and English (1971) used the polygraph ( l i e detector) to measure the 38 f r u s t r a t i o n l e v e l of students as they read progressively more d i f f i c u l t passages. Other studies by Ekwall, S o l i s and S o l i s (1973) and Ekwall (1974) showed that when a l l r e p e t i t i o n s were not counted as errors, students became p h y s i o l o g i c a l l y upset before they reached t h e i r f r u s t r a -t i o n l e v e l . Ekwall concluded a l l r e p e t i t i o n s must be counted as errors (1976, p. 267). For use at the secondary l e v e l the most serious d i f f i c u l t y i s that the miscue method i s inadequate for the evaluation of s i l e n t reading l a n -guage cues. B l u s t e i n (1977) i n detecting an important d i f f e r e n c e between o r a l and s i l e n t reading, suggested the importance of d i f f e r e n -t i a t i n g between str a t e g i e s used for each. Although miscue analysis illuminates the reading process, i t may not be s e n s i t i v e enough to semantic-syntactic in t e r a c t i o n s that can be accounted for by the unity of thought and language and which are probably e s s e n t i a l to the understanding of reading as reasoning. (p. 1871-A) Theobald (1973), therefore, proposed that the discrepancies be resolved by e s t a b l i s h i n g a method for analyzing the syn t a c t i c and semantic sys-tems by combining the Reading Miscue Inventory and cloze. Cloze Procedure Since Wilson L. Taylor developed the basic concept of the cloze technique i n 1953, i t has been used widely i n research, at f i r s t i n reading comprehension and r e a d a b i l i t y , but l a t e r i n language studies. The name "cloze" i s derived from the Gestalt psychology concept of "closure," the human tendency to complete a f a m i l i a r but incomplete pat-tern by mentally f i l l i n g i n the gaps. Taylor applied the p r i n c i p l e to language: 39 A cloze unit may be defined as: Any s i n g l e occurrence of a successful attempt to reproduce a part deleted from a "message" (any language product) by deciding from the con-text that remains, what the missing part should be. Cloze procedure may be defined as a method of intercepting a mes-sage from a "transmitter" (writer or speaker), mutilating i t s language patterns by d e l e t i n g i t s parts, and so administering i t to " r e c e i v e r s " (readers or l i s t e n e r s ) that t h e i r attempts to make the patterns whole again p o t e n t i a l l y y i e l d s a considerable number of cloze u n i t s (p. 416). Since the cloze procedure was introduced research has focused on many problems. Rupley (1973) surveyed the research l i t e r a t u r e i n the ERIC system i n three broad areas: methodological considerations, cloze as a measuring device, and cloze as a teaching t o o l . The author reported that cloze i s a v a l i d and r e l i a b l e measure of comprehension a b i l i t y . T h e o r e t i c a l Basis From the standpoint of measurement theory, cloze procedures are superior measures of general comprehension. C r i t i c s suggest that a cloze test lacks face v a l i d i t y , since a cloze test appears much l i k e the type of f i l l - i n - t h e - b l a n k exercises i n workbooks. A more serious c r i t -i c ism i s that cloze merely measures general verbal competency. Rankin (1978) presents f i v e reasons for the contention that cloze t e s t s do i n fact measure comprehension more " d i r e c t l y " than conventional t e s t s . F i r s t , cloze tests are i n t r i n s i c measures of the effectiveness of communication by sampling the degree of correspondence between a message source and a receiver. Substantially the same r e s u l t s are obtained whether the scoring i s done by exact word method or by the synonym method. (p. 151) Second, cloze measures comprehension i n process, not as a product a f t e r reading. Third, a l l cloze i s based on the p s y c h o l i n g u i s t i c process of 40 inference which i s i n t r i n s i c to a l l communication. Fourth, the cloze tests sample the choice points for p r e d i c t a b i l i t y within the passage i n a random fashion. F i f t h , cloze items can be p r e c i s e l y r e p l i c a t e d . A number of empirical studies substantiate these claims about r e l i a b i l i t y and v a l i d i t y . R e l i a b i l i t y A cloze t e s t of exact replacements can be made, administered, and scored i n a v a r i e t y of ways. Bormuth (1975) summarized his extensive work on fifth-word d e l e t i o n , the procedure accepted as the standard for r e l i a b i l i t y i n research. Bormuth (1967, 1968) and Rankin and Culhane (1969) established three c r i t e r i o n l e v e l s f o r the exact scoring of cloze responses. A score of- 58-100 percent suggests the student can read the passage independently; a score of between 43 and 57 percent means that the passage i s at the i n s t r u c t i o n a l l e v e l ; and a score below 43 percent i s at the l e v e l of f r u s t r a t i o n . Cloze tests based on exact responses measure " g l o b a l " aspects of comprehension (Hittleman, 1978; Rankin, 1978; Smith-Burke, Gingrich & Eagleeye, 1978). Jones and P i k u l s k i ' s (1974) study suggested that " . . . the cloze test gave a considerably more accurate reading l e v e l placement than did the standardized t e s t " (p. 437). In a study of a l t e r n a t i v e responses, McKenna (1976) demonstrated that the a b i l i t y of seventh graders to produce an exact answer did not vary greatly from the synonymic response. The r e s u l t s from a study of 446 seventh and eighth graders with f i v e cloze tests of f i f t h word del e t i o n patterns, indicated that random d e l e t i o n was more desirable than every f i f t h word, i f only one cloze test i s to be used; however, the greater d i f f i c u l t y of random pattern necessitates new c r i t e r i o n 41 l e v e l s . Bormuth (1971) determined c r i t e r i o n scores for cloze r e a d a b i l i t y tests based on four outcomes: information gain, a student's willingness to study a material, novelty of the material's content, and rate of read-ing. Recent findings suggest that the type of material a f f e c t s the c r i t e r i o n l e v e l . Dupuis (1976) tested the cloze procedure as a pre-d i c t o r of reading success with l i t e r a t u r e f or secondary students. Tenth grade English students read short s t o r i e s with a cloze exercise of stand-ard f i f t h word d e l e t i o n as a pre-test, and a multiple-choice comprehension test as a post-test. P r e d i c t i o n equations were developed to support 48 percent as a cut-off score for minimum comprehension. Dupuis concluded: "That the cloze i s f l e x i b l e , easy to develop and score, and responsible to i n d i v i d u a l differences adds to i t s p o t e n t i a l for matching students to appropriate l i t e r a t u r e " (p. 99). V a l i d i t y P r e d i c t i v e v a l i d i t y . Cloze tests c o r r e l a t e highly (.70 to .85) with standardized tests (Jenkinson, 1957; Ruddell, 1963; Bormuth, 1967). Content v a l i d i t y . In a test of factor v a l i d i t y , the " . . . data were interpreted as providing l i t t l e grounds f o r claiming that cloze tests measure anything other than what has been commonly labeled compre-hension s k i l l s " (Bormuth, 1969, p. 358). Construct v a l i d i t y . From a bibliography of over 600 references c o l l e c t e d over 20 years, Rankin (1978) asserted that the cloze test i s a superior measure of general comprehension, since i t has construct v a l i d i t y based on a p s y c h o l i n g u i s t i c theory of comprehension which makes 42 extensive use of Smith's (1978) concept of information processing and context redundancy. However, Rankin cautions that cloze " . . . does not lend i t s e l f to the measurement of s p e c i f i c a l l y defined language comprehension processes" (p. 152). Limitations of Cloze Hittleman (1978) concurs that cloze tests measure "g l o b a l " aspects of comprehension, but not s p e c i f i c substrategies. "Therefore the test cannot be used for an analysis of s p e c i f i c diagnostic information about a pupil's strategies and s k i l l s " (p. 138). A study by Smith-Burke, Gingrich and Eagleeye (1978) substantiates t h i s view. They examined the e f f e c t of contextual b u i l d up, s t y l e , and d e l e t i o n pattern on cloze comprehension on the basis of the Goodman model. Subjects were 53 eleventh grade and 40 twelfth grade students. In a 3 x 2 x 3 (Context x Style x Deletion) f a c t o r i a l analysis of variance, no s i g n i f i c a n t e f f e c t s of p r i o r context were found. In both narrative and expository s t y l e s , the t r a d i t i o n a l o n e - f i f t h cloze was more d i f f i c u l t than the function word cloze. The score for the narrative s t y l e (M = 54.75) was s i g n i f i c a n t l y greater than that for expository s t y l e (M = 48.43). A further analysis investigated the s t y l e d i f f e r e n c e . The r e s u l t s of 3 x 2 analysis of variance (Content x Style) indicated l e x i c a l items were s i g n i f i c a n t l y more d i f f i c u l t to replace i n narrative material than i n expository material. Function words, such as conjunctions which carry the l o g i c a l argument of text, were s i g n i f i c a n t l y more d i f f i c u l t to replace i n expository material. The authors suggested that d e l e t i o n patterns based on a l i n g u i s t i c model may be more accurate measures of contextual buildup i n connected discourse. 43 In s p i t e of i t s l i m i t a t i o n s , however, the evidence suggests that cloze i s a cognitive, coding task of p r e d i c t i o n from the context. The completion of the cloze gap samples the extent of likeness between the language patterns of w r i t e r and the p o s s i b i l i t i e s which represent what the reader guesses at what he thinks the writer meant (Theobald, 1973, p. 20). Analysis of Cloze Responses The evidence supporting the v a l i d i t y of exact scoring i s not unequivocal. The r e s u l t s of a study by Asher, Hymel and Wigfield (1978) indicated that: "the general bias i n the l i t e r a t u r e against accepting synonyms may be leading to a loss of valuable information when the i n d i v i d u a l rather than the passage i s the unit of a n a l y s i s " (p. 46). Part of the bias against accepting synonyms r e s u l t s from the claims that o b j e c t i v i t y of scoring i s decreased. Various attempts have been made to examine the q u a l i t a t i v e nature of cloze responses. Reading process. Vaughan and Meredith (1978) examined the r e l i -a b i l i t y of the cloze procedure as a measure of semantic awareness i n post-reading s i t u a t i o n s and the r e l i a b i l i t y of cloze scores as an index of students' syn t a c t i c fluency. Subjects were 298 grade eight students, randomly assigned to read two science-related s e l e c t i o n s , at the eighth grade r e a d a b i l i t y l e v e l s . Each blank was examined for exact-replacement, acceptable synonyms, and syntactic a c c e p t a b i l i t y . Synonyms were scored correct i f three of four evaluators agreed. A response was acceptable i n the sy n t a c t i c category i f i t agreed both with the part of speech and the proper use of context of the deleted item. Results indicated that a l l cloze scores are highly r e l i a b l e f o r p r a c t i c a l purposes. Internal consistency c o e f f i c i e n t s d i f f e r e d s i g n i f i c a n t l y from the p a r a l l e l form. Since cloze scores may be somewhat passage dependent . , multiple cloze s e l e c t i o n s should be used for accurate assessment. The research-ers noted the " i n t r i g u i n g prospect" that the s y n t a c t i c a l l y acceptable score " . . . may be the most accurate index of students' comprehension as measured by cloze scores because i t does include both syn t a c t i c and semantic awareness" (p. 179). To ignore s y n t a c t i c a l l y acceptable responses on a cloze test seems tantamount to discarding pertinent information about the students' understanding of the s e l e c t i o n , e s p e c i a l l y i f c e r t a i n psycho-l i n g u i s t i c assumptions are accepted. (p. 179) Zinck (1978) i n a study of 68 seventh grade students reading nar-r a t i v e passages, used three d i f f e r e n t procedures: exact replacement, and adaptations from the Goodman Taxonomy of Reading Miscues (Goodman, 1976b) and the Reading Miscue Inventory (Goodman & Burke, 1972a), to evaluate semantic and syn t a c t i c a c c e p t a b i l i t y . The r e s u l t s supported the use of the cloze procedure with the miscue analysis to evaluate the reading process. . However, i n looking to the future of cloze as a s i l e n t diagnostic t e s t , Zinck c a l l s f o r an examination of the r o l e of r e t e l l i n g of the passage a f t e r completion. Diagnostic Tools Jenkinson (1957), Hafner (1964), Bortnick and Lopardo (1972), and Rankin (1978) r e f e r to cloze for diagnosis. Following RMI procedure, Theobald (1973) constructed p r o f i l e s for the 24 subjects, which did not always match the cloze score or teacher ranking. The patterns, however, showed that good readers had better syntactic and semantic scores and les s loss of meaning than poor readers. Bortnick and Lopardo (1973) stated: " . . . a student's inc o r r e c t cloze responses can y i e l d abundant diagnostic information" (p. 114). 45 However, the scoring procedures lacked o b j e c t i v i t y and were unsystem-a t i c . Myers (1976) asserted: " I t i s clear that more research i n t h i s area w i l l provide teachers with a useful t o o l of informal a n a l y s i s " (p. 12). Cambourne (1977) devised a s i l e n t reading version of the RMI (Goodman & Burke, 1972a), on the assumption that the process of reading i n t a c t and mutilated text i s s i m i l a r . In a p i l o t study with a sample of 39 subjects, years 3 to 7, chosen by t h e i r teachers, the method i l l u s t r a t e d that the s i l e n t reading behaviour of above average readers appeared to be quite d i f f e r e n t from that of below average readers. Pro-f i c i e n t readers c o n t r o l l e d meaning at the whole story l e v e l by scanning both forward and back. Poorer readers r e s t r i c t e d focus to a much smaller unit of meaning, usually p r i o r to the cloze d e l e t i o n . Since knowledge and s k i l l i n the graphophonic domain played a minimal r o l e i n accounting for d i f f e r e n t reading a b i l i t y , the category was l a t e r omitted. Vaughan, Tierney, and Alpert (1977) examined the r e l a t i o n s h i p between the s y n t a c t i c a l l y and semantically acceptable responses on cloze tasks by 240 students i n grades 4, 5, 6, 8, and 11, i d e n t i f i e d as good, average, or.poor readers on the basis of reading t e s t scores. Each subject read two selections at a r e a d a b i l i t y l e v e l of t h e i r grade and responded to a cloze t e s t . Responses were scored for semantic and syn-t a c t i c appropriateness. Semantic a c c e p t a b i l i t y was based on three c a t e g o r i c a l d e f i n i t i o n s : exact replacement, synonym, or minimal change. The meaning, number, tense, and gender of the synonym had to match the deleted word. Sl i g h t s h i f t s i n meaning, through connotation, for example, were allowed. Interjudge r e l i a b i l i t y with three assessors was about 90 percent agreement. The authors concluded: "In terms of 46 assessing observed responses i n reading comprehension v i a cloze t e s t s , a meaningful r e l a t i o n s h i p exists between syntactic and semantic elements-' (p. 202). In summary, the review of research from error studies, miscue analysis, and cloze procedure indicates that the q u a l i t a t i v e analysis of cloze non-exact-replacements by a miscue inventory i s a f e a s i b l e pro-cedure to investigate responses to the word and sentence l e v e l s . The next section reviews the l i t e r a t u r e i n discourse analysis to extend the in v e s t i g a t i o n to the whole l e v e l of discourse, congruent with the purpose of the study: to examine the.comprehension process by comparing responses to two types of materials: narrative f i c t i o n and expository prose. Discourse Analysis Since the term discourse analysis can be used to designate both the cognitive structure of the language user and the conventions of the form of the discourse i t s e l f , one r e f l e c t s the psychological approach; the other, the l i n g u i s t i c or text-based. Recently writers i n psycholin-g u i s t i e s have addressed the i d e n t i f i c a t i o n of the processes involved i n the comprehension of en t i r e discourses. Strategy research associated with miscue analysis or cloze has focused on the l e v e l of the sentence. Now from l i n g u i s t i c s , cognitive psychology, and the f i e l d of a r t i f i c i a l i n t e l l i g e n c e , have come many studies, and although, as Winograd (1977) observes, at th i s stage i n the f i e l d i t i s not " . . . possible to lay out a precise u n i f y i n g theory . . ." (p. 63), i n t e r r e l a t i o n s h i p s are evident which provide twin frameworks for viewing discourse and methods for studying s t r a t e g i e s . The term "discourse" has been used to cover many only loosely 47 related problems. Winograd l i s t s over 60 technical terms, such as context, text, macro-structure, staging, template,-rhetoric, text grammar, textual coherence, theme, story grammar, frame, foreground, macro-rule system, s c r i p t , plan, d e f i n i t i o n . A c e n t r a l theme i n a l l the work i s the postu-l a t i o n of a pattern of organization that can be i d e n t i f i e d and explored and, of a l l the terms proposed, one from psychology seems p a r t i c u l a r l y pervasive: "schema" or p l u r a l "schemata." Introduced by Piaget (1926) and B a r t l e t t (1932), the term schema refe r s to the knowledge that i s "incorporated i n abstract structures that have c e r t a i n properties" (Ander-son, 1977, p. 67). In the study of the comprehension process the cogni-t i v e approach has p a r t i c u l a r u t i l i t y . Theories of Discourse Cognitive Approach Anderson (1977), Rumelhart (1977), Winograd (1977), and others view the problem of studying discourse as one of understanding the cognitive processes of production and comprehension. In language communication there are three independent sets of structures: the cognitive structures of the "producer," whether writer or speaker; those of the "comprehen-der," the reader or l i s t e n e r ; and the text i t s e l f . In each p a r t i c i p a n t , the form the discourse takes depends on h i s knowledge base of language, as w e l l as the broader cognitive processes, which include the perception and i n t e r p r e t a t i o n of the contextual s i t u a t i o n , the formation of goals, and expectations. The i n t e r a c t i n g knowledge structures or schemata are guides i n a process of "pattern.recognition." The comprehension process i s oriented towards i d e n t i f y i n g known discourse and reasoning patterns. Some feature of an utterance t r i g g e r s a hypothesis or p r e d i c t i o n . 48 Kintsch (1977) c a l l e d t h i s use of schema a "top-down" process. It follows that the more f a m i l i a r the patterns of the discourse, the greater the ease of comprehension. Patterns of Discourse Rhetorical schemas. Olson (1977a) cogently argued that the con-ventionalized, e x p l i c i t l o g i c a l patterns of text^book prose or the "language of schooling" are very d i f f i c u l t f o r many students. Text-books may have c o n t r o l l e d l e v e l s of d i f f i c u l t y , appropriate a c t i v i t i e s , and subject matter and even workbooks, and s t i l l f a i l , because the con-ventions employed and the meaning s p e c i f i e d are lodged i n the text alone. The conventions for laying out a reasoning sequence are c a l l e d r h e t o r i c a l schemas (Winograd, 1977, p. 83). There ris?. a number of language devices to d i s t i n g u i s h the patterns of reasoning. Some are s i g n a l l e d d i r e c t l y by function words that i n d i c a t e sequence such as " f i r s t " and "second"; c a u s a l i t y , "because"; summation, " i n conclusion"; change of ideas, "however." In addition, there are conventional schemas for organizing arguments and exposition, such as enumeration, c l a s s i f i c -a tion, problem s o l u t i o n , comparison, cause-effect, but i n each d i s c i p l i n e these may be used i n a d i f f e r e n t way (Robinson, 1975). Grimes (1975) noted that informational discourse, which has as i t s c e n t r a l function explanation, often involves premises that the writer f e e l s are generally accepted and, therefore, leaves unsaid.' A few key arguments are expected to a c t i v a t e a whole l o g i c a l structure. Such omissions cause comprehension problems, e s p e c i a l l y i n poor readers. P e r f e t t i and Lesgold (1977) postulated that i n d i v i d u a l d ifferences i n the use of discourse structure can mean a d i f f e r e n t s e n s i t i v i t y to ". . . information structure, to thematization, and to clause s t r u c t u r e " 49 (p. 149). Although the r e s u l t s of the empirical studies with c h i l d r e n and adults were inconclusive, s p e c i f i c a l l y about causation, the authors advocated further elaboration of the component processes of discourse comprehension and of i n d i v i d u a l differences (p. 179). I t appeared also that highly thematized elements are more quickly and automatically pro-cessed by a reader i n n a r r a t i v e . Narrative schemas. There i s a set of schemas for r e l a t i n g nar-r a t i v e s of events, either from the memory of a speaker or as a story: time sequence, c a u s a l i t y , plans, action, form the organization of s t o r i e s . Rumelhart (1975) developed a "story grammar" which divides a story into a l i n e a r sequence of episodes, states, and events which he c a l l e d problem-solving episodes. There are two schemata: one, the EPISODE, s p e c i f i e s the r e l a t i o n s h i p s among the i n i t i a t i n g event, the goal, and the attempt. EPISODE ABOUT PROTAGONIST P_ (1) EVENT E CAUSES P TO DESIRE GOAL G (2) P TRIES TO GET G UNTIL OUTCOME 0 OCCURS The other schemata s p e c i f i e s the structure of the attempt or TRY: AGENT A TRIES TO GET GOAL G (1) A SELECTS A METHOD M WHICH COULD LEAD TO G (2) FOR EACH PRECONDITION P OF M A TRIES TO GET P UNTIL OUTCOME 0 (3) A DOES M WHICH HAS CONSEQUENCE C. (p. 270) Grimes (1975) and Winograd (1977) considered that, f o r a l l but the simplest n a r r a t i v e , there are interwoven themes and changes of scene, so that any piece of narrative text i s the "product of i n t e r a c t i n g struc-tures" (p. 83), some dealing with time flow, some with c a u s a l i t y , and others with s p e c i f i c story conventions. Well-known to many are the schemas for the t e l e v i s i o n western, the mystery story, the f a i r y t a l e , so that the phrase "Once upon a time" evokes an immediate response, 50 whereas the schema of a s c i e n t i f i c report i s f a m i l i a r only to the i n i t i -ated. Some schema are l i t e r a r y conventions such as s e t t i n g , character, p l o t , development, c o n f l i c t , or t h e s i s , a n t i t h e s i s , and synthesis. Other schema cut across these general conventions, such as the flashback or suspense; these are highly culture-dependent (Kintsch & Greene, 1978). Within a language or culture are s p e c i f i c conventions for e s t a b l i s h i n g the point of view such as: the omniscient narrator, f i r s t person, or the dramatic. In Van Dijk's (1977) view, such semantic structures are the "macro-rules" which underlie the global i n t e r p r e t a t i o n and meaning of the discourse that cannot be accounted for by a simple summation of i s o l a t e d sentences or "micro-structures." Setting, or premise and"conclusion, are macro-level operations; whereas the use of pronouns, for example, are at the sentence l e v e l . Comprehension then takes place at several l e v e l s . To Broudy (1977) a pattern for construing the import and relevance of the constituent elements i s the context* Propositions have " l e x i c a l " meaning, but without the clue to context, the meaning can be equivocal and ambiguous. Context can be cognitive, a f f e c t i v e , aesthetic, moral, s o c i a l , r e l i g i o u s . Within each of these types, contexts can be thought of as more or l e s s precise, c l e a r , r e f i n e d , and 'educated,' that i s formulated i n the categories of an academic d i s c i p l i n e . This l a t t e r d i s t i n c t i o n i s of the utmost importance for formal school-ing, because the s o c i a l m i l i e u furnishes commonsense contexts for which l i t t l e or no formal t u i t i o n i s necessary. (p. 13) The d i f f i c u l t y of comprehension of a long text or discourse i s determined not only by the l o c a l e f f e c t s of the i n d i v i d u a l sentences or paragraphs, but also by the o v e r a l l organization of the text. Certain types of discourse have a conventional structure and i t has been argued that a knowledge of these aids i n comprehension (Kintsch & Greene, 51 1978, p. 1). Olson (1977b) distinguished between utterances, which include speech, poetry, and l i t e r a t u r e , and prose text. The meaning of conven-t i o n a l utterances l i e s i n shared commonsense knowledge and common i n t u i -t i o n . "Statements match, i n an often t a n t a l i z i n g way, the expectations and experiences of the l i s t e n e r " (p. 277). Prose text, however," appeals to the rules of l o g i c and the c r i t e r i o n f o r success i s the correctness of i t s formal structure. F a i l u r e to comprehend i s the reader's problem, since the meaning resides i n the text. From a number of points of view of discourse organization: schemata, macro-structures, context, i t appears a l o g i c a l assumption that the a b i l i t y to express the bases for the choices of comprehension s t r a t -egies are greater i n number for narrative f i c t i o n than for the l e s s f a m i l i a r r h e t o r i c a l patterns of expository prose. There remains the problem of i n v e s t i g a t i n g the possible reasons f o r the diff e r e n c e s . Methodology Pearson (1978) affirmed that the d i f f i c u l t task of the study of meaning i s complicated by a lack of good tools f o r prose-related research. A t r a d i t i o n of strategy research, based on responses, has evolved i n cognitive psychology. Free r e c a l l protocol and protocol analysis have been adapted to discourse research; introspection-retrospection, to many aspects of the reading process. Protocols There are two kinds of protocols: free r e c a l l and protocol anal-y s i s . In free r e c a l l d e s c r i p t i o n , the subject writes down everything he can remember about a passage. From the cognitive view of schemata. 52 Kintsch and Greene (1978) found that the errors of college students i n r e c a l l summaries, both d i s t o r t i o n s and omissions, occurred at points i n an Indian story where the text deviated from expectations based on a western story schema. The authors concluded: "The pattern of these r e s u l t s indicates the dependence of story comprehension and r e c a l l on the r i g h t schema" (p. 12). Mandler (1978) examined developmental d i f -ferences i n the use of story schemata with grade two, four, s i x , and college subjects. Subjects were tested on four normal s t o r i e s and four interleaved s t o r i e s , i n which the same episodes were rearranged. The tape-recorded r e s u l t s suggested quantitative d i f f e r e n c e s , but few q u a l i -t a t i v e differences i n the r e c a l l of well-formed s t o r i e s . The data also contrasted "rather sharply" with the many studies of children's r e c a l l of l i s t s of words. The use of a f a m i l i a r structure to access connected discourse was "dramatically" evident and, even i n the i r r e g u l a r material, the input was organized i n terms of a canonical story structure. Marshall and Glock (1978-79), designed a study to determine the e f f e c t s of varying c e r t a i n aspects of d e s c r i p t i v e text upon the structure and content of the written r e c a l l s of college students. The r e s u l t s i d e n t i f i e d two d i s t i n c t populations: the t r u l y fluent reader whose com-prehension was unaffected by the manipulations i n the text, and not-so-fluent reader who required e x p l i c i t text references i n the discourse and was unable to make inferences about meaning. Bridge, Tierney and Cera (1978) found s i m i l a r differences i n a t h i r d grade population who read a non-narrative informational passage. Pearson (1978), i n summarizing other discourse studies using r e -c a l l , i d e n t i f i e d the d i f f i c u l t i e s of inter-judge r e l i a b i l i t y r e s u l t i n g from a lack of agreement on the meaning ascribed to a s p e c i f i c text and 53 the use of d i f f e r e n t representations for the same meaning. An examina-t i o n of discourse l i t e r a t u r e suggests a plethora of schematic representa-tions of discourse structures from studies i n problem solving behaviour. Olshavsky (1976-77) adapted protocol analysis to examine the com-prehension s t r a t e g i e s of tenth grade subjects on a short story, on the basis that objective data analysis i s possible, i f the categories and processes are determined from the data, rather than imposed on the data. Each subject was required to "think aloud" a f t e r reading each clause of the story. Strategy usage was r e l a t e d to three f a c t o r s : reader pro-f i c i e n c y , high and low; i n t e r e s t , high and low; and w r i t i n g s t y l e , concrete and abstract. A l l readers used the same s t r a t e g i e s , but readers with high i n t e r e s t , readers with abstract material, and good readers, used c e r t a i n strategies s i g n i f i c a n t l y more frequently. Ten strategies were i d e n t i f i e d : three pertaining to the word l e v e l ; s i x to the clause l e v e l , but only one to the whole story l e v e l . In an exploratory study, Olshavsky (1978) investigated the hypo-thesis that good twelfth grade readers would use s t r a t e g i e s more often than poor readers as the material became more d i f f i c u l t . An analysis of variance showed no s i g n i f i c a n t d ifferences for p r o f i c i e n c y on eleven strategies i n f e r r e d from the protocols. The t o t a l number of strategies used by both groups decreased as the s t o r i e s increased i n d i f f i c u l t y . The most frequent strategy, inference, r e l a t e d to added i n t e r p r e t a t i o n of both clause and story. However, for cloze deleted passages of t h i s study, neither protocol method i s s u i t a b l e since the basis of protocol analysis i s the subjects' comments about the content and problems of short segments of reading and, for free r e c a l l , of an e n t i r e passage. 54 Introspection-Retrospection The i n t r o s p e c t i v e - r e t r o s p e c t i v e technique has been used to study the reading process since the turn of the century. Introspection, as used by Huey (1901), requires the subject to report h i s process as i t occurs; retrospection requires s i m i l a r responses a f t e r the en t i r e s e l -ection i s read. Despite other early investigations with reading (McCallister, 1930; Pickford, 1933) and the success of Piaget i n c h i l d psychology (1928), Harker (1974) reported that the majority of studies employing the technique were conducted i n the l a s t two decades, c o i n c i d -ing with a renewed i n t e r e s t i n the process of learning. The subjects of Swain's (1953) study, good, average, and poor c o l -lege students, a l l above average i n i n t e l l i g e n c e , were required to "think aloud" while reading a c a r e f u l l y structured problem-oriented reading t e s t . Results suggested that good readers focused on determin-ing the t o t a l meaning of the passage by attention to the language clues of the whole context. Pierkarz (1954), i n observing s i x t h grade sub-j e c t s , confirmed that good readers manipulated various clues to meaning as they made many tent a t i v e attempts to derive the t o t a l meaning of the passages. In both studies the data were analyzed according to a framework postulated i n advance. Several other studies employed the in t r o s p e c t i v e - r e t r o s p e c t i v e technique with the short story. Strang and Rogers (1965), i n comparing the responses of h i g h - l e v e l and low-level high school readers, found marked i n d i v i d u a l differences within, as we l l as between, groups. Both groups integrated new ideas and modified views and attitudes a f t e r read-ing, but h i g h - l e v e l readers exhibited a greater v a r i e t y of approaches to 55 understanding the s t o r i e s , had a greater a b i l i t y to d i s t i n g u i s h between l i t e r a l and implied meanings, and remained a l e r t to symbolism. Squire; (1964) concluded from a study of ninth and tenth grade students that poorer readers concentrated on the l i t e r a l meaning and f a i l e d to make the appropriate inferences. Rogers (1960) noted the d i f f i c u l t i e s of the i n t r o s p e c t i o n technique: the same text may not constitute the same problem for each reader; sub-j e c t s may not be aware of t h e i r mental processes, and d i f f e r i n the extent they possess relevant information and i n the e f f o r t they are w i l l i n g to make. Rogers also cautioned that responses must be undirected and unguided to be legitimate "process" data. . \ Directed questions d i s t o r t e d p i l o t r e s u l t s . In s p i t e of i t s l i m i t a t i o n s , the method revealed that the "high school subjects were passive readers, u n c r i t i c a l and unaware of implications i n a biased a r t i c l e . Fareed (1971) studied the retrospective verbalized responses of s i x t h grade pupils who read passages of h i s t o r y and biology. He found the process v a r i e d according to the type of material. The reading of h i s t o r y was characterized by more emotive reactions, more r a t i o n a l evolu-t i o n s , and a tendency to summarize. Reading b i o l o g i c a l content involved more i l l u s t r a t i o n s and expansions, and reconstructions of sensory images. Fareed concluded that the r e t r o s p e c t i v e method seemed to y i e l d further evidence of i t s value i n i n v e s t i g a t i n g the reading process. Two studies combined retrospection with cloze. Jenkinson (1957), i n a frequently c i t e d study, explored the product and processes of the comprehension of grade ten, eleven, and twelve students. Subjects were asked to r e c a l l why they selected cloze responses to a comprehension test of l i t e r a r y material. The categories f o r c l a s s i f i c a t i o n were formulated 56 from the responses. Jenkinson's f i n d i n g , i n support of Swain, indicated that better readers exhibited a greater knowledge of language structure and word meanings and were concerned with the i n t e r r e l a t i o n s h i p s and sequence of ideas. Laing (1974) used a s i m i l a r technique i n a developmental study of the context s t r a t e g i e s of grade 4, 6, and 8 subjects on cloze passages, as did Ames (1966) and Rankin and Overholser (1969). The conclusions of Harker (1974) are relevant. S t r i k i n g s i m i l a r -i t i e s existed between the nature of the reading process, as portrayed by the i n t r o s p e c t i v e - r e t r o s p e c t i v e case studies and the p s y c h o l i n g u i s t i c model of reading: comprehension i s represented as a dynamic complex process i n which the reader i s a c t i v e l y searching for meaning, the s e l e c -t i o n process depending to a great degree on the appropriate language clues. Harker concluded that the case-studies ". . . provide the inves-t i g a t o r with a more d i r e c t access to the covert processes of the reading process than does any other method currently a v a i l a b l e " (p. 93). Attitude "Attitude i s defined as the tendency to respond i n a favorable or unfavorable manner to s p e c i f i c questions about reading" (Summers, 1979). The importance of a t t i t u d e has been emphasized. Alexander and F i l l e r (1976) stated: Today's teacher of reading cannot a f f o r d to ignore the at t i t u d e s of h i s students since attitudes are important i n the a c q u i s i t i o n of reading and i n the continued use of reading for information and recreation. (p. 19) The s e l e c t i o n of the Estes Reading A t t i t u d e Scale (ERAS) "(Estes et a l . , 1975), was based on an extensive review of a t t i t u d e measures by Summers (1977). Estes (1971, 1972, 1975) reported the research r e l a t e d 57 to the development of the Estes Attitude Scales to Measure Attitudes  Toward School Subjects, including English, mathematics, reading, science, and s o c i a l studies. The scale consists of 15 L i k e r t or "summated r a t i n g s " items i n each subject ( t o t a l 75); responses are on a 5-point scale from "strongly agree" to "strongly disagree." Scores range from 15 to 75 with an approximate midpoint of 45, with a high score represent-ing a p o s i t i v e a t t i t u d e . It i s designed f o r grades three through twelve. The items were selected from a pool of statements contributed by elemen-tary and secondary teachers and sampled on a wide v a r i e t y of a b i l i t y l e v e l s . The v a l i d i t y i s well substantiated by Dulin and Chester (1974), Greene and Z i r k e l (1976), and Summers (1977), who stated that the Estes et a l . (1975), ". . . i s t e c h n i c a l l y and conceptually, the best developed reading a t t i t u d e scale to date" (p. 152). Summers, i n an undated study, presented data on the i n t e r n a l consistency and convergent v a l i d i t y of the Estes Reading A t t i t u d e Scale for intermediate grades, based on the responses of 1,403 c h i l d r e n i n grades 5, 6 and 7. It was concluded that the ERAS " i s useful i n obtaining a global r e a c t i v e assessment of school oriented a t t i t u d e toward reading i n the intermediate grades" (p. 8). On the basis of the study, Summers (1979) employed the ERAS i n an evaluation study of the e f f e c t s of a program of sustained s i l e n t reading (SSR) on reading achievement and a t t i t u d e . Roettger (1980) used the scale i n a study of 75 elementary students' at t i t u d e s toward reading. The r e s u l t s contradicted the b e l i e f that p o s i t i v e a t t i t u d e scores p a r a l l e l achievement scores. Oral responses showed d i f f e r e n t expectations of reading. High performance/low a t t i t u d e students viewed reading as a t o o l f o r s u r v i v a l and for success i n school. 58 Low performance/high a t t i t u d e students associated reading with a good self-concept and used reading for s p e c i a l i z e d i n t e r e s t s (p. 452). Summary The review of l i t e r a t u r e substantiated the problem of the assess-ment of reading comprehension i n the content areas i n secondary school. The inadequacy of sole dependence on the s k i l l s model and the d e s i r a b i l i t y of employing multiple measurement techniques were noted. The psycho-l i n g u i s t i c r a t i o n a l e f o r using the responses of students provides a promising approach. Methods of the assessment of o r a l responses by errors and by miscue analysis led:to the adaptation of techniques for the s i l e n t reading task. Scores of both exact replacement and non-exact-replacement cloze deletions are used to evaluate s i l e n t reading responses i n terms of the cue systems i n language to the l e v e l of the sentence. With the a p p l i c a t i o n of discourse theory, the analysis of responses may be extended to the whole organization of the passage. Retrospection research indicates another way of understanding how students comprehend both f i c t i o n and prose at the secondary l e v e l . CHAPTER III METHODOLOGY The design, which r e f l e c t s a d e s c r i p t i v e f i e l d study, was congru-ent with the main purpose of i n v e s t i g a t i n g the reading comprehension process by examining the responses of secondary school readers, at grade nine and grade twelve, i n r e l a t i o n to two f a c t o r s : l e v e l of pro-f i c i e n c y and mode of discourse. To make possible the exploration the i n v e s t i g a t i o n proceeded i n three stages, based on (1) the exact replace-ments of cloze responses; (2) the non-exact-replacements of cloze responses; and (3) the retrospective analysis of exact and non-exact-replacements of cloze responses. The purpose of t h i s chapter i s to discuss research design, proce-dures f o r s e l e c t i o n of the population sample, instrumentation, c o l l e c -t i o n of experimental data, and treatment of the data. Descriptions of the interviewing techniques and procedures are also included. The chapter concludes with the hypotheses tested. Research Design The research design was a 2 x 2 f a c t o r i a l design. Factor A: 2 modes of discourse, n a r r a t i v e f i c t i o n and expository prose. Fixed E f f e c t . Factor B: 2 l e v e l s of p r o f i c i e n c y , good and poor. Fixed E f f e c t . 59 60 Schematically, the design can be shown as follows: Mode of discourse (Factor A). 1 2 Narrative Expository P r o f i c i e n c y 1. Good (Factor B) Readers 2. Poor Readers The Population Sample  Selection of the Population Sample Since the major purpose of the study was to explore how secondary school students at d i f f e r e n t l e v e l s of maturity respond to d i f f e r e n t modes of discourse, two modes, narrative f i c t i o n and expository prose, and two grade l e v e l s , nine and twelve, were chosen. It was considered probable that patterns of performance might be revealed, according to the degree of experience with eacW-imode, i n p a r t i c u l a r expository prose, a f t e r one year i n secondary school at the ninth grade, and a f t e r four years at the twelfth grade. Since the study i s an in-depth i n v e s t i g a -t i o n , only subjects from Lord Byng Secondary School, Vancouver, B.C. were selected. From t h i s population three samples were drawn. Selection of Sample 1 To investigate exact cloze replacements, the e n t i r e population of each grade, N = 144 at grade 9 and N = 128 at grade 12, was tested. A subject was e l i g i b l e f o r the next phase of the study only i f English was the f i r s t language spoken i n the home, to eliminate possible second language interference i n the process of responding to cloze d e l e t i o n s . On that basis, the i n i t i a l sample was reduced to N = 107 i n grade 9 and 61 N = 100 i n grade 12. Exact cloze and a t t i t u d e scores only were inves-tigated. Selection Of the C r i t e r i o n Since the second major purpose was to explore the. performance of students at two l e v e l s of p r o f i c i e n c y , good and. poor, a further sample was drawn. The 80 subjects for d e t a i l e d study (40 at each grade l e v e l ) of both exact and non-exact-replacements of cloze deletions were selected on the basis of t h e i r performance on the screening t e s t , the Iowa S i l e n t  Reading Test (ISRT), administered by t h i s i n v e s t i g a t o r during regular English periods and scored by computer services. English 9, a Resource  Book for Teachers, states of the 1943 e d i t i o n of the test that: "These tests cover grades 9 through 13 and are useful for the diagnosis of i n d i v i d u a l a b i l i t i e s as they can be administered to a class i n a f o r t y -f i v e minute period" (p. 60). Level I I , Form E of the 1972 e d i t i o n was used because i t " r e f l e c t s the philosophy that guided e a r l i e r e d i t i o n s : that s i l e n t reading i s a multifaceted a b i l i t y best measured by tests that survey several reading behaviours" (p. 4). The tests measure f i v e areas: vocabulary, reading comprehension, use of reference materials, skimming and scanning for s p e c i f i c information, and speed of reading with comprehension. The combined score (50) from Comprehension Subtests A and B only had r e l e -vance to the study. The ISRT had been used i n the school i n the past and was a v a i l a b l e from Vancouver School Board. Selection of Sample 2 At each grade l e v e l , 9 and 12, poor readers (N = 20) were selected randomly from those students whose scores ranked between the 5th 62 pe r c e n t i l e and the 33rd p e r c e n t i l e on n a t i o n a l norms. At each grade l e v e l good readers (N = 20) were i d e n t i f i e d from those students whose scores ranked between the 65th p e r c e n t i l e and the 95th p e r c e n t i l e . The p e r c e n t i l e ranks corresponded to the ISRT norms f o r each grade. Selection of Sample 3 For the i n v e s t i g a t i o n of how students respond to cloze d e l e t i o n s , sample 3 was selected f o r the retrospective interviews of "loud-thinkers" (Cambourne, 1979). Six good readers and s i x poor readers were chosen randomly at each grade l e v e l (N = 12) from sample 2, then assigned randomly to the mode of discourse: narrative f i c t i o n and expository prose. Instrumentation Materials The research passages selected for cloze deletions were reproduced from tests used for B r i t i s h Columbia Reading Assessment, Grades 8 and 12, 1977. Grade 9 was tested with: a) "A Kind of Courage"—Narrative F i c t i o n b) "Australian C i t i e s " — E x p o s i t o r y Prose According to Report 1: Tests Results, both passages have a r e a d a b i l i t y l e v e l of approximately grade 8 established by the Fry Readability Formula and were written by Canadian authors. ' Grade 12 was tested with: a) "David Comes Home"—Narrative F i c t i o n b) " C u l t i v a t i o n of the Sea and i t s Present E x p l o i t a t i o n " — Expository Prose 63 Both passages have a grade 12 r e a d a b i l i t y l e v e l , as determined by the Fry Readability Scale (1976). Cloze Passages The test passages were typed according to the standard format with f i f t h word deletions (Bormuth, 1976, pp. 68-69). The 50 deletions were selected from the 250 words i n the f i n a l portion of each s e l e c t i o n at a point judged s u i t a b l e by t h i s i n v e s t i g a t o r . A ttitude Inventory Since an a n c i l l a r y purpose was to explore r o l e of a t t i t u d e i n r e l a t i o n to mode of discourse, the Estes Reading Attitude Scale was used. Procedures Testing Test booklets included: cover page, the a t t i t u d e scale, i n s t r u c -tions and sample test (Bormuth, 1976, p. 70), and the cloze t e s t s (Appendix A). The c o l l e c t i o n of the data was conducted i n three stages. In the f i r s t stage, the en t i r e population of each grade was administered the ISRT i n regular English classes by t h i s i n v e s t i g a t o r . In the second stage, during the same teaching block one week l a t e r , both research passages, n a r r a t i v e f i c t i o n and expository prose, a l t e r n a t i v e l y placed for order e f f e c t , were randomly assigned to a l l students. In the following week the c o l l e c t i o n of the interview data from the sample subjects completed the t h i r d stage. Interviews A l l interviews were conducted by t h i s i n v e s t i g a t o r i n pri v a t e appointments. Each subject was t o l d that he was part of a study on how 64 students read, and then instructed to read each sentence, and to t e l l how he completed each cloze d e l e t i o n on h i s unmarked t e s t . Responses from 20 subjects were taped on a Sony cassette recorder and transcribed by t h i s i n v e s t i g a t o r as soon as possible a f t e r the interview (Appendix B ) . Data Processing S t a t i s t i c a l procedures and post hoc examinations of the r e t r o -spective v e r b a l i z a t i o n s were used i n the analysis of the data from exact replacements of cloze and non-exact-replacements of cloze responses. Exact Cloze Scores S t a t i s t i c a l Procedures 1. C o r r e l a t i o n . A t titude scores, exact cloze scores on narrative f i c t i o n and on expository prose were correlated with scores on IRST, using the Pearson product-moment c o r r e l a t i o n c o e f f i c i e n t , N = 107 i n grade 9 and N = 100 i n grade 12. For good (N = 20) and poor readers (N = 20) at each grade l e v e l , c o r r e l a t i o n s with a t t i t u d e and ISRT were calculated on exact cloze scores, and on exact cloze plus complete d i s -course a c c e p t a b i l i t y scores. The percent of exact scores was corre-lated with the percent of complete discourse a c c e p t a b i l i t y scores. 2. Comparison of means. For exact cloze scores on each mode of discourse, means and standard deviations were computed for N = 107, grade 9 and N = 100, grade 12 and t values calculated for each pair of means. 3. Analysis of variance. Means of exact cloze scores of the s t r a t i f i e d random sample, N = 40 at each grade l e v e l , were examined by the analysis of variance (Table 3.1) based on the 2 x 2 f a c t o r i a l , completely randomized design, fixed e f f e c t s model (Kerlinger, 1973). 65 TABLE 3.1 SOURCES OF VARIANCE AND DEGREES OF FREEDOM FOR A 2 x 2 FACTORIAL RANDOMIZED DESIGN, FIXED EFFECTS MODEL Source df Modes of Discourse (Ai,A2) (p - 1) = 1 Levels of P r o f i c i e n c y (Bi,B2) (q - 1) = 1 Interaction (A,B) (p - 1)(q - 1) = 1 Within C e l l s N - pq = N-4 To t a l N-l Non-exact Cloze Replacements Each non-exact cloze replacement.was analyzed according to the four categories of the investigator's adaptation of the Cambourne Reading Assess- ment Procedure (C.R.A.P., 1978): grammatical function, syntactic accept-a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y , on each mode of discourse: n a r r a t i v e f i c t i o n and expository prose (Appendix C). Weighting system. The mean and standard deviation f or each cate-gory were calculated from the score obtained when the weighted score was divided b y the number of possible non-exact-replacements, including omissions, i . e . , the diffe r e n c e between the number of exact replacements and the t o t a l possible (50). For example, 2 x Yes + 1 x P a r t i a l + 0 x No non-exact-replacements For grammatical function category (Table 3.2) the weighted score of one subject was: 2 x 18 + 1 x 0 + 0 x 1 . — = 1.89 (Appendix C) 19 The equality of units f or an i n t e r v a l scale was thus assumed. 66 TABLE 3.2 WEIGHTING SYSTEM FOR FOUR VARIABLES OF NON-EXACT CLOZE REPLACEMENTS Grammatical Syntactic Semantic Discourse Function A c c e p t a b i l i t y A c c e p t a b i l i t y A c c e p t a b i l i t y 2 Yes 2 Yes 4 Yes 2 Yes 1 P a r t i a l ' 1 P a r t i a l 3 T-unit 1 P a r t i a l 0 No 0 No 2 P r i o r 0 No Cloze 1 Past Cloze 0 No S t a t i s t i c a l Procedures 1. Comparison of means. To determine the presence of s t a t i s -t i c a l l y s i g n i f i c a n t d i f f e r e n c e s , t values were calculated at each grade l e v e l : (a) between means of non-exact-replacements of cloze responses on modes of discourse, narrative f i c t i o n and expository prose, f o r good and poor readers; (b) between means of weighted scores, according to grammatical function, s y n t a c t i c a c c e p t a b i l i t y , semantic acceptability,' -Land discourse a c c e p t a b i l i t y , for good and poor readers on nar r a t i v e f i c -t i o n and expository prose. 2. Analysis of variance. To consider the dif f e r e n c e among means of the summated exact cloze and complete discourse a c c e p t a b i l i t y ( i . e . , acceptable on three v a r i a b l e s : syntactic a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y ) , analysis of variance was used, based on the 2 x 2 f a c t o r i a l design of the study, at each grade l e v e l , nine and twelve. The l e v e l of r e j e c t i o n for the n u l l hypotheses was p < •. 05. " i - — 67 R e l i a b i l i t y of the N.E.R. Scoring Four judges, three doctoral students i n reading education and the teacher i n the S k i l l s Development Centre at Lord Byng Secondary School, independently analyzed the non-exact-replacements of cloze responses on each mode of discourse. Two judges scored grade twelve, i n each of the four categories. Training sessions were conducted before the random assignment of the task. The r e l i a b i l i t y of the scoring was determined by i n t e r - s c o r e r agreement according to the Arrington Formula ( F e i f e l & Lorge, 1950). Retrospective Verbalizations A random sample of 24 subjects was scheduled for interview. Retrospective v e r b a l i z a t i o n s were completed from 11 subjects i n grade 9 and 9 subjects i n grade 12, as follows: Narrative Expository Grade 9 Good 3 3 Poor 2 3 Grade 12 Good 3 3 Poor 1 2 Responses were q u a l i t a t i v e l y analyzed on a post-hoc basis to determine how students use the four cue systems of language: grammatical function, syntax, semantic, and discourse. Categories were induced from the protocols, and s t a t i s t i c a l tests i n v o l v i n g frequencies and using Chi-square were performed. Hypotheses For the research question the following n u l l hypotheses were formu-lat e d Research Question 1 Are there c o r r e l a t i o n a l r e l a t i o n s h i p s among selected indices of 68 reading comprehension and a measure of a t t i t u d e for secondary school students i n grades nine and twelve? Hypothesis 1.1 There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on a reading comprehension measure and (1) the number of exact replace-ments of cloze responses i n nar r a t i v e f i c t i o n , (2) the number of exact replacements of close responses i n expository prose, or (3) a t t i t u d e scores, for either grade nine or twelve. i ) Hypothesis 1.11. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure (ISRT) and exact cloze narrative scores (ECN), for either grade nine or twelve. i i ) Hypothesis 1.12. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the scores on the c r i t e r i o n measure (ISRT) and the exact cloze expository score (ECE), for eit h e r grade nine or twelve. i i i ) Hypothesis 1.13. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure (ISRT) and the scores on the a t t i t u d e measure (ERAS), for either grade nine or twelve. Hypothesis 1.2 There are no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n s among (1) the number of exact replacements of cloze responses i n nar r a t i v e f i c t i o n , (2) the number of exact replacements of cloze responses i n expository prose, and (3) a t t i t u d e scores (ERAS), f or eit h e r grade nine or twelve. i ) Hypothesis 1.21. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the exact cloze narrative scores (ECN) and the exact cloze expository scores (ECE), for either grade nine or twelve. i i ) Hypothesis 1.22. There i s no s t a t i s t i c a l l y s i g n i f i c a n t 69 c o r r e l a t i o n between exact cloze n a r r a t i v e scores (ECN) and a t t i t u d e scores (ERAS), for either grade nine or twelve. i i i ) Hypothesis 1.23. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between exact cloze expository scores (ECE) and a t t i t u d e scores (ERAS), for either grade nine or twelve. Hypothesis 1.3 There are no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n s among (1) scores on a c r i t e r i o n measure, (2) the number of exact replacements of cloze responses i n narrative f i c t i o n , (3) the number of exact replace-ments of cloze responses i n expository prose, and (4) a t t i t u d e scores, over given p r o f i c i e n c y and grade l e v e l s . i ) Hypothesis 1.31. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure (ISRT) and exact cloze narrative (ECN) for good and poor readers at eit h e r grade nine or twelve. i i ) Hypothesis 1.32. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure (ISRT) and exact cloze expository scores (ECE), f o r good and poor readers, at either grade nine or twelve. i i i ) Hypothesis 1.33. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the exact cloze narrative scores (ECN) and the exact cloze expository scores (ECE) for good and poor readers, at either grade nine or twelve. iv) Hypothesis 1.34. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the exact cloze narrative scores (ECN) and a t t i t u d e scores (ERAS), f o r good and poor readers, at either grade nine or twelve, v) Hypothesis 1.35. There i s no s t a t i s t i c a l l y s i g n i f i c a n t i 70 c o r r e l a t i o n between exact cloze expository scores (ECE) and a t t i t u d e scores (ERAS), for good and poor readers at either grade nine or twelve. v i ) Hypothesis 1.36. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the c r i t e r i o n (ISRT) and at t i t u d e scores for good and poor readers, at either grade nine or twelve. Hypothesis 1.4 There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between exact cloze scores and complete discourse a c c e p t a b i l i t y scores, over modes of discourse and p r o f i c i e n c y l e v e l s at given grade l e v e l s . i ) Hypothesis 1.41. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the percent exact scores and the percent complete discourse a c c e p t a b i l i t y scores i n narr a t i v e f i c t i o n , f o r good and poor readers, at eit h e r grade nine or twelve. i i ) Hypothesis 1.42. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the percent exact scores and the percent complete discourse a c c e p t a b i l i t y scores i n expository prose f o r good and poor readers, at eit h e r grade nine or twelve. Research Question 2 Does subject comprehension vary with the mode of discourse? Hypothesis 2.1 There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e i n mean reading comprehension scores between the narr a t i v e f i c t i o n mode of discourse and the expository prose mode, as measured by (1) exact replacements of cloze responses and (2) non-exact-replacements, over given l e v e l s of grade and p r o f i c i e n c y . i ) Hypothesis 2.11. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between the mean scores over modes of discourse, as measured 71 by the number of exact replacements of cloze responses, f o r either grades nine and twelve. i i ) Hypothesis 2.12. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f -ference between mean scores over modes of discourse, as measured by the number of exact replacements of cloze responses, for subjects at good and poor l e v e l s of p r o f i c i e n c y , f o r either grade nine or twelve. i i i ) Hypothesis 2.13. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f -ference between weighted mean score over modes of discourse, as measured by the number of non-exact-replacements (N.E.R.) of cloze responses on each of four v a r i a b l e s : (1) grammatical function, (2) synt a c t i c accept-a b i l i t y , (3) semantic a c c e p t a b i l i t y , and (4) discourse a c c e p t a b i l i t y , for subjects at good and poor l e v e l s of p r o f i c i e n c y . Research Question 3 Do students at two l e v e l s of p r o f i c i e n c y , good and poor, vary i n comprehension? Hypothesis 3.1 There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e i n reading com-prehension between good and poor readers, as measured by the number of exact replacements of cloze responses, at given grade l e v e l s . i ) Hypothesis 3.11. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e i n mean scores over p r o f i c i e n c y l e v e l , as measured by the number of exact cloze responses, f o r eit h e r grades nine or twelve. i i ) Hypothesis 3.12. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between weighted score means over p r o f i c i e n c y l e v e l on eit h e r mode of discourse, as measured by the number of weighted scores on non-exact-replacements of cloze responses, at either grade nine or twelve. 72 Research Question 4 Is there an i n t e r a c t i o n between mode of discourse and p r o f i c i e n c y l e v e l ? Hypothesis 4.1 There i s no s t a t i s t i c a l l y s i g n i f i c a n t i n t e r a c t i o n between modes of discourse (expository prose and narrative f i c t i o n ) and p r o f i c i e n c y l e v e l s (good and poor) at either the grade nine or grade twelve l e v e l , based on the number of exact replacements of cloze responses. Summary In t h i s study, designed to explore how secondary students respond to cloze responses, c o r r e l a t i o n , parametric s t a t i s t i c s , and retrospective v e r b a l i z a t i o n techniques were used to analyze the data from exact and non-exact-replacements of the cloze deleted passages i n two modes of discourse: expository prose and nar r a t i v e f i c t i o n , over good and poor p r o f i c i e n c y l e v e l s , i n grades nine and twelve. CHAPTER IV RESULTS This chapter reports the r e s u l t s of the analysis of cloze responses for the purpose of i n v e s t i g a t i n g the comprehension process of secondary school students. S t a t i s t i c a l and q u a l i t a t i v e procedures were used to examine each research question posited, i n terms of n u l l hypotheses. The exact replacements of cloze responses of sample 1 (N = 107 i n grade 9 and N = 100 i n grade 12) were analyzed using c o r r e l a t i o n a l and i n f e r -e n t i a l s t a t i s t i c a l techniques. A l l subsequent s t a t i s t i c a l tests r e l a t e to sample 2 (20 good subjects and 20 poor subjects at each grade l e v e l ) . The analysis of a t t i t u d e scores, the a n c i l l a r y phase of the study, i s included i n the c o r r e l a t i o n a l studies. The post hoc analysis of the retro s p e c t i v e v e r b a l i z a t i o n s of sample 3 (six good and s i x poor subjects at each grade l e v e l ) completes the chapter. S t a t i s t i c a l Analyses Research Question 1 Are there c o r r e l a t i o n a l r e l a t i o n s h i p s among selected indexes of reading comprehension and a measure of a t t i t u d e f o r secondary students i n grade nine and grade twelve? Pearson product-moment c o r r e l a t i o n c o e f f i c i e n t s were computed among the following v a r i a b l e s : the c r i t e r i o n measure, the Iowa S i l e n t Reading  Test (ISRT), the cloze scores on nar r a t i v e f i c t i o n (ECN) and on 74; . expository prose OECE/) and on scores from the Estes Reading A t t i t u d e Scale (ERAS), to test four hypotheses. The following section reports the r e s u l t s i n terms of the sample tested. Hypothesis 1.1 There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on a reading comprehension measure and (1) the number of exact replacements of cloze responses i n nar r a t i v e f i c t i o n , (2) the number of exact replace-ments of cloze responses i n expository prose, and (3) a t t i t u d e scores, f o r e i t h e r grade nine or twelve. The hypothesis was tested for sample 1 (N = 107 i n grade nine and N = 100 i n grade 12). Complete r e s u l t s are presented i n Table 4.1. i ) Hypothesis 1.11. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure (IRST) and exact cloze n a r r a t i v e scores (ECN), for either grade nine or twelve. There was a moderate p o s i t i v e c o r r e l a t i o n , s t a t i s t i c a l l y s i g n i f i -cant at .01 l e v e l of s i g n i f i c a n c e (r = .45, grade 9; r = .46, grade 12). The n u l l hypothesis was rejected at each grade l e v e l . i i ) Hypothesis 1.12. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the scores on the c r i t e r i o n measure (ISRT) and the exact cloze expository score (ECE), for either grade nine or twelve. There were moderate p o s i t i v e c o r r e l a t i o n s (r = .55, grade 9; r = .53, grade 12), s t a t i s t i c a l l y s i g n i f i c a n t (p < .01). The n u l l hypo-thesis was rejected at each grade l e v e l . i i i ) Hypothesis 1.13. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure (ISRT) and the scores on the a t t i t u d e measure (ERAS), for either grade nine or twelve. In grade 12 the c o r r e l a t i o n (r = .40, p < .01) was moderate TABLE 4.1 INTERCORRELATIONS AMONG SELECTED INDEXES FOR GRADE NINE AND TWELVE READERSa Measure Iowa ECN ECE ATT Iowa S i l e n t Reading Test (ISRT) . 45** .55** .18 Exact Cloze Narrative (ECN) .46** .40** .14 Exact Cloze Expository (ECE) .53** .58** .20* Att i t u d e (ATT) .40** .21* .33**^ Upper r i g h t side of matrix l i s t s Pearson c o r r e l a t i o n c o e f f i c i -ents of grade nine readers (N = 107) and the lower l e f t side, grade twelve readers (N = 100). * p < .05 ** p < .01 p o s i t i v e . The n u l l hypothesis was accepted i n grade 9 and rejected i n grade 12. Hypothesis 1.2 There are no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n s among (1) the number of exact replacements of close responses i n narrative f i c t i o n , (2) the number of exact replacements of cloze responses i n expository prose, and (3) a t t i t u d e scores, f o r ei t h e r grade nine or twelve (Table 4.1). i ) Hypothesis 1.21. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between exact cloze n a r r a t i v e scores (ECN) and the exact cloze expository score (ECE), at ei t h e r grade nine or twelve. At each grade l e v e l the moderate p o s i t i v e c o r r e l a t i o n (r = .40, grade 9; r = .58, grade 12) was s t a t i s t i c a l l y s i g n i f i c a n t , at the .01 l e v e l of s i g n i f i c a n c e . The n u l l hypothesis was rejected at each grade l e v e l . i i ) Hypothesis 1.22. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between exact cloze narrative (ECN) and a t t i t u d e scores (ERAS), i n ei t h e r grade nine or twelve. • .. -There was a low p o s i t i v e c o r r e l a t i o n (r = .21, p < .05) i n grade 12. The n u l l hypothesis was accepted for grade 9 and rejected f o r grade 12. i i i ) Hypothesis 1.23. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between exact cloze expository scores (ECE) and a t t i t u d e scores (ERAS), i n either grade nine or twelve. The c o r r e l a t i o n (r = .20, p < .05) i n grade 9 and the c o r r e l a t i o n (r = .33, p < .01) i n grade 12 were low p o s i t i v e . The n u l l hypothesis was rejected at each grade l e v e l . 71 The c o r r e l a t i o n s for the following three hypotheses were computed for 20 good subjects and 20 poor subjects, at grade nine (Table 4.2) and grade twelve (Table 4.3). Hypothesis 1.3 There are no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n s among (1) scores on the c r i t e r i o n measure, (2) the number of exact replacements of cloze responses i n na r r a t i v e f i c t i o n , (3) the number of exact replacements of cloze responses i n expository prose, and (4) a t t i t u d e scores, over given p r o f i c i e n c y and grade l e v e l s . i ) Hypothesis 1.31. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure (ISRT) and exact cloze n a r r a t i v e (ECN) for good and poor subjects, at ei t h e r grade nine or twelve. No c o r r e l a t i o n achieved s t a t i s t i c a l s i g n i f i c a n c e . The n u l l hypo-thesis was accepted at both grade nine and twelve. i i ) Hypothesis 1.32. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between scores on the c r i t e r i o n measure•(ISRT) and exact cloze expository scores (ECE), for good and poor subjects, at ei t h e r grade nine or twelve. The moderate p o s i t i v e c o r r e l a t i o n (r = .60, p < .01) was s t a t i s -t i c a l l y s i g n i f i c a n t f o r poor grade 12 subjects. The n u l l hypothesis was rejected; i t was accepted for good and poor grade 9 subjects and good grade 12 subjects. i i i ) Hypothesis 1.33. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the exact cloze n a r r a t i v e scores (ECN) and the exact cloze expository scores (ECE) for good and poor subjects, at ei t h e r grade nine or twelve. 78 TABLE 4.2 INTERCORRELATIONS AMONG SELECTED INDEXES FOR GOOD AND POOR READERS3 IN GRADE NINE (N = 20) Measure Iowa ECN ECE ATT Iowa S i l e n t Reading Test (ISBSE) . .04 . .14 .25 Exact Cloze Narrative (ECN) .05 .36* .28 Exact Cloze Expository (ECE) .03 -.02 .56** Attitude (ATT) -.33 .35 -.35 Upper r i g h t side of matrix l i s t s Pearson c o r r e l a t i o n c o e f f i c i e n t s of good readers and the lower l e f t side of poor readers. * p < .05 ** p < .01 TABLE 4.3 INTERCORRELATIONS AMONG SELECTED INDEXES FOR GOOD AND POOR READERS3 IN GRADE TWELVE (N = 20) Measure Iowa ECN ECE ATT Iowa S i l e n t Reading. Test (ISR3?) \ ^ - . 1 0 .07 -.25 Exact Cloze Narrative (ECN) .30 .30 .06 Exact Cloze Expository (ECE) .60** .47* -.16 Atti t u d e (ATT) .56** .11 .18^"\. aUpper r i g h t side of matrix l i s t s Pearson c o r r e l a t i o n c o e f f i c i e n t s of good readers and the lower l e f t side of poor readers. * p < .05 ** p < .01 79 The low p o s i t i v e c o r r e l a t i o n (r = .36, p < .05) for good grade 9 subjects and the moderate p o s i t i v e c o r r e l a t i o n (r = .47, p < .05) for poor grade 12 subjects were s t a t i s t i c a l l y s i g n i f i c a n t . The n u l l hypo-thesis was rejected; i t was accepted for poor grade 9 subjects and for good grade 12 subjects. iv) Hypothesis 1.34. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the exact cloze n a r r a t i v e scores (ECN) and a t t i t u d e scores (ERAS) for good and poor subjects, at either grade nine or twelve. No c o r r e l a t i o n s were s t a t i s t i c a l l y s i g n i f i c a n t . The n u l l hypo-thesis was accepted at both grade nine and twelve. v) Hypothesis 1.35. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between exact cloze expository scores (ECE) and a t t i t u d e scores (ERAS), for good and poor subjects, at either grade nine or twelve. The moderate p o s i t i v e c o r r e l a t i o n (r = .56, p < .01) was s t a t i s -t i c a l l y s i g n i f i c a n t f o r good grade 9 subjects. The n u l l hypothesis was rejected; i t was accepted for poor grade 9 subjects and for both good and poor grade 12 subjects. v i ) Hypothesis 1.36. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the c r i t e r i o n (ISRT) and a t t i t u d e scores for good and poor readers, at either grade nine or twelve. The moderate c o r r e l a t i o n (r = .56, p < .01) was s t a t i s t i c a l l y s i g n i f i c a n t for poor subjects i n grade twelve. The n u l l hypothesis was rejected; i t was accepted for poor grade 12 subjects, and for good and poor grade 9 subjects. The percent exact cloze scores were correlated with the percent complete discourse a c c e p t a b i l i t y scores ( i . e . , acceptable on three v a r i -ables: s y n t a c t i c a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y ) . Results are presented i n Table 4.4. Hypothesis 1.4 There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between exact cloze scores and complete discourse a c c e p t a b i l i t y scores, over modes of discourse and p r o f i c i e n c y l e v e l s at given grade l e v e l s . i ) Hypothesis 1.41. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the percent exact scores and the percent complete discourse a c c e p t a b i l i t y scores i n n a r r a t i v e f i c t i o n , f o r good and poor subjects, at e i t h e r grade nine or twelve. The moderate p o s i t i v e c o r r e l a t i o n (r = .66, p < .01) was s t a t i s -t i c a l l y s i g n i f i c a n t f o r poor grade 12 subjects. The n u l l hypothesis was rejected; i t was accepted for good 12 subjects and f o r good and poor grade 9 subjects. i i ) Hypothesis 1.42. There i s no s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n between the percent exact scores and the percent complete discourse a c c e p t a b i l i t y scores i n expository prose for good and poor subjects, at e i t h e r grade nine or twelve. The negative c o r r e l a t i o n (r = -.51, p < .05) was s t a t i s t i c a l l y s i g n i f i c a n t f o r good grade 9 subjects. The n u l l hypothesis was rejected; i t was accepted f o r poor grade 9 subjects. The c o r r e l a t i o n (r = .53, r = .54) f o r good and poor grade 12 subjects, r e s p e c t i v e l y , were s t a t i s t i c a l l y s i g n i f i c a n t (p < .05). The n u l l hypothesis was rejec t e d . 81 TABLE 4.4 CORRELATION COEFFICIENTS OF PERCENT COMPLETE DISCOURSE ACCEPTABILITY SCORES AND PERCENT EXACT CLOZE SCORES FOR NARRATIVE AND EXPOSITORY PASSAGES BY GOOD AND POOR READERS IN GRADES NINE AND TWELVE3 Grade Nine Grade Twelve Narrative Expository Narrative Expository Good Readers .22 -.51* .14 .53* Poor Readers .16 .41 .66** .54* N = 20 for each c o e f f i c i e n t * p < .05 ** p < .01 82 Research Question 2 Does subject comprehension vary with the mode of discourse? Hypothesis 2.1 There i s no s t a t i s t i c a l l y s i g n i f i c a n t difference i n mean reading comprehension scores between the narrative f i c t i o n mode of discourse and the expository prose mode, as measured by (1) exact replacements of cloze responses and (2) non-exact-replacements, over given l e v e l s of grade and p r o f i c i e n c y . i ) Hypothesis 2.11. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between the mean scores over modes of discourse, as measured by the number of exact replacements of cloze responses, f o r either grade nine or twelve. The analysis involved the comparison of means for exact cloze scores on each mode of discourse (Table 4.5). For grade nine the mean ECN score was 23.27 and the mean ECE score was 13.91. They were s i g n i f i -cantly d i f f e r e n t (t = 16.87, p < .001). For grade twelve the narrative mean score of 24.59 was s i g n i f i c a n t l y greater than the expository mean score of 17.32. The t value was 15.61 (p < .001). The n u l l hypothesis that there i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f -ference between mean scores when comparing comprehension of modes of d i s -course, as measured by exact replacements of cloze scores, was rejected, at each grade l e v e l . i i ) Hypothesis 2.12. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between mean scores over modes of discourse, as measured by exact replacements of cloze responses, for good and poor l e v e l s of pr o f i c i e n c y . For good subjects i n grade nine (N = 20), the mean was 26.70, on TABLE 4.5 COMPARISON BETWEEN EXACT CLOZE NARRATIVE (ECN) AND EXACT CLOZE EXPOSITORY (ECE) MEANS FOR GRADES NINE AND TWELVE Variables Means S.D. t values Grade Nine (N = 107) ECN 23.27 5.41 16.87* ECE 13.91 5.12 Grade Twelve (N = 100) ECN 24.59 5.41 15.61* ECE 17.32 5.25 * p < .001 84 the narrative mode while on the expository, the mean was 17.75 (Table 4.6). This d i f f e r e n c e was s t a t i s t i c a l l y s i g n i f i c a n t (t = 8.76; p < .001). For poor subjects, the mean of 22.5 on the narrative mode was s i g n i f i c a n t l y greater than on the expository, with a mean of 11.55. The t value was 8.04 (p < .001). Similar r e s u l t s pertained for the grade twelve sample (N = 20). The cloze scores of good subjects had means of 29.60 and 22.15 for the n a r r a t i v e and expository modes, re s p e c t i v e l y . The d i f f e r e n c e was s t a t -i s t i c a l l y s i g n i f i c a n t at the .001 l e v e l (t = 7.42). Means of 21.65 and 15.95 were also s i g n i f i c a n t l y d i f f e r e n t for poor subjects (t = 5.46, p < .001). The n u l l hypothesis was rejected for p r o f i c i e n c y l e v e l , at each grade, nine and twelve, as measured by the exact cloze scores on each mode of discourse. i i i ) Hypothesis 2.13. There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between means over modes of discourse, as measured by non-exact-replacements (N.E.R.) of cloze responses on four v a r i a b l e s : (1) grammatical function, (2) s y n t a c t i c a c c e p t a b i l i t y , (3) semantic a c c e p t a b i l i t y , and (4) discourse a c c e p t a b i l i t y , for good and poor l e v e l s of p r o f i c i e n c y . To determine how p r o f i c i e n c y l e v e l s varied with the mode of d i s -course, t - t e s t s were performed on the means of the weighted scores for each v a r i a b l e . Good subjects at the ninth grade had s i g n i f i c a n t l y greater means (Table 4.7) on the narrative mode for a l l four v a r i a b l e s when com-pared with the expository mode; f o r example, on grammatical function, the means were 1.69 and 1.43 r e s p e c t i v e l y (t = 3.60; p < .001). The poor subjects showed s i m i l a r marked diff e r e n c e s , with values of 1.38 on the TABLE 4.6 COMPARISON BETWEEN MEANS OF EXACT CLOZE NARRATIVE (ECN) AND EXACT CLOZE EXPOSITORY (ECE) FOR GOOD AND POOR READERS IN GRADES NINE AND TWELVE (N = 20) Profic i e n c e s Variables Means S.D. t-values Grade Nine ECN 26.70 3.99 Good 8.76* ECE 17.75 4.10 ECN 22.50 4.35 Poor 8.04* ECE 11.55 4.19 Grade Twelve ECN 29.60 3.99 Good 7.42* ECE 22.15 3.59 ECN 21.65 4.89 Poor 5.46* ECE 15.95 4.11 * p < .001 TABLE 4.7 COMPARISON BETWEEN MEANS OF NON-EXACT-REPLACEMENTS OF CLOZE RESPONSES ON MODES OF DISCOURSE BY PROFICIENCY LEVEL IN GRADE NINE (N = 20) Modes of Grammatical Syntactic Semantic Discourse Discourse Function A c c e p t a b i l i t y A c c e p t a b i l i t y A c c e p t a b i l i t y Good Readers Narrative 1.69 1.84 3.40 Expository 1.43 1.56 2.45 t Value 3.60* 3.45* 5.98* 1.70 1.15 8.28* Narrative Expository t Value Poor Readers ,38 1.49 2.64 .93 .96 1.35 .58* 5.02* 7.25* 1.35 .60 8.07* .001 TABLE 4.8 COMPARISON BETWEEN MEANS OF NON-EXACT-REPLACEMENTS OF CLOZE RESPONSES ON MODES OF DISCOURSE BY PROFICIENCY LEVEL IN GRADE TWELVE (N = 20) Modes of Grammatical Syntactic Semantic Discourse Discourse Function A c c e p t a b i l i t y A c c e p t a b i l i t y A c c e p t a b i l i t y Good Readers Narrative 1.62 1.63 2.63 Expository 1.59 1.65 2.78 t Values 0.32 -0.27 -0.87 1.04 1.32 -3.89* Narrative Expository t Values Poor Readers 1.13 1.11 1.55 1.01 .96 1.32 0.92 1.19 1.10 .61 .61 -0.01 * p < .05 87 narrative mode and .93 on the expository discourse (t = 4.58; p < .001). The complete r e s u l t s indicated a r e j e c t i o n of the n u l l hypothesis f o r non-exact-replacements of cloze responses f o r good and poor readers at the ninth grade on both modes of discourse. In grade twelve, good subjects showed no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between scores on the narrative and expository modes on three v a r i a b l e s : grammatical function, s y n t a c t i c a c c e p t a b i l i t y , and semantic a c c e p t a b i l i t y (Table 4.8). In discourse a c c e p t a b i l i t y , the expository mode was both s i g n i f i c a n t l y d i f f e r e n t from the narrative mode (t = -3.89; p < .001) and reversed with expository higher. Poor subjects showed no s i g n i f i c a n t differences between scores on narrative f i c t i o n and expository prose on the four variables of non-exact-replacements of cloze responses. The n u l l hypothesis was therefore accepted on three v a r i a b l e s : grammatical function, syntactic a c c e p t a b i l i t y , and semantic a c c e p t a b i l i t y for good and poor subjects. On discourse a c c e p t a b i l i t y , the n u l l hypo-thesis was accepted f o r poor subjects; i t was rejected f o r good subjects in grade twelve. Research Question 3 Do students at two le v e l s of p r o f i c i e n c y , good and poor, vary i n comprehension? Hypothesis 3.1 There i s s t a t i s t i c a l l y no s i g n i f i c a n t difference i n reading compre-hension between good and poor readers, as measured by.exact replacements of cloze responses, at given grade l e v e l s . To consider how the number of exact responses varied i n terms of the two main e f f e c t s of pr o f i c i e n c y and of modes of discourse, analysis of variance was c a r r i e d out, at each grade l e v e l , on the exact scores. 88 i ) Hypothesis 3.11. There i s no s t a t i s t i c a l l y s i g n i f i c a n t main e f f e c t due to p r o f i c i e n c y , good and poor, as measured by the number of exact cloze responses for grades nine and twelve. Results indicated that the main e f f e c t s of p r o f i c i e n c y achieved s t a t i s t i c a l s i g n i f i c a n c e beyond the .001 l e v e l (F = 30.00, p < .001) i n grade nine, and i n grade twelve (F = 55.24, p < .001). i i ) Hypothesis 3.12. There i s no s t a t i s t i c a l l y s i g n i f i c a n t main ef f e c t due to mode of discourse, n a r r a t i v e f i c t i o n and expository prose, as measured by the number of exact cloze responses, for grades nine and twelve. At both grade l e v e l s the main e f f e c t s due to mode of discourse were s t a t i s t i c a l l y s i g n i f i c a n t , (F = 110.00, p < .001) i n grade nine and i n grade twelve (F = 47.71, p < .001). Narrative f i c t i o n scores were s i g n i f i c a n t l y .greater than expository prose scores. The n u l l hypotheses were rejected f o r p r o f i c i e n c y and mode of discourse. Complete r e s u l t s are presented i n Tables 4.9 and 4.10. Hypothesis 3.2 There are no s t a t i s t i c a l l y s i g n i f i c a n t differences i n comprehension over p r o f i c i e n c y l e v e l s on either mode of discourse, as measured by the number of weighted scores of non-exact-replacements of cloze responses at either grade nine or twelve. The means of the weighted scores of the good and poor readers were compared on each v a r i a b l e : grammatical function, s y n t a c t i c a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y , f o r narrative f i c -t i o n and for expository prose, using t - t e s t s . Hypothesis 3.21 for grades nine and twelve. There i s no s t a t i s t i c -a l l y s i g n i f i c a n t d i f f e r e n c e over p r o f i c i e n c y l e v e l on either mode of TABLE 4.9 ANALYSIS OF VARIANCE OF EXACT CLOZE SCORES OF NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE NINE Source SS d.f. M.S. F Between modes 1980. 05 1 1980. 05 110.00* Between p r o f i c i e n c i e s 540. 80 1 540. 80 30.00* Interaction 20. 00 1 20. 00 1.11 Within groups (residual)' 1313. 90 73 18. 00 To t a l 3854. 75 76 * p < .001 TABLE 4.10 ANALYSIS OF VARIANCE OF EXACT CLOZE SCORES OF NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE TWELVE Source SS d.f. M.S. F Between modes 864. 61 1 864. 61 47.71* Between p r o f i c i e n c i e s 1001. 11 1 1001. 11 55.24* Interaction 15. 32 1 15. 32 0.84 Within groups (residual) 1322. 85 73 18. 12 To t a l 3203. 89 76 * p < .001 90 discourse as measured by the weighted scores of non-exact-replacements. On each v a r i a b l e , the means of good subjects i n grade nine showed s t a t i s t i c a l l y s i g n i f i c a n t d ifferences from the means of poor subjects. The minimum obtained t value (t = 4.47; p < .001) on means of 1.70 and 1.35, was on the discourse a c c e p t a b i l i t y v a r i a b l e i n narrative f i c t i o n . The greatest value (t = 6.68; p < .001) was on the discourse accept-a b i l i t y of expository prose, with means of 1.15 for good subjects and .60 for poor subjects (Table 4.11). In grade twelve, good subjects achieved s t a t i s t i c a l l y s i g n i f i c a n t differences on a l l the four v a r i a b l e s . Means of 1.62 on grammatical function, n a r r a t i v e f i c t i o n , compared to 1.31 for poor readers, represented the lowest numerical difference, (t = 3.71; p < .001). The greatest numerical d i f f e r e n c e , (t = 8.54; p < .001) was on means of 1.32 and .62 for good and poor subjects r e s p e c t i v e l y , on the d i s -course a c c e p t a b i l i t y v a r i a b l e of expository prose (Table 4.12). Research Question 4 Is there a s i g n i f i c a n t i n t e r a c t i o n between mode of discourse and pr o f i c i e n c y l e v e l ? Hypothesis 4.1 There i s no s t a t i s t i c a l l y s i g n i f i c a n t i n t e r a c t i o n between mode of discourse (narrative f i c t i o n and expository prose) and pr o f i c i e n c y l e v e l s (good and poor) i n grade nine or i n grade twelve, i n the number of exact replacements of cloze responses. There i s no s t a t i s t i c a l l y s i g n i f i c a n t i n t e r a c t i o n (Tables 4.9 and 4.10) i n grade nine between p r o f i c i e n c y and modes of discourse, (F = 1.11) nor at grade twelve, (F = .84). Figures 4.1 and 4.2 i l l u s t r a t e the trends. The n u l l hypothesis was accepted. TABLE 4.11 MEANS OF WEIGHTED SCORES ACCORDING TO GRAMMATICAL FUNCTION (GF) SYNTACTIC ACCEPTABILITY (SA), SEMANTIC ACCEPTABILITY (SEM), AND DISCOURSE ACCEPTABILITY (DA) OF NON-EXACT-REPLACEMENT WORDS IN A NARRATIVE AND EXPOSITORY PASSAGE BY 20 GOOD (G) AND 20 POOR (P) READERS IN GRADE NINE Variable Readers Means Std. Dev. Narrative Passage GF SA SEM DA G P G P G P G P 1. 1. 1. 1. ,69 ,38 ,84 ,49 3.40 2.64 1.70 1.35 0.11 0.24 0.12 0.28 0.31 0.52 0.17 0.31 5.11* 5.05* 5.61* 4.47* Expository Passage GF SA SEM DA G P G P G P G P 1.43 0.93 1.56 0.96 45 35 1.15 0.62 0.30 0.37 0.35 0.37 0.64 0.59 0.24 0.28 4.74* 5.23* 5.65* 6.68* See text f o r weighting and scoring methods (p. 65) 'All c alculated t values are s i g n i f i c a n t at p < .001. TABLE 4.12 MEANS OF WEIGHTED SCORES3 ACCORDING TO GRAMMATICAL FUNCTION (GF), SYNTACTIC ACCEPTABILITY (SA), SEMANTIC ACCEPTABILITY (SEM), AND DISCOURSE ACCEPTABILITY (DA) OF NON-EXACT-REPLACEMENT WORDS IN A NARRATIVE AND EXPOSITORY PASSAGE BY 20 GOOD (G) AND 20 POOR (P) READERS IN GRADE TWELVE Variable Readers Means Std. Dev. t ^ Narrative Passage GF G 1.62 0.32 P 1.13 0.49 SA G 1.63 0.24 P 1.11 0.48 SEM G 2.63 0.54 P 1.55 0.76 DA G 1.04 0.22 P 0.61 0.34 Expository Passage GF G 1.59 0.23 P 1.01 0.31 SA G 1.65 0.22 P 0.96 0.29 SEM G 2.78 0.57 P 1.32 0.56 DA G 1.32 0.24 P 0.61 0.28 3.71* 4.35* 5.19* 4.70* 6.72* 8.28* 8.23* 8.54* aSee text for weighting and scoring methods (p. 65) ^ A l l c alculated t values are s i g n i f i c a n t at p < .001. 50 93 40 \-I I Poor Good Pr o f i c i e n c y Figure 4.1. Mean exact replacement cloze scores f or na r r a t i v e f i c t i o n and for expository prose passages by 20 poor and 20 good readers i n grade nine. 50 40 a) 0 Poor Good Pro f i c i e n c y Figure 4.2. Mean exact replacement cloze scores for narrative f i c t i o n and for expository prose passages by 20 poor and 20 good readers i n grade twelve. 95 Post Hoc Analysis In a post hoc analysis to determine the e f f e c t of synonyms, the number of exact cloze replacements were combined with the number of responses that were acceptable on three v a r i a b l e s : syntactic accept-a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y as measured by the t o t a l number of "yes" responses i n discourse a c c e p t a b i l i t y . Analysis of variance was used to consider s i g n i f i c a n t differences i n the le v e l s of p r o f i c i e n c y , and modes of discourse, and to determine i f i n t e r a c t i o n existed between the factors (Tables 4.13 and 4.14). The grade nine r e s u l t s indicated that main e f f e c t s achieved s i g n i f -icance. Good subjects exceeded poor subjects (F = 69.92; p < .001). Nar-r a t i v e f i c t i o n scores exceeded expository prose (F = 216.43; p < .001). Good subjects obtained means of 45.75 on the narrative and 32.45 on the expository mode. Poor subjects had means of 39.55 and 19.95 r e s p e c t i v e l y . The interactions (F = 15.26, p < .001) (Figure 4.3) showed that expository prose was more d i f f i c u l t f o r poor subjects than for good subjects. In grade twelve the r e s u l t s of the analysis indicated that the good readers were more p r o f i c i e n t than the poor subjects (F = 70.28; p < .001). For the main e f f e c t of mode of discourse there was no s i g -n i f i c a n t difference (F = 1.13). The means of the good subjects were 37.65 on the narrative and 37.55 on the expository; for poor subjects the means were 27.40 and 24.55, r e s p e c t i v e l y . No i n t e r a c t i o n was demonstrated (Figure 4.4). Analysis of variance was used to examine s i g n i f i c a n t differences i n the number of complete discourse a c c e p t a b i l i t y responses at each grade and p r o f i c i e n c y l e v e l f o r each mode of discourse, to determine i f i n t e r a c t i o n existed (Tables 4.15 and 4.16). 96 TABLE 4.13 ANALYSIS OF VARIANCE OF EXACT CLOZE PLUS COMPLETE DISCOURSE ACCEPTABILITY SCORES OF NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE NINE Source SS d.f. M.S. F Between modes 5412. 05 1 5412. 05 216. 43* Between p r o f i c i e n c i e s 1748. 45 1 1748. 45 69. 92* Interaction 381. 65 1 381. 65 15. 26* Within groups (residual) 1825. 40 73 25. 01 To t a l 9367. 55 76 * p < .001 TABLE 4.14 ANALYSIS OF VARIANCE OF EXACT CLOZE PLUS COMPLETE DISCOURSE ACCEPTABILITY SCORES OF NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE TWELVE Source SS d.f. M.S. F Between modes 43.51 1 43. 51 1. 13 Between p r o f i c i e n c i e s 2702.81 1 2702. 81 70. 28* Interaction 37.82 1 37. 82 0. 98 Within groups (residual) 2807.25 73 38. 45 T o t a l 5591.38 76 * p < .001 TABLE 4.15 ANALYSIS OF VARIANCE OF COMPLETE DISCOURSE ACCEPTABILITY SCORES OF NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE NINE Source SS d.f. M.S. F Between modes 756.45 1 756.45 46 .38** Between p r o f i c i e n c i e s 288.80 1 288.80 17 .71** Interaction 125.00 1 125.00 7 .66* Within groups (residual) 1190.50 73 16.31 To t a l 2360.75 76 * p < .01 ** p < .001 TABLE 4.16 ANALYSIS OF VARIANCE OF COMPLETE DISCOURSE ACCEPTABILITY SCORES OF NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE TWELVE Source SS d.f. M.S. F Between modes . 391.61 1 391.61 33. 61** Between p r o f i c i e n c i e s 400.51 1 400.51 34. 68** Interaction 117.62 1 117.62 10. 10* Within groups (residual) 850.75 73 11.65 Tot a l 1760.49 76 * p < .01 ** p < .001 98 cn S-i o o CO •H XI QJ 4-1 , | ft e o a co 3 ft CD N O . 1 o 4-1 c CD CO 6 <u QJ M CJ o to o CO ft a) ^  4-1 •H 4J T-H O •H CO • f l CO <D 4-1 ft c Q) CO O 01 CJ s CO 50 40 30 20 10 Narrative Expository Poor Good Pro f i c i e n c y Figure 4.3 Mean exact replacement cloze plus complete discourse a c c e p t a b i l i t y scores for narrative f i c t i o n and for expository prose passages by 20 poor and 20 good readers i n grade nine. 99 50 r CD cn yi O o CO •H X I CU •M QJ rH ft 0 O o CD ft 0) N O 40 Y 30 Narrative Expository CU CO e cu CD M u o cd o rH cn ft cu >, H 4-1 •H 4-1 PH O "ri Cfl .a CU 4J ft C cu «j o CU CJ 20 10 _1_ Poor Good Pro f i c i e n c y Figure 4.4. Mean exact replacement cloze plus complete discourse a c c e p t a b i l i t y scores for na r r a t i v e f i c t i o n and for expository prose passages by 20 poor and 20 good readers i n grade twelve. 100 The grade nine r e s u l t s indicated that the means of good subjects exceeded the means of poor subjects (F = 17.71; p < .001). Narrative f i c -t i o n scores exceeded expository prose (F = 46.38; p < .001). Good sub-j e c t s obtained means of 18.35 on narrative and 14.7 on expository. Poor subjects had means of 17.05 and 8.4 re s p e c t i v e l y . The i n t e r a c t i o n (F = 7.66; p < .01) showed that expository prose was more d i f f i c u l t f o r poor.readers than f o r good readers. In grade 12, the r e s u l t s showed means of good readers exceeded means of poor readers (F = 34.38; p < .001). Expository prose means exceeded na r r a t i v e f i c t i o n means (F = 33.61; p < .001). Good readers had means of 8.05 and 14.9 on narrative and expository, r e s p e c t i v e l y . Poor readers had means of 6.0 and 8.0, re s p e c t i v e l y . The i n t e r a c t i o n (F = 10.10; p < .01) showed the good.readers were more p r o f i c i e n t than poor readers i n expos-i t o r y prose. Results are presented i n Table 4.15 and 4.16, and i n Figures 4.5 and 4.6. R e l i a b i l i t y of Scoring The r e l i a b i l i t y of the scoring of the non-exact-replacements (N.E.R.'s) was determined by int e r - s c o r e r agreement. Judges 1 and 2 independently rescored grade 9 tests on the narrative and expository modes; judges 3 and 4 rescored grade 12 t e s t s . Ten words were randomly selected from each of f i v e tests i n grade 9 na r r a t i v e and i n grade 12 expository. Agree-ments were calculated i n terms of percentage using the Arrington Formula ( F e i f e l & Lorge, 1950) for scoring the r e l i a b i l i t y of q u a l i t a t i v e r e s -ponses. For each category—grammatical function, syntactic a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y — t h e responses of each observer's scoring that agreed with the others (doubling the agreements) was divided by t h i s t o t a l plus the disagreements. Poor Good Pr o f i c i e n c y Figure 4.5. Mean complete discourse a c c e p t a b i l i t y scores for narrative f i c t i o n and for expository prose passages by 20 poor and 20 good readers i n grade nine. 102 20 L Figure 4.6. Mean complete discourse a c c e p t a b i l i t y cloze scores for n a r r a t i v e f i c t i o n and for expository prose passages for 20 poor and 20 good readers i n grade twelve. 103 2 x agreements 2_ ^  Q • — *' 2 x agreements :+.ndisagreements The percentage of agreements for 200 items i n grade 9 ranged from 78 to 96.9 percent (Table 4.17). In grade 12 the percentage of agree-ments on 200 items was from 80.9 to 96.9 percent (Table 4.18). The mean i n t e r - r a t e r agreement was 91.6 percent. Retrospective Verbalizations This section of the r e s u l t s reports the q u a l i t a t i v e procedures which made possible the d e s c r i p t i o n of the processes used by subjects to obtain cloze responses, i n p a r t i c u l a r the four cue systems of language: grammatical function, syntax, semantics, and. discourse. The analysis was done i n three stages: (1) the induction of the categories from the protocols, (2) the assignments of protocols to the categories, and (3) the s t a t i s t i c a l analysis of the frequencies i n each category. Induction of the Categories The passage the subject read was transcribed with the cloze dele-t i o n numbered, the subject's response underlined, and the exact replace-ment i n brackets, as appropriate. Each d e l e t i o n was then matched to the transcribed protocol. Words read with s p e c i a l emphasis were enclosed i n s i n g l e quotation marks. For example, subject #18 read sentence 2 i n n a r r a t i v e f i c t i o n : "Davy tensed h i s 4. legs (muscles) ready for the dive" and reported: "I guess that tense h i s 'legs' be-cause you kind of bend your legs when you dive." (Complete t r a n s c r i p -tions are presented i n Appendix B.) After the e n t i r e sample was recorded, patterns of responses s i m i l a r to the categories of the taxonomy for the analysis of non-exact-replace-ments became apparent. Six categories were established, four r e l a t e to 104 TABLE 4.17 PERCENTAGE OF AGREEMENT BETWEEN INVESTIGATOR AND INDEPENDENT JUDGES 1 AND 2 IN THE SCORING OF NON-EXACT-REPLACEMENTS OF CLOZE RESPONSES IN THE NARRATIVE MODE BY GRADE NINE READERS Independent Judges Grammatical Syntactic Semantic Discourse Function A c c e p t a b i l i t y A c c e p t a b i l i t y A c c e p t a b i l i t y l a + 2 1 + 3 2 + 3 93.6 91.3 93.6 96.9 96.9 95.8 91.3 94.7 94.7 90.1 85.0 78.0 Investigator TABLE 4.18 PERCENTAGE OF AGREEMENT BETWEEN INVESTIGATOR AND INDEPENDENT JUDGES 3 AND 4 IN THE SCORING OF NON-EXACT-REPLACEMENTS OF CLOZE RESPONSES IN THE EXPOSITORY MODE BY GRADE TWELVE READERS Independent Judges Grammatical Function Syntactic A c c e p t a b i l i t y Semantic A c c e p t a b i l i t y Discourse A c c e p t a b i l i t y l a + 3 96.9 95.8 93.6 91.3 1 + 4 96.9 96.9 80.9 85.1 3 + 4 92.4 95.8 80.9 86.3 Investigator 105 the l i n g u i s t i c cue system, one to l i f e experience, and the l a s t included no reasons, corrections, miscues, and omissions. A d e s c r i p t i o n of the c r i t e r i a and examples for each category follow. 1. Grammatical Function. The response was assigned to the gram-matical function category i f the subject referred to a part of speech, such as "verb." For example, subject #18 read: "A crowd 1. had gathered . . . " and reported: "I got 'had' because i t had to be a verb." 2. Syntax. References to the rules of grammar or sentence struc-ture are syntax cues. Responses that a word " f i t s , " "sounds r i g h t " or " j o i n s " were assigned to the category. The use of punctuation and emphasis, that i s , intonation, were also i n f e r r e d to r e l a t e to syntax. For example, subject #13 read: " . . . begging him 10. to_ stop" and commented: " I t ' s probably the only word that could f i t there." 3. Semantic. The term semantic means the cues of meaning within the sentence. References to other words and statements about meaning or making sense appeared to. suggest that the subject used semantic cues to determine the replacement. For example, subject #15 read: "Later the railways 19. channeled (spread) t h e i r networks . . ." and reported: "I wasn't too sure but i t seemed to f i t with networks." 4. Discourse. The term discourse r e f e r s to cues of meaning i n the e n t i r e passage, i n c l u d i n g the i n t a c t and deleted portions. In the n a r r a t i v e mode, references were to the s t o r y - l i n e , s e t t i n g , and charac-t e r s . The main topic, t r a n s i t i o n words, or patterns of organization such as cause and e f f e c t , were used to replace cloze deletions i n the expository mode. S p e c i f i c words also cued responses. The following examples i l l u s t r a t e the placement of responses i n the discourse category. In n a r r a t i v e f i c t i o n , grade 12 subject #210 read: "And with 5. 106 Joseph (Father) there, i t was dark 6. and quiet and warm i n 7. the camp, and the big 8. trees whispered together t h e i r drowsy 9. melodies (talk) outside i n the dark." The comment was: "He's j u s t t a l k i n g about hi s Dad and there's only two people there. Both of them mentioned trees, l i k e Joseph men-tioned trees when he found the spinner and David mentioned trees before i n what heVwas saying i n the l e t t e r before that." In expository prose, grade 9 subject #12 read the sentence: "so the great c i t i e s 40. had (have) continued to grow r a p i d l y , 4. so (while) few country towns have . . ."and reported: "'had' meaning then, not any more, we're t a l k i n g about the past. 'So' means cause and e f f e c t , because of t h i s , t h i s has happened." 5. L i f e experience. Responses which drew on knowledge outside the passages were i n f e r r e d to be cued by the subject's experience i n a s i m i l a r s i t u a t i o n . To the phrase "chicken out" i n the grade 9 narra-t i v e passage, a t y p i c a l response was from subject #13: "'Chickening 1 out, 'copping' out, 'chicken' o u t — a n y t h i n g would work on that one. I guess when Cl i n t o n said t h i s i t bothered him a b i t , you know, when you get mad quite often people clench t h e i r f i s t s , to hide t h e i r aggravation." 6. No reason: The s i x t h category was established for those cloze replacements without responses, either because subjects were unable to give reasons or f a i l e d to do so. Separate categories for corrections, miscues i n o r a l reading, and omission were also made, and l a t e r , collapsed into category 6, because of low frequency. . Assignment to the Categories The numbered responses were assigned and coded: Grammatical Function (GF), Syntax (Sy), Semantics (Sm), Discourse (D), L i f e Experience 107 (LE) , No Reason (NR). The responses of grade 12 subject #211 to sentence 1 i n narrative f i c t i o n summarize the procedure: "And then, l a t e r when 1. the whole night was cool, 2. there was the dry wood 3. I_ (Father) could somewhere 4. near (for) the f i r e . " Response 1. "The" was the only word that would make any sense between "when" and "a whole night." Another adjective wouldn't have been needed (GF). The response was a r b i t r a r i l y assigned because of the reference to the grammatical term, "ad j e c t i v e , " however, awareness of semantics was apparent. Response 2. "There" was the only small word to make sense (Sm). The words "make sense" appeared to indic a t e that the response was cued by an awareness of semantics. Response 3. " I " — y o u have to have a subject somewhere (Sy). The use of "subject" suggested that the cue came from the sentence structure of syntax. Response 4. It would have to be "near" or close or showing about the f i r e , I guess (Sm). The subject's reference to the word " f i r e " i n the sentence placed the response i n the semantic category. Following the coding of the o r a l responses, frequency counts were made and percentages calculated to test the research questions of the study. S t a t i s t i c a l Analysis of the Frequencies  Research Question 2 Does the comprehension vary with the mode of discourse? Hypothesis 2.2 There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between the narra-t i v e f i c t i o n mode of discourse and the expository prose mode, as measured 108 by the proportions of responses i n s i x v a r i a b l e s , according to the cues the subject s p e c i f i e d : grammatical function, syntax, semantics, d i s -course, l i f e experience, and no reason, over grade nine and twelve. In grade 9 (Table 4.19), the differ e n c e between the frequencies of protocols of three good subjects on the narrative mode and three good subjects on the expository mode was s t a t i s t i c a l l y s i g n i f i c a n t at .05 l e v e l ( C h i 2 = 13.07). In grammatical function, l i f e experience, and discourse, the d i f -ferences were minimal. The frequency i n the no reason category increased s l i g h t l y from 12 percent to 16.6 percent. The greatest changes occurred i n the synta c t i c and semantic a c c e p t a b i l i t y categories. Good subjects obtained a frequency i n the syntax of 22.7 percent and i n the semantic of 32.7 percent with n a r r a t i v e , compared to 30 percent and 24 percent, r e s p e c t i v e l y , with the expository mode. Between two poor subjects i n na r r a t i v e f i c t i o n and three subjects i n expository prose, the di f f e r e n c e i n the frequency of responses was s t a t i s t i c a l l y more diverse, at .001 l e v e l of s i g n i f i c a n c e (Chi = 339.27). The grammatical function and l i f e experience categories had the lowest frequencies. From narrative to expository, the syntax cate-gory decreased from 19 percent to 6.7 percent; the semantic category increased from 23 percent to 30.1 percent; the discourse decreased from 41 percent to 13.3 percent; and the no reason category increased from 12 percent to 48 percent. The n u l l hypothesis that there was no s i g n i f i c a n t d i f f e r e n c e between the nar r a t i v e f i c t i o n and the expository prose mode was rejected for grade 9 subjects. 109 TABLE 4.19 FREQUENCIES AND PERCENTAGES FROM NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE NINE WITH CHI SQUARES3 Protocols Good(3) Narrative b % Poor(l) % Good(3) Expository % Poor(3) % Grammatical Function 2 1.3 1 1 0 0 Syntactic A c c e p t a b i l i t y 34 22.7 19 19 45 30 10 6. 7 Semantic A c c e p t a b i l i t y 49 32.7 23 23 36 24 46 30. 1 Discourse A c c e p t a b i l i t y 34 22.7 41 41 33 22 20 13. 3 L i f e Experience 13 8.7 4 4 11 7.3 2 1. 3 No reason 18 12 12 12 25 16.7 72 48 Tot a l 150 100 150 150 a N a r r a t i v e x Expository (Good) C h i 2 = 13.07* Narrative x Expository (Poor) C h i 2 = 339.27** Good x Poor (Narrative) C h i 2 = 69.68** Good x Poor (Expository) C h i 2 = 185.14** Number of subjects i n brackets * p < .05 ** p < .001 11D In grade 12 (Table 4.20), the difference between the frequencies of the responses of three good subjects on the narrative mode and three good subjects on the expository mode was s t a t i s t i c a l l y s i g n i f i c a n t ( C h i 2 = 65.05, p < .001). There was no change i n grammatical function and minimal change i n l i f e experience. From narrative to expository, the syntax category decreased from 27.3 percent to 18.7 percent and the semantic increased from 25.3:percentto 31.3; percent. The most marked changes were i n the use of discourse which decreased from 28.7 percent on n a r r a t i v e to 14.7 percent on expository, while the frequency of no reason responses almost doubled from 14 percent to 26 percent. The d i f f e r e n c e i n the frequencies of responses between one poor subject on narrative f i c t i o n and two poor subjects on expository prose was s t a t i s t i c a l l y s i g n i f i c a n t ( C h i 2 = 29.4, p < .001). There were no responses of grammatical function and a minimum change i n l i f e experi-ence from 0 to 1 percent. The syntax category decreased from 34 per-cent to 25 percent, discourse from 8 percent to 12 percent, and no reason from 50 percent to 43 percent. Semantics increased from 8 to 25 percent. The n u l l hypothesis that there was no s i g n i f i c a n t d i f f e r e n c e between the nar r a t i v e and expository mode was rejected for grade 12. In summary, the descriptions of the responses i n the discourse category corroborates the s t a t i s t i c a l data on the e f f e c t on comprehension of the mode of discourse. With one exception, grade 9 subjects were a r t i c u l a t e about the narrative f i c t i o n s e l e c t i o n , "A Kind of Courage" and appeared to use cues from the story. Comments such as "before," "early i n the morning," and " l a t e r " suggested the time element of se t t i n g ; references to the "boulder," "the rock," or the "ledge" I l l TABLE 4.20 FREQUENCIES AND PERCENTAGES FROM NARRATIVE AND EXPOSITORY MODES BY GOOD AND POOR READERS IN GRADE TWELVE WITH CHI SQUARES3 Protocols Good(3) Narrative b % Poor(l) % Good(3) Expository % Poor(3) % Grammatical Function 7 4.7 0 7 4.7 0 0 Syntactic A c c e p t a b i l i t y 41 27.3 17 34 28 18.7 25 25 Semantic A c c e p t a b i l i t y 38 25 4 8 47 31.3 25 25 Discourse A c c e p t a b i l i t y 43 28.7 4 8 22 14.6 6 6 L i f e Experience 0 0 7 4.7 1 1 No reason 21 14 25 50 39 26 43 43 Total 150 50 150 100 N a r r a t i v e x Expository (Good) C h i 2 = 65.05** Narrative x Expository (Poor) C h i 2 = 29.4** Good x Poor (Narrative) C h i 2 = 107.45** Good x poor (Expository) C h i 2 = 34.81 ** Number of subjects i n brackets ** p < .001 112 indicated a knowledge of the place element. A l l subjects used the names of the characters and s p e c i f i e d the r e l a t i o n s h i p s : Davy, the "main character" who was "scared," Clinton, the "challenger" and the "other c i t y boys" who "taunted," "tormented," and "laughed," and Ginny, the "only g i r l around," who "cared f o r " Davy. The c e n t r a l incident of a dangerous dive and the fear of "chickening out" on the dare was men-tioned by a l l f i v e subjects interviewed. On the grade 12 n a r r a t i v e f i c t i o n passage, "David Comes Home," a s i m i l a r use of the discourse cues was apparent. The four subjects men-tioned the time " e a r l i e r " i n the story and the place, "the woods" and "ploughing i n the f i e l d . " The names of characters: Joseph, the "fat h e r , " David, the "son," and E l l e n were used. One subject referred to the f i r s t person " I " narration and another to "writing thoughts." A l l subjects mentioned the "camping t r i p " and were aware the father was reading a l e t t e r from his son who had "gone away," although only one said that the son had died. The comments on discourse for both grades i l l u s t r a t e d an awareness of the patterns of expository prose, but d i f f i c u l t i e s i n comprehension were apparent. The s i x grade 9 subjects reported that "Australian C i t i e s " was about " c i t i e s on the coast," but only one mentioned "cause-e f f e c t , " " h i s t o r y , " and "past time." Subject #39 said the passage was "a story about how f a c t o r i e s come on the coast." One admitted he got "mixed up" and "guessed." In " C u l t i v a t i o n of the Sea and i t s Present E x p l o i t a t i o n " the f i v e grade 12 subjects used conventional r h e t o r i c a l terms such as "essay t o p i c " and "main t o p i c " and referred to the "ocean," "seas," " l i f e i n the ocean," the "sea world," as the subject of the "whole thing." The function of the word "however" to change ideas was 113 noted by subject #212: "They're obviously sort of saying, that although they are t h i s , they aren't good." Subject #207 reported: "You can t e l l — i t ' s l i k e they're contradicting, r i g h t , whatever you said before i s not going to be true." Subject #229 reported that sentence #4 ". . . was a hard one." Subject #207 commented: "I was getting a b i t desperate" and that "I went a l o t by biology and what you know." Subject #212 responded: It ' s so sort of technological. I don't r e a l l y know what goes on i n the ocean. And the a r t i c l e didn't r e a l l y say anything about i t . It said how they found out about i t , but they didn't say anything about what was keeping l i f e going, so you have to sort of guess at words that sound l i k e they'd go i n there. Other comments such as "I don't know" and "confusing" also indicated the e f f e c t on comprehension of expository passages compared to the nar r a t i v e f i c t i o n passages. Research Question 3 Do students at two l e v e l s of p r o f i c i e n c y , good and poor, vary i n comprehension? Hypothesis 3.3 There i s no s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e i n reading compre-hension between good and poor readers, as measured by the frequencies of protocols i n s i x v a r i a b l e s : grammatical function, syntax, semantics, discourse, l i f e experience, and no reason, over grade nine and twelve. In grade 9 (Table 4.19), there was a s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between the frequencies of the responses by three good sub-j e c t s and two poor subjects i n the narrative mode ( C h i 2 = 69.68, p < .001). There was no di f f e r e n c e (12 percent) i n the no reason category. Decreases were i n grammatical function from 2 percent to 1 percent; s y n t a c t i c a c c e p t a b i l i t y from 22.6 percent to 19 percent; semantic 114 a c c e p t a b i l i t y from 30.6 percent to 23 percent; l i f e experience from 8.7 percent to 4 percent. The most marked1 d i f f e r e n c e was i n discourse a c c e p t a b i l i t y which increased from 22.6 percent for good subjects to 41 percent for poor subjects. On the expository mode the C h i 2 value of 185.14 demonstrated a s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e between three good and three poor subjects at the .001 l e v e l of s i g n i f i c a n c e . Neither group used the grammatical function category. There was an increase from 24 percent to 30.7 percent i n the semantic category. Discourse decreased from 22 percent to 13 percent, and l i f e experience from 7.3 percent to 2 percent. The most marked changes were i n syntax, which decreased from 30 percent fo r good subjects to 6.7 percent for poor subjects, and i n the number of responses i n the no reason category, which increased from 16.6 percent to 48 percent, r e s p e c t i v e l y . The n u l l hypothesis that there was no s i g n i f i c a n t d i f f e r e n c e between good and poor subjects was therefore rejected, for grade nine. In grade 12 (Table 4.20), the di f f e r e n c e between the frequencies of the responses of three good subjects and one poor subject on the narrative mode was s t a t i s t i c a l l y s i g n i f i c a n t ( C h i 2 = 107.45, p < .001). No subjects used l i f e experience responses. Grammatical function decreased s l i g h t l y from 4.7 percent to no response; syntax increased from 27.3 percent to 34 percent. Marked changes were decreases i n the semantic category (from 25 percent to 8 percent) and i n discourse (28.7 percent to 8 percent); while the no reason category increased from 14 percent for good subjects to 50 percent for the poor subject. On the expository mode there was a • s t a t i s t i c a l l y s i g n i f i c a n t d i f -ference between three good subjects and two poor subjects ( C h i 2 = 34.81, 115 p < .001). Grammatical function decreased from 4.7 percent to 0; l i f e experience from 4.7 percent to 1 percent; semantics from 31.3 percent to 25 percent; discourse from 14.6 percent to 6 percent, for good and poor subjects, respectively. The most marked differ e n c e was i n the no reason category, which increased from 26 percent to 43 percent. The n u l l hypothesis that there was no s i g n i f i c a n t d i f f e r e n c e i n comprehension between good and poor subjects was rejected f o r grade twelve. Summary of Findings The analysis of the responses to cloze deletions i n two modes of discourse included s t a t i s t i c a l and q u a l i t a t i v e r e s u l t s , presented i n terms of the replacements: exact, non-exact, and retrospective v e r b a l -i z a t i o n s , over p r o f i c i e n c y l e v e l s (good and poor) and grade l e v e l s (nine and twelve). Moderate i n t e r c o r r e l a t i o n s were observed between selected indices of reading comprehension: the c r i t e r i o n measure, Iowa  S i l e n t Reading Test (ISRT), exact cloze scores on narrative f i c t i o n (ECN), and exact cloze scores on expository prose (ECE), over grade and over p r o f i c i e n c y . No c o r r e l a t i o n and low moderate c o r r e l a t i o n s were observed between the selected indices of comprehension and at t i t u d e , as measured by the Estes Reading Attitude Scale (ERAS). S t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e s , at the .001 l e v e l of s i g n i f -icance, i n mean reading comprehension scores were established between modes of discourse (narrative f i c t i o n and expository prose), over each grade and p r o f i c i e n c y l e v e l . The s t a t i s t i c a l analysis of the di f f e r e n c e between means indicated that exact cloze scores on narrative f i c t i o n exceeded exact cloze scores on expository prose, f o r a l l students tested i n grade nine and twelve. S t a t i s t i c a l analysis of the differences between means showed that the scores of 20 good subjects and 20 poor sub-j e c t s on na r r a t i v e f i c t i o n exceeded the scores on expository prose, as measured by exact cloze replacements. No s t a t i s t i c a l l y s i g n i f i c a n t i n t e r a c t i o n was noted. On the four v a r i a b l e s , grammatical function, s y n t a c t i c a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse accept-a b i l i t y , the means of the weighted scores of the non-exact-replacements on na r r a t i v e f i c t i o n were s t a t i s t i c a l l y greater than on the weighted scores of expository prose for good and poor grade 9 subjects (p < .001). No s t a t i s t i c a l l y s i g n i f i c a n t differences were noted for good grade 12 subjects, with the exception of discourse a c c e p t a b i l i t y , where the expository mode was s i g n i f i c a n t l y d i f f e r e n t from the narrative mode. There were no s t a t i s t i c a l l y s i g n i f i c a n t differences f o r the poor group. S t a t i s t i c a l l y s i g n i f i c a n t differences (p < .001) i n mean reading comprehension scores were found between good and poor subjects i n grade nine and grade twelve. As measured by the exact cloze replacements and by non-exact cloze replacements, the scores of good subjects exceeded the scores of poor subjects, at each grade, nine and twelve. In post hoc analysis the means of the exact cloze plus complete discourse a c c e p t a b i l i t y score on nar r a t i v e f i c t i o n exceeded the mean score on expository .prose, at a s t a t i s t i c a l l y s i g n i f i c a n t l e v e l of .001, for both good and poor grade 9 subjects. Interaction was noted (p < .001). In grade twelve the scores of good subjects exceeded scores of poor subjects (p < .001). No s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e i n means on modes of discourse and no i n t e r a c t i o n was found-In post hoc analysis of complete discourse a c c e p t a b i l i t y scores, the means on nar r a t i v e f i c t i o n exceeded the means on expository prose of 117 good and poor grade 9 subjects. Interaction was obtained (p < .01). In grade twelve, the scores of good subjects exceeded the scores of poor subjects (p < .001); however, expository prose means exceeded narrative f i c t i o n means (p < .001), with i n t e r a c t i o n demonstrated (p < .001). The mean r e l i a b i l i t y of int e r - s c o r e r agreement was 91,6 percent, f o r 200 items tested i n narrative f i c t i o n i n grade nine and for 200 items i n expository prose i n grade twelve. From the retrospective v e r b a l i z a t i o n s of eleven grade 9 subjects and nine grade 12 subjects, f i v e categories were induced by the i n v e s t i -gator: grammatical function (GF) , syntax (Sy), semantics (Sm), discourse (D), l i f e experience (LE). The s i x t h category, no reason (NR), was established for cloze replacements without responses, either because subjects were unable to give reasons or f a i l e d to do so. Corrections, miscues from o r a l reading, and omissions were l a t e r collapsed into category 6 because of low frequency. S t a t i s t i c a l l y s i g n i f i c a n t d i f f e r -ences i n frequency of protocols were obtained beyond the .05 l e v e l of s i g n i f i c a n c e between nar r a t i v e f i c t i o n and expository prose, and between good and poor subjects at grade nine and at grade twelve. Summary The treatment of the data from the cloze responses of secondary students to two modes of d i s c o u r s e — n a r r a t i v e f i c t i o n and expository p r o s e — i n c l u d e d both quantitative and q u a l i t a t i v e procedures. Corre-l a t i o n s were computed among the c r i t e r i o n measure: the Iowa S i l e n t Reading Test (IRST), exact cloze scores on narrative f i c t i o n (ECN), exact cloze scores on expository prose (ECE), and scores on the Estes  Reading Attitude Scale (ERAS). The c a l c u l a t i o n of t-values determined 118 s i g n i f i c a n t differences between exact cloze scores on modes of discourse and between the variables of non-exact-replacements: grammatical func-t i o n , s y n t a c t i c a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y . Analysis of variance procedures were used to c a l c u l a t e the s i g n i f i c a n t differences between means of the main e f f e c t s of mode of discourse and p r o f i c i e n c y f o r exact cloze scores. Analysis of variance was also used when the v a r i a b l e was exact cloze scores plus complete discourse a c c e p t a b i l i t y scores, that i s , the number of responses acceptable on three v a r i a b l e s : s y n t a c t i c a c c e p t a b i l i t y , semantic a c c e p t a b i l i t y , and discourse a c c e p t a b i l i t y , as measured by the t o t a l number of "yes" responses i n discourse a c c e p t a b i l i t y . Inter-rater agreement among four judges was calculated by the Arrington Formula ( F e i f e l & Lorge, 1950). Q u a l i t a t i v e d e s c r i p t i v e procedures were used to induce the categories of c l a s s i f i c a t i o n from the retr o s p e c t i v e v e r b a l -i z a t i o n s . The frequencies of protocols were tested f o r s i g n i f i c a n c e by Chi-square. CHAPTER V SUMMARY, CONCLUSIONS, DISCUSSION, AND IMPLICATIONS This exploratory study investigated the r o l e of exact and non-exact-replacements of cloze responses i n the assessment of reading com-prehension. Two modes of discourse, narrative f i c t i o n and expository prose, were investigated. Two t h e o r e t i c a l assumptions guided the study: from p s y c h o l i n g u i s t i c s , that reading involves responses to the grapho-phonemic, s y n t a c t i c , and semantic cue systems of language (Goodman, 1976a); and from discourse a n a l y s i s , that a "schema" or cognitive map d i r e c t s the reader i n the search for discourse cues (Winograd, 1977). Subjects were p r o f i c i e n t and le s s p r o f i c i e n t secondary school students at two l e v e l s of maturity. Attitude to reading was also examined. Operational d e f i n i t i o n s of discourse were: ( i ) narrative f i c t i o n or conventions of a story, and ( i i ) expository prose or coherent explanation of a t o p i c . Subjects were entering grades nine (N = 107) and twelve (N = 100) i n Lord Byng Secondary School, Vancouver, B r i t i s h Columbia. To examine exact replacements (E.R.'s) and a t t i t u d e s , only subjects whose primary language was English were e l i g i b l e . Exact and non-exact-replacements (N.E.R.'s) were examined for a random sample at each grade l e v e l of Good (N = 20) and Poor (N = 30) readers i d e n t i f i e d from scores on the compre-hension subtest of the Iowa S i l e n t Reading Test (1973) . To v e r i f y how l i n g u i s t i c cues t r i g g e r responses, s i x subjects were randomly drawn from 119 120 each p r o f i c i e n c y group for retrospective v e r b a l i z a t i o n interviews, which were taped and transcribed. Each subject (N = 207) completed the Estes  Reading Attitude Scale and two cloze tasks: a narrative f i c t i o n and an expository prose, from the B r i t i s h Columbia Reading Assessment 1977, Grades 8 and 12. Responses were tested f o r exact match to the author's word (Bormuth, 1975). To evaluate N.E.R.'s, the inv e s t i g a t o r adapted the Cambourne Reading Assessment Procedure (1978), based on the Goodman  Taxonomy of Reading Miscues (1969). Following two p i l o t studies, the c l a s s i f i c a t i o n scheme was made consistent with discourse theory and the coding s i m p l i f i e d . A synonym replacement for the exact response was acceptable i n three categories: syntax, semantics, and discourse. S t a t i s t i c a l procedures included c o r r e l a t i o n , independent t - t e s t s , and two-way analysis of variance. For the o r a l protocols, categories were induced from the t r a n s c r i p t i o n s . Frequency of response was analyzed using the chi-square s t a t i s t i c , supported by q u a l i t a t i v e d e s c r i p t i o n . A t t i t u d e to reading had a generally weak c o r r e l a t i o n with the selected indices of comprehension. For exact cloze scores, r e l a t i o n s h i p s with the standardized measure were s i g n i f i c a n t , p a r t i c u l a r l y with exposi-tory prose for poor grade twelve subjects. Narrative f i c t i o n scores exceeded expository prose scores. Good readers were d i f f e r e n t i a t e d from poor readers. The N.E.R. score discriminated between p r o f i c i e n c y l e v e l s . At grade nine, n a r r a t i v e f i c t i o n scores exceeded expository prose, but at grade twelve, differences were not s i g n i f i c a n t . The mean i n t e r - r a t e r agreement, calculated by the Arrington Formula ( F e i f e l & Lorge, 1950), was 91.6 percent. 121 The interviews demonstrated that three cue systems operated most frequently: syntax, semantics, and discourse; and two much less often: grammatical function and l i f e experience. S i g n i f i c a n t differences i n frequency were found between modes of discourse and p r o f i c i e n c y l e v e l s . For combined exact scores plus synonyms, i n grade nine, n a r r a t i v e f i c t i o n scores exceeded prose scores, but i n grade twelve the reverse occurred. Discrimination between p r o f i c i e n c y l e v e l s was noted; how-ever good readers were s i g n i f i c a n t l y superior with expository prose. Data based conclusions were (1) at t i t u d e i s not correlated with either p r o f i c i e n c y or comprehension, (2) comprehension scores d i f f e r e d f o r modes of discourse: narrative f i c t i o n and expository prose, (3) exact cloze score discriminated between p r o f i c i e n c y l e v e l s , (4) N.E.R. scores revealed differences i n the use of cue systems by a b i l i t y groups, (5) a l l readers used the same cue systems: syntax, semantics, and d i s -course, to gain meaning, but control of the set of cue systems, especi-a l l y with expository prose, distinguished the good reader, and (6) the addition of synonym scores to exact cloze scores d i f f e r e n t i a t e d between p r o f i c i e n c y l e v e l s , modes of discourse, and maturity l e v e l s . Discussion of the Findings Within the l i m i t a t i o n s described i n Chapter I, the findings are discussed i n r e l a t i o n to a t t i t u d e and to the kinds of responses to the cloze task. The r e s u l t s of the a t t i t u d e scores contradict the b e l i e f that p o s i t i v e attitudes are c l o s e l y associated with high performance. In grade nine, high a t t i t u d e scores were p o s i t i v e l y related to comprehension only when the expository scores for good readers were examined. In 122 grade twelve the only s i g n i f i c a n t r e l a t i o n s h i p was between a t t i t u d e and the standardized test scores f o r poor readers. These r e s u l t s , i n sup-port of Roettger's (1980) study with elementary students, indicate that secondary students view reading as a tool f o r learning and equate s k i l l i n reading with good grades. In accordance with the views of P e r f e t t i and Lesgold (1977) compre-hension scores are affected by mode of discourse. Findings from means on the variables of non-replacement-cloze scores, with the exception of good readers on the discourse a c c e p t a b i l i t y of expository prose, indicate the highly thematized elements of nar r a t i v e f i c t i o n are more e a s i l y processed than the r h e t o r i c a l schemas of expository mode. Interviews suggested the operation of a "story grammar" (Rumelhart, 1975), based on cues such as se t t i n g , p l o t and characters. Responses indicated a d i f f i c u l t y with the i m p l i c i t premises (Grimes, 1975) of expository prose. However, the inte r a c t i o n s of the post hoc analyses suggested that, while grade nine subjects found expository prose d i f f i c u l t , the differences were greater for poor readers than good readers. In grade twelve, where the scores on complete discourse a c c e p t a b i l i t y scores on expository exceeded narrative f i c t i o n , the i n t e r a c t i o n demonstrated the greater pro-f i c i e n c y of good readers on the expository mode. The f i n d i n g was substan-t i a t e d by the score, on the discourse a c c e p t a b i l i t y v a r i a b l e . It appears that greater c o n t r o l over the cue systems of the expository mode d i s t i n -guishes the good reader. The s u p e r i o r i t y of the more mature secondary reader on expository prose, regardless of p r o f i c i e n c y l e v e l , may be a r e s u l t of four years experience with the "language of schooling" (Olson, 1977), i n which the needs of "reading to lea r n " take precedence over reading f o r pleasure. 123 The broader cognitive map or "top-down" schema (Kintsch, 1977) of the older student further guides the process of pattern p r e d i c t i o n i n exposi-tory prose. The r e s u l t s may also be an a r t i f a c t of the passages chosen; further investigations with other passages i n s i m i l a r modes are needed. The moderate i n t e r c o r r e l a t i o n s of the comprehension measures sup-ports the r e l i a b i l i t y and v a l i d i t y of exact cloze scores reported by Rankin and Culhane (1969). The well-documented, questionable accuracy of standardized tests (Farr, 1969) recommends the cloze test to classroom teachers. Although the c o r r e l a t i o n s i n t h i s i n v e s t i g a t i o n were lower than those reported by such researchers as Jones and P i k u l s k i (1974), much of th i s d i f f e r e n c e may have resulted from the r e l a t i v e l y homogeneous nature of the sample from one school. Cloze tests on s i m i l a r materials and i n other subject areas are needed to substantiate the fin d i n g s . However, i n support • of Hittleman (1978) •, exact cloze scores have u t i l i t y as an i n i t i a l group screening device. The r e s u l t s of the analysis of non-exact-replacements substantiates Cambourne.'s (1978) fin d i n g s . At both grade l e v e l s , good readers were more e f f i c i e n t than poor readers in the use of grammatical function, s y n t a c t i c , semantic, and discourse cues. The taxonomy appears to integrate the advantages of- cloze procedure and miscue a n a l y s i s . The t r a d i t i o n a l anal-y s i s of o r a l responses provides information about i n d i v i d u a l comprehension problems, but f o r the secondary teacher i t i s both time-consuming and unsuitable f o r the evaluation of s i l e n t reading. Miscue. an a l y s i s , applied to the evaluation of acceptable synonyms, has value i n the diagnosis of strengths and weaknesses of the cue systems of language. The ease of administration, r e l i a b i l i t y of r e p l i c a t i o n , and r e l a t i v e s t a b i l i t y of scoring, recommend the use of the combination procedure of cloze and 124 miscue analysis as a group diagnostic t o o l i n the content area classroom. Such a t o o l i s presently lacking (Ekwall, 1976, p. 290). The r e s u l t s of the analyses from the taxonomy and from the r e t r o -spective v e r b a l i z a t i o n s , although the sample was small, support psycho-l i n g u i s t i c theory. Reading i s a process that involves an integrated response to the components of a l i n g u i s t i c cue system. In accordance with Goodman (1976a), a l l readers used the syntactic organization of the sentence and the semantic cues of meaning within the sentence. In the pr e d i c t i v e , non-visual process, they also searched for cues from the mode of discourse (Kintsch, 1978) . The lack of emphasis on the grammatical function of words suggests that the category be eliminated from the c l a s -s i f i c a t i o n scheme. Moreover, the close concurrence with the syntactic a c c e p t a b i l i t y category suggests i t s unimportance, at least f or native English speakers. The interviews further suggest that f o r teaching pur-poses, the evaluation of the a c c e p t a b i l i t y of synonyms f o r syntax, seman-t i c s , and discourse be s i m p l i f i e d to either "yes" or "no." The extra information did not warrant the d i f f i c u l t y i n scoring, p a r t i c u l a r l y i n the semantic category. F i n a l l y , the post hoc findings demonstrated the e f f e c t of synonyms i n the evaluation of reading by cloze tests (Vaughan, Tierney & Alport, 1977; Cambourne, 1978). Vaughan and Meredith (1978) postulated on "the i n t r i g u i n g prospect" that the s y n t a c t i c a l l y acceptable score " . . . may be the most accurate index of students' comprehension as measured by cloze scores because i t does include both.syntactic and semantic aware-ness" (p. 179). This taxonomy f o r determining the a c c e p t a b i l i t y of syno-nyms evaluates discourse awareness. For example, i n grade twelve, the addition of synonyms to the exact cloze scores showed that the e f f i c i e n c y 125 i n expository prose exceeded narrative f i c t i o n , regardless of a b i l i t y . In grade nine, even though the exact cloze scores of good readers were low, the high number of acceptable synonyms indicated an awareness of meaning i n expository prose. Furthermore,.regardless of mode of d i s -course, at both maturity and grade l e v e l s , as the i n t e r a c t i o n showed, good readers exhibited greater control over the use of cue systems to obtain suitable responses than poor readers. The addition of the complete discourse a c c e p t a b i l i t y scores to the exact cloze score, therefore, a f f e c t s the c r i t e r i o n l e v e l s established by Bormuth (1968). For exact scores only (Table 5.1), narrative f i c t i o n scores were at the i n s t r u c t i o n a l l e v e l (44 percent), with the exception of good grade twelve readers who were at the lower end of the independent l e v e l (58 percent). In expository prose, these good grade twelve read-ers were almost at the f r u s t r a t i o n l e v e l , while a l l other scores were well below 43 percent. The combined score (Table 5.2) altered the placement i n n a r r a t i v e f i c t i o n to at, or near, independent l e v e l . In expository prose, only poor grade nine readers remained at the f r u s t r a t i o n l e v e l . Good readers at both grades became independent. Poor grade twelve readers were at the i n s t r u c t i o n a l l e v e l . Implications Implications f o r teaching f o r content area reading materials and for the assessment of reading were i d e n t i f i e d from t h i s i n v e s t i g a t i o n of responses. The s i g n i f i c a n t e f f e c t s of mode of discourse on comprehension sug-gests that younger readers and the less p r o f i c i e n t may benefit from i n s t r u c t i o n i n the r h e t o r i c a l cues of expository prose and i n the patterns of organization, such as cause and e f f e c t . Discussion of the i m p l i c i t TABLE 5.1 COMPARISON BETWEEN CLOZE CRITERION LEVELS AND MEAN PERCENTAGES ON EXACT CLOZE SCORES ON MODES OF DISCOURSE BY PROFICIENCY AND GRADE LEVELS Grade N P r o f i c i e n c y Narrative Expository 9 107 46.5% 27.8** 12 100 49.2 34.6** 9 20 good 53.4 35.5** 20 poor 45.0 23.1** 12 20 good 59.2* 44.3 20 poor 43.3 31.9** TABLE 5.2 COMPARISON BETWEEN CLOZE CRITERION LEVELS AND MEAN PERCENTAGE EXACT CLOZE PLUS COMPLETE DISCOURSE ACCEPTABILITY SCORES ON MODES OF DISCOURSE BY PROFICIENCY AND GRADE LEVELS Grade Pro f i c i e n c y Narrative Expository 12 20 20 20 20 good poor good poor 91.5%* 79.1* 73.30* 54.80 64.90* 41.90*'' 75.10* 49.10 independent Level 58-100% I n s t r u c t i o n a l Level 44-57% ** F r u s t r a t i o n Level below 43% (Bormuth, 1968) 127 premises of the author i s desirable at a l l l e v e l s . The d e s c r i p t i v e data indicated the need to strengthen the a b i l i t i e s of.readers with devices such as parentheses, and p a r t i c i p i a l phrases which caused d i f f i c u l t i e s . The analysis of the cue systems implied a s h i f t i n emphasis from the h i e r a r c h i c a l " s k i l l s " of reading to strategies that c a p i t a l i z e on the l i n g u i s t i c .strengths of students. Cloze exercises, e s p e c i a l l y as a basis for group discussion, f o s t e r the conscious awareness of the patterns of meaning i n words, i n sentences, and i n whole passages. Pre-reading a c t i v i t i e s , such as s e t t i n g the purpose, p r e d i c t i o n exercises, and ques-tion i n g techniques of "what might happen" t r a i n the cognitive map or "schema." The value of the "loud-thinkers" in. discussing the language cues that triggered t h e i r responses i s indicated. The willingness of students to share t h e i r ideas and to discuss how they read suggest that such i n t e r -a c t i o n with teachers i s worthwhile, p a r t i c u l a r l y with content area materials. A multi-faceted approach to the assessment of comprehension i s i n d i -cated by the fi n d i n g s . Exact cloze scores discriminate between l e v e l s of p r o f i c i e n c y , while the examination of non-exact-replacements points up facets of i n d i v i d u a l a b i l i t y with language. The combination of the exact score with the number of acceptable, synonyms indicates i n s t r u c t i o n a l l e v e l s i n modes of discourse. Suggestions f o r Further Research In addition to the i n s t r u c t i o n a l implications, the findings of the exploratory study warrant the following suggestions f o r further study. 1. There i s a need to r e p l i c a t e the study with other samples to deter-mine whether the findings can be generalized. 128 2. The i n v e s t i g a t i o n should be r e p l i c a t e d with other passages from the content areas of t h i s study: s o c i a l studies and biology, and with materials from other d i s c i p l i n e s . 3. The study needs to be extended to other modes of discourse, such as the newspaper report or the argument. 4. Other discourse v a r i a b l e s , such as narrative schemata and r h e t o r i c a l schemas need to be investigated. 5. Further studies are needed to determine the e f f e c t on the use of cue systems of s t y l i s t i c features within passages, such as sentence patterns. A l t e r a t i o n of complex structures to follow speech pat-terns may a f f e c t comprehension. 6. Studies are needed with a population whose f i r s t language i s other than English. Possibly the "grammatical function" category of the taxonomy w i l l be relevant. 7. R e p l i c a t i o n of the study can determine the v a l i d i t y of the c l a s s i f i -cation scheme of the taxonomy. 8. To determine whether r e l i a b i l i t y of scoring i s improved, categories of the taxonomy can be s i m p l i f i e d to "yes" and "no" coding only. 9. Retrospective v e r b a l i z a t i o n , interview from a larger sample at d i f -f e r i n g a b i l i t y l e v e l s are needed to v e r i f y the c r i t e r i a induced from the data, p a r t i c u l a r l y i n the use of discourse category. 10. Analysis i s needed to determine the r e l a t i v e contributions of each protocol category to any differences detected between a b i l i t y groups or modes of discourse. Concluding Statement The value of the evaluation of reading comprehension with content area materials has been demonstrated by t h i s study. The in-depth 129 analysis of the cloze responses of secondary school students provides abundant information about the reading process. The ease and s i m p l i c i t y of t r a d i t i o n a l cloze procedure recommend i t for research and for c l a s s -rooms. The taxonomy for evaluation of the non-exact-replacements func-tions as the s i l e n t reading equivalent of miscue analysis. The assess-ment method employed by the investigator suggests that, j u s t as " a l l mis-cues are not equal" (Beebe., 1976), a l l cloze replacements are not "equal." Some r e t a i n s y n t a c t i c , semantic, and discourse a c c e p t a b i l i t y , and, there-fore, detract l i t t l e from t o t a l comprehension. The evaluation of both exact cloze replacements and non-exact cloze replacements as synonyms i s es s e n t i a l f o r accurate assessment. This i s the f i r s t task and the challenge of educators i n the goal to help students to greater p r o f i c i e n c y i n reading. 130 REFERENCES Alexander, J . , & F i l l e r , R. C. Attitudes and reading. Newark, Del.: International Reading Association, 1976. Ames, W. S. The development of a c l a s s i f i c a t i o n scheme of contextual aids. Reading Research Quarterly, 1966, 2_, 57-85. Anderson, R. C. Schema-directed processes i n language comprehension. In Lesgold, A. , P e l l e g r i n o , J . , Fokkema, S., & Glaser, R. (Eds.) Cognitive psychology of i n s t r u c t i o n . New York: Plenum Press, 1977, 67-82. Asher, S. R., Hymel, S., & Wigfield, A. Influence of topic i n t e r e s t on children's reading comprehension. Journal of Reading Behavior, 1978, 10, 35-47. Athey, I. A f f e c t i v e factors i n reading. In H. Singer & R. Ruddell (Eds.) Theoretical models and processes of reading. Newark, Del.: International Reading Association, 1970. Athey, I. Reading research i n the a f f e c t i v e domain. In R. B. Ruddell (Ed.) Theoretical models and processes Of reading. Newark, Del.: International Reading Association, 1976. B a r t l e t t , F. C. Remembering. Cambridge, England: Cambridge U n i v e r s i t y Press, 1932. Beebe, M. J. Case studies of grade-level e f f e c t s on children's miscues and reading comprehension. Canadian Journal of Education, 1976, 1(2), 51-62. Bleakley, J . J . , & Johnson, E. P. The comprehension of prose: Problems i n measuring learning outcomes. In P. D. Pearson & J . Hansen (Eds.) Reading: D i s c i p l i n e d inquiry i n process and p r a c t i c e . Twenty-seventh yearbook of the National Reading Conference, Clemson, South Carolina: National Reading Conference, 1978, pp. 166-174. Bl u s t e i n , E. S. Reading as a p s y c h o l i n g u i s t i c process: A study of the r e l a t i o n s h i p between problem solving and o r a l reading strategies i n sixth-grade c h i l d r e n (Doctoral d i s s e r t a t i o n , Columbia Un i v e r s i t y Teachers College, 1977). D i s s e r t a t i o n Abstracts International, 1977, 38, 1871A. (University Microfilms No. 77-22, 237.) Bormuth, J . R. Comparable cloze and multiple-choice comprehension te s t scores. Journal of Reading, 1967, 10, 291-99. Bormuth, J . R. Cloze t e s t r e a d a b i l i t y : C r i t e r i o n reference scores. Journal of Educational Measurement,, 1968, _5, 189-196. 131 Bormuth, John R. Factor v a l i d i t y of cloze tests as measures of reading comprehension a b i l i t y . Reading Research Quarterly, 1969, 4_, 358-365. Bormuth, J . R. Development of standards of r e a d a b i l i t y : Toward a r a t i o n a l c r i t e r i o n of passage performance, 1971 (ERIC Document Reproduction Service No. ED 054 233). Bormuth, J. R. The cloze procedure: L i t e r a c y i n the classroom. In W. D. Page (Ed.) Help for the reading teacher: New d i r e c t i o n s  i n research. Urbana, 111.: National Council of Teachers of English, 1975. Bortnick, R., & Lopardo, G. S. The cloze procedure: A multi-purpose classroom t o o l . Reading Improvement, 1973, 13, 113-117. Brazee, P. E. A q u a l i t a t i v e and quantitative d e s c r i p t i o n of eighth grade students' o r a l reading i n both narrative and expository materials (Doctoral d i s s e r t a t i o n , U n i v e r s i t y of Northern Colorado, 1976). D i s s e r t a t i o n Abstracts International, 1977, '37_, p. 3023A. (University Microfilms No. 76-23, 166.) Bridge, C. A., Tierney, R. J . , & Cera, M. J . I n f e r e n t i a l operations of c h i l d r e n involved i n discourse processing. In P. D. Pearson & J. Hansen (Eds.) Reading: D i s c i p l i n e d inquiry i n process and  p r a c t i c e . Twenty-seventh yearbook of the National Reading Con-ference, 1978: Clemson, S.C., National Reading Conference, 68-72. The B r i t i s h Columbia Reading Assessment, 1977, Report"-!: Test r e s u l t s . A report to the M i n i s t r y of Education, Province of B r i t i s h Colum-b i a , V i c t o r i a , 1977. (a) The B r i t i s h Columbia Reading Assessment, 1977, Report 2: I n s t r u c t i o n a l  P r a c t i c e s . A report to the M i n i s t r y of Education, Province of B r i t i s h Columbia, V i c t o r i a , 1977. (b) The B r i t i s h Columbia Reading Assessment, Report 3: Summary. A report to the M i n i s t r y of Education, Province of B r i t i s h Columbia, V i c t o r i a , 1977. (c) Brody, D. A. A p s y c h o l i n g u i s t i c comparison of o r a l reading behavior of p r o f i c i e n t and remedial readers (Master's t h e s i s , Rutgers Uni v e r s i t y , 1973). (ERIC Document Reproduction Service No. ED 085 659.) Broudy, H. S. Types of knowledge and purposes of education. In R. C. Anderson, R. J. Spiro & W. E. Montague (Eds.) Schooling and the  a c q u i s i t i o n of knowledge. New York: John Wiley & Sons, 1977, 1-17. Burke, C. L. Reading miscue research: A t h e o r e t i c a l p o s i t i o n . In P. D. A l l e n & D. J . Watson (Eds.) Findings i n research i n miscue a n a l y s i s : 132 Classroom implications. Urbana, 111. : National Council of Teachers of English, 1976. Burmeister, L. E. Reading strategies for secondary school teachers. Reading, Mass.: Addison-Wesley Pub. Co., 1974. Cambourne, B. L. Getting to Goodman: An analysis of the Goodman model of reading with some suggestions for evaluation. Reading Research  Quarterly, 1976-77, 12, 607-636. Cambourne, B. Some ps y c h o l i n g u i s t i c dimensions of the s i l e n t reading p r o c e s s — a p i l o t study. Paper presented at New Zealand/Australian Reading Conference, Melbourne, 1977. Cambourne, B. The processing of text-book prose: An exploratory study into the s i l e n t reading behavior of readers processing text-book prose. Paper presented at the Fourth A u s t r a l i a n Reading Associa-tion's Annual Conference, Brisbane, 1978. Carlson, K. L. A p s y c h o l i n g u i s t i c d e s c r i p t i o n of selected fourth grade c h i l d r e n reading a v a r i e t y of textual material (Doctoral d i s s e r t a -t i o n , Wayne State U n i v e r s i t y , 1970). D i s s e r t a t i o n Abstracts  International, 1971, j32_, 158A-159A. (University Microfilms No. 71-17243.) Chomsky, N. Syntactic structures. The Hague: Mouton, 1957. Dulin, K. L., & Chester, R. D. A v a l i d a t i o n study of the Estes A t t i t u d e Scale. Journal of Reading, 1974, 18^ , 56-59. Dupuis, M. M. The cloze procedure: Can i t be used with l i t e r a t u r e ? Reading Improvement, 1976, 13, 199-203. Ekwall, E. E. Should r e p e t i t i o n s be counted as er r o r s . The Reading  Teacher, 1974, 2_7, 365-67. Ekwall, E. E. Diagnosis and remediation of the disabled reader. Boston: A l l y n and Bacon, Inc., 1976. Ekwall, E. E., & English, J . Use of the polygraph to determine elemen-tary school students' f r u s t r a t i o n reading l e v e l . F i n a l r e p o r t — United States Department of Health, Education, and Welfare, Pro-j e c t No. OG078, 1971. Ekwall, E. E., S o l i s , J . K., & S o l i s , E. Investigating informal reading inventory scoring c r i t e r i a . Elementary English, 1973, 47, 666-670. Emans, R. Reading theory: Bringing points of view together. Journal  of Reading, 1979, 22, 690-698. English 9: A resource book f or teachers. D i v i s i o n of Public I n s t r u c t i o n , Curriculum Development Branch. M i n i s t r y of Education, Province of B r i t i s h Columbia, V i c t o r i a , B.C., 1978. 133 Estes, T. H. A scale to measure att i t u d e s toward reading. Journal of  Reading, 1971, 8, 134-318. Estes, T. H. Assessing attitudes toward reading. Paper presented at International Reading Association, D e t r o i t , Michigan, 1972. (ERIC Document Reproduction Service No. ED 063 077.) Estes, T. H. Attitude toward reading: A l t e r n a t i v e s i n assessment. Paper presented at Lehigh Reading Conference, 1975. (ERIC Docu-ment Reproduction Service No. ED 105 415.) Estes, T. H., Johnston, J . P., & Richards, H. C. Estes A t t i t u d e Scales. C h a r l o t t e s v i l l e , V i r g i n i a : V i r g i n i a Research Associates Ltd., 1975. Estes, T. H., & Vaughan, J. L. Reading and learning i n the content  classroom: Diagnostic and i n s t r u c t i o n a l s t r a t e g i e s . Boston: A l l y n and Bacon, Inc., 1978. Fareed, A. A. Interpretive responses i n reading h i s t o r y and biology: An exploratory study. Reading Research Quarterly, 1971, 6_, 493-532. Farr, R. Reading: What can be measured? Newark, Del.: International Reading Association, 1969. F e i f e l , H., & Lorge, I. Q u a l i t a t i v e differences i n the vocabulary responses of ch i l d r e n . Journal of Educational Psychology, 1950, 41, 1-18. Fry, E. B. Fry Readability Scale. Providence, R.I.: Jamestown Pub-l i s h e r s , 1976. Geyer, J . T. Comprehensive and p a r t i a l models rel a t e d to the reading process. Reading Research Quarterly, 1972, 7_, 541-587. Gibson, E. J . , & Levin, H. The psychology of reading. Cambridge, Mass.: MIT Press, 1975. Goodman, K. S. A l i n g u i s t i c study of cue and miscues i n reading. Elementary English, 1965, j42, 639-643. Goodman, K. S. The p s y c h o l i n g u i s t i c nature of the reading process. In K. S. Goodman (Ed.) The p s y c h o l i n g u i s t i c nature of the reading  process. D e t r o i t : Wayne State U n i v e r s i t y Press, 1968. Goodman, K. S. Analysis of o r a l reading miscues: Applied psycholin-g u i s t i c s . Reading Research Quarterly, 1969, J5, 9-30. Goodman, K. S. Behind the eye: What happens i n reading. In H. Singer & R. B. Ruddell (Eds.) Theoretical models and processes of reading. Newark, Del.: International Reading Association, 1976. (a) 134 Goodman, K. S. L i n g u i s t i c a l l y sound research i n reading. In R. Farr, S. Weintraub & B. Tone (Eds.) Improving reading research. Papers from the International Reading Association's Preconvention I n s t i t u t e on Research, New Orleans, 1976. (b) Goodman, K. S. Reading: A p s y c h o l i n g u i s t i c guessing game. In H. Singer & R. B. Ruddell (Eds.) Theoretical models arid processes of  reading. Newark, Del.: International Reading Association, 1976. (c) Goodman, K. S., & Burke, C. L. A study of o r a l reading miscues that  r e s u l t i n grammatical re-trarisformatibns. F i n a l report, June 1969. Project number 7-E-219, Contract number OEG-0-8-070219-2806(10), United States Department of Health, Education and Wel-fare, O f f i c e of Education, Bureau of Research. (ERIC Document Reproduction Service No. ED 039 101.) Goodman, Y. M. A p s y c h o l i n g u i s t i c d e s c r i p t i o n of observed o r a l reading phenomena i n selected young beginning readers (Doctoral d i s s e r -t a t i o n , Wayne State University, 1967). D i s s e r t a t i o n Abstracts  International , 1968, 2^9_, 60A. (University Microfilms No. 68-9961.) Goodman, Y. M. Developing reading p r o f i c i e n c y . In P. D. A l l e n & D. J . Watson (Eds.) Findings of research i n miscue a n a l y s i s : Classroom implications. Urbana, 111.: National Council of Teachers of English, 1976. Goodman, Y., & Burke, C. L. Reading miscue inventory manual: Procedures  for diagnosis and evaluation. New York: MacMillan, 1972. (a) Goodman, Y.,.& Burke, C. Reading miscue inventory: Readings for taping. New York: MacMillan Company, 1972. (b) Greene, J . F., & Z i r k e l , P. A. A review of instrumentation to assess reading a t t i t u d e s i n the elementary grades. Paper presented at the Annual Meeting of the American Educational Research Association, San Francisco, 1976. (ERIC Document Reproduction Service No. ED 131 088.) Grimes, J . The thread of discourse. The Hague: Mouton, 1975. Guzak, F. J . Dilemmas i n informal reading assessments. Elementary  English, 1970, 50, 271-274. Hafner, L. E. Relationships of various measures to the cloze. In E. L. Thurston & L. E. Hafner (Eds.) Thirteenth Yearbook Of the National  Reading Conference. Milwaukee, Wis.: National Reading Conference, 1964. 135 Harker, W. J . Introspective-retrospective case studies i n retrospect: Implications for modeling the reading process. In P. L. Nacke (Ed.) I n t e r a c t i o n : Research and p r a c t i c e f o r college-adult reading. Twenty-third yearbook of the National Reading Conference. Clemson, S.C.: National Reading Conference Inc., 1974. H a r r i s , A. J . (Ed.) Advanced s k i l l s i n reading. Toronto: C o l l i e r -MacMillan Canada Ltd., 1968. Herber, H. L. Teaching reading i n content areas. Englewood C l i f f s , N.J.': Prentice H a l l , Inc., 1970 . Hittleman, D. R. Developmental reading: A p s y c h o l i n g u i s t i c perspective. Chicago: Rand McNally College Publishing, 1978. Huey, E. B. The psychology and, pedagogy of reading (1908). Cambridge, Mass. : MIT Press, 1968. Jenkinson, M. E. Selected processes of d i f f i c u l t i e s i n reading compre-hension (Doctoral d i s s e r t a t i o n , U n i v e r s i t y of Chicago, 1957). Jones, M. B., & P i k u l s k i , E. C. Cloze for the classroom. The Reading  Teacher, 1974, 17, 432-438. Johnson, D. D., & Pearson, P. D. S k i l l s management systems: A c r i t i q u e . Reading Teacher, 1975, 28, 757-759. Kerlinger, F. N. Foundations of Behavioral Research. New York: Holt, Rinehart and Winston Inc., 1973. Kintsch, W. On comprehending s t o r i e s . In M. A. Just & P. A. Carpenter (Eds.) Cognitive processes i n comprehension. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates, 1977.' Kintsch, W., & Greene, E. The r o l e of c u l t u r e - s p e c i f i c schemata i n the comprehension and r e c a l l of s t o r i e s . Discourse Processes, 1978, I, 1-13. Kolczynski, R. G. A p s y c h o l i n g u i s t i c analysis of o r a l reading miscues i n selected passages from science, s o c i a l studies, mathematics, and l i t e r a t u r e (Doctoral d i s s e r t a t i o n , Ohio State University, 1973). D i s s e r t a t i o n Abstracts International, 1973, 34, 7108A. (University Microfilms No. 74-10989.) Laing, W. W. An exploratory study of the ways readers i n grades 4, 6, and 8 process the context to obtain word meaning (Doctoral d i s -s e r t a t i o n , University of Alberta, 1974). MacGinitie, W. Research suggestions from the l i t e r a t u r e search. Reading Research Quarterly, 1975-76, 11_, 7-35. Mandler, J . M. A code in the node: The use of story schema i n r e t r i e v a l . Discourse Processes, 1978, 1^, 14-35. 136 Marshall, N., & Glock, M. D. Comprehension of connected discourse: A study into the r e l a t i o n s h i p s between the structure of text and information r e c a l l e d . Reading Research Quarterly, 1978-79, 14, 10-56. M c C a l l i s t e r , J . M. Reading d i f f i c u l t i e s i n studying content subjects. Elementary School Journal, 1930, 31, 191-201. McKenna, M. Synonymic versus verbatim scoring of the cloze procedure. Journal of Reading, 1976, 20, 141-43. Menosky, D. M. A p s y c h o l i n g u i s t i c analysis of o r a l reading miscues generated during the reading of varying portions of text by selected readers from grades two, four, s i x , and eight: A d e s c r i p t i v e study (Doctoral d i s s e r t a t i o n , Wayne State U n i v e r s i t y , 1971). D i s s e r t a t i o n Abstracts International, 1972, J2^ 6108A. (University Microfilms No. 72-14600.) M i l l e r , J . W., & Isakson, R. L. The e f f e c t of syntactic and semantic v i o l a t i o n of high and low reading comprehenders. Paper presented at the S i x t y - f i r s t Annual Meeting of the American Research Associ-ation, San Francisco, 1976. (ERIC Document Reproduction Service No. ED 123 591.) Myers, P. C. The cloze procedures: Latest research and uses. Paper presented at the Twenty-first Annual Meeting of the International Reading Association, Anaheim, 1976. (ERIC Document Reproduction Service No. ED 123 556.) Olshavsky, J. E. Reading as problem solving: An i n v e s t i g a t i o n of s t r a t e g i e s . Reading Research Quarterly, 1976-77, JL2_, 654-674. Olshavsky, J . E. Comprehension p r o f i l e s of good and poor readers across material of increasing d i f f i c u l t y . In P. D. Pearson & J. Hansen (Eds.) Reading: D i s c i p l i n e d inquiry i n process and p r a c t i c e . Twenty-sixth Yearbook of the National Reading Conference, 1978. Clemson, S.C.: Nat i o n a l Reading Conference, 73—76. Olson, D. R. The languages of i n s t r u c t i o n : On the l i t e r a t e bias of schooling. In R. C. Anderson, R. J . Spiro & W. E. Montague (Eds.) Schooling and the a c q u i s i t i o n of knowledge. New York: John Wiley & Sons, 1977. (a) Olson, D. R. From utterance to text: The bias of language i n speech and w r i t i n g . Harvard Educational Review, 1977, 47, 257-281. (b) Otto, J . A. Reading: Cue u t i l i z a t i o n , achievement, and comprehension i n an adolescent population (Doctoral d i s s e r t a t i o n , Indiana State University, 1977). D i s s e r t a t i o n Abstracts International, 1977, 38, 3429A. (University Microfilms No. 77-27, 010.) 137 Page, W. D. A p s y c h o l i n g u i s t i c d e s c r i p t i o n of patterns of miscues generated by a p r o f i c i e n t reader i n second grade, an average reader i n fourth grade, and an average reader i n s i x t h grade encountering ten basal reader selections ranging from pre-primer to s i x t h grade (Doctoral d i s s e r t a t i o n , Wayne State U n i v e r s i t y , 1970). D i s s e r t a t i o n Abstracts International, 1971, 32, 164A. (University M i c r o f i l m No. 71-17296.) Page, W. D. Miscue research and diagnosis. In P. D. Page & D. J . Watson (Eds.) Findings of research i h miscue a n a l y s i s : Classroom  impli c a t i o n s . Urbana, 111.: National Council of Teachers of English, 1976. Palmer, W. S. Defining reading i n proper perspective.. In R. E. Shafer (Ed.) Applied l i n g u i s t i c and reading. Newark, Del.: Interna-t i o n a l Reading Association, 1979. Pearson, G. The a c q u i s i t i o n of knowledge from text. In P. D. Pearson & J . Hansen (Eds.) Reading: D i s c i p l i n e d inquiry i n process and  p r a c t i c e . Twenty-seventh Yearbook of the National Reading Confer-ence. Clemson, S.C.: National Reading Conference, 1978, 61-67. P e r f e t t i , C. A., & Lesgold, A. M. Discourse comprehension and sources of i n d i v i d u a l d i f f e r e n c e s . In M, A. Just & P. A. Carpenter (Eds.) Cognitive processes i n comprehension. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates, 1977. Piaget, J . The language and thought of the c h i l d . New York: Harcourt, Brace, 1926. Piaget, J . Judgement and reasoning i n the c h i l d . London: Routledge & Kegan Paul, 1928. Pickford, R. W. The tendency towards synthesis i n reading. B r i t i s h  Journal of Psychology, 1933, 24_, 50-66. Piekarz, J. A. Individual differences i n i n t e r p r e t i v e response i n reading (Doctoral d i s s e r t a t i o n , U n i v e r s i t y of Chicago, 1954). Poole, D. E. A p s y c h o l i n g u i s t i c d e s c r i p t i o n of comprehension as demon-strated by g i f t e d readers i n grades four and f i v e (Doctoral d i s -s e r t a t i o n , George State University, 1977). D i s s e r t a t i o n Abstracts  International, 1978, ^38, 6583A. (University Microfilms No. 7804938.) Rankin, E. C h a r a c t e r i s t i c s of the cloze procedure as a research t o o l i n the study of language. In P. D. Pearson & J . Hansen (Eds.) Reading: D i s c i p l i n e d inquiry i n process and p r a c t i c e . The Twenty-seventh Yearbook of the National Reading Conference. Clemson, S.C.: National Reading Conference, 1978. 138 Rankin, E. F., & Culhane, J . W. Comparable cloze and multiple-choice comprehension te s t scores. Journal of Reading, 1969, 3-3, 193-199. Rankin, E. F., & Overholser, B. M. Reaction of intermediate grade c h i l -dren to contextual clues. Journal of Reading Behavior, 1969, 1., 50-73. Reading: Language and learning across the curriculum, Report.4. Project B u i l d . Vancouver School Board, 1978. Robinson, H. A. Teaching reading and study s t r a t e g i e s : The content  areas. Boston: A l l y n & Bacon, 1975. Rode, S. S. Development of phrase and clause boundary reading i n ch i l d r e n . Reading Research Quarterly, 1974-5, 10_, 124-42. Roettger, D. Elementary students' a t t i t u d e s toward reading. The  Reading Teacher, 1980, _33, 451-453. Rogers, B. Directed and undirected c r i t i c a l reading responses of high school students (Doctoral d i s s e r t a t i o n , U niversity of Chicago, 1960). Rosen, H. The language of text-books. In A. Cashdan & E. Grugeon (Eds.) Language i n education: A source book. London: Routledge & Kegan Paul, 1972. Rousch, P. D., & Cambourne, B. L. A p s y c h o l i n g u i s t i c model of the reading process as i t re l a t e s to p r o f i c i e n t , average, and low  a b i l i t y readers. Wagga Wagga, N.S.W.: Educational Research and Development Committee, 1979. Ruddell, R. B. An i n v e s t i g a t i o n of the e f f e c t of the s i m i l a r i t y of o r a l and written patterns of language structure on reading compre-hension (Doctoral d i s s e r t a t i o n , Indiana U n i v e r s i t y , 1963). Rumelhart, D. E. Understanding and summarizing b r i e f s t o r i e s . In D. Laberge & S. J . Samuels (Eds.) Basic processes i n reading: Perception and comprehension. New Jersey: Launena Erlbaumo Associates, 1977, pp. 265-303. Rupley, W. H. The cloze procedure. Journal of Reading, 1973, 16, 496-502. Sawyer, D. J . The diagnostic mystique—a point of view. Reading Teacher, 1974, _27, 555-558. Secondary G u i d e — E n g l i s h 8-12. (Revised 1978.) D i v i s i o n of Educational Programs—Schools Curriculum Development Branch, V i c t o r i a . Province of B r i t i s h Columbia: M i n i s t r y of Education, Science and Technology, 1979. 139 Shafer, R. E. (Ed.) Success i n reading. Morristown, N.J.: S i l v e r Burdett Co., 1968. Shafer, R. E. W i l l p s y c h o l i n g u i s t i c s change reading i n secondary schools? Journal of Reading, 1978, 21, 305-316. Shepherd, D. L. Comprehensive high school reading methods. Columbus, Ohio: Charles E. M e r r i l l Pub. Co., 1973. Simons, H. D. Reading comprehension: The need for a new perspective. Reading Research Quarterly, 1970-71, 6^, 338-363. Smith, F. Understanding reading: A p s y c h o l i n g u i s t i c analysis of reading  and learning to read (2nd ed.) New York: Holt, Rinehart & Winston, 1978. Smith, H. K. The responses of good and poor readers when asked to read f o r d i f f e r e n t purposes. Reading Research Quarterly, 1967, 3^, 53-84. Smith, N. B. Be a better reader. Scarborough, Ont.: P r e n t i c e - H a l l , 1975. Smith-Burke, M., Gingrich, P. S., & Eagleeye, D. D i f f e r e n t i a l e f f e c t s of p r i o r context, s t y l e and d e l e t i o n pattern on cloze comprehen-sion. In P. D. Pearson & J . Hansen (Eds.) Reading: D i s c i p l i n e d  inquiry i n process and p r a c t i c e . The Twenty-seventh Yearbook of the National Reading Conference. Clemson, S.C.: National Reading Conference, 1978. Squire, J . The responses of adolescents while reading four short  s t o r i e s . Champaign, 111.: National Council of Teachers of English, 1964. S t a n s e l l , J. C. An exploratory study of perceptions of the reading process and c o n t r o l of that process i n narrative and expository material by selected 9th grade readers (Doctoral d i s s e r t a t i o n , Indiana U n i v e r s i t y , 1977). D i s s e r t a t i o n Abstracts International, 1977, 38, 3260A. (University Microfilms No. 77-27, 013.) S t a n s e l l , J . C , Harste, J . C , & De Santi, R. J. The e f f e c t s of d i f f e r i n g materials on the reading process. In P. D. Pearson & J . Hansen (Eds.) Reading: D i s c i p l i n e d i n q u i r y i n process and  p r a c t i c e . Twenty-seventh Yearbook of the National Reading Confer-ence. Clemson, S.C. : National Reading Conference, 1978. Stern, H. W. A p h i l o s o p h i c a l analysis of the use of comprehension i n an educational context. Reading World, 1973, ]_2(4) , 246-265. Strang, R., & Rogers, C. How do students read a short story? English  Journal, 1965, 54, 819-23, 829. 140 Summers, E. G. Instruments f o r assessing reading a t t i t u d e s : A review of research and bibliography. Journal of Reading Behavior, 1977, 9, 137-165. Summers, E. G. The v a l i d i t y of the Estes Reading A t t i t u d e Scale for intermediate grades. Vancouver: U n i v e r s i t y of B r i t i s h Columbia, (undated). Summers, E. G. An evaluation of the e f f e c t s of a program of sustained s i l e n t reading (SSR) on reading achievement and a t t i t u d e toward reading i n intermediate grades. F i n a l Report. Richmond SSR Project, February 1979. Swain, E. Conscious thought processes used i n the i n t e r p r e t a t i o n of reading materials (Doctoral d i s s e r t a t i o n , U n i v e r s i t y of Chicago, 1953). Taylor, W. L. "Cloze procedure": A new t o o l for measuring r e a d a b i l i t y . Journalism Quarterly, 1953, 30, 415-433. Theobald, B. Applying the technique of miscue analysis to the cloze procedure (Master's t h e s i s , U n i v e r s i t y of Chicago, 1973). Thomas, E. L., & Robinson, H. A. Improving reading i n every c l a s s : A  sourcebook f o r teachers, second e d i t i o n . Boston: A l l y n & Bacon, Inc., 1977. Van Dijk, T. A. Semantic macro-structures and knowledge frames i n discourse comprehension. In M. A. Just & P. A. Carpenter (Eds.) Cognitive processes i n comprehension. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates, 1977, pp. 3-32. Vaughan, J . L., J r . , & Meredith, K. E. R e l i a b i l i t y of the cloze pro-cedure as assessments of various langauge elements. In P. D. Pearson & J . Hansen (Eds.) Reading: D i s c i p l i n e d inquiry i n  process and p r a c t i c e . The Twenty-seventh Yearbook of the National Reading Conference. Clemson, S.C.: National Reading Conference, 1978. Vaughan, J. L., J r . , Tierney, R. J . , & Alpe r t , M. A p s y c h o l i n g u i s t i c analysis of cloze responses. In P. D. Pearson (Ed.) Reading:  Theory, research and p r a c t i c e . Twenty-sixth Yearbook of the National Reading Conference, 1977, pp. 200-202. Vorhaus, R. P. Analysis and comparison of o r a l miscues generated by f i r s t grade students on four d i f f e r e n t reading tasks (Doctoral d i s s e r t a t i o n , U n i v e r s i t y of Pennsylvania, 1976). D i s s e r t a t i o n  Abstracts International, 1976, _37_, 163A. (University Microfilms No. 76-15, 602.) Weber, R. The study of o r a l reading e r r o r s : A survey of the l i t e r a t u r e . Reading Research Quarterly, 1968, j>_, 96-119. 141 Weber, R. M. F i r s t graders' use of grammatical context i n reading. In H. Levin & J . P. Williams (Eds.) Basic studies i n reading. New York: Basic Books, 1970. Weber, R. Review of Findings of Research on Miscue Analysis, edited by P. D. A l l a n & D. Watson. Urbana, 111. : National Council of Teachers of English, 1976. Journal of Reading Behavior, 1977, 9, 416-419. Winograd, T. A framework for understanding discourse. In M. A. Just & P. A. Carpenter (Eds.) Cognitive processes i n comprehension. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates, 1977. Zinck, R. A. An i n v e s t i g a t i o n of semantic and syntactic language cues u t i l i z e d during o r a l and s i l e n t reading (Doctoral d i s s e r t a t i o n , U n i v e r s i t y of Georgie, 1977). D i s s e r t a t i o n Abstracts Inter- n a t i o n a l, 1978, 38, 4635A. (University Microfilms No. 7730523.) APPENDIX A Reading Tests 143 Lord Byng Reading Assessment Name ' Block Student No. ' Grade Boy G i r l Score % Iowa Test Score % Did you usually speak a language other than English before you started Grade 1? Yes _ No Is English the language usually spoken i n your home? Yes No _____ 144 Cloze Test Instructions On the next page i s a sample of a new kind of t e s t . Each of these tests i s made by copying a few paragraphs from a book. Every f i f t h word was l e f t out of the paragraphs, and blank spaces were put where the words were taken out. Your job w i l l be to guess what word was l e f t out of each space  and to write that word i n that space. It w i l l help you i n taking the test i f you remember these things: 1. Write only one word i n each blank. 2. Try to f i l l every blank. Don't be a f r a i d to guess. 3. You may skip hard blanks and come back to them when you have f i n i s h e d . 4. Wrong s p e l l i n g w i l l not count against you i f we can t e l l what word you meant. 5. Most of the blanks can be answered with ordinary words but a few w i l l be numbers l i k e 3,427 or $12 or 1954 contractions l i k e can t or weren t abbreviations l i k e Mrs. or U.S.A. parts of hyphenated words l i k e . s e l f — i n the word self-made (Bormuth, 1976, p. 70) 145 Sample Test Here i s a sample of one of these t e s t s . F i l l each blank with the word you think was taken out. You may check your paper when you f i n i s h i t by looking at the answers which are written upside down at the bottom of the page. Write neatly. The Beaver Indians c a l l s beavers the " l i t t l e men of the woods." But they r e a l l y so very l i t t l e . beavers grow to be or four feet long weigh from 30 to pounds. These " l i t t l e men the woods" are busy ' of the time. That why we sometimes say, " busy as a beaver." know how to b u i l d that can hold water. use t h e i r two front to do some of work. Cutting down a with t h e i r four sharp-teeth i s easy. A can cut down a four inches t h i c k i n 15 minutes. n n o q F "61 as-i} *8I : i 3 A B a q • i_\ paauxod *gx 'gj araqa 'yj sned "_x A3u_ 'ZT SUlBp " X T SJ3AB9a '01 S F *6 sx • g asoui • i_ _o - 9 0*7 "5 P U B >17 33iqa '£ 3SOft 'z 3 , U 3 J B *X :sjaasuy (Bormuth, 1975, p. 71) 146 Narrative F i c t i o n : Grade 9 to the uppermost j u t t i n g boulder about t h i r t y feet above the water. He would make i t — h e had to make i t . He had done i t once t h i s morning and he could do i t again. A crowd gathered, and the c i t y on the wharf were too. Davy tensed h i s , ready f o r the dive. he heard a g i r l " : "Don't—don't do i t , !" He looked down and Ginny holding out her to him, begging him stop. Davy stared at "Come down," she c r i e d . " , Davy, come down!" The i n her voice c a u s e d t o hesitate and then back. But when Clinto n , "What's the matter—you out?" he clenched h i s and stepped into diving again. He couldn't back now. He didn't want He knew he could the dive. "Davy ..." There fear i n Ginny's voice. " ; ; , please don't dive," she • He sat down to from d i v i n g , h i s head _ b i s hands. From below the laughter of the _ ; boys, the st r i d e n t hoot Clint o n louder than the . Davy s palms were wet be fought back the _ _ ; to leap up and , no matter what Ginny _• When he looked down, crowd was leaving. Only and Ginny stood watching he came down t h e — s l o w l y for he was ' exhausted. They walked t o w a r d — G i n n y pale and close to , Clinto n smiling condescendingly. looked l i k e a champ that one," Clinto n taunted. clenched h i s f i s t s , but ' l a i d her hand on arm and he slowly He wished he could her how i t was— ' i t was harder to l e t Clin t o n think he was yellow than i t would have been to dive. But he couldn't explain i t exactly, the d i f f e r e n t kind of courage i t had taken. Any kid could have taken the dare and dived o f f the c l i f f , but i t t o o k — w e l l — a man to l e t himself be r i d i c u l e d for something no one would understand. "I wasn't a f r a i d , " Davy said. "I wasn't seared of d i v i n g . " "I know," she said, and slipped her arms through h i s . "But what you did was braver." Answers: 1. had 2. boys 3. watching 4. muscles 5. then 6. cry 7. Davy 8. saw 9. arms 10. to 11. her 12. Please 13. anguish 14. him 15. step 16. shouted 17. chicken 18. f i s t s 19. p o s i t i o n 20. out 21. to 22. make 23. was 24. Davy 25. c a l l e d Narrative F i c t i o n : Grade 9 26. keep 27. i n 28. came 29. c i t y 30. of 31. rest 32. as 33. impulse 34. dive 35. wanted 36. the 37. Clint o n 38. as 39. rocks 40. suddenly 41. him 42. tears 43. You 44. on 45. Davy 46. Ginny 47. h i s 48. relaxed 49. t e l l 50. that 148 Expository Prose: Grade 9 Much of the explanation ' t h i s urban growth i n the h i s t o r y of the ' ' ' of A u s t r a l i a . The s e t t l e r s ... on ships, and n a t u r a l l y f i r s t at the chosen which soon developed i n t o and ports. As the and p a s t o r a l i s t s * pushed inland looked to the established f o r t h e i r supplies, and t h e i r produce back to ports to be shipped the overseas markets. This expanding trade i n the offered employment to new , who tended to s t a y t h e towns rather than ' ' inland. Later, the railways t h e i r networks out from p o r t - c a p i t a l s and accelerated the of t r a f f i c between town ' country, while the d e v e l o p m e n t m i n i n g , and such new ' occupations as wheat a n d f a r m i n g , added greatly to volume of goods h a n d l e d t h e coastal c i t i e s . These became even more important manufacturing began to develop the A u s t r a l i a n market. I t l o g i c a l to b u i l d f a c t o r i e s ' ' the places where there large numbers of workers, the l o c a l markets were , and where other markets be most e a s i l y reached, by use of the network or by coastal So the great c i t i e s continued to grow r a p i d l y , ' few country towns have than doubled i n population '  the l a s t t h i r t y years. means that the c i t i e s o u t over tremendous areas. f a m i l i e s demand s u f f i c i e n t not only to b u i l d houses but also to vegetable gardens, lawns, flowers, shrubs. Hence, Sydney and Melbourne cover areas which are l i t t l e smaller than that of Metropolitan London although t h e i r populations are much l e s s . The other A u s t r a l i a n c i t i e s are proportionately large. Most Australians l i v i n g i n c i t i e s must t r a v e l several miles to t h e i r work each morning. The thinness of settlement so obvious i n the A u s t r a l i a n countryside i s also to be seen i n the c i t i e s , so that transport and other communications, water, sewage, and other services are a l l very expensive to provide and maintain. *farmers Answers: Expository Prose: Grade 9 1. for 26. the 2. l i e s 27. by 3. development 28. centres 4. arr i v e d 29. when 5. s e t t l e d 30. for 6. anchorages 31. was 7. townships 32. i n 8. explorers 33. were 9. they 34. where 10. ports 35. largest 11. sent 36. could 12. the 37. ei t h e r 13. to 38. r a i l 14. s t e a d i l y 39. shipping 15. ports 40. have 16. immigrants 41. while 17. i n 42. more 18. s h i f t 43. i n 19. spread 44. This 20. the 45. spread 21. flow 46. Most 22. and 47. land 23. of 48. t h e i r 24. r u r a l 49. have 25. dairy 50. and 150 Narrative F i c t i o n : Grade 12 was the Z wood' ' ^ ^ W a S C O O l > f i r e . And with ^ f ™ - - w h e r e ^ J j ^ T h T -. c a m p T ^ n d - E h ^ " d<S q«iet and warm d r ° " s y outside i n t h e l ^ k W h l S P e r e d A e t h e r t h e i r .__ had slept i n the ' a ^ T T C r W a S t h e f i r s t n i § h t ^ h i n T T i k e i t . . i Tr' a n d t h e r e h a d n e v ^ ever was. I f W a S t h e b e s t there . day-TikTThaT. § 7 S h° U l d ^ t o i n ^ r j u a t — l y i n g awake a lone t-r,-,^  • . Dad was awake, too. § d T d ^ 7 ^ * A n d 1 - think he knew how 7~ ^ t f a y ^ t h i n g , but always knew how i t " S°meh°W 1 t h i n k "The horses," Joseph said i n a low voice. " forgot the horses." E l l e n seen him t i e the before he came over the house. But she say anything. David! David! said h i s son's name and over m h i s _ when he was again the f i e l d s . He S a l d w l t h a n e x a l t a t i o n , f o r s t i l l n e s s was a l l § o n e from h i s mind, and good kind of tears t i g h t i n h i s throat. had been David who t h i s place best of . And i f your son the place he went from, then he could leave i t I f h e e v e n f o r t h e s e things both loved, then you hear h i s voice i n t h e i r voice s t i l l . He took the spinner from h i s pocket and made a hole deep i n the ground. Then he covered i t over gently with earth where the plow could never reach i t . That would be David's spot, always. He stood for a minute, looking across the v a l l e y from mountain to mountain, and then he turned the horses again into the furrow. The f i e l d had looked long and wide. But now i t seemed an easy f i e l d to plow. It seemed as i f everywhere he looked, David had come home. Answers: 1. the 2. there 3. Father 4. for 5. Father 6. and 7. the 8. trees 9. t a l k 10. It 11. I 12. woods 13. been 14. guess 15. day 16. guess 17. s a t i s f i e d 18. one 19. remember 20. time 21. think 22. He 23. I 24. i t 25. Dad ive F i c t i o n : Grade 12 26. was 27. suddenly 28. I 29. had 30. horses 31. to 32. didn't 33. Joseph 34. over 35. mind 36. i n 37. i t 38. the 39. now 40. the 41. were 42. I t 43. loved 44. a l l 45. loved 46. away 47. never 48. died 49. you 50. could 152 Expository Prose: Grade 12 The Sea—Not L i m i t l e s s The immensity of oceans—covering two-thirds the earth s u r f a c e — a n d f i s h abundance i n some , undoubtedly explain many of exaggerations and o v e r s i m p l i f i -cations presented the p o t e n t i a l i t i e s of the and t h e i r f i s h r i c h e s . i s hardly an a r t i c l e a book on the which does not resort _ terms l i k e "inexhaustible" or " " i n describing the abundance the oceans. It i s , however, to keep i n that these words do ex i s t i n Nature's own . There i s i n the world no such thing ' l i m i t l e s s and inexhaustible. A l l i n the oceans i s t i e d to the primary , i n the top layer, seaweeds and phytoplankton with containing plant pigments capable trapping the sunlight energy into the upper regions the water masses (generally the photic layer, meaning l i g h t r e i g n s ) . Technical devices, ingenious they may be, l i t t l e to free :the and t h e i r l i v i n g organisms' t h i s dependence. The base t h e i r existence i s ulti m a t e l y . pigment-carrying plant organisms, to the l i m i t of penetration, down to a of one hundred feet, more. Here the crea t i v e take place which produce organic matter that sustains . Even the bottom animals, may l i v e thousands of below the surface, feed the "manna" r a i n i n g down ' above. This consists of remainders, animal excretions, dead , and other waste. Sometimes of t h i s organic matter caught by other scavengers on i t s way down and may be cycled once or several times by being converted into l i v i n g matter again before i t reaches the large depths. But a l l t h i s i s regulated by the unchange-able fundamental law of the b i o l o g i c a l world: a l l f l e s h i s grass. Answers: 1. the 2. of 3. the 4. waters 5. the 6. about 7. seas 8. There 9. or 10. subject 11. to 12. " l i m i t l e s s " 13. of 14. important 15. mind 16. not 17. dictionary-l S . l i v i n g 19. as 20. l i f e 21. ul t i m a t e l y 22. production 23. of 24. c e l l s 25. of Expository Prose: Grade 12 26. penetrating 27. of 28. termed 29. where 30. however 31. do 32. waters 33. from 34. for 35. these 36. spread 37. l i g h t ' s 38. depth 39. occasionally 40. processes 41. the 42. l i f e 43. which 44. feet 45. from 46. from 47. plankton 48. bodies 49. part 50. i s APPENDIX B Retrospective Verbalizations Key: 1. Grammatical function GF 2. Syntax Sy 3. Semantics Sm 4. Discourse D 5. L i f e experience LE 6. No reason NR 155 Narrat ive F i c t i o n : Grade 9 Subject #35: Poor Reader "A crowd 1. had gathered, and the c i t y 2. people (boys) on the wharf were 3. looking (watching) too." Responses 1,2,3. It f i t t e d i n with the rest of the story that a l l the c i t y people would be "looking" up with everyone else (D). "Davy tensed h i s 4. muscles ready to dive." Response 4. Well, you have to tense and usually when you dive you tense your "muscles" (Sm). "5. Then he heard a g i r l 6. cry: 'Don't—don't do i t , 7. Davy!' He looked down and 8. holding (saw) Ginny holding out her 9. arms to him, begging him 10. _to_ stop." Response 5. They seemed to complete the sentence—most of them. Well, when he was up on the ledge (D), Response 6. she would have probably c r i e d out or screamed out (Sm), Response 7. and through what she said "Don't—don't do i t , " I put i n "Davy," since h i s name was Davy, and i t f i t t e d i n that she'd by crying i t out (D). Response 8. What I put i n , "holding" didn't r e a l l y f i t (NR). Response 9. "Arms"—when i t said Ginny holding out her arms or some-thing, i t was j u s t that usually when you cry out, you hold out your arms or hands (Sm), Response 10. and begging him "to" stop j u s t f i t i n (Sy). "Davy started (stared) at 11. her." Response 11. He started up the next ledge or something and I decided that he would probably s t a r t going down and then decided he had to be brave and stay up (or a l miscue—NR). "Come down," she c r i e d 12. Please, Davy come down!" Response 12. When you're begging someone you usually say "Please" instead of ordering them to come down (LE). "The 13. tone (anguish) i n her voice caused 14. him to hes i t a t e and then 15. s t a r t (step) back." Response 13. The "tone" i n her voice r e a l l y seemed to f i t more than anything else would (Sm), 156 Response 14. and caused "him" to hes i t a t e (Sy), Response 15. and then " s t a r t " back f i t s i n (Sy) . "But when Clinto n 16. said (shouted), "What's the m a t t e r — you 17. chicken out?" Response 16. Well most things you'd say (Sm), Response 17. would be "chicken" out or ready to jump (LE). "And he clenched h i s 18. f i s t s and stepped into d i v i n g 19. p o s i t i o n again." Response 18. I put " f i s t s " because he'd be mad and usually when you're mad you clench something and i t would be your hands (D), Response 19. and into the di v i n g " p o s i t i o n " again where he's getting ready to j ump (D). "He couldn't back 20. out now." Response 20. The only thing that would r e a l l y f i t into the sentence i s "out" (Sy). "He didn't want 21. to." Response 21. I t ' s j u s t what i t said i n the beginning of the s t o r y — he'd didn't r e a l l y want to jump—they j u s t forced him to (D) . "He knew he could 22. make the dive." Response 22. That's because he had i n the morning (D). "'Davy ... ' There 23. was fear i n Ginny's voice." Response 23. "There"—when you have something l i k e that "was" was the only word that f i t i n (Sy). "24. Don't (Davy), please don't dive," she 25. cryed ( c a l l e d ) . " Response 24,25. Well she cr i e d out before and so she wouldn't change her voice a l l of a sudden (D). "He sat down (to) 26. scared (keep) from d i v i n g , h i s head 27. i n h i s hands." Response 26. Well, I didn't r e a l l y know what to f i t i n outside of the fact he was scared. (Oral miscue—NR) Response 27. His head " i n " h i s hands—usually you do s i t down sulking so (LE). 157 "From below 28. a l l (came) the l a u g h t e r o f the 29. o t h e r ( c i t y ) boys, t h e s t r i d e n t hoot 30. from ( o f ) C l i n t o n l o u d e r t h a n t h e 31. o t h e r s . Davy's palms were wet 32. and (as) he fought back t h e 33. earge (impulse) t o l e a p up and 34. what ( d i v e ) no matter what Ginny ( d i d ) 35. dear (wanted). Response 28. From " a l l " the l a u g h t e r , t h e r e would be l o t s o f p e o p l e t h e r e (D). Response 29. The o t h e r "boys" because t h e r e ' s more than j u s t one (D), Response 30. and the n t h e hoot "from" C l i n t o n l o u d e r — a n y o t h e r word wouldn't f i t ( S y ) , Response 31. than a l l t h e " o t h e r " boys ( S y ) . Response 32. Davy's palms were w e t — t h a t ' s the way th e y g e t , "and" he fought back t h e urge t o l e a p — i t ' s k i n d o f l i k e you combine two s e n t e n c e s ( S y ) , Response 33. and " u r g e " seemed t o f i t i n (Sm). Response 34,35. The next word "what." I d i d n ' t r e a l l y know what t o put i n , f o r t h e next two words, so I d e c i d e d I might as w e l l put "what" and so I thought " d e a r " (NR, NR). "When he l o o k e d down, 36. t h e crowd was l e a v i n g . " Response 36. "The" j u s t seemed t o f i t i n w i t h crowd ( S y ) . "Only 37. C l i n t o n and Ginny s t o o d w a t c h i n g 38. as he came down (the) v e r y ( r o c k s ) s l o w l y f o r he was 40. _o (suddenly) e x h a u s t e d . " Response 37. There was o n l y somebody and Ginny s t o o d w a t c h i n g and i t sounds l i k e i t would be C l i n t o n , s i n c e he won't get him to jump (D), Response 38. and t h e " a s " i s as he came d o w n — t h e y ' d p r o b a b l y be s t a n d i n g " a s " he came down (Sm), Response 39. and he came down " v e r y " s l o w l y , s i n c e he was exhausted ( O r a l m i s c u e — N R ) Response 40. and t h a t ' s t h e " s o " (Sm). "They walked toward 41. h i m — G i n n y p a l e and c l o s e t o 42. him ( t e a r s ) , C l i n t o n s m i l i n g c o n d e s c e n d i n g l y . " Response 41. Walking toward something so I s a i d w a l k i n g toward "him" s i n c e he got down (D), Response 42. and t h e "him" doesn't r e a l l y f i t i n , but i t seemed t o work out (NR). 158 "43. You looked l i k e a champ 44. at (on) that one, taunted C l i n t o n . " Response 43. He was t a l k i n g to Davy and usually when you t a l k to someone, you c a l l them "you" (LE). Response 44. The "at" j u s t f i t i n the sentence (Sy). "45. H_ (Davy) clenched h i s f i s t s , but 46. Ginny l a i d her hand on 47. h i s arm and he slowly 48. relaxed." Response 45. I t ' s l i k e C l i n t o n was tormenting so he, Davy, was kind of upset, so he'd clench h i s f i s t s (D), Response 46. and then "Ginny" because somebody l a i d her hands, and since she was the only g i r l around (D), Response 47. " h i s " arm not somebody else's since he clenched h i s f i s t (Sm), Response 48. and he slowly "relaxed" (NR). "He wished he could 49. t e l l her how i t was—50. But (that) i t was harder to l e t Clint o n think he was yellow than i t would have been to dive." Response 49. He could " t e l l " h e r — l i k e what i t was l i k e up there (D). Response 50. "But" seemed to f i t i n — i n order to make i t seem l i k e i t was hard to t e l l that he was chicken and hard not to jump (D). 159 Expository Prose: Grade 9 Subject #15: Good Reader "Much of the explanation 1. to_ (for) t h i s urban growth 2. i s ( l i e s ) i n the h i s t o r y of the 3. s e t t l e r s (development) of A u s t r a l i a . " Response 1. I put the "to" because i t f i t s numerically (Sy). Response 2. I put i n " i s " for the same reason (Sy). Response 3. I put i n " s e t t l e r s " because the next sentence s t a r t s out with them and i t f i t s i n (D). "The s e t t l e r s 4. s a i l e d (arrived) on ships, and n a t u r a l l y 5. landed (settled) f i r s t at the chosen 6. blank which soon developed i n t o 7. c i t i e s (townships) and ports." Response 4,5. I put "landed" because I figured they'd have to land on shore or whatever (LE). Response 6. Then I put t h i s blank here because I had no idea what t h i s meant (NR). Response 7. I put i n " c i t i e s " because i t seemed to f i t and that's mostly what i t ' s about (D). "And the 8. workers (explorers) and p a s t o r a l i s t i s pushed inland 9. and (they) looked to the established 10. stores (ports) f o r t h e i r supplies, and 11. gave (sent) t h e i r produce back to 12. the ports to be shipped 13. _to_ the overseas markets." Response 8. I put "workers" because workers and farmers (Sm), Response 9. I put "and" (NR), Response 10. I put " s t o r e s " — I wasn't too sure but I guess that's where they get t h e i r supplies. I was a b i t stuck there (LE), Response 11. and "gave" because i t seemed to f i t i n (Sy). Response 12,13. N.R. (omitted). "This 14. new (steadily) expanding trade i n the 15. c i t i e s (ports) offered employment to the new 16. immigrants, who tended to stay 17. i n the towns rather than 18. move ( s h i f t ) i n l a n d . " Response 14. I put "new" because the s e t t l e r s had j u s t come i n (D), Response 15. expanding trade i n the " c i t i e s " because that's what's i t a l l about (D), 160 Response 16. "immigrants" because they were a l l immigrants (Sm), Response 17,18. tended to stay " i n " the towns than "move" inland (Sy). "Later the railway 19. channeled (spread) t h e i r networks out from 20. the p o r t - c a p i t a l s and accelerated the 21. amount (flow) of t r a f f i c between town 22. and country, while the development 23. of mining and such new 24. inland (rural) occupations as wheat and 25. produce (dairy) farming, added greatly to 26. the volume of goods handled 27. _ t (by) the coastal c i t i e s . " Response 19. Well, made the railways " c h a n n e l l e d " — I wasn't too sure but i t seemed to f i t i n with networks (Sm), Response 20. from "the" p o r t ^ c a p i t a l s (Sy), Response 21. and accelerated the "amount"—more t r a f f i c was going i n between (Sm), Response 22. town "and" c o u n t r y — i t j u s t f i t s i n (Sy), Response 23. while the development "of" mining f i t s i n (Sy), Response 24. " i n l a n d " occupations which are things you have to grow on the dry p r a i r i e s (LE), Response 25. "produce" farming j u s t seemed to go with what was going on (D), Response 26. added greatly, to "the" volume of goods (Sy) , Response 27. handled "at" the coastal c i t i e s j u s t seemed to f i t i n (Sy). "These 28. c i t i e s (centres) became even more important 29. when manufacturing began to develop 30. i n (for) the A u s t r a l i a n market." Response 28,29,30. These obviously f i t i n except for " c i t i e s " which I am not too sure about (NR, Sy, Sy). " I t 31. was l o g i c a l to b u i l d f a c t o r i e s 32. i n the places where there 33. were large numbers of workers, 34. and (where) the l o c a l markets were 35. located (largest) and where other markets 36. could be most e a s i l y reached, 37. _o (either) by use of the 38. r a i l r o a d ( r a i l ) network or by coastal 39. ports (shipping)." Response 31. It "was" l o g i c a l to b u i l d f a c t o r i e s f i t s i n (Sy). Response 32. "In" the places seems to f i t i n (Sy). Response 33. There "were" large amounts i s mostly what seems to f i t i n (Sy), Response 34,35. "and" l o c a l markets were "located" I guess seems to f i t 161 i n very well (Sy, Sm), Response 36. and where other markets " c o u l d " — j u s t seemed to f i t i n (Sy), Response 37,38. "so" by use of the " r a i l r o a d , " seeing i t was t a l k i n g about the r a i l r o a d network (Sy, D), Response 39. by coastal "ports"--well, ports on the coast (D). "So the great c i t i e s 40. then (have) continued to grow r a p i d l y 41. for (while) few country towns have 42. grown (more) than doubled i n population 42.^in the l a s t t h i r t y years." Response 40. Great c i t i e s "then" continued to g r o w — l i k e a f t e r i t happened they grow (D), Response 41. " f o r " few country towns—I think I got a b i t mixed up h e r e — i t doesn't seem grammatically correct (NR) , Response 42. for few of the country towns have "grown" (NR), Response 43. then doubled i n population " i n " the l a s t t h i r t y y e a r s — I must have got mixed up (Sy). "44. That (this) means that the c i t i e s 45. spread out over tremendous areas." Response 44,45. That's about a l l you can do i n tremendous areas (Sm, Sm). "46. The (most) f a m i l i e s demand s u f f i c i e n t 47. t o o l s (land) not only to b u i l d 48. t h e i r houses but also to 49. grow (have) vegetable gardens, lawns, flowers, 50. and shrubs." Response 46. "The" f a m i l i e s seems to f i t i n (Sy). Response 48. not only to b u i l d " t h e i r " houses (Sy), Response 47. b u i l d i n g and Response 49. do (Sm), Response 50. 162 Narrative F i c t i o n : Grade 12 Subject #211: Good Reader "And then, l a t e r when 1. the whole night was cool, 2. there was the dry wood (and) 3. X (Father) could f i n d somewhere 4. near (for) the f i r e . " Response 1. "The" was the only word that would make any sense between "when" and "whole night." Another adje c t i v e wouldn't have been needed (GF). Response 2. Blank was the dry wood. I".couldn't s e e — " t h e r e " was the only small word to make sense (Sm). Response 3. Blank could always f i n d somewhere—"I"—you have to have a subject somewhere, so (Sy). Response 4. Somewhere "near" the f i r e — w e l l i t would have to be near or close or showing about the f i r e , I guess (Sm). "And with 5. Father there, i t was dark 6. and quiet and warm i n 7. our camp, and the b i g 8. trees whispered together t h e i r drowsy 9. language (talk) outside i n the dark." Response 5. And with "Father" t h e r e — w e l l , i t had to be with someone, and i n the story i t was Father (D). Response 6. It was dark and—dark blank quiet and warm—presumably i f i t ' s dark, quiet and warm, i t would have to be have to be dark "and" quiet (Sm), Response 7. i n blank camp—well, i t wasn't h i s camp, i t was t h e i r or "our" camp (Sm), Response 8. and the b i g somethings whispered t o g e t h e r — i t would have to be "trees" i n the forest (Sm). Response 9. Their drowsy "blank" outside i n the d a r k — i t could have been the song or something but a language seemed with whispers to make more sense (Sm). "10. that ( i t ) was f i r s t night 11. _I had slept i n the 12. woods, and there had never 13. been anything l i k e i t . " Response 10. That was the only construction. It could have been i t was the f i r s t n i g h t — " t h a t " was (Sy). Response 11. Blank had slept i n the woods. " I " — I mean who else? That's what he's saying (D). Response 12. He could have said the forest or woods but he referred to woods e a r l i e r on (D), 163 Response 13. and there never blank anything l i k e i t — " b e e n " i s the only word that made any sense there. I t ' s j u s t a grammar thing (Sy). "I 14. thought (guess) that was the best 15. night (day) there ever was." Response 14. I could have said I f e l t o r — i t would have to be thought or f e l t (Sm). Response 15. That was the best " n i g h t " — w e l l , i t could have been the best camping t r i p but night was the obvious kind of word to put there (Sm) . "16. I thought (guess) a guy should be 17. allowed ( s a t i s f i e d ) to have had j u s t 18. one day l i k e that." Response 16. I could have said f e l t again, but " t h o u g h t " — i t would have to be something l i k e that (Sm). Response 17. A guy should b e — I thought about able but I thought "allowed" seemed to s u i t what he meant more c l o s e l y (Sm). Response 18. And to have had j u s t — w e l l one day l i k e that made sense. The only thing that where I said one day, i t sort of goes against where I put night up e a r l i e r somewhere. That was the best night there ever was, but I figured, you could have made that kind of thing (Sm). "I 19. lay (remember) l y i n g (oh boy) awake a long 20. time that night." Response 19. I don't know why I put lay " l y i n g " — I suppose I wasn't thinking. You can have l a y l y i n g I suppose but i t ' s u n l i k e l y (NR). Response 20. A long "time" that n i g h t — w e l l what e l s e — t i m e , a long time (Sm). "And, I 21. knew (think) Dad was awake, too." Response 21. I could have said thought or f e l t but "knew" was as good as anything (Sm). "J. (He) didn't say anything, but 23. I think he know how 24. i t was." Response 22. I didn't say anything (Sm), Response 23. but I t h i n k — o n l y made sense (Sm). Response 24. I had trouble with the " i t " was. Somehow i t was hard to f i t anything but j u s t a small word l i k e " i t " between how and was. You had to express somehow the idea of the whole experience, only you can't use the experience, " i t " was the only word that would f i t (D). 164 "Somehow I think 25. he (Dad) always knew how i t 26. f e l t was" Response 25. Obviously i t ' s "he" that he's t a l k i n g about (D). Response 26. Always knew how i t — I could have put "was" again, but I thought " f e l t " would express i t (D). "The horses," Joseph said 27. suddenly i n a low voice." 28. I_ forgot the horses." Response 27. Joseph said i n a low voice f i t s . Obviously you j u s t have to modify the sai d . You could have said Joseph said worriedly i n a low voice, but since i t sort of i n t e r r u p t s , Ii thought suddenly would be a good way to put i t (Sm). Response 28. " I " forgot the horses i s obvious. What else, you know. It could have been you forgot the horses, but that would have been a whole new—there's no mention that David was i n charge of the horses. It seems l i k e Joseph was (D). " E l l e n 29. had seen him t i e the 30. horses before he came over 31. tjo the house." Response 28. That proves that i t was Joseph who had forgotten some-thing, (D) . Response 29. "Had" seen him i s j u s t grammar (Sy). Response 30. Tie the blank before he came over—presumably, he's s t i l l t a l k i n g about horses (D). Response 31. Before he came over "to"—what else would you put a f t e r over to? There's no other prep o s i t i o n that goes with over, i n that sentence anyway (GF). "But she 32. didn't say anything." Response 32. Again j u s t grammar, nothing els e . But she couldn't—why couldn't she. No i t had to be didn't say anything (Sy). "David! David! 33. Joseph said h i s son's name 34. over and over i n h i s 35. mind when he was again 36. ixi the f i e l d s . " Response 33. I t ' s obviously "Joseph" saying h i s son's name (Sm) . Response 34. "Over" and over i s the usual sort of thing to put (Sy). Response 35. In h i s "mind"—well he could have said i t , but don't think he said i t out loud somehow. It ' s not indicated. There's not enough quotation marks. I t ' s more as i f he was ju s t thinking i t (Sy). Response 36. "In" the f i e l d i s n ' t a very strong way of putting i t . I 165 started again, when he was again ploughing the f i e l d s , but l a t e r on he doesn't s t a r t ploughing u n t i l l a t e r , so he would j u s t have to be i n them for now (D). "He said 37. i_t with an e x a l t a t i o n , for 38. the s t i l l n e s s was a l l gone 39. away (now) from h i s mind, and 40. a_ (the) good kind of tears 41 f e l t (were) t i g h t i n h i s throat." Response 37. He said-—I don't have room f or "the word," so i t would have to be " i t " again (D). Response 38. With an e x a l t a t i o n for "the" s t i l l n e s s (Sy), Response 39. was a l l gone "away" from h i s mind. Gone away sounds a b i t strange but you think i t would be j u s t gone, but since there i s a blank between "gone" and "from" I thought gone away was j u s t another longer way of saying i t (Sm). Response 40. And "a" good kind of t e a r s — a n o t h e r adj e c f i v e wouldn't have been needed before good kind, so ju s t leave "a good kind of t e a r s " (Sy). Response 41. " f e l t " t i g h t i n h i s throat—how else are you going to put i t ? They didn't r e a l l y choke tightness. I suppose I could have put that, but that's u n l i k e l y . I r e a l l y had to guess i f that was the word, so i t was j u s t a general kind of word (Sm). "42. T_t had been David who 43. loved the place best of 44. a l l . " Response 42. " I t " had been i s sort of l i k e there was, a construction (Sy). Response 43. David who "loved" the p l a c e — a n d that was a guess but, since he t a l k s about love l a t e r on, I figured i t i t would love up there, (D). Response 44. Best of " a l l " i s j u s t , i s j u s t a phrase that goes i n there (Sy). "And i f your son 45. loved the place he went 46. away from, then he could 47. never leave i t . " Response 45. I s t i l l used the "loved" again (D). Response 46. "He went away from" i s the same as away from up higher or before (D). Response 47. Then he could "never" leave i t — h e had talked e a r l i e r i n the story part about how i f your son goes away angry, he can s t i l l be your son, but i f he goes away with no sign, then maybe he's gone f o r -ever. Whereas, since he does now, Joseph knows that David loved the place, so therefore he can never leave i t forever (D). 166 " I f he 48. leaves (dies) even for these things 49. you both loved, then you 50. can (could) hear t h e i r voice s t i l l . " Response 48. "Leaves" was from the sentence e a r l i e r , was the only verb that made any sense (D). Response 49. Even for these things "you" both l o v e d — w e l l , i t ' s p l u r a l obviously, and he's t a l k i n g about Joseph and h i s son (D). Response 50. When you "can" hear h i s voice i n t h e i r voice s t i l l . There's no room for then you w i l l be able so w i l l have to be you can hear h i s voice (Sm). * 167 Expository Prose: Grade 12 Subject #212: Good Reader "The immensity of the oceans—covering two-thirds 2.'of the earth s u r f a c e — a n d 3. large (the) f i s h abundance i n some 4. parts , " Response 1. "The" seemed to be the only thing that would sort of go (Sy), Response 2. and the same with "of" (Sy). Response 3. And "large " was j u s t a guess, cause I couldn't think of anything e l s e — l i k e , i t didn't make much sense why there was a blank there (Sm). Response 4. "Parts" was the only word I could think for sort of areas of the ocean, type thing (Sm). " . . . undoubtedly explain many of 5. the exaggerations and ov e r s i m p l i f i c a t i o n s presented 6. about the p o t e n t i a l i t i e s of the 7. ocean (seas) and t h e i r f i s h r i c h e s . " Response 5. That was again the only word to f i t i n (Sy) Response 6. and the same with "about" (Sy) Response 7. and the "ocean" because that's what they're t a l k i n g about (D). "8. There i s hardly an a r t i c l e 9. or_ a book on the 10. sea (subject) which does not resort 11. _to_ terms l i k e "inexhaustible" or 12. " l i m i t l e s s " i n describing the abun-dance 13. of the oceans." Response 8,9. "There" and "or" are ju s t sort o f — I j u s t put them i n there because they were the only prepositions that would f i t (GF). Response 10. The "sea" i s again the subject that they're t a l k i n g about (D). Response 11. And "to" I j u s t put i n (NR). Response 12. And " l i m i t l e s s " I read on further. If you read on i t says i t somewhere, i t repeats i t s e l f , so I put that i n (D). Response 13. And the abundance "of" the oceans i s the only word that would f i t (Sy). 168 " I t i s 14. necessary (important), however, to keep i n 15. mind that these words do 16. not exi s t i n Nature's own 17. world ( d i c t i o n a r y ) . " Response 14. " N e c e s s a r y " — i t j u s t sounded a l l r i g h t (Sy). Response 15. Keep i n "mind" I put. I had trouble with that one. I sort of got the idea what they meant (Sm), Response 16. but I didn't r e a l l y know what to put i n (NR). Response 17. And nature's own "wor l d " — I j u s t guessed at that (NR). "There i s i n the 18. natural ( l i v i n g ) world no such thing 19. _s l i m i t l e s s and inexhaustible." Response 18. Well, they're t a l k i n g about nature's world so I j u s t put again the "na t u r a l " world (D), Response 19. and "as" ju s t sort of went i n there (Sy). " A l l 20. l i f e i n the oceans i s 21. d i r e c t l y (ultimately) t i e d to the primary 22. organisms (production), i n the top layer 23. of seaweeds and phytoplankton with . . . " Response 20. " L i f e " I put because they're sort of t a l k i n g about the fish e s and the plants (D). Response 21. And " d i r e c t l y " — I don't know w h y — i t ' s hard to explain why I put i t (NR). Response 22. And "organisms" I read on f a r t h e r — I had no idea what to put for that one. I t ' s sort of technical and i t ' s t a l k i n g about organisms, so I j u s t put i t i n (D). Response 23. "of" seaweeds because I j u s t assumed they were giving examples of seaweed and phytoplankton (SM). "24. substances ( c e l l s ) containing plant pigments capable 25. of_ trapping the sunlight energy 26. a v a i l a b l e (penetrating) into the upper regions 27. _ f the water masses (generally 28. c a l l e d (termed) the photic layer, meaning 29. where l i g h t r e i g n s ) . Response 24. And "substances" was ju s t a guess (NR). This b i t here I had r e a l problems w i t h — i t was hard to figu r e out. I f you don't have the words, I didn't r e a l l y know what they were t a l k i n g a b o u t — i t i s so sort of technological. I don't r e a l l y know what what goes on i n the ocean. And the a r t i c l e didn't r e a l l y say anything a b o u t — i t said how they found out about i t , but they didn't say anything about what was keeping l i f e going; so you have to sort of guess at words that sound l i k e they'd go i n there. Response 25. N.R. 169 Response 26. I put " a v a i l a b l e " because I couldn't think of anything e l s e . I sort of thought of needed, but I wasn't r e a l l y sure, so I stuck a v a i l -able (Sm) . Response 27. And "of" j u s t sort of went i n there, because they were lacking a word l i k e that (Sy). Response 28. Generally " c a l l e d " was the only thing I could think and they've got the name there (Sm). Response 29. And "where" l i g h t r e i g n s — I didn't r e a l l y know i t meant, but I guess they're t a l k i n g about l i g h t and i t being at the top, I j u s t figured i t had something to do with that (Sm). "Technical devices, 30. however ingenious they may be, 31. do l i t t l e to free the 32. f i s h (waters) and t h e i r l i v i n g organisms 33. of_ (from) t h i s dependence." Response 30. "However" I stuck i n because they're obviously sort of saying that, although they are t h i s , they aren't good (Sm), Response 31. and so "do" l i t t l e (Sm). Response 32. And the " f i s h " because they've got organisms and the only other l i v i n g thing they were t a l k i n g about was f i s h (D). Response 33. And "from" t h i s dependence—it's obvious the only thing again to put i n (Sy). "The base 34. of_ (for) t h e i r existence i s ultimately 35. energy (these) pigment-carrying plant organisms, 36. extending (spread) to the l i m i t of 37. l i g h t penetration, down to a 38. depth of one hundred feet , 39. jor (occasionally) more." Response 34. The base "of" I put i n because again i t was the only thing that would go (Sy). Response 35. And "energy" I put i n because I didn't r e a l l y know what they were t a l k i n g about, t h i s pigment thing was the only thing I could think of and I was running out of time (Sm). Response 36. "Extending" I put i n because they're t a l k i n g about the l i m i t s so i t would have to be something l i k e that (Sm). Response 37. And I put i n " l i g h t " because that's what they're t a l k i n g about (D). Response 38. And to a "depth" of one hundred f e e t — t h e y ' r e obviously t a l k i n g about below, so i t would be depth (Sm), Response 39. "or" more (Sm). 170 "Here the creative 40. process (processes) take(s) place which produce 41. the organic material (read matter) that sustains 42. l i f e . " Response 40. I put "process" because they're j u s t sort of t a l k i n g about the way something i s done (D). Response 41. And "the"—you need a preposition i n there (GE). Response 42. " L i f e " I put i n because up here they're saying how l i f e depended on the sun (D). "Even the bottom animals, 43. which may l i v e thousands of 44. feet below the surface, feed 45. on (from) the "manna" r a i n i n g down 46. from above." Response 43. "Which" I put i n because I read a comma and that's the only thing that went i n that kind of sentence (Sy). Response 44. "Feet" I put i n because they're obviously t a l k i n g about some sort of measurement—I didn't r e a l l y know which ones so I took a guess and they also had " f e e t " up there (D). Response 45. Feed "on" was sort of the only thing that would come a f t e r that (Sy). Response 46. Raining down "from" above—because i t wouldn't be anything else (Sy). "This consists of 47. plant (plankton) remainders, animal excre-t i o n s , dead 48. organisms (bodies), and other waste. Sometimes 49. some (part) of t h i s organic matter 50. '_is_ caught by other scavengers on i t s way down and may be cycled." Response 47. These ones I sort of guessed at because i t doesn't r e a l l y t e l l you what the remainders are so I j u s t figured i t would be plants (NR) Response 48. and they got animals here and dead so I j u s t put "organisms" — i t seemed to be the only thing l e f t (Sm). Response 49. I put "some" i n because they needed a quantity and I didn't know how much (Sm). Response 50. " I s " I put i n because i t didn't have a verb there (GF). 171 APPENDIX C 172 CRAM Reading Assessment Method Narrative F i c t i o n : Grade 12: Subject #211 Cloze Replacements Grammar Syntax Semantics Discourse exact non-exact Y P N Y P N Y T Px P 2 N Y P N 1 2 3 4 5 . 6 7 8 9 10 11 the there Father for Father and the trees t a l k I t I 12 woods 13 been 14 guess 15 day guess s a t i s f i e d 18 one 19 remember 20 time 21 think 22 He I i t 16 17 23 24 25 Dad 26 was 27 28 30 31 36 37 38 suddenly I 29 had horses to 32 didn't 33 Joseph 34 over 35 mind i n i t the 39 now 40 the I near our language That thought night thought allowed lay knew I he f e l t away a / / / / / / 173 CRAM Reading Assessment Method (cont.) Cloze Replacements Grammar Syntax Semantics Discourse exact non-exact Y P N Y P N Y T Pi P 2 N Y P N 41 were f e l t / / / 42 It = 43 loved = 44 a l l = 45 loved -46 away = 47 never = / / / / 48 died leaves 49 you = / / / / 50 could can Totals E.R. 31 18 0 1 16 0 3 10 6 0 0 3 10 4 5 Weighting 2 1 0 2 1 0 4 3 2 1 0 2 1 0 Weighted Value 36 32 Weighted Score 3 1.89 1.68 58 3.05 24 1.26 Weighted Score Weighted Value 50 - Exact Replacements 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0078359/manifest

Comment

Related Items