@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Education, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Bryce, Joy Alberta"@en ; dcterms:issued "2010-02-26T04:41:51Z"@en, "1978"@en ; vivo:relatedDegree "Master of Arts - MA"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """This study examined differences in oral reading behaviour of 58 grade one children attending achool in Richmond, B.C. Half the subjects received initial reading instruction through a phonics approach; half the subjects received initial reading instruction through a language experience approach. Among the findings were that subjects taught by the phonics approach, which emphasized letter-sound correspondence, produced more oral reading errors, more nonwords, and more substitutions with graphic and sound similarity to the response word than did children instructed by the language experience approach. Subjects taught by the language experience approach produced fewer errors and more substitutions syntactically and semantically acceptable, and more substitutions that did not alter the meaning of the sentence than the children instructed by the phonics approach. An analysis of the children's substitution errors for high, middle and low achievement groups was also discussed."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/21083?expand=metadata"@en ; skos:note "AN ANALYSIS OF THE ORAL READING ERRORS OF GRADE ONE PUPILS IN TERMS OF TWO TEACHING EMPHASES by Joy Alberta Bryce A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of .. MASTER OF. ARTS in THE FACULTY OF GRADUATE STUDIES (Faculty of Education) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1978 © Joy Alberta Bryce, 1978 In presenting th i s thesis in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Univers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th is thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thesis fo r f inanc ia l gain sha l l not be allowed without my writ ten permission. Department of Education. The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date September, 1978 i i ABSTRACT This study examined differences i n oral reading behaviour of 58 grade one children attending achool i n Richmond, B.C. Half the subjects received i n i t i a l reading instruction through a phonics approach; half the subjects received i n i t i a l reading instruction through a language experience app-roach. Among the findings were that subjects taught by the phonics app-roach, which emphasized letter-sound correspondence, produced more oral reading errors, more nonwords, and more substitutions with graphic and sound similarity to the response word than did children instructed by the language experience approach. Subjects taught by the language experience approach produced fewer errors and more substitutions syntactically and semantically acceptable, and more substitutions that did not alter the meaning of the sentence than the children instructed by the phonics app-roach. An analysis of the children's substitution errors for high, middle and low achievement groups was also discussed. i i i TABLE OF CONTENTS Chapter Page I. INTRODUCTION 1 Need for the Study 2 THE PROBLEM k Definition of Terms h GENERAL PROCEDURES 7 SUMMARY 8 ORDER OF PRESENTATION 8 I I . RELATED RESEARCH 10 Background 10 Whole Word Approach 13 Phonics Approach • 17 Comparison of Approaches . • • • 18 SUMMARY 26 III. DEVELOPMENT AND ADMINISTRATION OF THE INSTRUMENTS 28 Selection of Subjects . . . . . . . . . 28 Selection of Subgroups • 29 Reading Instruction Provided 29 Instruments Used 30 Collection of Data • • 30 Classification and Coding of Data 31 Scoring of the Data • • • 36 Data Analysis 36 SUMMARY 37 i v Chapter Page IV. PRESENTATION AND INTERPRETATION OF DATA 38 Part 1: Oral Reading Behaviour . . . . . . . . . . . . 38 Part 2: Substitution Components • • • Ui SUMMARY 56 V. SUMMARY, FINDINGS, CONCLUSIONS, AND EDUCATIONAL IMPLICATIONS 59 SUMMARY 59 Administration of Instruments . . . . . . . . . . . . . 60 Treatment of the Data • 60 FINDINGS 61 Part 1: Reading Behaviour 6 l Part 2: Substitution Components • 62 CONCLUSIONS 6U RECOMMENDATIONS 65 SUGGESTIONS FOR FURTHER RESEARCH 68 BIBLIOGRAPHY . 70 V LIST OF TABLES Table Page 1. Means, Standard Deviation and t-value for Reading Behaviours for Phonics and Language Experience Beginning Readers . . • • hO 2# Comparison of Means, Standard Deviation and t-values for Substitution Components for the Phonics Group and the Language Experience Group U 5 3« Comparison of Means, Standard Deviation and t-value for Nonwords for High, Middle, and Low Achieving Phonics and Language Experience Beginning Readers • h9 km Comparison of Means, Standard Deviation, t-value for Graphic Similarity of Substitution Errors, Words and Nonwords, for High, Middle and Low Achieving Phonics and Language Experience Beginning Readers • . . . . . . 51 5. Comparison of Means, Standard Deviation and t-value for Sound Similarity of Substitution Errors, Words and Nonwords, for High, Middle and Low Achieving Phonics and Language Experience Beginning Readers . . 53 6. Comparison of Means, Standard Deviation and t-value for No Meaning Change and Contextually Appropriateness of Substitution Errors for High, Middle, and Low Achieving Phonics and Language Experience Beginning Readers . . . . . . . . . . 5 5 ACKNOWLEDGEMENT I would like to extent sincere thanks to my advisor, Dr. Tory Westermark, for his patience, understanding and guidance, and to Dr. Kenneth Slade and Dr. Florence Pieronek, members of the examining com-mittee, for their helpful advice. In addition, I would like to express my gratitude to Dr. Robert Sweet for his guidance and suggestions i n the analysis of the data and to the administrators, teachers and students who willingly cooperated i n the study. Chapter I INTRODUCTION Considerable research has focused on the relationship between read-ing achievement and the method of reading instruction employed (Robinson, 197Uj Chall, 1967s Bond & Dykstra, 1967j Weber, 1968). In recent years, however, some researchers have f e l t that i n order to understand the child's a b i l i t y to acquire reading s k i l l s , or his failure to do so, we should look not solely to the method of instruction and achievement but also to the strategies the child employs i n his attempt to read. Barr (197k) states: In order to understand the processes used by beginning readers, we . . . must obtain independent evidence of what beginning readers do by observing them as they learn to read. (pp. 13-lU.) It has been suggested that the processes used by children learning to read might be successfully investigated by an analysis of oral reading errors (Smith, Goodman & Meredith, 1970j Goodman & Goodman, 1977j Weber, 1970j Biemiller, 1970; Barr and Page, 197U; Hood, 1976). Such an analysis should not be simply a tabulation of errors (such as insertions, omissions, and substitutions) that the child makes. An evaluation of each of these types of error should be made to determine how closely the error approxi-mates the correct response i n categories such as syntactic, semantic, and graphophonics appropriateness. \"The number of miscues a reader makes i s much less significant than the meaning of the language which results when a miscue has occurred.\" (Y. Goodman, 1972, p. 32.) 1 2 Many educators (Weber, 1970; Weber, 1968; Goodman, 1969j Burke, 1973) believe that errors should not be treated simply as \"incorrect\" res-ponses. As Weber (1970) states: Even casual observation shows that, i n one way or another, an error i s pa r t i a l l y correct. The correct features of an error are significant because they reveal what the reader chose as the basis for his response i n a particular instance. More generally, correct features of errors can be seen to reflect the sorts of information that a reader regularly utilizes i n identifying words. He uses the same strategies, presumably, whether or not the response i s f u l l y accurate, (p. U29.) Goodman (1965, 1969) suggests that the term error might have nega-tive connotations. He prefers to use the term \"miscue\" to indicate any deviation from the written material. \"Miscues are the windows of the read-ing process at work.\" (Goodman & Goodman, 1977, p. 323.) I t would thus appear that an analysis of oral reading errors on a qualitative as well as quantitative basis w i l l provide important informa-tion for the teacher. Barr and Page (197U) state: Teachers who are able to observe, analyze, and interpret oral reading responses possess one of the most useful s k i l l s for assessing children's reading. From understanding gained about the students' reading processes, a teacher can plan appropriate instruction and evaluate i t s effectiveness, (p. 103.) Need for the Study A number of educators have emphasized the importance of studying the oral reading behaviour of children as i t relates to reading instruc-tion. Chall (1973) states: The implications of these kinds of error data for under-standing the beginning reading process and for diagnosis and teaching based on individual needs are enormous, (p. 189.) Weber (1968) reviewed the literature pertaining to oral reading 3 errors and concluded that the child's errors may be due to the type of reading instruction that child has received. A few studies have compared and analyzed the oral reading errors of beginning readers who have been taught by different reading methods (for example, Barr, 1972$ Elder, 1971j DeLawter, 197Uj Burke, 1973 j Norton, 1976). These studies found that children taught by a phonics approach tended to use oral reading strate-gies related to the letters. Many of their errors were nonwords and/or words not from their reading vocabulary. On the other hand, children taught by the sight word approach tended to use oral reading strategies related to the whole word. They used few or no nonword responses and tended to choose their substitutions from words being taught at that par-ticular time. Most research analyzing the oral reading errors of beginning read-ers has been concerned with a sight word basal reader approach and/or a phonics approach. There appears to have been only three studies which have mentioned language experience (Dank, 1976j DeLawter, 197Uj Ewoldt, 1976). However, in each of these studies the language experience approach did not refer to a teacher developed program using the child's natural language and self selected sight words. Instead, basal readers called language experience readers were used such as the Oinn 360 and the Chandler Language Experience Readers. As MaryAnne Hall (1978) has recently pointed out in her survey of language experience research: Investigations of children's oral reading performance, using miscue analysis procedures, could be conducted with language experience materials. . . . However, no study investigating children's reading performance in conjunc-tion with the language experience approach was located in this survey of the literature, (p. UO.) The intention of this study is to analyze and compare the oral k reading behaviour of children learning to read by a language experience method with children learning to read by a phonics method, THE PROBLEM The purpose of the present study i s to determine the extent to which children instructed i n i n i t i a l reading by a language experience approach d i f f e r i n their reading behaviours from children who have been instructed i n i n i t i a l reading by a phonics approach. This study seeks to answer specifically the following questions: 1, What i s the effect of alternative instruction on the following aspects of reading—oral reading errors, repetitions, self-corrections, comprehension and fluency? 2 , What i s the effect of alternative instruction on the following aspects of substitution errors—words/nonwords, graphic similarity of words, graphic similarity of nonwords, sound similarity of words, sound similarity of nonwords, no meaning change and contextual appropriateness? (a) What are the effects of treatment on the seven compon-ents for the children instructed through a phonics app-roach and for the children instructed through a language experience approach? (b) What are the effects of treatment on these seven compon-ents for achievement subgroups—high, middle and low phonics subgroups and high, middle and low language ex-perience subgroups? Definition of Terms For the purpose of this study the following terms were defined: 5 language Experience Approach. The Language Experience Approach i s a process of teaching reading which utilizes the child's own oral language and concepts to develop an i n i t i a l sight vocabulary and begin-ning word recognition s k i l l s . • • • (Dorchester, 197U, P. 3.) This approach to reading begins with the children choosing the words they want to l e a r n — t h e i r \"key vocabulary\" as developed by Sylvia Ashton Warner (1963). The children proceed from words to sentences, from dictating to writing their own stories and from reading their own composi-tions to reading commercial materials. The child's f i r s t introduction to reading i s via the whole word and reading for meaning i s stressed from the beginning. Phonetic analysis and structural analysis are also a part of this program. Phonics Approach. This approach to reading refers to the code emphasis approach developed i n the Language Patterns Series published by Holt, Rinehart and Winston. The pupils learn the names and sounds of the letters of the alphabet. Once the children have learned a few of these sound-symbol relationships and can blend the sounds into words they pro-gress through a series of readers. The vocabulary i n the readers i s con-trolled and i n i t i a l l y only phonetically regular words using short vowel sounds and consonants are introduced. Once the process of blending has been mastered, sight words and irregular spellings are introduced gradually. In summary, the child's f i r s t introduction to reading i s v i a the sounds and blending the sounds together to make words--the emphasis i s on the synthesis of words rather than upon memorization of whole words. High. Middle and Low Subgroups. Subjects were divided into sub-groups on the basis of their number of substitution errors. The abbreviations 6 used i n this paper for each subgroup and the number of subjects within each subgroup i s l i s t e d below: ing of the child differs from that i n the written text. Oral reading err-ors were analyzed according to the following categories: - No Response or Don't Know - Insertions - Omissions - Substitutions - Sounding Out Reading Behaviour. This refers to the various ways i n which child-ren process information when reading. Aspects considered w i l l be the type of oral reading errors produced, the number of repetitions and corrections produced, the speed at which they process information and the understand-ing of the information processed. Substitution Components. Substitution errors were analyzed for: - words/nonwords -graphic similarity of words -graphic similarity of nonwords -sound similarity of words -sound similarity of nonwords -no meaning change -contextual appropriateness high phonics subgroup middle phonics subgroup low phonics subgroup 9 subjects 10 subjects 9 subjects high LE subgroup middle LE subgroup low LE subgroup 9 subjects 10 subjects 9 subjects Oral Reading Errors. An error occurs when the observed oral read-Subjects, (a) The subjects for this study were randomly selected from grade one public school classes located i n the geographic area of 7 Richmond, 6. C. (b) A l l non-readers, repeaters, transfers i n and child-ren who could read before entering grade one were eliminated. (b) A l l subjects with observable visual and auditory defects were eliminated. GENERAL PROCEDURES The general procedures were as follows: 1. The literature was surveyed to find existing information on the subject, to note the research design used i n similar studies and to determine i f there was a need for further investigation. 2. A study was made of reading achievement tests and diagnostic tests to determine the s u i t a b i l i t y of using such. 3. The schools were selected i n consultation with the Supervisor of Early Childhood Instruction i n the c i t y of Richmond, B. C. They were located i n similar socioeconomic areas. U. The exceptional children, i . e . , those with observable visual and auditory defects, those who had repeated grade one, transfers i n , pre-grade one readers and non-readers were eliminated. 5. Twenty-eight children from each treatment group ( i . e . , 28 taught by the language experience approach and 28 taught by the phonics approach) were selected at random. 6. Student data—age, sex, and birthdate—were collected. 7» The following instruments were administered and data collected during the last two weeks of May, 1978. (a) Wide Range Achievement Test (Jastak, 1965). 8 (b) Spache Diagnostic Reading Scales (1972). Tests 1A, IC, 2A, 3A, and 3C were used. 8. The following data were tabulatedt (a) reading grade level, (b) oral reading errors according to the above-mentioned categories, (c) comprehension on each selection, (d) self-corrections, (e) time required to read each selection, (f) repetitions. 9 . The treatment of the data was as follows: The results of each child's oral reading were tabulated and mean scores computed for oral reading errors, repetitions, self-correc-tions, fluency and comprehension. Mean percentages were computed for sub-stitution components. The t-test for independent samples was used to de-termine differences between groups and between subgroups. SUMMARY Chapter one has introduced the purpose of the study which was to analyze the oral reading behaviours of beginning readers taught by either a phonics approach or by a language experience approach. The need for this study and a brief description of the general procedures were also included. ORDER OF PRESENTATION The content and organization of the chapters are: 1. Chapter I presents the problem, the need for the study, limits of the study, and general procedures. 9 2, Chapter II provides a survey of the pertinent literature on oral reading behaviour and method of reading instruction. 3, Chapter III describes the method employed i n this study-selection of subjects, instrumentation, collection, classification and analysis of data. U . Chapter IV i s concerned with the presentation and interpreta-tion of the data. 5 . Chapter V i s concerned with the findings, conclusions and recom-mendations for educational practice and future research. Chapter II RELATED RESEARCH Daring the past there has been increased research on children's oral reading behaviour* The studies covered such diverse variables as: (1) language—bilingualism (Hodes, 1977; Williamson and Young, 1976j Folman, 1977), speakers of Spanish (Hood, et al., 1976), (2) functionally illiterate adults (Russell, 1973), and (3) reflection and impulsivity (Waltz, 1977; Butler, 197U; Hood, 1975). This chapter, however, will concern itself primarily with research relating to the effect of reading instruction on oral reading errors of beginning readers. The pertinent research will be discussed in the follow-ing order: (a) background, (b) whole word approach, (c) phonics approach, and (d) comparison of approaches. Background Although researchers have analyzed oral reading errors in numerous ways, i t is useful to divide research in this area into two main categor-ies (Weber, 1968). The first category consists of research in which oral reading errors were considered to be indicative of deficient reading skill and were viewed in a negative sense (for example, Monroe, 1928; Payne, 1930; Madden and Pratt, 19Ul). In the second category lies the research that has attempted to analyze oral reading behaviour to gain better understanding of the pro-cesses or strategies which result in the error (for example, Goodman, 1965; 10 11 Clay, 1967} Weber, 1970), The errors are analyzed to see which part of the erroneous response i s correct. It i s this type of research that w i l l be the concern of this chapter. One of the earliest investigations concerned with the quality of the oral reading errors and the processes which contributed to those errors was conducted by Bennett (19^2). In an analysis of over 3U,000 errors made by retarded readers she found that letters or word parts were the two important areas; that context played a major role i n word recognition and that errors were usually the same part of speech as the written word. (The l a t t e r finding was later confirmed by T. Goodman (1967), who noted Ul percent of the errors were closely associated i n meaning with the writ-ten text.) MacKinnon (19\">9) analyzed the oral reading errors of grade one children who were being taught to read by a programme which emphasized sentences and had a controlled vocabulary. He discovered that the f i r s t graders attempted to read sentences as grammatical wholes rather than res-ponding to word-by-word stimuli. Their errors were not haphazard. Often the second error would be brought about by the grammatical constraints of the f i r s t error. MacKinnon also found that the children's errors developed from contextually constrained, to non-response, and f i n a l l y to graphically con-strained (as did K. Goodman, 1967j Y. Goodman, 1967} Weber, 1970} Clay, 1967). He noted also that once the children started using graphic cues, their reading became more \"word by word.\" Further evidence of the importance of contextual and syntactic cues i s provided i n Goodman's (1965) descriptive study i n which he studied the 12 errors of 1GG children i n grade one, two and three. He found that young readers recognized with greater accuracy words i n context than words pre-sented i n l i s t s * Therefore, he concluded that the syntactic and semantic constraints of language were used by the children when reading. Goodman noted that i n grade one half the errors were omissions, i n grade two the children tried to figure out the word and made more substitu-tions, and by grade three the children \"showed a pronounced increase i n the percent of substitutions among their l i s t errors.\" (p. 6Ul.) Another finding of the study was that \"virtually every regression • • • was for the purpose of correcting previous reading.\" (p. 6U2.) However, when reading the words on the l i s t , the children seldom regressed. Y. Goodman (1967) analyzed the miscues of three \"slow\" readers and three \"average\" readers as they progressed through grade one. She found that the slower readers make more Miscues Per Hundred Words (MPHW) than did the average readers, but no relationship was found between number of miscues and comprehension scores. This study was continued for four years (Goodman, 1971). In 1971 Goodman discovered that as the child's reading s k i l l increased so did his a b i l i t y to use grammatical constraints. The average readers seemed able to use a l l cues,—graphic, phonic, semantic and syntactic, while the slower readers produced primarily graphic respon-ses. This l a t t e r finding i s i n agreement with Biemiller (1970),. Clay (1968) and Weber (1970). Goodman identified three stages which are similar to Biemiller (1970) and MacKinnon (1959). In the f i r s t stage unknown words are omitted. In the second stage the children begin to sound out the words, and nonwords may result. The errors tend to have close graphic proximity to the stimu-lus word. In the third stage the children employ a variety of cues. 13 Goodman and Burke (1969) studied the oral reading errors made by proficient readers i n grade two, four and six using the Goodman Taxonomy of Reading MLscues. The errors were divided into two groups—non-trans-formation miscues (those which did not alter syntactic structure) and re-transformation miscues (those which did alter syntactic structure). The authors found that the errors often served the same grammatical function as the stimulus word. This tendency existed even when the children changed the text to a more familiar language pattern. This tendency to retain the same grammatical function i n retransformation errors increased from grade two to six. This was seen as an indication that the readers were develop-ing increasing control of the English language. The above-mentioned studies i l l u s t r a t e several important points about recent trends i n this area of research. Emphasis of the research has moved from solely an analysis of word errors to an attempt to under-stand the strategies the children bring to the reading process. Whole Word Approach A selection of recent studies discusses the effect of basal readers on beginners' oral reading behaviour. Clay's three related articles (1967, 1968, 1969) examined the oral reading errors of beginning readers for one year. The children were being instructed i n reading by a method which \"stressed fluency, meaning, and •learning as one reads'.\" (p. 12). Minimal attention was given to the teaching of sounds or the development of a sight vocabulary. At the end of the f i r s t year the children were divided into quar-t i l e groups on the basis of a word recognition test. A significant d i f f e r -ence between subgroups on number of errors was noted. The median child i n lU each group—high, high middle, low middle, and low, respectively made one error i n 37.29, 15.20, 7.86, and 2.58 words. I t was also noted that 72 percent of a l l substitution errors were \" l i n g u i s t i c a l l y equivalent\" (p. 22) to the written text. A grapheme/phoneme correspondence existed for Ul per-cent of the errors. The high group's substitutions were more graphically similar to the stimulus words than were those of the low group. (This i s similar to Bennett's 19U2 study.) The high group corrected more errors than the low group. Clay concluded that the high group's low error, high self-correction rates were the result of efficient processing of cues. Clay further analyzed the data i n two a r t i c l e s . In one a r t i c l e (1969) Clay noted that grammatical competency was the main reason for self-correction behaviour. In the other a r t i c l e Clay (1968) noted that guesses at uncertain words seemed to be the result of the syntactic aspects of the sentence rather than by the phoneme/grapheme relationship i n words. Biemiller (1970) examined the oral reading errors of k2 grade one children from October to May. The children were using a basal reader and were observed on the average of 23 times. Biemiller's purpose was to examine changes i n the use of context-ual and graphic information for word identification. He analyzed the err-ors i n terms of semantic and graphic constraints and non-response errors and thus developed three main phases of reading acquisition. In the f i r s t phase, Pre Non-Response, the children made predominant use of contextual information for anticipating or guessing unknown words. Their substitutions were appropriate to the sentence context but not app-ropriate graphically. Ninety-nine percent of the substitution and inser-tion responses came from sight words previously learned (as did those i n 15 Bennett's study, 19k2). Biemiller suggested that this heavy reliance on context was the result of the children avoiding the use of graphic infor-mation. In the second phase, Non-Response, there was a predominance of non-response errors (50 percent or more of a l l errors). The number of context-ually constrained errors decreased but there was a significant increase in graphically constrained errors. The children did not use both contextual and graphic information but tended to rely on one or the other. Ninety-four percent of the substitutions came from words previously learned. Biemiller interpreted this phase as the child becoming interested in and paying close attention to graphic information and realizing that \"one specific word is associated with each graphic pattern.\" (p. 93.) In the third phase, Post Non-Response, there was a drop in non-response errors to below 50 percent of al l errors. The children made sig-nificantly more substitution errors that were both contextually and graph-ically acceptable—82 percent. The children now seemed able to use both graphic and contextual information and there was an increase in speed of word recognition. Ninety-one percent of the substitutions came from words previously learned, Biemiller (in agreement with Weber, 1970) believed that poorer readers moved slowly from stage to stage because of difficulty in handling graphic information. Throughout the year, only the better readers prog-ressed through the three phases. He noted that the slower readers seemed unable to develop reading strategies. It seemed as i f they \"started off on the wrong track.\" (p. 95.) Weber (1970) studied the oral reading errors of a class of 21 grade one children (10 boys and 11 girls) in order to determine the 16 strategies used by beginning readers to identify words* The children had a mean age of 6.3 upon school entry and were taught to read from a basal reading series. The class was divided into high achievers and low achiev-ers. The errors recorded were reversals, insertions, omissions and sub-stitutions. They were analyzed for letter-sound correspondence, grammat-i c a l acceptability, semantic appropriateness and grammatical function. Eighty percent of the to t a l errors were substitutions while the remaining 20 percent were divided, almost equally, between omissions and insertions. About 9$ percent of the errors were words the children had encountered previously. (Bennett (19U2) observed similar behaviour.) Weber found that the substitutions of the better readers were more graphically similar to the text. For the total group, about two-thirds of the errors were grammatically acceptable to the whole sentence. Of those errors judged f o r semantic appropriateness, a l l of which were grammatic-a l l y acceptable, 92.8 percent were found to be \"consistent with the mean-ing of the rest of the sentence,\" (p. UU9) and that two-thirds conformed to the preceding context. In fact, Weber found almost complete overlap between semantic and syntactic appropriateness. (This finding i s supported i n the study of Y. Goodman, 1967.) Weber concluded that both high and low readers used semantic and grammatical constraints equally well i n reading. She f e l t that perhaps there exists an inverse relationship between the beginners' use of graphic cues and syntactic cues. She states that, \"Learning the optimal balance i n the use of graphic information and of structural constraints may i n fact be one of the main tasks for the novice reader.\" (p. kkl») It should be noted that these two groups read different stories 17 with different quantities of errors analyzed. Phonics Approach Only one recent study to date discusses the effect of the phonics approach on beginning reading strategies. Cohen (197b-75) studied the oral reading errors of 50 grade one children (2k boys and 26 girls) i n two heterogeneously grouped classes. The children were being instructed by a phonics approach. The study spanned the last eight months of grade one and i t focused on changes in' word recognition strategies when oral reading errors were analyzed accord-ing to type of error—word substitution, no response, sound out and self-correction. Also examined were graphic similarity of the error to the written word and grammatical acceptability within the sentence. The errors of good and poor readers were analyzed and compared as well as the total group's errors. The three most common errors noted were word substitutions, no response and nonsense. During the f i r s t four months no response errors occurred most frequently for a l l groups. After Janu-ary, however, the good readers' nonsense errors began to decline and by the end of the study were very low. On the other hand, the nonsense errors of the poor readers slowly increased throughout the eight months. In the self-correction category, a l l groups increased, although the increase was only \"slight\" for poor readers and \"substantial\" for good readers. The substitutions of a l l the groups showed a steady improvement i n graphic similarity to the stimulus word and by the la t t e r half of the study the non-systematic errors had nearly disappeared. The poorer readers continued to make substitutions with similar f i r s t and/or l a s t letters but for the better readers the substitutions were on small function words. 18 The grammatically acceptable substitutions increased for a l l groups and by the end of the study they exceeded substitutions which were not grammatic-a l l y correct. Cohen concludes that the good readers' large number of nonsense errors early i n the study appeared to be related to their a b i l i t y to learn to use phonics before they used semantic cues. The strategies for poor readers were less systematic. Comparison of Approaches In this section several studies, each dealing with two methods of instruction, are summarized. DeLawter (197U) examined the error patterns of 169 grade two child-ren from a low income area. A l l the children had participated i n a Begin-ning Reading Project for two and one-half years and had received reading instruction i n one of two reading systems. The decoding group received reading instruction i n the Miami Linguistic Readers and the Merrill Ling-ui s t i c Readers—a programme with a phonic emphasis, and the meaning oriented group received instruction with the Chandler Language Experience Readers which emphasized the content of the stories. The purpose of the study was to find out i f different patterns of errors resulted from different reading approaches. She found that most of the substitutions made by the children i n the decoding group were non-words—about twice as many nonwords as words—which closely resembled the words i n graphophonemic similarity to the stimulus word. Sixty-four per-cent were considered poor attempts at decoding. The children tended to respond hastily. DeLawter then studied the errors for syntactic and semantic 19 acceptability. Almost a l l the words were syntactically acceptable and half were semantically acceptable for both groups. DeLawter concluded that there appears to be a relationship between error pattern and beginning reading instruction, and that the error patt-erns \"demonstrate strategies that are predictable, given particular instruc-tional emphases.\" (p. U 8 . ) She also mentioned that after two years of in-struction these patterns are s t i l l evident and this \"reinforces the find-ing of the study.\" (p. U 8 . ) Elder (1971) analyzed the oral reading errors of h9 Scottish child-ren and 98 American children whose mean age was 91 months. The Scottish children were taught by a phonics method while the American children had been instructed by a sight word approach. The child-ren were analyzed on their performance on the Gray's Standardized Oral Reading Paragraphs. The Scottish children displayed fewer word recognition errors, in-cluding fewer word substitutions but significantly more nonword substitu-tions than the American children. The Scottish children were found to be \"highly accurate but comparatively slower readers,\" while the American children were \"fluent but relatively less accurate readers.\" (p. 220.) The Scottish children had a significantly higher percent of omiss-ions and repetitions than the American children. An analysis of the word substitutions revealed that the Scottish children made significantly more substitutions that changed the meaning, while the American children seemed more attentive to meaning. The children did not differ in self-corrections. Elder's conclusions were as follows: 1. Children who begin reading at age five rather than 20 age six can, on the average, be expected to gain at • least an initial advantage in ability to read orally. 2* A stress on phonics in beginning reading tends to promote accurate word recognition, but slows the process of recognition. 3* A reliance on sight recognition tends to de-crease accuracy, but increases rate of recognition. U* The oral reading of children who rely on sight recognition tends to have a high frequency of word substitutions and initially a need for much assistance. 5* The oral reading of children who rely on phonics tends to have a high frequency of mispronuncia-tions* 6* A reliance on phonics tends to divert children from the meaning of what they read* (p. 228.) Barr (1972) analyzed the word recognition errors made by pre-read-ing grade one children—18 urban subjects (9 boys and 9 girls), and 2h suburban subjects (12 boys and 12 girls). The mean age was 6 years 5 months. The children were taught by two reading methods, a sight word or a phonics. As well as examining the error patterns resulting from the two instructional conditions, Barr was also interested in comparing these error patterns with those found by Biemiller (1970) to see i f the changes were related to instructional method* Barr found that children instructed by the phonics approach made substitution errors that came from words other than those taught* A high proportion of the substitutions were nonwords and the substitutions were significantly more graphically constrained* They also made a significantly higher percent of non-response errors than the sight group* Children in-structed by a sight word approach made more substitutions that were the same as the words being taught at the same time. Rarely were the substitu-tions words the children had not learned, or words taught earlier, or non-words* Also, these children seldom made substitutions which reflected graphic cues* Barr noted that the error patterns for sight word learning resembled 21 the f i r s t phase of Biemiller*s study, while the error patterns for phonics learning resembled Biemiller's second phase. Barr concluded that these findings suggest that \"different instructional methods influence d i f f e r -entially the pattern of word recognition errors. Different instruction entails different strategies for word recognition.\" (p. 527.) It should be noted that the words were presented i n isolation and not i n context so this limits the strategies a child can use. Barr (197U-75), i n reviewing the literature, determined that find-ings to this point i n time strongly suggested that instructional method does influence the strategies of children i n translating \"the printed word to language.\" (p. 569.) In a l l of this research two questions had not been addressed according to Barr; namely, ( l ) to what extent can results obtained for groups be confidently applied to individuals, and (2) how do strategies alter over time. Consequently Barr undertook to study the oral reading responses of 32 grade one children i n December and May to determine word recognition strategies. She also studied individual children's strategies i n order to see i f they were determined by the class instructional method. Half the subjects received instruction by a phonics approach and the other half with an eclectic basal approach. Barr discovered that children who learn by a phonics approach pro-duced substitutions which were not cognizant of the constraints of the printed word as a representation of natural language and often their err-ors were nonsensical. However, children instructed by a sight word method provided responses which f i t t e d contextually although their substitutions were limited by their vocabulary. 22 Barr concluded that: It appears to be possible to determine the strategies that beginning readers use for translating print to speech [andj. . • the response patterns for groups of pupils instructed by particular methods are representa-tive of most members within the group rather than a function of the distinctive patterns of a few. (p. $77.) Since the subjects read l i s t s of words no analysis could be made with regard to semantic, syntactic or correction strategies. M.S. Burke (1973) examined the oral reading errors of 3 grade one children taught by a synthetic method (emphasized phoneme-grapheme corres-pondence) and 3 grade one children instructed by an analytic method (basal reader sight words). Burke noted that the children learning to read by a synthetic method made more errors and many of the errors stressed the phoneme-grapheme relationship, while few,were syntactically and semantic-a l l y acceptable. These children had lower comprehension ratings than the children taught with the analytic,approach. The children taught by the analytic approach made slightly fewer errors that displayed the phoneme-grapheme relationship. These children, however, tended to produce more varied patterns of errors and used seman-t i c , syntactic, and phoneme-grapheme constraints although the semantic system was s t i l l inadequate. Burke f e l t that the children taught by the analytic approach did not often resort to word-by-word processing because they seem to have an understanding of the interaction of the various cue-ing systems. The subjects of this study (like those of Y. Goodman, 1967) showed an inverse relationship between phoneme-grapheme correspondence and grammatical and semantic acceptability. Burke concluded that the method of reading instruction can affect reading behaviour. 23 Norton (1976) i n v e s t i g a t e d the e f f e c t of two reading approaches on the reading s t r a t e g i e s o f hO grade one and grade t h r e e p u p i l s o f hig h and low a b i l i t y . They found t h a t those c h i l d r e n i n s t r u c t e d by a s y n t h e t i c phonics emphasis developed higher phonic, graphic and s y n t a c t i c s t r a t e g i e s , more nonwords, but produced comparatively few s e l f - c o r r e c t i o n s . On the other hand, those c h i l d r e n taught by an a n a l y t i c - e c l e c t i c approach produced more miscues s e m a n t i c a l l y a c c e p t a b le, more miscues t h a t d i d not change meaning, and s i g n i f i c a n t l y more s e l f - c o r r e c t i o n s than the phonics emphasis group. No s i g n i f i c a n t d i f f e r e n c e s were found f o r grammatical f u n c t i o n o f miscues or f o r t h e i r s y n t a c t i c a c c e p t a b i l i t y . The comparative d i f f e r e n c e s h e l d a t both grade l e v e l s . The authors concluded t h a t the method of reading i n s t r u c t i o n d evel-oped i n the f i r s t grade seems t o r e s u l t i n l o n g range o r a l r e a d i n g behav-i o u r s as evidenced by the e r r o r s o f the grade three c h i l d r e n . However, the grade one and grade three c h i l d r e n were d i f f e r e n t samples, Norton, and Hubert (1977) examined the o r a l r e a d i n g e r r o r s of 60 grade one c h i l d r e n i n two comparable d i s t r i c t s i n Texas. W i t h i n each o f the two i n s t r u c t i o n a l g r o u p s — e c l e c t i c b a s a l and phonics emphasis—were three a b i l i t y g r o u p s — H i g h , Average, and Low. The purpose of the study was t o determine how c h i l d r e n taught by these two d i f f e r e n t approaches com-pared i n o r a l r e a d i n g s t r a t e g i e s . Norton found t h a t c h i l d r e n i n s t r u c t e d by the phonics approach pro-duced s i g n i f i c a n t l y more miscues w i t h a high graphic o r phonic p r o x i m i t y and more miscues t h a t were nonwords. The c h i l d r e n achieved higher word r e c o g n i t i o n grade scores than comprehension grade scores and s i g n i f i c a n t l y h i g h e r i n s t r u c t i o n a l word r e c o g n i t i o n l e v e l s . The c h i l d r e n i n s t r u c t e d by 2U the eclectic basal approach produced significantly more miscues that were syntactically acceptable, that were semantically acceptable, that caused no change i n meaning, and that were self-corrected. The children achieved higher comprehension grade scores than word recognition grade scores and significantly higher instructional comprehension grade levels. There was no significant difference for grammatical function. Norton concluded that the method of instruction does produce d i f f -erent oral reading strategies and that a l l a b i l i t y groups within each app-roach \"demonstrated very similar oral reading profiles.\" (p. 23») E. Burke (1976-77) studied the decoding strategies used by 216 seven, eight and nine year old children i n terms of graphic, syntactic, and semantic cues. Age, sex, school-type, and school emphasis on reading were the variables considered. Burke's study revealed that the quality of children's miscues im-proved with age, although not uniformly. In the graphic category the qual-i t y of miscues only slightly increased with age whereas i n the semantic category the increase was continuous. Por the syntactic category a large increase occurred at the 8 year level followed by a slight decrease at the age of 9# These differences i n pattern for these three categories were significant at the one percent l e v e l . No other effect was found to be significant although differences were evident. Burke stated that \"no firm conclusions as to the relative merits of the different approaches to the teaching of reading can be inferred.\" (p. Ul.) Ewoldt (1976) studied the oral reading errors of 73 grade three children i n the seventh month of grade three. The subjects were i n two groups—Follow Through and Non Follow Through. The Follow Through group 25 had participated in a programme since grade one that emphasized the lang-uage experience approach, while the Non Follow Through group had not. How-ever, during grade three only 8 of the Follow Through Group used any lang-uage experience. Basal readers were used by all 37 Non Follow Through sub-jects. The main purpose of the study was to \"identify differences between the two groups of readers which may be the result of the types of instruc-tion received... • .\" (p. 3.) Twenty-five miscues were analyzed for each child. She found that both groups of children employed effective use of strategies. The mean scores were similar for syntactic and semantic accept-ability and meaning change, graphic similarity, sound similarity and gram-matical function. She felt that the Non Follow Through readers appeared to be less efficient readers as evidenced by their correction strategies —they \"wasted too much time on unsuccessful attempts at corrections.\" (pp. 90-91.) Ewoldt concluded that: \"Differences in favor of the Follow Through readers may be attributable more to differences in the program in prior years than to the third grade program.\" (p. 92.) Dank (1976) analyzed and compared the oral reading errors and comprehension of 20 selected grade two children. The children received instruction with the McGraw Hill Programmed Reader which emphasized the grapheme-phoneme approach or with the Ginn 360. a language experience series. Dank found that the children taught by the approach that emphasized the letter sound relationships produced fewer omissions, more nonwords and more miscues with high graphic and sound proximity. Those children taught 26 by the language experience approach of the Ginn 360 programme produced more aemantically acceptable errors. Their understanding of what they had read, reflected i n their r e t e l l i n g of the story, was superior to the Programmed Reading Group. She concluded that the children i n both groups made errors that reflected the reading instruction they had received. SUMMARY The present chapter has summarized some of the research pertaining to oral reading behaviour of beginning readers. These studies seem to suggest that differences i n reading behaviour may be influenced by the method of reading instruction. Children taught by a whole word approach tended to use strategies related to the word and they seldom made nonword responses. Many of their substitutions were from words taught at that par-ticu l a r time. These children made satisfactory use of syntactic constraints and their substitutions often were related to the meaning of the sentence. Children taught by a phonics approach, however, produced substitu-tions that often distorted the meaning of the sentence. They produced many nonwords and their substitutions were usually graphically similar to the stimulus word. These children also made satisfactory use of syntactic constraints. These studies seemed to indicate that children who experience l i t t l e trouble learning to read quickly learn to use a l l cue systems, while the children having trouble learning to read tend to rely heavily on one cue system and neglect the others. To date most of the research i n this area has been conducted with children being taught to read by a phonics or a basal reader sight 27 vocabulary approach. There appears to be no research evaluating the effect of a teacher developed language experience programme using the child's natural language and self-selected sight vocabulary on the oral reading behaviour of beginning readers. Chapter III DEVELOPMENT AND ADMINISTRATION OF THE INSTRUMENTS The purpose of this chapter is to describe (l) the selection of subjects, (2) the selection of subgroups, (3) the reading instruction pro-vided, (k) the instruments used, (5) the collection of data, (6) the classification of data, (7) scoring of the data, and (8) the analysis of data. Selection of Subjects The subjects for the study were £6 grade one children attending four public schools in Richmond, B. C. Each school contained the same socio-economic levels, ranging from lower to upper middle class. Two of these schools taught beginning reading by a phonics approach; two taught beginning reading by a language experience approach. The individual children within the schools were randomly selected. Twenty-eight children (13 boys and 15 girls) were selected from the total population of grade one children using a phonics approach in the two schools, and 28 children (12 boys and 16 girls) were selected from the total population of grade one children using a language experience approach in the other two schools. The following children were excluded from the population before the random selection occurred—children with obvious auditory and visual defects, children who had transferred in during the school year, children who could read before they entered grade one, and 28 29 children who were non-readers. At the time of testing, chronological age of the children receiving instruction through phonics ranged from 6.5 years to 7.U years, with a mean age of 6.513. Chronological age of the children receiving instruction through language experience ranged from 6.5 years to 7.U years, with a mean age of 6.618. Selection of Subgroups Substitution errors were used as the criterion for the selection of achievement groups. Three achievement groups—high, middle and low-were separated out of the language experience treatment group, and three achievement groups—high middle and low—were separated out of the phonics treatment group, Reading Instruction Provided The 28 children being instructed by the phonics approach received heavy emphasis on decoding but minimal sight word development. Some supple-mentary reading took place i n books other than the phonetically controlled readers. The 28 children being instructed by the language experience app-roach began reading with sight words they had chosen. They were exposed to many commercial books as well as numerous and varied reading series. Instruction i n phonics and i n structural analysis formed a part of the read-ing programme. For convenience, throughout the rest of this paper the two groups w i l l be referred to according to the i n i t i a l teaching emphasis—phonics and language experience. 30 Instruments Used 1* Spache Diagnostic Reading Scales Six selections were chosen from this measure: 1A, IC, 2A, 2C, 3A and 3C The readability of the selections was grade 1.6, 1.8, 2.3, 2.8, 3.3 and 3.8 respectively. The two selections at the grade three level were necessary to provide difficult enough material so a l l subjects would produce at least 5 errors. The use of this instrument provided content validity. 2, The Wide Range Achievement Test (WRAT) The reading section was administered. It is a pronunciation test of words presented in isolation. This test provided criterion related validity. The correlation between the Wide Range Achievement Test and other achievement tests was as follows: WRAT reading (19U6) vs. New Stanford Paragraph Reading +.81 WRAT reading (19U6) vs. New Stanford Word Reading +.8U (Jastak, 1965, p» 15.) Collection of Data The information was collected during the last two weeks of May, 1978, The procedure was the same for each subject. First each subject read the words from the Wide Range Achievement Test. Then they read the six selections from the Spache Diagnostic Reading Scales and answered the comprehension questions. The children were tested individually in a separ-ate room. Prior to the testing the examiner told the child: \"Try to read this aloud to me as well as you can. I'm not going to t e l l you any words, because I want to see how well you read i t by yourself. If you meet some words you don't know, just try them and go on. I ' l l ask you some questions 31 about the story when you f i n i s h . \" (Spache, 1972, p. 16.) I f the child hesitated for more than 10 seconds, he was told to continue reading. The examiner recorded the errors on a transcript of each selection. The examiner recorded deviations from the written text i n pencil above the typed word of her copy of the selections. Following the reading of each selection, the child responded to the questions asked by the examiner. A l l performances were recorded onto tape for later analysis and verification of errors, comprehension, repetition and self-correction. The reading of each selection was timed using a stop watch. Classification and Coding of Data The data for this study was classified according to the following categories: types of errors, repetitions, self-corrections, comprehension, fluency, and substitution components. 1. Types of errors a) No response or don't know. The child hesitates before a word and i s unable to read i t or else says, \"I don't know,\" before proceeding with reading. This was indicated by a NR or DK above the text and crossing out the written word i n the text. DK Example: At n ^ i t she i s very t i r e d . b) Insertions. The child adds a word. This was indicated by a caret ( /\\) and the recording of the inserted word. Example: <* u i c k ly Mary saw the car and ran/\\the rest of the way. 3 2 c) Omissions. The child leaves out a word i n a sentence. This was indicated by drawing a line through the omitted word. Example: Bob stopped to watch-%tee-other animals. d) Substitutions. The child says a different word from the one i n the sentence. This includes nonwords as well as words. This was indicated by crossing out the word i n the text and recording the substitution above the original word. Example: word He pu)tffed the dog up the h i l l , ared nonword Mary was af^srfd but she was glad she wasn't hurt. e) Sounding out. The child unsuccessfully attempts to sound out the word. This was indicated by recording the letters sounded, followed by dashes. ' \" c a a — Example: The keeper didn't enter the cage. 2 . Repetitions. The child repeats a word or words. This was indicated by underlining the repetitions with a wavy l i n e . Each group of words repeated together counted as one repetition. Example: The keeper wasjfeeding the wolf from a p a i l of food. 3. Self-corrections. The child corrects the error without ass-istance. This was indicated by recording the f i r s t response and then drawing a c i r c l e around the correction. Example: ^J}®\" (On)the way out of the park Bob stopped to watch the other animals. 33 h. Comprehension. To check comprehension each child answered orally the comprehension questions following each selection on the Spache Diagnostic Scales. Each student began reading at the grade 1.6 level and read through to the grade 3»8 lev e l . Each question was worth one-half to one mark determined by the directions i n the manual. 5# Fluency. To check fluency each selection the child read was timed. This was recorded i n seconds. The child thus received six scores, one for each selection, which were then totalled. 6. Substitution Components. Each child's substitution responses were further analyzed according to the following categories: a) Words/Nonwords. In this category i t was determined whether or not the substitution error was a word or a nonword. Nonwords are nonsense words composed of a series of sounds. Some examples are: text: cages re sponse: eagers (nonword) text: feeding response: fenting (nonword) text: strong response: storing (word) b) Graphic similarity for words. In this category i t was determined whether the f i r s t l e t t e r of the substituted word was the same as the text. This was the same proced-ure Biemiller (1970) used when he noted \"whether the f i r s t l etter of the response matched the f i r s t l e t t e r of the stimulus word.\" (p. 80) An example i s : text: sound response: song 3U Graphic similarity for nonwords. In this category i t was determined whether the f i r s t l e t t e r of the nonsense word was the same as the text. An example i s : text: skunk response: skunt Sound similarity for words. In this category i t was noted whether the substituted words were similar i n sound to the text. The sounds of the words were considered and not necessarily the le t t e r s . For a word to be considered similar i n sounds, two sounds i n the word error had to be the same as i n the text word. Also the sounds had to be i n the same position i n both words. Digraphs (sh, th, wh, ch) and consonant blends (such as gr, s l , c l ) were consid-ered as one sound. Some examples are: text: walked response: worked text: greet response: green text: roar response: road Sound similarity for nonwords. In this category i t was noted whether the nonsense words were similar i n sound to the text word, using the same c r i t e r i a noted i n (d) above. Some examples as: text: week response: wenk text: cages response: cags 35 Contextual appropriateness. This category analyzed the child's a b i l i t y to use syntactic and semantic constraints. A sentence can be grammatically and semantically correct but not have the same meaning as the written text. In this category, as did Biemiller, the substitutions were consid-ered contextually appropriate i f they were \"grammatically and semantically acceptable up to and including the error.\" (p. 82.) Later errors were judged using the previous err-ors i n the context of the sentence. Some examples are: text: At night she i s very t i r e d , response: At night she was very t i r e d . text: One day Bob took. • . response: One day Bob looked. . . The following example i s not contextually appropriate up to and including the error: text: Then they rode down the h i l l , response: Then they rude down the h i l l . No meaning change. This category analyzed the error i n terms of whether or not i t altered the meaning of the text. Some examples of errors that resulted i n l i t t l e or no change are: text: Then she slowly comes home, response: Then she slowly came home. text: But the dog did not li k e to ride down, response: But the dog did not like the ride down. text: He was a l i t t l e frightened, response: He was a l i t t l e afraid. The following example resulted in-meaning loss: 36 text: He pulls i t slowly up the h i l l , response: He pulls i t softly up the h i l l . Scoring of the Data After testing the examiner listened to the tape recordings twice and checked her written record of the oral reading behaviour of each child on each selection. Then the data of ten randomly chosen subjects were analyzed by another scorer. The percent of agreement between the two scorers ranged from 96.05 to 100 percent. This was higher than the res-ults reported by Weber (1970) who found agreement of over 90 percent. Having two judges double score 10 percent of the papers was found to pro-vide adequate interjudge r e l i a b i l i t y . (Weber, 1970; Norton, 1976j Hood, 1975-76). Data Analysis In examining the effect of the alternative instructional treat-ments ( i . e . , phonic and language experience) on beginners* oral reading performance the data was examined or treated as follows. After the reading behaviour was recorded and coded for each child, a computer card, one for each child,was key punched. The data was then run through as SPSS (Statistical Package for the Social Sciences), per-formed on IBM 360/67 computer using the applicable sub routines of SPSS. The probability level of less than or equal to .05 was accepted as being indicative of a significant difference and w i l l be reported i n Chapter four for substantive discussion and interpretation. S t a t i s t i c a l significance w i l l be reported using a 2 tailed test of significance. The analysis was related to the two problems. 37 1. Group difference and oral reading behaviour* A frequency distribution was performed and mean scores were calculated for types of oral reading errors, repetitions, self-corrections, fluency and compre-hension* The s t a t i s t i c a l procedures employed i n examining the question of treatment group differences on reading behaviour of subjects was the t-test. The particular t-test used was the t-test of significant difference for independent samples (Glass and Stanley, 1970)* 2. Effect of treatment on substitution components* a) A frequency distribution was performed and mean percentages were calculated for words/nonwords, graphic similarity of words, graphic similarity of nonwords, sound similarity of words, sound similarity of nonwords, no meaning change and contextual appropriateness* To determine the effects of instruction on substitution components a t-test of significant difference for independ-ent samples was employed. b) Similarly, the effects of treatment on substitution errors between achievement subgroups were examined using the t-test f o r independent samples. SUMMARY This present chapter has presented information pertaining to selec-tion of subjects; teaching methods; instrumentation; and the collection, classification, coding, scoring and analysis of data. Chapter IV PRESENTATION AND INTERPRETATION OF DATA This chapter presents the results of the analysis of the data and interpretation of these results. The presentation will be in two sec-tions. The first section will present the data on the effect of two meth-ods of reading instruction on these types of reading behaviour—oral read-ing errors, self-corrections, repetitions, fluency and comprehension. The second section will present the data on the effects of the two methods of instruction on some aspects of substitution errors (words/nonwords, graphic similarity of words, graphic similarity of nonwords, sound similarity of words, sound similarity of nonwords, no meaning change, and contextual appropriateness) on the two treatment groups and between substitution achievement subgroups. Part It Oral Reading Behaviour 1, Types of oral reading errors. The total number of errors made by the children instructed through the language experience approach and the phonics approach were tabulated. The children instructed through the language experience app-roach made 1007 errors, which was one error per 13,U6 running words. The children instructed through the phonics approach made 1607 errors, which was one error per 8,1*3 running words. This finding supports earlier re-search by Burke (1973) who found that the phoneme-grapheme group made more errors than a basal reader sight vocabulary group. 38 39 The oral reading errors were categorized according to type, and the numbers for each category were tabulated for the two instructional groups. The mean score and the standard deviation were calculated for each group. The t-test for independent samples was applied to determine the differences between groups. The data are presented in Table 1. a) No Response/Don't Know. Por the children instructed through the phonics approach, the average number of errors in this cate-gory was 1*2$, or 2.18 percent of the total errors. Por the group instruc-ted through a language experience approach, the average number of errors was 1.39, or 3.87 percent of the total errors. There was no significant difference between the two groups. The findings do not support the find-ings of other studies. Barr (1972) found that children instructed by a phonics method made a significantly greater number of non-response errors than children instructed by a sight recognition method. Cohen (197U-75) found that non-response errors made by children taught by a phonics method were 29 percent of the total errors, far exceeding the percentage found in this study. b) Insertions. The children instructed through the phonics approach had an average of 1.82 insertions, or 3.17 percent of the total errors. The children instructed through the language approach had an av-erage of 1.96 insertions, or 5.U6 percent of the total errors. There was no significant difference between the two groups. c) Omissions. For the children Instructed through the phonics approach the average number of errors in this category was l» 6 l , or 2.80 percent of the total errors. For the group instructed through a language experience approach the average number of errors in this category was 1.11 Table 1 Means, Standard Deviation and t-value for Reading Behaviours for Phonics and Language Experience Beginning Readers Phonics Language Experience Reading Behaviour Category Mean Standard Deviation Mean Standard Deviation t-value A. Type of Error 1* No Response/ Don»t Know 1.25 2.012 1.39 3.370 - 0.19 2. Insertions 1,82 1.611 1.96 2.099 - 0.29 3. Omissions 1.61 3.083 1.11 1.3U3 .79 k. Substitutions 50.25 37.920 31.25 3U.171 1.97 * 5. Sounding Out 2.U6 3.707 .25 .701 3.11 * * B. Repetition 3.U6 2.603 iuOO 2.815 - 0.7U C. Self-Corrections 10.00 7.727 7.U3 5.928 l.Uo D. Fluency 113.60 50.39U 91.99 62.339 1.U3 E. Comprehension 29.6U 5.86U 32.U3 5.371 - 1.85 ^Significant at the .05 l e v e l . L H . Significant at the .01 l e v e l . l a or 3»08 percent of the total errors. There was no significant difference between the two groups. Two related studies, Dank (1976) and Elder (1971) report conflicting results. Dank found that children taught by a phonics method made fewer omissions than the language experience Oinn 360 group, while Elder (1971) found that the phonics group produced significantly more omissions than the basal reader sight vocabulary group. d) Substitutions. The children instructed through the phonics approach made a total of llt07 substitutions, with an average of 50.25 substitutions, while the children instructed through the language experi-ence approach made a total of 875 substitutions, with an average of 31.25 substitutions. The children instructed by the phonics approach produced a significantly greater number of substitutions than did children instruc-ted by the language experience approach. The percentage of total errors that were substitutions i s similar, with the children instructed through the phonics approach producing 87.55 percent and the children instructed through the language experience app-roach producing 86.89 percent. The findings of this study are higher than those noted by Barr (1972) who found that the phonics group produced 71.3k percent substitutions and the sight word group produced 76.21 percent sub-stitutions, and higher than the 79.9 percent that Weber (1970) found with her grade one children using a sight word approach. The present findings also d i f f e r from those of Elder (1971) who noted that the phonics group produced 38 percent substitutions while the sight word group produced U8 percent substitutions. e) Sounding Out. The children instructed through the phonics approach had an average of 2.U6 sounding out errors, or U.29 percent of U 2 the total errors. The children instructed through the language experience approach had an average of . 2 5 sounding out errors, or .70 percent of the total errors. The difference was significant. These findings support those of Elder (1971) who found that the children taught by a phonics method produced a significantly higher percentage of errors i n this cate-gory (2556) than the children taught by a sight recognition method (8$) , although the proportions for both groups i n his study were greater than those i n the present study. 2. Repetitions. Repetitions were not considered errors for as Clay (1967) com-mented, a \"repetition may be a form of hesitation—a f i l l e d pause—or an act of confirmation rather than an error.\" (pp. 101-102.) The children instructed through the phonics approach had a mean number of 3.U6 repetitions, whereas the children instructed through the language experience approach had a mean of luOO repetitions. There was no significant difference between the groups. These findings do not concur with Elder (1971) who found that the group taught by a phonics approach made a significantly greater number of repetitions. 3. Self-Corrections. For the children instructed through the phonics approach the average number of self-corrections was 10.00. For the children instructed through the language experience approach the average number of self-correc-tions was 7.U3. There was no significant difference between the groups. The children instructed through the phonics approach self-corrected 1U.8U percent** of the errors or one i n every 57.85 words. The group * » J. , , . , , number of self-corrections v percentage obtained by r • • • - A 100 = r 0 \" number of errors + number of number of errors + number of self-corrections 1*3 instructed through the language experience approach self-corrected 20.66 percent of the errors or one in every 1*2.98 words. These percentages are lower than those of Clay (1967) and Norton and Hubert (1977). In the study by Clay (1967) the children learning to read by a meaning emphasis corrected 26 percent of their errors. In the study by Norton and Hubert (1977) the eclectic readers corrected approximately 50 percent of their errors which was significantly more than the phonics group self-corrected. 1*. Fluency. The children instructed through the phonics approach took an average of 113.60 seconds to read a selection, whereas the group instruc-ted by the language experience approach took an average of 91.99 seconds to read a selection. There was no significant difference between the two groups. Elder (1971) found that the Scottish children instructed by a phonic approach were slower readers than the American children instructed by a sight recognition approach. DeLawter (197U) also noted that the phonics group \"frequently took longer attempting to figure out unknown words\" (p. 1*6) than did the meaning emphasis group. 5. Comprehension. The group taught through the phonics approach achieved a mean score of 29.61* in comprehension while the children taught through the language experience approach achieved a mean score of 32.1*3 in this area. There was no significant difference between the two groups. Dank (1976), Norton and Hubert (1977), and Burke (1973) noted that children taught by a phonics approach produced lower comprehension scores than children taught by a sight recognition approach. Part 2: Substitution Components The substitution errors were further examined and categorized for words/nonwords, graphic similarity of words, graphic similarity of non-words, sound similarity of words, sound similarity of nonwords, no meaning change, and contextual appropriateness. The number of substitutions for each component category were tabulated for the two instructional groups. The mean percentage score, the standard deviation were calculated for each group. The t-test for independent samples was applied to determine the differences between groups. The data are presented i n Table 2. Trends Between Groups 1. Nonwords. For children learning to read through the phonic approach, 2U.90 percent of their substitutions were nonwords. For children learning to read through a language experience approach 15.U5 percent of their sub-stitutions were nonwords. The difference was significant. This finding supports earlier research by Elder (1971), Dank (1976), Norton and Hubert (1977), Norton (1976), Cohen (197U-75), and DeLawter (197U). The percentages of nonwords i n DeLawter's study were considerably higher—65 percent for children being instructed by a phonics approach and U6 percent for children being instructed by a whole word approach, than i n the present study. On the other hand, Bennett (19U2) reported that none of the 3U,39U errors analyzed i n her study were nonsense words. 2. Graphic similarity of word substitutions. The children being instructed through the phonics approach had an average of 6$.29 percent graphically similar word substitutions. The group being instructed through the language experience approach had an Table 2 Comparison of means, standard deviation and t-values for Substitution Components for the Phonics Group and the Language Experience Group Substitution Component Phonics Language Experience Means Standard Deviation Means Standard Deviation t-value Nonwords .2U90 .11*8 .15U5 .115 2.67 ** Graphic Similarity of words .6529 .186 .6366 .231 .29 Graphic Similarity of nonwords .9317 .109 .9529 .109 -0 .73 Sound Similarity of words .6963 .139 .6023 .211 1.97 Sound Similarity of nonwords .9687 .053 .9623 .111 .27 No Meaning Change .2klh .113 .U966 .282 -U.U5 * * Contextual Appropriateness .6508 .156 .7U92 .2U8 -1.78 **Signif leant at the .01 level. U6 average of 63,66 percent graphically similar word substitutions. There was no significant difference between the two groups. Earlier research (Barr, 1972, 1975J DeLawter, 197Uj Dank, 1976j Norton and Hubert, 1977j Norton, 1976) found that children instructed by a phonics approach produced more graphically similar substitutions than did children instructed by a whole word approach. 3, Graphic similarity for nonword substitutions. The nonsense words that were substituted for text words were also analyzed for graphic similarity. For the children taught by the phonics approach the mean percentage was 93.17 percent and for the child-ren taught by the language experience approach the mean percentage was 95*29 percent. There was no significant difference between groups. U. Sound similarity of word substitutions. For children taught by the phonics approach the mean percent-age of words in this category was 69.63 percent. For the group taught by the language experience approach the mean percentage of words in this category was 60.23 percent. There was no significant difference between the two groups. Previous research (DeLawter, 197Uj Dank, 1976; Norton and Hubert, 1977j and Norton, 1976) reported that children taught by a phonic approach produced a higher percentage of word substitutions that were simi-lar in sound to the stimulus word. 5. Sound similarity of nonword substitutions. The nonwords produced by both the children taught by the phonics approach and the children taught by the language experience approach were closely related in sounds to the stimulus word. The mean percentage for the children taught by the phonics approach was 96.87 percent and for 1*7 children taught by the language experience approach 96.23 percent. There was no significant difference. 6. No meaning change. For children instructed by the phonics approach the mean per-centage of substitutions that did not change the meaning of the sentence was 2U.2U percent. For children instructed by the language experience approach the mean percentage of substitutions that did not change the mean-ing of the sentence was 1*9.66 percent. The difference between the groups was significant. This finding supports earlier research by Elder (1971), Burke (1973), Norton and Hubert (1976), and DeLawter (1971*) which noted that children instructed by a phonics approach produced more substitutions that changed the meaning of the sentence than did children being instructed by a sight recognition approach. 7. Contextual appropriateness. For children taught by a phonics approach 65.08 percent of their substitutions were contextually appropriate. For the children taught by the language experience approach 71*.92 percent of their substitutions were contextually appropriate with the preceding part of the sentence. The difference between the groups was not significant. The mean percentages i n the present study are somewhat lower than the 91 percent of the substitutions that Weber (1970) noted were grammatic-a l l y appropriate to the preceding context but higher than the 58 percent produced by the meaning emphasis group i n Clay's (1967) research. Trends Between Subgroups. The subjects i n each treatment group were divided into three groups: high, middle and low, based on the number of 1*8 substitution errors they had produced. The mean percentage score and the standard deviation were calculated f o r each subgroup—high, middle, and low subgroups for children taught by a phonics approach, and high, middle, and low subgroups for children taught by a language experience approach. The t-test for independent samples was applied to determine the differences between subgroups. The data are presented i n Table 3 for nonwords, Table 1* for graphic similarity of words and of nonwords, Table 5 for sound simi-l a r i t y of words and of nonwords, and Table 6 for no meaning change and for contextual appropriateness. The subgroups w i l l be referred to i n the following manner: The children instructed through the phonics approach w i l l be re-ferred to as the high phonics subgroup, the middle phonics subgroup and the low phonics subgroup. The children instructed through the language experience approach w i l l be referred to as the high LE subgroup, the middle LE subgroup and the low LE subgroup. The phonics achievement subgroups (high, middle, low) produced sig-nificantly more substitutions than the comparable high, middle, and low language experience subgroups. 1. Nonwords. The high phonics subgroup produced a mean percentage of 20.73 percent nonwords. The high LE subgroup produced a mean percentage of 6.22 percent nonwords. This difference was significant. For the middle phonics subgroup the mean percentage of nonwords was 29*86 percent. For the middle LE subgroup the mean percentage of non-words was 18.95 percent. There was no significant difference between the groups* Table 3 Comparison of Means, Standard Deviation and t-value for Nonwords for High, Middle, and Low Achieving Phonics and Language Experience Beginning Readers Standard Group Mean Deviation t-value high phonics ,2073 .lit? ? 2.U7 * high LE .0622 11.7}.097 J middle phonics .2986 .182 ) 1.63 middle LE .1895 .108j low phonics .2356 .099 ) .63 low LE .2079 .086/ *Significant at the .05 level. 50 The low phonics subgroup produced 23.56 percent nonwords and the low LE subgroup produced 20,79 percent nonwords. There was no significant difference between these two groups. In each achievement subgroup—high, middle, and low—the children receiving instruction by a phonics method produced more nonword substitu-tions than children receiving instruction by a language experience method, 1. Graphic similarity of word substitutions. For the high phonics subgroup the mean percentage of graphic-a l l y similar words was U9.66 percent. For the high LE subgroup the mean percentage of graphically similar words was U7.U1 percent. There was no significant difference between the subgroups. The middle phonics subgroup produced a mean percentage of 70.97 percent for this aspect of substitutions. The middle LE subgroup produced a mean percentage of 68.21 percent for this aspect of substitutions. There was no significant difference between the two subgroups. The mean percentage for the low phonics group was 7U.62 percent and 71**85 percent for the low LE group. The low LE subgroup had the high-est percentage of graphically similar words while the high LE subgroup had the lowest percentage of graphically similar words. The high subgroups of both treatment groups produced a lower percentage of graphically simi-l a r words than did the middle and low subgroups. Cohen (197U-75) also noted that the poorer readers at the end of the school year s t i l l produced many substitutions graphically similar to the stimulus word, while the better readers' substitutions were on small function words. The findings of this study do not agree with those of Weber ( 1 9 7 0 ) , Bennett (191*2), and Clay ( 1 9 6 7 ) , who noted that the high group made more 51 Table h Comparison of Means, Standard Deviation, t-value for Graphic Similarity of Substitution Errors, Words and Nonwords, for High, Middle, and Low Achieving Phonics and Language Experience Beginning Readers Group Mean Words Standard Deviation t-value Mean Nonwords Standard Deviation t-value high phonics high LE middle phonics middle LE low phonics low LE 1.000 .9296 .9018 .9213 .9593 .167] .0 j .0861 .061 ) .088 j -1.00 JUS -1.07 52 graphically similar substitutions than did the low group. 3. Graphic similarity of nonword substitutions. The mean percentages of graphically similar nonwords produced by the three phonics subgroups and the three language experience subgroups were all above 90 percent. There were no significant differences between subgroups of comparable achievement. U. Sound similarity of word substitutions. For the high phonics subgroup the mean percentage of word sub-stitutions in this category was 70.07 percent. For the high LE subgroup the mean percentage of word substitutions in this category was U7.U1 per-cent. This difference was not significant. A comparison of Table h and Table 5 shows that the high LE subgroup produced the same mean percentage of word substitutions similar in sound as graphically similar. The high phonics subgroup, by contrast, had a higher mean percentage of words simi-lar in sound than words graphically similar. The mean percentage for the middle phonics subgroup was 71.08 per-cent, and the mean percentage for the middle LE subgroup was 69.37 percent. There was no significant difference between the two subgroups. The low phonics subgroup had a mean percentage of 67.60 percent for this category while the low LE subgroup had a mean percentage of 62.90 percent* There was no significant difference between the two subgroups. The mean percentages in this study are higher than those mentioned in the study by Norton and Hubert (1977), in which they noted that the high phonic group produced 66.0 percent, the high eclectic produced UO.U per-cent, the low phonic produced 55*2 percent, and the low eclectic produced 32.8 percent. 53 Table 5 Comparison of Means, Standard Deviation and t-value for Sound Similarity of Substitution Errors, Words and Nonwords, for High, Middle and Low Achieving Phonics and Language Experience Beginning Readers Group Words Standard Mean Deviation t-value Mean Nonwords Standard Deviation t-value high phonics high LE middle phonics middle IE low phonics low LE 1.000 1.000 .9588 .9667 .9U81* .9198 . 0 ) 0. 0 .056 .105 o.o -.21 .50 5U 5. Soond similarity of nonword substitutions. The phonics subgroups and the language experience subgroups a l l attained mean percentages above 90 percent* The high phonics sub-group and the high LE subgroup produced mean percentages of 100 percent* The middle phonics subgroup produced an average of 95*88 percent nonword substitutions with sound similarity, and the middle LE subgroup produced an average of 96*67 percent. For the low phonics subgroup the mean percentage was 9U.8U percent, and for the low LE subgroup the mean percentage was 91*98 percent* There was no significant difference between comparable subgroups* 6. No meaning change. For the high phonics subgroup 3U.69 percent of their substitu-tions did not change the meaning of the sentence. For the high LE sub-group 79.U2 percent of the substitutions did not change the meaning of the sentence. This difference was significant. The mean percentage of substitutions that did not change the mean-ing of the sentence was 20.82 percent for the high phonics group and U5.19 percent for the high LE subgroup. This difference was significant. For the low phonics subgroup 17*29 percent of their substitutions did not change the meaning of the sentence. For the low LE subgroup 2U.87 percent did not change the meaning of the sentence. This difference was significant. For both the phonics subgroups and the language experience sub-groups the mean percentages decreased as the achievement level of the sub-groups decreased. Norton and Hubert (1977) also noted this decrease in mean percentages with the decrease in ability level of the subgroups. 55 Table 6 Comparison of Means, Standard Deviation and t-value for No Meaning Change and Contextually Appropriateness of Substitution Errors for High, Middle, and Low Achieving Phonics and Language Experience Beginning Readers Qroup No Meaning Change Mean Standard Deviation t-value Contextually Appropriate Standard Mean Deviation t-value high phonics high LE middle phonics middle LE low phonics low LE .31*69 .791*2 .2082 .1*519 .1729 .21*87 .119 ? -5.69 .201; J .081*) .203/ .01*2) .09l} -3.50 -2.27 .8018 .81*1*1* .6305 .8031* .5225 .5937 .11*61 -0. .328 / 36 .111) -2.59 .180/ .051) -1.37 .11*7 ( Significant at the ,05 l e v e l . Significant at the .01 l e v e l . 5 6 7. Contextual appropriateness. For the high phonics subgroup 80.18 percent of the substitu-tions were contextually appropriate, and for the high LE subgroup 8U.UU percent of the substitutions were contextually appropriate. There was no significant difference between the two subgroups. For the middle phonics subgroup 6 3 . 0 5 percent of the substitutions were contextually appropriateJ for the middle LE subgroup 80.3k percent of the substitutions were contextually appropriate. There was a significant difference between the two subgroups. The low phonics subgroup had a mean percentage of 5 2 . 2 5 percent contextually appropriate substitutions, and the low LE subgroup had a mean percentage of 59,37 percent contextually appropriate substitutions. There was no significant difference between these two subgroups. In this study the number of contextually appropriate errors diminished as the achieve-ment level of the subgroups diminished, unlike the study by Weber (1970), which found \"negligible\" difference between the high and low groups i n this category. SUMMARY The present chapter has presented and interpreted the data collec-ted. The oral reading behaviours were coded and subjected to s t a t i s t i c a l analysis to determine what similarities and differences existed between children taught by a phonics approach and children taught by a language experience approach. Behaviours examined were types of oral reading err-ors, repetitions, self-corrections, comprehension, and fluency. The sub-stitution errors were further analyzed for nonwords, graphic similarity of 57 words, graphic similarity of nonwords, sound similarity of words, sound similarity of nonwords, no meaning change and contextual appropriateness. The major findings are summarized below. 1. There was no significant difference between groups i n the error categories, no response/don't know, insertions, omissions. 2. The children taught by the phonics approach made a s i g n i f i c -antly greater number of substitutions and sounding out type errors. 3. There were no significant differences between groups for repe-t i t i o n s , self-corrections, fluency and comprehension. U. In the analysis of substitution components there were no s i g -nificant differences between groups for graphic similarity of words, graphic similarity of nonwords, sound similarity of words, sound similarity of non-words, and contextual appropriateness. 5. In the analysis of substitution components the children taught by the phonics approach produced significantly more nonwords than did the children taught by the language experience approach. 6. In the analysis of substitution components the children taught by the language experience approach produced significantly more substitu-tions that did not change the meaning of the sentence than did the children taught by the phonics approach. 7. There were no significant differences l n mean percentages bet-ween the high, middle, and low phonics subgroups and the comparable lang-uage experience subgroups for graphic similarity of words, graphic similar-i t y of nonwords, sound similarity of words and sound similarity of nonwords. 8. There was no significant difference between the middle phonics subgroup and the middle LE subgroup, and between the low phonics subgroup and the low LE subgroup i n the mean percentage of nonword substitutions, 58 but the high phonics subgroup produced significantly more nonwords than the high LE subgroup. 9» The high, middle and low language experience subgroups prod-uced a significantly greater number of substitutions that did not change the meaning of the sentence than the high, middle and low phonics subgroups. 10, There was no significant difference between the high phonics subgroup and the high LE subgroup, and between the low phonics subgroup and the low LE subgroup i n the mean percentage of contextually appropriate errors. The middle LE subgroup produced significantly more contextually appropriate substitutions than the middle phonics subgroup. Chapter V SUMMARY, FINDINGS, CONCLUSIONS, AND EDUCATIONAL IMPLICATIONS Several investigations i n the area of oral reading behaviour have been conducted. These studies have noted the effect of a phonics emphasis and/or a basal reader sight vord emphasis on the oral reading behaviour of children. Prior to the present study no study could be found which exam-ined and compared the oral reading behaviour of children being taught to read by a language experience emphasis—a teacher developed program using the child's natural language and self-selected sight vocabulary—with the oral reading behaviour of children being taught to read by a phonics empha-s i s . The present study adds evidence concerning the effect of a language experience emphasis on the oral reading behaviour of beginning readers and the effect of a phonics emphasis on the oral reading behaviour of begin-ning readers, SUMMARY The purpose of this study was to determine the extent to which children instructed i n i n i t i a l reading by a language experience approach differed i n their oral reading behaviours from children who had been i n -structed i n i n i t i a l reading by a phonics approach. The study sought ans-wers to the following questions: 1. What i s the effect of alternative instruction on the following aspects of reading behaviour—oral reading errors, self-corrections, 59 60 repetitions, comprehension and fluency? 2. What is the effect of alternative instruction on the following aspects of substitution errors—words/nonwords, graphic similarity of words, graphic similarity of nonwords, sound similarity of words, sound similarity of nonwords, contextual appropriateness, and no meaning change? a) What are the effects of treatment on these seven components for the group taught by the phonics emphasis and the group taught by the language experience emphasis? b) What are the effects of treatment on these seven components for achievement groups—high, middle, and low phonics sub-groups and high, middle, and low language experience sub-groups? Administration of Instruments The children were administered individually the Spache Diagnostic Reading Scales—tests 1A, IC, 2A, 2C, 3A, 3C during the last two weeks of May, 1978. Treatment of the Data The results of each child's oral reading of the Spache Diagnostic Reading Scales was recorded and coded and a computer card, one for each child, was key punched. The data was then run through an SPSS (Statistical Package for the Social Sciences), performed on IBM 360/67 computer using the applicable sub-routines of SPSS. Mean scores were computed for oral reading errors, repetitions, self-corrections, fluency and comprehension. Mean percentages were computed for substitution components. The t-test for independent samples was subsequently applied to determine differences, significant at the $ percent level of confidence, between groups and between subgroups. 61 FINDINGS Briefly, the questions raised at the beginning of this investiga-tion were answered in the following manner based on the data presented in Chapter four. Part 1: Reading Behaviour 1. Errors. Children taught by a phonics emphasis made substantially more errors (1607) than children taught by a language experience emphasis (1007). They produced a significantly higher mean for number of substitution errors and number of sounding out errors than did the children taught by a lang-uage experience approach. As many studies have reported, substitution errors comprised the largest percentage of errors produced for the group taught to read by the language experience approach, and for the group taught to read by the phon-ics approach. 2. Repetitions. No significant difference was found in this category between the group taught by the phonics approach and the group taught by the lang-uage experience approach. 3. Self-corrections. No significant difference was found in this category between the group taught by the phonics approach and the group taught by the lang-uage experience approach. The children taught by the phonics approach self-corrected LU.8U percent of the errors produced while the group taught by the language experience approach corrected 20.66 percent of the errors produced. 62 U. Fluency. No significant difference was found i n this category between the children taught by the phonics approach and the children taught by the language experience approach. 5. Comprehension. The children taught by the phonics approach produced lower mean comprehension scores than the children taught by the language experience approach. The difference was not significant. Part 2: Substitution Components !• Between treatment groups. Of the seven substitution components analyzed significant d i f f -erences were noted for two components. The children taught through the phonics approach made a significantly higher mean percentage of nonword substitutions than the children taught through the language experience app-roach. The children taught through the language experience approach pro-duced a significantly higher mean percentage of substitutions which did not change the meaning of the sentence. There was no significant difference between children taught by the phonics approach and children taught by the language experience approach for graphic similarity of words, graphic similarity of nonwords, sound similarity of nonwords, and contextually appropriate substitutions. The children taught through a phonics method made a higher mean score of word substitutions similar i n sound to the stimulus word than did the group taught through the language experience approach. The difference approached significance. 63 2. Between achievement sabgroups. The high, middle and low phonics subgroups produced s i g n i f i -cantly more substitutions than did the high, middle and low language experi-ence subgroups* The high phonics subgroup made a significantly higher mean percent-age of nonword substitutions than did the high LE subgroup. The middle and the low phonics subgroups made higher mean percentages of nonword substitu-tion than did the middle and low LE subgroups but the differences were not significant. There were no significant differences between the high, middle and low achieving phonics subgroups and the high, middle and low language ex-perience subgroups on the mean percentage of graphically similar words produced. The low phonics subgroup and the low LE subgroup had the highest scores for graphic similarity with the two high subgroups producing the lowest mean percentages of graphically similar words. There was no significant difference between the high, middle and low phonics subgroups and the high, middle and low LE subgroups i n mean percentages of graphically similar nonwords produced and nonwords similar i n sound to the stimulus word. There was no significant difference between the high, middle and low phonics subgroups and the high, middle and low LE subgroups i n mean percentage of words produced that were similar i n sound to the stimulus word. The high, middle and low phonics subgroups produced higher mean percentages of words similar i n sound than did the comparable language ex-perience subgroups and the difference between the high phonics subgroup and the high LE subgroup approached significance. The high, middle and low language experience subgroups produced 6k significantly higher mean percentages of substitutions that did not change the meaning of the sentence than did the high, middle and low phonics sub-groups. The middle LE subgroup made a significantly higher mean percentage of contextually appropriate substitutions than did the middle phonics sub-group. There were no significant differences between the high phonics sub-group and the high LE subgroup, or between the low phonics subgroup and the low LE subgroup, CONCLUSIONS The results of the present investigation seem to warrant the following conclusions: 1, Among children taught by either a phonics approach or a lang-uage experience approach, substitution of a word different from the ex-pected word was the most frequent error. 2, The phonics approach produces readers who make more errors than the readers taught by the language experience approach. This trend was consistent when the high, middle and low phonics subgroups were compared with the high, middle and low language experience subgroups. 3, The phonics approach produces readers who make many nonwords and many substitutions which alter the meaning of the sentence while the language experience approach, with i t s emphasis on meaning, develops read-ers who produce few nonwords and many meaningful substitutions that do not distort the meaning of the sentence. I t would appear that an approach which emphasizes the sound symbol relationship, as does the phonics app-roach, produces readers who f a i l to pay attention to what the author i s 65 saying. This trend was consistent when the high, middle and low phonics subgroups were compared with the high, middle, and low language experience subgroups, U, The phonics approach and the language experience approach pro-duces readers capable i n the use of graphic and sound elements, although the phonics approach produces readers who make more use of the sound elem-ents than readers taught by the language experience approach. This trend was consistent for the phonics and language experience subgroups as well; the exception being the high language experience subgroup. The failure of the high achieving readers taught by the language experience approach to use graphic and phonic elements i s indicative of the third stage i n reading development mentioned by Biemiller (1970), MacKinnon (1959), and Goodman (1971), i n which the reader pays less attention to the grapho-phonemic elements and more to the syntactic and semantic elements. 5, Although both the phonics and the language experience approaches produce readers aware of syntactic and semantic constraints, the language experience group used these constraints more effectively. This trend was consistent within the phonics and language experience achievement groups. The average achieving language experience readers seem more cognizant of syntactic and semantic constraints than those taught by the phonics approach, RECOMMENDATIONS 1, Oral reading analysis may be beneficial to the classroom teacher. It i s recommended that such analysis be used as a diagnostic tool by the classroom teacher i n order to discover the strengths and weaknesses of each child, Weber (1968) states: 66 • • . the study of reading errors can provide significant clues to the nature of the reading process and, in this way, can contribute to a substantive rationale for both basic and remedial instruction in reading, (p. 98.) Goodman (1971*) also notes that; Miscues are not simply errors. They show more about the learner's strengths than about his weaknesses. In reading they are the best possible indications of how efficiently and effectively the reader is using the reading process, (p. 61*.) 2, It is recommended that the classroom teacher use the results of the oral reading analysis to plan an individualized programme for the child that will strengthen the reading strategies the child uses effect-ively, and will teach him strategies he is deficient in. Yetta Goodman (197U) states that, \"Strategy lessons help readers focus on aspects of written language they are not processing effectively.\" (p. 36.) 3. Classroom teachers should be cognizant of the fact that, . . . i t is most likely that at least as many children are suffering from difficulties caused by overusing particular learning strategies in reading as are suffer-ing from a lack of such strategies. (Goodman, 1965, p. 61*3.) 1*. Teachers should ensure that children are taught to read for meaning. It is recommended that teachers using a heavy phonics emphasis program such as the Language Patterns Program, add to and strengthen the program by ensuring that children are made aware that the major purpose of reading is to understand what the author is saying, 5, It is recommended that all children who fail to read for mean-ing should receive instruction in doing so. Stauffer (1970) states: It is possible to direct the reading-thinking process in such a way that children will be encouraged to think when reading—to speculate, to search, to evaluate, and to use. . (p. 3U8.) The children should be taught to pause and to think about what they have 67 read. They can r e s t a t e i n t h e i r own words what they have read. A l l o w the c h i l d r e n t o d i c t a t e and w r i t e t h e i r own s t o r i e s i n order t h a t the language and content w i l l be t h e i r own and w i l l be f a m i l i a r t o them. I t i s import-ant t o l e t c h i l d r e n read about experiences they are f a m i l i a r w i t h f o r as P i k u l s k i (1976) s t a t e s : . . . the reader processes v i s u a l i n f o r m a t i o n on the b a s i s o f what he o r she already \"knows\" which r e f e r s both t o what i s known about the s t r u c t u r e of language and what i s known i n terms o f background o f informa-t i o n , ( p . 376.) The teacher must a l s o provide c h i l d r e n w i t h the ncessary experiences t o en-able them t o understand the content of the m a t e r i a l t o be read. C h i l d r e n can be taught t o read p a s t the d i f f i c u l t word t o see i f they can f i g u r e out the meaning of the word i n r e l a t i o n t o the r e s t o f the sen-tence. 6. I t i s recommended t h a t teachers a l l o w c h i l d r e n t o make mistakes, f o r as Prank Smith (1971) s t a t e s : . . . f l u e n t reading and l e a r n i n g t o read f l u e n t l y , r e q u i r e a w i l l i n g n e s s t o 'make mistakes.' And the extent t o which a c h i l d i s prepared t o r i s k mistakes i s d i r e c t l y r e l a t e d t o the t o l e r a n c e o f the teacher i n a c c e p t i n g them. (p. 230.) 7. I t i s recommended t h a t teachers a l l o w the c h i l d r e n t o f i g u r e out the word, i f p o s s i b l e , w i t h o u t prompting. I f an e r r o r i s made, the teacher should encourage the c h i l d t o re g r e s s and attempt t o c o r r e c t the e r r o r , f o r as Recht (1976) s t a t e s : The reader who i s encouraged t o r e g r e s s and attempt t o make sense o f an i n c o n s i s t e n c y encountered dur-i n g o r a l reading l e a r n s from the process, (p. 638.) 8. I t i s recommended t h a t a c h i l d who i s unsuccessful using one rea d i n g approach which emphasizes a p a r t i c u l a r s t r a t e g y should be helped t o i d e n t i f y new s t r a t e g i e s i n the hope he w i l l experience success a t r e a d i n g . 68 9. Teachers must be cautioned not to judge a child's reading a b i l i t y s t r i c t l y on the number of oral reading errors he makes because, as Yetta Goodman (197U) comments: The grammatical structure, style of writing or concept load of any particular part of a story a l l are involved i n the complex reasons which cause readers to produce miscues and which cause miscue numbers to vary from one part of the story to another, (p. 67.) Rather, look closely at oral reading errors i n order to see whether they distort the meaning of the story. Also, It i s not enough to say that the reader sometimes substituted a for the. It must be seen that such behavior can~only result from the l i n g u i s t i c compet-ence of the reader which makes i t possible for him to produce a determiner, where one i s needed. I t must also be seen that something more than word recognition or l e t t e r perception i s involved. (Goodman, 1969, p. 13.) SUGGESTIONS FOR FURTHER RESEARCH 1. A replication of the present study using grade one children i n other geographical areas of Canada would add to the present findings, and to the applicability of the findings to a wider population. 2. It i s recommended that a replication of the present study could be carried out longitudinally so as to determine whether strategies taught i n beginning reading are s t i l l evident i n subsequent grades. 3. It i s suggested that a longitudinal study be conducted i n order to examine children taught by a phonics approach and a language experience approach throughout their f i r s t year i n school, to note developmental trends. It. Further research i s needed to determine how children expand their i n i t i a l reading strategy and adopt different reading strategies. Is i t developmental? !>• I t i s recommended that an indepth study of the self-correction 69 strategies of children learning to read by a method emphasizing language experience be compared with children learning to read by a method emphasiz-ing phonics. When i s a child more prone to correct an error? 70 BIBLIOGRAPHY Ashton-Warner, Sylvia. Teacher. New York: Bantam Books, 1963. Barr, Rebecca. \"The Influence of Instructional Conditions on Word Recognition Errors,\" Reading Research Quarterly, 7:3 (Spring 1972), 509-529. Barr, Rebecca. \"Influence of Instruction on Early Reading,\" Interchange, 5:U (197U), 13-22. Barr, Rebecca. \"The Effect of Instruction on Pupil Reading Strategies,\" Reading Research Quarterly. 10:1* (197U-75), 555-582. Bennett, Annette. \"An Analysis of Errors i n Word Recognition Made by Retarded Readers,\" Journal of Educational Psychology, 33 (19l*2), 25-38. Biemiller, Andres. \"The Development of the Use of Graphic and Contextual Information as Children Learn to Read,\" Reading Research Quarterly, 6:1 ( F a l l 1970), 78-96. Bond, Guy L., and Robert Dykstra. \"The Cooperative Research Program i n First-Grade Reading Instruction,\" Reading Research Quarterly. 2:1*, (Summer 1967), 5-U*2. Burke, Carolyn. \"The Reading Process Oral Reading Analysis: A View of the Reading Process,\" i n William Page (ed.), Help for the Reading Teacher: New Directions i n Research. Urbana, I l l i n o i s : ERIC Clearinghouse on Reading and Communication S k i l l s , National Conference on Research i n England, 197i*, pp. 25-35, Burke, Elizabeth. \"A Developmental Study of Children's Reading Strategies,\" Birmingham University Educational Review, 29:1* (1976-77), 30-1*6. Burke, M.S. \"A Psycholinguistic Analysis of Oral Reading by Selected First-Graders Taught Analytically or Synthetically.\" Unpublished Master's Equivalency Thesis, Rutgers University, 1973. Butler, Lester G. \"A Psycholinguistic Analysis of the Oral Reading Behavior of Selected Impulsive and Reflective Second Grade Boys.\" (May 1971*), ED 09l* 325. Chall, Jeanne S. Learning to Read: The Great Debate. New York, N.Y.: McGrawrHill, 1967. Chall, Jeanne. \"Research i n Linguistics and Reading Instruction: Implica-tions for Further Research and Practice,\" i n J.A. Figural (ed.), Reading and Realism, Proceedings of the 13th Annual Convention of the International Reading Association, Part I. 13 (1969), 560-571, 71 Chall, Jeanne. \"Research i n linguistics and Reading Instruction: Implica-tions for Further Research and Practice,\" i n Robert Kalin (ed.), Perspectives on Elementary Reading. New York: Harcourt Brace Jovanovich, 1973, pp. 188-201. Clay, Marie M. \"The Reading Behaviour of Five Year Old Children: A Research Report,\" New Zealand Journal of Educational Studies, 2 (May 1967), 11-31. Clay, Marie M. \"A Syntactic Analysis of Reading Errors,\" Journal of Verbal Learning and Behavior, 7 (1968), l*3l*-l*38. Clay, Marie. \"Reading Errors and Self-Correction Behavior,\" British Journal of Educational Psychology. 39 (Feb. 1969), 1*7-56. Cohen, Alice Sheff. \"Oral Reading Errors of F i r s t Grade Children Taught by a Code Emphasis Approach,\" Reading Research Quarterly, 10:1* (197U-75), 616-6U9. Dank, Marion Edelson. \"A Study of the Relationship of Miscues to the Mode of Formal Reading Instruction Received by Selected Second Graders, 1976.\" ED 126 1*31. DeLawter, Jayne A. \"Miscue Patterns: The Relationship of Beginning Reading Instruction and Miscue Patterns,\" i n William Page (ed.), Help for the Reading Teacher: New Directions i n Research. Urbana, I l l i n o i s : ERIC Clearinghouse on Reading and Communication S k i l l s , National Conference on Research i n England, 197i*, 1*2-51. Dorchester County Board of Education. \"Right to Read Language Experience Program.\" Cambridge, Md.: ERIC Document No. ED 105 1*1*9, 197U. Elder, Richard D. \"Oral Reading Achievement of Scottish and American Children,\" Elementary School Journal, 71 (Jan. 1971), 216-230. Ewoldt, Carolyn. \"Miscue Analysis of the Reading of Third Grade Follow Through and Non-Follow Through Children i n Wichita, Kansas.\" ERIC Document ED 136 219, October, 1976. Folman, Shoshana. \"A Psycholinguistic Analysis of Miscues Generated by a Selected Group of Israeli Non-Native Speakers of English During the Oral Reading of an Unadapted American Story--A Descriptive Study,\" Dissertation Abstracts, 38:1* (October 1977), 19i*3A. Glass, Gene V., and Julian Stanley. S t a t i s t i c a l Methods i n Education and Psychology. Englewood C l i f f s , N.J.: Prentice-Hall, 1970. Goodman, Kenneth S. \"A Linguistic Study of Cues and Miscues i n Reading,\" Elementary English. 1*2 (1965), 639-61*3. Goodman, Kenneth S. \"Analysis of Oral Reading Miscues: Applied Psycho-lingu i s t i c s , \" Reading Research Quarterly. 1 ( F a l l 1969), 9-31. 72 Goodman, Kenneth S., and C. Burke. A Study of Oral Reading Miscues That Result in Grammatical Re-Transformation, U.S.O.E. Final Report, Project No. 7-E-219, Contract No. OEG-0-8-070219-28O6 (010), U.S. Department of Health, Education, and Welfare, June, 1969. Goodman, Kenneth. \"Reading: A Psycholinguistic Guessing Game,\" in H. Singer and R.B. Ruddell (eds.), Theoretical Models and Processes of Reading. Newark* Delaware:- International Reading Association, 1970, pp. 259-271. Goodman, Kenneth S. \"Reading: You Can Get Back to Kansas Anytime You*re Ready Dorothy,\" English Journal. 63:8 (Nov. 197b), 62-61*. Goodman, K.S., and Yetta M. Goodman. \"Learning About Psycholinguistic Processes by Analyzing Oral Reading,\" Harvard Educational Review, 1*7:3 (Aug. 1977), 317-333. Goodman, Yetta. \"A Psycholinguistic Description of Oral Reading Phenomena in Selected Young Beginning Readers.\" Unpublished Doctoral disser-tation, Wayne State University, 1967. Goodman, Yetta. \"Using Children's Miscues for Teaching Reading Strategies,\" Reading Teacher. 23 (Feb. 1970), 1*55-1*59. Goodman, Yetta. Longitudinal Study of Children's Oral Reading Behavior, Final Report, Project No. 9-E-062, Grant No. OEG-5-9-325062-001*6, U.S. Department of Health, Education, and Welfare, Office of Education, Bureau of Research, Sept. 1971. Goodman, Yetta. \"Reading Diagnosis—Qualitative or Quantitative?\" Reading Teacher. 26 (1972), 32-37. Goodman, Yetta. \"I Never Read Such a Long Story Before,\" English Journal, 63:8 (Nov. 197b), 65-71. Goodman, Yetta. \"Reading Strategy Lessons: Expanding Reading Effective-ness,\" in William Page (ed.), Help for the Reading Teacher: New Directions in Research. Urbana, Illinois: ERIC Clearinghouse on Reading and Communication Skills, National Conference on Research in England, 197U, pp. 36-1*3. Hall, Mary Anne, The Language Experience Approach for Teaching Reading, ERIC Clearinghouse on Reading and Communication Skills, National Institute of Education, International Reading Association, 800 Barksdale Road, Newark, Delaware: 1978, Hodes, Phyllis. \"Oral Reading of Bilingual Yiddish/English Children,\" ED 137 731, May, 1977. Hood, Joyce, and Janet Kendall. \"A Qualitative Analysis of Oral Reading Errors of fteflective and Impulsive Second Graders: A Follow-Up Study,\" Journal of Reading Behavior. 7 (1975), 269-281, 73 Hood, Joyce. \"Qualitative Analysis of Oral Reading Errors: The Inter-Judge Reliability of Scores,\" Reading Research Quarterly, 11 (1975-76), 577-598. Hood, Joyce, and Clara Gonzalez. \"The Oral Reading of Columbian Second-and Fourth-Graders: An Illustration of Issues in Cross-Cultural Oral Reading Research.\" ED 130 226, 1976. Jastak, J.F. Wide Range Achievement Test. Wilmington, Delaware: Guidance Associates of Delaware, 1965. Jastak, J.F., and S.R. Jastak. The Wide Range Achievement Test Manual. Wilmington, Delaware: Guidance Associates of Delaware, 1965. Linn, J.R., Mabel Bruce, Dorothy Donaldson, Jean Ellis, Anne Saunders, and Janet Trischuk. Language Patterns. Toronto: Holt, Rinehart and Winston of Canada, Ltd., 1967* MacKinnon, A.R. How Do Children Learn to Read? Vancouver: Copp Clark, 1959. Madden, Marie, and Marjorie Pratt. \"An Oral Reading Survey as a Teaching Aid,\" Elementary English Review. 18 (19kl), 122-26, 159. Monroe, Marion. \"Methods for Diagnosis and Treatment of Cases of Reading Disability,\" Genetic Psychology Monographs, 1* (1928), 335-U56. Norton, Donna E. \"A Comparison of the Oral Reading Errors of High and Low Ability First and Third Graders Taught by Two Approaches—Synthetic-Phonic and Analytic-Eclectic.\" Doctoral dissertation, University of Wisconsin, Madison, 1976. ED 136 191. Norton, Donna E., and Patty Hubert. \"A Comparison of the Oral Reading Strategies and Comprehension Patterns Developed by High, Average, and Low Ability First Grade Students Taught by Two Approaches—Phonic Emphasis and Eclectic Basal.\" ED 1U5 393, 1977. Page, William D., and Rebecca C. Barr. \"Use of Informal Reading Inventories,\" in William Page (ed.), Help for the Reading Teacher: New Directions in Research. Urbana, Illinois: ERIC Clearinghouse on Reading and Communication Skills, National Conference on Research in England, 197k, p. 103. Pikulski, John. \"Linguistics Applied to Reading Instruction,\" Language Arts. 53:lt (1976), 373-77, 381*. Payne, C.S. \"The Classification of Errors in Oral Reading,\" Elementary School Teacher. 31 (1930), ll*2-H*6. Recht, Donna. \"The Self-Correction Process in Reading,\" Reading Teacher, 29:7 (1976), 632-636. 7k Robinson, Helen. \"Insights From Research: Children's Behavior While Reading,\" i n William Page (ed.), Help for the Reading Teacher: New Direction i n Research. Urbana, I l l i n o i s : ERIC Clearinghouse on Reading and Communication S k i l l s , National Conference on Research i n England, 197k, pp. 9-22. Russell, Sheldon Noel. \"Error Pattern Relationship of Developmental Readers and Functionally I l l i t e r a t e Adults.\" ED 133 689, 1973. Smith, Brooks E., Kenneth S. Goodman, and Robert Meredith. Language and Thinking i n the Elementary School. New York: Holt, Rinehart and Winston, Inc., 1970. ' Smith, Frank. Understanding Reading. New York: Holt, Rinehart and Winston, 1971* Spache, . Spache Diagnostic Reading Scales, rev. ed. Monterey, C a l i f . : CTB/Tlc Graw H i l l , Delmonte Research Park, 1972. Stauffer, Russell. The Language Experience Approach to the Teaching of Reading. New York: Harper and Row, 1970. Waltz, Pennie A l i c e . \"Reflection-Impulsivity and Oral Reading Miscues Among Fourth-Grade Boys.\" ED 132 510, June 1977. Weber, Rose-Marie. \"The Study of Oral Reading Errors: A Survey of the Literature,\" Reading Research Quarterly, k (1968), 96-119. Weber, Rose-Marie. \"A Linguistic Analysis of First-Grade Reading Errors,\" Reading Research Quarterly. 3 (Spring 1970), Ij27-U5l. Williamson, Leon, and Freda Young. \"The Reading Performance of Mono-linguals and Bilinguals Compared.\" ED 130 252, 1976. "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0078350"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Education"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "An analysis of the oral reading errors of grade one pupils in terms of two teaching emphases"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/21083"@en .