Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of text organization and headings on grade 5 through 10 students’ written recall of expository… Stables, Roderick Gwyn 1985

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

Item Metadata

Download

Media
831-UBC_1985_A8 S73.pdf [ 5.02MB ]
Metadata
JSON: 831-1.0078303.json
JSON-LD: 831-1.0078303-ld.json
RDF/XML (Pretty): 831-1.0078303-rdf.xml
RDF/JSON: 831-1.0078303-rdf.json
Turtle: 831-1.0078303-turtle.txt
N-Triples: 831-1.0078303-rdf-ntriples.txt
Original Record: 831-1.0078303-source.json
Full Text
831-1.0078303-fulltext.txt
Citation
831-1.0078303.ris

Full Text

THE EFFECTS OF TEXT ORGANIZATION AND HEADINGS ON GRADE 5 THROUGH 10 STUDENTS' WRITTEN RECALL OF EXPOSITORY PROSE WITH EMPHASIS ON GRADES 5 AND 6 By RODERICK GWYN STABLES B.Ed., The University of British Columbia, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES, DEPARTMENT OF LANGUAGE EDUCATION We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA AUGUST, 1985 © Roderick Gwyn Stables, 1985 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British 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 or her representatives. It i s understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia 1956 Main Mall Vancouver, Canada \ V6T 1Y3 \ t ABSTRACT This study investigated the effects of headings and text organization on grade 5 through 10 students' written r e c a l l of expository prose passages written in a classification/description mode. Emphasis was placed on the results from students in grades 5 and 6. This study was a component of a three part study. The other two parallel studies emphasized grades 7 and 8 (King, 1985) and 9 and 10 (Gibbs, 1985). Each subject read and recalled two passages: one written at his or her grade level and one written at a low readability l e v e l . Performance on the written recalls from passages with headings and without headings was examined on the basis of the number of superordinate and subordinate ideas recalled, the superordinate and subordinate organization, and the format. Developmental trends were investigated by including the data from the two parallel studies (Gibbs, 1985; King, 1985). There was some evidence that headings had a significant positive effect on the number of superordinate ideas recalled from a passage of low readability. Some significant differences indicated negative effects by headings. The majority of differences, however, were not significant. Developmental trends in grades 5 through 10 were noted in the number of ideas recalled on a low readability passage and the format used on the written r e c a l l s . Implications for instruction and suggestions for further research are discussed. i i TABLE OF CONTENTS Chapter Page Abstract i f Table of Contents i f f Lis t of Tables v i i Lis t of Figures x Acknowledgement x i 1 STATEMENT OF THE PROBLEM 1 Introduction 1 The Problem 2 The Need for the Study 3 Definitions 6 Delimitations 7 General Procedures 8 The Construction and Selection of the Instruments 8 The Selection of the Subjects 9 Administration of the Instruments 10 The Collection of the Data 11 The Treatment of the Data 11 Summary 12 i i i Table of Contents, continued Chapter Page 2 REVIEW OF RESEARCH 13 Prose Organization and Study Methods: Pre 1975 14 The Influence of Prose Organization on Recall of Superordinate and Subordinate Ideas. 15 Developmental Aspects in Ability to Use the Author's Organizational Structure 19 Age Effects 19 Reading Level Effects 20 The Effectiveness of Headings in Aiding the Reader 25 Summary 29 3 METHODOLOGY 31 Subjects 31 Materials 34 Administration of Instruments. . 36 Scoring and Tabulation of Data 39 Design 40 Statistical Procedures 40 Summary 40 4 ANALYSIS OF DATA 41 Effect of Headings on Recall of Superordinate and Subordinate Ideas 42 iv Table of Contents, continued Chapter page Effect of Headings on the Organizational Structure of the Students' Written Recall . . . 51 The Relationship Between Organization and the Number of Superordinate and Subordinate Ideas on , the Recall i . . .60 The Effect of Headings on the Format of the Written Recall 62 Developmental Factors 71 Post Hoc Analysis of Developmental Factors Exhibited by Good and Poor Readers in Grades 5 through 10 76 Summary 80 5 SUMMARY AND CONCLUSIONS 81 Summary of the Study 81 Effect of Headings on Recall of Superordinate and Subordinate Ideas 85 Summary of the Effects of Headings on Recall of Superordinate and. Subordinate Ideas 87 Effect of Headings on the Organizational Structure of the Students' Written Recall 88 Summary of the Effect of Headings on the Organizational Structure of the Students' Written Recall 90 v Table of Contents, continued Chapter Page The Relationship Between Organization and the Number of Superordinate and Subordinate Ideas on the Recall 90 The Effect of Headings on the Format of the Written Recall 92 Developmental Factors 94 Summary of Developmental Factors 98 Summary of Research Findings .99 Post Hoc Analysis of Interrelationships of Research Variables 101 General Conclusions 103 REFERENCES 105 Appendix A: Sample Passages 112 Appendix B: Directions for Scoring and Sample Protocol Sheets . 115 Appendix C: Student Information Record Sheet 120 Appendix D: Sample of Recall From Parrot Passage With Explanation of Marking Procedures 122 vi LIST OF TABLES Table Page 1 Percentile Scores on Gates-MacGinitie Comprehension Subtest by District 32 2 Reading Scores on Gates-MacGinitie Comprehension Subtest by Grade 33 3 Readability of Passages 35 4 Effect of Headings on Recall of Superordinate and Subordinate Ideao by Grade 5 Students 43 5 Effect of Headings on Recall of Superordinate and Subordinate Ideas, by Grade 6 Students 44 6 Effect of Headings on Recall of Superordinate and Subordinate Ideas by Grade 5 Students by Reading Level 46 7 Effect of Headings on Recall of Superordinate and Subordinate Ideas by Grade 6 Students by Reading Level 48 8 Effect of Headings on Recall of Superordinate and Subordinate Ideas by Grade 5 through 10 Students 49 9 Effect of Headings on Organization for Grade 5 Students 52 10 Effect of Headings on Organization for Grade 6 Students 54 11 Effect of Headings on Organization for Grade 5 Students by Reading Level .55 12 Effect of Headings on Organization for Grade 6 Students by Reading Level 56 v i i List of Tables, continued Page 13 Effect of Headings on Organization for Grade 5 Through 10 Students 59 14 Correlation Between Organization and Amount of Recall for the Parrot Passage 61 15 Correlation Between Organization and Amount of Recall for the Grade Level Passage 61 16 Effect of Headings on the Number of Grade 5 Students Who Used Each Format Characteristic 63 17 Effect of Headings on the Number of Grade 6 Students Who Used Each Format Characteristic 65 18 Effect of Headings on the Number of Students in Grade 5 Through 10 Who Used Each Format Characteristic 66 19 Recall of Superordinate and Subordinate Ideas by Grade. . 73 20 Recall Organization by Grade 73 21 The Number of Students Who Used Each Format Characteristic in Grades 5 Through 10 75 22 Recall of Superordinate and Subordinate Ideas for Good Readers by Grade 78 23 Recall Organization for Good Readers by Grade 78 24 Recall of Superordinate and Subordinate Ideas for Poor Readers by Grade • . .79 v i i i L i s t of Tables, continued Page 25 Recall Organization for Poor Readers by Grade 79 ix LIST OF FIGURES Figure Page 1 Interaction of Grade and Heading for the Use of a Title on the Written Recall From the Grade Level Passage 68 x ACKNOWLEDGEMENT Special thanks to my family, friends, the advisory committee, and those who became the "Thesis Club" for a l l the support! R. G. S. xi CHAPTER 1 STATEMENT OF THE PROBLEM Introduction In the current educational system much of children's learning must be done by reading commercially written textbooks. The majority of the material to be learned i s unfamiliar to the student and i s written i n an expository format. In order to f a c i l i t a t e student learning, textbooks must be structured so that the student can be taught to process e f f i c i e n t l y the information presented. Current studies suggest that the structure of the discourse affects the students' comprehension. Also, the extent of the students' awareness of the author's structure may determine t h e i r a b i l i t y to r e c a l l ideas presented. Developmental patterns may exi s t which l i m i t the students' a b i l i t i e s to use paragraph headings that indicate the author's structure. Students' r e c a l l of material may be optimized i f textbooks are structured i n a manner which r e f l e c t s the most e f f i c i e n t patterns for r e c a l l . 1 The Problem The purpose of the study is to investigate the ability of students in grades 5 and 6: 1) to use paragraph headings to facilitate recall; 2) to recall information material written in a descriptive form; 3) to use the author's organizational structure to facilitate recall. This study is one component of a three part study. The other two parallel studies emphasized grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). The inclusion of data from the other grade levels enables the above points to be investigated in a developmental context. The study seeks answers to the following questions: 1.0 In a written recall task, are more superordinate or subordinate ideas recalled when the prose to be recalled has (a) headings or (b) no headings: 1.1 within each grade level? 1.2 within each grade level by reading level? 1.3 over grade levels 5, 6, 7, 8, 9 and 10? 2.0 In a written recall task, do the students' written recalls follow the author's organizational structure more closely when the prose to be recalled has (a) headings (b) no headings: 2.1 within grade levels? 2.2 within grade levels by reading level? 2.3 over grade levels 5, 6, 7, 8, 9 and 10? 3.0 Do students recall more superordinate and subordinate ideas 2 when they more c l o s e l y follow the author's organization i n t h e i r written r e c a l l s over grades 5t 6 , 7 , 8 , 9 and 10? 4 . 0 Are there d i f f e r e n c e s i n the format of the written r e c a l l when the prose to be r e c a l l e d has (a) headings, (b) no headings, and are there d i f f e r e n c e s over reading l e v e l ? 5 . 0 In a written r e c a l l task, i s there evidence of developmental e f f e c t s i n (a) number of ideas r e c a l l e d , (b) organization of the written r e c a l l , and (c) format, when the r e s u l t s of t h i s study are compared to s i m i l a r studies being done at the grade 7 , 8 , 9 and 10 l e v e l s (Gibbs, 1 9 8 5 ; King, 1985 )? The Need For The Study There are many f a c t o r s involved i n the reading comprehension of expository discourse. These f a c t o r s include the content and the structure of the passage read. Aspects of s t r u c t u r a l organization w i l l receive emphasis i n t h i s study. There appear to be two aspects of s t r u c t u r a l o r g a n i z a t i o n : 1. Text Organization, which i s defined, for purposes of t h i s study, as s y n t a c t i c a l and i d e a t i o n a l arrangements of information u n i t s through which the author conveys meaning. Text can be organized i n several d i f f e r e n t ways. 2. Typographic-Semantic Cues, which have, as t h e i r purpose, the conveyance or emphasis of meaning through a s t r u c t u r a l feature. Headings are i n t h i s category of s t r u c t u r a l o r g a n i z a t i o n . Before 1975 the l i t e r a t u r e on text organization and i t s influence 3 on reading comprehension was fairly extensive but only minimally based on empirical research. Robinson's writing about his SQ3R study method (1961, 1970), Smith's discussion of text writing patterns (1964) and ways to use them in reading comprehension (1968), and Niles' statement about the importance to recall of perceiving an author's text organization (1965), are typical items. The statements are a l l insightful and may be valid. They were not, however, the product of controlled research. Since 1975, Meyer's research (1975, 1977, 1984) has had an important impact on reading literature. The influence of text structure on the recall by university students of the superordinate and subordinate ideas in a passage was found to be significant, and that study led the way to many more. We are now much better informed in 1985 about the influence of text structure on reading comprehension than we were ten years ago. However, as always, the studies already done provide more questions to be answered. The importance of headings as cues for prior knowledge, as cues accentuating relationships among concepts and facts in the text, and as retrieval cues for recall has been examined (Alverman, 1982; Dansereau, Evans, Collins, Brooks and Larson, 1981). However, the limitation of existing studies involving headings is that the subjects have been restricted to college level students. There is a need for studies that involve students in upper elementary and junior secondary since this is the first period of emphasis on reading for information. There is also a need for studies which cut across these grades so that developmental patterns, i f present, would be evident. Variations in materials, differences in scoring units, and differences in 4 treatments across experimenters make i t difficult to compare studies. Therefore data over different experiments can be examined only for such general trends as increases in proportion recalled over age or reading ability (Drum, 1984). Drum's (1984) summary of recall data in eight similar studies showed differences in grades used, from grade 2 to adult; in number of subjects used from 20 to 128 per grade level; in number of passages given, from 2 to 12; in scoring units, and in methods utilitzed and in time between readings and recall, from immediate to fourteen days. Thus i t would be impossible to establish a developmental pattern from the existing data. There is much practical significance to the study of the problem presented. Evidence of a developmental sequence of student recognition and use of headings as cues to the author's organizational structure in text, would provide guidelines for constructing textbooks that would result in optimum information retention by students at each grade level. As more empirical knowledge in the area of reading comprehension is gained, reading educators come closer to understanding the way children process information material. As teachers and textbook publishers are become aware of the important findings emerging from this area of research, materials and instruction in the area of reading comprehension should improve. 5 Definitions Specific terms used in this study are defined as follows: 1. Heading: For the purposes of this study, a heading is a phrase set apart above the paragraph which cues the superordinate idea in the paragraph. 2. Superordinate idea: The main idea of a paragraph is the superordinate idea. It is the central thought, meaning, or gist of a passage; a statement which gives the stated or implied major topic of a passage and the specific way in which the passage is limited in content or reference (Harris & Hodges, 1981, P« 188). 3. Subordinate idea: A detail which is found under a superordinate idea in the hierarchical structure of a paragraph is a subordinate idea. 4. Written recall: Written recall is the task of writing down everything that is remembered from a previously read passage with no reference back to the passage and no aids to prompt memory. 5. Organization: The order of ideas and the extent to which the student's organizational pattern in the recall reflects the author's organizational pattern. 6. Format: In this study, the format is the general plan or 6 arrangement of the written recall. The presence of the following format characteristics will be studied: t i t l e , section headings, one grouping, point form, numbering, and paragraphs. A precise definition for each of these items is contained in the Directions for Scoring in Appendix B, Delimitations The size of the sample, 50 students from each grade level, limits the population over which the results can be generalized. The grade 5 and 6 results in this study are also limited to students in regular, non-streamed, grade 5 and 6 classrooms in the school districts of Richmond, Maple Ridge, and Surrey. The developmental trends discussed are limited to students in regular, non-streamed grade 5 through 10 classrooms in the above mentioned districts. The study measured recall performance on expository prose written in a description form and may not be generalizable to narrative materials or other forms of expository material organization. 7 General Procedures The general procedures to be followed in the investigation are: (a) the construction and selection of the instruments; (b) the selection of the subjects; (c) the administration of the instruments; (d) the collection of the data; and (e) the treatment of the data. Each step is discussed briefly. The Construction and Selection of the Instruments The Gates-MacGinitie Reading Test, Canadian Edition (1980), Comprehension subtest, was chosen because i t is appropriate for the purposes of this study. It contains Canadian content and was ncrmed over a Canadian population. The D and E forms of this test are appropriate for grades 5 and 6. The comprehension items have reliability scores which range from .85 to .92 (Teacher's manual, p. v). Appropriate forms of this test have also been selected for use in parallel studies at higher grades (Gibbs, 1985; King, 1985). This will enable comparison of the results of this study with results of the parallel studies. A battery of expository prose passages was written by one author, Megan Crowhurst (1984), in a classification/description form at grade levels ranging from approximately grade 4 to approximately grade 10. This study uses the grades 4, 5, and 6 passages. Appropriate passages from this battery were also selected for use in the parallel studies (Gibbs, 1985; King, 1985. This enables comparison of the results of 8 this study with results of the parallel studies. The classification/description form was used because i t is the style in which the majority of textbook material is written (Niles, 1965). The selections are typical of textbook passages on which students may receive reading assignments at their grade level. All passages have been carefully constructed to each contain 5 superordinate and 26 subordinate ideas. Each student was given a passage written at his approximate grade level and the grade 4 level passage. Grade levels of the passages were determined by the Fry readability formula (Fry, 1968). Versions of the passages were prepared with and without headings (see Appendix A). A pilot study was conducted in September, 1984. Adjustments were made to materials at that time. The Selection of the Subjects For this specific study, three classes of grade 5 students and three classes of grade 6 students were chosen to participate. They were selected from these schools: Kingswood Elementary, Richmond, B.C.; Whonnock Elementary, Maple Ridge, B.C.; and Georges Vanier Elementary, Surrey, B.C. The communities chosen are a l l suburbs of Vancouver and are comparable in their average socioeconomic status (Canada Census, 1981). The complete study included parallel studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). Three classes from each of these grades also participated. 9 Intact, regular, non-streamed, grade 5,6 and 7 classes were chosen for the study on the basis of availability. Students from grades 8,9 and 10 were chosen from intact, regular, non-streamed, English or Social Studies classes. Administration of the Instruments Three sessions were spent with each of the classes. In the first session the Comprehension subtest of the Gates-MacGinitie Reading Test, Canadian Edition (1980) was administered. According to the Teacher's Manual, the test takes 45 to 55 minutes of class time. The second session involved the administration of a recall passage. At each grade level, one-half of the students were randomly chosen to receive the passage written at the grade 4 level first and the other students received the passage written at their own grade level f i r s t . At each grade level, one-half of the students were randomly chosen to receive passages with headings. The other half received passages without headings. The students were given 10 minutes to study the passage after receiving the following instructions: "You will have 10 minutes to study this article. You must try to remember as much information as possible. In 10 minutes I will collect the articles and give you each a lined piece of paper. You are then to write down as much of the information as you can remember. You will have a maximum of 25 minutes in which to write down the information." After 10 minutes the articles were collected and foolscap sheets were distributed. Students were given 25 minutes to complete the recall task. All papers were collected at that time. Information 10 such as name, grade, and school was included on the papers. Early-finishers were expected to read silently for the remaining time. The third session was similar to the second except the student groups were reversed so that the grade passages not written the first time were written. All other procedures were the same. The Collection of the Data All sessions were conducted within a three week period during the months of November and December, 1984. Written recall was used as the method of obtaining a measure of comprehension of the passage because it provides for a purer measure than do questions (Clark, 1982). It is the common form of measure used in research of this type. A protocol form was developed for the marking of each of the passages (see Appendix B). The Treatment of the Data Analysis of variance was used to analyze the relationships among the factors measured in the study. This provided a basis for answering the research questions. 11 Summary This chapter has introduced the study, given the rationale, defined the terms, noted the delimitations, and outlined the general procedures. The remainder of the report i s organized as follows: Chapter 2: Review of research Chapter 3: Methodology Chapter 4: Analysis of data Chapter 5 : Summary and conclusions 12 CHAPTER 2 REVIEW OF RESEARCH The parameters of reading comprehension may never be fully delineated. However, this review is based on the premise that at least some part of reading comprehension requires the ability to follow and to use the organizational structure of a prose passage either to comprehend i t i n i t i a l l y or to recall its content later. The review is organized in 4 sections: (a) Prose organization and study methods: pre 1975; (b) The influence of prose organization on recall of superordinate and subordinate ideas; (c) Developmental factors in ability to use an author's organizational plan; and (d) The effectiveness of headings in aiding the reader. As has already been noted (p. 3) the literature of empirical research reviewed includes a l l studies that deal with text organization whether the researchers use the term "text organization" or "schema for structure" to mean the organizational pattern of a text passage. 13 Prose Organization and Study Methods: Pre 1975 Before 1975 the literature on prose organization as an influence on study methods was reasonably extensive but not empirically based. Writing by Robinson (1941, 1961, 1970), Smith (1964, 1968) and Niles (1965) provided some of the best samples of the kinds of statements and proposals made before 1975. The importance of the reader's approaching a passage with a mental framework to retain superordinate and subordinate ideas had a major educational impact with Robinson and Hall (1941). This idea formed the theoretical basis of the SQ3R study method (Robinson, 1961). SQ3R was designed to give the reader an orientation to what the chapter presented, to help recall what wa? already known about the material, and to facilitate subsequent readings. It allowed the reader to ensure that there was an understanding of the organizational pattern that was in place, and attempted to ensure that subordinate ideas were slotted under the appropriate superordinate ideas in the reader's memory as the material was learned. Robinson (1961) believed that his procedures would be useful to superior high school students. The SQ3R method has been recommended by many methodology text authors since its introduction. Smith's work is also worth noting. She suggested (1964) that school texts in Science, Social Studies, Mathematics and English followed certain predictable writing patterns and proposed that children be taught to use these patterns in reading to comprehend. She later published a set of workbooks (1968) that provided practice in reading the various patterns. 14 Niles (1965), in the article Organization Perceived, drew parallels between the system and order of expository prose and its probable organization in the mind of the reader for efficient recall. The respect with which her statement has been viewed is evident in the frequency with which i t is cited in the subsequent literature about study. The Robinson, Smith and Niles thinking, cited above, provide evidence that the influence of text structure had been seriously discussed by educators before 1975. It is clear, however, that their statements are those of creative methodologists rather than those of empirical researchers. The Influence of Prose Organization on Recall  of Superordinate and Subordinate Ideas Initial studies were concerned with the manipulation of content organization and its effect on comprehension and recall. Meyer (1975), investigated the effects of the hierarchical level of information in the content structure on information recall using two passages with the same content but different structure. The dependent variable in her study was the number of content units and relationship units recalled from three target paragraphs by each subject. These paragraphs were administered to 105 students enrolled in an introductory psychology course. The factors manipulated in the different treatments were: signaling phrases or words, order, group, and time. Subjects and passages were treated as random factors. After each passage was read a written free recall test for retention of the passage was administered. One week later a second written free 15 recall test was administered as well as a cued recall test which involved using cue words to generate information from the original passage. Meyer found that when the hierarchical structure of the text was manipulated, the amount and type of information recalled was affected. Ideas which were structurally superordinate were recalled better than ideas which were structurally subordinate. Meyer (1975) raised the question: How did different types of prose organization • affect recall of superordinate and subordinate ideas and was i t possible to manipulate the prose for maximum results? Danner (1976) investigated prose organization and its effects on memory with second, fourth and sixth grade students using descriptive passages. One passage described a fox and the other passage described a polar bear. Each passage had three main topics with four sentences for each topic. Each passage had two versions. In one version (organized), the sentences in each of the three topics appeared in order. In the second version (disorganized), each sentence in one topic was followed by a sentence from a different topic. The passages were administered to 12 girls and 12 boys from each of the grades. Danner found both recall and ability to cluster sentences by topic in a sorting exercise was greater for those students who received the topically organized passages. The effect of content organization on the recall of superordinate ideas was further supported by Waters (1978) in an experiment which showed that targeted information placed high in the hierarchical structure of the passages was recalled better than information placed low in the hierarchical structure. Waters had 64 third grade, 64 sixth grade, and 128 college students either generate passages to be 16 recalled two weeks later, or listen to passages generated by other subjects and recall them two weeks later. Each subject was given a word outline "prompt" to study while composing or listening to the passage. Each subject was given two minutes to look at the prompt sheet and organize his or her thoughts. Subjects then either read their generated passages into the tape recorder, or listened to the passage, depending on their experimental group. Two weeks later the subjects were asked to recall their passages. Students recalled more superordinate ideas than subordinate ideas in both the "generated" and "listen" treatments. Manipulation of the content organization of prose did not have any direct educational application to the improvement of reading comprehension. It showed, however, that manipulation of organizational aspects of prose was worth pursuing in research. Subsequent studies by Elliott (1980) and Meyer, Brandt and Bluth, (1980) focused on the use of the author's structural organization in the students' recall strategies. Meyer et al (1980) investigated the use of reading strategies by ninth grade students. One hundred and two ninth grade students were divided into good, average, and poor groups on the basis of reading tests. Two passages were used for each of the subjects. One passage was a comparison (adversative) structure and one was a problem/solution structure. Each passage had two versions; a with-signaling version and a without signaling version. The with-signaling version explicitly stated the top level structure, or the superordinate ideas, and had key words underlined, while the without signaling version did not. Students were grouped by 17 comprehension ability and sex. Then a random procedure for each group was used to determine the assignment of the with or without signaling versions of the passages and to determine the order of presentation of the two passages. After reading each passage and then placing i t out of sight, the students wrote down a l l they could remember. One week later they were again asked to write what they could recall from each passage, and then they took a recognition test on sentences from the passages. The study found that the signaling of top level structure did improve the good and average students' immediate recall but few poor readers employed the structure strategy. Elliott (1980) investigated what 102 sixth grade students knew about prose organization and whether the use of specific organizational patterns would facilitate their written recalls. Students were given one of two versions of a well organized passage of expository text. One was written in the adversative top-level structure and one was written in the attributive form. No student had a reading level lower than fifth grade. Students wrote down everything they could remember two days after reading the expository passage assigned to them. Meyer's method of scoring the recall protocols was used. The findings indicated that students' use of the top level structure to organize their written recalls resulted in the remembering of more idea units. The differences in amount recalled between the two types of top-level structure was not significantly different. These studies demonstrated that i t would be useful to examine the top level organization of the text. 18 Developmental Aspects in Ability to Use the Author's  Organizational Structure The developmental effects of age and reading level on ability to use the author's structure have not been studied extensively. Age Effects Research indicates that there may be a clear developmental pattern in the ability to follow and make use of the author's structural organization. These findings reflect the work of Piaget (1955), who suggested that there is a series of developmental stages in the acquisition of a child's language and complex thought processes, or schema. One of the few studies done with young children was that of Danner (1976), who tested second, fourth, and sixth grade students using descriptive passages. Recall of ideas increased with age and degree of paragraph organization. Danner found that older students made more use of the author's organizational patterns. Brown, Campione, and Day (1981) used a schema theory (structure) approach and suggested that strategies for dealing with author's organizational patterns begin to be used between five and seven years of age and continue to be refined throughout the school years. They state, "Also common to the developmental course of these strategies is an intermediate stage called a production deficiency, where the child does not produce the strategy spontaneously but can be prompted or instructed to do so quite readily." (p. 14). Brown et al also suggested that there were two types of problems that could impede 19 comprehension: inefficient application of rules and organizational strategies and a lack of background knowledge. Both of these problems are closely tied to the child's development. Their statements underscore the need to determine the student's developmental stage before instruction is given. Bos (1981) studied junior high school educably mentally retarded students reading at a grade three level and compared their recall abilities to average grade three students reading at their grade level. He found the older students had the greater ability to deal with the text structure. It can be hypothesized that the older student generally has had more background experiences and exposure to different styles of text organization and thus outperforms the younger student on the written recall of passages. Reading Level Effects This section deals with studies comparing the performance of good and poor readers' passage recall ability. The evidence presented here suggests that good and poor readers have different developmental patterns in ability to use the author's structural organization. In comparing the various studies i t should be noted that i t is possible to compare only overall trends rather than specific findings. Close comparisons are difficult because of problems in determining whether the variations in recall at particular grade levels are due to treatment differences, scoring differences, or both. Variation in the instructions or instruction given to the students and in the various 20 experimental designs also present problems. Some studies required written r e c a l l while others required oral r e c a l l ; and the passages given to the students were presented in a variety of written and oral formats. There are, however, three trends clearly identifiable from the research. They are: 1. good and poor readers seemed to use different r e c a l l strategies; 2. readers who used the author's schema format in organizing their recalls were able to remember more information in both immediate and delayed recalls than those who did not follow the author's organizational structure; 3. cues embedded in the prose structure helped less able readers follow the author's schema and increased their r e c a l l a b i l i t i e s . The f i r s t general trend the studies showed was that one difference between good and poor readers lay in their differing r e c a l l strategies. Doctorow, Wittrock, and Marks (1978), in a study of grade 6 children which compared the results of good and poor readers f i r s t reading passages without paragraph headings and then reading passages with headings, found paragraph headings improved the performance of both groups of students. However, the increase in performance was much greater for the good readers than the poor readers. This finding was supported by a study of grade three students which indicated that the poor readers differed from the good readers both in the extent to which they recalled superordinate and subordinate ideas and in their a b i l i t y to generate the author's organizational patterns in their recalls (Tierney, Bridge, and Cera, 21 1978-79). Earaon (1978-79) studied the differences between good and poor readers. He used the Iowa Silent Reading Test to pick the 14 highest and 14 lowest a b i l i t y students from 68 students in an introductory psychology class. He assigned the students the task of reading paragraphs and ranking the level of importance of statements about the topic or theme of the paragraphs. He found that the good readers evaluated information presented in the paragraphs with respect to i t s relevance to the topic but the poor readers were unable to identify the theme or topic of a paragraph effectively. A second experiment was carried out on a sample of 102 introductory psychology students. Scores on the Iowa Silent Reading  Test were obtained and the top 16 were chosen as the "good readers" and the bottom 16 were chosen as the "poor readers". The treatment involved reading six paragraphs. The subjects were then asked to write what they could r e c a l l . Each subject received a blank page with a cue word (either a thematic or non-thematic concept word) for each paragraph and was asked to write down what they could remember. Better readers remembered more from the thematic cues and remembered much more than the poor readers. The poorer readers actually recalled s l i g h t l y more from the non-thematic cues. Thus poor readers showed no evidence of using structural strategies. The second general trend that the studies show is that readers who used the author's structural patterns in organizing their recalls were able to r e c a l l more information than those who did not. Palmer, Slate, and Graves (1980) had grade nine students read two well organized passages of expository text with clearly identifiable 22 superordinate ideas and then write re c a l l s . Their analysis of the protocols indicated that most high a b i l i t y ninth graders used the same structure for organizing their recalls as did the authors in the original passages while most low a b i l i t y students did not. Students who employed the strategy of using the author's textual organization recalled more information from the passages than those who did not. They also found that simplifying the vocabulary and syntax of a passage did not necessarily change those outcomes. Meyer, Brandt, and Bluth (1980) stated that, "Use of the top-level structure was expected to relate to the amount of information remembered; organization has been shown to be a crucial variable in learning and memory." (p. 75). This was supported by findings from a study by McGee (1981), involving 120 third and f i f t h grade students. Forty grade 3 good readers were randomly selected from a l l third graders comprehending on a 3-0-4.5 grade level (S.R.A. Primary II). Forty grade 5 poor readers were randomly selected from a l l f i f t h graders comprehending on a 3.0-4.5 grade level (S.R.A., Multilevel Blue). McGee administered two attributive style passages written at a third grade level of readability and containing an identical three-level hierarchy of importance and a similar number of idea units. Oral recalls were tape recorded after the students had read the passages. She found that f i f t h grade good readers were more aware of text structure, and recalled proportionately more total and superordinate idea units than f i f t h grade poor or third grade good readers. Further, f i f t h grade poor readers displayed some awareness of text structure and recalled proportionately more superordinate ideas than third grade good readers. 23 The third general trend that the studies show, is that cues embedded in the structure of the text may help less able readers to follow the author's organizational structure. Winograd (1984) examined the differences between good and poor eighth graders' strategies in summarizing expository text. Students were f i r s t interviewed to determine what they knew about the summarizing process. Following the interview, the students read six different expository passages, answered five multiple choice questions for each passage, wrote a sixty word summary, rated the relative importance of each sentence in each passage and selected the five most important sentences from each passage. Winograd found that the poor readers used c r i t e r i a in determining what was important in a passage that were different from those of the good readers, who followed the author's organizational pattern. Poor readers judged importance based on whatever captured their interest. These results suggested that the a b i l i t y to identify important text information i s an underlying strategic s k i l l that enhances the summarization process. Berkowitz and Taylor (1981) conducted a study in which sixth grade good and poor readers read material with which they were familiar. The students were given prior cues as to the superordinate ideas. The results of the study suggest that poorer readers can perform similarly to more competent readers on r e c a l l , provided advance cues are given and the material is written at below grade level readability to accommodate the s k i l l s of the poorer readers. Hartley, Morris and Trueman (1981) adapted a. passage of about 300 words. It described food preparation and consumption during the Middle Ages. Two versions were prepared. Version A contained 24 headings in the margin written in the form of questions. Version B contained headings in the margin written in statement form. The headings were cues to the main idea of each of the five paragraphs in each passage. Twenty-one children from a remedial class took part in the study. The average age was 12 years 8 months and their reading ages ranged from 7 years 3 months to 10 years 8 months. The children were given 15 minutes to read the appropriate passage and then were tested on their r e c a l l . One week later the same test was administered again. Hartley et al reported that cues in the form of advance questions, headings, or embedded questions a l l helped the less able children identify the superordinate and subordinate aspects of the author's organization. They speculated that headings written in the form of questions may be the best type of cue. This, of course, was an important part of the SQ3R method described by Robinson (1961). In conclusion, the evidence suggests that good and poor readers have different developmental patterns and propensity to use the author's structure. The Effectiveness of Headings in Aiding the Reader Robinson (1961) stated, "Superior students given selections with headings in and headings omitted read the former no faster nor comprehend them any better, although evidence indicates that such headings can with training be used to increase speed and comprehension" (p. 29). Robinson and Hall (1941) had suggested that, "A higher-level reading s k i l l greatly emphasized in how-to-study work 25 is the use of headings in non-fiction material. Since these headings indicate the ideas to be developed, their effective use should make for more eff i c i e n t reading" (p. 251). They found in fact that few college students had the higher level s k i l l s to make use of headings. Christensen and Stordahl (1955) tested Air Force basic trainees' r e c a l l on passages with and without headings, including headings phrased as questions, and found no significant differences between groups. Aulls (1975), in a study of the effects of prose structure on sixth grade readers' l i t e r a l r e c a l l scores, found that the presentation of a main idea statement appeared to have a significant influence on r e c a l l . Meyer (1975) completed a study which indicated that not only did the presentation of a main idea statement have a positive effect, but the closer i t was to the beginning of the passage, the more lik e l y i t was to be recalled. Studies were needed to determine i f headings which were a restatement of the main idea (or a question which could cue the main idea) would aid poorer readers to focus on important aspects of the author's structural organization. Doctorow, Wittrock and Marks (1978), in a study involving 488 sixth graders, found that students reading stories with paragraph headings which restated or cued the main ideas of the paragraphs produced a significantly greater r e c a l l than did those reading the same stories without the paragraph headings on tests used in the experiments with the high-ability readers and with the low-ability readers. They also found that student generated topic sentences produced even greater r e c a l l . Dee-Lucas and DiVesta (1980) also found that student generated 26 topic sentences produced the greatest r e c a l l . Their study compared the effects of topic sentences, headings, additional sentences restating the main idea, and distractor sentences which restated a de t a i l . These features were investigated both when they were generated by the reader or provided in text. The subjects were 133 female students enrolled in introductory psychology courses. A l l students read a passage which consisted of 15 paragraphs. Each paragraph consisted of 3 sentences describing a mineral in terms of physical property, mode of processing, and use. Recall consisted of a free r e c a l l test, a test for knowledge of passage structure, and a matching test which involved sorting general and specific topics. A l l the features tested had significant effects only i f they were generated by the subject. Subject generated topic sentences produced by far the greatest r e c a l l . These findings supported the use of headings which would involve the student with the author's structural organization. Hartley, Kenely, Owen and Trueman (1980) streamed 200 grade two students into groups of high, medium, and low a b i l i t y on the basis of primary school records and first-year examination results. The children studied a short passage in one of four conditions: control, passage with the t i t l e , passage with headings in the form of statements, and passage with headings in the form of questions. The results showed that the headings had a marked effect for every a b i l i t y l e v e l . Overall i t made no difference whether or not the headings were in the form of statements or questions, but low-ability pupils did significantly better with headings in the form of questions. Holley, Dansereau, Evans, Collins, Brooks, and Larson (1981) 27 conducted a study which involved ninety-five college students. Students read passages with and without headings and half the students were given instruction on the use of headings as processing aids. The results showed that headings accounted for an increase of 11 percent in immediate r e c a l l and 44 percent in delayed r e c a l l five days later. The scores for the groups with or without training were not significantly different. Alverraan (1982), in a study involving thirty grade ten students, found that the use of headings made a significant contribution to students' a b i l i t y to r e c a l l details and main ideas after one week. Brooks, Dansereau, Spurlin and Holley (1983) reported on two experiments involving 132 psychology students. The f i r s t experiment showed that embedded headings reliably inproved delayed test scores. In the second experiment i t was found that instruction in the use of headings as processing aids improved test performance. Presentation of headings without instruction, however, failed to increase test scores on the second experiment. Meyer (1984) concludes, "Titles and subtitles can be employed to focus on the macropropositions and e x p l i c i t l y signal the top-level structure of the text." (p. 133). In summary, evidence is not yet conclusive as to the relative merits of headings, with or without prior instruction. However, headings are shown in most studies to serve as an aid to memory. 28 Summary The influence of text structures had been seriously discussed by educators for a considerable time before 1975. Empirical research by Meyers (1975) and findings of other studies at that time showed that information placed higher in the hierarchical structure of passages was recalled better than information placed lower in the hierarchical structure. Subsequent studies demonstrated that i t would be useful to examine the a b i l i t y of students at various grade levels to use top level organization of text. Until Danner (1976), empirical research had not been done to indicate that the use of the author's structure to f a c i l i t a t e r e c a l l was affected by developmental factors. It was speculated that readers who were able to use the author's structure in formulating r e c a l l of the information presented by the author would be able to remember more information in both immediate and delayed testing. It became evident in current empirical research that this was the case. It i s evident, from current research, that good and poor readers use different r e c a l l strategies. Poor readers have d i f f i c u l t y in following the author's organization in that they are unable to distinguish the relative importance of each idea presented by the author. This suggests that less able readers would be aided by cues which indicate the relative importance of the ideas in a passage, while these cues would be helpful, but not necessary, for good readers. Finally, evidence indicates that headings are the most effective means of cueing for less able readers to "access" the author's 29 structure. It i s not yet clear whether prior instruction i s necessary for headings to be useful in increasing r e c a l l , as results from recent empirical studies are conflicting. Headings were shown, in most studies, to serve as cues to memory. 30 CHAPTER 3 METHODOLOGY This study was designed to investigate the effects of headings in expository text on the written r e c a l l of students in grades 5 and 6. It i s one component of a three part study. The other two parallel studies emphasized grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). The selection of the subjects i s discussed, the instruments described, and the administration and scoring procedures are outlined. Subjects The participants in the study were students in grades 5 and 6. The complete study included parallel studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). One class of each grade from each of the Brit i s h Columbia school d i s t r i c t s of Richmond, Maple Ridge, and Surrey was chosen. The subjects were from intact, regular, non-streamed grade 5,6 and 7 classes and from intact, regular, non-streamed English or Social Studies classes at grades 8,9 and 10. The communities chosen are a l l suburbs of Vancouver, B.C. and are comparable in their socioeconomic status (Canada Census, 1981). An analysis of variance done between the three school d i s t r i c t s on the percentile scores achieved on the reading comprehension test 31 indicate no significant difference between students in the three school d i s t r i c t s . F(2,297)=.0975, p=.3785. See Table 1 for the percentile scores by d i s t r i c t . Table 1 Percentile Scores on Gates-MacGinitie Reading Test, Canadian Edition  (1980), Comprehension Subtest, By D i s t r i c t Percentile Scores D i s t r i c t n Mean S.D. Maple Ridge 102 53-35 30.69 Richmond 99 54.31 26.74 Surrey. 99 58.63 27.62 An analysis of variance done between grades on the reading comprehension test percentile scores indicate no significant differences between grades. F(5,294)=0.617, p=0.6867. See Table 2 for the mean percentile scores, T-scores, and grade equivalents at each grade l e v e l . From the total of the 68 grade 5 and 68 grade 6 students completing the reading comprehension test and the two written recalls, 50 were randomly chosen from each grade level, using a table of random numbers, to participate in the study. In the two parallel studies, 50 students at each grade level were also randomly chosen from a total of 70 at grade 7, 62 at grade 8, 76 at grade 9, and 54 at grade 10. 32 Table 2 Reading Scores on Gates-MacGinitie Comprehension Subtest by Grade Scores Grade Grade Equiv. T-Score Percentile n 5 Mean 6.18 53.22 59.14 50 S.D. 2.48 9.83 29.04 6 Mean 6.85 5:.52- 54.00 50 S.D. 2.51 9.58 28.32 7 Mean 7.96 52.78 58.22 50 S.D. 2.31 8.66 26.41 8 Mean 8.22 49.82 51.04 50 S.D. 2.62 9.88 30.84 9 Mean 9.37 50.78 53.16 50 S.D. 2.48 9.24 28.82 10 Mean 10.54 52.52 56.90 50 S.D. 2.16 9.61 27.43 33 Materials The Gates-MacGinite Reading Test, Canadian Edition (1980), Comprehension subtest was selected to provide a standard measure of reading comprehension for a l l subjects participating in the study. This test was selected for several reasons. It contains Canadian content and i t was norraed over a population of Canadian students. Also, several forms of the test are available so that appropriate forms could be used in parallel studies at the grade 5,6,7, and 8 levels (Gibbs, 1985; King, 1985) thus enabling comparisons to be made across the six grades. The r e l i a b i l i t y of the comprehension items range from .85 to .92 (Teacher's Manual, p. v). Forms D and E of this test encompass grades 5 and 6. The hand-scored version of the test was used. A battery of seven expository prose passages of approximately 300 words each, was written in a descriptive format. The passages were authored by Megan Crowhurst (1984), and revised by the researchers. The passages range in readability from approximately grade 3 to grade 11 as determined by the Fry Readability Formula (Fry, 1968). See Table 3. 34 Table 3 Readability of Passages Grade Level Topic Readability level 4 Parrots 3 5 Kwakiutl Indians 6 6 Venus Flytrap 7 7 Haida Indians 7 8 Anorexia Nervosa 10 9 Painkilling Drugs 11 10 Earthquakes 10 The readability levels of the grade 8 and 9 passages are high because a key multisyllabic word in each passage, "anorexic" in the grade 8 passage and "anesthesia" in the grade 9 passage, was repeated several times in each passage. The Fry Readability Formula, based on sentence and word length, does not take word repetition into account. Thus with the resultant high number of syllables, an a r t i f i c i a l l y high readability level i s obtained. However, when the Fry Readability Formula was again applied to the passages counting the syllables in "anorexia" and "anesthesia" only once, the grade 8 passage had a readability level of grade 9 and the grade 9 passage had a grade 10 readability l e v e l . The seven passages each contain. 5 superordinate ideas and 26 subordinate ideas in 5 paragraphs, as determined by the three researchers. The passages are well organized with a clearly defined 35 superordinate idea stated in the f i r s t sentence of each paragraph. Versions of each passage were prepared with and without headings. Headings were not indented and were set apart from each paragraph by a double space. They are a restatement of the superordinate idea or a cue to that idea. The prose in the passages with and without headings is identical. See Appendix A for two sample passages. Copies of the other passages can be obtained from Dr. T. Westermark, Department of Language Education, University of B r i t i s h Columbia. Each passage was printed with a dot-matrix Apple Computer printer onto one side of one sheet of 8 1/2 x 11 inch paper. The passages written for the grades 4, 5, and 6 levels were chosen for use in this study. "Appropriate passages from the same battery were selected for use in para l l e l studies at the grade 7,8,9 and 10 levels (Gibbs, 1985; King, 1985). A l l participants in this study read the grade 4 level passage to ensure that less able readers had the opportunity to demonstrate a b i l i t y to follow the author's organizational structure without being handicapped by passage readability l e v e l . Administration of Instruments A pil o t study was conducted with 50 grade 5 to 10 students in October, 1984. Students were assigned to read either the grade 4 level passage or their grade level passage. They were also assigned to read the passage either with headings or without headings. The treatment was administered to the students following the procedures 36 outlined in chapter 1. The students' written recalls were scored but the resultant data was not statistically analyzed. Students participating in the pilot study did not participate in the actual study. Results of the pilot study prompted a revision of the passages to eliminate ambiguities, to ensure that the superordinate ideas were stated in the topic sentences, and to revise the headings to reflect the main idea. Protocols were also revised to correspond with the changes made to the passages. Although the instructions to the students remained unchanged, responses to student questions were standardized to avoid giving any verbal cues. Scoring procedures were also revised to include a format evaluation. The instruments were administered in a three week period in November and December, 1984, during regular English or Social Studies classes in the high schools and in the mornings in the elementary schools. The research was done by the author and by the authors of parallel studies at the grade 7,8,9 and 10 levels (Gibbs, 1985; King, 1985). Students were numbered in alphabetical order on class l i s t s . Using a table of random numbers, half of the students at each grade level were randomly assigned to receive the passages with headings, the remaining students received passages with no headings. Half of each class was also randomly chosen to receive the passage at the grade 4 level f i r s t , the remaining students receiving their grade level passage f i r s t . Students assigned the passages with headings read both the grade 4 and the grade level passages with headings. The remaining students read both the grade 4 level and the grade level 37 passages without headings. Three sessions were spent with each class: a testing session, and two reading-recall sessions. A l l students were f i r s t given the appropriate level Comprehension subtest of the Gates-MacGinite Reading Test, Canadian Edition (1980). The test was administered by the researchers at the secondary school and by the classroom teachers or the researchers at the elementary schools, following the procedures outlined in the Teacher's Manual. Appropriate forms of the same test were selected for use in paral l e l studies at higher grade levels (Gibbs, 1985; King, 1985). In each of the second and third sessions, each student received the passage that had been assigned to the student through the random procedure conducted prior to the experimental session. These sessions were a l l conducted by the researchers at both the secondary and elementary levels. Students were instructed to study the passage for 10 minutes after which they were to write down a l l that they could remember from the passage without referring back to the passage read. At this point no further explanations were given. After 10 minutes had elapsed the passages were collected and students received lined foolscap. They were told to put their name, grade and school at the top of the paper. They were then told to write down everything they could remember about the passage they had just read. No discussion was allowed. They were given 25 minutes to complete the task. Questions from students regarding method or format of the r e c a l l were answered with, "Do what you think i s best". 38 Scoring and Tabulation of Data A l l marking was done by the author and the authors of the p a r a l l e l studies (Gibbs, 1985; King, 1985). Ten percent of the written r e c a l l s were then remarked by the authors. The mean of the in t e r r a t e r correlations was .838, ranging from .976 for grade 4 l e v e l (parrot passage) subordinate ideas to .743 for grade l e v e l superordinate ideas. This was considered to be an acceptable l e v e l of co r r e l a t i o n . Recall protocols were developed l i s t i n g the superordinate and subordinate ideas i n the order that they appeared i n the passage. Two adjacent boxes were drawn to the l e f t of each of the ideas on the protocols. (See Appendix B). The number of ideas r e c a l l e d , the organization of the r e c a l l , and the format of the r e c a l l were evaluated. Scoring procedures augmented evaluation procedures from Niles (1955). See Appendix B for the Directions for Scoring. A covering score sheet was completed for each student p a r t i c i p a t i n g i n the study. See Appendix C. This score sheet contains the student i d e n t i f i c a t i o n number; student grade; sex; reading score i n percentiles, T-scores and grade equivalents; the version of the passage read, headings or no headings; and the number of superordinate and subordinate ideas r e c a l l e d , the organization" of the superordinate and subordinate ideas and the format scores of both the grade 4 and the grade l e v e l passages. This completed the scoring procedures used. 39 Design In order to test the effects of headings on students' organization and r e c a l l of ideas a "Posttest-Only Control-Group Design" was used (Borg and Gall, 1979, p. 548). St a t i s t i c a l Procedures The data were processed using analysis of variance between mean posttest scores of the experimental (with headings) group and the control (without headings) group and between groups of different grade or reading a b i l i t y l e v e l . Pearson product-moment correlations were calculated to investigate the relationships between ideas and organization. The Scheffe^test was used to detect developmental patterns from grade 5 through 10. Summary This chapter has outlined the methodology of the study. It included a description of the subjects and materials; outlined the administration of the instruments; reviewed the scoring and the tabulation of data; and reported the design and s t a t i s t i c a l procedures. 40 CHAPTER 4 ANALYSIS OF DATA In this chapter the results of the analysis of the data w i l l be presented. The research questions w i l l be restated followed by a discussion of the relevant data. The effects of the independent variables of headings, grade level and reading a b i l i t y level on a variety of dependent measures w i l l be described. Analysis of variance (Anova) was conducted comparing the various effects. Differences are noted as significant i f p<.05. Pearson product-moment correlations have been calculated in an effort to investigate the relationship between the number of ideas recalled and organization of the written r e c a l l . For the developmental issues, the Scheffe' test, was used to discover significant patterns among the grades. The analyses in this chapter are based on the results from grade 5 and 6 students along with data from the parallel studies of grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). Data from 300 students were used in the entire study. This included data from 50 students at each grade level who were chosen from a larger population in order to achieve equal c e l l sizes. Using a table of random numbers, 50 students at the grade 5 level were chosen from a total possible of 68 who participated in the study, and 50 grade 6 students were chosen from 68. In the two parallel studies, 50 students at each grade level were also randomly chosen from a total of 70 at grade 7, 62 at grade 8, 76 at grade 9, and 54 at grade 10. 41 Effect of Headings on Recall of Superordinate and Subordinate Ideas Research Question 1.1 In a written r e c a l l task, are more superordinate and subordinate ideas recalled when the prose to be recalled has (a) headings or (b) no headings, within each grade level? Grade 5 Students It was found that there were no significant differences between mean scores (p<.05) when an analysis of variance was conducted (see Table 4) on the Grade 5 scores for superordinate and subordinate idea re c a l l when they were categorized as recalled from passages with headings and passages without headings Grade 6 Students It was found that there were no significant differences between mean scores (p<.05) when an analysis of variance was conducted (see Table 5) on the Grade 6 idea r e c a l l scores. 42 Table 4 Effect of Headings on Recall of Superordinate and  Subordinate Ideas by Grade 5 Students MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. Parrot passage Superordinate Ideas 1.76 1 .01 : 1.78 1.01 .02 Subordinate Ideas 8.94 5.53- 10.22 5.42 1.28 Grade Level passage Superordinate Ideas 2.12 1.42 1.84 1.30 .28 Subordinate Ideas 8.84 n=25 4.81 9.52 n=25 4.75 .68 «p<.05 43 Table 5 Effect of Headings on Recall of Superordinate and  Subordinate Ideas by Grade 6 Students MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. Parrot passage Superordinate Ideas 1.78 .97 1.38 1.19 .40 Subordinate Ideas 9.88 4.51 10.28 4.21 .40 Grade Level passage Superordinate Ideas 1.12 1.20 1.44 1.02 .32 Subordinate Ideas. 6.16 3-77 5.78 3-36 .38 n=25 n=25 *p<.05 44 Research Question 1.2 In a written r e c a l l task, are more superordinate and subordinate ideas recalled when the prose to be recalled has (a) headings or (b) no headings, within each grade by reading level? Grade 5 and 6 Good and Poor Readers Subpopulations of the grade 5 and 6 students were next taken on the basis of their Gates-MacGinite Reading comprehension subtest percentile scores. Students who had scored above the 50th percentile were labelled good readers while those who had scored below the 50th percentile were' labelled poor readers. There was no significant difference between the mean scores of grade 5 good readers who read the parrot or grade level passages with headings and the mean scores of grade 5 good readers who read the parrot or grade level passages without headings. Significant differences were found between mean scores of grade 5 poor readers. For the parrot passage, the mean number of subordinate ideas for the parrot passage recalled by the grade 5 poor readers reading passages with headings was 7.09 out of a possible 25.00. Those who had passages without headings had a mean of 3-10 (see Table 6). This difference of 3.99 was significant at the .05 leve l . F(1,14)=4.863, P<.05. There was no significant difference between the mean scores of 45 Table 6 E f f e c t of Headings on R e c a l l of Superordinate and  Subordinate Ideas by Grade 5 Students by Reading Level MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. GOOD READERS n=12 n=20 Parrot passage Superordinate Ideas 1.67 1.05 1.93 .96 .26 Subordinate Ideas 9.75 5.94 12.00 4.49 2.25 Grade Level passage Superordinate Ideas 2.38 1.56 2.08 1.34 .30 Subordinate Ideas 10.29 4.89 10.63 4.39 .34 POOR READERS n=11 n=5 Parrot passage Superordinate Ideas 1.64 .92 1.20 1.10 .44 Subordinate Ideas 7.09 3.90 3.10 1.19 3.99* Grade Level passage Superordinate Ideas 1.95 1.21 .90 .55 1.05 Subordinate Ideas 6.91 4.28 5.10 3.66 1.81 »p<.05 46 grade 6 good readers who read the passages with headings and the mean scores of grade 6 good readers who read the passages without headings. Similarly, there was no significant difference between the mean scores for the grade 6 poor readers reading passages with and without headings (see Table 7). Research Question 1.3 In a written r e c a l l task, are more superordinate and subordinate ideas recalled when the prose to be recalled has (a) headings or (b) no headings, over grade levels 5,6,7,8,9, and 10? Grades 5 through 10 Results for the effect of headings on idea r e c a l l for grades 5 and 6 were compiled with results from the two studies at grades 7 and 8 (King, 1985), and grades 9 and 10 (Gibbs, 1985). For this combined set of data, 300 subjects were used. An Anova was run which showed the results following. The significant effects due to grade w i l l be mentioned here but discussed in response to question 5. For superordinate ideas from the parrot passage there were significant effects due to grade, F(5,294)=8.197, p<.000, and due to heading, F (1,298)=5.011, p<.05. Students who read passages with headings had a mean r e c a l l score of 2.30 out of a possible 5.00 while those without headings scored a mean of 2.03 (see Table 8). No 47 Table 7 Effect of Headings on Recall of Superordinate and Subordinate Ideas by Grade 6 Students by Reading Level MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. GOOD READERS n= 15 n= 13 Parrot passage Superordinate Ideas 1.90 .85 1.76 1.30 .14 Subordinate Ideas 11.67 3.64 11.00 3.91 .67 Grade Level passage Superordinate Ideas 1.37 1.39 1.81 1.03 .44 Subordinate Ideas 7.37 3.95 7.00 3.21 .37 POOR READERS n=10 n= :12 Parrot passage Superordinate Ideas 1.60 1.15 .96 .94 .36 Subordinate Ideas 7.20 4.49 9.50 4.55 .30 Grade Level passage Superordinate Ideas .75 .75 1.04 .86 .29 Subordinate Ideas. 4.35 2.74 4.46 3.12 .11 *p<.05 Table 8 Effect of Headings on Recall of Superordinate and Subordinate  Ideas by Grade 5 Through 10 Students MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. Parrot passage Superordinate Ideas 2.30 1.11 2.03 1.10 .27* Subordinate Ideas 13.67 5.90 14.62 : 5.58 .95 Grade Level passage Superordinate Ideas 1.76 1.32 1.83 1.21 .07 Subordinate Ideas 9.54 5.12 9.58 4.77 .04 n=150 n=150 *p<.05 49 significant effects were found due to heading by grade F(5,288)=0.378, p=0.864. For subordinate ideas from the parrot passage there were significant effects due to grade, F(5,294)=26.875, p<.000. No significant effects were found due to heading, F(1,298)=2.915, p=.089, or due to heading by grade, F(5,288)=0.681, p=0.638. For superordinate ideas from the grade level passage, there were significant effects due to grade F(5,294)=8.771, p<.000. No significant effects were found due to heading, F(1,298)=0.288, p=0.592, or due to heading by grade F(5,288)=1.643, p=0.148. For subordinate ideas from the grade level passage, there was a significant effect due to grade, F(5,294)=9.342, p<.000. No significant effects were found due to heading, F(1,298)=0.006, p=0.936, or due to heading by grade, F(5,288)=0.316, p=0.903. Subpopulations from the total data were taken on the basis of reading test percentiles. As before, students who scored above the 50th percentile were labelled good readers while those who scored below the 50th percentile were labelled poor readers. There were no significant differences in the re c a l l measures between those who read passages with headings (H) and those who read passages with no headings (NH) for either the good readers (n=174: H=87; NH=87) or the poor readers (n=121: H=59; NH=62). Since there were no differences at this level i t was decided to form 2 smaller subpopulations which would represent very good readers (>84th percentile on comprehension test), and very poor readers (<16th percentile on comprehension test). Once again, no significant differences were found for either good or poor readers. 50 Effect of Headings on the Organizational Structure  of the Students' Written Recall Research Question 2.1 In a written r e c a l l task, do the students' written recalls follow the author's organizational structure more closely when the prose to be recalled has (a) headings or (b) no headings, within grade levels? Grade 5 Students The presence or absence of headings did not significantly affect the grade 5 students' written r e c a l l in following the author's organizational structure. Mean scores for superordinate organization by those students who read passages with headings were identical to those students who read passages without headings. Differences in their subordinate organization were also not significant (see Table 9). Grade 6 Students The presence or absence of headings did not significantly affect the grade 6 students' written r e c a l l in following the author's organizational structure. Mean scores for superordinate organization 51 Table 9 Effect of Headings on Organization for Grade 5 Students MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. Parrot passage Superordinate Org. .44 .51 .52 .51 .08 Subordinate Org. 1.96 1.34 2.28 1.34 .32 Grade Level passage Superordinate Org. .56 .50 .56 .50 .00 Subordinate Org. 2.48 1.50 2.40 1.55 .08 n=25 n=25 »p<.05 52 by those students who read passages with headings were identical to those students who read passages without headings. Differences in their subordinate organization were also not significant (see Table 10). Research Question 2.2 In a written r e c a l l task, do the students' written recalls follow the author's organizational structure more closely when the prose to be recalled has (a) headings or (b) no headings within grade levels by reading levels? Grade 5 and 6 Good and Poor Readers The presence or absence of headings did not significantly affect the students' written r e c a l l in following the author's organizational structure for good (>50th percentile) or poor (<50th percentile) readers. There were no significant differences or patterns in the mean scores for the organization measures of grade 5 good readers (see Table 11). There were also no significant differences or patterns in the mean scores for the organization measures of grade 6 good readers (see Table 12). There were no significant differences or patterns in the mean scores for the organization measures of grade 5 poor readers (see 53 Table 10 Effect of Headings on Organization for Grade 6 Students MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D, MEANS S.D. Parrot passage Superordinate Org. .72 1.02 .52 .51 .20 Subordinate Org. 1.60 1.12 1.84 1.37 .24 Grade Level passage Superordinate Org. .48 .51 .40 .50 .08 Subordinate Org. 1.56 1.29 1.28 1.21 .28 n=25 n=25 p<.05 54 Table 11 Effect of Headings on Organization for Grade 5 Students by  Reading Level MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. GOOD READERS n=12 n=20 Parrot passage Superordinate Org .42 .51 .55 .51 .13 Subordinate Org 2.16 1.75 2.40 1.43 .24 Grade Level passage Superordinate Org .58 .51 .55 .51 .03 Subordinate Org 3.00 1.71 2.65 1.60 .35 POOR READERS n=11 n=5 Parrot passage Superordinate Org .45 .52 .40 .55 .05 Subordinate Org 1.82 .87 1.80 .84 .02 Grade Level passage Superordinate Org .63 .50 .60 .55 .03 Subordinate Org 2.00 1.18 1.40 1.89 .60 *p<.05 55 Table 12 Effect of Headings on Organization for Grade 6 Students by  Reading Level MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. n=15 n=13 GOOD READERS Parrot passage Superordinate Org Subordinate Org 1.00 1.73 1.20 1.16 .54 1.69 .52 1.25 .46 .04 Grade Level passage Superordinate Org .47 .52 .38 .51 Subordinate Org 1.60 1.12 1.62 1.35 POOR READERS Parrot passage Superordinate Org Subordinate Org n=10 .30 1.40 .48 1.08 n=12 .50 2.00 .52 1.54 .09 .02 .20 .60 Grade Level passage Superordinate Org .50 .52 .41 .51 .09 Subordinate Org 1.50 1.58 .92 1.00 .58 »p<.05 56 Table 11). There were also no significant differences or patterns in the mean scores for the organizational measures for grade 6 poor readers (see Table 12). Research Question 2 . 3 In a written r e c a l l task, do the students' written recalls follow the author's organizational structure more closely when the prose to be recalled has (a) headings or (b) no headings, over grade levels 5,6,7,8 9 and 10? Grades 5 Through 10 Results for the effect of headings on organization for grades 5 and 6 were compiled with the results from the two studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). An Anova was run which showed the results following. The significant effects by grade w i l l be mentioned here but discussed in answer to question 5. For superordinate organization from the parrot passage there were no significant effects due to grade, F(5,294)=0.929, p=0.462, due to heading, F(1,298)=0.095, p=0.758, or due to heading by grade, F(5,288)=1.33M, p=0.250. For subordinate organization from the parrot passage there were significant effects due to grade, F(5,29*0=10.219, p>.000. No significant effects were found due to heading, F(1,298)=0.948, 57 P=0.331, or due to heading by grade, F(5,288)=1.554, p=0.173. For superordinate organization from the grade level passage, there were no significant effects due to grade, F(5,294)=1.657, p=0.145 but there were for heading, F(1,298)=4.505, p<.05. Without headings, 61 percent (a mean of .61 out of 1.00) of the students organized the superordinate ideas of their recalls in the same order as the author. This compares with a lower total of 49 percent (mean of .49) for those students who had headings on their passage (see Table 13). There were no significant effects due to heading by grade, F(5,288)=1.869, p=0.100. For subordinate organization from the grade level passage, there were significant effects due to grade, F(5,294)=8.815, p<.000, but not due to heading, F(1.298)=0.241, p=0.624, or due to heading by grade, F(5,288)=1.412, p=0.220. As with idea r e c a l l , subpopulations were chosen to represent good readers (>50th percentile) and poor readers (<50th percentile). There were no significant differences attributed to the presence or absence of headings. Further subpopulations were selected for very good readers (>84th percentile) and very poor readers (<l6th percentile). Only one significant difference was found (p<.05). This was in the superordinate organization of the grade level r e c a l l . Very good readers in the without headings group had a mean of .83 which represented 83 percent or 24 out of 29 students who organized the r e c a l l in a similar way to the author. This compared to a mean of .50 for the with headings group which represented 15 out of 30 very good readers who followed the author's organizational pattern. 58 Table 13 Effect of Headings on Organization for Grade 5 Through 10 Students MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. MEANS S.D. MEANS S.D. Parrot passage Superordinate Org. .54 .62 .52 .50 .02 Subordinate Org. 2.48 1.41 2.63 1.40 .15 Grade Level passage Superordinate Org. .49 .50 .61 .49 .18* Subordinate Org.. 2.15 1.36 2.23 1.40 .08 n=150 n=150 *p<.05 59 F(1,59)=7.750, p<.01. No significant differences were found betweend the mean organization scores of very poor readers reading passages with headings and very poor readers reading passages without headings. The Relationship Between Organization and the Number of  Superordinate and Subordinate Ideas on the Recall Research Question 3 Do students r e c a l l more superordinate and subordinate ideas when they more closely follow the author's organization in their written recalls, over grades 5,6,7,8,9 and 10? Grades 5 through 10 There does appear to be a relationship between the organization of the r e c a l l and the number of superordinate ideas recalled, over grades 5 through 10. Pearson product moment correlations were computed to see i f there was a relationship between the amount of r e c a l l and the way i t was organized. A l l correlations presented in tables 14 and 15 were significant (p<.05) on the basis of 111" values calculated for 298 degrees of freedom on a two-tailed test. Correlations for the parrot passage are indicated in Table 14. Superordinate organization had low correlations with superordinate 60 Table 14 Correlation Between Organization and Amount of Recall for the Parrot Passage Superordinate Subordinate Ideas Ideas Superordinate Organization .1624 .1222 Subordinate Organization .3810 .5031 Table 15 Correlation Between Organization and Amount of Recall for  the Grade Level Passage Superordinate Subordinate Ideas Ideas Superordinate Organization .3923 .2504 Subordinate Organization .3403 .5876 61 ideas (.1624) and subordinate ideas (.1222). Subordinate organization, however, had better correlations: .3810 with superordinate ideas and .5031 with subordinate ideas. Correlations for the grade level passage are shown in Table 15. These correlations are higher than the correlations for the parrot passage. Superordinate organization correlated .3923 with superordinate ideas and .2504 with subordinate ideas. Subordinate organization correlated .3403 with superordinate ideas and .5876 with subordinate ideas. This shows that a relationship exists between r e c a l l organization and number of superordinate and subordinate ideas recalled. The Effect of Headings on the Format of the Written Recall  Research Question 4 Are there differences in the format of the written r e c a l l when the prose to be recalled has (a) headings, (b) no headings, and are there differences over reading levels? Grade 5 Students For the grade 5 students, the only significant differences found were for the use of section headings in the recalls of each of the passages and also for students using one grouping (see Table 16). 62 Table 16 Effect of Headings on the Number of Grade 5 Students Who Used Each Format Characteristic MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. Parrot Passage Title 15 16 1 Section Headings 8 0 8* One Grouping 6 13 7* Point Form 7 4 3 Numbering 6 3 3 Paragraphs 10 11 1 n=25 n=25 Grade Level Passage Title 13 12 1 Section Headings 10 1 9* One Grouping 8 12 4 Point Form 5 6 1 Numbering 7 4 3 Paragraphs 12 11 1 n=25 n=25 *p<.05 63 r Eight students who read passages "with headings" used section headings in their parrot recalls while none of those did from the "without headings" group. F(1,48)=11.294, p<.05. With the grade level passage ten of the "with headings" group used section headings while, one of the "without headings" group did. F(1,48)=11.172, p<.01. For the parrot passage, six students who read passages "with headings" used one grouping while 13 of those from the "without headings" group used one grouping. The difference of 7 was significant. F(1,48)=4.356, P<.05. Grade 6 Students Significantly more students who read the passages with headings included section headings in their written r e c a l l on the parrots passage than students reading the parrot passages without headings. F(1,48)=7.579, p<.01. The "with headings" group also had significantly fewer recalls done using point form than those in the "without headings" group on the grade level passage. Four of the "with headings" group used point form while eleven of the "without headings" used point form. The difference of seven was significant. F(1,48)=4.941, p<.05. (See Table 17). Grades 5,6,7,8,9,10 Results for the effect of headings on format for grades 5 and 6 were compiled with the results from the studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). See Table 18 for the 64 Table 17 Effect of Headings on the Number of Grade 6 Students Who Used Each Format Characteristic ~~~ MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. Parrot Passage Title 15 16 1 Section Headings 6 0 6* One Grouping 13 15 2 Point Form 6 10 4 Numbering 2 2 0 Paragraphs 11 8 3 n=25 n=25 Grade Level Passage Title 12 13 1 Section Headings 3 1 2 One Grouping 16 13 3 Point Form 4 11 7* Numbering 1 3 2 Paragraphs , 9 8 1 n=25 n=25 *p<.05 65 Table 18 Effect of Headings on the Number of Students in Grades 5 Through 10 Who Used Each Format Characteristic MEASURE WITH HEADINGS WITHOUT HEADINGS DIFF. Parrot Passage T i t l e 92 95 3 Section Headings 30 0 30* One Grouping 54 58 4 Point Form 39 35 4 Numbering 14 8 6 Paragraphs 75 82 7 n=150 n=150 Grade Level Passage T i t l e 86 83 Section Headings 30 2 28* One Grouping 57 56 1 Point Form 31 38 7 Numbering 11 9 2 Paragraphs 77 81 4 n=150 n=150 *p<.05 66 number of students who used each format characteristic. An Anova was run which showed the following results. The significant effects by-grade w i l l be mentioned here but discussed in answer to question 5. For the use of a t i t l e on the parrot r e c a l l , there were no significant effects due to grade, F(5,294)=0.355, p=0.879, heading, F(1,298)=0.128, psO.721, or heading by grade, F(5,288)=1.966, p=0.084. For the use of a t i t l e on the grade level r e c a l l there were no significant effects due to grade, F(5,294)=0.789, p=0.558, or heading F(1,298)=0.123, p=0.726. There were significant effects due to heading by grade, F(5,288)=2.395, p<.05. Mean scores for each of the grades have been graphed in Figure 1. Points connected with the solid line represent the mean scores at each grade of students who read the passages with headings. Points connected with the dotted line represent the mean score at each grade of students who read the passages without headingsi Of special note i s the apparent positive effect by headings at the grade 8 level which surpasses the effect at any of the other grades. For the use of section headings on the parrot r e c a l l , there was no significant effect due to grade, F(5,294)=0.789, p=0.558, however there was a significant effect due to heading, F(1,298)=36.986, p<.000. Twenty percent of a l l students who had headings on their passages put headings on their r e c a l l s . None of those who read passages without headings used headings on their recalls for the parrot passage. There were no significant effects due to heading by grade, F(5,288)=0.789, p=0.558. For the use of section headings on the grade level r e c a l l , there were no significant effects due to grade, F(5,294)=2.039, p=0.073, but 67 Figure 1 Interaction of Grade and Heading For the Use of  a T i t l e on the Written Recall From the Grade Level Passage 1.0 .9 .8 .7 .6 Mean . 5 Score . 4 .3 .2 .1 0.0 5 6 7 8 9 10 Grade Recall from passages with headings _ _ Recall from passages without headings 68 there were significant effects due to heading, F(1,298)=30.745, p<.000. Once again twenty percent of a l l students who had headings on their passages put headings on their r e c a l l s . Only one percent of those who did not have headings on their passages used headings on their r e c a l l s . There were no significant effects due to heading by grade, F(5,288)=1.475, p=0.198. For the use of one grouping on the parrot r e c a l l , there were significant effects due to grade, F(5,294)=3.031, P<.05, but no significant effects due to heading, F(1,298)=0.236, p=0.628, or due to heading by grade, F(5,288)=1.403, p=0.223. For the use of one grouping on the grade level r e c a l l , there were significant effects due to grade, F(5,294)=4.899, p<.000, but no significant effects due to heading, F(1,298)=0.015, p=0.902, or due to heading by grade, F(5,288)=1.209, p=0.305. For the use of point form on the parrot r e c a l l , there were no significant effects due to grade, F(5,294)=1.060, p=0.383, due to heading F(1,298)=0.287, p=0.593, or due to heading by grade, F(5,288)=1.189, P=0.314. For the use of point form on the grade level r e c a l l , there were no significant effects due to grade, F(5,294)=0.757, p=0.581, due to heading, F(1,298)=0.923, P=0.337, or due to heading by grade, F(5,288)=1.511, p=0.186. For the use of numbering on the parrot r e c a l l , there were significant effects due to grade, F(5,294)=3.528, p<.01, but no significant effects due to heading, F(1,298)=1.846, p=0.175, or due to heading by grade, F(5,288)=1.231, p=0.295. For the use of numbering on the grade level r e c a l l , there were 69 significant effects due to grade, F(5,294)=5.604, p<.000, but no significant effects due to heading, F(1,298)=0.230, p=0.632, or due to heading by grade, F(5,288)=1.056, p=0.385. For the use of a paragraph format on the parrot r e c a l l , there were significant effects due to grade, F(5,294)=2.907, p<.05, but no significant effects due to heading, F(1,298)=0.671, p=0.413, or due to heading by grade, F(5,288)=0.868, p=0.503. For the use of a paragraph format on the grade level r e c a l l , there were significant effects due to grade, F(5,294)=4.414, p<.01, but no significant effects due to heading, F(1,298)=0.224, p=0.637, or due to heading by grade F(5,288)=0.626, p=0.680. Subpopulations of good (percentile >50) and poor (percentile <50) readers were selected. Once again, the only significant differences were positive effects by headings on the use of section headings in both r e c a l l s : for the good readers on the parrot passage, F(1,172)=30.906, p<.001; on the grade level passage for the good readers, F(1,172)=18.453, p<.001; for the poor readers on the parrot passage, F(1,119)=6.903, p<.01; on the grade level passage for the poor readers, F(1,119)=10.976, p<.01. Further subpopulations of very good (percentile >84) and very poor (percentile <16) readers were selected. Significantly positive (p<.05) effects for headings were found in the use of t i t l e s on the parrot recalls of the very good readers as well in the use of section headings on both the parrot and grade level r e c a l l s : for t i t l e s on the parrot passage, F(1,57)=4.583, p<.05; for parrot section headings, F(1,57)=14.008, p<.001; and for grade level section headings, F(1,57)=10.188, p<.01. With the very poor readers there 70 were significantly positive (p<.05) effects with headings for the use of point form on the parrot r e c a l l and for the use of section headings on the grade level r e c a l l . For parrot point form, F(1,29)=7.484, p<.05. For parrot section headings, F(1,29)=5.052, p<.05. Developmental Factors Research Question 5 In a written r e c a l l task, i s there evidence of developmental effects in (a) number of ideas recalled, (b) organization of the written r e c a l l , and (c) format, when the results of this study are compared to similar studies done at the grade 7,8,9 and 10 levels (Gibbs, 1985; King, 1985)? Grades 5 through 10 In the written r e c a l l task, there was evidence of developmental effects over grades 5 through 10 in (a) number of ideas recalled, (b) organization of the written r e c a l l , and (c) format, when results of recalls from passages with and without headings were combined. The data from this study were combined with the data from two parallel studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). An Anova was run which showed significant effects due to grade for many of the variables. The effect of each of these 71 variables were then investigated more closely. Multiple comparisons were made with the data, using Scheffe"' s method to determine i f there were significant differences between the mean scores of the grades. Unless otherwise noted the lower grades in each comparison have the lower means. See Tables 19 and 20 for the mean scores for ideas and organization from the parrot passage and the grade level passage. For superordinate ideas from the parrot passage, there were significant effects due to grade, F(5,294)=8.197, p<.000. Scheffe'1 s method showed that there was a significant difference (p<.05) between mean scores of grade 5 and grade 10 students and between mean scores of grade 6 and each of grades 8,9 and 10. For subordinate ideas from the parrot passage, there were significant effects due to grade, F(5,294)=26.875, p<.000. There were significant differences between mean scores of grade 5 and each of grades 7,8,9, and 10, between mean scores of grade 6 and each of grades 7,8,9, and 10, and between mean scores of grade 7 and grade 10. For superordinate ideas from the grade level passage there were significant effects due to grade, F(5,294)=8.771, p<.000. There were significant differences between mean scores of grade 7 and each of grades 6 and 8. In this instance, the grade 7 mean was higher than the grade 8 mean as well as the grade 6 mean. For subordinate ideas from the grade level passage, there were significant effects due to grade, F(5,294)=9.342, p<.000. The grade 6 mean was significantly lower than the means from each of grades 5,7,8,9, and 10. In the superordinate organization of the parrot and grade level passages there were no significant effects due to grade. 72 Table 19 Recall of Superordinate and Subordinate Ideas by Grade GRADE MEASURE 5 6 7 8 9 10 Parrot Passage Superordinate Ideas Means 1.77 1.58 2.18 2.29 2.42 2.73 S.D. 1.00 1.09 1.13 1.08 .91 1.06 Subordinate Ideas Means 9.58 10.08 14.70 15.69 16.83 17.98 S.D. 5.46 4.32 4.84 4.98 4.24 4.88 Grade Level Passage Superordinate Ideas Means 1.98 1.28 2.53 1.20 1.81 1.98 S.D. 1.36 1.12 1.07 1.05 1.35 1.15 Subordinate Ideas Means 9.18 5.97 10.23 9.44 11.96 10.57 S.D. 4.75 3-54 4.24 4.89 5.11 5.02 n 50 50 50 50 50 50 Table 20 Recall Organization by Grade GRADE MEASURE 5 6 7 8 9 10 Parrot Passage Superordinate Org. Means .48 .62 .42 .50 .60 .56 S.D. .50 .81 .50 .51 .49 .50 Subordinate Org. Means 2.12 1.72 2.38 2.76 3.06 3-28 S.D. 1.33 1.25 1.24 1.29 1.36 1.39 Grade Level Passage Superordinate Org. Means .56 .44 .66 .44 .60 .58 S.D. .50 .50 .48 .50 .49 .50 Subordinate Org. Means 2.44 1.42 2.30 1.62 2.72 2.64 S.D. 1.51 1.25 1.23 1.03 1.43 1.27 n 50 50 50 50 50 50 73 For subordinate organization in the parrot r e c a l l , there were significant effects due to grade, F(5,294)=10.219, p<.000. There were significant differences between mean scores of grade 6 and each of grades 8,9 and 10, between mean scores of grade 5 and both grades 9 and 10, and between mean scores of grades 7 and 10. For subordinate organization in the grade level passage, there were significant effects due to grade, F(5,294)=8.815, p<.000. There were significant differences between mean scores of grade 6 (displaying the lowest mean) and each of grades 5,7,9 and 10, and between means of grade 8 and both grades 9 and 10. For format characteristics there were some significant effects due to grade. See Table 21 for the number of students who used each format characteristic at each grade. For the use of one grouping on the parrot r e c a l l , there were significant effects due to grade, F(5,294)=3.031, p<.05. For the use of one grouping on the grade level r e c a l l , there were significant effects due to grade, F(5,294)=4.899, p<.000. On both passages, significantly more grade 6 than grade 10 subjects used a "one grouping" format on their r e c a l l s . For the use of numbering on the parrot r e c a l l , there were significant effects due to grade, F(5,294)=3.528, p<.01. There was a significant difference between the numbers of grade 5 and grade 10 students using numbering on the parrot r e c a l l . For the use of numbering on the grade level r e c a l l , there were significant effects due to grade, F(5,294)=5.604, p<.000. There was a significant difference between numbers of grade 5 and grades 7,8,9 and 10 students using numbering on the grade level r e c a l l . In each case there was 74 Table 21 The Number of Students Who Used Each Format Characteristic  in Each of Grades 5 Through 10 GRADE MEASURE 5 6 7 8 9 10 Parrot Passage T i t l e 31 31 29 29 34 33 Section Headings 8 6 3 4 5 4 One Grouping 19 28 20 19 16 10 Point Form 11 16 16 8 11 12 Numbering 9 4 6 1 2 0 Paragraphs 21 19 24 27 33 33 Grade Level Passage T i t l e 25 25 27 30 32 30 Sectiou Headings 11 4 3 4 6 4 One Grouping 20 29 17 25 12 10 Point Form 11 15 14 8 10 11 Numbering 11 4 3 1 0 1 Paragraphs 23 17 28 21 35 34 75 greater use by the grade 5 subjects. For the use of paragraphs on the parrot r e c a l l the Anova indicated significant effects due to grade, F(5,294)=2.907, p<.05. The Scheffe method indicated no significant differences between the grades. For the use of paragraphs on the grade level r e c a l l , there were significant effects due to grade, F(5,294)=4.414, p<.01. There was a significant difference between number of students in grade 6 and each of grades 9 and 10 using paragraphs on the grade level r e c a l l . The students in the higher grades made more use of paragraphs in their r e c a l l s . Significant differences between the grades, relating to the use of other format characteristics were not found between the grades. Post Hoc Analysis of Developmental Factors Investigated in  Good and Poor Readers in Grades 5 through 10 The data from the three studies were analyzed further by choosing subpopulations of good readers (percentile >50) and poor readers (percentile <50). Multiple comparisons were again made between the means at each grade level using Scheffe's method. Unless otherwise noted, the lower grades in each comparison have the lower means. Good Readers The idea and organization r e c a l l mean scores for the good readers 76 at each grade level are shown in Tables 22 and 23. Scheffe's method showed that for superordinate ideas from parrots there was a significant difference between the mean scores of grade 5 and grade 10 and between the mean scores of grade 6 and grade 10. For subordinate ideas from parrots there were significant differences between mean scores of grade 5 and each of grades 7,8,9, and 10 and between mean scores of grade 6 and each of grades 7,8,9, and 10. For superordinate ideas from the grade level passage there was a significant difference between grade 6 and grade 7 mean scores. For subordinate ideas from the grade level passage a significant difference was found between mean scores of grade 6 and each of grades 8,9, and 10. In the superordinate organization of the parrots and grade level passage there were no significant differences. For subordinate organization in the parrot r e c a l l there were significant differences between grade 5 and grade 10 mean scores and between mean scores of grade 6 and each of grades 8,9, and 10. For subordinate organization in the grade level passage there were significant differences between grade 6 and grade 9 mean scores and between grade 8 and grade 9 mean scores. It i s also interesting to note that significantly more grade 6 good readers than grade 8 good readers used point form in their r e c a l l s . Poor Readers The idea and organization r e c a l l means for the poor readers at each grade level are shown in Tables 24 and 25. Scheffe's method 77 Table 22 Recall of Superordinate and Subordinate Ideas for Good  Readers by Grade GRADE MEASURE 5 6 7 8 9 10 Parrots Passage Superordinate Ideas Means 1.83 1.84 2.40 2.35 2.39 3-03 S.D. .99 1.06 1.12 1.04 .96 1.02 Subordinate Ideas • Means 11.16 11.36 15.72 17.38 17.89 18.92 S.D. 5.11 3.71 5.27 4.59 4.14 4.39 Grade Level Passage Superordinate Ideas Means 2.19 1.57 2.68 1.60 2.05 2.22 S.D. 1.41 1.24 1.05 1.08 1.52 1.14 Subordinate Ideas Means 10.50 7.20 10.85 12.10 14.32 12.03 S.D. 4.51 3.57 4.64 3.98 4.83 4.78 n 32 30 26 28 30 Table 23 Recall Organization for Good ] Readers by Grade GRADE MEASURE 5 6 7 8 9 10 Parrots Passage Superordinate Org. Means .50 .79 .47 .65 .57 .60 S.D. .51 .96 .51 .49 .50 .50 Subordinate Org. Means 2.31 1.71 2.57 3.08 3.18 3.67 S.D. 1.53 1.18 1.25 1.32 1.47 1.18 Grade Level Passage Superordinate Org. Means .56 .43 .70 .58 .61 .57 S.D. .50 .50 .47 .50 .50 .50 Subordinate Org. Means 2.78 1.61 2.37 1.77 3.11 2.90 S.D. 1.62 1.20 1.19 1.14 1.47 1.21 n 32 28 30 26 28 30 78 Table 24 Recall of Superordinate and Subordinate Ideas for Poor  Readers by Grade GRADE MEASURE 5 6 7 8 9 10 Parrot Passage Superordinate Ideas Means 1.50 1.25 1.87 2.23 2.40 2.28 S.D. .97 1.07 1.12 1.14 .88 .98 Subordinate Ideas Means 5.84 8.45 12.87 13.85 15.43 16.58 S.D. 3.76 4l57 3.57 4.83 4.25 5.34 Grade Level Passage Superordinate Ideas Means 1.63 .91 2.26 .77 1.45 1.63 S.D. 1.15 .83 1.10 .85 1.07 1.10 Subordinate Ideas Means 6.34 4.41 9.26 6.56 8.73 8.38 S.D. 4.07 2.89 3.57 4.14 3.85 4.65 n 16 22 19 24 20 20 Table 25 Recall Organization for Poor Readers by Grade GRADE MEASURE 5 6 7 8 9 10 Parrot Passage Superordinate Org. Means .44 .41 .32 .33 .60 .50 S.D. .51 .50 .48 .48 .50 .51 Subordinate Org. Means 1.81 1 .73 2.16 2.42 2.75 2.70 S.D. .83 1 .35 1.21 1.18 1.16 1.49 Grade Level Passage Superordinate Org. Means .63 .45 .63 .29 .55 .60 S.D. .50 .51 .50 .46 .51 .50 Subordinate Org. Means 1.81 1 .18 2.26 1.46 2.20 2.25 S.D. 1.12 1 • 30 1.33 .88 1.28 1.29 n 16 22 19 24 20 20 79 showed that for superordinate ideas from parrots there was a significant difference between the mean scores of grade 6 and grade 9. For subordinate ideas from parrots there were significant differences between mean scores of grade 5 and each of grades 7,8,9, and 10 and between mean scores of grade 6 and each of grades 8,9, and 10. For superordinate ideas from the grade level passage there were significant differences between grade 6 and grade 7 mean scores and between grade 1 and grade 8 mean scores. The grade 7 mean was significantly higher in each comparison. For subordinate ideas from the grade level passage a significant difference was found between mean scores of grade 6 and each of grades 7 and 9. There were no significant differences between any of the grade level means for organization of the parrots passage or the grade level passage. Summary The results of the analysis of the data have been presented in this chapter. Each of the research questions has been s p e c i f i c a l l y addressed. A post hoc analysis was also conducted to note differences between the mean scores of the grades for good and poor readers. A l l of these results w i l l be discussed in Chapter 5. 80 CHAPTER 5 SUMMARY AND CONCLUSIONS In this chapter a summary of the study and the results w i l l be given. The summary of the study w i l l review the purpose, materials used, collection of data, design and data analysis used. Following this, each question w i l l be addressed and the findings reviewed. Conclusions w i l l be drawn after discussion of each of the research questions. Discussion w i l l include results from similar studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). Results from the post hoc analysis of data w i l l be discussed and conclusions drawn. Finally, overall results at the grade 5 and 6 level, results throughout grades 5 to 10, general conclusions, as well as some implications for instruction and further research w i l l be discussed. Summary of the Study The purpose of this study was to investigate the a b i l i t y of students in grades 5 and 6: to use paragraph headings to f a c i l i t a t e r e c a l l , to re c a l l information material written in a classification/description form, and to use the author's organizational structures. The inclusion of data from two parallel 81 studies at grades 7 and 8 (King, 1985), and grades 9 and 10 (Gibbs, 1985), enabled the above points to be investigated in a developmental context of grades 5 through 10. Specifically the study sought to measure grade 5 and 6 students' written recalls of passages with and without headings. Recalls were evaluated on the number of superordinate and subordinate ideas recalled, the superordinate and subordinate organization of the recal l s , as well as the format of the recalls from both the grade level and grade 4 level passages. Materials The Gates-MacGinite Reading Test, Canadian Edition (1980), Comprehension subtest was used to determine the reading a b i l i t i e s of the students. A battery of expository prose passages was written by one author, Megan Crowhurst (1984), in a classification/description form at grade levels ranging from approximately grade 4 to approximately grade 10. This study used the grades 4,5 and 6 passages. Appropriate passages from this battery were also selected for use in parallel studies (Gibbs, 1985; King, 1985). This enabled comparison of the results of this study with results of the parallel studies. A l l passages were carefully constructed to each contain 5 superordinate and 26 subordinate ideas. Each student was tested on a passage written at his approximate grade level and on the grade 4 passage. Grade levels were determined by the Fry readability formula (Fry, 1968). Versions of the passages 82 were prepared with and without headings (see Appendix A for sample passages). A p i l o t study was conducted in September, 1984. Adjustments were made to materials at. that time. Subjects For this specific study, the subjects were three classes of grade 5 students and three classes of grade 6 students. They were selected from these schools: Kingswood Elementary, Richmond, B.C.; Whonnock Elementary, Maple Ridge, B.C.; and Georges Vanier Elementary, Surrey, B.C. The communities chosen are a l l suburbs of Vancouver and are comparable in their average socioeconomic status (Canada Census, 1981). The complete study included parallel studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985). Three classes from each of these grades also participated. Intact, regular, non-streamed, grade 5,6, and 7 classes were chosen for the study on the basis of a v a i l a b i l i t y . Students from grades 8 through 10 were chosen from intact, non-streamed, regular English and Social Studies classes. 83 Collection of the Data Three sessions were spent with each of the classes. In the f i r s t session the Comprehension subtest of the Gates-MacGinitie Reading Test, Canadian Edition (1980) was administered. The second session involved the administration of a re c a l l passage. At each grade l e v e l , one-half of the students were randomly chosen to receive the passage written at the grade 4 level f i r s t and the other students received the passage written at their own grade level f i r s t . At each grade l e v e l , one-half of the students were randomly chosen to receive passages with headings. The other half received passages without headings. The students were given 10 minutes to study the passages in each session. After 10 minutes the articles were collected and foolscap sheets were distributed. Students were given 25 minutes to complete the task. A l l papers were collected at that time. Information such as name, grade, and school was included on the papers. Early finishers were expected to read s i l e n t l y for the remaining time. The third session was similar to the second except the student groups were reversed so that the grade passages not written the f i r s t time were written. A l l other procedures were the same. A l l sessions were conducted within a three week period during the months of November and December, 1984. A protocol form was developed for the marking of each of the passages (see Appendix B). 84 Design A posttest-only control-group design was used (Borg and Gall, 1979, p. 548). Data Analysis Analysis of variance was used to analyze the ;relationships among the factors measured in the study which provided a basis for answering the research questions. Pearson product moment correlations were calculated to determine the relationships between organization and idea r e c a l l . For the developmental issues, the Scheffe test was used to discover significant patterns among the grades. Effect of Headings on Recall of Superordinate and  Subordinate Ideas Research Question 1.1 In a written r e c a l l task, are more superordinate and subordinate ideas recalled when the prose to be recalled has (a) headings or (b) no headings, within each grade level? The presence or absence of headings does not appear to affect the number of superordinate and subordinate ideas recalled at the grade 5 or 6 lev e l . No significant differences (p<.05) were found at grades 5 85 or 6 when an analysis of variance was conducted on scores for superordinate and subordinate idea r e c a l l from passages read with and without headings. Research Question 1.2 In a written r e c a l l task, are more superordinate and subordinate ideas recalled when the prose to be recalled has (a) headings or (b) no headings, within each grade by reading level? Grade 5 and 6 students were categorized into good readers (>50th percentile) and poor readers (<50th percentile) according to their scores on the Gates-MacGinite Reading Comprehension subtest. No significant differences were found in the amount of r e c a l l of superordinate Ideas and subordinate ideas from passages read with and without headings in scores from the grade 5 good readers and the grade 6 good and poor readers. A significant difference was found in the scores for the grade 5 poor readers. Headings appeared to have a positive effect on the number of subordinate ideas recalled from the parrot passages by the poor readers. 86 Research Question 1.3 In a written r e c a l l task, are more superordinate and subordinate ideas recalled when the prose to be recalled has (a) headings or (b) no headings, between grade levels 5,6,7,8,9 and 10? Using the grade 5 and 6 results and those from similar studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985) overall means were determined. Only one significant difference was found in the amount of idea r e c a l l when passages with and without headings were read. Headings seemed to positively affect the amount of superordinate ideas recalled from the parrot passage. Thus, over grades 5 to 10, students reading the passages with headings generally recalled more superordinate ideas from the parrot passage than students reading the passages without headings. Summary of the Effect of Headings on Recall of Superordinate and  Subordinate Ideas Over grades 5 to 10, headings seemed to positively affect the rec a l l of superordinate ideas in the grade 4 level passage. Headings appeared to hinder the re c a l l of subordinate ideas of grade 8 good readers. Popular reading strategies such as SQ3R use headings and other text features to increase comprehension of expository prose. These strategies are learned and students must be guided through the processes in order to benefit from them. Teaching these students 87 strategies of how to follow the author's organizational structure may-help to improve their comprehension. Effect of Headings on the Organizational Structure  of the Students' Written Recall Research Question 2.1 In a written r e c a l l task, the students' written recalls follow the author's organizational structure more closely when the prose to be recalled has (a) headings or (b) no headings, within grade levels? Headings appeared to have no significant effect on the grade 5 students' written r e c a l l in following the author's organizational structure. They also appeared to have no significant effect on the grade 6 students written r e c a l l in following the author's organizational structure. This was true for both the parrot and the grade level passage. 88 Research Question 2.2 In a written r e c a l l task, do the students' written recalls follow the author's organizational structure more closely when the prose to be recalled has (a) headings or (b) no headings, within grade levels by reading levels? Good and poor readers at the grade 5 and 6 levels who read passages with headings do not organize their written recalls to follow the author's structure more closely than those reading passages without headings. There was no evidence in the structure of the written recalls that indicated these groups used headings as a strategy to f a c i l i t a t e r e c a l l of the author's organizational structure. Research Question 2.3 In a written r e c a l l task, do the students' written recalls follow the author's organizational structure more closely when the prose to be recalled has (a) headings or (b) no headings, over grade levels 5,6,7,8,9 and 10? Including results from similar studies at grades 7 and 8 (King, 1985) and grades 9 and 10 (Gibbs, 1985), headings seemed to have a significantly negative effect on students following the author's superordinate organizational structure on the grade level passage. 89 There were no significant differences noted for other superordinate or subordinate measures. Summary of the Effect of Headings on the Organizational Structure of the Students' Written Recall This study has shown that students do not spontaneously use headings to r e c a l l the author 's organizational structure. The readability of the passage seems to have some effect on the r e c a l l of the author's organizational structures. Students appeared to have more d i f f i c u l t y following the author's organizational structure in grade level passages than in the grade 4 level passage when headings were present. The Relationship Between.Organization and the Number of  Superordinate and Subordinate Ideas on the Recall Research Question 3 Do students r e c a l l more superordinate and subordinate ideas when they more closely follow the author's organization in their written r e c a l l , over grades 5,6,7,8,9, and 10? There does appear to be a relationship between the organization of r e c a l l and the number- of superordinate and subordinate ideas recalled over grades 5 through 10. Results over grades 5 to 10 show moderate positive correlations 90 between subordinate organization and the number of superordinate and subordinate ideas recalled from the parrot passage. Positive moderate correlations are also found between the superordinate organization and the number of superordinate ideas recalled, and the subordinate organization and the number of subordinate ideas recalled, from the grade level passages. This indicates the existence of a relationship between the variables. Thus i t appears that at both the grade 4 and grade level passage, students who recalled more subordinate ideas also followed the author's subordinate organization more closely. On the grade 4 level passage i t appears that students who followed the author's subordinate organization more closely also recalled more superordinate ideas. On the grade level passage students recalled more superordinate ideas as they more closely followed the author's superordinate organization. 91 The Effect of Headings on the Format of the Written Recall Research Question 4 Are there differences in the format of the written r e c a l l when the prose to be recalled has (a) headings, (b) no headings, and are there differences over reading levels? At the grade 5 level a significant number of students reading passages with headings used section headings on their written recalls of the parrot and grade level passages, as opposed to those reading passages without headings. Students who read passages without headings made significantly more frequent use of one grouping on their parrot recalls than did those who read the passages with headings. Grade 6 students reading passages with headings, wrote their recalls of the parrot passage significantly more often using section headings when compared to grade 6 students reading passages without headings. The grade 6 students who read the grade level passage without headings made significantly greater use of point form than those who read the grade level passage with headings. Over grades 5 to 10, significantly more students reading passages with headings used section headings in their written recalls of both the parrot and grade level, passages than students reading passages without headings. Very good readers (>84th percentile), reading passages with headings, used t i t l e s on their parrot passage r e c a l l and section 92 headings on both parrot and grade level recalls more often than those reading passages without headings. As well, there was a significant effect due to heading by grade for the use of a t i t l e on the written r e c a l l from the grade level passage. Headings appeared to have a far greater positive effect at the grade 8 level than at any other grade. Very poor readers (<l6th percentile) reading passages with headings tended to use point form in their r e c a l l of the parrot passage and section headings in their grade level passage r e c a l l . Headings in a prose passage appear to affect the students' written r e c a l l . Although few students used section headings in their r e c a l l s , the results were s t a t i s t i c a l l y significant. Section headings in a r e c a l l , either quoted exactly from the passage or in a shortened form, seemed affected by headings in the given passages. Very few students reading passages without headings made up headings to f a c i l i t a t e the organization of their r e c a l l . The inclusion of a t i t l e on the written r e c a l l also seemed affected by headings. Although a l l passages with and without headings had t i t l e s , significantly more grade 8 students and very good readers (>84th percentile) over grades 5 to 10 reading passages with headings, included a t i t l e in their written r e c a l l . This may indicate more of an awareness of an overall text feature such as a t i t l e when the prose to be read i s sectioned by headings. 93 Developmental Factors Research Question 5 In a written r e c a l l task, is there evidence of developmental effects in (a) number of ideas recalled, (b) organization of the written r e c a l l , and (c) format, when the results of the study are compared to similar studies done at the grade 7,8,9 and 10 levels (Gibbs, 1985; King, 1985)? Results over grades 5 to 10 indicate that there are some developmental effects in the number of superordinate and subordinate ideas recalled from the grade 4 level passage when results of recalls of passages with and without headings were combined. The developmental aspect of comprehension, and r e c a l l of superordinate and subordinate ideas from a passage of low readability, was investigated by analyzing the results of a common task using materials common to a l l subjects at a l l grade levels studied. Students in grade 6 recalled significantly fewer superordinate ideas than students in grades 8,9, and 10. Thus, on a passage of low readability, older students were better able to r e c a l l main ideas of the passage read. Grade 5 and 6 students also recalled significantly fewer subordinate ideas from the grade 4 level passage than students in grades 7,8,9 and 10. Older students seemed better able to r e c a l l subordinate ideas on a passage of low readability than younger students. There did not seem to be any developmental effects in the number 94 of superordinate or subordinate ideas recalled from the grade level passages when results of recalls of passages with and without headings were combined. On the grade level passages the grade 7 group scored higher than any other grade and significantly higher than grade 6 and 8 students in recalling superordinate ideas. The grade 7 students may have been familiar with some of the content in the passage on Haida Indians since North American Indians are part of the grade 5 as well as the grade 7 curriculum. There were no significant differences between any other grades in number of superordinate grade level ideas recalled. The grade 6 students recalled significantly fewer subordinate ideas from the grade level passage than students in grades 5,7,8,9 and 10. The grade 6 group seemed to have some d i f f i c u l t y with the Venus Flytrap passage perhaps because the s c i e n t i f i c processes and some vocabulary (ie. digestion, c i l i a ) may have been unfamiliar and possibly caused comprehension problems. There were no significant differences in the superordinate organization r e c a l l of either the parrot or grade level passage. However, there was a developmental trend seen in the r e c a l l of the parrot subordinate organization. Significant differences were found between the grade 5 scores and the grades 9 and 10 scores; the grade 6 scores and the grades 8,9, and 10 scores; and the grade 7 and grade 10 scores. This may indicate that older students are better able to follow the author's schema within a paragraph when recalling a passage of low readability. On the grade level passages there were significant differences between the r e c a l l of subordinate organization of grade 6 students and the r e c a l l scores of grade 5,7,8,9 and 10 students; and between the grade 8 scores and the grade 9 and 10 95 scores. However there did not seem to be a developmental trend in this variable as the grade 6 and 8 scores were considerably lower than scores from any of the other grades. It i s interesting to note that many students at a l l grade levels organized the superordinate ideas from the parrot passage in a similar manner that was different from the author's organizational structure. The author f i r s t introduced the topic, gave a physical description of the parrot, then outlined the parrot's natural habitat. Many students recalled the introduction, the parrot's natural habitat, then described the parrot's physical features. It seemed clear that these students, although perhaps following the author's subordinate organization, used their own schema in recalling the superordinate ideas. A number of developmental, trends could be seen in the format of the recalls from grades 5 to 10. Grade 6 students used the "one grouping" characteristic in their recalls of the parrot passage significantly more often than the grade 10 students. Grade 5 students used "numbering" significantly more often than grade 10 students on their recalls from the parrot passage. On the recalls from the grade level passage a significant difference was found in the use of "numbering" between grade 5 and 7,8,9 and 10, with grade 5 students using "numbering" significantly more often than the other grades. Although no other significant differences were found between the grades in r e c a l l format, i t can be concluded that students' writing seemed to evolve from a r e c a l l format written as one grouping to a format consisting of two or more paragraphs. Similarly, younger students seemed to depend more on numbering to organize their 96 information than did older students. One grade 6 student recalled the passage information in reverse order but numbered the facts to coincide with the author's organizational structure. When the subjects were grouped into good readers (>50th percentile) and poor readers (<50th percentile) the findings were the same as for the total group on the six variables of superordinate ideas, subordinate ideas, and the superordinate organization of both the parrot and grade level passages. Scores of the subordinate organization do d i f f e r . On both the grade level and parrot passages no significant differences were found between any grade level group of poor readers. However significant differences were found between subordinate organization scores of grades 6 and 7 good readers and grade 9 good readers reading grade level passages. On the parrot passage significant differences were found between subordinate organization scores of good grade 5 and good grade 10 readers; and between grade 6 and grade 8,9 and 10 good readers. In both cases the lower grade had the lower mean. This may indicate that the difference in age and reading experience and i t s affect on the a b i l i t y to follow the author's subordinate organization i s greater in good readers than in poor readers. 97 Summary of Developmental Factors In this study many of the significant differences found between grade levels seem to group the subjects into two categories f a i r l y consistently. The mean scores of grade 5 and 6 students seemed to make up the lower group and the mean scores of grade 8,9 and 10 students seemed to make up the higher group. The grade 7 students f i t into either the high or low 'group depending on the variable examined. These groupings may be explained i f one accepts Piaget's (1955) description of the preadolescent stage (11-12 years of age) when the child starts to move from a concrete to an abstract level of thinking. The child cannot "deal c r i t i c a l l y with his own thinking" (Inhelder and Piaget, 1958, p.340) and i s thus unable to build his own theories or schemata effectively. The adolescent (age 14-15 years) i s capable of analyzing his own thinking and building his own theories based on more abstract rather than concrete ideas (Inhelder and Piaget, 1958). The child at grade 5 or 6 level may not be aware of the author's organization, particularly for expository material, since this structure i s a rather abstract concept. Students in grades 8,9 and 10, however, have had more experience with expository prose and, along with greater a b i l i t y to perceive the author's organizational structure, are able to r e c a l l more ideas and to r e c a l l the author's organizational structure more closely than the younger students on a passage of low readability. Older students also seemed more apt to group recalled ideas in paragraph form, a task requiring the more abstract concept of joining related ideas. Younger students tended to l i s t recalled ideas in one 98 grouping or to number the facts to indicate organization, a more concrete representation. Summary of Research Findings Findings from the study seem to indicate that headings do not f a c i l i t a t e students' r e c a l l of superordinate or subordinate ideas, or superordinate or subordinate organization of a passage of low or grade level readability. These findings were consistent within the grade 5 and 6 levels and within grades 5 to 10. At some levels headings seemed to be detrimental to the students' comprehension and r e c a l l of the passage, showing a lack of understanding of the use of the passage structure for study purposes. There did appear to be a relationship between the number of ideas recalled and the organization r e c a l l score over grades 5 to 10. It seemed, then, that students who followed the author's organization more closely were also able to r e c a l l more ideas. Those students who were aware of the author's organization seemed to be able to use this structure to f a c i l i t a t e their idea r e c a l l . Headings did seem to affect the format of students' written r e c a l l s . At a l l grade levels i t was found that students did not make up their own headings to f a c i l i t a t e their r e c a l l . Some students reading passages with headings used the author's headings or modified versions of the author's headings in their r e c a l l s . Some of these students l i s t e d a l l of the headings f i r s t , then wrote the subordinate ideas as one group. Others recalled the headings and subordinate 99 ideas following the format of the passage. Students reading passages without headings did not create written headings to help them r e c a l l the information from the passages. Grade 8 students and very good readers (>84th percentile) reading passages with headings seemed to include t i t l e s in their recalls more often than did those reading passages without headings. The presence of headings may have cued these students to the overall passage stucture including the t i t l e . Developmental effects were seen in the number of superordinate and subordinate ideas and in the subordinate organization of the recalls from the parrot passage written by students in grades 5 to 10. Older students scored higher in these variables than younger students. This may indicate that grade 8,9 and 10 students were able to use the author's subordinate organization to assist them in recalling more subordinate ideas from a passage of low readability. In contrast, the grade 5 and 6 students did not seem to use this strategy for r e c a l l . It would appear that the subjects of this study had not been instructed in the use of typographic-semantic cues as study strategies, either as part of a study technique such as SQ3R, or as features of text. This seemed evident from the analysis of the written results. As students get older and gain in reading a b i l i t y and experience they seem to become more aware of the concept of an organizational structure within a passage. However, without instruction, the older students' recognition and re c a l l of the structure and i t s use in f a c i l i t a t i n g r e c a l l of ideas seems limited to text written at a low readability l e v e l . 100 Post Hoc Analysis of Interrelationships of Research Variables Data from grades 5 through 10 were combined and analyzed to determine i f patterns of organization or trends could be determined on a post hoc basis. Eight major variables were examined as to their relationships with the format scores on the parrot and grade level passages. Headings and no headings means were combined. The eight major variables involved the number of superordinate ideas recalled, the number of subordinate ideas recalled, the superordinate organization score, and the subordinate organization score for both the parrot and grade level passages. Common trends were observed in the r e c a l l format. The greater the number of superordinate and subordinate ideas recalled and the higher the score achieved on superordinate and subordinate organization on both the grade 4 and grade level passages, the less frequent the use of the one grouping format and the greater the use of the paragraph format on the r e c a l l s . As well, the higher the score on the subordinate organization variable the less frequent the use of the point form format. Thus i t seemed that students who better comprehended the passages tended to organize their recalls in a higher level format that was most similar to the format of the passages read. An investigation was conducted to see i f the independent variable of reading a b i l i t y played a significant role in the dependent measures of amount and organization of r e c a l l for the passages. Pearson product-moment correlations were computed to see the relationship i between T-scores as obtained from the comprehension subtest of 101 Gates-MacGinitie and the scores for number of ideas and the organization of the r e c a l l . Correlations ranged from .1783 for superordinate ideas from the parrot passage to .2881 for subordinate ideas from the parrot passage, •3385 for superordinate ideas at grade level, and .4444 for subordinate ideas at grade l e v e l . Reading a b i l i t y did not have a high correlation with any of the measures but there was a moderate positive correlation between reading a b i l i t y level and the number ;of subordinate ideas recalled from the grade level passages. This indicates the existence of a relationship between the variables. These levels of correlations were predictable. As one would expect, passages of higher readability serve to highlight differences between students' reading a b i l i t y better than passages of lower a b i l i t y . It was determined that r e c a l l of grade level subordinate ideas can be used as a rough predictor of reading a b i l i t y . Correlations between the parrot passage measures and reading a b i l i t y were low. Two distinct trends could be determined from the post hoc analysis. One i s that as more ideas were recalled, the closer the written recalls followed the author's organization of the passage read. The second i s the positive correlational relationship between reading a b i l i t y level and the amount of subordinate idea r e c a l l from a passage of higher readability. 102 General Conclusions In conclusion, i t appeared that students in this study did not have the higher level s k i l l s to make use of headings in the writing of their passage r e c a l l s . This concurred with Robinson and Hall (1941) who had the same findings. The results also were similar to recent research findings by Brooks et a l (1983). Studies with contrary findings, which indicated that headings had a positive result, were Doctorow et al (1978) and Hartley et al (1980). Meyer (1984) indicated that t i t l e s and subtitles can be used to signal top-level structure in text. Brooks et al (1983) found that instruction in the use of headings improved written r e c a l l . A reasonable assumption to make would be that the students involved in the present study did not employ the s k i l l s necessary to use the typographic-sematic cues provided. This suggests that future studies would need to add an instructional component in order to c l a r i f y the role of headings as a useful typographic-semantic cue. Previous studies had not taken into account the developmental range of grade levels encompassed by this study. The significance of the developmental aspects of the written recalls from the experimental passages indicate the need for future studies to u t i l i z e or expand the range of grade levels. It was noted that there was a change in reasoning strategies from concrete to abstract at the preadolescent stage (Inhelder and Piaget, 1958), about 11-12 years of age. There i s a clear instructional implication. Study s k i l l s should be taught at the grade 6/7 level, as these students would be able to u t i l i z e the abstract reasoning s k i l l s necessary to internalize study s k i l l 103 strategies that involve the overview of text organization and awareness of the author's schema. It i s recommended that future studies u t i l i z e a pretest before administering the passages in order to control for the effects of previous knowledge. Passages may have to be revised i f a great deal of prior knowledge i s indicated. This may have been a factor on the grade 7 passage in the current study. Observation also indicated that students interpreted, the word "study", from the i n i t i a l instructions, in a variety of ways. Some students read the passage once, some appeared to read i t several times, and others appeared to be attempting to memorize the material verbatim. In summary, i t can be concluded that future studies in this area should: encompass a wide range of grades; control for the factor of prior knowledge; and u t i l i z e some form of instructional component. The instructional component should provide training in "accessing" the author's structure and using i t to advantage for study purposes. Particular note should be made of developmental aspects in the 11 to 12 year age group. Only after an instructional component i s included in a study of this type can i t be determined i f headings do indeed positively affect comprehension and r e c a l l of expository prose. 104 REFERENCES Alverraan, D. (1982) Restructuring text f a c i l i t a t e s written r e c a l l of main ideas. Journal of Reading, 25, 8, May, 754-758. Aulls, M.W. (1975). Expository paragraph properties that influence l i t e r a l r e c a l l . Journal of Reading Behaviour, 7, 4, 391-400. Berkowitz, S., & Taylor, B. (1981). The effects of text type and familiarity on the nature of information recalled by readers. In M.L. Kamil (Ed.), Directions in Reading: Research and Instruction. Clemson: N.R.C., 157-161. Borg, R. & Gall, M. (1978). Educational Research, Third Edition. Longman, New York. Bos, C.S. (1981). Remembering information from text: a comparison of the text structures of educable mentally retarded and average students' re c a l l s . In M.L. Kamil (Ed.), Directions in Reading;  Research and Instruction. Clemson: N.R.C. Brooks, L.W., Dansereau, B.F., Spurlin, J.E. & Holley, CD. (1983). Effects of headings on text processing. Journal of Educational  Psychology, 75, 2, 292-302. Brown, A.L., Carapione, J.C, & Day, J.D. (1981). Learning to-learn: 105 on training students to learn from texts. Educational Researcher, 10,2, 14-17. Census of Canada (1981). Population, Occupied, Private Dwellings,  Private Households, and Census and Economic Families in Private  Households. Selected Social and Economic Characteristics, B r i t i s h  Columbia. Statistics Canada. Christensen, CM., & Stordahl, K.E., (1955). The effect of organizational aids on comprehension and retention. The Journal of  Educational Psychology, 46, 2, 65-74. Clark, CH. (1982). Assessing Free,Recall. Reading Teacher, 35,4, 434-39. Crowhurst, M. (1984). Seven unpublished expository passages. University of B r i t i s h Columbia, Vancouver. Danner, F.W. (1976). Children's understanding in intersentence organization and the r e c a l l of short descriptive passages. Journal  of Educational Psychology, 68, 2, 174-183. Dee-Lucas, D., & Di Vesta, F.J. (1980). Learner-generated organizational aids; effects on learning from text. Journal of  Educational Psychology, 72, 3, 304-311. Doctorow, M., Wittrock, M.C., & Marks, C. (1978). Generative 106 processes in reading comprehension. Journal of Educational  Psychology, 70, 2, 109-118. Drum. P.A. (1984). Children's understanding of passages. In J. Flood (Ed.), Promoting Reading Comprehension. Newark: I.R.A., 61-78. Eamon, D.R. (1978-79). Selection and r e c a l l of topical information in prose by better and poorer readers. Reading Research Quarterly, 14, 2, 244-257. E l l i o t t , S.N. (1980). Children's knowledge and use of organizational patterns of prose in recalling what they read. Journal of Reading  Behaviour, 12, 3, 203-212. Fry, E. (1968). "A readability formula that saves time." Journal of  Reading, No. 11, 513-516. Gibbs, R.S. (1985). The Use of Headings and Text Organization as Aids to Recall of Expository Prose by Students in Grades 5 Through  10 With an Emphasis on Grades 9 and 10. Unpublished master's thesis, University of Br i t i s h Columbia, Vancouver. Harris, T.L., & Hodges, R.E. (1981), (Eds.). A Dictionary of Reading  and Related Terms. Newark: I.R.A. Hartley, J., Kenely, J., Owen, G., & Trueman, M. (1980). The effect of headings on children's r e c a l l from prose text. The B r i t i s h 107 Journal of Educational Psychology, 50, 504-507. Hartley, J., Morris, P, & Trueman, M. (1981). Headings in text. Remedial Education, 16, 1, 5-7. Holley, CD., Dansereau, D.F., Evans, S.H., Collins, K.W., Brooks, L., & Larson, D. (1981). U t i l i z i n g intact and embedded headings as processing aids with nonnarrative text. Contemporary Educational  Psychology, 6, 227-236. Inhelder, B. & Piaget, J (1958). The Growth of Logical Thinking. Basic Books Inc.: London (translated by A. Parsons and S. Milgram) King, CM. (1985). Effects of Headings on the Written Recall and Organization of Expository Text in Grades 5 Through 10  with Emphasis on Grades 7 and 8. Unpublished master's thesis, University of Br i t i s h Columbia, Vancouver. MacGinite, W.H. (1980). Gates-MacGinite Reading Tests: Canadian Edition, Teacher's Manuals and Student Booklet Forms for Levels D, E, F. Don M i l l s : Nelson. McGee, L.M. (1981). Good and poor readers' a b i l i t y to distinguish among and r e c a l l ideas on different levels of importance. In M.L. Kamil (Ed.), Directions in Reading: Research and Instruction. Clemson: N.R.C. 108 Meyer, B.J.F. (1975). The Organization of Prose and i t s Effects on  Memory. Amsterdam: North-Holland Publishing Company. Meyer, B. (1977). Organization in prose and memory: research with application to reading comprehension. In P. David Pearson (Ed.), Reading: Theory, Research and Practice. Clemson: N.R.C. Meyer, B., Brandt, D., & Bluth, G. (1980). Use of top-level structure in text: key for reading comprehension of ninth grade students. Reading Research Quarterly, 16, 72-103. Meyer, B.J.F.(1984). Organizational aspects of text: effects on reading comprehension and applications for the classroom. In J. Flood (Ed.), Promoting Reading Comprehension. Newark: I.R.A., 113-138. Niles, O.S. (1955). Unpublished informal written r e c a l l test battery. Boston University. Niles, O.S. (1965). Organization Perceived, in H.L. Herber (Ed.), Perspectives in Reading: Developing Study S k i l l s in Secondary  School• Newark, I.R.A. Palmer, R.J., Slater, W.H., & Graves, M.F. (1980). The effect of passage d i f f i c u l t y on good and poor readers' use of author's schema in written re c a l l protocols. In M.L. Kamil (Ed.), Perspectives on  Reading Research and Instruction. Washington: N.R.C, 38-41. 109 Piaget, J. (1955). The Language and Thought of the Child. New York: The New English Library Ltd. Robinson, F.P. (1941). Diagnostio and Remedial Procedures  in Effective Study. New York: Harper and Row. Robinson, F.P. (1961). Study s k i l l s for superior students in secondary school. The Reading Teacher, 15, 1. Robinson, F. P. (1970). Effective Study (Fourth Edition) New York: Harper and Row. Robinson, F.P. & Hall, P. (1941). Studies of higher-level reading a b i l i t i e s . Journal of Educational Psychology, 32, 241-252. Smith, N.B. (1964). Patterns of writing in different subject areas. Journal of Reading, 7, 2, 97-102 Smith, N.B. (1968). Be a Better Reader. Englewood C l i f f s , N.J.: Prentice H a l l . Tierney, R.J., Bridge C , & Cera, M.J. (1978-79). The discourse processing operations of children. Reading Research Quarterly, 14, 4, 539-573. Waters, H.S. (1978). Superordinate-subordinate structure in semantic 110 memory: the roles of comprehension and retrieval processes. Journal of Verbal Learning and Verbal Behaviour, 17, 587-591. Winograd, P. (1984). Strategic d i f f i c u l t i e s in summarizing texts. Reading Research Quarterly, 19, Summer, 404-425. 111 APPENDIX A SAMPLE PASSAGES 112 Parrots Beautiful and Interesting Birds Few birds are as interesting and beautiful as parrots. There are more than 700 kinds of birds in the parrot family. People like parrots because they can teach them to talk. They are also admired for their bright colours. How Parrots are Different Parrots are very different from other birds. The parrot always has a large beak like a hook. This beak is very strong. The bird uses i t to help him climb about. Parrots also use their feet to hold food and to help them climb. Parrots Live Where i t i s Warm Parrots can be found in a l l the warmer parts of the world. South America and Australia have the greatest number of different kinds. Many of them nest in trees. Some nest in c l i f f s . S t i l l others nest on the ground. Parrots Like Cages More than Other Birds Of a l l the birds kept as pets, parrots seem to l i k e cages the most. The parrot's cage should be large enough for him to move easily about without breaking his feathers.. Now-a-days most cages are made of stainless steel. This metal i s very strong and i s easy to clean. Sand or gravel should cover the bottom of the cage. The cage should be cleaned once a week. Parrots Eat Nearly Anything Parrots w i l l eat nearly anything that is given to them. Many things are not good for them, though. They may eat some f r u i t , but not a l o t . The best food for the parrot is a mix of seeds and nuts. Liquid vitamins should be added to the parrot's food. Following these simple rules w i l l help keep your parrot healthy and happy for many years. 113 The Venus Flytrap The Venus flytrap i s a plant that has interested people for hundreds of years because of the unusual way in which i t gets i t s food. The flytrap grows wild only in a small part of North and South Carolina in the United States. The s o i l there i s very poor and does not contain the nourishment the plant needs. Since the flytrap cannot get food from the earth as most plants do, i t catches insects and eats them instead. The structure of the flytrap leaf i s quite unusual. The leaf of the venus- flytrap i s made up of a leaf and a stem-like structure. The leaf is divided into two halves. In the middle of each half, several hairs stick into the a i r . The plant traps i t s prey. When an insect touches one of these hairs, the two halves quickly close on i t . Around the outside of the leaf about thirty-six finger-like growths called c i l i a are arranged in a row, like the teeth of a comb. When the leaf closes these c i l i a come together and keep the insect from escaping. The closing leaf, called a trap, looks like a pair of hands clasped so that the fingers form an interlocking pattern. The flytrap digests the trapped insects. Digestion begins once an insect has been caught, the edges of the trap form an airtight seal. Then juices from glands in the leaf f i l l the trap. These juices break down a l l the soft parts of the victim's body, but cannot digest the hard shells of insects lik e wasps and beetles. Once the juices have been reabsorbed by the plant, the trap opens and these hollow shells are l e f t for the wind to blow away. The whole process can take up to two weeks, depending on the size of the insect. A few years ago the venus flytrap was an endangered species. The venus flytrap was so popular that large numbers were dug up. They became very scarce and nearly died out. Today, however, thousands of flytraps are raised in greenhouses and sold around the world as house plants. They are a very popular plant for young people. 114 APPENDIX B DIRECTIONS FOR SCORING AND SAMPLE PROTOCOL SHEETS 115 Directions for Scoring I. To score the number of superordinate and subordinate ideas. Step 1. Record, on the protocol, the numerical order of the superordinate and subordinate ideas as they appear on the student's written r e c a l l 1) record the order of a l l ideas regardless of accuracy 2) write the numbers in the f i r s t column of boxes on the protocol Step 2. To determine credit for accuracy of recalled ideas. 1) record the second column of boxes on the protocol 2) mark credit as follows: (+) f u l l credit for f u l l ideas as presented on the protocol (1/2) one-half credit for partial ideas including from one-third to two-thirds of the idea as presented on the protocol (-) no credit for unacceptable ideas Step 3. Total the number of superordinate ideas correctly recalled and record the number in the appropriate box on the Student Information Record sheet. 1) maximum score i s 5 Step 4. Total the number of subordinate ideas correctly recalled and record the number in the appropriate box on the Student Information Record sheet. 1) maximum score i s 26 II. To score the organization of the superordinate and subordinate ideas Step 1. Evaluate the superordinate organization 1) For students who did not re c a l l a l l of the superordinate ideas, the assumption was made that i f three or more consecutive subordinate ideas were recalled within a paragraph, the gist of the superordinate idea was understood. The student receives no credit for recalling the superordinate idea, but credit may be given for superordinate organization. a) i f no superordinate idea was recalled from a paragraph, the paragraph must contain at least three consecutively recalled subordinate ideas in order to qualify for the following "averaging" procedure. b) on the protocol, write the middle number of the consecutively recalled subordinate ideas to the l e f t of the f i r s t column of the superordinate idea that was missing from the r e c a l l . This w i l l then represent the numerical order of that superordinate idea. 2) at least two superordinate ideas must be represented in order to receive credit for superordinate organization. 3) score 1 i f a l l of the represented superordinate ideas, 116 regardless of accuracy, are i n consecutive numerical order. Step 2 . Evaluate the subordinate organization 1) at l e a s t two subordinate ideas must be present i n a paragraph to r e c e i v e c r e d i t for subordinate organization 2) score 1 i f a l l of the r e c a l l e d subordinate ideas i n a paragraph are i n consecutive numerical order (n, n+1, n+ 2 . . . ) 3) t o t a l the score for subordinate organization from a l l paragraphs and record t h i s i n the appropriate box on the Student Information Record sheet 4 ) maximum score i s 5 I I I . To score the format of the written r e c a l l Step 1. Evaluate each of six format features. 1) score 1 i f the feature i s present 2) score 0 i f the feature i s absent 3) record each score i n the appropriate box on the Student Information Record sheet C r i t e r i a f o r scoring format features 1) t i t l e - an o v e r a l l heading r e l a t e d to the passage t o p i c , written at the top of the r e c a l l , and d i s t i n g u i s h e d from the r e s t of the r e c a l l by indenting, u n d e r l i n i n g , or spacing was considered a t i t l e 2) section headings - one to several words set apart from a section of written material by spacing, indenting, or u n d e r l i n i n g was considered a section heading. Recalls had to contain at l e a s t two section headings i n order to score one. 3) one grouping - i f a r e c a l l was written i n one d i s t i n c t grouping with no d i s t i n c t breaks, gaps, paragraphs, or sections i t was considered to be written i n one grouping. 4 ) point form - a l i s t i n g of f a c t s i n incomplete sentences, or s i n g l e words was considered to be i n point form. These f a c t s may or may not have been preceded by some i d e n t i f y i n g marking. 5) numbering - i f the r e c a l l was generally numbered so that the numbers applied to the o v e r a l l arrangement rather than being s p e c i f i c to one section or paragraph, the r e c a l l was considered to be numbered. 6) paragraphs - i n order to score one for "paragraphs" a written r e c a l l must have had at l e a s t two paragraphs. Each paragraph must have contained at l e a s t one complete sentence. Indenting at the beginning of the paragraph was not necessary, but a d i s t i n c t break, gap, or space had to be evident between paragraphs. The number of ideas contained i n each paragraph on the r e c a l l was not considered i n the evaluation. 117 Parrots ( )( )I. Few birds are as beautiful and Interesting as parrots ( )( )a)more than 700 different kinds in parrot family ( )( )b)people like them because they can teach them to talk ( )( )c)they are also admired for their bright colours ( )( )II. Parrots are different from other birds ( )( )a)large beak like a hook ( )( )b)very strong beak ( )( )c)uses his beak to help him climb about ( )( )d)use feet to hold food ( )( )e)use feet to help climb ( )( )III. Parrots l i v e where i t i s warm ( )i )a)South America ( )( )b)Australia ( )( )c)S.A. and Aus. have the greatest number of different kinds ( )( )d)many nest in trees ( )( )e)some nest in c l i f f s ( )( ) f ) s t i l l others nest on the ground ( )( )IV.Pel; parrots are kept in cages ( )( )a)parrots lik e cages more than other pet birds ( )( )b)cage should be large enough to move easily without breaking feathers ( )( )c)cages made of stainless steel now-a-days ( )( )d)this metal i s strong and easy to clean ( )( )e)sand or gravel should cover the bottom of the cage ( )( )f)cage should be cleaned once a week ( )( )V. Parrots eat nearly anything given to them ( )( )a)many things not good for them ( )( )b)may eat some f r u i t , but not a lot ( )( )c)mix of seeds and nuts are best food for them ( )( )d)liquid vitamins should be added to food ( )( )e)following simple rules keep parrots healthy ( )( )f)keep happy for many years 118 The Venus Flytrap ( )( )I.An interesting plant for hundreds of years ( )( )a)people interested in i t because of way i t gets food ( )( )b)only grows wild in a small part of North and South Carolina in the United States ( )( )c)so i l i s poor and doesn't contain the nourishment the plant needs ( )( )d)plant catches insects and eats them because i t can't get food from the earth as most plants do ( )( )II.The structure of the flytrap's leaf i s unusual ( )( )a)leaf i s made up of a leaf and stem structure ( )( )b)leaf divided into two halves ( )( )c)in middle of each half several hairs stick into air ( )( )III.The plant traps i t s prey ( )( )a)an insect touching the hairs causes the halves to quickly close on i t ( )( )b)36 finger-like c i l i a arranged around the outside of the leaf ( )( ) c ) c i l i a in a row like teeth on a comb ( )( )d)when the leaf closes, c i l i a come together to stop the insect from escaping ( )( )e)closed leaves called a trap ( )( )f)looks li k e a pair of hands clasped with fingers in an interlocking pattern ( )( )IV.The flytrap digests insects ( )( )a)digestion begins when insect i s caught ( )( )b)edges form airtight seal ( )( )c)juices from glands in leaf f i l l trap ( )( )d)juices break down a l l the soft parts of the victim's body ( )( )e)cannot digest the hard shells of insects like wasps and beetles ( )( )f)juices reabsorbed by plant ( )( )g)the trap opens and the hollow insect shells are l e f t for the wind to blow away ( )( )h)process can take two weeks, depending on the size of the insect ( )( )V.Was an endangered species a few years ago ( )( )a)because they were so popular large numbers were dug up ( )( )b)became scarce and nearly died out ( )( )c)thousands now raised in greenhouses ( )( )d)sold around the world as houseplants ( )( )e)very popular plant for young people 119 APPENDIX C STUDENT INFORMATION RECORD SHEET 1 2 0 STUDENT INFORMATION RECORD SHEET Student Number Grade Sex ( ) ( ) ( ) ( ) ( ) ( ) Gates-McGinitie Rdg. Comp. Scores Percentile Rank ( ) ( ) T-Score ( ) ( ) Grade Equiv. ( ) ( ) ( ) Headings Indicator Headings ( ) Parrots Content and Organization Superord. Ideas ( ) ( ) Subord. Ideas Superord. Org. Subord. Org. ( ) ( ) ( ) ( ) ( ) Parrots Format T i t l e ( ) Section Head. ( ) One Grouping ( ) Point Form ( ) Numbering ( ) Paragraphs ( ) Grade Level Content and Organization Superord. Ideas ( ) ( ) Subord. Ideas Superord. Org. Subord. Org. Grade Level Format ( ) ( ) ( ) ( ) < ) T i t l e ( ) Section Head. ( ) One Grouping ( ) Point Form ( ) Numbering ( ) Paragraphs ( ) 121 APPENDIX D SAMPLE OF RECALL FROM PARROT PASSAGE WITH EXPLANATION OF MARKING PROCEDURES 122 T r a n s c r i p t of Sample Re c a l l With Ideas Numbered* Parrots ® T h e r e are 700 d i f f e r e n t types of parrots .^They are an enjoyabe pet because you can teach them to t a l k . ^ T h e y are also admired for t h e i r many b e a u t i f u l c o l o r s . @ P a r r o t s are d i f f e r e n t than most b i r d s because o f ^ t s hook l i k e beak and its'Qtrong claws wich are used to hold food ancf^climb. ou can f i n d parrots i n most of the warmer areas of the world such as ^ S o u t h America a n d ^ u s t r a l i a . Most p a r r o t s ^ e s t i n trees or on the side o f ^ l i f f s . Parrots also seem tc$?njoy cages as long as t h e i r $ l o t s of room to move around without breaking any feathers. @bn the bottom of the cage t h e i r should be a mixture of sand and gravel P^k parrot i s known to eat j u s t about anything. ® T h e y enjoy a l i t t l e f r u i t once and awhile b u t ^ good mixture for them i s seeds and nuts.^These points w i l l help keep your parrot healthy and^iappy for many years. * numbers i n s e r t e d by marker. 123 Parrot Protocol Marking Sheet For Sample R e c a l l Parrots ( )( ) I . Few bi r d s are as b e a u t i f u l and I n t e r e s t i n g as parrots (I )(-r)a)more than 700 d i f f e r e n t kinds i n parrot family (2) (+)b)people l i k e them because they can teach them to t a l k (3) (-r)c)they are also admired for t h e i r bright colours (*!-)(•+)II. Parrots are d i f f e r e n t from other b i r d s C5)(+)a)large beak l i k e a hook ( )( )b)very strong beak ( )( )c)uses h i s beak;to help him climb about (£)(+)d)use feet to hold food (7) (-10e)use feet to help climb (8)0+)III. Parrots l i v e where i t i s warm (^)(+)a)South America (lO)(-r)b)Australia ( )( )c)S.A. and Aus. have the greatest number of d i f f e r e n t kinds 01) (+)d)many nest i n trees (.ll) Of) e) some nest i n c l i f f s ( )( ) f ) s t i l l others nest on the ground ( )( )IV.Pet parrots are kept i n cages (13) (X.) a)parrots l i k e cages mor» than other pot b i r d s (MOf )b)cage should be large enough to move e a s i l y without breaking feathers ( )( )c)cages made of s t a i n l e s s s t e e l now-a-days ( )( ) d ) t h i s metal i s strong and easy to clean 05) (+-)e)sand or gravel should cover the bottom of the cage ( )( )f)cage should be cleaned once a week (tW(+)V. Parrots eat nearly anything given to them ( )( )a)many things not good for them (17)0+)b)may eat some f r u i t , but not a l o t (}8)(+)c)mix of seeds and nuts are best food for them ( ) ( ) d ) l i q u i d vitamins should be added to food (tt) 0+)e) following simple r u l e s keep parrots healthy (20) (+)f )keep happy for many years 124 

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}]}"
                            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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0078303/manifest

Comment

Related Items