A PSYCHOLINGUISTIC ANALYSIS OF THE ORAL READING BEHAVIOR OF SELECTED PROFICIENT, AVERAGE AND WEAK READERS READING THE SAME MATERIAL Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY LOUISE JONES JENSEN 1972 TfiESI!‘ Date (II-L’IITI)’ 0-7639 I I II IZIIIILIII I III II II III I This is to certify that the thesis entitled A PSYCHOLINGUISTIC ANALYSIS OF THE ORAL READING BEHAVIOR OF SELECTED PROFICIENT, AVERAGE AND WEAK READERS READING THE SAME MATERIAL presented by Louise Jones Jensen has been accepted towards fulfillment of the requirements for aWcIII/ZV Mnmnmfiaun Michigan State we.” “J‘Z. LIBRARY University IIflAG & sons? 5 Iauux emnm me II LIBRARY amoaas ‘IIIl I SIMON" Inc-Inn ‘I'T ABSTRACT A PSYCHOLINGUISTIC ANALYSIS OF THE ORAL READING BEHAVIOR OF SELECTED PROFICIENT, AVERAGE AND WEAK READERS READING THE SAME MATERIAL By Louise Jones Jensen Through a descriptive analysis of the oral reading behavior of three groups of readers, this study compares the oral reading behavior of proficient readers with that of readers who use less effective reading strategies. Discovery of those characteristics which are manifested by effective readers should enable researchers to develop a theory and a related model of the reading process which will influence the design of instructional programs in reading. The subjects are five proficient second grade readers, five weak sixth grade readers and five highly proficient sixth grade readers. Each of the subjects read the same third grade story orally. The reading was tape recorded and analyzed by means of the Goodman Taxonomy of Reading Miscues. Miscues or mismatches between text and oral response are considered to be "cued" or caused and are generally explicable because they are principled, motivated and rule-governed. The Goodman Taxonomy of'Reading Miscues provides a number of questions to be asked about each miscue‘ to enable the researcher to analyze the interaction of grapho-phonic, grammatical and semantic information as it is processed by the reader. Louise Jones Jensen The results of the study were as follows: 1. 10. 11. When the proficient readers miscued, their substitutions showed less graphic similarity to the text item than did those of the weaker readers. The miscues of the proficient readers resulted in a higher percentage of syntactically acceptable sentences. ' The miscues of the proficient readers involved a higher percentage of re-transformations and alternate options while retaining acceptable deep structure. The miscues of the proficient readers changed syntax on a. higher percentage of occasions than did those of weaker readers. The degree of success in retaining meaning was much higher in the proficient readers. The number of miscues involving intonation was not significant. Mbst differences at the bound morpheme level involved the substitution of alternate inflections due to dialect and as such, were not serious. The proficient readers made fewer substitutions but a higher percentage of omissions, insertions and reversals. This is directly related to their freer use of Optional transformations. The proficient readers substituted fewer non-words, reflecting their concern with meaning and their conceptual experience. The proficient readers made changes at the phrase level with a larger percentage of their miscues than did the other groups. As with phrase change, clause level change was higher for the proficient readers, but the difference between groups was not so great. This category involves deep structure to a greater Louise Jones Jensen extent, and a larger percentage of the proficient readers' miscues were at the surface level. 12. The proficient readers made much better use of correction strategy than did the other two groups. They were aware of the structures which required correction and were usually successful in making that correction. 13. The word level substitutions of the proficient readers showed a much higher relationship to the text than did those of the weaker readers. The data suggest an instructional model directed toward meaning rather than the processing of visual information. The weak readers over-used grapho-phonic skills to the detriment of meaning. A program for beginning reading would be built around a total language program with the role of the school being that of providing substantial amounts of data from which children could develop symbol to sound generaliza- tions inmuch the same way that their oral language was acquired. All reading should take place in a natural language context. Materials should be interesting and meaningful. Children should be encouraged to read a great deal and the experience should be as rewarding and free from threat as possible. The premium on accuracy should be reconsidered. Reading, like language, "is learnt in operation, not by dummy. runs" (John Dixon 1967:13). A PSYCHOLINGUISTIC ANALYSIS OF THE ORAL READING BEHAVIOR OF SELECTED PROFICIENT, AVERAGE AND WEAK READERS READING THE SAME MATERIAL By Louise Jones Jensen A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of English 1972 ACKNOWLEDGMENTS I wish to thank the members of my committee, Dr. James Pickering, Dr. Fred Carlisle and Dr. Steven Judy for their assistance in the preparation of this thesis. I am particularly appreciative of the valuable suggestions and prompt attention received from my committee chairman, Dr. James Stalker. I wish to express my gratitude to Drs. Kenneth and Yetta Goodman, Dr. Carolyn Burke and the research team at Reading Miscue Research for their help, encouragement, and above all, friendship to me while I conducted my research with them. Love and thanks to my husband, Lee, for his patience and encourage- ment, and for his technical assistance in the preparation of the tables and graphs. ii Chapter I II III IV TABLE OF CONTENTS INTRODUCTION . O C C O O O O O Rationale . . . . . . . Design. . . . . . . . . Procedure . . . . . . . Related research. . . . THE GRAPHO-PHONIC SYSTEM . . . Introduction. . . . .'. Grapho-phonic proximity Grapho-phonic proximit Non-words . . . . . .'. Phonics generalizations THE SYNTACTIC SYSTEM . . . . . Introduction. . . . . . Syntactic acceptability Syntactic change. . . . Transformations . . . . Intonation. . . . . . . Dialect . . . . . . . . Structural levels . . . Morphology. . . . . . . Word level. . . . . . . Phrase level. . . . . . Clause level. . . . . . Conclusions . . . . . . THE SEMANTIC SYSTEM. . . . . . Introduction. . . . . . Semantic acceptability. comprehendi g Syntactic and semantic acceptability. Syntactic and semantic change . . . Comprehending . . . . . Miscues per hundred words . Semantic word relationships . Conclusions . . . . . . iii Page \DUI-‘H 12 12 13 22 25 26 43 43 44 48 66 67 76 77 '78 90 - 96 101 103 103 105 107 110 115 116 116 125 Chapter Page V CORRECTION STRATEGY. . . . . . . . . . . . . . . . . . . 126 Introduction. . . . . . . . . . . . . . . . .'. . 126 Correction data . . . . . . . . . . . . . . . .'. 126 Correction and acceptability. . . . . . . . . .'. 133 VI CONCLUSIONS AND INSTRUCTIONAL IMPLICATIONS . . . . . . . 139 Conclusions . . . . . . . . . . . . . . . . . . . 139 Implications of the study . . . . . . . . . . . . 144 A profile of instruction. . . . . . . . . . . . . 147 Beginning reading. . . . . . . . . . . . . 148 The older reader . . . . . . . . . . . . . 152 BIBLIOGMPHY O O O O O O I O O O O O O C O O O O O O O O O O Q ‘0 O 155 APPENDIXES o o o o o o o o o o o a o o o o o o o o o o o o o I. o o 159 A BILLY WHITEWONO O O O 0 O O C O O O O O C O C O O O O C 158 B GOODMAN TAXONOMY 0F READING MISC'UES - SHORT FORM . . . . 166 C STANFORD DIAGNOSTIC MING TEST 0 O O O O O O O O O O I O 177 iv Table 10 11 12 13 14 15 16 17 18 19 20 21 22 23 LIST OF TABLES PERCENT OF GRAPHIC PROXIMITY, GROUP 6L . . . . . . PERCENT OF GRAPHIC PROXIMITY, GROUP 2A . . . . . .~ PERCENT OF GRAPHIC PROXIMITY, GROUP 6H . . . . . . PERCENT OF PHONEMIC PROXIMITY, GROUP 6L. . . . . . PERCENT OF PHONEMIC PROXIMITY, GROUP 2A. . . . . . PERCENT OF PHONEMIC PROXIMITY, GROUP 6H. . . . . . MEAN GRAPHIC SCORES FOR EACH SUBJECT . . . . . . . RELATIONSHIP OF MISCUES TO SYNTACTIC ACCEPTABILITY. .‘ SYNTACTIC CHANGE, GROUP 6L SYNTACTIC CHANGE, GROUP 2A. SYNTACTIC CHANGE, GROUP 6H TRANSFORMATIONS, GROUP 6L. TRANSFORMATIONS, GROUP 2A. TRANSFORMATIONS, GROUP 6H. INTONATION........ DIALECT, GROUP 6L. . . . . DIALECT, GROUP 2A. . . . . DIALECT, GROUP 6H. . . . . DIALECT / COMPREHENDING RELATIONSHIP . . . . . . .' MORPHOLOGY, GROUP 6L . . . MORPHOLOGY, GROUP 2A . . . MORPHOLOGY, GROUP 6H . . . GENERALIZATION-GOVERNED Page 14 15 16 17 18 19 20 34 45 50 51 52 62 63 64 68 71 72 73 75 79 80 81 Table 24 25 26 27 28 29 3O 31 32 33 34 35 36 37 38 39 4O 41 42 43 44 45 46 47 48 FREE MORPHEMES, GROUP 6L . . . . . . . . . FREE MORPHEMES, GROUP 2A . . . . . . . . FREE MORPHEMES, GROUP 6H . . FREE MORPHEME TYPES, GROUP 6L. . . . . . . FREE MDRPHEME TYPES, GROUP 2A. . . . . . .'. FREE MORPHEME TYPES, GROUP 6H. . . . . . . PHRASE, GROUP 6L . . . . . . . . . . . .'. PHRASE, GROUP 2A . . . .'. . . . . . . . . PHRASE, GROUP 6H . . . . . . . . . . . . . CLAUSE, GROUP 6L . . . . . . . . . .4. . . (mmms,cmmP2A. ... .. ... .. .. CLAUSE, GROUP 6H . . . . . . . . . . . . . SEMANTIC ACCEPTABILITY . . . . . . . .-. . SEMANTIC CHANGE, GROUP 6L. . . . . . . . . SEMANTIC CHANGE, GROUP 2A. . . . . . . SEMANTIC CHANGE, GROUP 6H. . . . . . . . . MISCUES PER HUNDRED WORDS / COMPREHENDING. SEMANTIC RELATIONSHIPS, GROUP 6L'. . . . . SEMANTIC RELATIONSHIPS, GROUP 2A . . . SEMANTIC RELATIONSHIPS, GROUP 6H . . . . . PERCENT OF MISCUES CORRECTED, GROUP 6L . . PERCENT OF MISCUES CORRECTED, GROUP 2A . . PERCENT OF MISCUES CORRECTED, GROUP 6H . . PERCENT OF CORRECTIONS FOR SYNTACTIC ACCEPTABILITY PERCENT OF CORRECTIONS FOR SEMANTIC ACCEPTABILITY. vi Page 84 85 86 ' 87 88 89 92 93 '94 98 .'99 100 106 111 11 2 113 118 120 121 122 128 129 130 134 135 Figure 10 ll 12 13 14 15 16 17 18 19 LIST OF FIGURES GRAPHIC PHONEMIC RANGE AND MEANS . . . . GRAPHIC AND PHONEMIC MEANS . . . . . . .’. COMPREHENDING SCORES . . . . . . . . . .‘. GRAPHIC MEANS vs COMPREHENDING MEANS . .'. SYNTACTIC ACCEPTABILITY. . . . . . .'. . . MEANS OF TOTALLY ACCEPTABLE AND TOTALLY UNACCEPTABLE SYNTAX O o o A. O o O o O O O .O o o O a o O SYNTACTIC CHANGE . . . . . . . . TRANSFORMTIONS O O O O O C O O O I .0 O O O 4. DIALECT / COMPREHENDING RELATIONSHIP . . . PHRASE LEVEL MISCUES . . . . . . . . SEMANTIC ACCEPTABILITY . . . . . . . . . . DEGREE OF SYNTACTIC AND SEMANTIC ACCEPTABILITY DEGREE OF SYNTACTIC AND SEMANTIC CHANGE - MEANS. COMPREHENDING RANGE AND MEAN . . . . . . . SMMIC WTIONSHIPS O O O O O O O O O O PERCENT OF CORRECTIONS RANGES AND MEANS. . UNCORRECTION AND CORRECTION FOR SYNTAX . . UNCORRECTION AND CORRECTION FOR SEMANTICS. COMPOSITE GRAPH OF DATA (PARTS 1 THROUGH 4). vii Page 21 21 22 24 47 49 53 65 75 95 108 109 114 117 123 132 136 '137 '140- 143 CHAPTER I: INTRODUCTION Rationale: Through a descriptive analysis of the oral reading behavior of three groups of readers, this study compares the oral reading behavior of proficient readers with that Of readers who use less effective reading strategies (here termed weak readers). Discovery of those characteristics which are manifested by effective readers should enable researchers to develop a theory and a related model of the reading process which will influence the design of instructional pro- grams in reading. Design: MOst research studies deal with a few variables over relatively. large groups. This study, which attempts to describe all the possible variables in reading miscues, is a depth study and as such, it is limited to a small number of subjects. Even though the subject number is low, the results are statistically valid, since fifteen variableB for one subject generates the same volume of data as one variable for fifteen subjects. A total of 1662 miscues was recorded and analyzed; basic statistical procedures were handled through computer program. A study such as this introduces the problem of comparing reading behavior across stories. One developmental trend was the greater percent of' alternate options which the average readers produced, 52 and 42, than the slower readers, .52 and 22. This may be related to the different types of material read 2 by the two groups. The material read by the average readers represents a much greater variety of syntax which allowed for alternate Optional transformations to a greater extent than the primer and first grade material read by the slow readers. (Goodman, Y. 1971:50) In order to control for difference in structure and conceptual load of the material, the present study investigates proficient and weak readers reading the same story, "Billy Whitemoon" from Along Friendly ROads, 1963 (Appendix A). The grade level suggested by the publisher is third grade, second month. The subjects were Obtained by teacher selection rather than by standardized tests,none of which examine reading as a language process involving the simultaneous interaction Of phonological, syntactic and semantic systems (see p. 5 ). Teachers were asked to rank order their students in the presence of the researcher with no opportunity to con- sult test scores. This selection is somewhat arbitrary but the research instrument enables detection of readers whose characteristics differ from others placed within the same group. The weak readers are the bottom five from the list of a sixth grade teacher at Sampson School, Detroit, Michigan. The average readers are a group of second graders, also from Sampson School. They are the sixth through tenth in the rank ordering of the second grade class. Since the story is approximately a third grade level, the average readers were chosen from second grade so that the task would be sufficiently difficult to generate adequate miscueing information. A third group of students has been included in the study. These. are proficient sixth grade readers from Kinawa Middle School in Okemos, Michigan. Since they are proficient readers reading a simple story, 3 their reading behavior in the given task might be termed "highly proficient." They probably epitomize excellent reading behavior. None of these students produced enough miscues to provide a significant sample for analysis, consequently, individual statistics are not dependable. However, the group data provide some interesting phe- nomena of excellent reading against which to compare the reading of the other two groups. Following is a list of subjects by name and number: 6L 2A 6B 188- Bennie 121- Graeme 601- Kathy 189 - Thurman 122- Kevin 602- Nancy 190- LeRoy 123- Deborah 603- Warren 191- Stanley 124- Russell 604- Marlene 192- Danetta 125- Cheryl 605- Kurt Procedure: Each subject read a complete story into a tape recorder after which he freely retold what he had read. Emphasis was placed on having the material at a level which would initiate some reading difficulty without causing the subjects to give up on the task. Each miscue was then analyzed for its relation to the text and the reading process. The instrument by means of which the oral reading was studied is the Goocbnan Taxonomy of Reading Miscues (Appendix B). The Taxonomy is based on the following psycholinguistic model of the reading process. III! from Get rasp II sDel the Author's Author's Inferred Structure (graphic display) rules Structure Deep rules Surface Structure inferred Deep €525h422>>maat from QZHZ>MK Encoding in Oral Reading Reader's Surface Deep Structure Structure (oral display) A writer starts with meaning. He then assigns a deep underlying grammatical structure. Using the transforma- tional rules, he then generates a surface structure. Finally, he utilizes the rules of English orthography (spelling, punctuation) to produce the graphic display. The reader must infer from that graphic display the rules that have produced it and its underlying deep structure. Only then can he reconstruct the writer's message, that is, comprehend the meaning. If he is reading orally, the reader must then- encode the message as oral output producing an oral surface structure. (Goodman, K. 1972:147) The psycholinguistic model of the reading process differs basically "traditional" and "linguistic" views. Traditionally, reading has been considered to be a process of "decoding" print into speech, either vocally or sub-vocally, and responding to the speech signal thereby deriving meaning. ... reading a word involves looking at a group of marks, thinking the word-sound that those marks stand for, and recognizing the meaning for that word sound. (Durr 1965:75) This view of reading leads directly to instruction devised to teach Spelling to sound relationships and various programs were devised.with. the intention of doing this. 5 Leonard Bloomfield (1925) became concerned that exiSting phonics programs were not based on sound linguistic principles. Our schools are conducted by persons who, from professors of education down to teachers in the classroom, know nothing of the results of linguistic science, not even the relation of writing to speech or of standard language to dialect. (Bloomfield l925:5) From the work of Bloomfield and later Fries (1963), various "linguistic" approaches emerged which were essentially phonics programs but based on scientific phonological principles. A view of reading which is based on symbol to sound relationships results in instructional methods and testing techniques designed to deal with the child's abilities to relate letter to sound. Chapter- Two, "Grapho-phonics," provides a detailed analysis of the degree of success attained by the subjects of the present study as they dealt with print to sound relationships. However, a psycholinguistic point of view, one which maintains that reading is much more than grapho-phonic processing, necessitates investigation of the function of all the language systems: phonology, syntax and semantics. Standardized reading tests do not deal adequately with reading as a total language process, but rather with fractionated particles of language taken out of context, such as phonics relationships, vocabulary, etc.' Reading comprehension tests provide short paragraphs which do usually contain language in context, however, there is no attempt to discover exactly how the reader has handled the material but only how he is able to answer a set of questions which Often deal with quantitat- ive details. 6 One widely used reading instrument, The Stanfbrd Diagnostic Reading Test, Level II, consists of the following six sections: 1) reading comprehension, 2) vocabulary, 3) syllabication, 4) sound discrimination, 5) blending, 6) rate of reading (see Appendix C for examples of test items). Test 1), reading comprehension, consists of a series of short paragraphs with blanks which students are required to fill by choosing a suitable word from a list of four. This is essentially a vocabulary. test. Test 2), vocabulary, is similar with the teacher reading a sentence and the student choosing an apprOpriate completion. Test 3), syllabi- cation, tests the student's ability to divide words into syllables according to some pre-taught arbitrary conventions established for printing. Test 4), sound discrimination, requires the student to match sounds across words. Usually the sound in the sample word is matched with the same sound in the answer word spelled in quite a different way, e.g., they, tail; jump, bridge; dinngg, pictugg. Test 5), blending, gives scrambled syllables and asks students to put them together into words. This is an extremely unnatural process and it is difficult to see its relationship to reading. 2.8081: ()0 Obb 0 pr 0 i O 11 Answer: spill Test 6) is a test of reading rate. Students are asked to read a piece of material but have normal eye movements and syntax disrupted by choices which are based on knowledge rather than reading ability. 7 Many years ago peOple thought that when North America was discovered by 0 Columbus 0 Edison 0 Washington there were wild horses here. We now know that this was not true. The Indians had always 0 often 0 never seen horses before the Spaniards arrived. NO‘U’IbWNH On line 6, the student is required to adapt one of two actions, neither of which is part of fluent reading, 1) make a choice of always, often~ never, before he knows what is to follow, or, 2) proceed to the next line and regress to find the correct word. It is not the purpose of this study to investigate in detail the characteristics of existing reading tests, however, it is necessary that the instrument chosen to describe and analyze the reading behavior of the selected subjects be one that examines reading as a total language process. When a miscue, or mismatch between the text and the oral response occurs, it is necessary to look at all aspects of the reading process to determine the way in which each is involved in the observed behavior. This the Goodman Taxonomy was specifically designed to do. The Goodman Taxonomy of'ReadingIMiscues provides a number of questions to be asked about each miscue, since the reader has, in every case, produced his response through the use of the*wide range of information available to him in the reading process. Each question is answered on its own merits and the researcher does not have to choose between possible cues and causes. (Goodman, Y 1971:3) The term "miscue" is used rather than "error" or "mistake" because these deviances from the text are not random but caused. For example, the reader may substitute a word graphically similar to the text item or make a faulty prediction based on the frequency of certain patterns and combinations. Any observed response (OR) which differs from the expected response (ER) is marked as a miscue. 8 After the student's reading has been recorded on tape, a copy of the text is marked and from this marked cOpy all the miscues are coded by working through the twenty-eight categories of the Taxonomy. Because some of the marking conventions appear in examples given throughout the present work, they are included here. Basically four types of miscues occur: substitutions, insertions, omissions and reversals. 1. Substitution - A little £338? 2. Insertion - He tied thgzesticks to the broken l_g. 3. Omission - TheLfound a £008) place to camp. 4. Reversal — "He's my deerfi, said Billy, v Sometimes readers substitute nondwords in place of words they do not recognize. These nonrwords are indicated by a dollar sign. $frawn e.g. Poor little fawn. A brief illustration of the operation of the Taxonomy is presented here by proceeding through the various categories with one sample miscue. Russell: They would spend days picking;the ripe ciggggsgiégs which theyAput in boxes and sent to the city. 46 - Correction- 0 - miscue not corrected. 47 - Dialect - O - not a dialect alternative. 48 - Graphic similarity - 7 - beginning, middle and and similar. 49 - Phonemic similarity - 7 - beginning, middle and end similar. 50 - Allolog - 0 - not an allolog. 51 - Syntactic acceptability - 4 - totally acceptable syntactically. The non-word is inflected appropriately so acceptable syntax is maintained. 52 53 55 56 57 58-59 60-61 62 63 65-69 99999 70-71 9 Semantic acceptability - O - The use of a non-word necessitates a decision that the meaning has been lost. The researcher cannot determine that the reader.has a meaning for the non- word he uses. Transformation - O - the reader has maintained the grammatical structure of the writer. Syntactic change - 9 - no syntactic change. Semantic change - blank - categories 54 and 55, syntactic' and semantic change, are coded only when the sentence is acceptable syntactically and/or semantically. Intonation - O - no intonation change. Submorphemic level - 5 - multiple minor variations. Bound morpheme level - 00 - no change. The plural inflection has remained intact. Word level - l7 - substitution of a non—word. Phrase level - O — no change. Clause level - O - no change. Blank Grammatical function - the presence of the inflectional ending indicates that the grammatical function is still noun. represents unchanged grammatical function. Semantic word relationships - blank. The semantic relationship between a real word and a nondword cannot be determined. Related research: Other studies have been conducted using the Goodman Taxonomy of' Reading.Miscues but none has compared various groups reading the same material. 10 Y. Goodman (1967) analyzed the oral reading of a group of beginning readers over a period of one year. The results of her research demon- strated that a depth study of the oral reading behavior of a group of children using a miscue analysis is highly productive of knowledge about the total language process. In general, miscues affected meaning change more than syntactic change, and they resulted in increased semantic and syntactic acceptability with time. The study by Goodman and Burke (1968) confirmed the interplay of semantic, syntactic and grapho-phonic information in the reading process of proficient fourth and fifth graders. All of the childTen seemed to have a solid control of the grammatical structures of the language and tended to correct or not correct, depending on whether or not the miscue resulted in grammatical patterns which were totally unacceptable, and a very high percentage Of their miscues produced fully acceptable grammatical patterns. Dorothy Menosky, 1971, analyzed reading behavior in various portions of text by readers in grades two, four, six, and eight. The study revealed that quantity and quality of miscueing changes as readers progress through a given piece of material. It indicated that reading becomes easier as context is developed. Some involvement in the plot provides a basis for prediction. Menosky's study has direct relevance to the construction of comprehension tests which are usually based on one or two short paragraphs. A study conducted by William Page, 1970, involved a proficient second grade reader, an average fourth grade reader and an average. sixth grade reader encountering ten basal reader selections ranging from pre-primer to sixth grade. The major finding of Page's study was 11 that as the material became more difficult, the readers processed the graphic information more accurately. Peter Rausch, 1972, conducted research to determine the influence of prior conceptual knowledge of the subject matter of the text on reading behavior. His study revealed that those readers with a clear understanding of the subject matter of a piece of material performed- much more successfully than those for whom the material was unfamiliar. The effectiveness of materials written in Black English for early reading instruction of speakers of the dialect was investigated by Rudine Sims, 1972. The study showed that the dialect material had no influence on the quality or quantity of miscues produced by the sub- jects who were selected for the similarity of their spoken dialect to that of the texts used. Rose-Marie Weber, 1970, analyzed reading-errors according to level of sounds and letters, word structure, grammatical structure, and semantic consistency. Her conclusions indicated that*the structure and meaning of language influence how children-read‘and-that children grow in their efficient use of letter-sound patterns. CHAPTER II: THE GRAPHO-PHONIC SYSTEM Introduction: Reading at its proficient best is a smooth, rapid, guessing game in which the reader samples from available language cues, using the least amount of available informa- tion to achieve his essential task of reconstructing and comprehending the writer's meaning. It can be regarded as a systematic reduction of uncertainty as the reader starts with graphic input and ends with meaning. The reader need not use all the graphic cues available in the printed page, nor is he restricted to them. As a user of language, he has both syntactic and semantic input to relate to graphic cues and interact with them. He uses graphic cues, per— haps supported by related phonological cues, to help predict grammatical sequences; he uses graphic and grammatical cues to trigger the search of his memory for related meaning; and he uses all, in turn, to predict subsequent input. He is, then, at all times utilizing three sources of information interdependently: l. Grapho-phonic information 2. Syntactic information 3. Semantic information (Goodman, K 1972:154) Although the reader is at all times making use of all three language systems, grapho-phonic, syntactic, and semantic, it is the grapho- phonic which is the most obvious. Reading is, after all, a processing of visual information. Common sense tells us that reading is a matter of converting written symbols to spoken symbols and from there to meaning. Response to print in terms of simple conversion of letter to sound is observable behavior, grammatical competence and cognitive structure are not. Consequently, reading instruction and measurement have tended to deal with that observable behavior. Any indication of reading weakness, then, implies ineffective knowledge or use of 12 13 grapho-phonic skills. Remediation programs, naturally, concentrate on the upgrading of phonics skills. Grapho-phonic proximity: The present study investigates the relative effectiveness of the the processing of grapho-phonic information by the subjects in the three groups. The Goodman Taxonomy of'ReadingIMiscues examines each substitution of a single word Observed response for a single word expected response in terms of graphic and phonemic proximity. This analysis reveals the effectiveness with which readers are making use of graphic and phonemic cues available to them. The substitutions are graded on a ten point scale increasing in similarity from 0 - no graphic or phonemic similarity to 9 - homograph or homophone (see Taxonomy, Appendix B). These categories are not coded if the miscue is an' omission or an insertion. The nine point scale may be reduced to four categories: 1) no similarity, 2) slight similarity, 3) moderate similarity, 4) high similarity. Tables 1 to 6 show the percentage of miscues falling into each of the four categories. In order to interpret the above findings more easily, means were calculated for each subject. For each group, graphic proximity is higher than phonemic proximity. These figures reflect the relationship between the phonemes of the language and the written symbols used to represent them as well as the reader's preference for graphic cueing. When Operating on graphic and phonemic cues only, the reader must move through the graphic symbols to reach the phonemic realization. For this reason, the miscues are more likely to look like the expected response rather than sound like them. 14 PERCENT OF GRAPHIC PROXIMITY GROUP 6L MHHM8HM_l_1_8_, E, _G_u§_, _t_h_i_§_, with voiceless /s/, also w_a_s_, hi3 with voiced /z/. Therefore, Egg, hi§_were classified in the NG (no generali- zation applicable) category. The generalizations involving syllables were both inconsistent and unenlightening as far as predicting pronunciation is concerned; e.g. 38. If the first vowel sound in ayword is followed by two consonants, the first syllable usually ends with the first of the two consonants. 39. If the first vowel sound in a word is followed by a sigglg consonant, that consonant usually bggins the second syllable. It seems incredible that young children would be expected to under- stand and remember such abstract generalizations. In the event the 33 children did learn such rules there is no prediction of pronunciation involved in stating how a word may be divided into syllables. The obvious conclusion to be drawn from a study of phonics and syllable generalizations is that one must be able to pronounce the word before determining which generalization applies. One page of Billy Whitemoon contains the words, move, love, drove. After pronounc- ing each of these three words one is able to state that the word drove follows #4. When there are two vowels, one of which is final 3, the first vowel is long and the g is silent. However, if the reader were to depend on rule #4, he would be wrong 67% of the time with just these three words. The same is true of stress placement in English. Consider the x / / following words: inaccurate, inconceivable, infancy. After a lesson on prefixes, a student pronounced the last word infdhcy and defined it as 'not fancy' even though the word infancy was part of his vocabulary. To examine the relationship between generalization-governed and deviant words in the text and reading behavior, the miscues of two readers from each of the study groups (6L, 2A, 6H) have been classified. Groups were organized containing rule governed words and non-rule governed words. Table 8 will illustrate the relationship of miscues to generalization-governed words. The column labelled 3953; gives the total number of substitution miscues made by each subject. 'Rglg gives the percentage of the miscues made by the subject that were made on words which were consistent with. the generalizations given. Non-rule indicates the percentage of miscues made on words which did not conform to the generalizations. Bennie, therefore, made a total of 75 word-level substitution miscues. Seventy percent of these miscues were on words which were predictable from the 34 Total Rule Non-Rule Bennie 75 70% 30% Stanley 104 60% 40% Deborah 44 63% 37% Graeme 51 52% 48% Kathy 10 100% 0% Nancy 7 67% 33% Table 8 generalizations and 30% of Bennie's miscues were on non-conforming words. If Bennie's problem in reading was with words which do not follow the rules he has been taught, it would be expected that he would have more than 45% of his miscues on words which do not follow the generaliza- tions. Interestingly, however, Bennie makes more miscues on the regular words. One might conjecture, perhaps, that this is an indication that Billy needs to learn the rules, but this would cause him to miscue 45% - of the time because only 55% of the words in Billy Whitemoon follow the generalizations given. If’Bennie were completely devout in his con- formity to the rules, he would miscue on 45% of the text, or 495 words. Bennie would have made 495 substitution miscues instead of 75 if he were true to the forty-five generalizations. There seems to be no relationship between the reading proficiency of the students and their rule or non-rule oriented miscues. The range of miscues in the non-rule words for the 6L group is from 30% to 40% while the range in the 2A group is from 37% to 48%.. This appears to be a trend toward better performance on the rule-governed words but the highly proficient group invalidates that trend. One hundred percent of Kathy's substitution miscues were made on 35 rule-governed words. The number of miscues in this group is so small that the statistics are not very meaningful but it does indicate that the more proficient readers are not operating on the phonics generali- zations more proficiently. Nancy (6H) looks more like Bennie (6L) than anyone else. It is obvious that the Spelling regularity or irregularity has no relationship to the miscueing behavior of the subjects in the study. What, then, does cause students to miscue? The first ten miscues made by Bennie, Deborah, and Kathy are described below with suggested causes. Bennie: Winnebago - was omitted. This word appears on the list of words for which no generalization was given. There is no generalization to predict stress. Other than that, quite a regular graphic-sound relationship exists. Bennie probably skipped it because it was a long word and because it was not a word for which he had a concept. It was not in his oral vocabulary. Most of the readers had problems with this word. Qghig - is an easy word to read. Bennie substituted £322. The text read, Billy lived with his father and mother in a cabin near the river. Very likely Bennie's concept of Indians has them living in teepees not cabins. Perhaps, also, he thinks of camping near the river. It is Obvious that Bennie is processing meaning here and has translated his output into something more in keeping with his conceptual experience. .EiEEEu‘ Bennie substituted ligk for $2539. This shows some probable secondary dialect involvement as Bennie's realization of licked would be lick in most cases. There is not enough context to help Bennie 36 choose between likgg and licked. He does retain the identical gram- matical function. In subsequent occurrences Bennie reads likgd correctly. The context allows him to succeed. In the next line, One of the things Billy liked most..., the word likgd is highly pre- dictable and is read correctly. $5223 - All the men and women and children of the tribe wentfigo the cranberry swamp .... The word Eggpg is obviously not in Bennie's vocabulary. It occurs several times in the story and Bennie never is able to read it. He omits the first occurrence, next he tries §££ih_ which is very close grapho-phonically, next he tries §trawbel. Obviously. his problem is experiential and his apparent attempts at following phonics rules do not bridge the gap. Cranbergy - This word was placed on the non-rule governed list because of the post-vocalicIE. However, Bennie read the 'pg££y_morpheme with ease. Nearly all the inner-city black children miscued on cranberry. Obviously it is not a word in their active vocabularies. Cranberries are rather a white American custom. Picking - Because of the problem with cranberr , Bennie omitted the next word picking. The activity of cranberry picking meant nothing to him. Two lines later, the word picking occurs again and Bennie reads it successfully. 35525 - see Eggpg above. Cranberries - This time Bennie tries cackleberries. He is still not correct but the word probably sounds more familiar to him. There is a breakfast cereal called cackleberries and a joking reference to eggs as cackleberries is often made. 37 Syggp_- Another word outside the conceptual experience of the pavement- oriented inner-city child. Bennie omitted it. It should be easy to read. Eigy — Bennie substituted country, a very interesting miscue. ‘ngy follows the phonics generalizations while country does not. Country is a longer word and as such may appear to be more.difficult. The semantic features on both words are identical except for one feature @rban) perhaps. Point of view is obviously operating here. Bennie lives in the city. The story has the cackleberries (cereal from a factory?) put into boxes and sent to the city. Since Bennie is in the city and they are being sent not brought, Bennie sends them to the country. A word attack problem? Obviously again, a conceptual problem. Deborah - 2A Winnebago - Deborah substitutes $Winebugg1e. It is obvious she is working quite efficiently with graphic cues. Her problem, like Bennie's, is that Winnebago is not in her vocabulary. Hip - Deborah omitted the word hip in this case. Obviously a high average reader like Deborah is not having trouble with an easy word like hi3. She reads it correctly every other time it appears in the text. She is anticipating the following word Epipg_which she does not know. Anticipation sometimes causes regression in order to 'take a run at it.' Sometimes it causes additional words to be missed as happened in this case. IElRE.’ Deborah's attempt was $trible. Like Bennie, Deborah is unfamiliar with Indians and tribes. 38 Cranbergy - Deborah substituted the non-word $carabergy.- She is another of the inner-city children who have not had experience with cranberries. ‘Iglpg - This time she comes slightly closer grapho-phonically with the non-word‘§£££p but she is still not processing meaning. Deborah never does read Epipg. Sgggp_- Again a conceptual experiential problem as Deborah does an excellent job of handling graphic cues. She produces EEEE;+ p. Rips - Deborah should be able to read this word. It is likely in her vocabulary. However, here again, it is like hyg. She miscues on ripe because it comes before cranberries which she produces as $canberries this time. If she were orally familiar with the word cranberries, and arrived at as close a pronunciation as $canberries she would have recognized the word and produced it correctly. Sent - Deborah substituted set. The text read ...which they put in boxes and sent to the city. Deborah was likely predicting something like ...and set on the table. As soon as she encountered the next word, 53, she realized she had misread and corrected §g£_to £325. .Thg - Deborah substituted hp. Again one would suspect a good reader such as Deborah to know the difference between Egg and hp and she does. This is the only time that this particular substitution occurs. Discourse analysis is required to find the cause for this miscue. The previous sentence begins, When he wasn't in school .... The sentence in question continues but (indicating a compound sentence often con- taining parallel constructions) when the heavy snows.... Deborah was predicting a similar pattern. 39 Heayy - This word is involved with the previous miscue in a complex miscue. After producing But when he it was impossible to use heayy. But when he heayy would offend Deborah's excellent grammatical sense. So while the first miscue, hg_for Ehg, was caused by a faulty predic- tion built on a structural sense, the second having for hggyy arises from a need to maintain a possible grammatical sequence. Neither of these miscues would be eliminated by drill in phonics. Kathy 12'- Kathy substituted Ii due to a prediction that a small word follow- ing Ehgpg and beginning with 3 is likely to be ig. Actually the probability of her guess being correct is very high. As soon as Kathy saw the following phrase she corrected. There in the dead damp leaves.... Went - Here Kathy is using her creativity and dramatic sense to supple- ment her semantic and grammatical strengths. The story concerns the finding of a fawn by a young Indian boy. The text read, Billy went closer (to the fawn). Kathy transformed it to the more colorful Billy_ crept closer. Absolutely no similarity grapho-phonically, yet semantically and grammatically an excellent substitution. That - He was surprised that the little fawn didn't run away becomes He was sugprised when the little fawn didn't run awgy. Again a perfectly acceptable structure semantically and grammatically with no grapho- phonic similarity. Kathy could not be said to be having trouble process- ing graphic information. Tpgp - Kathy substitutes yhgp. This is a very natural thing to do as both words are clause markers and thus fill the-same grammatical function. They look and sound alike differing in only one phoneme and one grapheme and have some semantic similarity both referring to time. 40 This - Then he noticed that this one's lgg... becomes Then he noticed that his, correction. Kathy predicts a likely word hlg but corrects when she sees the following word. His one's is not grammatically acceptable. £222.‘ Again Kathy substitutes yhgp, As Ehgp introduced an independent clause and ghgp is a dependent clause marker, Kathy put two sentences together to create one which was grammatically acceptable. Then Billy and his father built a summer house. Thgy covered it with deer hides... became When Billy and his father built a summer house they covered it with deer hides.... Her miscue was semantically acceptable and she made it grammatically acceptable by using a conjoining transformation. Would - The original sentence, Lightfoot was so much bigger now that the hunters would surely shoot him'became Lightfoot was soypuch bigger now that the hunters could surely shoot him. Again semantically and grammatically acceptable. Kathy is a highly proficient reader making thesame types of miscues that adults do. ‘Kpg! - Kathy substituted looked. A proficient reader such as Kathy. in not having difficulty distinguishing between these two words and a phonics program is not indicated. The text was Billy knew how he could... and Kathy read Billy looked then corrected; The context led her to expect looked conceptually. It fit grammatically but the next word how broke English co-occurrence restrictions and forced Kathy to correct. What - The original sentence was: Billy was so_p1eased by the hunter's words that he told his father and mother what had happened. Kathy read: Billy was so pleased by the hunter's words that he told his father and mother that he and then corrected. Again she is predicting, an important 41 part of the reading process. When her prediction is tested and fails, she regresses and corrects. The above examples indicate two main causes of miscue behavior, neither of which is phonics failure. 1. Students miscue on words for which they have no concept, e.g., cranberry, swamp. The word rustling occurs three times in Billy Whitemoon. The children familiar with the word orally had no trouble producing it from the written text. Those who were unfamiliar with it either skipped it or resorted to phonics. Those who 'sounded-out' tried rusting, rusty, resting. They did not have any information about the silent 't'. One line was particularly interesting: Bllly_ feasted on roast corn and baked fish. Each word follows regular spelling to sound rules, yet nearly every one of the average and weak readers had problems with that sentence. Roast corn and baked fish are obviously outside their experience, also the concept of feasting. Peter Rausch (1972) has done a study where he pre-tested students on the concepts contained in a particular story before they read it. He divided them into two groups, the high and low concept groups. He discovered that the high- concept group did significantly better in terms of semantically acceptable miscues than the low concept group. These students were controlled for I.Q. and had varied standardized reading test scores. He found that their knowledge of the concepts was more predictive of their success in reading performance than were their test scores. In other words, a student with 42 a high concept rating but a low standardized test score per- formed better than a reader with the Opposite characteristics. 2. Students miscue when they predict unsuccessfully. All readers make predictions. These predictions are tested and either confirmed or rejected. A good reader will correct a faulty prediction. (See Chapter V.) The phonics generalizations are inconsistent and non-predictive. Students' miscueing behavior is not determined by knowledge or lack of knowledge of the generalizations. All readers in the present study are arriving at high grapho-phonic proximity probably on the basis.of generalizations developed themselves on the basis of their experience with reading. CHAPTER III: THE SYNTACTIC SYSTEM Introduction: Old insights about reading based on an over-emphasis. on words must be carefully reconsidered as the view of words is placed in prOper perspective. At the same time, new concerns are emerging whose significance was previously overlooked or only dimly seen. Grammar, as the system of language, emerges as one such colossal oversight. Whenever any language user attempts to derive meaning from language he must treat it as gram- matical sequences, and be aware of grammatical interdependencies. This is true when a reader deals with a simple sequence like Tom saw Betty. He must know that 329 is subject and BettyI is object in order to comprehend. In a much more complex sequence, such as See Flip run, he must be aware that the subject ygp is not present inIthe surface structure; that Flip run is an embedding Of an underlying structure, ygg, see (Flip runs), and that the clause functions as the object of the verb 233. If he cannot process this information, he will not comprehend the message See Flipyrun. Both examples are three word sentences. The task of reading each sentence depends largely on the processing of grammatical information. Thus, when viewed from a psycholinguistic base, what has appeared to be a word recognition problem is a very dif- ferent phenomenon. (Goodman, K. 1972:144) The beginning reader has already mastered the syntax of spokenx language. Even though teachers often feel that they teach language there is very little a teacher can do to change the grammatical system a child has developed by the age of five or six. The task of the reading teacher is to help the student to use his grammatical competence in. reading since written language makes use of the same grammatical rules as spoken language. There are some stylistic differences, of course, but these are largely surface level options. The syntactic information possessed by all users of language consists of: 43 44 A. Sentence Patterns: The grammatical sequences and inter- relationships of language. The 3 ed the s, is an example of a sentence pattern common in English. B. Pattern Markers: The markers which outline the patterns. 1. Function Words: Those very frequent words which, though themselves relatively without definable meaning, signal the grammatical function of the other elements. Examples; the, was, not, do, in, very, why, but. 2. Inflections: Those bound morphemes (affixes) which convey basically grammatical information. Examples; ing,‘gd,.§. 3. Punctuation - Intonation: The system of markings and space distribution and the related intonation patterns. Pitch and stress variations and variable pauses in speech are represented to some extent by punctuation in writing. C. Transformational Rules: These are not characteristic of the graphic input itself, but are supplied by the reader in response to what he perceives as its surface structure. They carry him to the deep structure and meaning. If he is to recognize and derive meaning from a graphic pattern, he must bring these grammatical rules into the process. (Goodman and Niles 1970:15) Syntactic acceptability: The present study investigates the degree of success of the various subjects in dealing with the grammatical structure of Billy Whitemoon. Table 9 shows the degree to which each subject's miscues disrupted the syntax of the text. There are five categories: 0 - Unacceptable - Miscues have resulted in a sentence totally unacceptable grammatically. e.g. Bennie: When he wasn't in school, hefskatedjwith u his friends on the river ice . The omission of the verb results in a completely unacceptable structure. 45 SYNTACTIC ACCEPTABILITY muthe_youggptrees into lqufstriggg. The deep structure of both sentences consists of the pronoun .ghg plus the past tense of the verb pgppd plus a direct object. Leroy uses a different transformational rule to derive past tense. As a speaker of Black English the reali- zation of the past tense for Leroy contains a zero inflection. In the bound morpheme category (see p. 78) such a miscue is coded as a substitution of one inflection (O) for another, (/ad/). 3 - The reader anticipates the same deep structure as the author, but selects an alternate surface structure which is available through the use of Optional transformations. e.g. Nancy: Billy was very happy, He knewfthzflspring had U seas... The deep structure contains that but an Optional transformation in standard English provides for the deletion of phat between two complete sentences. 4 - The fourth category is reserved for miscues in which the structure is so badly garbled it is impossible to ascertain- the transformational process because the deep structure is lost to the researcher. e.g. Stanley: Billy knew how he could prove Lightfoot wasv his.--€>2But know how he Lightfoot was his. The transformation miscues of one reader from each group will be examined to see if any differences may be seen. 60 6L Bennie (188): Bennie made a total of 77 miscues which involved re-transformation Of the original deep structure. Twenty-three of his transformation miscues resulted in different deep structures. e.g. There in the dry,(§§§§)leaves he saw a little fawn. The omission Of dead_changes the deep structure as the adjective represents an underlying clause which is deleted. An even larger number (31) of miscues were ones in which dialect involvement took Bennie through different rules. e.g. She made her ownpaianDfrom the rooqglthat Billy gatheréa from the swampé} This one sentence contains four miscues in-which zero morphemes are substituted for the inflectional morphemes of the text. Since these are consistent with Bennie's grammatical and semantic systems, they-are not considered tO be disruptive in any way. Bennie did not make any alternate Optional transformations. This is revealing. In order to re-structure successfully, the reader must correctly infer the meaning and deep structure Of the author but proceed by choosing an Optional transformational rule. Bennie is not that proficient a reader.~ In 24.72 of cases he changes deep structure and 24.7% Of his miscues cause the deep structure to be lost completely. 2A Graeme (121): Graeme made a total Of 41 transformation mis- cues. Of these, 25 represent changes in deep structure. This total is higher than Bennie's and since Graeme had fewer miscues, the percent- .age.is much higher, 24.7% to 45.5% for Bennie and Graeme, respectively. Transformations coded 1, although involving a change in deep structure, ch: retain a syntax which is acceptable or at least acceptable with the portion of the sentence preceding or following it. The revealing 61 category here, is category 4, in which the deep structure is completely lost. Graeme had only 10 such miscues whereas Bennie had 23. Graeme's completely disrupting miscues, while much lower than Bennie's 23, represent the highest number in this category in the 2A group. Graeme has 5 dialect related transformations. He uses one alternate Option, category 3. 6H Nancy (602); Nancy made a total of fourteen re—transformations. Of these, 47.1% did change deep structure. However, none of Nancy's miscues resulted in a complete loss of deep structure. Although the members Of this group were all white middle class students whose dialect matched very closely that of the text, Nancy did make one dialect miscue. deers e.g. All deer look alike. This was coded g_idiolect. Nancy had 29.42 Of her miscues in the Optional transformation category. She is handling the meaning well and processing deep structures. This allows her to be free in her use Of Optional transformations. Tables 13, 14 and 15 show the raw scores and percentages for the subjects across the five transformation categories. Figure 8 presents the mean percentages in each transformation category in graphic form. Conclusions: 1. All subjects made more miscues which involved transformations. than miscues which did not. 2. The 6H group made a higher percentage Of transformations which changed deep structure than did the other two groups. 62 smog MMDHUDmHm mmma onHmo MH [grizi/ /tes/ test ——-> /test/ All readers pronounce written English in a manner consistent with. their dialect. NO American reading a British novel reads with British Received Pronunciation. In fact, an attempt to duplicate the pronunciation 68 ¢.~ 0.0 0.0 0.0 m.n In TVNINHHL ox HONELNHS o.nm o.o0H c.00H c.00H H.ao>zH aa>ao>zH Hoz £.1~ Huafiaam 28 62 188 0.0 3.2 0.0 30.1 66.7 43 189 0.0 1.9 0.0 15.4 7 82. 21 94 190 0.8 3.3 0.0 5 17. 78.3 107 191 0.0 1.7 0.0 5.2 93.0 24 82 192 0.0 6.2 0.0 21.2 72.6 17.9 0.0 3.2 0.2 78.7 MEAN I Table 17 72 Hamaaaoa mamazooam HoaaoHaH DIALECT GROUP 2A aa>ao>zH aa>ao>zH 902 area Hoahaam 49 121 0.0 0.0 0.0 10.9 89.1 75 122 0.0 0.0 0.0 0.0 100.0 47 123 0.0 2.0 0.0 4.0 94.0 47 124 0.0 2.0 0.0 2.0 95.9 15 31 125 2.1 2.1 0.0 31.3 64.6 1.2 0.4’ 0.0 9.6 88.7 MEAN 2 Table 18 73 aamfiaaoa wmHo>zH aa>ao>ZH 802 #z Hoahaam 17 601 0.0 0.0 0.0 0.0 100.0 16 602 0.0 0.0 5.9 0.0 94.1 17 603 0x0 0.0 0.0 0.0 100.0 at, 10 604 0.0 0.0 0.0 0.0 100.0 34 605 0.0 0.0. 0.0 0.0 100.0 0.0 0.0 1.2 0.0 98.8 MEAN 2 Table 19 74 The 2A group had a mean dialect involvement Of 11.3% while the mean for the 6L group was 24.88%. There seems to be little correla- tion between dialect involvement and comprehending score. Table 20 and Figure 9 include data from groups 6L and 2A. The dialect category is not relevant to the 6H group. The scattergram, Figure 9, shows the random nature of the relationship. In the 2A group, Cheryl (125), the student with the highest degree of dialect involvement,obtained the highest comprehending score, but this relationship is not general, across subjects. 1. Phonological dialect variations occur between any reader and a given text. The goal Of reading is to Obtain meaning and this is best accomplished if the reader uses his own pronunciation system. In silent reading we do not know what, dialect pronunciation the reader uses, if any. 2. Grammatical dialect variation in the present study was limited almost entirely to morphological features. The Black dialect speakers made use of 0 plural, 0 possessive, and 0 past tense morphemes in alternation with standard English morphology. There was an occasional use of a double negative construction but this was the only structural change that went beyond the morphological level that could be attributed to dialect. Thus, even grammatical dialect features appear to be Of little significance. 3. The L group had more dialect influenced miscues than did the A group. However, there was no relationship between dialect involvement and comprehending scores within groups. 'Bennie (188) had the highest degree Of dialect involvement in the L group and the second highest comprehending score, while SUBJECT 188 188 190 191 192 HOML“>HU 40 30 20 10 75 DIALECT / COMPREHENDING RELATIONSHIP U 0 Z Z S H a E H in [-4 H LT: U a U U a [:1 m a a 3 E a 3 E Q U U) Q O 33.3 37.6 121 11.1 40.7 17.3 46.2 122 0.0 45.3 21.7 17.5 123 5.9 43.1 7.0 33.9 124 4.1 61.2 27.4 29.2 125 35.4 75.0 Table 20 X X X X X X ix 5 10 15 20 25 30 35 40 45 50 55 60 COMPREHENDING Figure 9 65 76 Stanley (191) had the lowest dialect involvement with the third lowest comprehending score. It would be too simplistic, then, to claim that dialect divergence inhibits effective reading. The question of why the weaker group had the higher degree of dialect involvement is one which bears investigation. It could be that teacher evaluation plus standardized reading tests, both of which are biased toward the standard dialect, have effectively, over a period of six years, sifted these children to the bottom track.and the children now match the lower expectations. 4. There seems to be no indication of a need for special dialect materials. The second grade group, all of whom live in inner-city Detroit, are processing a third grade story written in standard English quite effectively. The sixth graders are not so successful but there is no indication that dialect is responsible for their weakness. On the other hand, both inner-city groups had difficulty with the cultural aspects of the story. Their miscues indicated that they do not have the conceptual framework to handle Indians, wild- life and other concepts foreign to their experience. Any change in material should focus on experiential familiarity rather than dialect. Structural levels: The next four categories involve four structural levels: morpho- logical, word, phrase, and clause levels. The Goodman Taxonomy of'Reading.Miscues enables analysis of the specific structural levels involved in each miscue. Miscues may involve 77 morphological, word, phrase and clause level changes. There is a high degree of interrelationship between the categories. An omission or insertion at one level is usually a substitution at another. deers e.g. All deer look alike. The miscue consists of an insertion at the morphological level and a substitution at the word level. e.g. They drove until they found a ggod place to camp. The omission of good is coded as an omission at the word level, a sub- stitution at the phrase level - a place for a_good place - and an. omission at the clause level. Adjectives are considered to be derived from embedded clauses. e.g. They found a place. The place wasygood. Morphology: Morphology is handled by the Goodman Taxonomy of'Reading.Miscues in two categories. The first deals with process: substitution, insertion, omission and reversal of bound morphemes. The second determines the type of morpheme involved: inflectional, derivational, contraction, etc. Any change at the morphological level requires analysis in both sub-categories. e.g. He livqfilwith his father and mother. Substitution of inflectional affix paints She made paint. Insertion of inflectional affix beautiful What a beauty. Insertion of derivational affix 78 He packed carefu® Omission of derivational affix Tables 21, 22 and 23 present the number and percent of miscues involving various types of morpheme change. They also present two columns of data concerning the type Of morphemes involved in the given change. Of these, column one shows the number and percent of miscues in which morphology was not involved.’ Column two gives the number and percent of miscues involving inflectional endings. Since. these two columns comprise at least 90% of miscues in most cases, the other types of morphemes involved have not been included in the table. The tables indicate that morphology is not a severe problem for any of the readers. A mean of 69.1% of the miscues made by the 6L group did not involve the bound morpheme level. This mean increased to 74.12 and 94.12 for the A and H groups, respectively. The H group was nearly free of bound morpheme miscues. Of the possible changes, substitution accounted for the highest. number of bound morpheme miscues. The L group had a total of 108 substitution miscues at this level. The A had 40 and the H only 4. This category relates very directly to dialect. Dialect alternatives with 0 morpheme inflections were coded as substitutions of inflectional morphemes. If all dialect substitutions were removed, the figures for the 6L group would look much more like those of the other two groups. Since dialect seems not to be a major concern, it appears that morphology is not a problem for readers. Word level: Most miscues involve the word level in some way. The high degree of involvement at this level causes most reading instruction to center 79 «.mu n.mN mm N.NH on o.o~ am «.mH 0H N.Hm Ch N NOIIOHTJNI «.mo o.co mu «.mn Hm m.nn Hm «.mo an «.00 on ION ORATOANI amNH azaamaoz «.0 0.0 o.o o.o ¢.H o.o TVSXHAHX HN OHAoH m.¢ n.m w.n ¢.m m.n ~.m NOISSINO aw macaw Nooaoammaz o.m .NeN m.¢ N.¢ m.n H.H H NOIIXHSNI aoz baskets 4 - A single morpheme word is substituted for a multiple morpheme word. e.g. sticks-——> .EEESE 5 - A word in a larger word is substituted. e.g. cranberries-4> cackleberries 6 - A word in a compound is substituted. e.g. summggyhouse-%> summertime 7 - A non-word is substituted for a real word. e.g. ‘fggpg—4> §frawn 8 - A dialect alternative is used. ’ e-s- lam—9 .1392, The choice between 8 and 4 depends largely on the frequency of dialect responses in the individual's reading. 83 Tables 24, 25 and 26 itemize the number and percentage of miscues involving no change, substitution, insertion, omission and reversal at the word level. Tables 27, 28 and 29 indicate the free morpheme types involved in the changes. The data show: 1. Most miscues involving words were substitution miscues. Both the L and A groups had 84.6% substitution miscues at this level, the H group only 47.82. The biggest difference between groups was in the insertion category with L - 1.1%, A - 5.72 and H - 21.1%. The inser- tions are related to the transformation category. The H readers made a larger percentage of this type because of their application of optional transformation which allowed for the insertion or deletion of Optional elements. 0f the two lower groups, omissions were much more frequent in. 6L. They omitted 57 words compared with 22 for the A group. These weak readers omitted the words they did not know rather than attempt to predict. Most often, these omissions resulted in severe damage to the syntax. e.g.' Bennie: All the men and women and children of the $carberry .yent to theycranberg- near the ~ Winnebago Ian. The H group made 14 omissions for a mean of 14.7%. They made a high number of omissions for a very different reason from that of the L group. Omissions, like insertions, for the H group are related to optional transformations. The words 84 FREE MORPHEMES GROUP 6L Haa onmmHzo onHaamzH onHaHHHmaam aozaa onmmHZo onHaamzH ZOHHDHHHmaDm aazaa ZOHmmHZO onHaamzH ZOHHDHHHmaDm auz "Poor little fawn!" Billy said. 91 Tables 30, 31 and 32 present the number and percentage of phrase level miscues made by the subjects. Figure 10 shows the relationships graphically. 1. The H group had a significantly higher percentage of miscues at the phrase level. These figures reflect the involvement of the proficient readers in units larger than the word. They go directly to deep structure, extract meaning, and restructure. The higher frequency of optional transforma- tions is reflected again at this level. The H group made a larger percentage of phrase level substitutions than the other groups. Many of the omissions and insertions at the word level are substitutions at the. phrase level. e.g. Billy took the sticks off.-—4>- Billy took the_two sticks off. They drove until they found a(good)place to cagp. W White men from the cities came to hunt in the forest near(TEEDWinnebago land. 1‘_y, Most of these miscues for the H group changed meaning very little. The larger context of the discourse allowed such miscues. In the first example, the insertion of £33 is natural as the paragraph has supplied the information that it was two sticks that Billy used. The 2A group are handling a larger percentage of structures beyond the word than the L group are. They are on their way to becoming proficient readers. It must be kept in mind that they are only second graders dealing with a story at least a 92 Aaa onmmHzo onHaamzH PHRASE GROUP 6L ZOHHDHHHmaDm auzaa onmmHzo ZOHHaamZH PHRASE GROUP 2A ZOHHDHHHmaam aozaa onmmHZO ZOHHaamZH PHRASE GROUP 6H onHaHHHmaam aozaa do we onmmHSo aa aw "Poor little fawn!" Billy said. He picked up the fawn.-—4> He picked the fawn up. One of the things he liked most was cranberry picking. '-4> One of the things he liked most was picking cranberries. Clause level: Clause level miscues are handled in six taxonomy subécategories. 0 - NO change - the miscue does not affect the clause level. 1 - Substitution - one clause is substituted for another. e.g.- Kurt: Next year when the Winnebggo_Dance Time came, Billy sang for all the tribe. Nextyyear when Billy came to the Winnebago Dance Timel he sang for all the tribe. 2 - Insertion - a deep structure clause is inserted. little e.g. Kurt: Billy knew thatflfawns were always very shy. 3 - Omission - a deep structure clause is omitted. e.g. Kathy: He was surprised that theflIEEI23fawn.didn't‘ V run away. The interrelationship of the various levels may be seen in the- insertion and omission categories above. The insertion of little, is 97 an insertion at the word level, a substitution at the phrase level, and an insertion at the clause level. 4 - Clause dependency is changed within the sentence. e.g. Deborah: Billy wished he could singysome of the and songsdhe was always making up. The dependent clause introduced by the deep structure clause marker that is lost by the replacement of and. 5 - Clause dependency is changed between sentences. e.g. Bennie: Then he and the fawn would race together through the forest. Billy named his_pet. Lightfoot because he could run so fast.--> Then he and the fawn would race together. Through the forest Billy named his_pgt Lightfoot because he could run so fast. The data indicate that there is not much difference between groups at the clause level (Tables 33, 34 and 35). While the H group had slightly more clause level change than the other two groups - 22.8% vs. 21.2% and 17.1% - the difference is not so great as at the phrase level. Change at the clause level, as at the phrase level, represents the processing of larger units than the word. The proficient readers are more adept at handling the larger units than are the weaker readers who are still concerned with identifying words. However, the Taxonomy deals with clauses at the deep structure level whereas phrases represent surface structure phenomena. In the transformation category it was seen that the H group made Optional transformations at the surface level. The deep structure level was not involved in this type of transformation. 98 maozaHzam zaa3Haa wuzaazamaa auzafizam zHaHH3 wozmnzmmmn onmmHzo L E 6 S u w m o c m zogmmmzm onaayoommam muz The omission of the last word prohibits the miscue from being acceptable with prior or with after. 106 SEMANTIC ACCEPTABILITY ammmmdm zH aaa .figg Sequential association, the relationship between two words is sequential - they "go togehter." e.g. He spoke carefully-—— He epoke clearly Shift to generic from specific. e.g. winter-—%> 'weather Common attribute or confusion between characters. e.g. leaping'-€> limping Antonyms. e.g. summer-€> winter Semantic pair. e.g. father——) mother Variant form of the same word, inflectional or derivational. e.g. carefully-—%> careful Similar name e.g. Whitemoon-—%> Whitman. Synonym within the text. house e.g. On theig way back to their winter home. Synonyms in other texts but not in this text. e.g. They (guilt One for Lightfoot, too. Some semantic association. e.g. Poor littleypet --9' Poor little friend Tables 41, 42 and 43 show the percentage of miscues falling into each sub-category for individual subjects. Figure 15 shows the group means for the various sub-categories of semantic relationship. The low group were much higher in miscues which were completely unrelated semantically. In all of the sub-categories where the miscue SEMANTIC RELATIONSHIPS, GROUP 6L NOIIVIOOSSV EROS SIXHINOD XHHIO NI NANONAS 1X31 NHL NIHIIM WANONAS HNVN HVTINIS W805 INVIUVA XIVd NI NHHIO NANOINV EIHHIHIIV NOWWOD OIdIDZdS NOHJ OIHHNHO OI NOIIVIOOSSV TVIINHHOHS JIHSNOIIVTHH OIIOVINAS OHIVTHHNH ass: IDHFHHS 26 188 0.0 0.0 2.4 0.0 19.5 7.3 0.0 0.0 7.3 0.0 0.0 63.4 16 189 120 14.3 0.0 0.0 2.9 0.0 0.0 25.7 11.4 0.0 0.0 0.0 45.7 47 190 5.9 0.0 0.0 0.0 1.5 1.5 11.8 2.9 1.5 0.0 5.9 69.1 13 59 191 8.7 0.0 0.0 1.1 2.2 0.0 14.1 2.2 2.2 1.1 4.3 64.1 10 44 192 0.0 0.0 3.0 4.5 0.0 15.2 4.5 1.5 0.0 1.5 3.0 66.7 0.0 0.0 1.4 2.1 0.3 17.3 5.3 1.0 0.2 3.8 6.4 61.8 MEAN Z' Table 41 SEMANTIC RELATIONSHIPS, GROUP 2A NOIIVIOOSSV EROS SIXEINOD EHHIO NI WANONLS 1X31 3H1 NIHIIM NANONAS HNVN HVTIWIS H301 INVINVA HIVd NI NHHIO NANOINV HIflHIHIIV NONNOD OIJIOEJS NONE OIXHNHO OI NOIIVI oossv 'mmz-mbss dIHSNOIIVTHH DIIOVINAS GHIVTHHNH ease 133F808 27 121 0.0 0.0 0.0 0.0 0.0 19.0 4.8 2.4 0.0 2.4 7.1 64.3 11 37 122 121 1.8 0.0 3.5 0.0 1.8 0.0 19.3 3.5 1.8 1.8 1.8 64.9 24 123 11.1 0.0 0.0 0.0 2.8 0.0 16.7 2.8 0.0 0.0 0.0 66.7 12 124 0.0 0.0 2.9 2.9 26.5 8.8 0.0 0.0 0.0 2.9 20.6 35.3 14 125 16.0 0.0 0.0 4.0 20.0 4.0 0.0' 0.0 0.0 0:0 ‘0.0 56.0 8.9 0.6 .0.7 0.0 '1.3 0.6 20.3 "7.2 0.8 0.4 1.6 57.4 Table 42 SEMANTIC RELATIONSHIPS, GROUP 6H NOILVIDOSSV EROS SIXHINOD HHHIO NI RANONAS 1X31 NHL NIHIIM RLNONAS HRVN HVTIRIS RHOJ INVIHVA HIVJ NI HHHIO RLNOINV HIHHIHIIV NORROD OIHIOHdS ROHJ OIHENHO 0L NOIIVIDOSSV TVIINEHOES dIHSNOIIVTHX OIIOVINAS OHIVTHHNO *3! IDHFHHS 601 0.0 0.0 10.0 0.0 0.0 0.0 0.0 30.0 0.0 0.0 20.0 40.0 602 122 0.0 0.0 20.0 0.0 0.0 -0.0 0.0 0.0 40.0 0.0 0.0 40.0 603 50.0 0.0 0.0 0.0 0.0 0.0 16.7 0.0 16.7 0.0 0.0 16.7 604 20.0 20.0 0.0 0.0 0.0 0.0 0.0 0.0 40.0 0.0 0.0 20.0 605 33.3 "0.0 0.0 4.8 0:0 0.0 23.8 4.8 4.8 0.0 0.0 28.6 14 0.0 1.0 0;0 0.0 22.1 9.0 10.3 0.0 ' 0:0 14.7 29.1 Table 43 ma shaman onadHoomm< aZOO HNaH zH zwzozwm mafia amaHZHw azaom Hz UHHO cottage semis-9 5292 g y -€> scared baggage —) packages winter-—€> weather caught sight of-€> seen Sometimes the readers substituted words generally called antonyms. It is difficult to define antonym. If a semantic feature analysis is made of such words it is usually found that they differ in only one feature. e.g. winter summer <+season +seaso€> <-cold> city countpy _ “r r. l <+geographic> +g eographic) location location high concen- -high concen- <:+ tration of < tration of humans humans 1— _ — _4 125 mother father +human; human +adu1t adult (+parent) +parent <+fema1e> -female _ -1 L A This type of substitution also shows the reader has dealt with meaning. He has merely switched one feature. Conclusions: 1. 3. The semantic category is the most important for two major reasons: (a) Reading must result in meaning or it is not reading. (b) It is largely through the use of the semantic system that the visual information is processed. The differences between the proficient readers and the weak readers are most pronounced in this category. The average readers are progressing in this direction. The task of the novice reader is to learn to go from deep structure to the surface, to predict from meaning the visual configuration. (Smith, F. 1971:222) 4. A comparison of comprehending and grapho-phonic data shows that the weaker readers are relying much too heavily on grapho- phonic information. The proficient readers are using their semantic system to predict the visual configuration. There- fore, their graphic predictions are not always entirely accurate but the meaning is preserved. The highly proficient readers are freer with syntax than the slower readers but much more faithful to meaning. CHAPTER V: CORRECTION STRATEGY Introduction: Reading is a process involving sampling and predicting. The reader is able to predict due to his grammatical ability and because of the meaning symbolized by the text. There is a certain degree of redundancy in language which allows the reader or listener to sample rather than to attend to every available signal. e.g. Those three boys were stealing cars. There are four cues in the above sentence which indicate that the subject is plural: ppoee - a plural determiner; pppee - a number more than one; PQZE,' the e plural inflection; yepe - the plural form of the verb. It is not necessary for the reader to use all four signals. After the first one or two which he processes he is able to predict to others. Elements of meaning present as the plot of the story evolves pro- vide semantic redundancy. e.g. Billy Whitemoon was a Winnebego Indian. As the story unfolds the author is able to say, "The Winnebagps feasted... The reader knows the Winnebagos are Indians. Correction data: Most readers are successful in their predictions most of the time. Even Leroy who made 19.3 miscues per hundred words is successful with 80% of his reading. The more proficient the reader, the better his predictions will be. 126 127 If any of the reader's predictions are demonstrated to be incorrect by further reading, he can regress for additional cues and correct his original reaponse. All readers in the study did this. e.g. Bennie: Soon he returned with two straight sticks and strong some striug. In this case Bennie's grammatical ability enabled him to see that his prediction of strong after EEEE.W33 faulty as the sentence termination followed. He had predicted strong on the basis of the previous quantifier - adjective - noun structure two straight sticks, but when he encountered the period he regressed and corrected. e.g. Bennie: One spring day Billy was 5§1§1g§_through the woods. In this case“ Bennie's miscue wasperfectly grammatical. He substituted a present participle for a present participle and did nothing to change the syntax. However, semantically, talkipg_through the woods, while possible, is rather unlikely. Bennie's semantic sense caused him to go and pick up the cues more accurately and correct. The fact that he did not have to orally finish the sentence before discovering his miscue indicates how far ahead Bennie was processing. Tables 44, 45 and 46 are tables of the percentages of miscues corrected by the subjects in the three groups. Column one shows the percentage of miscues left without any attempt at correction. Column two shows the percentage of successful corrections by each subject. In column three are the percentages of miscues in which the initial response of the reader was correct but the reader abandoned his correct response in favor of an incorrect one. Column four gives the percentage of miscues on which an unsuccessful attempt at correction was made. 128 PERCENT OF MISCUES CORRECTED GROUP 6L mZOHHUaaaoo Hammmaooawza Hoaaaoo mzoaz\\\\\\\\\ l W W Ill \\\V _ [LII k\\\>\\\\\\\fi - vl [ l l l _E If 100 O O\ O 00 O C.) O O O O 00 r\ «3 Ln <2- m N H 6L 2A 6H 6L 2A 6H 6L 2A 6H 6L 2A 6H UNACCEPTABLE SEMANTICALLY .PHRASE CHANGE 6L 2A 6H CLAUSE CHANGE NON-WORD RELATED WORDS STRUCTURE CORRECTED 19 (cont'd.) Figure 144 Intonation — The number of miscues involving intonation is not significant. Bound morpheme level - Most differences at the bound morpheme level involve the substitution of alternate inflections due to dialect and as such are not serious. Word level - The proficient readers make fewer substitutions but a higher percentage of omissions, insertions and reversals. This is directly related to their freer use of optional transformations. Non-words - The proficient readers substitute fewer non-words, reflecting their concern with meaning and their conceptual experience. Phrase change - The proficient readers make changes at the phrase level with a larger percentage of their miscues than do the other groups. Clause change - As with phrase change, this category is higher for the proficient readers, but the difference between groups is not so great. This category involves deep structure to a greater extent, and a larger percentage of the proficient readers' miscues are at the surface level. Corrections - The proficient readers are making much better use of correction strategy. They know which structures require correction and are usually successful in making that correction. Semantic word relationships - The word level substitutions of the proficient readers show a much higher relationship to the text than do those of the weaker readers. Implications of the study: The data suggest an instructional model directed toward meaning rather than the processing of visual information. The weak readers are over-using grapho-phonic skills to the detriment of meaning. 145 In the category of syntactic acceptability (p. 47) the largest percentage of miscues for all groups fell in the totally acceptable category, which indicates that even the weakest readers are making some use of their grammatical ability, however, they are not nearly so effective as the proficient readers. All speakers have a well- develOped grammar but some students are more successful in applying this knowledge to reading. An over-emphasis on phonics and on identi- fying words to the neglect of larger syntactic units has likely con- - tributed to their lack Of awareness that reading must sound like language and must have meaning. The second highest percentage of miscues in the syntactic accepta- bility category was in acceptable withgprior for the more proficient groups but in totally unacceptable for the low group. Acceptable with 'prigr‘miscues occur with prediction; the reader predicts a word on the basis of the previous structure. Since the successful readers make~ predictions often, this strategy should be encouraged in the weaker readers. They should be encouraged to guess or predict in terms of grammar and meaning when they encounter an unknown word. Traditional "word-attack" training provides only one mode of operation - "sounding- out." In order to encourage prediction, the high premium on accuracy should be abandoned, providing a low threat situation for readers. A high-threat situation forces readers to wait for help from authority or peers or to omit unfamiliar words in fear of being wrong. An atmosphere *which encourages guessing will enable grammatical abilities to Operate. The data show that most non—words - guesses - retain the grammatical function of the expected response. While non-words do not provide meaning, their use is an improvement over omissions. 146 e.g. ER: He skated on the river-ice. 0R (1): He-on the river-ice. 0R (2): He $skatted on the river-ice. 0R (1) results in a completely unacceptable sentence whereas OR (2) results in meaning loss but retains grammatical acceptability and the grammatical function of the verb. If the student is able to retain the grammatical flow, he is well on the way to effective reading as the response at least sounds like language. Most instructional programs discourage or actually forbid correction, which is termed "regression," a highly negatively loaded term. Effect- ive correction strategy must accompany encouragement to guess, or to predict. Prediction can be made on the basis of meaning and high frequency syntactic patterns. Freedom to guess provides a greater chance of being correct than does refusal to make an attempt but the chance of being wrong is implicit in the gamble. Being wrong is not a problem if the reader can recognize his mistake and correct it.' The successful readers in the study were very proficient in their use of correction strategy. They corrected only when the miscue resulted in loss of meaning or ungrammaticality. Many of the miscues of the weak readers were related to their inadequate conceptual framework for the given material. Reading material must present settings and experiences familiar to the student. Any improvement in material with reference to the culture of the weak readers should be in the direction of content rather than dialect. The entire view of the nature of reading must be made-clear to the slower readers. They Operate on the assumption that-reading is a skill which involves deciphering sounds from the graphic display - a 147 natural result of the skill-oriented programs they have experienced. Their attention must be re-focused toward meaning. A profile of instruction: The present study has provided a close-up view of the proficient reader and a comparison between that successful reader, the weak reader, and the developing reader. Bear in mind that the purpose of the inquiry method is to help learners increase their competence as learners. It hOpes to accomplish this by having students db what effective learners do. Thus, the only reasonable kind of logic or structure that can be applied in this environ- ment is that which is modeled after the behavior of'good learners. Good learners, like everyone else, are living, squirming, questioning, perceiving, fearing, loving, and languaging nervous systems, but they are good learners precisely because they believe and db certain things that less effective learners do not believe and do. And therein lies the key. (Postman and Weingartner 1971:31) Whether it has ever been proven that it is possible to successfully teach some to do what others do naturally is not known at this point, but surely we can help them to come closer. The professional golfer, the accomplished musician, the talented cook, all these experts analyze their methods of Operation and attempt to pass this successful process on to those who would learn. And so it should be with reading. It is likely that all the readers in the present study were taught in a similar manner, yet some have been successful and others have not. But it is not possible to know how a person has learned. Did he learn because of the method or in spite of the method? Perhaps good learners subvert the system while the slower learners try desperately to do as they are told, while the information they are given leads them astray. 148 The following is an attempt to apply the strategies of the good reader to l) the beginning reader, 2) the older unsuccessful reader. Beginning reading: Reading is a complex process by which a reader recon- structs to some degree, a message encoded by a writer in graphic language. (Goodman, K. l969:2) Since meaning is not only the end product of reading, but also essential to efficient processing of visual information, meaning should be the focal point of an instructional program. Reading is a language process. Written language is a symbolic code by means of which meaning is transferred from sender to receiver. Oral language is learned, not as an end in itself, but rather as a means to an end. That end is communication, the sharing of information, feelings, etc. A successful beginning reading program will focus on the communicative aspect of reading. Reading should not be "taught" as a skill, but rather, "learned" as a portal to exciting literature, sharing of experiences, and acquisition of information. As an alternative language process, learning to read should be as natural as learning to speak. The native speaker learns his language by being exposed to a considerable amount of data from which he selects, predicts, and develops rules which allow him to be productive, to create structures which are new to him. Now let's think about word recognition and word analysis. You probably are wondering how children are going to learn sight words and word attack skills when you are not required to pre-teach them. The pre-teaching of words is the traditional method of teaching children to read. It's as if they must be spoon-fed the vocabulary, word by word, ad infinitum. This is not the way the children learned the language. You do realize, don't you, that they already are speaking the language, framing sentences, constructing intricate meanings, and reacting to a multiplicity of sentence sounds which were never pre—taught? 149 Children learned the spoken language by continuous exposure to it. Learning to use the language orally is probably the greatest intellectual feat they will achieve in a lifetime, and they did it before the age of five. The frustrated mother who tells a three-year- old, "0h, stOp your arguing!" does not reason whether he has encountered this particular vocabulary before. She takes it for granted that the intonations, the emo- tional overtones, the sentence sounds, and the specific context will tell him what she means. This is how language is learned. Since reading is a part of language, isn't it reasonable to assume that this is how reading can be learned? By exposing children to memorable and oral language, by letting them see the printed form of that oral language, you will find them coming to you with book in hand, saying, "I know this. This word is friends. I know what this says. It says, 'We are your friends.'" How much better, more insightful, to encourage a child to discover what he knows, rather than to tell him, "You know this word. We had it before. This word is friends." (Martin l966:9) A good deal of listening, or receptive control, is accomplished before the young child produces language himself. As far back as 1935, teams of investigators such as Gesell and Thompson or Buhler and Hetzer reported that when children learn their first language, listening comprehension of many complex utterances is demonstrated before these children produce any intelligible speech. In the field of second language teaching, many educators are now stressing that students should be submerged in oral data and encouraged to listen and try to comprehend. The student is not encouraged to attempt to speak until he has an uncontrollable urge to do so (Asher, 1972). It would seem worthwhile to provide the same kind of submersion in data for beginning readers. Various means of providing the data are available. 150 1. Reading to children The child learns to speak by being spoken to. He can learn to read by being read to (Pyle, 1964). The first few months of school should contain as much reading to children as they can enjoy without becoming bored. As often as possible, they should be provided with copies of the text which the teacher is reading. Reading materials are available which provide recordings of children's literature with accompanying text. The Weston Woods Collection of records and related books is an excellent example, as is the Read Along Library by Scott Foresman. 2. Singing - Children may be provided with song sheets to follow as they and the teacher sing. A program could be developed in which folk-song lyrics are projected on a screen and a "follow-the-bouncing- ball" technique used to help them match sound with symbol. There are many advantages to this suggestion. l) The rhythm and rhyme of folk songs provide a pattern which will allow children to make predictions. 2) Most folk music has a definite story line so meaning is prominent. 3) There is usually a good deal of repetition to provide reinforcement. 4) The children can invent new verses, thus developing their creativity and interest. 5) Children love to sing (if this is not ruined by making it a competitive issue). The activity of reading will become associated with pleasure not pain. Children learn songs quite quickly and so will be involved as a group from the very early stages. There will be no pressure of having to read orally, individually, to be evaluated by peers and teacher. The group experience will provide a de-individualized, thus unthreatening situation. At least one hour each day should be spent in this way. 151 3. Labels Printed stimuli should be abundant within the classroom. Bulletin boards, animal pens, pictures, etc., all should be accompanied by labels in full sentences. I am a hamster. My name is Mr. Big, etc. The labels should be changed often to maintain interest. A miniature store could be set up with products bearing familiar labels: Crest, All, Kellogg's, etc. 4. Language experience After a good deal of input, the children will eventually be anxious to try to produce written language. They will share experiences which the teacher can write. The students will make hypotheses about sound to spelling correspondences, etc., as they associate what they dictate with what the teacher writes. Language experience stories may then be shared with peers (Lee and Allen, 1968). 5. Literature As early as possible, children should be encouraged to read literature. There is a wealth of good children's literature available today. "The Sounds of Language" series and the new Scott Foresman series are worthwhile collections. Given this kind of saturation in written and oral language, most children will learn to read. Undoubtedly, some will be slower than others. Maturation always proceeds at different rates in different individuals. Some babies begin to talk before their first birthday while others are much slower. It would be alarming to see a test given at nine months of age to all babies to determine whether or not they will be "problem speakers." Those who were "would-be" problems would be separated out, their mothers told their babies are slow and the babies given special drills to promote "readiness." 152 Readiness is involved in learning to read but it should be the type of readiness which causes a child to spontaneously begin reading when he feels the desire. The types of activities mentioned do not require the stratification of children into ability groups. Ability grouping exaggerates and enlarges any differences in learning ability or learning rate which may exist between children. This type of segregation can only be~ harmful to the self—concept of the child. Small group work is essential in order to give children plenty of Opportunity to verbalize but the grouping should be random and flexible. There is no need for formal drill in phonics. There is no doubt that readers using an alphabetic system do operate to a great extent on sound-symbol correspondences. However, the rules are so complex (see Ch. ‘11) and abstract (Chomsky 1968:50) that it is ineffective to attempt to teach them directly. Children can formulate their own rules as they do when learning to speak (Smith 1971:226). Formal drill in phonics puts the emphasis on the analytical rather than the meaning process. Spoken language is not learned by a step-by-step sequencing of sounds but rather in the total context of meaning. The efficient readers in the present study operate with meaning and grammar to pre- dict to the visual. An emphasis on the visual with a premium on accuracy reverses this process. In order to help children to read for meaning, we must begin with meaning, not with abstract symbol to sound relationships. The older reader: A beginning reading program such as the one utilized above should prevent the type of problem reader the study revealed. No doubt some children will learn more quickly than others but the mode of reading 153 should be the same. The second grade readers in the study, although certainly not proficient readers as yet, are moving in the direction of the proficient sixth graders. They produced more semantically and syntactically acceptable structures than the 6L group and were less concerned with accurate processing of visual information. The standard remedial reading program has ineffective readers placed in phonics skills activities. The data reveal that phonics work is certainly not needed for the weak readers in this study. On the contrary, they are functioning too heavily with grapho-phonics and not deriving meaning. Further work in phonics can only repeat what has failed to be effective with them in previous years. They need to become aware that reading must have meaning and must sound like language. Many students are very quickly labelled "problem readers" and problem readers they become (Rosenthal, 1968). Once assessed as problems they are given phonics drills, workbook exercises, even spelling practice in order to help their reading. But these activities are not reading. One learns to read by reading and being read to. All of the activities suggested for beginning readers should be used with older readers. Production should be de-emphasized for a period of time and the student allowed to listen and follow in his own book. Reading to students should continue throughout all the school years. All people enjoy being read to and those students for whom reading is a difficult task will benefit greatly from the involvement with exciting literature they obtain in this way. All readers, even the L group in the present study, have certain, strengths: they all use their syntactic system effectively; they all Inake good use of grapho-phonic cues; they all derive some meaning from 154 their reading. These strengths should be pointed out to them. They need to be encouraged to make corrections where appropriate, to predict, to guess, to keep going in order that their syntax be as natural as possible. They need to be free to make hypotheses and test them; to make mistakes. Remedial or develOpmental reading at the secondary level often focuses on helping students to learn to read content material from their other subjects. In order to read effectively, the reader must bring to the task a rich experiential background.' The semantic system is a large contributor to the processing of the visual information and without a conceptual framework, reading becomes a very difficult Opera- tion. Add to this the poor writing style of most text-book writers and students become presented with a formidable task. Teachers in all subjects need to be made aware of this and temper their expectations of reading. Concept develOpment should precede reading. Meaning is both input and output in reading. Any selection will be understood only to the extent that the reader brings to it the prerequisite concepts and experiences. Even in reading to learn, the new concepts can only be slightly beyond the reader's prior attainments, and he must be able to relate the vicarious experience to real experience in order to make any use of it. (Goodman, K. 1972:159) All reading should take place in a natural language context. Materials should be interesting and meaningful.“ Children should be encouraged to read a great deal and the experience should be as reward- ing and free from threat as possible. Reading, like language, "is learnt in operation not by dummy runs" (Dixon 1967:13). BIBLIOGRAPHY BIBLIOGRAPHY Asher, James J. 1972. "Children's First Language as a Model for Second Language Learning," Modern Language JOurnal, March 1972. Baratz, Joan, and Stewart, William S. 1970. Dialect Reader: Friends. Education Studies Center, Washington, D.C. Bloomfield, Leonard. 1925. "Why a Linguistic Society?" Language, Vol. 1, pp. 1-5. Bfihler, C., and Hetzer, H. 1935. Testing Children's Development from Birth to School Age. Farrer and Rinehart, New York. Chafe, Wallace L. 1970. Meaning and the Structure of’Language. The University of Chicago Press, Chicago. Chomsky, N., and Halle, Morris. 1968. The Sound Pattern of’English. Harper and Row, New York. Clymer, Theodore. 1963. "The Utility of Phonic Generalizations in the Primary Grades," from The Reading Teacher, 16: 252-258; January,_ 1963. Dixon, John. 1967. Growth Through English. National Association for the Teaching of English, Oxford. Durr, William K. 1965. "Developing Reading Independence in the Primary Grades," from Todhy's Challenges in the Teaching of’Reading, SCIRA Conference Proceedings, Scholastic Magazines, Inc., New York, pp. 75-80. Fries, Charles C. 1963. Linguistics and Reading. Holt, Rinehart and Winston, New York. Gesell, A., and Thompson, H. 1935. Infant Behavior. McGraw-Hill, New York. Goodman, Kenneth S. 1967. "Reading: A Psycholinguistic Guessing Game," JOurnal of'the Reading Specialist, May 1967. Goodman, Kenneth S. 1970. Taxonomy of'Reading Miscues. Unpublished manual. Goodman, Kenneth S. 1972. "The Reading Process: Theory and Practice," in Hodges and Rudorf, 1972. 155 156 Goodman Kenneth 8., and Burke, Carolyn. 1968. Study Of'chiZdan'B Behavior while reading orally. Final Report, United'States Department of Health and Welfare, Office of Education. Goodman, Kenneth S., and Burke, Carolyn L. 1969. A study of'oral reading miscues that result in grammatical re-transfbrmations. U.S. Department of Health, Education and Welfare, Office. of Education. Goodman, Kenneth S., and Fleming, James T. (eds.) 1968. Psycho- linguistics and the Teaching of'Reading. International Read- ing Association, Newark, Delaware. Goodman, Kenneth S., and Niles, Olive 8. 1970. Reading: Process and Program. National Council of Teachers of English, Champaign, Illinois. Goodman, Yetta M. 1967. A psycholinguistic description of observed oral reading phenomena in selected young beginning readers. Unpublished doctoral dissertation, Wayne State University. _ Goodman, Yetta M. 1971. Longitudinal Study of’Children's Oral Reading Behavior. Final Report, U.S. Department of Health, Education and Welfare, Office of Education. Hodges, Richard E., and Rudorf, E. Hugh. 1972. Language and Learning to Read. Houghton Mifflin Company, Boston. Kolers, Paul A. 1966. "Reading is Only Incidentally Visual," in Goodman and Fleming, 1968. Lee, Doris and Allen, Roach V. 1963. Learning to Read Through Experience. Appleton-Century-Crofts, New York. Lefevre, Carl A. 1964. Linguistics and the Teaching of’Reading. McGraw-Hill, New York. Martin, William, Jr. 1966. Sounds of'Numbers. Holt, Rinehart and Winston, Inc., New York. - Menosky, Dorothy M. 1971. A psycholinguistic analysis of the oral reading miscues generated during the reading of varying portions of text by selected readers from grades two, four, six and eight. Unpublished doctoral dissertation, Wayne State University. Page, William D. 1970. A psycholinguistic description of patterns of miscues generated by a proficient reader in second grade, an average reader in fourth grade, and an average reader in sixth grade encountering ten basal reader selections ranging from pre-primer to sixth grade. Unpublished doctoral dissertation, Wayne State University. ‘1 f1“; I 157 Postman, Neil, and Weingartner, Charles. 1971. Teaching as a Subversive Activity. Delacorte Press, New York. Pyle, Wilma J. 1964. An exploratory study in reading on the first grade level using a combination of trade books and their cor- responding phonograph records. Unpublished doctoral disserta- tion, Wayne State University. Rausch, Peter D. 1972. A psycholinguistic investigation into prior conceptual knowledge, oral reading miscues, silent reading, and post-reading performance. Unpublished doctoral dissertation, Wayne State University. Rosenthal, Robert, and Jacobson, Lenore F. 1968. "Teacher Expectations for the Disadvantaged," Scientific American, April 1968, pp. 1923. Sims, Rudine. 1972. A psycholinguistic description of miscues generated by selected young readers during the oral reading of text material in black dialect and standard English. Unpublished doctoral dissertation, Wayne State University. Smith, Frank E. 1971. Understanding.Reading. Holt, Rinehart and Winston, New York. Weber, Rose-Marie. 1970. "A Linguistic Analysis of First-Grade Reading Errors," Reading Research Quarterly, 5, Spring 1970, pp. 427—451. APPENDIXES APPENDIX A BILLY WHITEMOOIV 158 I'm—x! _ 0101 0102 0103 0104 0105 0106 0107 0108 0109 0110 0111 0112 0113 0114 0115 0116 0117 0201 0202 0203 0204 0205 159 BILLY WHITEMOON Billy Whitemoon was a Winnebago Indian boy. He lived with his father and mother in a cabin near the Black River. Billy liked to take part in the work of his tribe. One of the things he liked most was cranberry picking in the fall. All the men and women and children of the tribe went to the cranberry swamp near the Winnebago lands. They would spend days picking the ripe cranberries, which they put in boxes and sent to the city. Billy liked the winter, too. It was fun to go to school. When he wasn't in school, he skated with his friends on the river ice. But when the heavy snow was gone from the Winnebago lands, Billy was very happy. He knew that spring had come. One spring day Billy was walking through the woods. He heard a little moaning cry. There in the dry, dead leaves he saw a little fawn. Billy went closer. He was surprised that the 0206 0207 0208 0209 0210 0211 0212 0213 0214 0215 0216 0217 0218 0219 0220 0221 0222 0301 0302 0303 0304 0305 0306 0307 160 little fawn didn't run away. Billy knew that fawns were always very shy. Then he noticed that this one's leg was broken! "Poor little fawn!" said Billy. "You just wait here. I'll be back soon." Billy hurried to his cabin. Soon he returned with two straight sticks and some string. He tied the sticks to the broken leg. Then he picked up the fawn and carried it home. When his father saw the fawn, he said, "What a beauty! He will make a good pet." Billy loved all wild animals, but he loved the shy little fawn best of all. When the broken leg was better, Billy took the sticks off. Then he and the fawn would race together through the forest. Billy named his pet Lightfoot betause he could run so fast. Every spring Billy helped his father cut down young trees, which his mother used in making baskets. Mother Whitemoon made baskets the way. all Winnebago women did. She pounded the young trees into long strings. From the strings she made beautiful baskets. 0308 0309 0310 0311 0312 0313 0314 0315 0316 0317 0401 0402 0403 0404 0405 0406 0407 0408 0409 0410 0411 0412 0413 0414 161 Some of the baskets she colored red or blue or orange. She made her own paints from the roots that Billy gathered from the swamps. She had taught him to know the kind of roots used by Winnebago Indians for many, many years. This spring Billy was delighted that the roots had made such beautiful colors. He knew that the baskets would sell well at their summer camp. When warm weather came, the Whitemoons moved to their summer camp. They packed their kettles, blankets, clothes, and other baggage into their old car. They packed Mbther Whitemoon's baskets carefully. Then they pushed Lightfoot into the car. When everything was loaded, they started down the highway. They drove until they found a good place to camp for the summer. Then Billy and his father built a summer house. They covered it with deer hides to keep the family dry in rainy weather. When their house was done, they built one for Lightfoot, too. 0415 0416 0417 0418 0419 0501 0502 0503 0504 0505 0506 0507 0508 0509 0510 0511 0512 0513 0514 0515 0516 0517 0518 162 Every day Mother Whitemoon would put on a bright cotton dress and pretty earrings. Then she would sit in front of the summer house and sell her baskets. She let travelers who bought them take her picture. When summer ended, the Whitemoons packed their belongings again. Then they crowded into the car with Lightfoot, who was much bigger now. On their way back to their winter home, they stOpped for a week to take part in the Winnebago Dance Time. At this season of the year all the Winnebago Indians camped near the river. They built campfires and danced every day. Billy feasted on roast corn and baked fish.‘ He listened to the stories and the songs of their tribe. Billy wished he could sing some of the songs he was always making up. But he was too shy to sing in front of peOple. Only Lightfoot, his pet fawn, knew the songs that Billy could sing. 0601 0602 0603 0604 0605 0606 0607 0608 0609 0610 0611 0612 0613 0614 0615 0616 0617 0618 0619 0620 0621 0622 0701 0702 163 After the Dance Time was over, all the tribe returned to their winter cabins. Now it was the season for deer hunting. White men from the cities came to hunt in the forests near the Winnebago land. Billy was glad that there was a law saying that no white man could hunt on Winnebago land. Lightfoot was so much bigger now that the hunters would surely shoot him. One afternoon Billy was walking through the forest on his way home from school. He heard a rustle in the leaves. A short way ahead of him he saw Lightfoot coming to meet him! The sight of his pet frightened Billy, for Lightfoot was off Winnebago land! If a hunter should see him, he would have the right to shoot. Billy looked around quickly to see if there was any danger. He heard the rustling of leaves! His eyes caught sight of a red jacket. There was a hunter looking at Lightfoot. The man lifted his gun to his shoulder. Billy shook with fear. Then in a flash he stepped between the hunter and Lightfoot. 0703 0704 0705 0706 0707 0708 0709 0710 0711 0712 0713 0714 0715 0716 0717 0718 0719 0720 0721 0722 0801 0802 0803 0804 164 "Get out of the way, boy!" shouted the hunter angrily. "You might get hit!" "Oh, please!" Billy cried. "Don't shoot that deer! He's mine! He's mine!" "How do I know he is your deer?" the hunter asked. "All deer look alike." "Oh, but he is mine!" Billy insisted. "You can't prove it!" the hunter said. He was still angry. Billy knew how he could prove Lightfoot was his. If he sang, Lightfoot would come to him. NO one had ever heard Billy's songs. The man might laugh at him, but he had to save Lightfoot. Billy smiled shyly. Then he began to sing. "Come, Lightfoot, come here, come here. Come to me, my little deer!" There was a rustling sound. Lightfoot came leaping through the woods toward Billy. He put his soft nose on his master's shoulder. "You win!" said the hunter. "You have proved the deer does belong to you. I liked your song, too. You sing very well." Billy was so pleased by the hunter's words 0805 0806 0807 0808 0809 0810 0811 0812 0813 0814 0815 0816 165 that he told his mother and father what had happened. Then he sang for them, too. Next year when the Winnebago Dance Time came, Billy sang for all the tribe. He was no longer shy as he sang his songs about the big world and the blue sky. He sang of the stars and the moon, and the brook that flows over the stones in the forest. He sang of the seasons of the year, and of Lightfoot, his wonderful pet deer. He sang so well that the tribe called him "Billy Whitemoon, Maker of Beautiful Songs." Along Friendly Roads - 3-2 APPENDIX B GOODMAN TAXONOMY 0F READING MISCUES - SHORT FORM 166 46 47 48* 167 READING MISCUE RESEARCH - CODING SHEET Correction CRECT 0 l 2 9 no yes abandons correct unsuccessful Dialect Involved DILCT 0 9 no yes idiolect (M.A.T.) super correct (M.A.T.) secondary involvement in miscue (sub-studies) foreign language influence (M.A.T.) doubtful Graphic GRAPH blank 0 no similarity 1 letters in common 2 any key letter in common or the middle portions similar 3 end Add 1 for configuration (0-6) 4 beginning 5 beginning, middle 6 beginning, end/middle, end 7 beginning, middle, end or reversals of three letters * or more Category involves use of both blanks and zeros. 49* 50 * 8 9 168 single grapheme difference or reversals of two letters or all but punctuation homographs Phonemic PHONM blank 0 no similarity 1 some common sounds 2 single key elements in common 3 final portions in common 4 common beginning 5 common beginning and middle portions 6 common beginning, end/middle, end 7 beginning, middle and end similar 8 differ in single vowel or consonant or morphOphonemic or intonation shift (including schwa) 9 homOphones Allologs ALLOG 0 1 no contraction/full full/contraction contraction not rep. in print long and short forms or syllable deletion/insertion shift to idiomatic form shift from idiomatic form misarticulation Category involves use of both blanks and zeros. 51 52 53* 54* 169 Syntactic Acceptability SYNAC 0 1 2 3 4 0 no only with prior only with after in sentence in total passage Semantic Acceptability SEMAC ' (This category cannot be scored higher than 51.) Fifi no only with prior R 1 2 3 4 only with after in sentence in total passage Transformation TRANS 0 l 2 3 4 9 no transformation through different deep structures same deep structure - through alternate or compulsory rules alternate options deep structure lost or garbled (51 marked 0) doubtful Syntax SYNTX (mark when 51 is 3 or 4) (blank when 51 is O, 1, 2) blank 0 unrelated 1 single element in common 2 key element in common Category involves use of both blanks and zeros. 55* 56 9 170 major change in sentence pattern minor change in sentence pattern a major change within structure of phrase minor change within structure of phrase change in person, tense or number change in choice of function word or other minor shift unchanged Semantic SMANT (mark when 52 is 3 or 4) (blank when 52 is 0, l, 2) blank 0 completely anomalous to rest of story 1 change or loss affecting plot in basic sense or creates major anomalies 2 change or loss involving key aspects or seriously interfering with sub-plots 3 change or loss resulting in inconsistency of major incident, major character or major aspect of sequence 4 change or loss resulting in inconsistency of minor incident, minor character or minor aspect of sequence 5 change or loss of aspect which is significant but does not create inconsistencies 6 change or loss of unimportant detail 7 change in person, tense, number, comparative, etc., which is noncritical 8 slight change in connotation/or similar name which doesn't confuse cast 9 no change Intonation INT¢N 0 1 no within words *Category involves use of both blanks and zeros. 57 58-59 171 between words within one phrase structure relative to phrase or clause structure of the sentence end of phrase or sentence (terminal) conjunction substituted for terminal or vice versa intonation involving direct quotes Sub morphemic Level SUBMR 0 l 5 no substitution insertion omission reversal multiple minor variations Bound and Combined Morpheme BNDMR 0 1 no 0 no substitution 1 inflectional suffix insertion 2 non-inflected form omission 3 contractional suffix reversal 4 derivational suffix 5 prefix 6 miscue across affix types 7 miscue involving base 60-61 62 63 172 Word and Free Morpheme WORDL 0 1 no 0 no substitution 1 multiple morpheme word (O.R.) for multiple morpheme word (E.R.) insertion 2 single morpheme word (O.R.) for omission single morpheme word (E.R.) reversal 3 multiple morpheme word (O.R.) for single morpheme word (E.R.) 4 single morpheme word (O.R.) for multiple morpheme word (E.R.) 5 word or free morpheme in longer word 6 word in compound 7 non-word 8 dialect alternative Phrase PHRSL no substitution insertion omission reversal Clause CLAUS no substitution insertion omission reversal without change in dependency clause dependency is altered within sentence clause dependency is altered across sentences 65-69* blank (65-66 Gram- matical Category) 1. 2. 3. Noun Verb Noun Modifier 173 (67-68 Gramr matical Filler) 0 1 common prOper pronoun verb derived phrasal unit word as word name quantifiers adjective in noun position be forms transitive intransitive infinitive pro-verbs adjective noun adjunct verb derived possessive noun Grammatical Category and Surface Structure of O.R. GFOBR (69 Grammatical Function) 0 1 subject 2 direct object 3 indirect object 4 appositive 5 address 6 noun in adverb- ial phrase or other preposi- tional phrase 7 subject comple- ment 8 object comple- ment 9 intensifica- tion 0 1 active 2 passive 3 imperative 4 subjunctive 0 1 subject comple- ment 2 embedded 3 object comple— ment 3. 4. 5. Noun Modifier (cont'd.) Verb Modifier Function Word 174 10 ll 12 possessive pronoun titles adverbial ordinal number phrasal unit 0 pro-adverb 1 place adverb 2 manner noun form 3 time 4 reason 5 other noun marker verb marker verb particle question marker clause marker phrase marker intensifier conjunction negative quantifier other adverb particle 6. 7. Indeterminate Contractions 175 interjection words out of syntactic context defies classifica- tion/ambiguous greetings (left) pronoun verb marker. be let question marker/ clause marker it/there adverb no 1111 transitive verb (have) (right) verb marker be trans verb (have) negative pronoun (us) When the E.R. and the O.R. are the same grammatical function, 65-69 can be coded 99999. 70-7l* blank 0 unrelated 1 Category involves use of both blanks and zeros. Semantic Word Relationships SMWRD association to homOphone or homograph primarily syntactic relationship with minor semantic association strong sequential semantic association to prior/subsequent word or to word itself 10 11 12 13 14 176 shift to generic from specific shift to specific from generic common attribute or confusion between characters antonym other in a pair variant form of same word: inflected or derivational slight difference in connotation similar name synonym within the text synonym in other contexts some semantic association between E.R. and O.R. APPENDIX C STANFORD DIAGNOSTIC READING TEST 177 178 STANFORD DIAGNOSTIC READING TEST* TEST 1: READING COMPREHENSION Directions: Find the one word that belongs in each space and make a cross in the circle in front of that word. Do not write in the spaces. Samples The mouse ran away when it saw the A . The mouse was B . A ® cat 0 hole 0 cheese 0 table B O hungryo happy 0 afraid O glad The cactus is C that grows in the desert. It can survive with almost no D . C O a plant Oa hill C) an animal C) an insect D 0 sun 0 heat 0 water 0 air TEST 2: VOCABULARY Directions: Make a cross in the circle beside the word which best completes the sentence that the teacher reads. Samples A 0 fish X bird 0 airplane B O nap O bite 0 picture TEST 3: SYLLABICATION Directions: Look at the first word in each line. Find the first syllable of that word. Then find it at the right and make a cross in the circle in front of the syllable. Samples A winter 0 wi ® win 0 wint B different 0 dif 0 diff O differ * Harcourt, Brace & World, Inc., New York, 1966. 179 TEST 4: SOUND DISCRIMINATION Directions: One of the last three words in each line has the same sound as the sound which is underlined in the first word of the line. Make a cross in the circle beside the word that has the same sound. A g3 Cam (2 no 0 do B dgy Oby 0 dog 0 ate TEST 5: BLENDING Directions: Make crosses as your teacher tells you. Samples “D d 08 ® 8 Om (3)0 O f BOor Oee 0 se 0 ... Q... Q at COon Om O den 01. Q... o ...... TEST 6: RATE OF READING Directions: This is a test to see how fast and how accurately you can read. Make a cross in the circle next to the word in every third line that best fits in with the sentence. 1 Many years ago people thought that when 2 North America was discovered by 3 OColumbus 0 Edison 0 Washington 4 there were wild horses here. We now know 5 that this was not true. The Indians had 0 always 0 often 0 never 0‘ 10 11 12 13 14 15 l6 l7 l8 19 20 21 180 seen horses before the Spaniards arrived. The horses they later tamed were descendants of O boats 0 horses 0 wagons that had escaped from the Spaniards. Soon herds of wild horses were roaming the O plains Ostreets O seas. Those brought by the Spaniards, however, were not the first to exist on the North 0 African 0 American 0 EurOpean continent. Some 50 million years ago, North America's swampy forests were home to the 0 last 0 earliest 0 biggest horses, called "dawn horses,‘ which were no larger than a small dog, or perhaps a O fox 0 bear 0 cow.