UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Grammatical implication and anaphora as determinants of text comprehension : English as a second language Walsh, Daniel Michael 1978

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

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

Item Metadata

Download

Media
831-UBC_1978_A8 W38.pdf [ 7.09MB ]
Metadata
JSON: 831-1.0094464.json
JSON-LD: 831-1.0094464-ld.json
RDF/XML (Pretty): 831-1.0094464-rdf.xml
RDF/JSON: 831-1.0094464-rdf.json
Turtle: 831-1.0094464-turtle.txt
N-Triples: 831-1.0094464-rdf-ntriples.txt
Original Record: 831-1.0094464-source.json
Full Text
831-1.0094464-fulltext.txt
Citation
831-1.0094464.ris

Full Text

GRAMMATICAL IMPLICATION AND ANAPHORA AS DETERMINANTS OF TEXT COMPREHENSION (E n g l i s h as a Second Language) by DANIEL MICHAEL WALSH B.A., U n i v e r s i t y of Hawaii, 1967 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES Department of E n g l i s h Education We accept t h i s t h e s i s as conforming to the re q u i r e d standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1978 © D a n i e l Michael Walsh, 1978 In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Univers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f r ee l y ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th is thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives . It is understood that copying or pub l i ca t ion of th is thes is for f inanc ia l gain sha l l not be allowed without my wri t ten permission. Department of E n g l i s h Education The Univers i ty of B r i t i s h Columbia 2075 Wesbrook P l a c e Vancouver, Canada V6T 1W5 Date September 12. 1978 i i ABSTRACT The present study seeks to assess the contribution of two aspects of l i n g u i s t i c competence, recognition of grammatical i m p l i c a t i o n and i d e n t i f i c a t i o n of anaphoric referents, to a p r e c i s e l y defined measure i n accord with a model that sets c e r t a i n requirements for acts of text comprehension. The f i r s t part of the research instrument presents items containing one or more sentences i n i s o l a t i o n to test understanding of twelve syntactic transformations and four types of anaphora. The second part examines comprehension of sentences within a continuous text. The c r i t e r i o n measure f o r text comprehension i s defined by two equally weighted components: i n d i c a t i o n of textual locus and judgment of truth value. Sample populations include native speakers of English, Chinese speakers, and other non-native speakers learning English as a second language (E.S.L.). In each category, students at elementary, junior secondary, and senior secondary grade i n t e r v a l s are compared. Analysis of the data reveals a developmental trend toward augmented performance on a l l v a r i a b l e s . Native speakers as a group s i g n i f i c a n t l y outperform E.S.L. students only on the measure of recog-n i z i n g grammatical implication r e l a t i o n s which, i n conjunction with the task of i d e n t i f y i n g anaphoric referents, contributes a greater proportion of variance to c r i t e r i o n scores than i s observed i n the case of either E.S.L. sample. I t i s concluded that native speakers and E.S.L. students a t t a i n equal p r o f i c i e n c y i n text comprehension by means of d i f f e r e n t s t r a t e g i e s . i i i A comparison of the rank order of d i f f i c u l t y among the s i x t e e n t r a n s f o r m a t i o n a l and anaphoric types i n d i c a t e s that the scores of a m u l t i -e t h n i c sample of E.S.L. students more c l o s e l y approximate the p a t t e r n set by n a t i v e speakers than do those of the Chinese speaking group, i n d i c a t i n g that n a t i v e language may i n f l u e n c e performance on t h i s s o r t of task. The rankings f o r a l l three groups g e n e r a l l y support the n o t i o n that t r a n s f o r -m a t i o n a l l y simpler s t r u c t u r e s are more e a s i l y i n t e r p r e t e d . Regarding the components of t e x t comprehension, n a t i v e speakers perform s i g n i f i c a n t l y b e t t e r on the task of judging the t r u t h value of an item statement than on i n d i c a t i n g the locus of r e l e v a n t statements i n the accompanying passage. E.S.L. students encounter s i m i l a r d i f f i c u l t y w i t h both tasks. The locus i n d i c a t i o n component c o n t r i b u t e s somewhat more variance to the c r i t e r i o n scores of n a t i v e speakers and m u l t i e t h n i c E.S.L. students than to those of the Chinese speakers. I t i s suggested that locus i n d i c a t i o n i n i t s e l f may be a p r a c t i c a l and r e l i a b l e measure of te x t comprehension. Some d i r e c t i o n s are o f f e r e d f o r f u r t h e r research, recommendations are made f o r t e s t i n g s y n t a c t i c comprehension, and p o s s i b l e i n s t r u c t i o n a l i m p l i c a t i o n s a r i s i n g from t h i s and other research are discussed. i v TABLE OF CONTENTS Chapter I INTRODUCTION 1 Purpose of the study 1 Statement of the problem 1 Importance of the i n v e s t i g a t i o n 3 Overview of the study 5 L i m i t a t i o n s of the study 6 I I REVIEW OF THE LITERATURE 9 Sentence comprehension i s something more than word comprehension 9 S y n t a c t i c u n i t s are used to organize sentence p e r c e p t i o n 11 S y n t a c t i c comprehension i s c h a r a c t e r i z e d by developmental trends 12 School aged c h i l d r e n are unable to comprehend many s y n t a c t i c s t r u c t u r e s 12 Some s y n t a c t i c patterns are e a s i e r to comprehend than others 13 S y n t a c t i c comprehension c o n t r i b u t e s to more gen e r a l i z e d measures of reading comprehension 16 Syntax i s an important v a r i a b l e i n determining r e a d a b i l i t y 18 I I I DEVELOPMENT OF THE INSTRUMENT 19 Testing s y n t a c t i c comprehension 19 Grammatical i m p l i c a t i o n 20 Previous attempts to measure understanding of grammatical i m p l i c a t i o n 21 The present instrument 24 Anaphora 27 Previous attempts to measure understanding of anaphora 27 The present instrument 28 Testi n g t e x t comprehension 28 Previous attempts to measure understanding of a t e x t 28 The present instrument 30 V Chapter IV RESEARCH DESIGN AND HYPOTHESES 34 V PROCEDURES 40 Preliminary appraisal of test items 40 Format of the tests 41 Description of the sample populations 45 Administration of the instrument 49 Scoring of the tests 51 VI RESULTS AND DISCUSSION 54 St a t i s t i c a l properties of the instrument 54 Problems in s t a t i s t i c a l analysis of the data 63 Comparison of samples by Language group and Grade interval 71 Performance on tasks of syntactic and text comprehension Hypothesis I : Results 77 Hypothesis I : Conclusions 78 Hypothesis I : Discussion 79 Hypothesis II : Results 81 Hypothesis II : Conclusions..... 82 Hypothesis II : Discussion ...84 Relative d i f f i c u l t y of subtasks of syntactic comprehension Hypothesis III : Results 86 Hypothesis III : Conclusions 87 Hypothesis III : Discussion 87 Comparison of samples by Language group 88 Performance on tasks of syntactic and , text comprehension Hypothesis IV : Results 89 Hypothesis IV : Conclusions 102 Hypothesis IV : Discussion 102 Relative d i f f i c u l t y of subtasks of syntactic comprehension Hypothesis V : Results 104 Hypothesis V : Conclusions 106 Hypothesis V : Discussion 106 v i Chapter Performance on c r i t e r i o n measure of t e x t comprehension and component tasks compared Hypothesis VI : Results 109 Hypothesis VI : Conclusions 109 Hypothesis VI : D i s c u s s i o n I l l C o n t r i b u t i o n of s y n t a c t i c tasks to c r i t e r i o n measure of t e x t comprehension Hypothesis V II : Results 112 Hypothesis V II : Conclusions 116 Hypothesis V II : D i s c u s s i o n 116 C o n t r i b u t i o n of component tasks to c r i t e r i o n measure of t e x t comprehension... Hypothesis V I I I : Results 118 Hypothesis V I I I : Conclusions 118 Hypothesis V I I I : D i s c u s s i o n 120 Supplementary a n a l y s i s of the data 121 VII SUMMARY AND CONCLUSIONS 130 D i r e c t i o n s f o r f u r t h e r research 132 Recommendations f o r t e s t i n g 134 Im p l i c a t i o n s f o r i n s t r u c t i o n . . 137 BIBLIOGRAPHY 140 APPENDICES A: Model f o r sentence comprehension w i t h i n a w r i t t e n t e x t 145 B: The research instrument 146 v i i L i s t of Tables 1 Test s t a t i s t i c s f o r Task 1: Native Speakers and E.S.L. Students compared 64 2 Test s t a t i s t i c s f o r Task 2: Native speakers and E.S.L. Students compared 65 3 Test s t a t i s t i c s f o r Task 3: Native Speakers and E.S.L. Students compared 66 4 Test s t a t i s t i c s f o r Task 4: Native speakers and E.S.L. Students compared 67 5 Test s t a t i s t i c s f o r Task 5: Native Speakers and E.S.L. Students compared. 68 6 Comparison of performance on Task 1 by Language group and Grade i n t e r v a l 72 7 Comparison of performance on Task 2 by Language group and Grade i n t e r v a l . . 73 8 Comparison: of performance on Task 3 by Language group and Grade i n t e r v a l 74 9 Comparison of performance on Task 4 by Language group and Grade i n t e r v a l 75 10 Comparison of performance on Task 5 by Language group and Grade i n t e r v a l . . . . 76 11 Task 1 and Task 2 Subtask mean scores, Language group by Grade i n t e r v a l 83 12 Kendall's c o e f f i c i e n t of concordance by Language group 86 13 Kendall's c o e f f i c i e n t of concordance by Grade i n t e r v a l 86 14 Number of ^ 3s i n each Language group by Grade i n t e r v a l 88 14a Number of ^ s i n adjusted Native Speaker sample by Grade i n t e r v a l 89 15 Comparison of performance on Task 1 by Language group 91 15a Comparison of performance on Subtask A by Language group 92 v i i i 15b Comparison of performance on Sub task B by Language group 92 15c Comparison of performance on Subtask C by Language group 93 15d Comparison of performance on Subtask D by Language group . 93 15e Comparison of performance on Subtask E by Language group 94 15f Comparison of performance on Subtask F by Language group 94 15g Comparison of performance on Subtask G by Language group 95 15h Comparison of performance on Subtask H by Language group 95 15i Comparison of performance on Subtask J by Language group 96 15j Comparison of performance on Subtask K by Language group 96 15k Comparison of performance on Subtask L by Language group 97 151 Comparison of performance on Subtask M by Language group 97 16 Comparison of performance on Task 2 by Language group.. 98 16a Comparison of performance on Subtask P by Language group 98 16b Comparison of performance on Subtask R by Language group 99 16c Comparison of performance on Subtask S by Language group 99 16d Comparison of performance on Subtask T by Language group 100 17 Comparison of performance on Task 3 by Language group 100 18 Comparison of performance on Task 4 by Language group 101 i x 19 Comparison of performance on Task 5 by Language group 101 20 Comparative rankings of subtask d i f f i c u l t y by Language group . 105 21 K e n d a l l tau c o e f f i c i e n t s f o r p a i r e d comparisons among Language groups 106 22a C o r r e l a t e d t - t e s t s f o r Native Speakers across three measures of t e x t comprehension...... 110 22b C o r r e l a t e d t - t e s t s f o r E.S.L.(A) Students across three measures of t e x t comprehension 110 22c C o r r e l a t e d t - t e s t s f o r E.S.L.(B) Students acroos three measures of t e x t comprehension. 110 23 Mean item d i f f i c u l t i e s of s i n g l e sentence and m u l t i p l e sentence items f o r e n t i r e sample p o p u l a t i o n . 112 24a C o r r e l a t i o n c o e f f i c i e n t s f o r Native Speakers completing a l l tasks 113 24b C o r r e l a t i o n c o e f f i c i e n t s f o r E.S.L.(A) Students completing a l l tasks 113 24c Co r r e l a t i o n c o e f f i c i e n t s f o r E.S.L.(B) Students completing a l l t a s k s . . . . 113 25a M u l t i p l e r e g r e s s i o n analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by s y n t a c t i c measures to c r i t e r i o n Task 5 scores of Native Speakers 115 25b M u l t i p l e r e g r e s s i o n analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by s y n t a c t i c measures to c r i t e r i o n Task 5 scores of E.S.L.(A) Students 115 25c M u l t i p l e r e g r e s s i o n analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by s y n t a c t i c measures to c r i t e r i o n Task 5 scores of E.S.L.(B) Students 115 26a M u l t i p l e r e g r e s s i o n analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by component measures to c r i t e r i o n Task 5 scores of Native Speakers 119 26b M u l t i p l e r e g r e s s i o n analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by component measures to c r i t e r i o n Task 5 scores of E.S.L.(A) Students 119 26c M u l t i p l e r e g r e s s i o n analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by component measures to c r i t e r i o n Task 5 scores of E.S.L.(B) Students 119 X 27 P r o p o r t i o n of c o r r e c t responses f o r graphic length clue items and ( a l l items) by Language group and Grade i n t e r v a l 123 28 P r o p o r t i o n of correct, responses f o r " s u p e r f i c i a l c l u e " items and ( a l l items) by Language group and Grade i n t e r v a l . 125 29a C o r r e l a t i o n c o e f f i c i e n t s f o r 17 t r u e / f a l s e items and ( a l l items) f o r Native Speakers completing a l l tasks 127 29b C o r r e l a t i o n c o e f f i c i e n t s f o r 17 t r u e / f a l s e items and ( a l l items) f o r E.S.L. (A) Students completing a l l tasks 127 29c C o r r e l a t i o n c o e f f i c i e n t s f o r 17 t r u e / f a l s e items and ( a l l items) f o r E.S.L. (B) Students completing a l l t a s k s . . 127 30 Combined variance c o n t r i b u t e d by Tasks 1 and 2 to c r i t e r i o n Task 5 scores comprising 17 t r u e / f a l s e items and ( a l l items) by Language group 129 XX Acknowledgments I wish to convey my s i n c e r e a p p r e c i a t i o n to those who c o n t r i b u t e d to the completion of the research and the p r e p a r a t i o n of t h i s r e p o r t , the l i m i -t a t i o n s of which I am s o l e l y r e s p o n s i b l e . Dr. B. A. Mohan f o r h i s c o n t i n u i n g encouragement and r e c e p t i v e guidance i n p r o v i d i n g ample opportunity to explore the research t o p i c i n a manner that advanced my p r o f e s s i o n a l and e d u c a t i o n a l growth N. M. Ashworth f o r arousing my i n t e r e s t and confidence to pursue s t u d i e s l e a d i n g to t h i s research which has been enriched by her per-s p e c t i v e of p r a c t i c a l a p p l i c a t i o n s and s o c i a l value Dr. K. Reeder f o r c o n s t r u c t i v e i n s i g h t s that broadened the scope of the study Dr. C. Pennock and J . Booth f o r o f f e r i n g p e r s p e c t i v e s from research i n reading education A. Moodie, Research Studies and T e s t i n g , Vancouver School Board, f o r prompt a s s i s t a n c e i n arranging access to s u i t a b l e subjects Teachers f o r t h e i r generous cooperation i n a d m i n i s t e r i n g the research instrument at a busy time of the school year Dr. T. Rogers f o r extensive advice on the s t a t i s t i c a l a n a l y s i s of the data F. Ho and S. K i t a f o r a s s i s t a n c e i n accessing the computer programmes R. Berwick f o r advice on i n t e r p r e t a t i o n of some r e s u l t s and p r e p a r a t i o n of the manuscript as w e l l as c a r e f u l l y executed graphics T. Komori f o r a s s i s t a n c e i n p r e p a r a t i o n of the data M. Seesahai f o r a c a r e f u l proofreading of a lengthy manuscript M. Bjornson and S. Haney f o r enduring support and a s s i s t a n c e i n my teaching assignment without which t h i s research could not have been sustained 1 Chapter I INTRODUCTION Purpose of the study The purpose of this study i s to examine comprehension of a text in terms of one's a b i l i t y to recognize grammatical implication. Specifi-cally, such a b i l i t y is manifest when one correctly judges whether or not a syntactic transformation is semantically equivalent to a given sentence or when one correctly judges whether or not the propositional content of a sentence can be deduced from that of a preceding sentence. In some contexts, either of these operations may require one to accurately identify anaphoric referents. The present investigation i s intended to .assess the specific contribution of two linguistic s k i l l s — (1) recognition of semantic equivalence between syntactic transformations or a grammatical implication relation from one sentence to another, and (2) identification of anaphoric referents — to a criterion measure of text comprehension. The study also seeks to determine whether the same order of relative d i f f i c u l t y among designated transformations and anaphoric types prevails for younger and older students and for native speakers and those who have acquired English as a second language. Statement of the problem What one perceives in attending to a text is the surface structure of sentences. To comprehend a sentence one must understand, at the very least, the relation of any logical subject and logical object (arguments) 2 to a p r e d i c a t e . Such un d e r l y i n g r e l a t i o n s h i p s are o f t e n expressed only i n the deep st r u c t u e of a sentence ( c f . Chomsky, 1965). Hence, the comprehen-s i o n process may be described as an act of deep s t r u c t u r e recovery. I t i s only be recovering the deep s t r u c t u r e that one i s able to recognize i m p l i c a t i o n s which are dependent upon the p r o p o s i t i o n s that a r i s e from the grammatical (argument - predicate) r e l a t i o n s of elements w i t h i n a sentence. Deep s t r u c t u r e recovery i s t h e r e f o r e a r e q u i s i t e f o r comprehen-s i o n of sentences and consequent r e c o g n i t i o n of grammatically based i m p l i -c a t i o n r e l a t i o n s . I t can be f u r t h e r claimed that r e c o g n i t i o n of i m p l i c a t i o n i s necessary f o r t e x t comprehension. T y p i c a l l y , t e s t s of text comprehension seek to assess one's a b i l i t y to d e r i v e v a l i d conclusions which are, i n f a c t , i m p l i c a t i o n s of s t a t e d p r o p o s i t i o n s of sentences w i t h i n the t e x t . In non-constructed response formats, one i s confronted w i t h a statement (e.g., Jack was t o l d to h u r r y . ) , the t r u t h value of which must be evaluated i n l i g h t of the p r o p o s i t i o n a l content of e a r l i e r sentences (e.g., The woman c a l l e d to Jack and J i l l , "Hurry!"). This task c l e a r l y n e c e s s i -t a t e s the r e c o g n i t i o n of grammatically based i m p l i c a t i o n i n order to compare the deep s t r u c t u r e s of two sentences that d i f f e r i n surface s t r u c t u r e . Given t h i s view of comprehension of a t e x t as r e c o g n i z i n g gramma-t i c a l i m p l i c a t i o n through deep s t r u c t u r e recovery, to what extent do the l i n g u i s t i c s k i l l s of r e c o g n i z i n g grammatical i m p l i c a t i o n and i d e n t i f y i n g anaphoric r e f e r e n t s c o n t r i b u t e to performance, f o r i n s t a n c e , on a t e s t of reading comprehension i n which each item i s t r a n s f o r m a t i o n a l l y l i n k e d to a s p e c i f i c sentence, or sentences, i n the accompanying passage? The format proposed r e q u i r e s that one judge a statement to be s e m a n t i c a l l y c o n s i s t e n t , 3 contradictory, or indeterminate to the truth value of the related sentence(s) in the text. It w i l l be assumed throughout that one's failure to understand a sentence results from d i f f i c u l t i e s of language comprehension in general rather than of reading comprehension in particular. In appraising the contribution of the designated linguistic s k i l l s — grammatical implication and anaphora — to the proposed measure of text comprehension, two questions of practical interest might be addressed: Are students who have undertaken an E.S.L. programme in the elementary or secon-dary school as proficient in these linguistic s k i l l s as their native speaker peers? If not, does this place the E.S.L. student at a disadvantage on tests of text comprehension of the sort proposed to reflect deep structure recovery? Further, by systematically defining a set of transformational and anaphora recognition tasks, i t is possible to establish a hierarchical order of d i f f i c u l t y among the forms selected. On the basis of empirical findings, can such a hierarchy be accounted for by reference to transformational complexity? Are there notable differences in prevailing hierarchies for native speakers and E.S.L. students? Can these differences be explained in terms of current curricula in the teaching of English as a second language? The research described in this report is intended to provide some tentative answers to these questions — findings which may inform the design of more effective programmes to develop text comprehension. Importance of the investigation It is generally recognized that the a b i l i t y to comprehend discourse i s crucial to academic success. While, for example, innumerable components of the reading process have been proposed and examined at length, i t is only 4 r e c e n t l y that educators have begun to focus upon the r o l e of syntax as a s i g n i f i c a n t f a c t o r i n determining comprehension d i f f i c u l t y . Studies i n t h i s area should e s t a b l i s h the fundamental importance of r e c o g n i t i o n of grammatical i m p l i c a t i o n and anaphora r e l a t i o n s i n the o v e r a l l comprehension s t r a t e g y . I f s y n t a c t i c complexity l i m i t s the n a t i v e speaker's a b i l i t y to comprehend a sentence, i t i s l i k e l y that the problem i s more acute f o r the E.S.L. student. In the context of immigrant education, there a r i s e s a grave moral i s s u e whereby the r e a l i z a t i o n of s o c i a l and economic a s p i r a -t i o n s of ethnic m i n o r i t i e s i s l a r g e l y dependent on the a b i l i t y to under-stand discourse c o n t a i n i n g i n i t s s t r u c t u r e l o g i c a l r e l a t i o n s h i p s — i m p l i c a t i o n s , p r e s u p p o s i t i o n s , entailments — which the n a t i v e speaker by v i r t u e of h i s language competence comprehends i n t u i t i v e l y . Adequate understanding of the content of discourse i s a l s o consequential to the f o r e i g n student whose career goals demand that he read p r o f e s s i o n a l l i t e r a t u r e published i n E n g l i s h and be able to use the language to.confer w i t h n a t i v e speaker colleagues. Both the immigrant and the f o r e i g n student may be hindered from a t t a i n i n g some important t e x t comprehension s k i l l s through a language teaching methodology that r e f l e c t s the concentrated e f f o r t s of e a r l i e r l i n g u i s t s and language teachers to b r i n g the student to greater l e v e l s of f l u e n t performance. Worthy as t h i s o b j e c t i v e may be, such an approach g e n e r a l l y f a i l s to help the student appreciate how numerous s y n t a c t i c patterns presented i n i s o l a t i o n f o r mastery are s e m a n t i c a l l y r e l a t e d to each other. I t has been asserted i n the opening statement that t h i s know-ledge i s necessary f o r deep s t r u c t u r e recovery, i s c e r t a i n l y a r e q u i s i t e f o r r e c o g n i t i o n of grammatical i m p l i c a t i o n , and cannot be ignored when 5 a n a l y z i n g the act of comprehending a t e x t . More pragmatic c o n c e p t u a l i z a t i o n s of second language l e a r n i n g w i l l need to address t h i s problem by f o c u s i n g student a t t e n t i o n on the semantic i n t e r p r e t a t i o n s of s y n t a c t i c s t r u c t u r e s . Teachers who have come to appreciate the importance of s y n t a c t i c -semantic r e l a t i o n s to comprehension have f u r t h e r reason to be d i s t r e s s e d by the s t a t e of current t e s t i n g procedures that seldom define which grammatical r e l a t i o n s are being examined. Lacking t h i s d e f i n i t i o n , o rdinary comprehen-s i o n t e s t s must be seen as c o l l e c t i o n s of items whose d e f i n i t i o n s are rooted e n t i r e l y i n the mind of the t e s t author and, t h e r e f o r e , not o b j e c t i v e l y v e r i f i a b l e . Such t e s t s are useless to a teacher who wants to know which s o r t s of s y n t a c t i c a l l y based r e l a t i o n s h i s student has f a i l e d to comprehend, inf o r m a t i o n e s s e n t i a l f o r remedying d e f i c i e n c i e s i n the fundamental s k i l l of sentence comprehension. One aim of the present study, i s to extend e a r l i e r research i n developing d i a g n o s t i c instruments that promote purposeful i n s t r u c t i o n . In arguing the importance of a heretofore neglected element of English, language l e a r n i n g , t h i s study c i t e s weaknesses i n i n s t r u c t i o n a l and t e s t i n g p r a c t i c e s . Through i t s examination of the c o n t r i b u t i o n of c e r t a i n l i n g u i s t i c s k i l l s , a b e t t e r understanding of some of the components of comprehension, may be achieved. This knowledge, based upon data from E.S.L. students and n a t i v e speakers, may provide i n s i g h t s that could e v e n t u a l l y improve the q u a l i t y of i n s t r u c t i o n a v a i l a b l e to both. Overview of the study In a review of the l i t e r a t u r e , Chapter I I , evidence f o r the r e l a -t i o n s h i p between syntax and reading comprehension i s c i t e d from a number of d i f f e r e n t p e r s p e c t i v e s . This i s followed i n Chapter I I I by a d e s c r i p t i o n 6 of each facet of the present instrument drawing comparisons with previous attempts to measure understanding of syntactic and anaphoric r e l a t i o n s . Requirements for comprehension testing are proposed through a model which seeks to. r u l e which cognitive acts are i n d i c a t i v e of comprehending the text at hand. In Chapter IV, the research design i s presented and the hypo-theses stated. A r a t i o n a l e i s offered for t e s t i n g each hypothesis. Chapter V outlines the procedures taken to assemble the t e s t s , s e l e c t s u i t a b l e Ss, administer and score the instrument. Af t e r reviewing the s t a t i s t i c a l properties of the instrument, Chapter VI examines each hypothesis, draws a conclusion, and b r i e f l y discusses the possible causes and p r a c t i c a l implications of the fi n d i n g s . Results of the study are summarized i n the f i n a l Chapter VII where are noted d i r e c t i o n s for further research, recommendations for a l t e r n a t i v e t e s t i n g procedures, and some implications for second language i n s t r u c t i o n . Limitations of the study Circumstances i n the research s e t t i n g forced the investigator to r e l y l a r g e l y upon teacher report for the s e l e c t i o n of s u i t a b l e Ss. The s p e c i f i c c r i t e r i a employed are outlined i n the section describing sample populations.. Information regarding I. Q. scores, measured achievement i n reading comprehension and other English language s k i l l s , and records of past performance i n other academic areas i s not usually a v a i l a b l e to researchers. To administer such measures i n addition to the study i n s t r u -ment would have placed an unreasonable demand upon the time of the p a r t i -cipants at the close of the school year. 7 Another d i f f i c u l t y encountered by the i n v e s t i g a t o r was that of access to an ample number of Ss i n a l l three n a t i v e language c a t e g o r i e s . As a r e s u l t , the sample s i z e of n a t i v e speakers of E n g l i s h at the elemen-t a r y and j u n i o r secondary i n t e r v a l s and of non-Chinese E.S.L. l e a r n e r s at a l l grade i n t e r v a l s i s admittedly s m a l l . A t t e n t i o n i s given to comparing Ss w i t h regard to the f a c t o r s of n a t i v e language and grade placement. One expectation i s that n a t i v e speakers w i l l perform s u p e r i o r l y to E.S.L. students on a l l t a s k s . Another i s that both f i r s t and second language l e a r n e r s w i l l e x h i b i t a p a t t e r n of augmented scores as a f u n c t i o n of higher grade placement. The l a t t e r i n c u r s a problem of v a l i d comparison across grade i n t e r v a l s . I d e a l l y , a l o n g i t u d i n a l study i s best s u i t e d to i n v e s t i g a t i n g any p a t t e r n of growth. The present study uses a c r o s s - s e c t i o n a l model which may be v a l i d i n the case of n a t i v e speakers as i t may be assumed t h a t , i n ge n e r a l , Ss at higher grade i n t e r v a l s possessed, at an e a r l i e r p o i n t i n t h e i r language develop-ment, a c o n s t e l l a t i o n of l i n g u i s t i c s k i l l s approximating that of Ss at lower grade i n t e r v a l s . On t h i s b a s i s , one may draw inferences from the data regarding the gradient of l e a r n i n g between those grade i n t e r v a l s s e l e c t e d and defined f o r t h i s study. The foregoing assumption, however, becomes untenable when a p p l i e d to an E.S.L. popu l a t i o n . T y p i c a l l y , immigrant students enter the host school system at v a r i o u s ages, spend approximately one year i n a s p e c i a l language t r a i n i n g c l a s s , and are then placed i n r e g u l a r c l a s s e s w i t h t h e i r . p e e r s . I t i s do u b t f u l that a l l immigrant students gain a s i m i l a r c o n f i g u r a t i o n of language l e a r n i n g through t h i s s o r t of experience. Another f a c t o r accounting f o r v a r i a t i o n i n l e a r n i n g among any sample of immigrants i s the extent and nature of the i n d i v i d u a l ' s e d u c a t i o n a l e x p e r i -ence i n h i s country of o r i g i n which may, or may not, have included E n g l i s h 8 as e i t h e r a subject or medium of i n s t r u c t i o n . Previous s c h o o l i n g of students from d i v e r s e backgrounds i s d i f f i c u l t to assess i n constant and o b j e c t i v e terms and the r e f o r e could not be c o n t r o l l e d i n t h i s study. Since the m a j o r i t y of E.S.L. p a r t i c i p a n t s had r e s i d e d i n Canada f o r l e s s than two years, i t cannot be s a i d that Ss at higher grade i n t e r v a l s would have been accepted as s u i t a b l e Ss f o r lower grade i n t e r v a l s a t a p o i n t e a r l i e r i n t h e i r c h r o n o l o g i c a l growth. Consequently, one must be wary of any attempt to draw conclusions about the gradients of change i n performance across grade i n t e r v a l s w i t h i n a non-native p o p u l a t i o n . A f i n a l l i m i t a t i o n to be kept i n mind a r i s e s from the p r a c t i c a l need to r e s t r i c t the number of transformations and anaphoric types to be test e d and to s e l e c t a p a r t i c u l a r format to t e s t t e x t comprehension. Conclusions can be based only upon the r e l a t i o n s h i p between f a c i l i t y i n the tasks i n c l u d e d i n the present instrument and performance on t e x t compre-hension tasks of a s p e c i f i e d nature which may depart from other, more conventional, means of measuring t e x t comprehension. 9 Chapter I I REVIEW OF THE LITERATURE In t h i s chapter, a number of w e l l e s t a b l i s h e d p r i n c i p l e s are r e c a l l e d i n order to place the present study i n the context of recent research i n the area of s y n t a c t i c comprehension as r e l a t e d to r e a d a b i l i t y and growth i n reading p r o f i c i e n c y . U n l i k e the present study whose domain i s language comprehension i n general, sources c i t e d h e r e i n are p a r t i c u l a r l y concerned w i t h reading comprehension. Nevertheless, i t i s considered that many i n s i g h t s gained from these i n v e s t i g a t i o n s can be g e n e r a l i z e d to compre-hension of both w r i t t e n and spoken t e x t s . Further a p p l i c a b i l i t y l i e s i n the f a c t that the problem at hand i s examined through the use of p r i n t e d mate-r i a l s . How the r e s u l t s of the present study r e l a t e to c e r t a i n of these e a r l i e r i n v e s t i g a t i o n s w i l l be o u t l i n e d i n the f i n a l chapter of t h i s r e p o r t . Sentence comprehension i s something more than word comprehension. P s y c h o l i n g u i s t s have given considerable a t t e n t i o n to the funda-mental r o l e of syntax i n comprehension. Perhaps the most obvious t h i n g we can say about the s i g n i f i c a n c e of a sentence i s that i t i s not given as the l i n e a r sum of the s i g n i f i c a n c e of the words that comprise i t . . . . t h e words i n a sentence i n t e r a c t . ( M i l l e r , 1965:16) The r u l e s f o r t h i s i n t e r a c t i o n of words i n sentences are set by the syntax of a given language; the outcome of t h i s i n t e r a c t i o n i s meaning. (Cooper and Petrosky, 1976:187) I m p l i c a t i o n s f o r d e r i v i n g meaning from the reading process can be drawn from the foregoing p r o p o s i t i o n s : 10 Since p s y c h o l i n g u i s t i c s a s s e r t s that the goal of f l u e n t reading i s the i d e n t i f i c a t i o n of meaning... which r e l i e s h e a v i l y on the b r a i n ' s a b i l i t y to bridge surface s t r u c t u r e and deep s t r u c t u r e w i t h s y n t a c t i c a l r u l e s , i t i s obvious that great weight i n t h i n k i n g about the reading process must be put on the r e l a t i o n s h i p of words i n sentences. (Cooper and Petrosky, i b i d . ) E m p i r i c a l support f o r t h i s p o i n t of view i s presented i n an e a r l y study by Gibbon (1941) which, through the use of a "disarranged [phrase t e s t " , e s t a b l i s h e d among a Grade 3 p o p u l a t i o n a high c o r r e l a t i o n (.89) between the a b i l i t y to perceive r e l a t i o n s h i p s between p a r t s of a sentence and the a b i l i t y to understand the sentence, when i n t e l l i g e n c e i s p a r t i a l l e d out. A l s o , a s i g n i f i c a n t c o r r e l a t i o n was demonstrated between the a b i l i t y to see s y n t a c t i c r e l a t i o n s h i p s and t o t a l reading achievement. More recent i n v e s t i -gations conducted by Cromer (1970); Oakan, Wiener, and Cromer (1971); and S t e i n e r , Wiener, and Cromer (1971) i n d i c a t e that the poor comprehension of some readers i s not due to weak s k i l l s i n word i d e n t i f i c a t i o n but to an i n a b i l i t y to i n t e g r a t e the meanings of separate words '•.to., a r r i v e at the meaning of an e n t i r e sentence. The relevance of these e m p i r i c a l observations, p a r t i c u l a r l y to the i n s t r u c t i o n of E.S.L. students, should be c l e a r : A method of teaching reading which stops w i t h r e c o g n i t i o n of words... assumes that the p u p i l s have acquired the a b i l i t y to supply the proper grammatical components of meaning as they have learned to speak the language. When c h i l d r e n are already f l u e n t speakers of E n g l i s h and are f a m i l i a r w i t h the standard usages, word recog-n i t i o n may be enough...Something more than word r e c o g n i t i o n i s indicated...when c h i l d r e n progress to w r i t t e n m a t e r i a l which i s s t r u c t u r a l l y d i f f e r -ent from c o n v e r s a t i o n a l E n g l i s h . (Ives, 1964:180) 11 To develop the reading comprehension of the E.S.L. student whose exposure to E n g l i s h i s p r i m a r i l y i n the domain of peer i n t e r a c t i o n and/or i n s t r u c t i o n a l m a t e r i a l s that emphasize the spoken language, teachers must come to recog-n i z e the c r u c i a l r o l e of syntax i n d e r i v i n g accurate meaning from w r i t t e n discourse as w e l l . S y n t a c t i c u n i t s are, used to organize sentence perception. M i l l e r and I s a r d (1963) demonstrated that words i n grammatical sentences are e a s i e r to perceive a u r a l l y than words connected i n ungrammat-i c a l . s t r i n g s . . Morton (1964) demonstrated a steady increase i n speed of reading as a s e r i e s of words approximates normal s y n t a c t i c p a t t e r n s . E p s t e i n (1961) showed that " s y n t a c t i c s t r u c t u r e f a c i l i t a t e s v e r b a l l e a r n i n g apart from the c o n t r i b u t i o n s of meaningfulness, f a m i l i a r i t y , and s e q u e n t i a l p r o b a b i l i t y " (p. 85). Wisher (1976) discovered that beforehand knowledge of the s y n t a c t i c s t r u c t u r e to be used f a c i l i t a t e s performance on a memory task and decreases the time r e q u i r e d to read the sentence. K o l e r s (1970) reported that 70% of the o r a l reading s u b s t i t u t i o n s of a d u l t s conform to the same par t of speech as the c o r r e c t word i n the t e x t . This i s i n d i c a t i v e of an i n t u i t i v e awareness by the reader of s y n t a c t i c c o n s t r a i n t s . In the context of l e a r n i n g to read, Goodman and Burke (1969) have noted that more than 60% of the o r a l miscues of elementary school c h i l d r e n observed could be described as retransformations of the t e x t sentence r a t h e r than simply anomalous s t r i n g s . A study by Weber (1970) among Grade 1 p u p i l s recorded that 90% of the o r a l reading e r r o r s d i d not c o n t a i n s y n t a c t i c v i o l a t i o n s , suggesting that beginning readers may be over r e l i a n t on t h e i r s t i l l l i m i t e d knowledge of syntax. Further evidence of the i n f l u e n c e of syntax on reading behaviour i s o f f e r e d by Rode (1974) who showed that the 12 eye-voice span — the number of words that the eye i s ahead of the v o i c e i n o r a l reading — among p u p i l s i n Grades 3, 4, and 5 i s i n some ways de t e r -mined by s y n t a c t i c s t r u c t u r e . S y n t a c t i c comprehension i s c h a r a c t e r i z e d by developmental trends. The extensive research of O'Donnell, G r i f f i n , and N o r r i s (1967) a f f i r m s the general n o t i o n of a developmental sequence of s y n t a c t i c a c q u i -s i t i o n throughout the elementary grades. These i n v e s t i g a t o r s found that some transformations (e.g., r e l a t i v e clauses) were used much more f r e q u e n t l y by kindergarten youngsters w h i l e other s t r u c t u r e s (e.g., noun m o d i f i c a t i o n by a p a r t i c i p l e ) appear i n the language of older c h i l d r e n . Ruddell (1969:11) i n t e r p r e t s such f i n d i n g s as l o g i c a l from the standpoint of t r a n s f o r m a t i o n a l -generative grammar i n that many of the l a t e r c o n s t r u c t i o n s are derived from more complex d e l e t i o n r u l e s . Marcus (1971) devised "A Test of Sentence Meaning" f o r Grades 5 through 8. Results show a p a t t e r n of improved performance as a f u n c t i o n of grade l e v e l . Such f i n d i n g s are c o n s i s t e n t w i t h those reported by C a r r o l l (1970), Smith (1970), and Tatham (1970). School aged c h i l d r e n are unable to comprehend many s y n t a c t i c s t r u c t u r e s . Bormuth et a l . (1970), having tested more than 240 Grade 4 students' comprehension of a wide v a r i e t y of sentence s t r u c t u r e s ( i n c l u d i n g n o m i n a l i -z a t i o n , r e l a t i v i z a t i o n , s u b o r d i n a t i o n , and anaphora) concluded that t h e i r sample p o p u l a t i o n "showed an unexpectedly low l e v e l of performance on these s k i l l s which seemed both very simple and very b a s i c " (p. 349). Marcus (1971) and Takahashi (1975) have demonstrated that among students i n upper elementary 13 and j u n i o r secondary grades, no group as a whole appears to have completely mastered an understanding of the t r a n s f o r m a t i o n a l types i n c l u d e d i n "A Test of Sentence Meaning"; pronominal :reference, d e l e t i o n s , embeddings, and conjunctions are c i t e d as p a r t i c u l a r sources of d i f f i c u l t y . In a more s p e c i a l i z e d study, Stoodt (1970) a l s o found Grade 4 students f r e q u e n t l y encountered d i f f i c u l t y i n t e r p r e t i n g sentences c o n t a i n i n g c e r t a i n conjunctions. A l l of the foregoing s t u d i e s g e n e r a l l y confirm the e a r l i e r obser-v a t i o n s of Chomsky (1969) who determined t h a t , at a given age, not a l l c h i l d r e n are able to demonstrate the same l e v e l of mastery of s e l e c t e d s y n t a c t i c s t r u c t u r e s . Some s y n t a c t i c p a t t e r n s are e a s i e r to comprehend than others. Research i n t h i s area i s extensive and r e q u i r e s c a r e f u l a n a l y s i s and i n t e r p r e t a t i o n . For developmental and s o c i o l i n g u i s t i c reasons, the syntax of o r a l language can be considered to be both more f a m i l i a r and t r a n s f o r m a t i o n a l l y simpler than many of the patterns encountered i n w r i t t e n d i s c o u r s e . There-f o r e , many i n v e s t i g a t o r s have attempted to r e l a t e aspects of o r a l syntax to reading comprehension. Tatham (1970) concluded that Grade 2 and Grade 4 c h i l d r e n are b e t t e r able to comprehend s y n t a c t i c patterns i n reading i f those patterns are f r e q u e n t l y used i n t h e i r o r a l language. Ruddell (1969) compared the e f f e c t on reading comprehension of w r i t t e n patterns of language s t r u c t u r e which occur w i t h high and low frequency i n c h i l d r e n ' s o r a l l a n g -uage. Among a sample of Grade 4 p u p i l s , reading comprehension scores on passages w r i t t e n w i t h high frequency p a t t e r n s were found to be s i g n i f i c a n t l y s u p e r i o r to comprehension scores on passages that contained low frequency p a t t e r n s . S i m i l a r l y , Reid (1972) rewrote b a s a l reader sentences to match 14 the spoken syntax of her 7 and 8 year o l d S_s. A comparison of comprehension between the o r i g i n a l and the r e w r i t t e n m a t e r i a l revealed the l a t t e r to be s i g n i f i c a n t l y e a s i e r f o r these c h i l d r e n to understand. F i n a l l y , Vogel (1975) c o n c l u s i v e l y demonstrated that d y s l e x i c c h i l d r e n are f r e q u e n t l y d e f i c i e n t i n o r a l language syntax. The f a c i l i t a t i n g e f f e c t s of o r a l syntax w e l l e s t a b l i s h e d , there remains the question of why t h i s i s so. Evidence can be found i n the l i t e r -ature to support the case f o r e i t h e r t r a n s f o r m a t i o n a l s i m p l i c i t y or f a m i l i -a r i t y . A touchstone study by Coleman (1964) e s t a b l i s h e d that reading a "detransformed" t e x t i n which n o m i n a l i z a t i o n s , p a s s i v i z a t i o n s , r e l a t i v e c l a u s e s , and grammatical d e l e t i o n s were not present r e s u l t e d i n s i g n i f i c a n t l y higher c l o z e t e s t scores among a group of 48 c o l l e g e students. Fagan (1969) a l s o observed s i g n i f i c a n t l y higher c l o z e t e s t performance when elementary students i n Grades 4, 5, and 6 read t e x t s that d i d not contai n r e l a t i v e clauses or grammatical d e l e t i o n s . Evans (1973) demonstrated the s u p e r i o r comprehension of s i m p l i f i e d (de-transformed) prose on m u l t i p l e choice t e s t s as w e l l as c l o z e measures. I t i s worth n o t i n g , however, that the Grade 12 £>s i n t h i s study were p r e v i o u s l y i d e n t i f i e d as reading at three to f i v e years below grade l e v e l . (Goodman and Burke, 1973, c l a i m that d i f f e r e n c e s i n a b i l i t y to handle complex syntax disappear among readers of moderate to high p r o f i c i e n c y . Evans recognizes the s i g n i f i c a n c e of t h i s f a c t o r i n hypothe-s i z i n g that problem readers w i l l r a i s e t h e i r comprehension by reading t r a n s -f o r m a t i o n a l l y simpler prose.) Dealing w i t h a more s p e c i f i c problem i n s y n t a c t i c comprehension, Richek (1976) r e q u i r e d p u p i l s i n Grades 3, 4, and 5 to i d e n t i f y subordinate clause subjects w i t h two l e v e l s of MDP (minimal distance p r i n c i p l e ) , conforming and v i o l a t i n g , and complexity, f o l l o w i n g and i n t e r r u p t i n g statements. That i n v e s t i g a t o r r e p o r t s : " S i g n i f i c a n t main e f f e c t s were found f o r MDP (conforming^ sentences were e a s i e r ) and complex-i t y ( f o l l o w i n g statements were e a s i e r ) . The MDP and complexity v a r i a b l e s formed a s i g n i f i c a n t i n t e r a c t i o n . The MDP v i o l a t i n g sentences produced performance c h a r a c t e r i s t i c of short term memory tasks, making complex sentences which separate subject and subordinate clause by s e v e r a l words, d i f f i c u l t to process." (p. 800). Results of the four preceding s t u d i e s support the case f o r t r a n s -f o r m a t i o n a l s i m p l i c i t y as a key f a c t o r i n comprehension. More problematic i s an i n v e s t i g a t i o n by P e l t z (1974) which r e q u i r e d 34 Grade 10 ^ s to w r i t e a page on s o c i a l s t u d i e s content. Their w r i t i n g was compared w i t h the syntax of t h e i r textbook and a t e x t passage was r e w r i t t e n to conform to the students' syntax. j[s' comprehension of the o r i g i n a l and the " s i m p l i f i e d " passages were then compared. Findings i n d i c a t e a s i g n i f i c a n t d i f f e r e n c e i n favour of the s i m p l i f i e d v e r s i o n when comprehension was measured by means of a c l o z e t e s t . Use of a m u l t i p l e choice format, however, showed no s i g n i f i -cant d i f f e r e n c e i n comprehension between the o r i g i n a l and r e w r i t t e n v e r s i o n s . Reviewing the methodology used i n t h i s study, i t i s debatable whether one v e r s i o n may have been e a s i e r because i t s grammatical transformations were simpler or because i t s syntax was more f a m i l i a r to the Ss. At l e a s t two other s t u d i e s , however, tend to support the case f o r f a m i l i a r i t y of s y n t a c t i c s t r u c t u r e e f f e c t i n g b e t t e r comprehension. Smith (1970) presented students i n Grades 4 through 12 w i t h four c l o z e t e s t s e x h i b i t i n g s y n t a c t i c c h a r a c t e r i s t i c s of the w r i t t e n productions of students i n Grades 4, 8, and 12 and s k i l l e d a d u l t s . R e sults i n d i c a t e d that students i n Grades 4, 5, and 6 read Grade 4 syntax best; Grade 11 students read i t w i t h l e a s t f a c i l i t y . Students i n Grades 8 through 12 found Grade 8 s y n t a c t i c 16 patterns easier to read than e i t h e r Grade 4 w r i t i n g or the more d i f f i c u l t passages. Pearson (1974) gathered evidence from the performance of c h i l d r e n i n Grades 3 and 4 which tends to refute the recommendation that the d i f f i -c u l t y of written discourse can be reduced by eliminating subordinating constructions or reducing sentence length. When the semantic r e l a t i o n i s held constant and when the test question i s relevant to the r e l a t i o n whose form i s v a r i e d , e i t h e r comprehension i s equally e f f i c i e n t across forms or else the more subordinate and longer sentence forms e l i c i t better comprehension, (p. 189) Further research seems necessary to determine more: p r e c i s e l y how s y n t a c t i c f a m i l i a r i t y and complexity i n t e r a c t i n comprehending a text and to define those conditions under which one f a c t o r becomes more prominent than the other. Syntactic comprehension contributes to more generalized measures of reading  comprehension. Studies conducted at every grade l e v e l sustain t h i s proposition. Harris (1975) established that f o r Grade 2 pupils performance on an o r a l syntax test showed a high c o r r e l a t i o n with scores on a standardized measure of reading achievement (.70). This c o r r e l a t i o n was s i g n i f i c a n t l y greater than the c o r r e l a t i o n between measures of reading achievement and i n t e l l i -gence (.57). The importance of syntactic attainment as a predictor of reading achievement at t h i s age l e v e l has also been demonstrated by Vogel (1975) i n which the proportion of unique variance contributed by nine measures of syntax to a c r i t e r i o n standardized test of reading achievement amounts to 53.1% for mormal (non-dyslexic) Ss, an impressive r e s u l t consi-dering the need of young readers to consciously attend to other tasks — 17 e.g., decoding, word r e c o g n i t i o n . Stoodt (1970) observed that among Grade 4 students, there i s a s i g n i f i c a n t r e l a t i o n s h i p between reading comprehension and understanding conjunctions. Hart (1971) administered a sentence combining t e s t to Grade 6 p u p i l s i d e n t i f i e d as reading at a Grade 3 l e v e l and determined that reading comprehension scores were r e l a t e d to the a b i l i t y to produce sentences that c a r r y the i n f o r m a t i o n a l l o a d of s e v e r a l k e r n e l sentences. S i m i l a r l y , Kuntz (1975) demonstrated a c l o s e r e l a t i o n s h i p between Grade 7 students' perform-ance on a w r i t t e n sentence c o n s t r u c t i o n t e s t and t h e i r scores on a widely used standardized reading achievement t e s t . D i f f e r e n c e s between t y p i c a l and d e b i l i t a t e d readers w i t h respect to s y n t a c t i c c a p a b i l i t y have been e s t a b l i s h e d i n a number of i n s t a n c e s . Vogel (1975) has shown the . s y n t a c t i c comprehension of Grade 2 male d y s l e x i c s to be i n f e r i o r to t h e i r normal achi e v i n g peers. Takahashi (1975) used "• "A Test of Sentence Meaning" (Marcus, 1971) to compare able and poor Grade 9 readers on t h e i r a b i l i t y to i n t e r p r e t sentences. On the b a s i s of a s i g n i f i -cant d i f f e r e n c e between these groups, she concluded that "comprehension of s y n t a c t i c s t r u c t u r e i s an element i n t o t a l reading comprehension." (p. 60) Cox (1976) developed a t e s t of s y n t a c t i c complexity and found that a d u l t beginning readers e n r o l l e d i n a " b a s i c education" programme performed s i g n i -f i c a n t l y i n t e r i o r l y to l i t e r a t e a d u l t s . A study conducted by van Metre (1974) compared some of the l i n g u i s t i c competencies of b i l i n g u a l Grade 3 p u p i l s drawn from the top and bottom q u a r t i l e s on a standardized reading achievement t e s t w i t h matched groups of monolinguals. In o r a l i n t e r v i e w s , a l l Ss were teste d f o r comprehension of four s y n t a c t i c s t r u c t u r e s described by Chomsky (1969) — ask (query) / t e l l , promise / t e l l , easy to see, and pronominal-i z a t i o n . Findings revealed that greater d i f f e r e n c e s occur between high and low reading achievers,than between b i l i n g u a l s and monolinguals. These r e s u l t s are p a r t i c u l a r l y i n t e r e s t i n g not only because they confirm the importance of s y n t a c t i c comprehension, f o r o v e r a l l reading achievement but a l s o f o r t h e i r p r a c t i c a l i m p l i c a t i o n s i n suggesting p o s s i b l e causes of a second language student's u n s a t i s f a c t o r y . p r o g r e s s i n l e a r n i n g to :. read E n g l i s h . Syntax i s an important v a r i a b l e i n determining r e a d a b i l i t y . The preceding arguments concerning the d i s t i n c t i o n between sentence comprehension and- word comprehension and the o r g a n i z i n g f u n c t i o n of syntax have not h i s t o r i c a l l y i n f l u e n c e d the c o n s t r u c t i o n of r e a d a b i l i t y formulas. B o t e l and Granowsky (1972) note that sentence l e n g t h i s the only s y n t a c t i c measure i n many wid e l y used formulas — e.g., D a l e - C h a l l (1948), Spache ( 1 9 5 3 ) — and argue that a s y n t a c t i c a n a l y s i s based on t r a n s f o r -mational grammar i n d i c a t e s the complexity of a sentence should not be judged s o l e l y on a word count of the surface s t r u c t u r e of the sentence. By i d e n t i -f y i n g s p e c i f i c l i n g u i s t i c v a r i a b l e s , Bormuth (1966) has claimed to have accounted f o r a f a r greater p r o p o r t i o n of vari a n c e i n comprehension d i f f i -c u l t y than was p o s s i b l e w i t h e a r l i e r formulas. Granowsky (1973) a l s o .. reports an a p p l i c a t i o n of t r a n s f o r m a t i o n a l grammar theory to the development of a s y n t a c t i c complexity formula which promises to be a more r e l i a b l e and v a l i d guide to determining r e a d a b i l i t y . Chapter I I I DEVELOPMENT OF THE INSTRUMENT This chapter describes the und e r l y i n g l i n g u i s t i c concepts which form the b a s i s f o r each task comprising the present instrument. TESTING SYNTACTIC COMPREHENSION The f i r s t p a r t of the instrument examines understanding of s y n t a c t i c r e l a t i o n s w i t h i n a s i n g l e sentence or between two sentences i s o l a t e d from any l a r g e r context. Studies reviewed i n the preceding chapter c l a i m that s k i l l i n s y n t a c t i c comprehension c o n t r i b u t e s to success i n more gen e r a l i z e d measures of reading comprehension. This aim r a i s e s the question of p r e c i s e l y how i t can be demonstrated that one has understood a sentence. In responding to comprehension t e s t items, i t has been suggested that the reader bases h i s guesses on as few l e x i c a l , s t r u c t u r a l , and graphic clues as p o s s i b l e , aided by the f a c t that language i s redundant and s e q u e n t i a l (Goodman and Burke, 1969). Hence, a p u p i l may generate or s e l e c t a c o r r e c t response c o i n c i d e n t a l l y upon such minimal cl u e s . More l i m i t e d than the n a t i v e speaker i n h i s knowledge of the language, the E.S.L. student may be even more dependent on s u p e r f i c i a l c l u es from the surface of the t e x t . A growing a p p r e c i a t i o n of t h i s problem has given r i s e to new demands f o r o p e r a t i o n a l l y based t e s t s which c l e a r l y demonstrate what l i n g u i s t i c p r o p e r t i e s are being tes t e d (Bormuth, 1970; Mohan, 1973). Grammatical i m p l i c a t i o n In the context of t h i s study, the term "grammatical i m p l i c a t i o n " i s r e s t r i c t e d to those instances wherein the p r o p o s i t i o n a l content of one sentence i s i m p l i e d by that of another, preceding, sentence by v i r t u e of a r e o r d e r i n g of s y n t a c t i c surface elements. D e l e t i o n or i n s e r t i o n of elements may occur. Thus, r e c o g n i t i o n of i m p l i c a t i o n may be viewed as a matter of a c c u r a t e l y r e c o v e r i n g the deep s t r u c t u r e of each of two sentences which d i f f e r i n t h e i r surface s t r u c t u r e but have a defined t r a n s f o r m a t i o n a l r e l a t i o n s h i p . In the case of two-way i m p l i c a t i o n r e l a t i o n s the surface s t r u c t u r e of both sentences r e f l e c t s a n . i d e n t i c a l deep s t r u c t u r e , e.g., Harry ate the cake. The cake was eaten by Harry. The sentences i n a one-way i m p l i c a t i o n are c h a r a c t e r i z e d by both d i f f e r e n t surface s t r u c t u r e s and d i f f e r e n t deep s t r u c t u r e s . The i m p l i c a t i o n r e l a t i o n r e s t s on the deep s t r u c t u r e of the second sentence being c l o s e l y r e l a t e d to that of the f i r s t , e.g., Harry ate the cake. The cake was eaten. A n a t i v e speaker recognizes that i f the f i r s t sentence i s t r u e , then the second sentence i s t r u e , but not v i c e v e r s a . On the b a s i s of the foregoing s t i p u l a t i o n s , given the sentence (1) The b l a c k cat jumped over the fence, only a s y n t a c t i c paraphrase such as (2) The cat which i s black jumped over the fence. i s here considered a "grammatical i m p l i c a t i o n " although i t i s e q u a l l y necessary to recover the deep s t r u c t u r e of 21 (3) The black f e l i n e jumped over the fence, to recognize i t to be a l e x i c a l paraphrase of (1) or (4) The b l a c k cat went to the other s i d e of the fence, to judge i t a l o g i c a l i m p l i c a t i o n of (1). By l i m i t i n g "grammatical i m p l i c a t i o n " to s y n t a c t i c paraphrases i t becomes p o s s i b l e , through the c l a s s i f i c a t i o n s o f f e r e d by transforma-t i o n a l grammar, to define the l i n g u i s t i c p r o p e r t i e s any item purports to t e s t . Previous attempts to measure understanding of grammatical i m p l i c a t i o n The i n v e s t i g a t o r examined three recent research instruments which r e l a t e to grammatical i m p l i c a t i o n and may be considered as attempts to meet the requirements of s t a t i n g what aspects of l i n g u i s t i c competence are to be t e s t e d . A l l c l a i m t h e i r a n a l y s i s to be based on transforma-t i o n a l - generative grammar and seek i n various ways to evidence the.S_'s c a p a b i l i t y to determine the deep s t r u c t u r e of a given sentence. Simons (1970) devised the "Deep St r u c t u r e Recovery Test". Twenty-five items t e s t understanding of transformations mainly concerned w i t h s y n t a c t i c - semantic c o n t r a s t s among a set of t r a n s i t i v e and i n t r a n s -i t i v e verbs and the ask / t e l l d i s t i n c t i o n ( c f . Chomsky, 1969). Ss were asked to i n d i c a t e which of three .sentences..differs i n meaning from the other two. The f o l l o w i n g i s a sample item: * a) What the boy would l i k e i s f o r the g i r l to leave. b) For the boy to leave i s what the g i r l would l i k e . c) What the g i r l would l i k e i s f o r the boy to leave. The t e s t was administered to 87 Grade 5 p u p i l s w i t h a mean I Q of 117. On the average, these students c o r r e c t l y answered approximately 75% of the items. 22 On the b a s i s of a p o s i t i v e c o r r e l a t i o n between scores on t h i s t e s t and performance on a standardized t e s t of reading achievement (.48) and c l o z e measures (.73), Simons concluded that the a b i l i t y to recover the deep s t r u c t u r e of sentences i s an important aspect of reading comprehension. Marcus (1971) constructed "A Test of Sentence Meaning" to determine intermediate grade students' understanding of s y n t a c t i c clues to l i t e r a l meaning (p. 50). While s t r u c t u r a l i s t c a t e g o r i e s were used to i s o l a t e the types of s y n t a c t i c s t r u c t u r e s to be included i n the t e s t , t r a n s f o r m a t i o n a l theory was used i n developing the items r e l a t e d to s p e c i f i c s k i l l s of i n t e r p r e t a t i o n . From ca t e g o r i e s of m o d i f i c a t i o n , p r e d i c a t i o n , and c o o r d i n a t i o n , seventeen s y n t a c t i c s t r u c t u r e s were i d e n t i f i e d and s i x r e l a t e d items were presented f o r each. A number of formats are used. In the f i r s t example, the student has to f i n d the transformation that has the same meaning as the underlined sentence. The man gave the boy a puppy. a) The man gave away the boy's puppy. * b) The man gave a puppy to the boy. c) The boy gave a puppy to the man. d) The man gave a puppy away f o r the boy. An a l t e r n a t i v e format r e q u i r e s the student to s e l e c t one of four sentences that u t i l i z e s the vocabulary of the other three but d i f f e r s from them i n meaning. S t i l l other formats r e q u i r e the reader.; to ^ analyze a given s t r u c t u r e i n t o i t s b a s i c k e r n e l sentences. In the f o l l o w i n g example, one i s to choose the two sentences that combine to give the complete meaning of the underlined sentence. 23 Bob and Don ate the bread and j e l l y . * a) Bob and Don ate the bread. b) Bob ate the j e l l y . c) .Don ate the bread. * d) Bob and Don ate the j e l l y , e) Don ate the j e l l y . The t e s t was administered to 487 students i n Grades 5 through 8 i n both "disadvantaged" and middle c l a s s area schools. Results show a trend of .improved performance as a f u n c t i o n of higher grade l e v e l . Grade 5 p u p i l s averaged 60% c o r r e c t responses; Grade 8 students reached an average of 80%. An a n a l y s i s of e r r o r s r e v e a l s that "some students mistakenly thought that a c o i n c i d e n t a l noun - verb - noun sequence of words was a subject - verb - o b j e c t sequence and thus a k e r n e l sentence of the l a r g e r sentence." (p. 58). Such an e r r o r would be an instance of f a i l i n g to recover the deep s t r u c t u r e of e i t h e r the o r i g i n a l sentence or one or more of the options suggested to be equivalent i n meaning. Another i n t e r e s t i n g o b s e r v a t i o n , from the viewpoint of grammatical i m p l i c a t i o n as defined i n the present study, i s that some students apparently " d i d not d i s t i n g u i s h between denotated l i t e r a l meanings and i m p l i e d meanings." ( i b i d . ) No attempt was reported to c o r r e l a t e t e s t r e s u l t s w i t h performance on other measures of reading comprehension. O'Donnell (1973) recognized the need f o r a t e s t to measure aware-ness of s y n t a c t i c s t r u c t u r e without r e l y i n g on the terminology of grammar. To t h i s end, the "Perception of A l t e r n a t e S t r u c t u r e s Test" was devised using nonsense vocabulary to encourage r e l i a n c e on s y n t a c t i c , r a t h e r than l e x i c a l , cues to s t r u c t u r e . According to;;the author: 24 Of the t h i r t y items on the t e s t , two measure perc e p t i o n of the a c t i v e - passive a l t e r n a t i v e s , two of the i n d i r e c t o b j e c t - p r e p o s i t i o n a l phrase o p t i o n s , s i x the r e l a t i v e clause - reduced r e l a t i v e v a r i a t i o n s (prenominal a d j e c t i v e s , p a r t i -c i p l e phrase and a p p o s i t i v e ) and two the a d v e r b i a l clause - abridged a d v e r b i a l a l t e r n a t i v e s . S i x items...deal w i t h noun clauses - i n f i n i t i v e -gerund phrase v a r i a t i o n s , and the remainder of the items t e s t v a r i o u s combinations of the options l i s t e d above, (pp. 3-4). Each item contains three sentences, two of which are s i m i l a r i n underlying meaning. The student must i n d i c a t e which sentence i s d i f f e r e n t from the others. A sample item f o l l o w s : a) The b i r t l e scared the i l b i d . b) The i l b i d was scared by the b i r t l e . * c) The i l b i d scared the b i r t l e . The t e s t was administered to 87 Grade 9 students and 62 Grade 10 students, approximately h a l f of whom scored below the 3 5 % i l e on the c o g n i -t i v e a b i l i t i e s (verbal) s e c t i o n of a w i d e l y used standardized t e s t . On the average, Grade 9 students answered 44.3% of the items c o r r e c t l y , Grade 10 students, 50.5%. The study d i d not e s t a b l i s h any c o r r e l a t i o n between t h i s t e s t and other measures of reading comprehension. The present instrument Referred to elsewhere i n t h i s report as Task 1, t h i s p a r t of the present instrument invokes two d i s t i n c t s e r i e s of questions. Test No. 1 sets a task of s y n t a c t i c paraphrase r e c o g n i t i o n s i m i l a r to that r e q u i r e d by the "Deep S t r u c t u r e Recovery Test" (DSRT) on which some items were modelled and the "Perception of A l t e r n a t e S t r u c t u r e Test" (PAST) as w e l l as c e r t a i n of the formats used i n ".A Test of Sentence Meaning" (ATSM) . Following Mohan (1973:97), paraphrase i s here defined as a two-way i m p l i c a t i o n r e l a t i o n : "An a s s e r t i o n A^ i s a paraphrase of A_. i f A^ i m p l i e s A_. and v i c e v e r s a . " Test No. 2 sets a task of r e c o g n i t i o n of one-way i m p l i c a t i o n r e l a t i o n s s i m i l a r to that r e q u i r e d by c e r t a i n formats of ATSM. U n l i k e the l a t t e r , however, i t does not d i r e c t l y show a recovery of u n d e r l y i n g k e r n e l sentences; r a t h e r , the ^ i s merely asked to judge i f A. i m p l i e s A.. In determining the optimal format f o r Tests No. 1 and No. 2 (q.v. Appendix "B"), i t was considered p r e f e r a b l e to r e q u i r e Ss to compare only two sentences at a time f o r semantic equivalence as defined above. By doing so, the memory burden, p a r t i c u l a r l y f o r younger p u p i l s , might be reduced thereby a f f o r d i n g a t r u e r measure of l i n g u i s t i c competence. For example, i n an e a r l i e r study conducted by t h i s i n v e s t i g a t o r , the f o l l o w i n g item was adapted from the sample of the DSRT c i t e d above: What the boy would l i k e i s f o r the g i r l to leave. For the boy to leave i s what the g i r l would l i k e , yes no 96.3% of a sample of twenty-seven average Grade 6 students answered t h i s item c o r r e c t l y i n c o n t r a s t to only 70.9% c o r r e c t responses among Grade 5 p u p i l s of s u p e r i o r i n t e l l e c t u a l a b i l i t y to the o r i g i n a l DSRT three sentence item. To accomplish the purpose of the present i n v e s t i g a t i o n , c e r t a i n other features of the aforementioned t e s t s were examined and considered u n s u i t a b l e . The uneven sampling of transformations i n the DSRT and the PAST does not f a c i l i t a t e d i r e c t comparisons of d i f f i c u l t y among t r a n s f o r -mational types. The use of nonsense vocabulary i n the PAST presents a task which appears not only a r t i f i c i a l but also-..may be not e s p e c i a l l y a p p l i c a b l e to i n v e s t i g a t i n g the v a r i a b l e s defined i n the research problem as that t e s t ' s author concedes: Those t e s t s that u t i l i z e nonsense vocabulary to encourage r e l i a n c e on s y n t a c t i c cues have a low c o r r e l a t i o n w i t h reading comprehension t e s t s , w h i l e those that u t i l i z e conventional vocabulary and a l l o w r e l i a n c e on semantic as w e l l as s y n t a c t i c cues have n o t i c e a b l y higher c o r r e l a t i o n s w i t h reading. (O'Donnell, 1976:4). ATSM, whi l e l b a s e d on a more comprehensive s y n t a c t i c typology than e i t h e r of the other two, does not t e s t the transformations of greatest i n t e r e s t to t h i s i n v e s t i g a t o r . As w e l l , m u l t i p l e response formats may have proved too complicated f o r younger Ss. L a s t l y , the length of the t e s t 102-four.:;and f i v e o p t i o n items — would have s u b s t a n t i a l l y increased the time re q u i r e d f o r a d m i n i s t r a t i o n s i n c e the research design sought to assess th c o n t r i b u t i o n of the s y n t a c t i c t e s t s to measures of t e x t comprehension. This i n v e s t i g a t o r t h e r e f o r e a e l e c t e d to construct f o r the f i r s t p a r t of the instrument, a group of t h i r t e e n f o u r - i t e m subtests to apprais understanding of p a s s i v i z a t i o n , r e l a t i v i z a t i o n , and transformations i n v o l v i n g the "minimal distance p r i n c i p l e " as w e l l as paraphrases of i n d i r e c t speech and pseudoimperatives. 27 Anaphora Apart from the sentence transformations h i g h l i g h t e d i n the instruments described thus f a r , :-another s y n t a c t i c paraphrase device — anaphora — i s thought to be f r e q u e n t l y a s s o c i a t e d w i t h comprehension d i f f i c u l t i e s encountered by second language l e a r n e r s . Anaphora i s the term used to denote a s t r u c t u r e i n a sentence,—-..e.g., a pronoun, a pro-verb, or a clause demonstrative — that d e r i v e s i t s meaning from another part of the present sentence or from another sentence, u s u a l l y one that occurs p r e v i o u s l y - i n .the '.text. Previous attempts to measure understanding of anaphora From a taxonomy of anaphoric s t r u c t u r e s developed by Bormuth (1970) and Menzel (1970), a t e s t was devised by Bormuth e t . a l . (1970) to check comprehension of anaphoric r e l a t i o n s . A m u l t i p l e choice format was used. These i n v e s t i g a t o r s e s t a b l i s h e d a rank order of d i f f i c u l t y among fourteen anaphoric types f o r t h e i r sample of 240 Grade 4 students. Oddly, the e a s i e s t s t r u c t u r e proved to be pro-clause forms, the most d i f f i c u l t , common personal pronouns. Lesgold (1973) examined the comprehension of 80 Grade 3 and 4 c h i l d r e n on fourteen v a r i e t i e s of anaphoric s t r u c t u r e s i n c l u d i n g s e v e r a l examined by Bormuth et.. a l . Using a "wh— question, constructed response" format, the r e s u l t s obtained i n d i c a t e a marked d i s p a r i t y between the two s t u d i e s w i t h respect to the ranking..of those anaphoric s t r u c t u r e s t e s t e d by both i n v e s t i g a t o r s . I t was found, f o r example, that the e a s i e s t of a l l forms t e s t e d i n t h i s l a t e r study were personal pronouns, the pro-clause being one of the most d i f f i c u l t . Neither Bormuth et a l . nor Lesgold r e p o r t c o r r e l a t i o n s between t h e i r anaphora t e s t s and other measures of reading comprehension. The present instrument The present i n v e s t i g a t o r defined four general c a t e g o r i e s of anaphora on the b a s i s of the d e s c r i p t i v e c l a s s i f i c a t i o n o f f e r e d by H a l l i d a y and Hasan (1976) to which was added one type (Subtask R) adapted from Chomsky (1969) . The p r e d i c t e d order from e a s i e s t to most d i f f i c u l t was: Subtask P: Pronominal reference (7 items) < Subtask R: Pronominal reference v i o l a t i n g the minimal d i s t a n c e p r i n c i p l e (5 items) Subtask S: Nominal s u b s t i t u t e s (5 items) Subtask T: C l a u s a l s u b s t i t u t e s (3 items) A t o t a l of twenty items as i n d i c a t e d above were prepared f o r Test No. 3 using a "constructed response s u b s t i t u t i o n " format (q.v. Appendix "B") to comprise t h i s p o r t i o n of the instrument, designated as Task 2. TESTING TEXT COMPREHENSION For the purposes of t h i s study, t e x t comprehension i s intended to mean the r e c o g n i t i o n of t e s t item statements as c o n t a i n i n g p r o p o s i t i o n s c o n s i s t e n t , c o n t r a r y , or indeterminate to those o v e r t l y s t a t e d i n one or adjacent sentences of the accompanying t e x t . Previous attempts to-measure understanding of a t e x t Teachers and researchers w i l l recognize that the foregoing o b j e c t i v e appears to motivate many, but by no means a l l , of the items 29 included i n the "paragraph comprehension" s e c t i o n s of standardized reading achievement t e s t s and i n the l e s s formal i n v e n t o r i e s contained i n published m a t e r i a l s which are intended to develop comprehension of t e x t s . In reviewing past research l i t e r a t u r e , a number of s t u d i e s were c i t e d which claimed to demonstrate that s y n t a c t i c understanding i s c r u c i a l to reading comprehension. U s u a l l y , the c r i t e r i o n measure employed was some w e l l known standardized t e s t . Future i n v e s t i g a t i o n s designed to t e s t the r e l a t i o n between syntax and o v e r a l l comprehension must invoke more ri g o r o u s d e f i n i t i o n s of comprehension than those operative i n most published t e s t s . The present study subscribes to a d e f i n i t i o n set f o r t h by Bormuth: ...comprehension a b i l i t y i s thought to be a set of g e n e r a l i z e d knowledge - a c q u i s i t i o n s k i l l s which permit people to acquire and e x h i b i t i n f o r m a t i o n gained as a consequence of reading p r i n t e d language. (1969:50). I f t h i s d e f i n i t i o n i s f u r t h e r r e s t r i c t e d to i n c l u d e only e x h i b i t i o n s of i n f o r m a t i o n gained from a reading of the m a t e r i a l immediately at hand and not from e a r l i e r reading experiences, the dubious v a l i d i t y of popular standardized t e s t s of reading comprehension should become apparent. Comprehension i s defined as the a b i l i t y to acquire i n f o r m a t i o n from a passage, but one t r i e s to measure i t by f i n d i n g out how many questions the person can answer on a t e s t given him only a f t e r he has read the passage This procedure ignores the f a c t s that i t i s almost impossible to f i n d a passage d e a l i n g w i t h i n f o r m a t i o n about which a person knows 30 a b s o l u t e l y nothing and that he could probably have used t h i s i n f o r m a t i o n to answer some of the questions even before he had read the passage. (Bormuth, 1969:52). E m p i r i c a l support f o r t h i s contention may be found i n a number of s t u d i e s on passage dependency. One of the most notable (Tuinman, 1973) examined f i v e major t e s t s of reading comprehension. Approximately 1,800 students i n Grades 4, 5, and 6 p a r t i c i p a t e d i n t h i s study. Results i n d i c a t e d that none of these widely used t e s t s provided s u f f i c i e n t guarantees against the examinee's answering items on the b a s i s of i n f o r -mation other than that presented i n the passage. Average p r o b a b i l i t i e s of c o r r e c t responses, even when the passage was not present, were w e l l above chance scores. P r i o r knowledge, e l i m i n a t i o n of i r r e l e v a n t d i s t r a c t o r s , and the use of i n f o r m a t i o n embedded i n preceding questions are a l l suggested as p o s s i b l e causes f o r unexpectedly high scores. The present instrument To cope w i t h such problems i n t e s t i n g , a t t e n t i o n must be given to an important d i s t i n c t i o n . ...scores on comprehension t e s t s given i n the usual way have two components: those questions the student could have answered without reading the passage and those questions he was able to answer only as a consequence of reading a passage. Only the l a t t e r may be d e f i n i t e l y s a i d to repre-sent knowledge gained through reading. (Bormuth, 1969:52). This c l a r i f i c a t i o n prompted the i n v e s t i g a t o r to construct a model f o r t e s t i n g comprehension of a. w r i t t e n t e x t , given c e r t a i n q u a l i f i c a t i o n s , that seeks to d i s c r i m i n a t e between te r m i n a l behaviours that c o n s t i t u t e acts of text, comprehension and those that do not. Having considered through the model (q.v. Appendix "A") ... a number of p o s s i b l e s t r a t e g i e s that could l e a d t o t t h e s e l e c t i o n of the c o r r e c t response o p t i o n , the e s s e n t i a l problem i n e v a l u a t i n g an examinee's set of responses i s a matter of r e c o g n i z i n g those l e g i t i -mately derived from an immediate reading of the t e x t and e l i m i n a t i n g those c o r r e c t responses a t t a i n e d through e a r l i e r r e c o l l e c t i o n s or f a l s e comprehension s t r a t e g i e s . Given the proposed format and the s t a t e d q u a l i f i c a t i o n s , the model p o s i t s two d i s t i n g u i s h a b l e aspects of any act of reading comprehension — (1) the a c t u a l l o c a t i o n of the statement i n the passage that i s r e l e v a n t to an i n t e r p r e t a t i o n of the item (or the c a p a b i l i t y to do so on demand) and (2) an accurate recovery of the deep s t r u c t u r e of both the passage and the item statements i n order to determine, i f p o s s i b l e , the t r u t h value of the item statement. Any a s s e r t i o n that a student has comprehended must be supported by a demon-s t r a t i o n of s k i l l i n both aspects. Conventional t e s t s , however, have ignored the p o s s i b i l i t y of asessing the f i r s t aspect, i d e n t i f i c a t i o n of t e x t u a l l o c u s , thereby r e l y i n g s o l e l y upon an apparent i n d i c a t i o n of the second aspect, judgment of t r u t h value. Comprehension t e s t s constructed i n t h i s f a s h i o n are inadequate f o r the reason that w h i l e locus i d e n t i f i -c a t i o n i s a p r e r e q u i s i t e to an act of t e x t comprehension, i t i s not e s s e n t i a l to the c o i n c i d e n t a l s e l e c t i o n of the c o r r e c t response to m u l t i p l e choice items as s t u d i e s on passage dependency have shown. In order to make more d i r e c t observations of s k i l l i n both aspects, the i n v e s t i g a t o r set two separate tasks f o r Test No. 4, the second p a r t of the instrument s p e c i f i c a l l y intended to t e s t sentence comprehension w i t h i n a w r i t t e n t e x t . Twenty-three item statements r e l a t i n g to an accompanying passage were presented. In each i n s t a n c e , the was asked to i d e n t i f y the numbered sentence(s), i f any, i n the passage judged to provide i n f o r m a t i o n appropriate to determining the t r u t h value of the item statement. This locus i n d i c a t i o n task i s r e f e r r e d to i n t h i s r e p o r t as Task 3. The other task was to appraise z the item statement's t r u t h value through s e l e c t i o n of one of three response options: " t r u e " , " f a l s e " , or "cannot say". This l a t t e r task i s r e f e r r e d to i n t h i s r e p o r t as Task 4. The l a b e l l i n g of these two separate measures as Tasks 3 and 4 i s not intended to suggest that the always performs the locus operation p r i o r to judging the t r u t h value of an item statement although the a c t u a l format of the t e s t might have i n v i t e d him to do so. Depending on stre n g t h of memory, a S^  might more r e a d i l y r e c a l l what had been s t a t e d i n the passage than the p o i n t at which the r e l e v a n t statement occurred. Since t h i s f a c t o r could not be c o n t r o l l e d by the i n v e s t i g a t o r , Ss were not given s p e c i f i c d i r e c t i o n to carry out one task before the other. In the context of the present i n v e s t i g a t i o n , an act of t e x t comprehension i s considered to have occured only when a S o f f e r e d a c o r r e c t response to both the Task 3 (Locus) and the Task 4 (Truth Value) components of any given item. This combined performance i s r e f e r r e d to i n t h i s r e p o r t as Task 5 (Locus + Truth Value). Task 5 was not an a d d i t i o n a l o p e ration f o r the Ss; r a t h e r , i t i s an e s s e n t i a l c onstruct used by the investigator in scoring Test No. 4, the text comprehension part of the instrument. It is Task 5 against which Ss1 performance on Task 1 (Grammatical Implication) and Task 2 (Anaphora) is compared. 34 Chapter IV RESEARCH DESIGN AND HYPOTHESES For p r e l i m i n a r y a n a l y s i s , t h i s study employs a 3 x 3 f a c t o r i a l design to e f f e c t comparison among sample populations w i t h respect t o : (1) a language f a c t o r (N.S. = n a t i v e speakers of E n g l i s h ; E.S.L.(A) = E n g l i s h as a second language students who are n a t i v e speakers of Chinese;;; E.S.L.(B) = E n g l i s h as a second language students who are n a t i v e speakers of a language other than Chinese) and (2) a grade placement f a c t o r (Elem = Elementary, i . e . , Grades 4, 5, and 6; JrSec = J u n i o r Secondary, i . e . , Grades 7, 8, and 9; SrSec = Senior Secondary, i . e . , Grades 10, 11, and 12). Elem JrSec SrSec "" ' N.S. E.S.L.(A) E.S.L.(B) Comparisons are conducted f o r each of f i v e task v a r i a b l e s : (1) r e c o g n i t i o n of grammatical i m p l i c a t i o n ; (2) i d e n t i f i c a t i o n of anaphoric r e f e r e n t s ; (3) i n d i c a t i o n of t e x t u a l l o c u s ; (4) judgment of t r u t h v a l u e ; and (5) a c r i t e r i o n measure of comprehension based d i r e c t l y on performance of the l a t t e r two s k i l l s . The f o l l o w i n g experimental hypotheses are o f f e r e d f o r an i n i t i a l examination of the data. 35 Hypothesis I At given levels of educational attainment as determined by grade placement, (1) the performance of native speakers is superior to that of students for whom English is a second language, and (2) the performance of Chinese speakers does not differ from that of other E.S.L. students on each task included in the present instrument. Rationale After examining the instrument, a l l participating E.S.L. teachers agreed that even the most proficient students were not li k e l y to perform at the same level as native speaker peers. Certain syntactic structures were frequently identified by these teachers as not having an instructional emphasis in the E.S.L. programme. The most readily accessible E.S.L. population was comprised, in large part, of Chinese speakers. Since a l l E.S.L. students selected for inclusion in this study were required to meet the same specified c r i t e r i a , there is no evident reason why one linguistic group should be more pro-ficient than others on any of the task variables. Any significant findings might reveal particular d i f f i c u l t i e s of syntactic comprehension for Chinese learners not experienced by a comparable multiethnic group of students. In other words, the present study is concerned to identify, where possible, qualitative as well as quantitative differences among groups of subsamples. 36 Hypothesis II In the case of both native speakers and students for whom English is a second language, performance on each.;task included in the present instrument is augmented at increasing levels of educational attainment as determined by grade placement. Rationale Research conducted on native speakers cited in the literature review suggests-this to be a plausible hypothesis. Because several E.S.L. students may have started to learn English after their elementary years of schooling, i t is considered less probable that second language learners also exhibit an augmented pattern similar to that of the native speakers. Hypothesis III The rank order of d i f f i c u l t y of the sixteen subtasks of Task 1 (recognition of grammatical implication) and Task 2 (identification of anaphoric referents) as determined by subtask mean scores does not vary among any subsamples. Rationale A l l students, regardless of their native-language, are at some point in developing their comprehension of written English. It i s expected that those transformational and anphoric types which are among the most d i f f i c u l t for those less proficient w i l l also be relatively more d i f f i c u l t than other types for the more accomplished students. Since comparison^:of subsamples by n a t i v e language f a c t o r i s of primary i n t e r e s t i n t h i s study, the a n a l y s i s by grade i n t e r v a l being conducted merely to confirm a probable p a t t e r n , the remaining hypotheses are t e s t e d across three grouped samples: n a t i v e speakers of Englisht^N. s T ) Chinese speakers fE.S.L.(A)3 , and speakers of other languages £E.S.L.(B)3 N.S. E.S.L.(A) . E.S.L.(B) In t h i s manner, l i m i t a t i o n s of sample s i z e may be p a r t i a l l y overcome w h i l e d i r e c t i n g a t t e n t i o n to broad c o n t r a s t s . Hypothesis IV Combined across grade i n t e r v a l s , (1) the performance of n a t i v e speakers i s s u p e r i o r to that of students f o r whom E n g l i s h i s a second language, and (2) the performance of Chinese speakers does not d i f f e r from t h a t of other E.S.L. students on each task and subtask included i n the present instrument. The r a t i o n a l e f o r t h i s hypothesis i s i d e n t i c a l to that f o r Hypothesis I . 38 H y p o t h e s i s V The r a n k o r d e r o f d i f f i c u l t y o f the s i x t e e n s u b t a s k s o f T a s k 1 ( r e c o g n i t i o n o f g r a m m a t i c a l i m p l i c a t i o n ) and T a s k 2 ( i d e n t i f i c a t i o n o f a n a p h o r i c r e f e r e n t s ) as d e t e r m i n e d by s u b t a s k mean s c o r e s does n o t v a r y among any subsamples combined a c r o s s grade i n t e r v a l s . The r a t i o n a l e f o r t h i s h y p o t h e s i s i s i d e n t i c a l to t h a t f o r H y p o t h e s i s I I I . H y p o t h e s i s V I A c o m p a r i s o n o f the t h r e e measures o f t e x t c o m p r e - ; h e n s i o n e x h i b i t s a p a t t e r n o f d i m i n i s h e d p e r f o r m a n c e w h e r e i n T a s k 4 ( judgment o f t r u t h v a l u e ) s c o r e s a r e g r e a t e r t h a n Task 3 ( i n d i c a t i o n o f t e x t u a l l o c u s ) s c o r e s w h i c h a r e g r e a t e r t h a n T a s k 5 s c o r e s . R a t i o n a l e I f , as p a s s a g e dependency s t u d i e s have shown, i t i s p o s s i b l e f o r s t u d e n t s to r e s p o n d c o r r e c t l y to m u l t i p l e c h o i c e i t e m s w i t h o u t r e f e r e n c e to t h e r e l a t e d t e x t a t g r e a t e r t h a n chance p r o b a b i l i t i e s , t h e n i t i s r e a s o n a b l e to e x p e c t t h a t T a s k 4 i s e a s i e r t h a n T a s k 3 w h i c h r e q u i r e s , f o r e a c h i t e m , t h a t the s t u d e n t . e v a l u a t e the p r o p o s i t i o n a l c o n t e n t o f a number o f s e n t e n c e s i n the t e x t to d e c i d e w h i c h one o r two p r o v i d e i n f o r m a t i o n s u f f i c i e n t to j u d g e the t r u t h v a l u e o f the i t e m s t a t e m e n t . Because T a s k 5 demands p r o f i c i e n c y on b o t h T a s k 3 and T a s k 4, i t i s i n e v i t a b l e t h a t i f T a s k 4 p r o v e s to be e a s i e r t h a n T a s k 3 as s u g g e s t e d , t h e n an i n d i v i d u a l ' s s c o r e f o r T a s k 5 w i l l be l o w e r t h a n t h a t a t t a i n e d f o r e i t h e r o f i t s component t a s k s . Hypothesis VII Task 1 ( r e c o g n i t i o n of grammatical i m p l i c a t i o n ) and Task 2 ( i d e n t i f i c a t i o n of anaphoric r e f e r e n t s ) are each b e t t e r p r e d i c t o r s of Task 5 (the combined measure o f . l o c u s i n d i c a t i o n and t r u t h value judgment) than of Task 4 (judgment of t r u t h value only.) Task 1 and Task 2 are e q u a l l y good p r e d i c t o r s of Task 5 scores. R a t i o n a l e The model proposed to define which sequences of behaviour c o n s t i t u t e acts of t e x t comprehension i n the present t e s t format (q.v. Appendix "A") c i t e s the l o c a t i n g of an item paraphrase i n the t e x t as c r u c i a l when short term memory f a i l s to prompt a judgment of t r u t h value. The sentence comparison aspect of Task 1 and the r e s o l u t i o n nature of Task 2 are thought to be e q u a l l y and more c l o s e l y r e l a t e d to the paraphrase search embodied i n Task 3 ( i n d i c a t i o n of t e x t u a l locus) which i s a r e q u i s i t e f o r the suggested c r i t e r i o n of comprehension, Task 5, but not f o r Task 4. Hypothesis V I I I Task 3 ( i n d i c a t i o n of t e x t u a l locus) i s a b e t t e r p r e d i c t o r than Task 4 (judgment of t r u t h value) of the c r i t e r i o n f o r comprehension, Task 5. Rationale I t w i l l be r e c a l l e d that Task 5 i s a combined measure of p r o f i c i e n c y on Task 3 and Task 4. Previous d i s c u s s i o n of passage dependency a s s e r t s that m u l t i p l e choice t r u t h value t e s t s (e.g., Task 4) cannot be considered as dependable measures of what f o r t h i s study has been defined as an act of t e x t comprehension. Because Task 4 i s b e l i e v e d to have a high r i s k toward spurious scores, success on Task 5 i s more l i k e l y to be l i m i t e d by performance on Task 3. 40 Chapter V PROCEDURES P r e l i m i n a r y a p p r a i s a l of t e s t items The i n v e s t i g a t o r s e l e c t e d t h i r t e e n t r a n s f o r m a t i o n a l c o n t r a s t s and four anaphoric types f o r examination. Items constructed f o r each of these subtasks were reviewed by two l i n g u i s t s who judged t h e i r a p p r o p r i -ateness to the category s p e c i f i e d . F ollowing r e v i s i o n s , seventy-two items were i n c l u d e d i n the f i n a l v e r s i o n of Test.Nos. 1, 2, and 3 ( i . e . Task 1 and Task 2). Because the aforementioned t e s t s are intended to measure s y n t a c t i c understanding, vocabulary content, an e s t a b l i s h e d major deter-minant of reading comprehension scores, had to be r i g o r o u s l y c o n t r o l l e d i n order to assure l e x i c a l access f o r younger and non-native Ss. E a r l i e r s t u d i e s had concluded the use of nonsense words to a f f o r d a "purer measure" of s y n t a c t i c comprehension to be an i n e f f e c t i v e s t r a t e g y owing to the i n t e r a c t i o n of syntax and semantics. A c c o r d i n g l y , items were constructed and reviewed to assure the p r o p o s i t i o n a l content of each sentence to be p l a u s i b l e to the Ss. Sentence t o p i c s were g e n e r a l l y r e s t r i c t e d to concrete objects or observable .events considered to be w i t h i n the realm of p u p i l experience and c o g n i t i v e development. P r i o r to.;the p r e p a r a t i o n of t e s t b o o k l e t s , a l l r e v i s e d items and w r i t t e n d i r e c t i o n s f o r both the s y n t a c t i c a n d t e x t comprehension parts of the instrument were evaluated according to the Dale - C h a l l formula, e s s e n t i a l l y an index of vocabulary d i f f i c u l t y and sentence l e n g t h . A l l m a t e r i a l analyzed y i e l d e d a " c o r r e c t e d r e a d a b i l i t y l e v e l of Grade 4.0 and below". Format of the t e s t s The instrument was ty p e w r i t t e n and photocopied to provide uniformly c l e a r copies to a l l Ss who i n d i c a t e d t h e i r responses d i r e c t l y i n the t e s t booklets according to d i r e c t i o n s provided i n p r i n t or by the t e s t a d m i n i s t r a t o r . To f a c i l i t a t e e x p l a n ation of d i r e c t i o n s as w e l l as provide ;S_s wi t h s u i t a b l e sample items, the instrument was d i v i d e d i n t o four ^ s ections Test No. 1 (36 items) This t e s t was designed to assess r e c o g n i t i o n of two-way i m p l i c a t i o n r e l a t i o n s between p a i r e d sentences. j>s were given d i r e c t i o n s followed by two sample items: D i r e c t i o n s : Read both sentences. I f the two sentences mean the same t h i n g , c i r c l e "yes". I f the two sentences DO NOT mean the same t h i n g , c i r c l e "no". Sample A: The boy h i t the g i r l . The g i r l was h i t by the boy. i ^ e ^ no Sample B: The boy looked at the b i g dog^. The b i g dog looked at the boy. yes ^no^ The t h i r t y - s i x items comprise nine subtasks of four items each which are intended to determine the S_rs f a c i l i t y w i t h the f o l l o w i n g t r a n s f o r m a t i o n a l types: A. P a s s i v i z a t i o n B. P a r t i c i p l e m o d i f i e r s C. Wh— f r o n t i n g D. R e l a t i v i z a t i o n contrasted w i t h c l a u s a l conjunction E. R e l a t i v i z a t i o n by pronoun d e l e t i o n F. Double transformation ( R e l a t i v i z a t i o n + P a s s i v i z a t i o n ) G. Ask (query) contrasted w i t h T e l l H. Easy to see J . Promise contrasted w i t h T e l l 42 Items i n Test No. 1 were arranged to correspond w i t h the above subtasks i n a r o t a t i n g sequence, i . e . , Item No. 1 belongs to Subtask A, Item No. 2 to Subtask B, Item No. 3 to Subtask C, e t c . Test No. 2 (16 items) Because c e r t a i n transformations of i n t e r e s t to the i n v e s t i g a t o r could not be presented f o r c o n s i d e r a t i o n as p o s s i b l e two-way i m p l i c a t i o n r e l a t i o n s , an a l t e r n a t i v e format was devised to examine the S^ 's f a c i l i t y w i t h the more r e s t r i c t i v e one-way i m p l i c a t i o n r e l a t i o n . . Again, d i r e c t i o n s and two sample items were provided: D i r e c t i o n s : Read the f i r s t sentence c a r e f u l l y . Then read the second sentence and decide i f i t i s true or f a l s e . Sample A: i f : Ann and Helen walk, to school together. does i t mean: Ann and Helen walk to school at the same time. (ye^) no Sample B: i f : Mother s a i d , "You must come home e a r l y . " does i t mean: Mother must come home e a r l y . ' ,—-v yes ( no) The s i x t e e n items comprise four subtasks of four items each d e a l i n g w i t h the f o l l o w i n g transformations: K. I n d i r e c t speech L. Ask (request) contrasted w i t h T e l l M. Pseudoimperatives N. Agentless p a s s i v i z a t i o n Items i n Test No. 2 were arranged c o n s e c u t i v e l y i n c l u s t e r s , i . e . , the f i r s t four items comprise Subtask K, the second four items, Subtask L, e t c . , i n the b e l i e f that t h i s format would minimize the c o g n i t i v e s h i f t i n g expected to occur when one must i n t e r p r e t f i r s t one t r a n s f o r m a t i o n a l type, then ot h e r s , only to r e t u r n to another item of the f i r s t type. The i n v e s t i g a t o r was concerned to reduce the time r e q u i r e d f o r response so as to maximize the opportunity f o r a l l Ss to complete the e n t i r e set of tasks i n a reasonable amount of time. Since the i n v e s t i g a t o r had no reason to hypothesize that r e c o g n i t i o n of one-way i m p l i c a t i o n r e l a t i o n s i s a l i n g u i s t i c s k i l l d i s t i n c t i n i t s d i s t r i b u t i o n from that of r e c o g n i t i o n of two-way i m p l i c a t i o n r e l a t i o n s , both Test No. 1 and Test No. 2 are combined to measure a s i n g l e g e n e r a l i z e d a b i l i t y , r e c o g n i t i o n of grammatical i m p l i -c a t i o n . This i s r e f e r r e d to elsewhere i n t h i s r e p o r t as Task 1. Test No. 3 (20 items) This t e s t was designed to assess the S/s c a p a c i t y to i d e n t i f y anaphoric r e f e r e n t s . Types chosen f o r the. present instrument were c i t e d i n e a r l i e r .discussion of anaphora. For each item, the r e l e v a n t pro-form i s u n derlined i n the context. I t i s then repeated followed by a l i n e provided f o r the S/s constructed response. The task was considered to be s e l f - e x p l a n a t o r y through the one example o f f e r e d . Sample: I had an apple f o r lunch. I_t was good. I t apple  I t was decided that any a d d i t i o n a l d i r e c t i o n could be b e t t e r provided o r a l l y by the t e s t a d m i n i s t r a t o r than by w r i t t e n d i r e c t i o n s or f u r t h e r samples. Items r e l a t e d to each of four subtasks were placed randomly throughout the t e s t . Test No. 3 i s r e f e r r e d to elsewhere i n t h i s r e p o r t as Task 2. Test No. 4 (23 items) This t e s t comprises two measures of t e x t comprehension separately and i n combination. A l l items are contained on a s i n g l e page which i s accompanied by another s i n g l e page c o n t a i n i n g the reading passa The l a t t e r page detaches from the t e s t booklet so as to a f f o r d Ss c o n t i n uous easy reference w h i l e responding to items. The passage„consists_of twelve sentences which are p r i n t e d s u c c e s s i v e l y , one sentence to a l i n e , r e gardless of sentence l e n g t h . Each sentence i s preceded by i t s number f o r i d e n t i f i c a t i o n as r e q u i r e d by the f o l l o w i n g d i r e c t i o n s : D i r e c t i o n s : Read the s t o r y on the short paper f i r s t . Read each question sentence below. In the parentheses ( ) w r i t e the number r . of the sentence, or sentences, i n the s t o r y that t e l l s you the answer. I f the question sentence i s t r u e , c i r c l e "T". I f the question sentence i s f a l s e , c i r c l e "F". I f none of the sentences i n the s t o r y t e l l you the answer to the question, put X i n the parentheses ( ) and c i r c l e " ?". Sample_iA: The cave men d i d not use the s k i n s of animals. ( 5 ) T 0 Sample B: They l i v e d i n caves on the sides of h i l l s where they could keep dry and warm. ( 2+3)(T) F Sample C: The cave men b u i l t f i r e s i n f r o n t of t h e i r homes. ( X . ) T F The i n d i c a t i o n of t e x t u a l locus p o r t i o n of Test No. 4 defines Task 3. The judgment of t r u t h value p o r t i o n defines Task 4. The combi^ n a t i o n of these two components generates that measure r e f e r r e d to e l s e -where i n t h i s r e p o r t as Task 5. 45 D e s c r i p t i o n of the sample populations Comparison of performance among the sample populations i s motivated p r i m a r i l y by the question: At the c o n c l u s i o n of an i n t e n s i v e programme i n E n g l i s h language t r a i n i n g , how s i m i l a r are E.S.L. students to t h e i r n a t i v e speaker peers w i t h respect to the l i n g u i s t i c v a r i a b l e s under i n v e s t i g a t i o n ? Because the E.S.L. populations to which the i n v e s t i g a t o r had access were dominated by Chinese speaking students, i t was decided to separate t h i s group at each grade i n t e r v a l to determine i f the performance of Chinese speakers d i f f e r s , e i t h e r q u a n t i t a t i v e l y or q u a l i t a t i v e l y , from that of other E.S.L. students r e p r e s e n t a t i v e of a number of n a t i v e languages and who had experienced s i m i l a r i n s t r u c t i o n s i n c e a l l c l a s s e s p a r t i c i p a t i n g i n the study included both Chinese and non-Chinese speakers. I t was considered appropriate to e s t a b l i s h the f o l l o w i n g c r i t e r i a f o r admission i n t o the r e s p e c t i v e samples. For n a t i v e speakers: 1. A c q u i s i t i o n of E n g l i s h p r i o r to any other language.^ 2. No r e p o r t of marked d i f f i c u l t y i n growth i n reading comprehension or of any other observed l e a r n i n g d i s a b i l i t y . For E.S.L. students: 1. Completion of a minimum of 800 hours of E n g l i s h language t r a i n i n g i n the Vancouver schools. This was a f i r m requirement, apart from 1. This c r i t e r i o n e l i m i n a t e d the i n c l u s i o n : o f s e v e r a l students e n r o l l e d i n r e g u l a r c l a s s e s who immigrated to Canada s e v e r a l years ago and now c l a i m to be more p r o f i c i e n t i n E n g l i s h than i n t h e i r o r i g i n a l language. 46 any p r i o r study of E n g l i s h the student may have undertaken i n h i s own country. 2. Assessment by the E.S.L. teacher that the student had made at l e a s t average progress i n the programme, given the d u r a t i o n of h i s attendance. 3. Recommendation by the E.S.L. teacher that the student be placed i n a r e g u l a r c l a s s w i t h E n g l i s h speaking peers on the reopening 2 of schools the f o l l o w i n g September. (The instrument was admini-stered at the end of June.) A l l _Ss were e n r o l l e d i n p u b l i c schools under the j u r i s d i c t i o n of the Vancouver School Board from whom approval was granted to conduct t h i s study. F o l l o w i n g i s a d e s c r i p t i o n of each subsample. 1. Native speakers — Elementary: 12 ^ s , 6 males and 6 females, ranging i n age from 9 y r 1 mo to 10 y r 5 mo, destined to enter a Grade 5 c l a s s on reopening of school. School "A" i s s i t u a t e d i n what the i n v e s t i g a t o r would describe as a middle c l a s s neighbourhood w i t h an e t h n i c a l l y d i v e r s e p o p u l a t i o n . 2. Native speakers — Junior Secondary 10 Ss, 4 males and 6 females, ranging i n age from 12 yr 10 mo to 14 y r 2mo, destined to enter a Grade 8 c l a s s on reopening of school by t r a n s f e r r i n g 2. In the case of s e n i o r secondary E.S.L, students, the Sjs were already e n r o l l e d i n some r e g u l a r courses w i t h n a t i v e speaker peers. The sample was drawn from a group of students who were r e q u i r e d to i n c l u d e a s p e c i a l " T r a n s i t i o n a l E n g l i s h " course i n t h e i r programmes. from School "A" to the nearest secondary school." 3 3. Native speakers — Senior Secondary 21 S^ s, 9 males and 12 females, ranging i n age from 14 yr 3 mo to 15 yr 10 mo, destinded to e n r o l l i n "Grade 10 English" on reopening of school. School "B" i s located i n an upper middle c l a s s neigh-bourhood which includes a small minority of immigrants. 4. E.S.L. students — Elementary 16 Ss, 9 males and 7 females, ranging i n age from 10 yr 0 mo to 12 yr 5 mo, destined for placement i n regular Grade 4, Grade 5, or Grade 6 classes on reopening of school. 10 Ss were i d e n t i f i e d to be Chinese speakers. The remaining 6 S^ s include one speaker each of French, Portugese, P o l i s h , Hungarian, Yugoslavian, and Korean. Students i n t h i s sample were drawn from three E.S.L. reception classes at School "C" and one such class at School "D". In a reception c l a s s , usually l i m i t e d to an enrollment of f i f t e e n , the student spends most of h i s i n s t r u c t i o n a l time i n a self-contained classroom under the d i r e c t i o n of one teacher and, i n some instances, a trained "teaching a s s i s t a n t " . Arrangements vary among classes but, generally, a student does not spend more than 20% of h i s i n s t r u c t i o n a l time i n other s e t t i n g s . Two of the three p a r t i c i p a t i n g classes at School "C" were taught by teachers who, for the f i n a l three months of the school year, had adopted a team approach enabling them to group students according to English language p r o f i c i e n c y . One cl a s s was designated "Intermediate", the other 3. In B r i t i s h Columbia, what i s customarily the f i r s t year of junior secondary, Grade 7, i s conducted as the f i n a l year of the elementary school programme. "Advanced". The t h i r d p a r t i c i p a t i n g c l a s s at School "CV may be c h a r a c t e r i z e d as m u l t i l e v e l . The E.S.L. programme at School "D" was d i v i d e d i n t o four phases, each under the d i r e c t i o n of a d i f f e r e n t teacher, a l l of whom j o i n t l y developed a c u r r i c u l u m intended to maximize i n s t r u c t i o n a l c o n t i n u i t y . The c l a s s p a r t i c i p a t i n g i n t h i s study was designated as the most advanced l e v e l . Both School "C" and School "D" are l o c a t e d i n lower - lower middle c l a s s neighbourhoods. However, one must bear i n mind that the m a j o r i t y of students i n the E.S.L. r e c e p t i o n c l a s s e s r e s i d e d outside the immediate neighbourhood of the school which they attended. E.S.L. students — J u n i o r Secondary 30 Ss, 18 males and 12 females, ranging i n age from 11 y r 7 mo to 15 y r 7 mo, destined f o r placement i n r e g u l a r Grade 7, Grade 8, or Grade 9 c l a s s e s on reopening of school. 23 Ss were i d e n t i f i e d to be Chinese speakers. The remaining 7 Ss i n c l u d e one speaker each of. French, Spanish, I t a l i a n , Vietnamese, H i n d i , and two Korean speakers. In a d d i t i o n to the E.S.L. c l a s s e s described above i n 4., two other m u l t i l e v e l r e c e p t i o n c l a s s e s at School "C" provided S_s f o r t h i s sample. E.S.L. students — Senior secondary 18 £>s, 10 males and 8 females, ranging i n age from 14 y r 10 mo to 18 yr 6 mo. 12 Ss were i d e n t i f i e d to be Chinese speakers. The remaining 6 Ss i n c l u d e two speakers of Portugese, one speaker of P a n j a b i , and three Tagalog speakers. These students were e n r o l l e d i n a supplementary " T r a n s i t i o n a l E n g l i s h " course i n a d d i t i o n to other courses i n the r e g u l a r secondary programme. On reopening of s c h o o l , students i n t h i s sample would be 49 classified as either Grade 10 or Grade 11. School "E" is situated in a lower middle class, ethnically diverse, area of the city and, in most cases, is the f a c i l i t y closest to the student's residence. It w i l l be noted that the numberiof Ss in each subsample is unequal. While the investigator would have preferred to increaseLthe sample size of native speakers, this was not feasible as i t would have necessitated accessing a number of classes at a busy time of the school year. It is also considered that the linguistic variables examined by the instruments are aspects of language ..competence which, fotia native speaker population of a given school grade interval, would not be greatly influenced by curriculum content or methods of teaching experienced by the student during the school year. On the other hand, considerable effort was expended to procure a sizable E.S.L. sample from several different sources within the school system so as to randomize the effects of strengths and weaknesses of specific programmes since one.might reasonably propose that the methodology and curriculum selected by the E.S.L. teacher throughout the school year could contribute substantially to the variance in students' performance on measures of the particular linguistic variables under investigation. Administration of the instrument In a l l but one case the tests were administered by the investi-gator to Ss in the elementary and junior secondary samples in their usual classroom setting.. The one exception, an elementary / junior secondary E.S.L. class from which eight pupils were deemed to be suitable Ss for 50 t h i s study, and the s e n i o r secondary samples completed the t e s t s under the s u p e r v i s i o n of t h e i r r e g u l a r E n g l i s h teacher who had been b r i e f e d by the i n v e s t i g a t o r on the purpose of the study, what each t e s t was designed to measure, and s p e c i f i c problems which might a r i s e i n a d m i n i s t r a t i o n . A l l a d m i n i s t r a t i o n s were conducted i n June, 1977, during the f i n a l three weeks of the school year. Scheduling r e s t r a i n t s and the number of p a r t i c i p a t i n g c l a s s e s i n dispersed l o c a t i o n s n e c e s s i t a t e d completion of a l l tasks i n one s i t t i n g . This c o n d i t i o n may have been a c o n t r i b u t i n g f a c t o r to the poor performance of some Ss. However, 90.7% of the n a t i v e speakers and 96.9% of the E.S.L. students who met the c r i t e r i a f o r admission i n t o the sample responded to a l l , or n e a r l y a l l , items on the f i n a l t a s k s . P i l o t t e s t i n g of the instrument on a small sample of elementary students (not p a r t i c i p a n t s i n t h i s study) who were known to vary widely i n reading comprehension s k i l l s i n d i c a t e d the p r o b a b i l i t y that most Ss would be able to complete a l l tasks i n l e s s than f i f t y minutes. A c c o r d i n g l y , a l l Ss were advised that t h e i r performance would not be timed and that they would be granted adequate time to complete a l l tasks c a r e f u l l y . Very few S_s r e q u i r e d more than f i f t y minutes f o r completion. Owing to i n d i v i d u a l d i f f e r e n c e s i n r a t e of response to v a r i o u s types of items, i t was decided to present d i r e c t i o n s and review examples of a l l tasks before a l l o w i n g Ss to commence the f i r s t t e s t . Supplementary o r a l d i r e c t i o n s appropriate to each of the four t e s t s i n the instrument were as f o l l o w s : Tests No. 1, No. 2, and No. 3 A l l c l a s s e s w i t h the exception of the s e n i o r secondary n a t i v e speakers were pr e - t e s t e d f o r r e c o g n i t i o n of gender of the common personal 51 names used i n t e s t items. E.S.L. c l a s s e s were posted a l i s t of such names. Test No. 3 Ss were cautioned that despite-.a s i n g l e short l i n e on which to w r i t e t h e i r responses, the most appropriate r e f e r e n t f o r the underlined pro-form might p o s s i b l y comprise s e v e r a l words. An example, a d d i t o n a l to that presented i n the t e s t b o o k l e t , was presented: Jack w i l l not come to school today. He s a i d so yesterday. so = that Jack w i l l not come to school Scoring of the t e s t s Each t e s t booklet was hand coded to prepare input data f o r a LERTAP 2.0 computer programme (Nelson, 1974) to compute and t a b u l a t e raw scores, present d e s c r i p t i v e s t a t i s t i c s , and f u r n i s h item analyses. Coding of booklets was double checked and a random sampling of booklets repre-s e n t i n g 20% of the e n t i r e sample revealed no c l e r i c a l e r r o r s when v e r i f i e d by an e x t e r n a l source. A l l keypunching was l a t e r v e r i f i e d e i t h e r v i s u a l l y or mechanically. In the case of Tests No. 1 (Task 1), No. 2 (Task 1), and the Truth Value p o r t i o n of No. 4 (Task 4 ) , answer keys were prepared i n advance of s c o r i n g . Test No. 3 (Task 2) i n which Ss were r e q u i r e d to i d e n t i f y anaphoric r e f e r e n t s i s c h a r a c t e r i z e d by a ^constructed response" format. The i n v e s t i -gator evaluated a l l responses f o r t h i s t e s t i n two concentrated sessions approximately two days apart i n order to e s t a b l i s h a c o n s i s t e n t standard i n s e t t i n g minimum c r i t e r i a f o r acceptableoresponses. As no key could be s e t , responses were judged somewhat s u b j e c t i v e l y f o r evidence of grasp of the e s s e n t i a l idea r a t h e r than f o r p r e c i s i o n i n a r t i c u l a t i n g an utterance that could be d i r e c t l y substituted into the pro-form p o s i t i o n . By way of example, consider Item No. 5: The boys played b a l l very hard. This i s what won the game. This T y p i c a l responses judged acceptable: playing b a l l very hard playing very hard playing hard played hard Typical responses judged unacceptable: playing b a l l very hard An inspection of responses to the textual locus portion of Test No. 4 (Task 3) indicated the need to depart from a strict.adherence to the predetermined key as the invest i g a t o r concluded the truth value of the item statement could i n c e r t a i n cases be derived from a combination of sentences other than what was suggested by the key. The following example i s i l l u s t r a t i v e of the problems incurred i n constructing and keying a test of text comprehension for which there i s no prototype i n the l i t e r a t u r e . Item No. 4 The picture s t o r i e s about cave men t e l l us new things. ( _) T F Keyed response: (12) They t e l l us things about cave men that we never knew before. Alternate responses: (10) The picture s t o r i e s are s t i l l there. (12) They t e l l us things about cave men that we never knew before. ( 8) They drew t h e i r p i c t u r e s on the stone walls of caves. (12) They t e l l us things about cave men that we never knew before. A f t e r s c o r i n g t h i s t e s t , the i n v e s t i g a t o r rechecked a l l t e s t booklets to v e r i f y that a c o n s i s t e n t standard f o r e v a l u a t i n g responses was maintained f o r a l l subsamples. Chapter VI RESULTS AND DISCUSSION S t a t i s t i c a l p r o p e r t i e s of the instrument This p r e l i m i n a r y d i s c u s s i o n i s intended to o u t l i n e those steps necessary to assess the v a l i d i t y of the instrument i n i t s present form. Task 1 Task 1 ( r e c o g n i t i o n of grammatical i m p l i c a t i o n ) comprises nine subtasks p e r t a i n i n g to two-way i m p l i c a t i o n r e l a t i o n s (Test No. 1) and four subtasks p e r t a i n i n g to one-way i m p l i c a t i o n r e l a t i o n s (Test No. 2). Each subtask contains four items f o r a t o t a l of 52 items. The p r e l i m i n a r y output of the LERTAP 2.0 programme i n d i c a t e d that one subtask, N: Agentless p a s s i v i z a t i o n , has a low c o r r e l a t i o n of .015 w i t h Task 1 as a whole thereby i n d i c a t i n g that the subtask makes no substantive c o n t r i b u t i o n to the v a r i a b l e under i n v e s t i g a t i o n . Further evidence of the inappropriateness of Subtask N l i e s i n the observation that two items bear negative p o i n t b i s e r i a l c o r r e l a t i o n s w i t h t o t a l task scores. This means t h a t , as a group, those Ss who answered the items c o r r e c t l y earned lower scores on Task 1 than those who answered i n c o r r e c t l y This p a r t i c u l a r subtask y i e l d e d a mean score of 2.04 f o r the e n t i r e sample, n o t i c e a b l y lower than that of any other subtask a s s o c i a t e d w i t h Task 1. Moreover, the highest scores were earned by E.S.L. students a t the lower grade i n t e r v a l s . Consequently, the d e c i s i o n was made to remove Subtask N from any f u r t h e r a n a l y s i s of the instrument. The items are included i n Appendix "B" and the subtask i s discussed i n a l a t e r s e c t i o n of t h i s r e p o r t The remainder of Task 1 contains twelve subtasks, a t o t a l of 48 items. The mean task score obtained by the e n t i r e sample p o p u l a t i o n i s 40.04, approximately 83.4% of the maximum p o s s i b l e score. The standard d e v i a t i o n i s 5.17 and the range i s 25 to 48, thus i n d i c a t i n g a negative skewness c h a r a c t e r i s t i c of a measure of mastery l e a r n i n g as might be expected of any aspect of l i n g u i s t i c competence. Since the v a l i d i t y of any t e s t i s l i m i t e d by i t s r e l i a b i l i t y , i t i s important to a s c e r t a i n that measures.:,of r e l i a b i l i t y are reasonably high. This i s to say that a researcher wants to be c e r t a i n that each item on h i s t e s t c o n t r i b u t e s to a measure of the same v a r i a b l e . The LERTAP 2.0 programme provides a number of s t a t i s t i c s to monitor i n t e r n a l r e l i a b i l i t y . Four have been s e l e c t e d f o r use i n the a n a l y s i s of data. The p o i n t b i s e r i a l c o r r e l a t i o n of each item w i t h the subtask  score (PB-ST) o f f e r s some i n d i c a t i o n of the extent to which a p a r t i c u l a r item a c t s i n concert w i t h the other items of the subtask to measure what i s conceived to be a s p e c i f i c v a r i a b l e . PB-ST c o r r e l a t i o n s among the v a r i o u s subtasks of Task 1 range from .20 to .82 w i t h a median PB-ST of .60 The p o i n t b i s e r i a l c o r r e l a t i o n of ceach..:item .with .the t o t a l task  score (PB-TT) o f f e r s some i n d i c a t i o n of the extent to which a p a r t i c u l a r item acts i n concert w i t h the other items appearing on the t o t a l task to measure what i s u s u a l l y conceived to be a more ge n e r a l i z e d v a r i a b l e that l o g i c a l l y subsumes the more p a r t i c u l a r i z e d v a r i a b l e s upon which the separate subtasks are^constructed. I f t h i s i s i n f a c t the case, the PB-TT c o r r e l a t i o n s w i l l encompass a somewhat lower range than t h e i r a s s o c i a t e d PB-ST c o e f f i c i e n t s . The 48 PB-TT c o e f f i c i e n t s f o r Task 1 range from —.05 to .57 w i t h a median PB-TT of .28. / 56 The Hoyt estimate of r e l i a b i l i t y i s designed to measure the i n t e r n a l consistency of a t e s t and i s comparable to the a l s o popular Kuder - Richardson KR 20 s t a t i s t i c . Nelson (1974:260) e x p l a i n s : I n t e r n a l consistency i s an estimate of the extent to which each t e s t item taps whatever the~ t e s t i s measuring. We might consider each t e s t item as a. sample t e s t from the t o t a l domain; then the i n t e r n a l consistency i s roughly equivalent to the average c o r r e l a t i o n between a l l p a i r s of items (or sample t e s t s ) . Hoyt estimates f o r the subtasks of Task 1 are g e n e r a l l y low, ranging from .00 to .68 w i t h a median r e l i a b i l i t y c o e f f i c i e n t of .31. This i s to be expected i n view of the f a c t that each subtask contains only four items. The Hoyt estimate f o r the t o t a l task i s a more dependable i n d i c a t o r as i t i s based upon a t o t a l of 48 items. For Task 1, t h i s c o e f f i c i e n t i s .80. Cronbach's alpha provides a more rig o r o u s v e r i f i c a t i o n of the p r o p o s i t i o n that the subtasks each c o n t r i b u t e to the measurement of a s i n g l e g e n e r a l i z e d v a r i a b l e . C o e f f i c i e n t alpha i s an index of the consistency of the su b t e s t s , o r , the degree to which the subtests tend to measure the same t h i n g . Cronbach suggests that t h i s alpha i s an index of how much the t o t a l t e s t score r e f l e c t s "common elements r a t h e r than a hodgepodge of elements each s p e c i f i c to one subtest." (Nelson, 1974:280). The Cronbach's alpha f o r the composite of the subtasks of Task 1 i s .75. Reviewing the four aforementioned i n d i c e s i n accord w i t h g e n e r a l l y accepted standards f o r t e s t c o n s t r u c t i o n and i n s p e c i f i c comparison w i t h s t a t i s t i c s reported f o r the p r e v i o u s l y c i t e d instruments p u r p o r t i n g to t e s t r e c o g n i t i o n of semantic equivalence of s y n t a c t i c s t r u c t u r e s , i t i s reason-able to conclude that the s t a t i s t i c a l p r o p e r t i e s of the t e s t s f o r Task 1 are adequate. 57 One of the most u s e f u l features of the LERTAP 2.0 programme i s the c o r r e l a t i o n matrix assembled from scores on a l l subtasks, the t o t a l task, and an e x t e r n a l c r i t e r i o n . Inter-subtask c o r r e l a t i o n s f o r Task 1 range from .033 to .370. The c o r r e l a t i o n s between each of the subtasks and the t o t a l task vary from a low of .340 to a high of .713; only one c o e f f i c i e n t l i e s i n the range of the i n t e r - s u b t a s k c o r r e l a t i o n s . From t h i s r e s u l t i t i s conceivable that the twelve subtasks of Task 1 each measure a unique aspect of l i n g u i s t i c competence, a l l of which combine to c o n t r i b u t e to a g e n e r a l i z e d language s k i l l , r e c o g n i t i o n of grammatical i m p l i c a t i o n . A c o r r e l a t i o n c o e f f i c i e n t i s computed between each subtask score and an e x t e r n a l c r i t e r i o n ; i n t h i s case, the v a r i a b l e i s age recorded i n months. The c o r r e l a t i o n s r e l a t e d to Task 1 are a l l very moderate, ranging from .009 to .280. The c o r r e l a t i o n between age and t o t a l task score i s .278. These r e s u l t s support the contention that w h i l e Ss1 performance may appear to improve a t higher grade i n t e r v a l s and, t h e r e f o r e , w i t h i n c r e a s i n g age, there i s s t i l l a wide v a r i a t i o n i n s y n t a c t i c comprehension among a group of Ss of a given age. One f i n a l p o i n t of i n t e r e s t i s the matter of p o s s i b l e sex d i f f e r e n c e s i n performance. A t - t e s t of independent means produced the f o l l o w i n g values: N Mean s t dev t value Males 56 39.32 5.14 Females 51 40.82 5.14 "1.51 I t i s t h e r e f o r e concluded t h a t , f o r the e n t i r e sample t r e a t e d as a s i n g l e group, sex i s not a s i g n i f i c a n t f a c t o r i n performance on Task 1. Task 2 Task 2 ; ( i d e n t i f i c a t i o n of anaphoric r e f e r e n t s ) comprises four subtasks (Test No. 3) and contains a t o t a l of 20 items. The mean task score obtained...by the e n t i r e sample pop u l a t i o n i s 14.10, approximately 70.5% of the maximum p o s s i b l e score. The standard d e v i a t i o n i s 5.27 and the range i s 0 to 20. As w i t h Task 1, the d i s t r i b u t i o n i s n e g a t i v e l y skewed. P o i n t b i s e r i a l c o r r e l a t i o n s among the subtasks of Task 2 range from .43 to .84 w i t h a median PB-ST of .74. Poin t b i s e r i a l c o r r e l a t i o n s r e l a t i n g each item to the t o t a l task range from .34 to .80 w i t h a median PB-TT of .62. Hoyt estimates of r e l i a b i l i t y f o r the subtasks of Task 2 range from .63 to .79. These tend to be much higher than the estimates f o r the subtasks of Task 1. The Hoyt estimate f o r the t o t a l task based on 20 items i s .91. The Cronbach's alpha f o r the composite of the subtasks of Task 2 i s .87. I t , t h e r e f o r e , appears that the s t a t i s t i c a l p r o p e r t i e s of Task 2 are adequate and su p e r i o r to those of Task 1. Inter-subtask c o r r e l a t i o n s f o r Task 2 range from .439 to .794. The c o r r e l a t i o n s between each of the subtasks and the t o t a l task vary from a low of .674 to a high of .930. As w i t h Task 1, only one c o e f f i c i e n t l i e s i n the range of the i n t e r - s u b t a s k c o r r e l a t i o n s . Hence, i t again seems conceivable that each subtask measures a somewhat unique., aspect of l i n g u i s t competence which when combined w i t h the other subtasks c o n t r i b u t e s to a more gen e r a l i z e d language s k i l l , the a b i l i t y to c o r r e c t l y i d e n t i f y anaphoric r e f e r e n t s . 59 The correlatioriobetween age i n months and scores on each of the subtasks of Task 2 are a l l moderate, ranging from .276 to .381. The c o r -r e l a t i o n between age and t o t a l task score i s .369. These c o e f f i c i e n t s are somewhat higher than those reported f o r Task 1. Nevertheless, the suggestion of wide v a r i a t i o n i n l i n g u i s t i c competence at a given age i s supported. F i n a l l y , w i t h regard to the p o s s i b i l i t y of sex d i f f e r e n c e s i n performance, a t - t e s t of independent means produced the f o l l o w i n g v a l u e s : N Mean s t dev t value prob Males 56 12.95 5.58 Females 51 15.37 4.64 "2.43 <.025 I t i s ther e f o r e concluded t h a t , f o r the e n t i r e sample t r e a t e d as a s i n g l e group, sex i s a s i g n i f i c a n t f a c t o r favouring the performance of female ^Ss on Task 2. Task 3 Task 3 contains 23 items and c o n s t i t u t e s the i n d i c a t i o n of t e x t u a l locus component of Test No. 4. The mean task score obtained by the e n t i r e sample pop u l a t i o n i s 15.96, approximately 69.4% of the maximum p o s s i b l e score. The standard, d e v i a t i o n i s 3.50 and the range i s 2 to 21. This d i s t r i b u t i o n of scores, u n l i k e those f o r Task 1 and Task 2, more c l o s e l y appraoches no r m a l i t y . The p o i n t b i s e r i a l correlations:, for-.the 23 items of Task 3 range from .08 to .60 w i t h a median PB-TT of .39. The Hoyt estimate of r e l i a b i l i t y f o r Task 3 i s .72. The s t a t i s t i c a l p r o p e r t i e s of Task 3, w h i l e not as strong as those f o r Tasks 1 and 2, may nevertheless be considered adequate. The c o r r e l a t i o n between Task 3 scores and the e x t e r n a l c r i t e r i o n of age i s .466, a moderate r e l a t i o n s h i p . In the matter of p o s s i b l e sex d i f f e r e n c e s i n performance, a t - t e s t of independent means produced the f o l l o w i n g v a l u e s : prob  N Mean St dev t value Males 53 15.98 3. 19 Females 48 15.94 3. 85 0.06 not s i g . I t i s ther e f o r e concluded t h a t , f o r the e n t i r e sample tr e a t e d as a s i n g l e group, sex i s not a s i g n i f i c a n t f a c t o r i n performance on Task 3, Task 4 Task 4 contains 23 items and c o n s t i t u t e s the judgment of t r u t h value component of Test No. 4. The mean task score obtained by the e n t i r e sample p o p u l a t i o n i s 16.42, approximately 71.4% of the maximum p o s s i b l e score. The standard d e v i a t i o n i s 3.27 and the range i s 7 to 22. This d i s t r i b u t i o n of scores, u n l i k e those f o r Task 1 and Task 2, more c l o s e l y approaches n o r m a l i t y . The p o i n t b i s e r i a l c o r r e l a t i o n s f o r the 23 items of Task 4 range from ~.05 to .60 w i t h a median PB-TT of .39. The Hoyt estimate of r e l i a b i l i t y f o r Task 4 i s .66. The s t a t i s t i c a l p r o p e r t i e s of Task 4, w h i l e not as strong as those f o r Tasks 1 and 2, may nevertheless be considered adequate. The c o r r e l a t i o n between Task 4 scores and the e x t e r n a l c r i t e r i o n of age i s .304, a very moderate r e l a t i o n s h i p . In the matter of p o s s i b l e sex d i f f e r e n c e s i n performance, a t - t e s t of independent means produced the f o l l o w i n g values: N Mean St dev t value Males 56 16.45 2 .83 Females 50 16.38 3 .73 0.10 I t i s ther e f o r e concluded t h a t , f o r the e n t i r e sample tre a t e d as a s i n g l e group, sex i s not a s i g n i f i c a n t f a c t o r i n performance on Task 4. Task 5 Task 5 contains 23 items and i s based on the combination of t e x t u a l locus i n d i c a t i o n (Task 3) and t r u t h value judgment (Task 4 ) . This combination of both tasks has been designated as the c r i t e r i o n measure of t e x t comprehension. The mean task score obtained by the e n t i r e sample pop u l a t i o n i s 14.58, approximately 63.0% of the maximum p o s s i b l e score. The standard d e v i a t i o n i s 3.65 and the range i s 1 to 21. This d i s t r i b u t i o n of scores i s s i m i l a r to those f o r Task 3 and Task 4 and, u n l i k e the d i s t r i b u t i o n s f o r Task 1 and Task 2, more c l o s e l y approaches n o r m a l i t y . The p o i n t b i s e r i a l c o r r e l a t i o n s f o r the 23 items of Task 5 range from .08 to .58 w i t h a median PB-TT of .39. The Hoyt estimate of r e l i a b i l i t y f o r Task 5 i s .72. The s t a t i s t i c a l p r o p e r t i e s of Task 5, wh i l e not as strong as those f o r Tasks 1 and 2, may nevertheless be considered adequate. The c o r r e l a t i o n between Task 5 scores and the e x t e r n a l c r i t e r i o n of age i s .414, a moderate r e l a t i o n s h i p . 62 In the matter of p o s s i b l e sex d i f f e r e n c e s i n performance, a t - t e s t of independent means produced the f o l l o w i n g v a l u e s : prob N Mean St dev t value Males 53 14.47 3. 23 Females 48 14.52 4. 07 ~0.07 not sxg. I t i s ther e f o r e concluded t h a t , f o r the e n t i r e sample t r e a t e d as a s i n g l e group, sex i s not a s i g n i f i c a n t f a c t o r i n performance on Task 5. Having completed a p r e l i m i n a r y survey of the general s t a t i s t i c a l p r o p e r t i e s of the research instrument u t i l i z i n g the e n t i r e sample p o p u l a t i o n some a t t e n t i o n must next be focused upon these same p r o p e r t i e s as they r e l a t e to s p e c i f i c subsamples w i t h i n the l a r g e r sample. Given the l i m i t a t i o n s of sample s i z e , i t was considered appropriate to examine only two subsamples: a l l n a t i v e speakers and a l l E.S.L. students. An exception i s the matter of a c c u r a t e l y r e p o r t i n g estimates of r e l i a b i l i t y . To present a s i n g l e c o e f f i c i e n t f o r each of these two subsamples, i t was considered necessary to equate the number of ^ s at each grade i n t e r v a l so as not to bi a s the Hoyt c o e f f i c i e n t by a l l o w i n g any one grade i n t e r v a l to be over-represented i n the c a l c u l a t i o n . This r e d u c t i o n i n sample s i z e was accom-p l i s h e d through e l i m i n a t i o n of Ss based on a t a b l e of random numbers. More accurate i s the r e p o r t i n g of separate i n t e r n a l r e l i a b i l i t y estimates f o r each of s i x sample populations. In t h i s way, the comparative s u i t a b i l i t y of d i f f e r e n t tasks f o r s p e c i f i e d populations can be r e a d i l y a s c e r t a i n e d . This should be noted as an important p r e c a u t i o n against m i s a p p l i c a t i o n of a t e s t upon an unsuited p o p u l a t i o n . C o e f f i c i e n t s reported h e r e i n are, however, t e n t a t i v e . I n t e r n a l r e l i a b i l i t y estimates obtained f o r c e r t a i n subsamples may be improved when a p p l i e d to a l a r g e r sample s i z e . For example, the Hoyt c o e f f i c i e n t f o r Native Speakers, Task 3 i s .81 when a l l 39 Ss completing the task are included i n the sample. By reducing the number of Ss to 21 to maintain equal d i s t r i b u t i o n across grade i n t e r v a l s , the estimate of r e l i a b i l i t y drops to .47. Two p o i n t s of i n t e r e s t are to be observed from the comparative data: (1) the only s i g n i f i c a n t sex d i f f e r e n c e i n performance i s confined to that of E.S.L. students on Task 2, and (2) age - task c o r r e l a t i o n s f o r n a t i v e speakers are c o n s i s t e n t l y higher than f o r E.S.L. students i n accord w i t h the suggestion made elsewhere that more l i n e a r developmental patt e r n s should be apparent f o r n a t i v e speakers. The l i m i t e d magnitude of the c o r r e l a t i o n s , however, tends to v e r i f y that considerable v a r i a t i o n i n performance i s to be encountered at any given age. Relevant i n d i c e s as o u t l i n e d are set out i n Tables 1 - 5 to f a c i l i t a t e d i r e c t comparisons between the n a t i v e and non-native samples. Problems i n s t a t i s t i c a l a n a l y s i s of the data Having evaluated the s t a t i s t i c a l p r o p e r t i e s of each task i n r e l a t i o n to the e n t i r e sample p o p u l a t i o n , the i n v e s t i g a t o r wished to make c e r t a i n comparisons among subsamples. I t was o r i g i n a l l y intended to apply a standard two-way a n a l y s i s of variance to the data as a f i r s t step to l o c a t i n g any s i g n i f i c a n t d i f f e r e n c e s . One mathematical assumption upon which the a n a l y s i s of variance procedure i s based i s that of homogeneity of sample v a r i a n c e s . Tests of homogeneity were ther e f o r e a p p l i e d to the data at hand: maximum variance / minimum vari a n c e r a t i o and B a r t l e t t -Box F, the l a t t e r g e n e r a l l y considered to be the most appropriate of a l l 64 N.S. E.S.L. Mean . .. 42.86 .. .38.23 St dev, 4.52 4.73 Range of scores 25 to 48 28 to 48 PB - ST range .00 to .91 .00 to .84 Median PB - ST .64 .57 PB - TT range ~.03 to .74 ~.01 to .56 Median PB - TT .35 .27 Hoyt r e l i a b i l i t y .69 .73 Cronbach's alpha .63 .69 Inter-subtask c o r r s ".207 to . 608 ".053 to .417 Subtask - T o t a l task c o r r s .293 to .734 .370 to .707 Age - T o t a l task c o r r .507 .302 * based on equal number of Ss at each Grade i n t e r v a l Hoyt estimate of r e l i a b i l i t y (Cronbach's alpha) f o r s i x sample populations Elem JrSec SrSec N.S. .86 (.82) .64 (.58) .32 (.21) E.S.L. .79 (.79) .64 (.55) .66 (.64) ** based on a l l Ss completing the task t - t e s t f o r sex d i f f e r e n c e s N Mean st dev t value prob. N.S. Males 19 43.16 3.61 Females 24 42.63 5.19 0.38 not s i g . E.S.L. Males 37 37.35 4.69 Females 27 39.22 4.61 "1.59 not s i g . Table 1: Test s t a t i s t i c s f o r Task 1: Native Speakers and E.S.L. Students compared N.S. E.S.L. Mean 14.72 13.82 , St dev 6.53 4.26 Range of scores 0 to 20 1 to 20 PB - ST range .53 to .93 .38 to .80 Median PB - ST .85 .70 PB - TT range .49 to .92 .25 to .77 Median PB - TT .81 .51 Hoyt r e l i a b i l i t y .96 .84 Cronbach's alpha .92 .80 Inter-subtask c o r r s .540 to .916 .318 to .639 Subtask - T o t a l task c o r r s .720 to .970 .655 to .869 Age - T o t a l task c o r r .685 .132 * based on equal number of Ss at each Grade i n t e r v a l Hoyt estimate of r e l i a b i l i t y (Cronbach's alpha) f o r s i x sample populations Elem JrSec SrSec N.S. .97 (.94) .91 ( .85) .10 (".37) E.S.L. .88 (.83) .72 ( .71) .88 (.82) ** based on a l l Ss completing the task t - t e s t f o r sex d i f f e r e n c e s N Mean s t dev t value prob. N.S. Males 19 13.42 6.83 Females 24 15.54 6.23 ~1.06 not s i g E.S.L. Males 37 12.70 4.91 Females 27 15.22 2.65 "2.64 < .025 Table 2: Test s t a t i s t i c s f o r Task 2: Native Speakers and E.S.L. Students compared 66 N.S. E.S.L. Mean 15.56 16.29 St dev 4.26 3.03 Range of scores 2 to 21 10 to 21 PB - Task range ~.02 to .82 .02 to .63 Median PB - Task .47 .35 Hoyt r e l i a b i l i t y .47 .62 Age - Task c o r r .596 .365 * based on equal number of Ss at each Grade i n t e r v a l Hoyt estimate of r e l i a b i l i t y f o r s i x sample populations Elem JrSec SrSec N.S. . .87 .29 .16 E.S.L. .23 .68 .67 ** based on a l l j>s completing the task t - t e s t f o r sex d i f f e r e n c e s N Mean N.S. Males 18 15.61 Females 21 15.52 E.S.L. Males 35 16.17 Females 27 16.26 s t dev t value prob 4.05 4.54 0.06 not s i g . 2.70 3.28 ~0.12 not s i g . Table 3: Test s t a t i s t i c s f o r Task 3: Native Speakers and E.S.L. Students compared 67 N.S. E.S.L. Mean 17.26 16.22 St dev 3.27 3.13 Range of scores 7 to 22 8 to 21 PB - Task range ~.09 to .71 ~.08 to .61 Median PB - Task .40 .35 * Hoyt r e l i a b i l i t y .70 .57 Age - Task c o r r .470 .310 * based on equal number of Ss at each Grade i n t e r v a l Hoyt estimate of r e l i a b i l i t y f o r s i x sample populations Elem JrSec SrSec N.S. .73 .70 .46 E.S.L. .43 .60 .75 ** based on a l l Ss completing the task t - t e s t f o r sex d i f f e r e n c e s N Mean N.S. Males 19 17.84 Females 23 16.26 E.S.L. Males 37 15.73 Females 27 16.48 s t dev t value prob 3.29 3.62 1.47 not s i g . 2.29 3.89 ~0.90 not s i g . Table 4: Test s t a t i s t i c s f o r Task 4: Native Speakers and E.S.L. Students compared 68 Mean St dev Range of scores PB - Task range Median PB - Task Hoyt r e l i a b i l i t y Age - Task c o r r N.S. 14.62 4.34 1 to 21 ,15 to .78 .47 .46 .602 14.54 3.25 8 to 20 .01 to .63 .35 .80 .288 * based on equal number of Ss at each Grade i n t e r v a l Hoyt estimate of r e l i a b i l i t y f o r s i x sample populations N.S. E.S.L. Elem .85 .45 JrSec .12 .64 SrSec .42 .77 ** based on a l l Ss completing the task t - t e s t f o r sex d i f f e r e n c e s N Mean N.S. Males 18 14.94 Females 21 14.29 E.S.L. Males 35 14.23 Females 27 14.70 st dev t value prob 4.43 4.41 0.47 not s i g . 2.53 3.87 ~0.58 not s i g . Table 5: Test s t a t i s t i c s f o r Task 5: Native Speakers and E.S.L. Students compared such t e s t s i n the case of unequal c e l l s . (Winer, 1962:95). As can be noted i n Tables 6 - 10, the homogeneity of variances requirement i s c l e a r l y v i o l a t e d w i t h respect to a l l tasks on which the subsamples are to be compared. Ferguson (1976:234) warns: "Gross departure from homo-geneity may lead to r e s u l t s which are s e r i o u s l y i n e r r o r . " and advises t h a t , under c e r t a i n c o n d i t i o n s , mathematical transformations of the data may be a p p l i e d to reduce the v a r i a t i o n among va r i a n c e s . A search f o r a s u i t a b l e transformation, however, d i d not prove f r u i t f u l i n t h i s i n s t a n c e . C o n s i d e r a t i o n was given to adding or d e l e t i n g Ss to e f f e c t u n i f o r m i t y of sample s i z e . This course was r e j e c t e d f o r two reasons, one s t a t i s t i c a l , the other e m p i r i c a l . F i r s t , a proper s e l e c t i o n of a d d i t i o n a l or r e t a i n e d Ss would most probably resemble the present sub-samples and thereby f a i l to reduce the range of va r i a n c e s . While the c o n d i t i o n of equal sample s i z e does g e n e r a l l y abate the demand f o r homo-geneity, a r a t i o as great as 20:1 among variances cannot be considered to l i e w i t h i n the tol e r a n c e of the a n a l y s i s of vari a n c e procedure. Secondly, i t has been w e l l e s t a b l i s h e d t h a t , f o r n a t i v e speakers of E n g l i s h , a c q u i -s i t i o n of s y n t a c t i c s t r u c t u r e s , p a r t i c u l a r l y some of those s e l e c t e d f o r t h i s study, occurs over a broad age range (Chomsky, 1969) . An i n s p e c t i o n of Tables 6 and 7 r e l a t i n g to performance on Tasks 1 and 2, r e s p e c t i v e l y , w i l l r e v e a l a p a t t e r n of decreasing variances across grade i n t e r v a l s (grade placement being h i g h l y c o r r e l a t e d w i t h age) f o r n a t i v e speakers. This, phenomenon i s c o n s i s t e n t w i t h e a r l i e r r e p o rts which suggest that one might expect to encounter greater v a r i a t i o n s i n l i n g u i s t i c a b i l i t i e s among younger p u p i l s . C h i l d r e n l e a r n i n g E n g l i s h as a second language might be bound by s i m i l a r c o n s t r a i n t s i n t h e i r a b i l i t y to grasp the und e r l y i n g meaning of 70 certain syntactic forms which may, or may not, be presented in the programme of second language instruction. Like their native speaker counterparts, the E.S.L. subsamples exhibit a pattern of decreasing variance in the case of Task 1 (Table 6). This trend, however, i s not evident in a comparison among the E.S.L. subsamples with respect to Task 2 (Table 7). It therefore seems reasonable to postulate that, in some instances, a lack of homogeneity of variances is an inevitable occurance when one chooses to make comparisons among widely disparate populations with respect to any variable that may be characterized as an^aspect ..of linguistic competence. In view of these problematic results, the investigator elected to proceed with an analysis of the data by use of appropriate non-parametric tests. 71 Comparison of samples by Language group and Grade i n t e r v a l The f o l l o w i n g sequence i s used i n presenting summary comparisons of the performance of a l l subsamples; (1) Mean scores, v a r i a n c e s , and number of Ss, l i s t e d i n t h i s format: 39.17< Subsample mean score 12 39.42-« Subsample var i a n c e i n scores f 1 Number of Ss i n the subsample (2) Indices of homogeneity of variances (3) Mean ranks: K r u s k a l - W a l l i s non-parametric one-way a n a l y s i s of v a r i a n c e , c o r r e c t e d f o r t i e s (4) Summary of s i g n i f i c a n t d i f f e r e n c e s c a l c u l a t e d according to a procedure developed by Dunn (1964) f o r the purpose of s i m u l -taneously conducting m u l t i p l e comparisons from a s i n g l e set of ranked data. Elem JrSec SrSec N.S. 39.17 43.20 44.81 12 39.42 10 10.84 21 4.06 E.S.L.(A) 37.50 36.74 40.17 10 19.83 23 20.29 12 13.97 E.S.L.(B) 37.83 36.86 42.33 6 49.37 7 13.48 6 15.47 Maximum s / Minimum s = 12.154 B a r t l e t t - Box F = 2.907, p 4.01 Elem JrSec SrSec 50.38 73.10 85.12 37.10 32.54 52.58 45.17 32.43 67.75 K r u s k a l - W a l l i s ANOVA, % 2 = 44.323, p 4.001 JrSec N.S. v. JrSec E.S.L. (A) .001 JrSec N.S. v. JrSec E.S.L. (B) .01 SrSec N.S. V. SrSec E.S.L. (A) P < .01 Elem N.S. V. JrSec N.S. .05 Elem N.S. V. SrSec N.S. v<c .001 JrSec E.S.L. (A) V. SrSec E • S • L • (A) P < .05 JrSec E.S.L. (B) V. SrSec E.S.L. (B) P < .025 N.S. E.S.L.(A) E.S.L.(B) Table 6: Comparison of performance on Task 1 by Language group and Grade i n t e r v a l Elem JrSec SrSec N.S. 7.50 15.70 18.14 12 57.18 10 23.57 21 2.93 E.S.L.(A) 13.80 15.22 14.33 1 0 1 2 . 4 0 2 3 4 . 7 2 12 28.61 E.S.L.(B) 9.00 13.29 12.33 6 34.80 7 26.24 6 17.07 Maximum s / Minimum s = 19.525 B a r t l e t t - Box F = 5.797, p<.001 Elem JrSec SrSec N.S. 28.21 66.10 84.60 E.S.L.(A) 43.90 52.67 56.46 E.S.L.(B) 24.33 45.36 35.08 K r u s k a l - W a l l i s ANOVA,"*-2 = 40.035, p^.001 SrSec N.S. v. SrSec E.S.L. (A) p<.01 SrSec N.S. v. SrSec E.S.L.(B) p 4.001 Elem N.S. v. JrSec N.S. p <.01 Elem N.S. v. SrSec N.S. p<.001 Table 7: Comparison of performance on Task 2 by Language group and Grade i n t e r v a l E l e m JrSec SrSec N.S. 12.00 14.71 18.00 12 28 •55 7 5.24 20 3.58 E .S.L.(A) 15.40 16.30 17.45 10 2. •93 23 10.31 11 3.87 E •S.L.(B) 14.00 15.83 17.50 6 6, .80 6 11.77 6 15.50 2 Maximum s / Minimum s 2 = 9.73-1 B a r t l e t t - Box F = 3.125, p<.01 Elem JrSec SrSec N.S. 26.17 34.21 69.67 E. S.L.(A) 39.50 53.50 65.73 E. S.L.(B) 29.75 46.83 66.00 K r u s k a l - W a l l i s ANOVA, 1C 2 = 28.735, p<.001 Elem N.S v. SrSec N.S. P < .001 JrSec N.S v. SrSec N.S. P < .001 Elem E.S .L.-(A) v. SrSec E.S.L.(A) P < .025 Elem E.S .L.(B) v. SrSec E.S.L.(B) P < .025 Table 8: Comparison of performance on Task 3 by Language group and Grade i n t e r v a l Elem JrSec SrSec N.S. 15.00 16.00 18.65 12 16.73 10 13.56 20 4.77 E.S.L.(A) 15.40 16.39 16.83 10 8.27 23 7.98 12 8.33 E.S.L.(B) 13.83 15.71 16.83 6 4.17 7 12.57 6 20.97 Maximum s / Minimum s = 5.03 2 B a r t l e t t - Box F =. 1.263, not s i g . Elem JrSec SrSec N.S. 42.25 50.20 74.25 E.S.L.(A) 42.45 52.59 57.13 E.S.L.(B) 25.83 46.50 62.83 K r u s k a l - W a l l i s ANOVA, X.2 = 18.323, p<.025 Elem N.S. v. JrSec N.S. v. Elem E.S.L.(B) v. SrSec N.S. SrSec N.S. SrSec E.S.L.(B) P<.01 p <.025 p < .025 Table 9: Comparison of performance on Task 4 by Language group and Grade i n t e r v a l Elem JrSec SrSec N.S. 10.83 14.43 16.90 12 27.42 7 4.29 20 5. 67 E. S.L.(A) 14.10 14.57 15.09 10 6.10 23 9.44 11 7. 89 E. S.L.(B) 12.17 14.17 15.50 6 7.37 6 15.37 6 24. 30 Maximum s 2 2 / Minimum s = 6.399 B a r t l e t t - Box F = 1.975, p <.05 Elem JrSec SrSec N.S. 29.54 46.36 71.07 E. S.L.(A) 45.10 50.43 56.45 E. S.L.(B) 28.17 46.83 61.42 K r u s k a l - W a l l i s ANOVA, X 2 = 21 .538, p< .01 Elem N.S. v. SrSec N.S. p < .001 JrSec N.S. v. SrSec N.S. p<.025 Elem E.S.L, .(B) v. SrSec E.S.L.(B) p <.025 Table 10: Comparison of performance on Task 5 by Language group and Grade i n t e r v a l 77 Hypothesis I : Results At given l e v e l s of educational attainment as determined by grade placement, (1) the performance of native speakers i s superior to that of students for whom English i s a second language, and ( 2 ) the performance of Chinese speakers does not d i f f e r from that of other E.S.L. students on each task included i n the present instrument. An inspection of Task 1 (recognition of grammatical implication) r e s u l t s shows that at each grade i n t e r v a l the performance of native speakers i s superior to that of either sample of E.S.L. students. Of s i x possible comparisons, only three, at the secondary grade i n t e r v a l s , are s t a t i s t i c a l l y s i g n i f i c a n t . Chinese speakers at each grade i n t e r v a l score s l i g h t l y lower than other E.S.L. students but .these differences are not s i g n i f i c a n t . Task 2 ( i d e n t i f i c a t i o n of anaphoric referents) r e s u l t s i n d i c a t e native speakers superior to E.S.L. students only at the secondary i n t e r v a l s , with the two contrasts f or senior secondary subsamples s t a t i s t i c a l l y s i g n i -f i c a n t . . Native speakers at the elementary i n t e r v a l appear to be markedly i n f e r i o r i n t h i s s k i l l although no s i g n i f i c a n t difference can be established with e i t h e r of the E.S.L. samples. Chinese speakers at a l l grade i n t e r v a l s c l e a r l y score higher than other E.S.L. students but none of the differences prove s i g n i f i c a n t . On Task 3 ( i n d i c a t i o n of textual l o c u s ) , native speakers are superior to the E.S.L. samples only at the senior secondary i n t e r v a l . Chinese speakers score somewhat higher than other E.S.L. students only at the two lower i n t e r v a l s . 78 None of these d i f f e r e n c e s c i t e d f o r Task 3 are s t a t i s t i c a l l y s i g n i f i c a n t . Task 4 (judgment of t r u t h value) r e s u l t s mark only s e n i o r secondary n a t i v e speakers s u p e r i o r to both E.S.L. samples. At the two lower i n t e r v a l s n a t i v e speakers are su p e r i o r only to the non-Chinese group of E.S.L. students. Chinese speakers at the two lower i n t e r v a l s score higher than other E.S.L. students, there being a t i e between the two s e n i o r secondary samples. None of these d i f f e r e n c e s c i t e d f o r Task 4 are s t a t i s t i c a l l y s i g n i f i c a n t . F i n a l l y , the c r i t e r i o n Task 5 r e s u l t s show a p a t t e r n wherein n a t i v e speakers are i n f e r i o r to e i t h e r E.S.L. sample at the elementary i n t e r v a l , rank between Chinese and the other E.S.L. students at the j u n i o r secondary i n t e r v a l , and score higher than e i t h e r E.S.L. sample at the s e n i o r secondary i n t e r v a l . Chinese speakers at the two lower i n t e r v a l s score over other E.S.L. students, the trend being reversed at the se n i o r secondary i n t e r v a l . None of these d i f f e r e n c e s c i t e d f o r Task 5 are s t a t i s t i c a l l y s i g n i f i c a n t . Hypothesis I : Conclusions Regarding the t e s t s of s y n t a c t i c comprehension outside of a t e x t , r e s u l t s of Task 1 and Task 2 do not f u l l y support the hypothesis of n a t i v e speaker s u p e r i o r i t y across grade i n t e r v a l s . The hypothesis of no d i f f e r e n c e between Chinese speakers and other E.S.L. students on measures of these s k i l l s i s f i r m l y supported. 79 With respect to the measures of t e x t comprehension, r e s u l t s of Task 3, Task 4, and Task 5 a l l f a i l to confirm the hypothesis of n a t i v e speaker s u p e r i o r i t y w h i l e c l e a r l y supporting the hypothesis of no d i f f e r e n c e between Chinese speakers and other E.S.L. students. Hypothesis I : D i s c u s s i o n A n a l y s i s of the comparative data f o r n a t i v e speakers and second language students i n d i c a t e s t h a t , a f t e r 800 hours of i n s t r u c t i o n , E.S.L. students at the secondary grade i n t e r v a l s are more l i k e l y to underperform t h e i r n a t i v e speakers peers than E.S.L. p u p i l s i n the elementary school. For the f i r s t two tasks of s y n t a c t i c comprehension, these d i f f e r e n c e s are s t a t i s t i c a l l y s i g n i f i c a n t i n most cases. This observation lends some support to the contention that younger c h i l d r e n tend to progress b e t t e r i n l e a r n i n g a second language than do o l d e r students. I t may be more e f f i c i e n t i n a c h i e v i n g n a t i v e - l i k e competencies to promote entry i n t o an E.S.L. programme at an e a r l y age. Comparisons across grade i n t e r v a l s r e v e a l that even a f t e r a ttending an E.S.L. programme over a prolonged period of time, students do not recognize grammatical i m p l i c a t i o n r e l a t i o n s as r e a d i l y as t h e i r n a t i v e speaker peers. Results f o r the anaphoric i d e n t i f i c a t i o n task are l e s s c l e a r . An apparent anomaly i s evident i n the comparative a b i l i t y of the elementary subsamples, the E.S.L. samples outperforming n a t i v e speakers. Some suggestion was made e a r l i e r that young c h i l d r e n may operate under severe memory c o n s t r a i n t s which i n h i b i t anaphoric r e s o l u t i o n . This a s s e r t i o n may be true but whether the age advantage of the elementary E.S.L. subsample over the n a t i v e speaker counterpart, 11 y r s 1 mo and 10 y r s 0 mo, r e s p e c t i v e l y , can account f o r the former's higher (though 80 not s t a t i s t i c a l l y s i g n i f i c a n t ) mean score on Task 2 i s open to question. A more p l a u s i b l e explanation may l i e i n a conscious r e c o g n i t i o n , o f t e n r e f l e c t e d i n E.S.L. c u r r i c u l a , of the need f o r focused i n s t r u c t i o n i n anaphoric s t r u c t u r e s . Regarding the measures of t e x t comprehension — Tasks 3, 4, and 5, i t i s noteworthy that no s i g n i f i c a n t d i f f e r e n c e s between n a t i v e speakers and E.S.L. students can be e s t a b l i s h e d at any grade i n t e r v a l . This f i n d i n g s t r o n g l y i m p l i e s that a high l e v e l of r e c o g n i t i o n of grammatical i m p l i c a t i o n r e l a t i o n s and the a b i l i t y to i d e n t i f y anaphoric r e f e r e n t s are not necessary to the attainment of n a t i v e - l i k e p r o f i c i e n c y i n t e x t comprehension tasks. I t does not, however, preclude the p o s s i -b i l i t y that these l i n g u i s t i c s k i l l s may have a f a c i l i t a t i n g e f f e c t on such tasks. As a f i n a l p o i n t of d i s c u s s i o n , i t i s i n t e r e s t i n g to compare c e r t a i n r e s u l t s of t h i s study w i t h those reported by van Metre (1974) who found f o r Grade 3 p u p i l s that a b i l i t y to comprehend s t r u c t u r e s examined by Chomsky (1969) does not d i s c r i m i n a t e between n a t i v e (monolingual) and non-native ( b i l i n g u a l ) p u p i l s but r a t h e r i s a b e t t e r p r e d i c t o r of reading comprehension scores. When comparison of the present r e s u l t s i s r e s t r i c t e d to the elementary grade i n t e r v a l o n l y , which j3s are most s i m i l a r to those of van Metre, i t i s noted that the s u p e r i o r performance of n a t i v e speakers on Task 1 i s not s t a t i s t i c a l l y s i g n i f i c a n t . A d d i t i o n a l a n a l y s i s of the data i n d i c a t e s that f o r secondary students, Task 1 d i s c r i m i n a t e s between high and low scorers on t e x t comprehension (Task 5) as w e l l as between n a t i v e speakers and E.S.L. students. By i s o l a t i n g the Chomsky based items — Subtasks G, H, J , L, and R — from the remainder of Tasks 1 and 2, i t can be 81 observed that these f u n c t i o n to d i s c r i m i n a t e between, high and low f comprehenders (Task 5) among the elementary subsamples without separating the E.S.L. students from the n a t i v e speakers. These r e s u l t s are remarkably c o n s i s t e n t w i t h van Metre, e s p e c i a l l y when i t i s r e c a l l e d that her sample comprised p u p i l s at high and low extremes of measured reading comprehension i n c o n t r a s t to the present study whose Ss are more homogeneous i n t e x t comprehension a b i l i t y as evidenced by the unimodal d i s t r i b u t i o n of Task 5 scores. Hypothesis I I : Results In the case of both n a t i v e speakers and students f o r whom E n g l i s h i s a second language, performance on each task included i n the present instrument i s augmented at i n c r e a s i n g l e v e l s of e d u c a t i o n a l attainment as determined by grade placement. The a n t i c i p a t e d gradient i s evident i n the Task 1 ( r e c o g n i t i o n of grammatical i m p l i c a t i o n ) mean scores of n a t i v e speakers, the elementary subsample performing s i g n i f i c a n t l y below e i t h e r of the secondary i n t e r v a l s . The gradient f o r E.S.L. samples i s i n t e r r u p t e d by depressed means f o r the j u n i o r secondary i n t e r v a l s whose performance i s s i g n i f i c a n t l y i n f e r i o r only to t h e i r s e n i o r secondary counterparts. Results f o r Task 2 ( i d e n t i f i c a t i o n of anaphoric r e f e r n t s ) are very s i m i l a r to those f o r Task 1 i n the case of n a t i v e speakers — a p o s i t i v e gradient w i t h the performance of the elementary subsample s i g n i f i c a n t l y i n f e r i o r to the secondary i n t e r v a l s . Results f o r E.S.L. students d i f f e r from the preceding task i n s o f a r as scores f o r both of the j u n i o r secondary i n t e r -v a l s are elevated from a l i n e a r g r a d i e n t . No s i g n i f i c a n t d i f f e r e n c e s are reported between any i n t e r v a l s of e i t h e r E.S.L. sample. 82 Task 3 ( i n d i c a t i o n of t e x t u a l locus) shows unbroken gradients f o r a l l three samples. In each case, the s e n i o r secondary i n t e r v a l i s s i g n i f i c a n t l y s u p e r i o r to the elementary; f o r n a t i v e speakers, there i s a l s o a s i g n i f i c a n t d i f f e r e n c e between the j u n i o r secondary and the s e n i o r secondary i n t e r v a l s . Task 4 (judgment of t r u t h value) gradients are defined f o r a l l three samples. Performance of s e n i o r secondary n a t i v e speakers i s s i g n i f i c a n t l y s u p e r i o r to that of the lower i n t e r v a l s . The only s i g n i -f i c a n t d i f f e r e n c e : f o u n d across grade i n t e r v a l s f o r E.S.L. students i s that between the elementary and s e n i o r secondary non-Chinese l e a r n e r s . Gradients f o r Task 5, the comprehension c r i t e r i o n , a l s o show no d e v i a t i o n s f o r any of the samples grouped by n a t i v e language f a c t o r . Again, s i g n i f i c a n t d i f f e r e n c e s are confined to n a t i v e speakers, s e n i o r secondary outperforming both of the lower i n t e r v a l s , and a c o n t r a s t between elementary and s e n i o r secondary E.S.L. speakers of languages other than Chinese. Hypothesis I I : Conclusions The hypothesis of augmented performance across grade i n t e r v a l s i s g e n e r a l l y confirmed f o r n a t i v e speakers on a l l tasks although not a l l c o n t r a s t s are s t a t i s t i c a l l y s i g n i f i c a n t . Regarding E.S.L. students, s i m i l a r r e s u l t s o b t a i n f o r the tasks of t e x t comprehension. Discounting small d i f f e r e n c e s i n mean scores which are not s t a t i s t i c a l l y s i g n i f i c a n t , l i k e gradients may evolve f o r the f i r s t two tasks of s y n t a c t i c compre-hension as w e l l . I t w i l l be r e c a l l e d , however, that c e r t a i n d i f f e r e n c e s between f i r s t and second language a c q u i s i t i o n could c o n t r i b u t e to d e v i -a t i o n s i n a gradient curve f o r an E.S.L. sample across grade i n t e r v a l s w h i l e a p o s i t i v e progression would be a d e f i n i t e e xpectation f o r n a t i v e speakers. N. S. E.S.L.(A) E.S.L.(B) Elem JrSec SrSec Elem JrSec SrSec Elem JrSec SrSec A. P a s s i v i z a t i o n 3.08 3.70 3.95 3.80 3.65 3.75 3.50 3.57 4.00 B. P a r t i c i p l e modif 2.50 3.40 3.76 1.20 1.70 2.33 2.17 0.86 2.33 C. Wh— f r o n t i n g 3.50 3.80 3.86 3.30 3.44 3.58 3.83 3.57 4.00 D. Re l t v z / Claus conj 2.92 3.40 . 3.57 2.90 3.04 3.25 2.83 3.43 3.50 E. R e l t v z / Pron d e l e t 3.17 3.40 3.71 2.70 2.91 3.17 3.00 3.14 3.50 F. Double transform 3.75 3.50 3.95 3.90 3.83 3.75 3.83 3.71 3.83 G. Ask (q) / T e l l 2.92 3.50 3.57 3.10 2.70 3.17 2.83 3.14 3.17 H. Easy to see 3.83 4.00 3.81 3.30 2.61 3.08 3.00 2.00 4.00 J . Promise / T e l l 2.75 3.30 3.33 2.60 3.09 3.50 2.67 2.86 2.83 K. I n d i r e c t speech 3.42 3.70 3.71 3.40 3.65 3.42 3.17 3.57 3.50 L. Ask (r) / T e l l 3.75 3.80 3.71 3.80 3.04 3.58 3.50 3.29 4.00 M. Pseudoimperatives 3.58 3.70 3.86 3.50 3.09 3.58 3.50 3.71 3.67 P. Pron r e f 1.81 3.26 3.86 3.26 3.11 3.10 2.10 2.69 3.05 R. Pron r e f (MDP) 1.27 3.28 3.73 3.04 3.17 2.60 1.60 2.97 2.13 S. Nominal subst . 1.47 3.36 3.73 1.84 2.85 2.60 2.13 2.74 1.87 T. C l a u s a l subst 1.22 2.27 3.05 2.67 3.01 3.22 0.89 1.90 2.67 (s. to cores on Subtasks: P, R, co i n c i d e w i t h the s c a l e S, T are converted to an f o r a l l other subtasks.) equivalent of a maximum of 4.00 Table 11: Task 1 and Task 2 Subtask mean scores, Language group by Grade i n t e r v a l 84 Hypothesis I I : D i s c u s s i o n A p p r e c i a t i o n of co n t r a s t s across grade i n t e r v a l s may be gained by examining the patterns of subtask mean scores f o r Task 1 and Task 2 (Table 11). Departures from a curve of augmented growth i n any of these s k i l l s of l i n g u i s t i c i n t e r p r e t a t i o n are i n most cases minimal, p a r t i c u -l a r l y f o r n a t i v e speakers. The more frequent i n t e r r u p t i o n s i n the curves f o r the E.S.L. samples may be due to the l i m i t a t i o n s of sample s i z e . B a r r i n g the p o s s i b l e e f f e c t s of d i f f e r e n c e s i n f a m i l i a r i t y , there i s no conceptual reason why ol d e r l e a r n e r s should f i n d p a r t i c u l a r s t r u c t u r e s more d i f f i c u l t to comprehend than do younger p u p i l s . D i s c u s s i o n w i t h p a r t i c i p a t i n g teachers o f f e r e d no in f o r m a t i o n to suggest the content of i n s t r u c t i o n experienced by any one grade i n t e r v a l d i f f e r e d from that of the other i n t e r v a l s w i t h i n the E.S.L. samples r e s p e c t i n g the v a r i a b l e s examined. I n s o f a r as p o s s i b l e , £[s were drawn from a number of programmes so as to randomize such i n s t r u c t i o n a l e f f e c t s . These f a c t o r s , combined w i t h the observation that d e v i a t i o n s from a p o s i t i v e curve occur w i t h n e a r l y equal frequency a t the lower and upper ends of the grade i n t e r v a l range, tend to suggest the p r o b a b i l i t y of fewer instances of such v i o l a -t i o n s i f the sample s i z e of E.S.L. students were increased. The anaphora subtasks evidence a p a t t e r n that s u b s t a n t i a t e s the Bormuth et a l . (1970) c l a i m that school age c h i l d r e n are unable to compre-hend many common s y n t a c t i c s t r u c t u r e s . This i s r e f l e c t e d i n the la r g e gains between elementary and j u n i o r secondary n a t i v e speakers, an obser-v a t i o n that can a l s o be g e n e r a l l y a p p l i e d to the E.S.L. samples where the gradient i s l e s s steep, p o s s i b l y as a r e s u l t of d i f f e r e n c e s i n c u r r i c u l u m content f o r n a t i v e and non-native p u p i l s i n the elementary grades:. These s u b s t a n t i a l gains by secondary students a l s o support the contention that Grade 4 p u p i l s ' c a p a c i t y f o r anaphora r e s o l u t i o n may be constrained by short term memory l i m i t a t i o n s (Lesgold, 1972). In d i s c u s s i n g r e s u l t s r e l a t e d to the preceding hypothesis, i t was suggested that anaphora s t r u c t u r e s are o f t e n featured i n E.S.L. c u r r i c u l a . I f E.S.L. teachers do cons c i o u s l y i d e n t i f y t h i s i n s t r u c -t i o n a l need, i t appears that there i s a tendency to b r i n g non-native students of a l l ages to a f a i r l y uniform c r i t e r i o n , evidenced by the narrower range of anaphora subtask mean scores f o r E.S.L. students across grade i n t e r v a l s i n c o n t r a s t to the broad developmental gradient e x h i b i t e d by n a t i v e speakers. P a r t i c u l a r l y , s e n i o r secondary E.S.L. students seem to be i n need of more i n s t r u c t i o n i n anaphoric i n t e r p r e -t a t i o n . Not only do they l a g behind j u n i o r secondary students, but t h i s i s the only i n t e r v a l at which the E.S.L. samples perform s i g n i f i c a n t l y below n a t i v e speaker counterparts. Regarding the measures of t e x t comprehension — Tasks 3, 4, and 5, i t i s perhaps i n t e r e s t i n g to note t h a t , f o r n a t i v e speakers, s i g n i f i c a n t d i f f e r e n c e s are found i n the narrow range between j u n i o r secondary and s e n i o r secondary i n t e r v a l s but not between the elementary and j u n i o r secondary. In c o n t r a s t , where a s i g n i f i c a n t d i f f e r e n c e occurs on any of these three tasks between E.S.L. i n t e r v a l s , i t i s only f o r a broad range comparison between elementary and s e n i o r secondary i n t e r v a l s . These observations, of course, need to be repeated on l a r g e r samples before the p o i n t of major gain i n these s k i l l s can be designated w i t h a reasonable degree of c e r t a i n t y . 86 Hypothesis I I I : Results The rank order of d i f f i c u l t y of the s i x t e e n subtasks of Task 1 ( r e c o g n i t i o n of grammatical i m p l i c a t i o n ) and Task 2 ( i d e n t i f i c a t i o n of anaphoric r e f e r e n t s ) as determined by subtask mean scores does not vary among any subsamples. The appropriate s t a t i s t i c to t e s t t h i s hypothesis i s Kendall's c o e f f i c i e n t of concordance (W). T o t a l agreement among a group of three or more subsamples i s manifest i f W = 1.000; a maximum d i s a r r a y i s expressed by W = .000. An i n s p e c t i o n of Table 11 shows a few instances wherein subtask mean scores d i f f e r by l e s s than .05. I t was considered that t h i s narrow margin was i n s u f f i c i e n t to merit separate rankings; r a t h e r , a l l subtask mean scores were rounded to one decimal place to t r e a t s m all v a r i a t i o n s as t i e d rankings before computing the c o e f f i c i e n t s which, i n each case, c o n t a i n a c o r r e c t i o n f o r t i e s . An o v e r a l l comparison f o r nine subsamples r e s u l t e d i n W = .663 (p<.001). A d d i t i o n a l comparisons were made by language group and by grade i n t e r v a l and are here summarized. N.S. E.S.L.(A) E.S.L.(B) .736 .780 .845 Table 12: Kendall's c o e f f i c i e n t of concordance by Language group (p-d.Ol) N.S. E.S.L.(A) E.S.L.(B) .859 .620 .681 Table 13: Kendall's c o e f f i c i e n t of concordance by Grade i n t e r v a l (p<.025) 87 These r e s u l t s c l e a r l y i n d i c a t e that even the most c o n s i s t e n t grouping f a l l s short of t o t a l agreement. Hypothesis I I I : Conclusions The hypothesis of no d i f f e r e n c e i n the rank order of subtask d i f f i c u l t y among subsamples cannot be accepted as s i g n i f i c a n t c o e f f i c i e n t s demonstrate s i m i l a r i t i e s among subsamples to f l u c t u a t e w i t h i n a moderate to high range of c o r r e l a t i o n . Hypothesis I I I : D i s c u s s i o n Two f i n d i n g s of some i n t e r e s t a r i s e from the t e s t of t h i s hypothesis based on the assumption that r e l a t i v e d i f f i c u l t y of l i n g u i s t i c tasks should remain constant f o r both f i r s t and second language l e a r n e r s and at v a r y i n g stages of language development. The f i r s t i s an i n d i c a t i o n of a somewhat greater degree of s i m i l a r i t y across grade i n t e r v a l s among E.S.L. students than f o r n a t i v e speakers.(Table 12). One p o s s i b l e reason f o r t h i s phenomenon may be that the conscious E n g l i s h language experience of the non-native student i s of n e c e s s i t y more c o n t r o l l e d through a narrow focus of E.S.L. i n s t r u c t i o n i n c o n t r a s t to the greater v a r i e t y of language forms to which the n a t i v e speaker can attend. The second observation to be made i s that the c o e f f i c i e n t s across language groups are lower f o r secondary than f o r elementary students (Table 13). In r e t r o s p e c t , from a developmental view, t h i s trend i s . t o be expected as n a t i v e speakers have, over s e v e r a l years of growth, i n c r e a s i n g opportunity to d i v e r s i f y t h e i r language experience while more mature E.S.L. students might be s e l e c t i v e i n a c q u i r i n g those s y n t a c t i c s t r u c t u r e s perceived to most adequately meet t h e i r l i n g u i s t i c needs. Whatever the cause of t h i s p a t t e r n , an important i n s t r u c t i o n a l c o n s i d e r a t i o n i s i m p l i c i t i n the r e s u l t s . Based on the sampling c r i t e r i a employed f o r t h i s study, i t would appear that elementary E.S.L. students about to be admitted to i n t e g r a t e d ( n a t i v e and non-native speaker). c l a s s e s i n which a f u l l e d u c a t i o n a l programme i s o f f e r e d are more s i m i l a r to n a t i v e speaker peers i n the nature of t h e i r language competence as measured by s y n t a c t i c comprehension than t h e i r counterparts at the secondary grade i n t e r v a l s . Comparison of samples by Language group In accord w i t h the plan of t h i s study as o u t l i n e d i n Chapter IV, the grade i n t e r v a l s of each s a m p l e — n a t i v e speakers N.S. , Chinese l e a r n e r s of E n g l i s h E.S.L.(A) , and E.S.L. students of other n a t i v e languages E.S.L.(B) — are next combined to form three s i n g l e groups. Consideration of the number of ^s at each grade i n t e r v a l i n the combined samples (Table 14), however, suggested that the three groups may not be comparable. While a symmetry i s apparent i n the E.S.L. samples wherein one h a l f of the Ss are at the j u n i o r secondary i n t e r v a l or below and the other h a l f are at that i n t e r v a l or above, i t was noted that the n a t i v e speaker sample i s skewed by a greater number of Ss at the s e n i o r secondary i n t e r v a l . Elem JrSec SrSec N.S. 12 10 21 E.S.L.(A) 10 23 12 E.S.L.(B) 6 7 6 Table 14: Number of Ss i n each Language group by Grade i n t e r v a l 89 To adjust f o r t h i s b i a s , nine Ss at t h i s i n t e r v a l were e l i m i n a t e d from the c o l l e c t i v e n a t i v e speaker sample through a t a b l e of random numbers i n order to gain a symmetry across grade i n t e r v a l s s i m i l a r to that of the E.S.L. samples (Table 14a). Elem JrSec SrSec 12 10 12 Table 14a: Number of Ss i n adjusted Native Speaker sample by Grade i n t e r v a l A check on the data p e r t a i n i n g to these combined subsamples i n d i c a t e s that the r e q u i s i t e s f o r the parametric a n a l y s i s of vari a n c e procedure — a reasonable approximation to no r m a l i t y and homogeneity of variances — have been g e n e r a l l y s a t i s f i e d . Hypothesis IV : Results Combined across grade i n t e r v a l s , (1) the performance of n a t i v e speakers i s s u p e r i o r to that of students f o r whom E n g l i s h i s a second language, and (2) the performance of Chinese speakers does not d i f f e r from that of other E.S.L. students on each task and subtask included i n the present instrument. A one-way a n a l y s i s of variance was conducted on each set of task and subtask scores. 'A p o s t e r i o r i ' m u l t i p l e range t e s t s were s e l e c t e d to l o c a t e s i g n i f i c a n t d i f f e r e n c e s among sample mean scores: (1) Scheffe (alpha = .05), and (2) Duncan (alpha = .05). These two t e s t s may be viewed as complementary checks against Type I and Type I I e r r o r s , r e s p e c t i v e l y , 90 i . e . , at a s p e c i f i c alpha l e v e l , the Scheffe procedure y i e l d s the fewest s i g n i f i c a n t d i f f e r e n c e s , the Duncan, the most, when a p p l i e d to the same data base (Ferguson, 1976:300). In each instance where a s i g n i f i c a n t d i f f e r e n c e i s claimed, i t i s s u b s t a n t i a t e d by both procedures w i t h one notable exception f o r Task 2. Results depicted i n Tables 15 - 19 may be summarized. Task 1 as a whole s i g n i f i c a n t l y d i s c r i m i n a t e s between n a t i v e speakers and both groups of E.S.L. students. Associated subtasks c o n t r i -b u t i n g to t h i s r e s u l t are B: P a r t i c i p l e m o d i f i e r s and H: Easy to see, and, to a l e s s e r extent as they only separate the Chinese speakers from the other two groups, E: R e l a t i v i z a t i o n by pronoun d e l e t i o n and M: Pseudo-imperatives . Task 2 r e s u l t s are l e s s c e r t a i n , there being a discrepancy between the two m u l t i p l e range procedures: the Duncan (.05) t e s t f i n d s a s i g n i f i c a n t d i f f e r e n c e between Chinese and the other E.S.L. students; the Scheffe (.05) appraises a l l three groups to be one homogeneous subset. To r e s o l v e t h i s c o n f l i c t , an a d d i t i o n a l procedure, Student-Newman-Keuls (.05), described as a "compromise" measure between Type I and Type I I e r r o r s (Ferguson, i b i d . ) , was conducted, l e a d i n g to no s i g n i f i c a n t d i f f e r -ences. This i n d i c a t i o n and the observation that the one-way a n a l y s i s of variance f o r Task 2 y i e l d s an i n s i g n i f i c a n t F r a t i o (p = .111) prompts the i n v e s t i g a t o r to conclude that Task 2 as a whole cannot be s a i d to d i s c r i -minate among any of the language groups. One a s s o c i a t e d subtask, T: C l a u s a l s u b s t i t u t i o n , nevertheless r e v e a l s a c l e a r l y s u p e r i o r performance by Chinese speakers over that of e i t h e r the other E.S.L. students or n a t i v e speakers. 91 Comparison of samples by Language group The f o l l o w i n g sequence i s used i n presenting summary comparisons of the performance of subsamples combined across Grade i n t e r v a l s : (1) Mean scores, v a r i a n c e s , and number of Ss, l i s t e d i n t h i s format: 42.32^ Subsample mean score 34 24.41^ Subsample variance i n scores t I Number of S!s i n the subsample (2) Indices of homogeneity of variances (3) M u l t i p l e range t e s t s (1) N.S. E.S.L,.(A) E.S.L. (B) 42.32 37.82 38.89 34 24.41 45 19.83 19 28.43 F = 8.801, p <.01 2 2 (2) Maximum s / Minimum s = 1.43,4 B a r t l e t t - Box F = 0.458, not s i g . (3) Scheffe (.05) Duncan (.05) N.S. v. E.S.L.(A),E.S.L.(B) N.S. v. E.S.L.(A),E.S.L.(B) Table 15: Comparison of performance on Task 1 by Language group N.S. E.S.L.(A) E.S.L.(B) 3.56 3.71 3.68 34 0.678 45 0.301 19 0.228 F = 0.564,. .not s i g . 2 2 Maximum s / Minimum s = 2.974 B a r t l e t t - Box F = 4.642, p^.01 Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 15a: Comparison of performance on Subtask A by Language group _ N_._S. _ E.S.L. (A) E.S.L. (B) 3.18 1.76 1.74 34 1.059 45 1.462 19 1.982 F = 15.906, p^.OOl 2 2 Maximum s / Minimum s = 1.872 B a r t l e t t - Box F = 1.197, n o t , s i g . Scheffe (.05) Duncan (.05) N.S. v. E.S.L.(A),E.S.L.(B) N.S. v. E.S.L.(A),E.S.L.(B) Table 15b: Comparison of performance on Subtask B by Language group N.S. E.S.L.(A) E.S.L.(B) 3.68 3.44 3.79 34 0.407 45 0.571 19 0.287 F = 2.134, not s i g . 2 2 Maximum s / Minimum s = 1.992 B a r t l e t t - Box F = 1.509, not s i g . Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 15c: Comparison of performance on Subtask C by Language group N.S. E.S.L.(A) E.S.L.(B) 3.29 3.07 3.26 34 0.578 45 0.245 19 0.649 F = 1.316, not s i g . 2 2 Maximum s / Minimum s = 2.645 . B a r t l e t t - Box F = 4.489, p-<.025 Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 15d: Comparison of performance on Subtask D by Language group N.S. E.S.L.(A) E.S.L.(B) 3.41 2.93 3.21 34 0.613 45 0.427 19 0.509 F = 4.446, p<.025 2 2 Maximum s / Minimum s = 1.435 B a r t l e t t - Box F = 0.613, not s i g . Scheffe (.05) Duncan (.05) N.S. v. E.S.L.(A) N.S. v. E.S.L.(A) Table 15e: Comparison of performance on Subtask E by Language group N.S. E.S.L.(A) E.S.L.(B) 3.76 3.82 3.79 34 0.185 45 0.149 19 0.175 F = 0.195, not s i g . 2 2 Maximum s / Minimum s = 1.240 B a r t l e t t - Box F = 0.231, not s i g . Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 15f: Comparison of performance on Subtask F by Language group N.S. E.S.L.(A) E.S.L.(B) 3.35 2.91 3.05 34 .1.205 45 0.992 19 0.941 F = 1.807, not s i g . 2 2 Maximum s / Minimum s = 1.280 B a r t l e t t - Box F = 0.245, not s i g . Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 15g: Comparison of performance on Subtask G by Language group N.S. E.S.L.(A) E.S.L.(B) 3.82 2.89 2.95 34 0.210 45 0.874 19 0.941 F = 14.272, p^.001 2 2 Maximum s / Minimum s = 4.476 B a r t l e t t - Box F = 9.082, p<^.001 Scheffe (.05) Duncan (.05) N.S. v. E.S.L.(A),E.S.L.(B) N.S. v. E.S.L.(A),E.S.L.(B) Table 15h: Comparison of performance on Subtask H by Language group N.S. E.S.L.(A) E.S.L.(B) 3.24 3.09 2.79 34 0.731 45 0.855 19 1.398 F = 1.326, not s i g . 2 2 Maximum s / Minimum s = 1.912 B a r t l e t t - Box F = 1.349, not s i g . Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 1 5 i : Comparison of performance on Subtask J by Language group N.S. E.S.L.(A) E.S.L.(B) 3.59 3.53 3.42 34 0.431 45 0.436 19 0.591 F = 0.368, not s i g . 2 2 Maximum s / Minimum s = 1.369 B a r t l e t t - Box F = 0.362, not s i g . Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 15j: Comparison of performance on Subtask K by Language group N.S. E.S.L.(A) E.S.L.(B) 3.71 3.36 3.58 34 0.456 45 1.098 19 0.591 F = 1.578, not s i g . 2 2 Maximum s / Minimum s = 2.406 B a r t l e t t - Box F = 3.703, p<T.025 Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 15k: Comparison of performance on Subtask L by Language group N.S. E.S.L.(A) E.S.L.(B) 3.74 3.31 3.63 34 0.261 45 0.537 19 0.690 F = 4.014, p<r.025 2 2 Maximum s / Minimum s = 2.642 B a r t l e t t - Box F = 3.276, p«C.05 Scheffe (.05) Duncan (.05) N.S. v. E.S.L.(A) N.S. v. E.S.L.(A) Table 151: Comparison of performance on Subtask M by Language group N.S. E.S.L.(A) E.S.L.(B) 13.56 14.67 11.63 34 48.14 45 12.41 19 26.69 F = 2.253, not s i g . 2 2 Maximum s / Minimum s = 3.879 B a r t l e t t - Box F = 8.452, p<;.001 Scheffe (.05) Duncan (.05) no s i g . d i f f . E.S.L.(A) v. E.S.L.(B) Table 16: Comparison of performance on Task 2 by Language group N.S. E.S.L.(A) E.S.L.(B) 5.24 5.49 4v58 34 6.246 45 1.710 19 3.702 F = 1.511, not s i g . 2 2 Maximum s / Minimum s = 3.652 B a r t l e t t - Box F = 7.740, p<.001 Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 16a: Comparison of performance on Subtask P by Language group N.S. E.S.L.(A) E.S.L.(B) 3.41 3.73 2.84 34 3.765 45 1.109 19 3.362 F = 2.173, not s i g . 2 2 Maximum s / Minimum s = 3.394 B a r t l e t t - Box F = 7.547, p<£.001 Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 16b: Comparison of performance on Subtask R by Language group N.S. E.S.L.(A) E.S.L.(B) 3.47 3.20 2.84 34 3.590 45 2.073 19 2.474 F = 0.909, not s i g . 2 2 Maximum s / Minimum s = 1.732 B a r t l e t t - Box F = 1.451, not s i g . Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 16c: Comparison of performance on Subtask S by Language group 100 (1) N.S. E.S.L.(A) E.S.L.(B) 1.59 2.24 1.37 34 1.462 45 0.598 19 1.023 F = 7.019, p .01 2 2 (2) Maximum s / Minimum s = 2.444 B a r t l e t t - Box F = 3.754, p4.025 (3) Scheffe (.05) Duncan (.05) E.S.L.(A) v. N.S.,E.S.L.(B) E.S.L.(A) v. N.S.,E.S.L.(B) Table 16d: Comparison of performance on Subtask T by Language group (1) N.S.. E.S.L.(A) E.S.L.<B) 15.06 16.39 15.78 31 20.39 44 7.31 18 12.18 F = 1.264, not s i g . 2 2 (2) Maximum s / Minimum s = 2.789 B a r t l e t t - Box F = 4.659, p<.01 (3) Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 17: Comparison of performance on Task 3 by Language group N.S. E.S.L.(A) E.S.L.(B) 16.68 16.29 15.47 34 13.80 45 8.03 19 12.71 F = 0.809, not s i g . 2 2 Maximum s / Minimum s = 1.719 B a r t l e t t - Box F = 1.516, not s i g . Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 18: Comparison of performance on Task 4 by Language group N.S. E.S.L.(A) E.S.L.(B) 14.13 14.59 13.94 31 20.78 44 8.06 18 15.82 F = 0.250, not s i g . 2 2 Maximum s / Minimum s = 2.578 B a r t l e t t - Box F = 4.080, p <.025 Scheffe (.05) Duncan (.05) no s i g . d i f f . no s i g . d i f f . Table 19: Comparison of performance on Task 5 by Language group 102 The measures of t e x t comprehension — Tasks 3, 4, and 5, are a l l c h a r a c t e r i z e d by no s i g n i f i c a n t d i f f e r e n c e s across language groups. Considering each language group i n t u r n , n a t i v e speakers s i g n i f i c a n t l y outperform both groups of E.S.L. students on Task 1 and a s s o c i a t e d subtasks, B: P a r t i c i p l e m o d i f i e r s and H: Easy to see, only. On an a d d i t i o n a l two subtasks, E: R e l a t i v i z a t i o n by pronoun d e l e t i o n and M: Pseudoimperatives, n a t i v e speaker performance i s s i g n i f i c a n t l y s u p e r i o r to that of Chinese speakers. Chinese speakers s i g n i f i c a n t l y outperform n a t i v e speakers as w e l l as other E.S.L. students on Subtask T: C l a u s a l s u b s t i t u t i o n of Task 2. Non-Chinese E.S.L. students i n no instance s i g n i f i c a n t l y out-perform e i t h e r of the other two groups. Hypothesis IV : Conclusions The hypothesis of n a t i v e speaker s u p e r i o r i t y cannot be e s t a b l i s h e d by the present r e s u l t s , except f o r Task 1, and by the f a i l u r e to a t t a i n s i g n i f i c a n t d i f f e r e n c e s on twelve out of s i x t e e n subtasks. The hypothesis of no d i f f e r e n c e between the E.S.L. groups i s supported by the r e s u l t s of a l l tasks, d e s p i t e any question concerning Task 2, and by the p a t t e r n i n g of scores on f i f t e e n of the s i x t e e n subtasks. Hypothesis IV : D i s c u s s i o n As s t a t e d e a r l i e r , even a f t e r a lengthy period of i n s t r u c t i o n , E.S.L. students remain g e n e r a l l y i n f e r i o r to n a t i v e speakers i n t h e i r comprehension of most s y n t a c t i c forms t e s t e d . Only i n a few i n s t a n c e s , notably i n connection w i t h p a s s i v i z a t i o n transforms, do E.S.L. mean scores exceed those of the n a t i v e speaker sample on subtasks of grammatical i m p l i -c a t i o n . 103 A review of the r e l a t i v i z a t i o n subtasks: C, D, and E, shows Chinese speakers to be at a n o t i c e a b l e ..disadvantage'-from other; E.S...L. students who compare r a t h e r favourably w i t h the n a t i v e speaker sample. For both groups of E.S.L. l e a r n e r s , however, c e r t a i n of the items i n d i c a t e a l a c k of f a c i l i t y i n i n t e r p r e t i n g "nested" f o r m s — e.g., The boy (the g i r l h i t ) f e l l down. S i m i l a r l y , Chinese speakers seem to have more d i f f i c u l t y than other l e a r n e r s i n d i s c e r n i n g c o n t r a s t s i n sentence meaning that i n v o l v e the minimal d i s t a n c e p r i n c i p l e , although the other E.S.L. group a l s o c o n s i s t e n t l y scores below the n a t i v e speaker sample on the four r e l a t e d subtasks: G, H, J , and L. Among the twelve subtasks of Task 1, two stand out as e f f e c t i v e d i s c r i m i n a t o r s between n a t i v e speakers and e i t h e r of the E.S.L. groups — B: P a r t i c i p l e m o d i f i e r s and H: Easy to see. F a m i l i a r i t y w i t h E.S.L. curr i c u l u m m a t e r i a l s w i l l r e a d i l y suggest why t h i s should be so. Neither s t r u c t u r e i s given much emphasis i n b a s i c programmes, e s p e c i a l l y i n terms of d e l i b e r a t e l y manipulating, the s y n t a c t i c patterns to i l l u s t r a t e c o n t r a s t s i n meaning. Subtask M: Pseudoimperatives d i s c r i m i n a t e s only against the Chinese l e a r n e r s , and to a l e s s e r extent than the aforementioned two, even though i t a l s o t e s t s a seldom taught c o n s t r u c t i o n . Subtask N: Agentless p a s s i v i z a t i o n i s of some i n t e r e s t i n that the great m a j o r i t y of Ss, n a t i v e speakers and E.S.L. students a l i k e , may have given too broad an i n t e r p r e t a t i o n to the lexeme, "someone", a l l o w i n g the i n c l u s i o n of non-human r e f e r e n t s . P r e l i m i n a r y item a n a l y s i s data i n d i c a t e d a d i s p a r i t y between t h i s and the other subtasks of Task 1 f o r . which reason i t was removed from a l l subsequent analyses i n the study. Consequently, no comparative mean scores are reported. 104 Comparative data f o r the anaphora subtasks are counter to i n i t i a l expectations. While n a t i v e speaker mean scores are higher than those f o r the m u l t i e t h n i c E.S.L. group on a l l subtasks, they f a l l below those a t t a i n e d by the Chinese speakers on three of the four subtasks. I t i s d i f f i c u l t to p o s t u l a t e a reason f o r these r e s u l t s . Ss f o r both E.S.L. samples were drawn from the same c l a s s e s thereby sug-gesting e q u a l i t y of i n s t r u c t i o n a l opportunity. While the anaphora sub-tasks were c a s t i n a "constructed response" format, the i n v e s t i g a t o r endeavoured to apply c o n s i s t e n t standards of a c c e p t a b i l i t y to the responses of a l l Ss. A c o n t r a s t i n language typology does not r e a d i l y account f o r the super i o r performance of the Chinese l e a r n e r s nor can i t e x p l a i n the i n f e r i o r scores of the n a t i v e speakers. The l a c k of s i g n i f i c a n t d i f f e r e n c e s among the three samples on the measures of t e x t comprehension — Tasks 3, 4, and 5 has already been observed i n comments on Hypothesis I where some of the i m p l i c a t i o n s of t h i s f i n d i n g were noted. Hypothesis V : Results The rank order of d i f f i c u l t y of the s i x t e e n subtasks of Task 1 ( r e c o g n i t i o n of grammatical i m p l i c a t i o n ) and Task 2 ( i d e n t i f i c a t i o n of anaphoric r e f e r e n t s ) as determined by subtask mean scores does not vary among any subsamples combined across grade i n t e r v a l s . A f t e r a d j u s t i n g subtask mean scores l i s t e d i n Table 20 by rounding to one decimal place to reduce the e f f e c t of numerous s m a l l , and p o s s i b l y spurious, d i f f e r e n c e s i n rankings, Kendall's c o e f f i c i e n t of concordance, corrected f o r t i e s , was computed upon the three language groups, r e s u l t i n g i n W = .809 (p<:.01). N. S. E.S.L.(A) E.S.L.(B) H. Easy to see 3. 82 F. Double transform 3. 82 C. Wh— f r o n t i n g 3. 79 F. Double transform 3. 76 A. P a s s i v i z a t i o n 3. 71 F. Double transform 3. 79 M. Pseudolmperatlves 3. 74 K. Indirect-speech 3. 53 A. P a s s i v i z a t i o n 3. 68 L. Ask (r) / T e l l 3. 71 C. Wh— f r o n t i n g 3. 44 M. Pseudoimperatives 3. 63 C. Wh— f r o n t i n g 3. 68 L. Ask (r) / T e l l 3. 36 L. Ask (r) / T e l l 3. 58 K. I n d i r e c t speech 3. 59 M. Pseudoimperatives 3. 31 K. I n d i r e c t speech 3. 42 A. P a s s i v i z a t i o n 3. 56 P. Pron r e f 3. 14 D. Re l t v z / Claus conj 3. 26 E. R e l t v z / Pron d e l e t 3. 41 J. Promise / T e l l 3. 09 E. Re l t v z / Pron d e l e t 3. 21 G. Ask (q) / T e l l 3. 35 D. R e l t v z / Claus conj 3. 07 G. Ask (q) / T e l l 3. 05 D. R e l t v z / Claus conj 3. 29 T. C l a u s a l subst 2. 99 H. Easy to see 2. 95 J. Promise / T e l l 3. 24 R. Pron r e f (MDP) 2. 98 J. Promise / T e l l 2. 79 B. P a r t i c i p l e modif 3. 18 E. R e l t v z / Pron d e l e t 2. 93 P. Pron r e f 2. 62 P. Pron r e f 2. 99 G. Ask (q) / T e l l 2. 91 R. Pron r e f (MDP) 2. 27 S. Nominal subst 2. 78 H. Easy to see 2. 89 S. Nominal subst 2. 27 R. Pron r e f (MDP) 2. 73 s. Nominal subst 2. 56 T. C l a u s a l subst 1. 83 T. C l a u s a l subst 2. 12 B. P a r t i c i p l e modif 1. 76 B. P a r t i c i p l e modif 1. 74 (Scores on Subtasks: P, R, S, T are converted to an equivalent of a maximum of 4.00 to c o i n c i d e w i t h the s c a l e f o r a l l other subtasks.) Table 20: Comparative rankings of subtask d i f f i c u l t y by Language group o Cn 106 Of f u r t h e r i n t e r e s t i s the degree of s i m i l a r i t y i n rank order of subtask d i f f i c u l t y between the two E.S.L. samples and between each E.S.L. sample and the n a t i v e speaker sample. Kendall's tau, a non-para-metric c o r r e l a t i o n c o e f f i c i e n t , was s e l e c t e d as the index most s u i t e d to comparisons c o n t a i n i n g s e v e r a l t i e d ranks ( c f . Nie et a l . , 1975:289). R e s u l t s , c o r r e c t e d f o r t i e s , are summarized. E.S.L. (A) w i t h E.S.L.(B) .655 E.S.L. (A) w i t h N.S. .382 E.S.L.(B) w i t h N.S. .687 Table 21: K e n d a l l tau c o e f f i c i e n t s f o r p a i r e d comparisons among Language groups (p<.025) Hypothesis V : Conclusions While i t appears that the three language groups possess a r a t h e r high degree of agreement i n t h e i r rankings of subtask d i f f i c u l t y as i n d i -cated by the c o e f f i c i e n t of concordance, a c l o s e r examination of the data u t i l i z i n g a more conservative measure, Kendall's tau, y i e l d s more moderate c o r r e l a t i o n s between the two E . S . L . groups and the m u l t i e t h n i c students tE.S.L.(B)l and the n a t i v e speakers. By c o n t r a s t s , Chinese l e a r n e r s ( J E . S . L . ( A ) ^ show a much lower degree of agreement w i t h n a t i v e speakers on the r e l a t i v e d i f f i c u l t y of subtasks. Therefore, the hypothesis of no d i f f e r e n c e among the combined subsamples cannot be accepted. Hypothesis V : D i s c u s s i o n Having e s t a b l i s h e d a degree of d i s p a r i t y i n the rankings between Chinese and non-Chinese l e a r n e r s (tau = .655), i t i s of some i n t e r e s t to note that the two E . S . L . groups, g e n e r a l l y comparable i n t h e i r o v e r a l l 107 4 performance on the subtasks , show such marked d i f f e r e n c e s i n t h e i r degrees of agreement w i t h the rank order of subtask d i f f i c u l t y estab-l i s h e d f o r n a t i v e speakers. The procedure of p a i r e d comparisons was undertaken to provide some i n d i c a t i o n as to whether or not a s p e c i f i c n a t i v e language determines which s y n t a c t i c forms of E n g l i s h are the most d i f f i c u l t to comprehend. The s t r i k i n g c o n t r a s t i n c o e f f i c i e n t s obtained by comparisons w i t h n a t i v e speakers (tau = .382 and .687 f o r Chinese and non-Chinese students, respec-t i v e l y ) does s t r o n g l y suggest that the order of d i f f i c u l t y of t r a n s f o r - . mations and anaphoric types tes t e d i s not independent of the students' n a t i v e language. As might be expected, the group of which h a l f are speakers of European languages performs i n a way more s i m i l a r to n a t i v e speakers of E n g l i s h than do Chinese students of equal p r o f i c i e n c y . The remainder of t h i s d i s c u s s i o n w i l l examine some notable trends among the s p e c i f i c subtasks. Because the present study was not conceived upon any norms of l i n g u i s t i c f a m i l i a r i t y f o r s p e c i f i e d p o p u l a t i o n s , no d e f i n i t i v e statement can be made regarding the e f f e c t of t h i s f a c t o r as discussed e a r l i e r i n the l i t e r a t u r e review. With respect to t r a n s f o r m a t i o n a l complexity, however, a t t e n t i o n can be drawn to a number of i n t e r e s t i n g observations. O'Donnell, G r i f f i n , and N o r r i s (1967) r e p o r t that younger c h i l d r e n use r e l a t i v e clauses much more f r e q u e n t l y than noun m o d i f i c a t i o n by a p a r t i -c i p l e . The present r e s u l t s i n d i c a t e that a l l three language groups f i n d 4. Of the 68 items comprising the s i x t e e n subtasks, the E.S.L.(A) sample mean i s 52.49 (77.2%); the E.S.L.(B) sample mean i s 50.53 (74.3%). t = 0.94, not s i g . 108 s t r u c t u r e s c o n t a i n i n g a p a r t i c i p l e as a noun m o d i f i e r (Subtask B) more d i f f i c u l t to i n t e r p r e t than any of the forms of r e l a t i v i z a t i o n t e s t e d . Another expectation that i s confirmed supposes a subtask that t e s t understanding of two s t r u c t u r e s , both of which conform to the minimal di s t a n c e p r i n c i p l e , i s e a s i e r than.a subtask i n v o l v i n g a c o n t r a s t between a conforming and a non-conforming s t r u c t u r e . As i n d i c a t e d i n Table 20, Subtask L: Ask (request) / T e l l e l i c i t s a b e t t e r performance from a l l groups of Ss than e i t h e r Subtask G: Ask (query) / T e l l or Subtask J : Promise / T e l l . L a s t l y , from the viewpoint of t r a n s f o r m a t i o n a l complexity, i t might be p r e d i c t e d that Subtask F i n v o l v i n g both r e l a t i v i z a t i o n and p a s s i -v i z a t i o n would be more d i f f i c u l t than subtasks c h a r a c t e r i z e d by only one of these transformations, i . e . , Subtasks A, D, and E. Contrary to such expec-t a t i o n s , the p r e d i c t e d order of d i f f i c u l t y i s v i o l a t e d by a l l groups. This r e s u l t may be explained p a r t l y by the presence i n most items of Subtasks D and E of r e l a t i v e pronoun d e l e t i o n , a transform which has been reported to be e s p e c i a l l y d i f f i c u l t f o r elementary p u p i l s (Fagan, 1969). One other p r e d i c t i o n , a l s o unconfirmed, i s that Subtask M: Pseudoimperatives would be among the most d i f f i c u l t f o r the E.S.L. samples. Even i f the somewhat i d i o m a t i c usage of these s t r u c t u r e s precludes them from an E.S.L. cur r i c u l u m , t h i s subtask ranks as f a i r l y easy f o r both n a t i v e and non-native speakers. Performance on the anaphora subtasks must be i n t e r p r e t e d w i t h c a u t i o n ; i t can be argued that scores f o r Subtasks P, R, S, and T are not d i r e c t l y comparable to those of other subtasks s i n c e a "constructed response" format was employed f o r the former, a "yes/no choice" f o r the l a t t e r ; hence, 109 the tendency f o r anaphora subtask scores to f i l t e r toward the bottom ranks. I f a "yes/no choice" format does g e n e r a l l y f a c i l i t a t e performance over "constructed responses", the marked d i f f i c u l t y of Subtask B: P a r t i -c i p l e m o d i f i e r s f o r E.S.L. students becomes a l l the more apparent. F i n a l l y , these four subtasks were thought to represent a hie r a r c h y of complexity, p a r t i c u l a r l y i n the context of the S^  having to form h i s own response. The p r e d i c t e d order of subtask mean scores was: P: Pronominal reference R: Pronominal reference (Minimal Distance P r i n c i p l e ) S: Nominal s u b s t i t u t i o n T: C l a u s a l s u b s t i t u t i o n . R e s u l t s , however, do not r e f l e c t t h i s p a t t e r n across the three language groups. The c o n t r a s t i n d i f f i c u l t y between Subtask P and Subtask R i s nevertheless supported. Hypothesis VI : Results A comparison of the three measures of t e x t compre-hension e x h i b i t s a p a t t e r n of diminished performance wherein Task 4 (judgment of t r u t h value) scores are greater than Task 3 ( i n d i c a t i o n of t e x t u a l locus) scores which are greater than Task 5 scores. This hypothesis was tested by means of a s e r i e s of c o r r e l a t e d t - t e s t s , r e s u l t s of which are presented i n Tables 22a, b, c. Hypothesis VI : Conclusions While the hypothesis i s supported by the data f o r n a t i v e speakers, d i f f e r e n c e s between Task 3 and Task 4 performance do not occur as p r e d i c t e d f o r e i t h e r of the E.S.L. samples. I t i s a l s o to be noted t h a t , f o r each group, h i g h l y s i g n i f i c a n t d i f f e r e n c e s are evident between each of the component tasks and the combined measure (Task 5), thereby i n d i c a t i n g that n e i t h e r Task 3 nor Task 4 can Mean s t dev t value Task 4 17.00 3.45 Task 3 15.06 4.52 Task 5 14.13 4.56 * Task 4 / Task 5, t value Table 22a: Co r r e l a t e d t - t e s t s across three measur 4.76 p 4.001 5.40 p<.001 = 8.22, p < .001 or Native Speakers (N = 31 s of t e x t comprehension Mean s t dev t value Task 4 16.39 2.79 Task 3 16.39 2.70 0.00 not s i g Task 5 14.59 2.84 7.69 p < .001 * Task 4 / Task 5, t value = 10.92, P < .001 Table 22b: Co r r e l a t e d t - t e s t s f o r E.S.L.(A) Students (N = across three measures of t e x t comprehension Mean s t dev t value Task 4 15.50 3.67 Task 3 15.78 3.49 0.57 not s i g Task 5 13.94 3.98 7.46 p<.001 * Task 4 / Task 5, t value = 4.18, p <.001 Table 22c: C o r r e l a t e d t - t e s t s f o r E.S.L.(B) Students (N = across three measures of t e x t comprehension I l l stand alone as an adequate measure of comprehension as e a r l i e r defined f o r t h i s study. Hypothesis VI : D i s c u s s i o n The suggested r a t i o n a l e f o r the p r e d i c t e d order of d i f f i c u l t y f o r the three measures of t e x t comprehension was based on r e l a t i v e proba-b i l i t i e s of c o r r e c t responses. Where the p r e d i c t e d order i s upheld, however, a l t e r n a t i v e explanations are p o s s i b l e ; where r e s u l t s c o n t r a d i c t p r o b a b i l i t i e s , other accounts must be sought. Native speaker performance observes the p r e d i c t e d p a t t e r n . One reason Task 3 ( i n d i c a t i o n of t e x t u a l locus) scores are lower than those f o r Task 4 (judgment of t r u t h value) may be that the Ss, p a r t i c -u l a r l y at the secondary i n t e r v a l s , found the passage r e l a t i v e l y easy to comprehend on f i r s t reading. I f so, inform a t i o n storage, be i t accurate or not, would have been f a c i l i t a t e d , rendering the task of c i t i n g the locus f o r one's response unusual and bothersome. By c o n t r a s t , the E.S.L. samples appear to have found one compo-nent task as d i f f i c u l t as the other. I t may be that second language l e a r n e r ' s comprehension s t r a t e g i e s are l e s s g l o b a l than those of n a t i v e speakers. Consequently, they would be somewhat accustomed to s c r u t i -n i z i n g a t e x t f o r i n d i v i d u a l sentences that provide i n f o r m a t i o n appro-p r i a t e to a given item statement, such as appears i n Test No. 4. The foregoing explanations are admittedly s p e c u l a t i v e . Because the a d m i n i s t r a t i o n of the instrument d i d not c a l l f o r the enforcement of a s t r i c t time l i m i t , i t i s not p o s s i b l e to present evidence here to suggest that a higher incidence of response v e r i f i c a t i o n through locus searching d i d occur among the E.S.L. samples. As an a d d i t i o n a l note of i n t e r e s t , Frase and Washington (1970) 112 found that, for elementary school pupils, comprehension errors increase sharply i f more than one sentence has to be processed. A comparison of the item d i f f i c u l t i e s of eleven single sentence locus items (Nos. 1, 3, 5, 7, 11, 12, 16, 17, 18, 20, 21) with six multiple sentence locus items (Nos. 2, 4, 9, 13, 14, 22) — q.v.. Table 23 — establishes that while for the entire sample of this study indicating the locus of multiple sentence items i s noticably more exacting a task, judging their truth value i s not much more d i f f i c u l t than for single, sentence items. single multiple sentence sentence Task 3 .781 .445 Task 4 .755 .684 Task 5 .679 .426 Table 23: Mean item d i f f i c u l t i e s of single sentence and multiple sentence items for entire sample population (N = 107) Hypothesis VII : Results Task 1 (recognition of grammatical implication) and Task 2 (identification of anaphoric referents) are each better predictors of Task 5 (the combined measure of locus indication and truth value judgment) than of Task 4 (judgment of truth value only). Task 1 and Task 2 are equally good predictors of Task 5 scores. Correlations indicate that Task 1 and Task 2 are each better predictors of Task 5 than of Task 4 only for the native speaker' sample. In the case of the E.S.L.(A) sample, Task 2, but not Task 1, better predicts Task 5 than Task 4 for these Chinese speakers. Neither Task 1 nor Task 2 predicts Task 5 better than Task 4 for the multiethnic E.S.L.(B) sample, (cf. Tables 24a, b, c) Task 1 Task 2 Task 3 Task 4 Task 5 Task 1 .581 .713 .661 .703 Task 2 .694 .664 .709 Task 3 .872 .977 Task 4 .919 Task 5 Table 24a : Correlation coefficients completing a l l tasks (N = for 31, Native Speakers p <.01) Task 1 Task 2 Task 3 Task 4 Task 5 Task 1 .481 .275 .417 .390 Task 2 .380 .598 .625 Task 3 .797 .845 Task 4 .925 Task 5 Table 24b: Correlation coefficients for. E. S...L..(A) . completing a l l tasks (N = 44, p<.0l) Student: Task 1 Task 2 Task 3 Task 4 Task 5 Task 1 .484 .436 .540 .451 Task 2 .255 .201 .189 Task 3 .837 .969 Task 4 .918 Task 5 Table 24c: Correlation coefficients for E.S.L.(B) Students completing a l l tasks (N = 18, p<.01) 114:. C o r r e l a t i o n c o e f f i c i e n t s between each p r e d i c t o r and the c r i t e r i o n are n e a r l y equal i n the case of n a t i v e speakers. Data p e r t a i n i n g to second language students i d e n t i f i e s Task 2 as a b e t t e r p r e d i c t o r than Task 1 f o r the E.S.L.(A) sample, the converse being true f o r the E.S.L.(B) sample. Zero order c o r r e l a t i o n s , however, do not e x p l a i n the f u l l s i g n i -f i c a n c e of the r e l a t i o n s h i p of each task to the c r i t e r i o n measure of compre-hension. In order to adequately assess the r e l a t i v e importance of the l i n g u i s t i c s k i l l s t e s t e d by each task to the act of t e x t comprehension as defined at the outset of t h i s study, a more complex s t a t i s t i c a l procedure known as m u l t i p l e r e g r e s s i o n a n a l y s i s i s r e q u i r e d . M u l t i p l e r e g r e s s i o n analyses of two s o r t s are a p p l i e d to the present data. Type I sets an i d e n t i c a l i n c l u s i o n l e v e l (Nie et a l . , 1975:344) f o r a l l independent v a r i a b l e s — i n t h i s case, Task 1 and Task 2 — to be considered i n any one a n a l y s i s . This means that a l l independent v a r i a b l e s enter the a n a l y s i s on an equal f o o t i n g . The p r a c t i c a l e f f e c t of t h i s proce-dure i s to f i r s t i d e n t i f y the l e a d i n g independent v a r i a b l e and r e p o r t the amount of variance i t c o n t r i b u t e s to the dependent v a r i a b l e , Task 5. This variance i s r e f e r r e d to h e r e i n as "independent v a r i a n c e " since i t comprises that v a r i a n c e uniquely c o n t r i b u t e d by the l e a d i n g v a r i a b l e plus any pro-p o r t i o n of v a r i a n c e c o n t r i b u t e d i n common w i t h the other independent v a r i -a b le. In other words, the "independent v a r i a n c e " a s s o c i a t e d w i t h a given task i s that p r o p o r t i o n of variance i n Task 5 scores that can be explained without reference to any other task. I t i s the same as would r e s u l t i f only that p a r t i c u l a r task were entered i n t o the r e g r e s s i o n equation. Next, the l e s s c r u c i a l independent v a r i a b l e i s i d e n t i f i e d and the amount of a d d i t i o n a l variance i t c o n t r i b u t e s to the dependent c r i t e r i o n measure i s reported. 115 Type I Type I I Task 2 + (Task 1) Task 1 + (Task 2) 50.3% + (12.7%) 49.4% + (13.6%) T o t a l v a r i a n c e explained = 63.0% Table 25a: M u l t i p l e Regression Analyses: independent var i a n c e and ( a d d i t i o n a l variance) c o n t r i b u t e d by s y n t a c t i c measures to c r i t e r i o n Task 5 scores of Native Speakers (N = 31, p«£.01) Type I Type I I Task 2 + (Task 1) Task 1 + (Task 2) 39.1% .+ (1.0%) 15.2% + (24.9%) T o t a l variance explained = 40.1% Table 25b: M u l t i p l e Regression Analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by s y n t a c t i c measures to c r i t e r i o n Task 5 scores of E.S.L. (A) Students (N = 44, p^.01) Type I Type I I Task 1 + (Task 2) Task 2 + (Task 1) 20.4% + (0.1%) 3.6% + (16.9%) T o t a l variance explained = 20.5% Table 25c: M u l t i p l e Regression Analyses: independent var i a n c e and ( a d d i t i o n a l variance) c o n t r i b u t e d by s y n t a c t i c measures to c r i t e r i o n Task 5 scores of E.S.L.(B) Students (N = 18, not s i g . ) 116 In c o n t r a s t to the foregoing procedure, Type I I analyses s p e c i f y the sequence i n which the independent v a r i a b l e s enter the a n a l y s i s thereby e s t a b l i s h i n g a h i e r a r c h y of i n c l u s i o n l e v e l s . This procedure was conducted subsequent to o b t a i n i n g r e s u l t s f o r Type I analyses i n order to assess the t o t a l amount of variance c o n t r i b u t e d by the l e s s c r u c i a l task. The r e s u l t s of separate analyses f o r each language group are summarized i n Tables 25a, b, c. Hypothesis V I I : Conclusions The hypothesis that the l i n g u i s t i c s k i l l s embodied i n the f i r s t two tasks b e t t e r p r e d i c t Task 5 than Task 4 scores i s supported only i n the case of n a t i v e speakers, the data from the E.S.L. samples b r i n g i n g mixed or contrary r e s u l t s . While Task 1 and Task 2 appear to be e q u a l l y good p r e d i c t o r s of Task 5 scores f o r n a t i v e speakers, t h i s does not prove to be true among E.S.L. students where one task p r e d i c t s b e t t e r than the other depending on the p a r t i c u l a r sample. Marked d i f f e r e n c e s i n the amount of v a r i a n c e explained by Task 1 and Task 2 are noted across language groups. In p a r t i c u l a r , n a t i v e speaker performance on Task 5 can be more adequately, accounted f o r by reference to these two tasks than can be the variance i n E.S.L. students' scores on the same c r i t e r i o n measure. I t i s to be noted, however, that the r e g r e s s i o n c o e f f i c i e n t s f o r the E.S.L.(B) sample are not s t a t i s t i c a l l y s i g n i f i c a n t . Hypothesis VII : D i s c u s s i o n In order to b e t t e r i n t e r p r e t the foregoing r e s u l t s , i t should be r e c a l l e d that i t has already been demonstrated that there i s no s i g n i f i c a n t d i f f e r e n c e among the three language groups i n t h e i r performance on the c r i t e r i o n Task 5. This f a c t makes i t p o s s i b l e to conduct comparisons among samples of equal p r o f i c i e n c y to determine the extent to which various populations are able to u t i l i z e the s k i l l s measured by the f i r s t two tasks to perform the c r i t e r i o n task. Only through such a componential examination of language behaviour can the primary purpose of t h i s study be accomplished. Results i n d i c a t e that where s i g n i f i c a n t d i f f e r e n c e s occur among groups on the s y n t a c t i c measures, a greater p r o p o r t i o n of va r i a n c e i n the c r i t e r i o n scores of the su p e r i o r group can be explained by those v a r i a b l e s . S p e c i f i c a l l y , Task 1 r e s u l t s d i s c r i m i n a t e between n a t i v e speakers and the E.S.L. samples. I t i s observed that the vari a n c e c o n t r i b u t e d by Task 1 to Task 5 scores i s more than twice as great f o r n a t i v e speakers than f o r e i t h e r of the E.S.L. samples. While the i n v e s t i g a t o r p r e f e r s to consider that no s i g n i f i c a n t d i f f e r e n c e s are present among the three language groups on Task 2, one m u l t i p l e range t e s t i n d i c a t e s that the two E.S.L. samples are not from the same subset. In t h i s case, the a d d i t i o n a l v a r i a n c e c o n t r i -buted by Task 2 to Task 5 scores of the su p e r i o r E.S.L.(A) group i s probably f a r greater than that f o r the E.S.L.(B) sample even though the r e g r e s s i o n c o e f f i c i e n t s f o r the l a t t e r are not s i g n i f i c a n t . In summation, i t appears that d i f f e r e n t populations u t i l i z e the l i n g u i s t i c s k i l l s t e s t e d to va r i o u s extents i n accord w i t h t h e i r demon-s t r a t e d a b i l i t y i n those s k i l l s . Of greater consequence, however, i s the f i n d i n g that w h i l e the c a p a c i t y to recognize grammatical i m p l i c a t i o n r e l a -t i o n s and i d e n t i f y anaphoric r e f e r e n t s i s advantageous, the two s k i l l s accounting f o r most of the va r i a n c e i n the c r i t e r i o n scores of the a b l e s t group, such a b i l i t y i s not c r u c i a l to success on what have been proposed to be r e q u i s i t e s f o r demonstrating sentence comprehension w i t h i n a t e x t , 118 i . e . , i n d i c a t i o n of t e x t u a l locus and judgment of t r u t h value. This l a s t outcome may be b e t t e r appreciated by bearing i n mind that v a r i a t i o n s across populations i n the importance of s y n t a c t i c v a r i a b l e s are commonly i n c u r r e d i n language research. For example, Takahashi (1975) found d i f f e r e n c e s between Grade 6 and Grade 9 p u p i l s i n the extent to which s y n t a c t i c comprehension a f f e c t e d scores on a standardized t e s t of reading comprehension w h i l e Baines (1975) r e p o r t s that s y n t a c t i c complexity i n w r i t t e n composition i s not an e q u a l l y good p r e d i c t o r of reading compre-hension f o r elementary and secondary students. Hypothesis V I I I : Results Task 3 ( i n d i c a t i o n of t e x t u a l locus) i s a b e t t e r p r e d i c t o r than Task 4 (judgment of t r u t h value) of the c r i t e r i o n f o r comprehension, Task 5. Zero order c o r r e l a t i o n s f o r the three measures of t e x t compre-hension are included i n Tables 24a, b, c. The r e s u l t s of the a s s o c i a t e d m u l t i p l e r e g r e s s i o n analyses are summarized i n Tables 26a, b, c. Hypothesis V I I I : Conclusions The hypothesis that Task 3 b e t t e r p r e d i c t s Task 5 scores than does Task 4 i s supported by the data from the n a t i v e speaker and the m u l t i -e t h nic E.S.L. samples, having i n c u r r e d contrary r e s u l t s f o r the Chinese speaking E.S.L. group. M u l t i p l e r e g r e s s i o n analyses i d e n t i f y Task 3 as l e a d i n g Task 4 i n c o n t r i b u t i n g to Task 5 scores f o r the n a t i v e speaker and m u l t i e t h n i c E.S.L. samples w h i l e Task 4 e x p l a i n s more variance than Task 3 on the c r i t e r i o n measure f o r the Chinese speaking E.S.L. group. 119 Type I Type I I Task 3 + (Task 4) Task 4 + (Task 3) 95.6% + (1.8%) 84.4% + (13.0%) T o t a l variance explained = 97.4% Table 26a: M u l t i p l e Regression Analyses: independent variance and ( a d d i t i o n a l variance) c o n t r i b u t e d by component measures to c r i t e r i o n Task 5 scores of Native Speakers (N = 31, p4.01) Type I Type I I Task 4 + (Task 3) Task 3 + (Task 4) 85.6% + (3.2%) 71.4% + (17.4%) T o t a l variance explained = 88.8% Table 26b: M u l t i p l e Regression Analyses: independent v a r i a n c e and ( a d d i t i o n a l variance) c o n t r i b u t e d by component measures to c r i t e r i o n Task 5 scores of E.S.L. (A) Students (N = 44, p 4 . 0 l ) Type I Type I I Task 3 + (Task 4) Task 4 + (Task 3) 94.0% + (3.8%) 84.2% + (13.6%) T o t a l variance explained = 97.8% Table 26c: M u l t i p l e Regression Analyses: independent v a r i a n c e and ( a d d i t i o n a l variance) c o n t r i b u t e d by component measures to c r i t e r i o n Task 5 scores of E.S.L. (B) Students (N = 18, p<.01) 120 Hypothesis VIII : Discussion Considering the cited inadequacies of Task 4 as a measure of comprehension, the high correlations obtained with the more rigorous Task 5 were unexpected. Despite the reservations raised earlier, the empirical results of this investigation find Task 4 a reasonably accurate predictor of what has been argued to be a conceptually adequate measure of sentence comprehension within a text. This outcome may be attributed to the selection c r i t e r i a which precluded students observed to have d i f f i c u l t i e s in reading comprehension. Differences between Task 3 and Task 4 correlations with the criterion measure are probably more lik e l y to occur among samples of less able students more prone to random response strategies. While the results of the multiple regression analyses for two of the three groups would seem to offer evidence that Task 3 is far more conse-quential to Task 5 scores than is Task 4, an anticipated outcome, additional analyses revealed that most of the variance attributable to either Task. 3 or Task 4 is in a l l probability common variance shared between these two variables. Since the correlations for both Task 3 and Task 4 with the criterion Task 5 f a l l within a narrow range, the question arises as to which of the two component tasks is a better measure of text comprehension. This is directed to a practical preference for a single measure that does not require the combining of component subscores. For design of a test of l i t e r a l comprehension to be administered to a sampling of students from the three language populations represented in this study, a certain choice is not indicated by the present data. Choosing to use a measure of indicating textual locus only would seem to offer the advantage of greater ease and f l e x i b i l i t y in constructing items; one merely 121 s e l e c t s a sentence or two from the passage and a p p l i e s the transformations or introduces the anaphoric s t r u c t u r e on which the students are to be t e s t e d . R e l a t i v e l y l i t t l e concern need be given to the p l a u s i b i l i t y of the item statement. The disadvantages of t h i s format l i e i n i t s n o v e l t y to the examinees and some p o s s i b l e v a r i a t i o n i n c r i t e r i a f o r s c o r i n g . By c o n t r a s t the more f a m i l i a r measure of judgment of t r u t h value only, although e a s i e r to score i n a very c o n s i s t e n t manner, u s u a l l y r e q u i r e s more time to c o n s t r u c t s i n c e item statements c o n t a i n i n g p l a u s i b l e but f a l s e p r o p o s i t i o n s must be devised. In formats c a l l i n g f o r a:'cannot say" o p t i o n , items must be included f o r which i t i s d i f f i c u l t , i f not i m p o s s i b l e , to s t a t e what aspect of language competence i s being t e s t e d . P r e c i s e l y because of t h i s l a s t c o n s i d e r a t i o n , i t i s here proposed that s i n c e e m p i r i c a l f i n d i n g s o f f e r no c l e a r i n d i c a t i o n , conceptual c o n s i d -e r a t i o n s should govern the d e c i s i o n as to which component task, locus or t r u t h v a l u e , to adopt as a s i n g l e measure of sentence comprehension w i t h i n a t e x t . In t h i s l i g h t , the argument that knowledge of s p e c i f i c l i n g u i s t i c v a r i a b l e s might be more r e a d i l y a t t r i b u t a b l e to responses c a l l i n g f o r i n d i -c a t i o n of t e x t u a l locus than to those based s o l e l y on a judgment of t r u t h value remains unrefuted. Supplementary a n a l y s i s of the data Throughout the a n a l y s i s of the data i n the present study, two i n d i c a t o r s of concurrent v a l i d i t y were c o n t i n u a l l y monitored: (1) s u p e r i o r performance of n a t i v e speakers i n comparison to E.S.L. students, and (2) f o r n a t i v e speakers, augmented performance as a f u n c t i o n of higher grade i n t e r -v a l . Beyond such perfunctory checks, i t i s necessary to examine the p o s s i -b i l i t y that i n a p p r o p r i a t e response s t r a t e g i e s may have determined performance 122 before concluding that scores r e f l e c t competence i n the l i n g u i s t i c v a r i a b l e s under i n v e s t i g a t i o n . Task 1 Three s t r a t e g i e s are considered: (1) random responses, (2) responses based on a s i m i l a r i t y of sentence l e n g t h , and (3) responses based on s i m i l a r i t y of surface s t r u c t u r e . Random responses occur when the S^  can n e i t h e r i n t e r p r e t one or both sentences of a p a i r nor e s t a b l i s h any p l a u s i b l e device to suggest whether or not two sentences may be se m a n t i c a l l y e q u i v a l e n t . In t h i s i n s t a n c e , a may c o n s i s t e n t l y respond e i t h e r "yes" or "no" to a l l items or he may set a p a t t e r n of responses -— e.g., yes, yes, no, yes, yes, no... In preparing the t e s t booklets f o r mechanical s c o r i n g , no such patterns were observed. Further, since the key f o r Tests No. 1 and No. 2 c a l l s f o r 23 "yes" and 25 "no" responses, mean scores r e s u l t i n g from t o t a l l y random response choices should approach approximately 24.00. A l l 107 jSs p a r t i c i -p a t i n g i n the study rec e i v e d scores higher than 24. The lowest mean score among the nine subsamples i s 36.74. P r o b a b i l i t y of a t t a i n i n g t h i s high a score on the b a s i s of random response choices i s low (p < . 125) . S i m i l a r i t y of sentence length might conceivably i n f l u e n c e a S_'s d e c i s i o n as to whether or not two sentences bear the same und e r l y i n g meaning. D e f i n i n g " s i m i l a r l e n g t h " as the instance wherein the two sentences of a p a i r do not d i f f e r by more than f i v e t y p e w r i t t e n spaces i n o v e r a l l l e n g t h , the 48 items of Tests No. 1 and No. 2 were d i v i d e d i n t o two c a t e g o r i e s . In the f i r s t category were placed sentence p a i r s whose s i m i l a r i t y of graphic length might p o s s i b l y be a clue to s i m i l a r i t y of meaning. Included were sentence p a i r s of s i m i l a r graphic length that do have the same meaning and sentence p a i r s of d i f f e r e n t graphic length that do not have the 123 same meaning. These two types form .a set of 25 items: Test No. 1 — 4, 7, 8, 11, 12, 15, 16, 17, 18, 19, 23, 24, 27, 28, 29, 30, 31, 32, 33, 35; Test No. 2 — 2, 7, 8, 9, 12. The remaining 23 items comprise the second category i n which r e l i a n c e upon s i m i l a r i t y and d i f f e r e n c e i n graphic l e n g t h would lead to an i n c o r r e c t response. f i r s t category i s su p e r i o r to that f o r the second category. The d i f f e r e n c e i s s i g n i f i c a n t (p«s£.025) f o r Native Speakers: Senior Secondary and a l l grade i n t e r v a l s of the E.S.L.(A) sample, thereby suggesting the p o s s i b i l i t y t h a t some Ss may have r e l i e d upon graphic length clues to determine t h e i r responses r a t h e r than any syntactic-semantic i n t e r p r e t a t i o n of the sentences. of items thought to o f f e r some cl u e to s i m i l a r i t y or d i f f e r e n c e of meaning, however, y i e l d s a p a t t e r n of r e s u l t s very s i m i l a r to the f i n d i n g s reported f o r Tests No. 1 and No. 2 as a whole (Table 27). While s i m i l a r i t y of graphic length between sentences may be a f a c t o r i n enhancing performance, i t i s concluded not to be a c r u c i a l v a r i a b l e c o n t r i b u t i n g to d i f f e r e n c e s i n mean scores among the subsamples. I t i s most probable that responses to items i n Tests No. 1 and No. 2 are g e n e r a l l y based on some a t t r i b u t e other than v i s u a l d i s c r i m i n a t i o n of sentence l e n g t h . A c o r r e l a t e d t - t e s t revealed that performance on items i n the A comparison of the performance of the nine subsamples on the set Elem JrSec SrSec N.S. 83.3% (81.6%) 91.5% (90.0%) 94.8% (93.4%) E.S.L.(A) 84.4% (78.1%) 79.0% (76.5%) 86.7% (83.7%) E.S.L.(B) 80.0% (78.8%) 80.0% (76.8%) 88.7% (88.2%) Table 27: P r o p o r t i o n of c o r r e c t responses f o r graphic length c l u e items and ( a l l items) by Language group and Grade i n t e r v a l 124 One purpose of the present study i s to assess Ss' p r o f i c i e n c y i n recovering deep s t r u c t u r e through an accurate r e c o g n i t i o n of the f u n c t i o n of s y n t a c t i c elements i n a sentence. I t i s , t h e r e f o r e , important i n estab-l i s h i n g the v a l i d i t y of Tests No. 1 and No. 2 as a measure of r e c o g n i t i o n of grammatical i m p l i c a t i o n that i t be demonstrated that performance on the t e s t s cannot be a t t r i b u t e d to S_s' st r a t e g y of r e l y i n g s o l e l y upon surface s t r u c t u r e patterns to a t t a i n a c o r r e c t response. I t i s conceivable that some Ss might examine a p a i r of sentences to determine i f content lexemes — nouns, verbs, a d j e c t i v e s , and adverbs — occur i n the same s e q u e n t i a l order i n both sentences. I f so, the sentences are judged to be se m a n t i c a l l y e q u i v a l e n t ; otherwise, the S^  concludes that the two sentences d i f f e r i n meaning. For example, The g i r l i s easy to see. The g i r l i s e a s i l y seen. are deemed to be paraphrases of each other. Another s t r a t e g y based on the surface s t r u c t u r e of the sentences would consider the a d d i t i o n of elements i n one sentence to preclude i t s being a paraphrase of another. Applying t h i s p r i n c i p l e , Mary t o l d Jack to come here today. Mary t o l d Jack that she should come here today. are judged to be se m a n t i c a l l y not e q u i v a l e n t . To determine the prevalence of these two s t r a t e g i e s , i t i s p o s s i b l e to d i v i d e the 36 items of Test No. 1 i n t o two groups based on whether or not s i m i l a r i t i e s and d i f f e r e n c e s i n surface s t r u c t u r e provide s u p e r f i c i a l clues to deep s t r u c t u r e i d e n t i t y between sentences. The f o l l o w i n g 21 items are thought to provide such c l u e s : 3, 4, 5, 7, 11, 12, 15, 16, 18, 20, 21, 22, 23, 24, 26, 27, 29, 30, 32, 33, 35. 125 A c o r r e l a t e d t - t e s t revealed that on Native Speakers: Senior Secondary a c t u a l l y performed s i g n i f i c a n t l y b e t t e r on those items c o n t a i n i n g s u p e r f i c i a l c l u e s . This i s the one sample whose n a t i v e language and grade i n t e r v a l f a c t o r s would i n d i c a t e the l e a s t dependency on gleaning surface s t r u c t u r e c l u e s . A s m a l l d i f f e r e n c e f a v ouring the s u p e r f i c i a l clue items f o r the E.S.L.(A): Senior Secondary subsample i s not s i g n i f i c a n t . Moreover, the p a t t e r n of r e s u l t s i s very s i m i l a r f o r both the " s u p e r f i c i a l c l u e s " only and the e n t i r e set of items comprising Test No. 1 (Table 28). I t i s to be noted that a l l but the aforementioned subsample of E.S.L. students performed b e t t e r , i n some cases s i g n i f i c a n t l y so, on that set of items b e l i e v e d not to c o n t a i n surface s t r u c t u r e c lues to u n d e r l y i n g meaning. I t may, t h e r e f o r e , be concluded that the r e s u l t s .of t h i s phase of the i n v e s t i g a t i o n do r e f l e c t the comparative competencies of the nine subsamples i n recovering the deep s t r u c t u r e of item sentences. Elem JrSec SrSec N.S. 80.6% 89.5% 94.5% (78.9%) (88.9%) (93.1%) E.S.L.(A) 73.8% 73.3% 82.9% (74.4%) (74.9%) (82.2%) E.S.L.(B) 73.8% 72.8% 84.1% (76.8%) (73.0%) (86.6%) Table 28: P r o p o r t i o n of c o r r e c t responses f o r " s u p e r f i c i a l c l u e " items and ( a l l items) by Language group and Grade i n t e r v a l Task 2 The "constructed rsponse" format of Test No. 3 g e n e r a l l y precludes a p a t t e r n i n g of responses based upon f a l s e s t r a t e g i e s . Since the c o r r e c t anaphoric r e f e r e n t s occurred i n v a r i o u s p o s i t i o n s w i t h i n item sentences, 126 the use of p o s i t i o n a l c lues could not be expected to augment scores. I n t e r n a l r e l i a b i l i t y estimates of Test No. 3 and i t s subtests are r a t h e r high f o r most subsamples, f u r t h e r suggesting the p o s s i b i l i t y of a v a l i d measure of anaphoric r e s o l u t i o n . Tasks 3, 4, and 5 These ta s k s , embodied i n Test No. 4, are intended to t e s t sentence comprehension w i t h i n a continuous t e x t . I t w i l l be r e c a l l e d that Task 3 r e q u i r e s the £ to i d e n t i f y the sentence(s) i n the t e x t that provide the b a s i s f o r a " t r u e " or " f a l s e " response to the item statement. The judgment of such t r u t h value c o n s t i t u t e s Task 4. To minimize the e f f e c t s of random responses which may have y i e l d e d s p u r i o u s l y high scores on Task 4, Ss were presented w i t h an a d d i t i o n a l response o p t i o n f o r each item — "cannot say". Of the twenty-three items, s i x were keyed "cannot say". In these i n s t a n c e s , no t r a n s -formation of the item sentence i s contained i n the t e x t . Given the nature of Task 1 ( r e c o g n i t i o n of grammatical i m p l i -c a t ion) and Task 2 ( i d e n t i f i c a t i o n of anaphoric r e f e r e n t s ) , i t would seem reasonable that these two tasks might be b e t t e r p r e d i c t o r s of the c r i t e r i o n Task 5 and i t s components, Tasks 3 and 4, when only those items i n v o l v i n g the v a r i a b l e s defined f o r Tasks 1 and 2 are taken i n t o account. An . increased p r o p o r t i o n of v a r i a n c e i n Task 5 scores explained by Tasks 1 and 2 may a l s o be expected. To v e r i f y these p r o p o s i t i o n s , zero order c o r r e l a t i o n s were estab-l i s h e d and m u l t i p l e r e g r e s s s i o n analyses were conducted based on the group of seventeen items keyed e i t h e r " t r u e " or " f a l s e " which remained a f t e r the s i x items keyed "cannot say" were removed from the s t a t i s t i c a l a n a l y s i s . 127 Task 3 Task 4 Task 5 Task 1 .758 .666 .750 (.717) (.663) (.706) Task 2 .737 .644 .754 (.711) (.665) (.720) Table 29a: C o r r e l a t i o n c o e f f i c i e n t s f o r 17 t r u e / f a l s e items and ( a l l items) f o r Native Speakers completing a l l tasks (N = 39, p 4.01) Task 3 Task 4 Task 5 Task 1 .334 .452 .454 (.275) (.417) (.390) Task 2 .404 .417 .500 (.380) (.598) (.625) .Table 29b: C o r r e l a t i o n c o e f f i c i e n t s f o r 17 t r u e / f a l s e items and ( a l l items) f o r E.S.L.(A) Students completing a l l tasks (N = 44, P4.01) Task 3 Task 4 Task 5 Task 1 .427 .520 .401 (.436) (.540) (.451) Task 2 .359 .296 .226 (.255) (.201) (.189) Table 29c: C o r r e l a t i o n c o e f f i c i e n t s f o r 17 t r u e / f a l s e items and ( a l l items) f o r E.S.L.(B) Students completing a l l tasks (N = 18, p4.01) 128 A comparison of these r e v i s e d analyses w i t h the o r i g i n a l examination of the data which included a l l twenty-three items (Tables 29a, b, c) i n d i c a t e s that performance on Tasks 1 and 2 d e f i n i t e l y tends to be a b e t t e r p r e d i c t o r of i n d i c a t i o n of t e x t u a l locus (Task 3) on a t e s t that contains only items whose t r u t h value can be determined d i r e c t l y from the t e x t than to a t e s t that contains at l e a s t some items whose t r u t h value cannot be determined from the accompanying passage. This observation lends f u r t h e r support to the p s y c h o l o g i c a l r e a l i t y of c e r t a i n aspects of the model f o r comprehending sentences i n a t e x t as proposed e a r l i e r i n t h i s r e p o r t (q.v., Appendix "A"). The e f f e c t of removing the indeterminate "cannot say" items from Test No. 4, however, i s e r r a t i c w i t h respect to p r e d i c t i n g judgment of t r u t h v a l u e , i n d i c a t i n g that Task 4 i s not as s y s t e m a t i c a l l y r e l a t e d to the l i n g u i s t i c v a r i a b l e s which define Tasks 1 and 2 as i s Task 3. A f i n a l c o n s i d e r a t i o n i s the e f f e c t the removal of the i n d e t e r -minate items from Test No. 4 has upon the amount of v a r i a n c e Tasks 1 and 2 c o n t r i b u t e to the c r i t e r i o n measure. For n a t i v e speakers, whose performance on Task 5 can be accounted to a greater extent by Tasks 1 and 2, the expected increase i n variance explained by s a i d tasks i s r e a l i z e d (Table 30). In the case of E.S.L. students, f o r whom i t has already been demonstrated that Tasks 1 and 2 c o n t r i b u t e notably l e s s v a r i a n c e to scores on the f u l l Task 5, i t can be seen that even l e s s v a r i a n c e i s explained by the two tasks on the abbreviated and more r i g o r o u s l y defined v e r s i o n of the c r i t e r i o n measure. This may w e l l be a f u r t h e r i n d i c a t i o n that w h i l e n a t i v e speakers u t i l i z e t h e i r p r o f i c i e n c y i n r e c o g n i z i n g grammatical i m p l i c a t i o n and anaphoric r e l a t i o n s i n responding to items i n the format s e l e c t e d f o r Test No. 4, E.S.L. students r e l y l e s s upon such c a p a b i l i t i e s , employing other comprehension s t r a t e g i e s to a t t a i n comparable scores on the c r i t e r i o n Task 5. N.S. E.S.L.(A) E.S.L.(B) 70.4%*" 30.9%* 16.2%* (63.3%)* (40.1%)* (20.4%)* Table 30: Combined variance c o n t r i b u t e d by Tasks 1 and 2 to c r i t e r i o n Task 5 scores comprising 17 t r u e / f a l s e items and ( a l l items) by Language group (*p<.01, ** not s i g . ) 130 Chapter V II SUMMARY AND CONCLUSIONS The major f i n d i n g s of the present study may be summarized: There i s a trend toward augmented performance on a l l tasks as a f u n c t i o n of higher grade i n t e r v a l . (The l i t e r a t u r e reviewed i n Chapter I I c i t e s evidence f o r developmental trends i n s y n t a c t i c comprehension.) Native speakers as a whole tend to s i g n i f i c a n t l y outperform E.S.L. students only on Task 1 ( r e c o g n i t i o n of grammatical i m p l i c a t i o n ) . A n a l y s i s by grade i n t e r v a l s i n d i c a t e s that s i g n i f i c a n t d i f f e r e n c e s i n Task 1 scores occur only a t the secondary i n t e r v a l s . No one subsample, even at the highest grade i n t e r v a l , demonstrates complete mastery of a l l transformations and anaphoric types t e s t e d . ( E a r l i e r i n v e s t i g a t i o n s have discovered that school aged c h i l d r e n are unable to comprehend many s y n t a c t i c s t r u c t u r e s . ) No s i g n i f i c a n t d i f f e r e n c e occurs at any grade i n t e r v a l between n a t i v e speakers and E.S.L. students on the c r i t e r i o n measure of t e x t compre-hension or i t s component tasks. The rank order of d i f f i c u l t y of the grammatical i m p l i c a t i o n and anaphora subtasks d i f f e r s f o r each subsample. The performance of the m u l t i e t h n i c E.S.L. students more c l o s e l y approximates the p a t t e r n of d i f f i c u l t y experienced by n a t i v e speakers than does that of the Chinese E.S.L. l e a r n e r s , suggesting that n a t i v e language may i n f l u e n c e the order of s y n t a c t i c d i f f i c u l t y f o r the second language l e a r n e r . 131 The rank order of d i f f i c u l t y v a r i e s across grade i n t e r v a l s moreso f o r n a t i v e speakers than f o r E.S.L. students. The rankings f o r elementary p u p i l s across language groups i s more c o n s i s t e n t than f o r students at the secondary i n t e r v a l s . (These f i n d i n g s a l l confirm the e s t a b l i s h e d p r i n c i p l e that some s y n t a c t i c patterns are e a s i e r to comprehend than others although one p o p u l a t i o n may d i f f e r from another i n t h i s respect.) In some in s t a n c e s , rankings appear to be a f f e c t e d by t r a n s f o r m a t i o n a l s i m p l i c i t y . (Several researchers have claimed s y n t a c t i c complexity to be a c r u c i a l f a c t o r i n reading comprehension and r e a d a b i l i t y . ) Regarding the components of t e x t comprehension, n a t i v e speakers f i n d the task of judging the t r u t h value of item statements to be s i g n i f i -c a n t l y e a s i e r than i d e n t i f y i n g t h e i r locus i n the t e x t . For E.S.L. students these two tasks appear to be e q u a l l y d i f f i c u l t . A l l language groups i n c u r s i g n i f i c a n t l y lower mean scores on the c r i t e r i o n measure than on e i t h e r of i t s component tasks. The measures of s y n t a c t i c comprehension are b e t t e r p r e d i c t o r s of the c r i t e r i o n f o r t e x t comprehension and i t s component tasks f o r n a t i v e speakers than f o r E.S.L. students. (The l i t e r a t u r e review and other d i s c u s s i o n r e f e r to a number of previous s t u d i e s which have shown that s y n t a c t i c comprehension c o n t r i b u t e s to more ge n e r a l i z e d measures of reading comprehension to v a r y i n g degrees depending on the sample population.) I n d i c a t i o n of t e x t u a l locus accounts f o r a greater p r o p o r t i o n of the variance i n c r i t e r i o n scores than does judgment of t r u t h value i n the 132 case of n a t i v e speakers and the m u l t i e t h n i c E.S.L. students. The converse p r e v a i l s f o r Chinese speakers. Foremost among the present f i n d i n g s , however, i s the evidence that n a t i v e speakers and second language l e a r n e r s d i f f e r i n the s t r a t e g i e s used to a t t a i n equal p r o f i c i e n c y on each of the component tasks of t e x t comprehension. S p e c i f i c d i f f e r e n c e s can be r e l a t e d to a group's demon-s t r a t e d l i n g u i s t i c knowledge. Those who possess one or another s k i l l of s y n t a c t i c comprehension draw upon that resource moreso than those who l a c k that s k i l l . Hence, more than twice as much vari a n c e .in. the t e x t compre-hension c r i t e r i o n scores of n a t i v e speakers can be accounted f o r by r e f e r -ence to Task 1 performance than can be explained f o r e i t h e r of the E.S.L. samples. Conversely, more of the variance i n the Chinese speaking E.S.L. group's c r i t e r i o n scores i s explained by i t s performance on Task 2 which exceeds that of e i t h e r of the other language groups. Results of t h i s nature serve to f u r t h e r e s t a b l i s h the p r i n c i p l e of v a r i a t i o n s i n the psycho-l i n g u i s t i c processes through which the content of discourse i s comprehended. D i r e c t i o n s f o r f u r t h e r research Because the present study i s of an e x p l o r a t o r y nature, s e l e c t i o n of t r a n s f o r m a t i o n a l equivalents and anaphoric forms, w h i l e not e n t i r e l y random, was, to some extent, e c l e c t i c . In reviewing other research i n s t r u -ments, comment was made on the degree to which each appeared to be motivated by a system of i d e n t i f y i n g and c a t e g o r i z i n g s y n t a c t i c s t r u c t u r e s . The pau c i t y of s i g n i f i c a n t d i f f e r e n c e s between n a t i v e and non-native speakers among the present subtasks suggests that an extended i n v e s t i g a t i o n might be 133 f a c i l i t a t e d by f i r s t a t t e m p t i n g to b u i l d a c o m p r e h e n s i v e taxonomy o f a p p r o p r i a t e s y n t a c t i c r e l a t i o n s w i t h i n a h i e r a r c h i c a l framework i n i t i a l l y c o n c e i v e d upon c o n c e p t u a l n o t i o n s o f t r a n s f o r m a t i o n a l c o m p l e x i t y . T h r o u g h a p p l i e d r e s e a r c h , the p r o p o s e d schema can be r e v i s e d i n s u c h a manner as to ex tend beyond a mere c a t e g o r i z e d i n v e n t o r y o f E n g l i s h s y n t a c t i c p a t t e r n s to d e r i v e an e m p i r i c a l l y d e v e l o p e d s c h e d u l e o f a c q u i s t i o n d i f f i c u l t y . Once a sequence i s c l e a r l y e s t a b l i s h e d f o r n a t i v e s p e a k e r s , i t becomes p r a c t i c a l to commence a s y s t e m a t i c s e a r c h f o r s y n t a c t i c c o m p r e h e n s i o n t a s k s w h i c h d i f f e r e n t i a t e s t u d e n t s a t advanced s t a g e s o f l e a r n i n g E n g l i s h as a s econd l a n g u a g e f r o m t h o s e o f s i m i l a r g r a d e p l a c e m e n t who p o s s e s s n a t i v e l i n g u i s t i c competence . The l a t t e r p u r s u i t may be u s e f u l on two c o u n t s : (1) i t c o u l d l e a d to more v a l i d t e s t s . o f E n g l i s h l a n g u a g e competence t h a n are . p r e s e n t l y a v a i l a b l e , and (2) h a v i n g d e f i n e d the d e f i c i e n c i e s i n s y n t a c t i c u n d e r -s t a n d i n g o f the s e c o n d l a n g u a g e s t u d e n t , a b a s i s m i g h t be p r o v i d e d f o r the d e s i g n o f c o m p e n s a t o r y i n s t r u c t i o n . The b r o a d e r p r o b l e m r e m a i n s , however , o f e s t a b l i s h i n g the r e l a -t i o n s h i p between s y n t a c t i c t a s k s i n i s o l a t i o n and t h o s e i n a l a r g e r c o n t e x t . The p r e s e n t s t u d y i s b a s e d upon a s p e c i f i c mode l f o r s e n t e n c e c o m p r e h e n s i o n w i t h i n a t e x t w h i c h demands t h a t each t e s t i t e m have a s y n t a c t i c r e l a t i o n to one o r more s e n t e n c e s i n t h e accompany ing p a s s a g e . I t has been a r g u e d h e r e and e l s e w h e r e t h a t an adequate t e s t o f c o m p r e h e n s i o n must be a b l e to demon-s t r a t e w h i c h a s p e c t s o f l a n g u a g e competence a r e b e i n g t e s t e d by e a c h i t e m . S t u d i e s o f p a s s a g e dependency s t r o n g l y s u g g e s t t h a t c o n v e n t i o n a l i n s t r u m e n t s , even t h o s e f o r . w h i c h e x t e n s i v e norms have been d e v e l o p e d , f a i l to meet t h i s c r i t e r i o n . T h e r e f o r e any a t t e m p t to c o r r e l a t e measures o f s y n t a c t i c compre -h e n s i o n d e s i g n e d to r e f l e c t u n d e r s t a n d i n g o f s e l e c t e d a s p e c t s o f l a n g u a g e competence w i t h w h a t , by c o m p a r i s o n , a r e l o o s e l y c o n s t r u c t e d i n v e n t o r i e s o f 134 "reading comprehension s k i l l s " i s not l i k e l y to r e v e a l the f u l l s i g n i -f i c a n c e of one's knowledge of syntax i n comprehending continuous d i s c o u r s e . Research l i n k i n g s y n t a c t i c comprehension as examined i n t h i s study w i t h performance on popular standardized t e s t s i s , i n the o p i n i o n of t h i s i n v e s t i g a t o r and others, predicated on f a u l t y t e s t i n g p r a c t i c e s and there-f o r e not worth the e f f o r t of c a r e f u l and d e t a i l e d study. I t may be of some i n t e r e s t , however, to determine how tasks of grammatical i m p l i c a t i o n and anaphoric r e s o l u t i o n r e l a t e to a c l o z e t e s t appropriate to the student's l e v e l of language competence or, i n the case of the second language l e a r n e r , a t a r g e t . l e v e l of competence. On the b a s i s of the present r e s u l t s , i t would seem to be a productive course, from the standpoint of assaying the importance of s y n t a c t i c knowledge, to f u r t h e r explore the r e l a t i v e c o n t r i b u t i o n of locus i n d i c a t i o n and t r u t h value judgment to the c r i t e r i o n measure of t e x t compre-hension submitted by t h i s study. Research employing l a r g e r samples repre-s e n t a t i v e of a number of p o p u l a t i o n s , p a r t i c u l a r l y those considered to be below average i n language comprehension as measured by conventional means, and using c a r e f u l l y constructed items based on a wide v a r i e t y of s y n t a c t i c v a r i a b l e s may b e t t e r evaluate the p o t e n t i a l of the locus i d e n t i f i c a t i o n component and c l a r i f y a s t i l l ambiguous r e l a t i o n s h i p . Recommendations f o r t e s t i n g A number of concerns a r i s e from the present study. The need f o r a comprehensive taxonomy of transformations and anaphoric forms has been mentioned i n connection w i t h d i r e c t i n g motivated research. Testing that i s motivated w i l l a l s o seek to s e l e c t from an e s t a b l i s h e d taxonomy those gram-m a t i c a l i m p l i c a t i o n r e l a t i o n s appropriate to the requirements of a p a r t i c u l a r 135 assessment, i . e . , l e v e l of s y n t a c t i c complexity, f o r m a l i t y of usage, subject of d i s c o u r s e , e t c . Once a taxonomy becomes f a m i l i a r to those engaged i n the e v a l u a t i o n of language l e a r n i n g , i t becomes p o s s i b l e f o r a p u b l i s h e r to describe i n meaningful l i n g u i s t i c terms what areas of the s y n t a c t i c system of E n g l i s h are examined by s p e c i f i c t e s t s thereby enabling an evaluator to choose the instrument most appropriate to the s i t u a t i o n and i n t e r p r e t the r e s u l t s i n terms of s p e c i f i c d e f i c i e n c i e s i n the examinee's language competence. In c o n s t r u c t i n g t e s t s intended to measure s y n t a c t i c comprehension, two cautions must be kept i n mind. F i r s t i s a r e c o g n i t i o n of the i n t e r -a c t i o n of syntax and semantics. T y p i c a l of recent experimental i n v e s t i -gations i n p s y c h o l i n g u i s t i c s i s that of H e r r i o t (1969) which found that the time r e q u i r e d to i n t e r p r e t a s y n t a c t i c r e l a t i o n ( i . e . , i d e n t i f y the subject of a clause) i s a f f e c t e d by the absence of semantic c l u e s . S i m i l a r l y , O'Donnell (1976) has commented on the inappropriateness of using nonsense vocabulary i n an attempt to gain a 'pure' measure of s y n t a c t i c comprehension. I t would seem t h a t , s i n c e s y n t a c t i c comprehension cannot occur unless one i s able to recover the deep s t r u c t u r e of a sentence, such recovery i s h a l t e d by the i n a b i l i t y to r e a d i l y a s s i g n a l e x i c a l meaning to u n f a m i l i a r formatives. One i s t h e r e f o r e w e l l advised to devise only item statements that bear p l a u s i b l e p r o p o s i t i o n s . Care was taken i n the p r e p a r a t i o n of items f o r the present instrument to avoid complications introduced by semantic f a c t o r s . The second c o n s i d e r a t i o n i s the importance of t e s t format as an i n t e r v e n i n g v a r i a b l e . Examples of c o n f l i c t i n g r e s u l t s have been c i t e d between m u l t i p l e choice and constructed responses i n connection w i t h i n v e s t i -gations i n t o the comparative d i f f i c u l t y of anaphora forms (Bormuth et a l . , 1970 vs. Lesgold, 1973) and between m u l t i p l e choice and c l o z e e x e r c i s e s i n 136 e v a l u a t i n g the e f f e c t of t r a n s f o r m a t i o n a l s i m p l i c i t y on comprehension ( P e l t z , 1974). This suggests the need f o r c a u t i o n i n comparing performances on t e s t s using d i f f e r e n t response formats. Admittedly such a problem a r i s e s i n i n t e r p r e t i n g c e r t a i n r e s u l t s of the present study. While comparisons across subsamples on the tasks of s y n t a c t i c comprehension may be v a l i d , i t i s debatable whether the c l u s t e r of low rankings f o r the anaphora subtasks i s a r e f l e c t i o n of a c t u a l comparative d i f f i c u l t y or a consequence of the v a r i a t i o n between a m u l t i p l e choice format..in Task 1 and constructed responses i n Task 2.~* Toward improving present p r a c t i c e , the task of i n d i c a t i n g a t e x t u a l locus f o r an item statement presents a new approach to comprehension t e s t i n g , one that i s based on observable l i n g u i s t i c f eatures and, on the b a s i s of the present evidence, one that i s as v a l i d as the conventional p r a c t i c e of asking the reader to make a judgment as to the t r u t h value of a statement. While the need remains to f u r t h e r v a l i d a t e t h i s task through a d d i t i o n a l research and a n a l y s i s , i t s adoption by t e s t c o n s t r u c t o r s and classroom teachers enables w r i t i n g items to t e s t r e c o g n i t i o n of p o s s i b l e grammatical i m p l i c a t i o n r e l a t i o n s between any two sets of one or more sentences. A l l that i s re q u i r e d i s to choose a sentence, or sentences, i n a passage on which to perform a s e l e c t e d transformation or anaphorization. The sentence r e s u l t i n g from the operation becomes the item statement. This approach e l i m i n a t e s the need to devise p l a u s i b l e f a l s e or indeterminate staements to balance the d i s t r i b u t i o n of keyed responses. I f the arrange-ment f o r numbering sentences i n the passage i s followed as; i n Test No. 4 of 5. C o r r e l a t e d t - t e s t s between Task 1 and Task 2 conducted f o r each language group produced the f o l l o w i n g values: N.S.: t value = 3.96, p <£.00l; E.S.L. (A): t value = 2.42, p<.025; E.S.L. (B) : t value = 4.53, p<.001. 137 the present instrument, items can be keyed f o r r a p i d s c o r i n g . With p r a c t i c e and c a u t i o n i n a s s u r i n g that the keyed answers i n c l u d e a l l p o s s i b l e c o r r e c t a l t e r n a t i v e s f o r the a s s o c i a t e d transformations of a p a r t i c u l a r item statement, the language teacher w i l l acquire a measure that i s l i k e l y to be both a v a l i d i n d i c a t o r of achievement i n E n g l i s h language l e a r n i n g and d i a g n o s t i c of s p e c i f i c i n s t r u c t i o n a l needs i n order to lead the student to n a t i v e - l i k e competence. I m p l i c a t i o n s f o r i n s t r u c t i o n The present study i s designed to i d e n t i f y some s p e c i f i c d i f f e r -ences i n language competence among i n s t r u c t i o n a l populations and to deter-mine the extent to which the p a r t i c u l a r aspects of competence t e s t e d p r e d i c t the a b i l i t y to l o c a t e i n a passage sentences r e l a t e d to a t e s t item and to judge the t r u t h value of the item statement. There i s no i n t e n t i o n here to p r e s c r i b e programme o b j e c t i v e s or teaching methods. In l i g h t of the present r e s u l t s , the second language teacher must c r i t i c a l l y assay the importance of focusing a t t e n t i o n upon t r a n s f o r m a t i o n a l equ i v a l e n t s and anaphoric r e l a t i o n s to i n s t r u c t i o n i n discourse compre-hension. E.S.L. teachers p a r t i c i p a t i n g i n t h i s study apparently prepare t h e i r students through present i n s t r u c t i o n a l methods to perform tasks of t e x t comprehension as adequately as n a t i v e speaker peers. These tasks i n c l u d e both f a m i l i a r o b j e c t i v e s (Task 4) and unconventional demands (Task 3). Examining a group of Grade 12 n a t i v e speakers, O'Donnell concludes that "the c o r r e l a t i o n between awareness of s t r u c t u r a l r e l a t i o n s h i p s of words i n sentences and a b i l i t y i n reading comprehension .44 . . . i s not s u f f i x , c i e n t l y high to give c o n c l u s i v e evidence to support the teaching of l i n g u i s -t i c s t r u c t u r e as a major means of developing reading comprehension."(1963:316) 138 I f one a p p l i e s t h i s argument to developing s k i l l i n deep s t r u c t u r e recovery, the present data would suggest that such i n s t r u c t i o n would be appropriate only f o r n a t i v e speakers. Before accepting such a c o n c l u s i o n , however, one ought to consider the r e p o r t s of some classroom a p p l i c a t i o n s . Hughes (1975) claims success i n r a i s i n g comprehension l e v e l s of Grade 7 students, p a r t i c u l a r l y lower and middle a b i l i t y groups, through p r a c t i c e i n sentence combining. Implementing a c u r r i c u l u m to enhance comprehension of s y n t a c t i c s t r u c t u r e s , Stedman (1971) was able to show s i g n i f i c a n t gains i n a c l o z e performance f o r Grade 4 black students who may resemble some second language l e a r n e r s i n t h e i r u n f a m i l i a r i t y w i t h s y n t a c t i c forms prevalent i n standard w r i t t e n E n g l i s h . In other attempts, O'Donnell and King (1974) were unable to improve the reading comprehension of Grade 7 students ranking below the 20th p e r c e n t i l e on a standardized t e s t through focused i n s t r u c t i o n i n deep s t r u c t u r e recovery. They do concede, however, that "DSRT Simons, 1970 a b i l i t i e s s i g n i f i c a n t l y i n f l u e n c e reading comprehension t e s t a b i l i t i e s " (p. 337). I t i s suggested that t h i s s o r t of e f f o r t might be more productive w i t h l e s s d e b i l i t a t e d students who show more p o s i t i v e a t t i t u d e s toward a programme of remediation. Indeed, i n a subsequent study, O'Donnell and Smith (1975) by designing a set of programmed e x e r c i s e s succeeded i n i n c r e a s i n g awareness of s y n t a c t i c s t r u c t u r e as measured by the PAST (O'Donnell, 1973) f o r some, but not a l l , of a group of Grade 9 students w i t h i n four weeks of supplementary i n s t r u c t i o n . While i t i s most probable that i n s t r u c t i o n a l content and d e l i v e r y are c r u c i a l i n t e r v e n i n g v a r i a b l e s i n a l l programme s t u d i e s , some evidence e x i s t s to support the contention that deep s t r u c t u r e recovery s k i l l s as manifest i n r e c o g n i z i n g grammatical i m p l i c a t i o n r e l a t i o n s can i n f l u e n c e 139 t e x t comprehension and t h a t , f o r at l e a s t some po p u l a t i o n s , such s k i l l s can be s u c c e s s f u l l y taught. I f on t h i s b a s i s the E.S.L. teacher chooses to d i r e c t h i s students' a t t e n t i o n to r e c o g n i z i n g t r a n s f o r m a t i o n a l equivalents and anaphoric r e l a t i o n s , e i t h e r because any such s k i l l s t hat i d e n t i f y n a t i v e competence are worthy second language l e a r n i n g o b j e c t i v e s i n t h e i r own r i g h t or f o r what these a b i l i t i e s may c o n t r i b u t e to more gen e r a l i z e d measures of comprehension, he i s w e l l advised to be wary of using a model f o r t e s t i n g as a model f o r i n s t r u c t i o n . Tests such as the DSRT, PAST, and the f i r s t two tasks of the present instrument are designed to sample from a l a r g e inventory the s y n t a c t i c comprehension'of s e l e c t e d forms by examinees at a given p o i n t i n time. R e f e r r i n g to these t e s t s , one cannot expect to l i s t a l l r e l a t i o n s that should be included i n a programme; hence, the need to organize a a E.S.L. cu r r i c u l u m upon a comprehensive i n s t r u c t i o n a l taxonomy. Furth e r , i t does not n e c e s s a r i l y f o l l o w that p r a c t i c e on items adapted from the t e s t s w i l l over a p e r i o d of time produce the l i n g u i s t i c understandings sought. Rather, the p r a c t i c i o n e r ' s task i s to develop e f f e c t i v e techniques and m a t e r i a l s to provide l e s s o n content, p r e f e r a b l y i n a context of graded t e x t s and e x e r c i s e s which, u n l i k e present p u b l i c a t i o n s , are ordered i n t h e i r s e l e c t i o n of c l e a r l y designated s y n t a c t i c comprehension tasks. 140 BIBLIOGRAPHY Baines, Helen. An assessment and comparison of s y n t a c t i c complexity and word a s s o c i a t i o n s of good and poor readers i n Grades 4, 8, and 12. 1975. ERIC m i c r o f i c h e ED 119 156 Bormuth, John R. R e a d a b i l i t y — a new approach. Reading Research Q u a r t e r l y , 1966, 1_, 79 - 132. An o p e r a t i o n a l d e f i n i t i o n of comprehension i n s t r u c t i o n . In Goodman, Kenneth S. and James T. Fleming (eds.) P s y c h o l i n g u i s t i c s  and the teaching of reading. Newark, D e l . , I n t e r n a t i o n a l Reading A s s o c i a t i o n , 1969. pp. 48 - 60. On the theory of achievement t e s t items. Chicago, U n i v e r s i t y of Chicago Press, 1970 , John Manning, J u l i a n Carr, and David Pearson. Children's compre-hension of between- and w i t h i n - sentence s y n t a c t i c s t r u c t u r e s . J o u r n a l of Educational Psychology, 1970, 61_, 5, 349 - 357. B o t e l , Morton and A l v i n Granowsky. A formula f o r measuring s y n t a c t i c com-p l e x i t y : A d i r e c t i o n a l e f f o r t . Elementary E n g l i s h , 1972, 4j?, 4, 513 - 516. C a r r o l l , John B. Comprehension by 3rd, 6th, and 9th graders of words having m u l t i p l e grammatical f u n c t i o n s . F i n a l r e p o r t . 1970. ERIC micro-f i c h e ED 048 311 Chomsky, C a r o l . The a c q u i s i t i o n of syntax i n c h i l d r e n from 5 to 10. Cambridge, Mass., M. I . T. Press, 1969 Chomsky, Noam. Aspects of the theory of syntax. The Hague, Mouton, 1965 Coleman, E. B. The c o m p r e h e n s i b i l i t y of s e v e r a l grammatical transformations. J o u r n a l of A p p l i e d Psychology, 1964, 48, 186 -190. Co oper, Charles R. and Anthony R. Petrosky. A p s y c h o l i n g u i s t i c view of the f l u e n t reading process. J o u r n a l of Reading, 1976, 20, 3, 184 - 207. Cox, Judy R. A study of the s y n t a c t i c competence of a d u l t beginning readers. 1976. ERIC m i c r o f i c h e ED 120 713 Cromer, W. The d i f f e r e n c e model: A new e x p l a n a t i o n f o r some reading d i f f i -c u l t i e s . J o u r n a l of Educational Psychology, 1970, 6_1, 471 - 483. Dale, E. and J . S. C h a l l . A formula f o r p r e d i c t i n g r e a d a b i l i t y . E d u cational  Research B u l l e t i n , 1948, 2_7, 11 - 20, 37 - 54. 141 Dunn, O l i v e J . M u l t i p l e comparisons using rank sums. Technometrics, 1964, 6, 3, 241 - 252. E p s t e i n , W i l l i a m . The i n f l u e n c e of s y n t a c t i c s t r u c t u r e on l e a r n i n g . American Journa l of Psychology, 1961, 7_4_, 80 - 85. Evans, Ronald V. The e f f e c t of t r a n s f o r m a t i o n a l s i m p l i f i c a t i o n on the reading comprehension of s e l e c t e d high school students. J o u r n a l of Reading Behavior, 1973, .5 4, 273 - 281. Fagan, W. An i n v e s t i g a t i o n i n t o the r e l a t i o n s h i p between reading d i f f i c u l t y , and the number and types of sentence t r a n s f o r -mations. Doctor's t h e s i s . U n i v e r s i t y of A l b e r t a . 1969 Ferguson, George A. S t a t i s t i c a l a n a l y s i s i n psychology and education. 2nd ed. New York, McGraw - H i l l , 1976 Gibbon, Helen D. Reading and sentence elements. Elementary E n g l i s h  Review, 1941, 18_, 42 - 46. Goodman, Kenneth S. and Carolyn L. Burke. A study of o r a l reading miscues that r e s u l t i n grammatical re t r a n s f o r m a t i o n s . 1969. ERIC micro-f i c h e ED 039 101 T h e o r e t i c a l l y based s t u d i e s of patterns of miscues i n o r a l reading performance. F i n a l . r e p o r t . 1973. . ERIC, m i c r o f i c h e ED 079 708 Granowsky, A l v i n . Background f o r a new s y n t a c t i c complexity formula. 1973. ERIC m i c r o f i c h e ED 083 566 H a l l i d a y , M. A. K. and R. Hasan. Cohesion i n E n g l i s h . London, Longman's, 1976 H a r r i s , Mary M. Second grade syntax attainment and reading achievement. 1975. ERIC m i c r o f i c h e ED 106 764 Hart, Margaret M. S. An assessment of the a b i l i t y to manipulate s y n t a c t i c s t r u c t u r e s as described by t r a n s f o r m a t i o n a l grammar theory. 1971. ERIC m i c r o f i c h e ED 080 995 H e r r i o t , P. The comprehension of a c t i v e and passive sentences as a f u n c t i o n of pragmatic expectations. J o u r n a l of Verbal Learning and Verbal  Behavior, 1969, 8, 166 - 169. Ives, Sumner. Some notes on syntax and meaning. The Reading Teacher, 1964 18, 3, 179 - 183 + 222. K o l e r s , P., A. Three stages i n reading. In H. L e v i n and J . P. W i l l i a m s (eds.) Basic s t u d i e s i n reading. New York, Basic Books, 1970 142 Kuntz, M i l d r e d L. The r e l a t i o n s h i p between w r i t t e n s y n t a c t i c attainment and reading a b i l i t y i n 7th grade. 1975. ERIC m i c r o f i c h e ED 113 710 Lesgold, Alan M. E f f e c t s of pronouns on c h i l d r e n ' s memory f o r sentences. 1972. ERIC m i c r o f i c h e ED 068 974 V a r i a b i l i t y i n c h i l d r e n ' s comprehension of s y n t a c t i c s t r u c t u r e s . 1973. ERIC m i c r o f i c h e ED 086 986 Marcus, A l b e r t D. The development of a d i a g n o s t i c t e s t of s y n t a c t i c meaning clues i n reading. In R. L e i b e r t (ed.) D i a g n o s t i c viewpoints i n  reading. Newark, D e l . , I n t e r n a t i o n a l Reading A s s o c i a t i o n , 1971 Menzel, Peter. L i n g u i s t i c bases of the theory of w r i t i n g items f o r i n s t r u c -t i o n s t a t e d i n n a t u r a l language. In J . R. Bormuth. On the theory  of achievement t e s t items. Chicago, U n i v e r s i t y of Chicago Press, 1970 M i l l e r , George A. Some p r e l i m i n a r i e s to p s y c h o l i n g u i s t i c s . American  P s y c h o l o g i s t , 1965, 20, 15 - 20. and S. Isard. Some perceptual consequences of l i n g u i s t i c r u l e s . Journal of Verbal Learning and Verbal Behavior, 1963, 2, 217 - 228, Mohan, Bernard A. Comprehension t e s t i n g as semantics plus induction. L i n g u i s t i c s , 1973, No. 115, 93 - 104. Morton, J . The e f f e c t s of context upon speed of reading, eye movements, and the eye-voice span. Quarterly Journal of Experimental Psychology, 1964, 16, 340 - 354. Nelson, Larry R. Guide to LERTAP use and i n t e r p r e t a t i o n . Dunedin, N. Z., Un i v e r s i t y of Otago, Education Dept., 1974 Nie, Norman H., Hadlai H u l l , Jean G. Jenkins, Karin Steinbrenner, and Dale H. Bent.., SPSS: S t a t i s t i c a l Package for the S o c i a l Sciences. 2nd. ed. New York, McGraw - H i l l , 1975 Oakan, R., M. Weiner, and W. Cromer. I d e n t i f i c a t i o n , organization, and reading comprehension for good and poor readers. Journal of  Educational Psychology, 1971, 62_, 71 - 78. O'Donnell, Roy C. A test of perception of syntactic a l t e r n a t i v e s . Studies i n Language Education, Report No. 2, 1973. ERIC microfiche ED 077 025 On l i n g u i s t i c structure and reading comprehension. 1976. ERIC microfiche ED 123 592 , William J . G r i f f i n , and Raymond C. Norris. Syntax of kinder-garten and elementary school c h i l d r e n . Champaign, 111., National Council of Teachers of English, 1967 143 and F. J. King. An exploration of deep structure recovery and reading comprehension s k i l l s . Research i n the Teaching of  English, 1974, 8., 3, 327 - 338. and William L. Smith. Increasing 9th grade students' awareness of syntactic structures through d i r e c t i n s t r u c t i o n . Research i n  the Teaching of English, 1975, 9_, 3, 257 - 262. Pearson, P. David. The e f f e c t s of grammatical complexity on children's comprehension, r e c a l l , and conception of c e r t a i n semantic r e l a t i o n s . Reading Research Quarterly, 1974, H), 2, 155 - 192. P e l t z , Fillmore K. The e f f e c t upon comprehension of repatterning based on students' w r i t i n g patterns. Reading Research Quarterly, 1974, 9_, 4, 603 - 621. Reid, Jessie F. Children's comprehension of syntactic features found i n some extension readers. In Jessie F. Reid (ed.), Reading: Problems and- P r a c t i c e . London, Ward Lock, 1972. pp. 394 - 403. Richek, Margaret Ann. E f f e c t s of sentence complexity on the reading compre-hension of syntactic structures. Journal of Educational Psychology, 1976, 68_, 6, 800 - 806. Rode, Sara S. Development of phrase and clause boundary reading i n c h i l d r e n . Reading Research Quarterly, 1974, 10_, 1, 124 - 142. Ruddell, Robert B. Language a c q u i s i t i o n and the reading process. 1969. ERIC microfiche ED 033 819 Simons, Herbert D. L i n g u i s t i c s k i l l s and reading comprehension. 1970. ERIC microfiche ED 047 927 Smith, William L. The e f f e c t of transformed syntactic structures on reading. 1970. ERIC microfiche ED 044 247 Spache, G. A new r e a d a b i l i t y formula for primary grade reading materials. Elementary School Journal, 1953, 53, 410 - 413. Stedman, Nathan A. I I I . The e f f e c t of a curriculum teaching syntactic embedding upon the reading comprehension of 4th grade students. 1971. ERIC microfiche ED 072 426 Steiner, R., M. Weiner, and W. Cromer. Comprehension t r a i n i n g and i d e n t i -f i c a t i o n of poor and good readers. Journal of Educational Psycho-. .. logy, 1971, 62, 506 - 513. Stoodt, Barbara D. The r e l a t i o n s h i p between understanding grammatical conjunctions and reading comprehension. 1970. ERIC microfiche ED 060 010 144 Takahashi, Barbara L. Comprehension of w r i t t e n s y n t a c t i c s t r u c t u r e s by good readers and slow readers. 1975. ERIC m i c r o f i c h e ED 117 655 Tatham, S. M. Reading comprehension of m a t e r i a l s w r i t t e n w i t h s e l e c t o r a l language p a t t e r n s : A study at Grades 2 and 4. Reading Research  Qua r t e r l y , 1970, 3 , 406 - 425. Tuinman, J . Jaap. Determining the passage dependency of comprehension questions i n f i v e major t e s t s . Reading Research Q u a r t e r l y , 1973, 9_, 2, 206 - 223. van Metre, P a t r i c i a D. S y n t a c t i c c h a r a c t e r i s t i c s of s e l e c t e d b i l i n g u a l c h i l d r e n . Claremont Reading Conference, 38th yearbook, 1974, 102 - 108. Vogel, Susan A. S y n t a c t i c a b i l i t i e s i n normal and d y s l e x i c c h i l d r e n . B a l t i m o r e , Md., U n i v e r s i t y Park Press, 1975 Weber, R. M. F i r s t graders' use of grammatical context i n reading. In H. L e v i n and J . P. Wi l l i a m s (eds.) Basic s t u d i e s i n reading. New York, Basic Books, 1970 Winer, B. J . S t a t i s t i c a l p r i n c i p l e s i n experimental design. New York, McGraw - H i l l , 1962 Wisher, Robert A. The e f f e c t s of s y n t a c t i c expectations during reading. J o u r n a l of Educational Psychology, 1976, 68, 5, 597 - 602 Appendix "A" MODEL FOR SENTENCE COMPREHENSION WITHIN A WRITTEN TEXT Test format: items in statement form, followed by three response options: true, false, cannot say Assumption: propositional content of all item statements keyed "true" or "false" can be demonstrated to be derived from a specific locus in an accompanying passage through a series of syntactic transformations or anaphoric relations Having read the passage in its entirety, the examinee Reads an item Successfully recalls from short term memory underlying meaning of a related statement in passage Determines truth value of item proposition Compares proposition of such passage statement with item proposition Selects response option most appropriate Fails to recall an underlying meaning for any statement in passage related to item proposition Re-reads passage, relying upon lexical clues, in search for most probable related statement Successfully locates a statement in passage whose underlying meaning is related to item proposition Compares proposition of such passage statement with item proposition Determines truth value of item proposition Selects response option most appropriate Fails to locate a statement in passage whose underlying meaning is related to item proposition Successfully recalls from long term memory propositional content of a statement presented elsewhere considered related to item proposition Compares propositional content of recalled statement with that of item Determines truth value of item proposition Selects response option most appropriate Fails to recall from long term memory any statement presented elsewhere whose propositional content can be related to item proposition Uses principles of logical reasoning to select a response option OR Randomly selects a response option Does not respond to item * a terminal behaviour that does not demonstrate comprehension of the text at hand 146 Appendix "B" THE RESEARCH INSTRUMENT Test No. 1 Test No. 2 Test No. 3 Test No. 4 147 154 158 163 Item d i f f i c u l t y Elem JrSec SrSec N.S. E.S.L. • N.B. -In a l l cases, the item d i f f i c u l t y index i s computed on the basis of the t o t a l number of Ss responding to the p a r t i c u l a r item. Not a l l Ss responded to a l l items or to both scales of Test No. 4, Test No. 1 D i r e c t i o n s : Read both sentences. If the two sentences mean the same thing, c i r c l e "yes". If the two sentences DO NOT mean the same thing, c i r c l e "no1 Sample A: The boy h i t the g i r l . The g i r l was h i t by the boy. Sample B: The boy looked at the big dog. The big dog looked at the boy. yea no yes (no) Subtask A: P a s s i v i z a t i o n 1. A book was given to the g i r l by the boy. The boy gave a book to the g i r l . yes no .917 1.00 1.00 .938 1.00. 1.00 10. The buses were cleaned by the men. The men cleaned the buses. yes no .833 1.00 1.00 1.00 .967 .947 19. The cat w i l l chase the dog. The cat w i l l be chased by the dog. yes no .500 .700 .952 .813 .733 .895 28. The fireman saw the dog. The fireman was seen by the dog. yes no .833 1.00 1.00 .938 .933 1.00 148 Subtask B: P a r t i c i p l e m o d i f i e r The man saw that h i s car was s t o l e n . The man saw h i s car that was s t o l e n . yes no .750 .800 .952 .313 .067 .368 11. The man saw h i s s t o l e n car. The man saw that h i s car was s t o l e n . yes no .667 .800 1.00 .313 .533 .684 20. The man saw h i s car s t o l e n . The man saw h i s s t o l e n car. yes no .750 .900 .952 .438 .433 .632 29. The man saw h i s car s t o l e n . The man saw h i s car being s t o l e n . yes no .333 .900 .857 .500 .467 .737 Subtask C: Wh— f r o n t i n g 3. What John saw was a box. John saw a box. yes no 1.00 1.00 1.00 .875 .900 .895 12. What the boy would l i k e i s f o r the g i r l to leave. * The g i r l would l i k e the boy to leave. yes no .917 1.00 1.00 .938 .867 .947 149 21. The man taught the boy to use a hammer. What the man taught the boy was to use a hammer. yes no .667 .900 .952 .875 .933 .947 30. The boy wants the g i r l to f i n d the b a l l . A What the g i r l wants i s f o r the boy to f i n d the b a l l . yes no .917 .900 .905 .813 .767 .947 Subtask D: R e l a t i v i z a t i o n contrasted w i t h c l a u s a l c o n j u n c t i o n 4. The woman c a l l e d the policeman and he came down the h a l l . The woman the policeman c a l l e d came down the h a l l . yes no 1.00 .700 .762 .938 .967 1.00 13. The g i r l h i t the boy and he f e l l down. The boy the g i r l h i t f e l l down. yes no .167 .700 .810 .125 .267 .316 22. Helen drew a p i c t u r e of a clown and went home. Helen went home and drew a p i c t u r e of a clown. yes no .833 1.00 1.00 . .813 .900 1.00 31. Betty i s happy and she l i k e s her new school. •k Betty l i k e s her new school and she i s happy. yes no .917 1.00 1.00 1.00 1.00 1.00 » 150 Subtask E: R e l a t i v i z a t i o n by pronoun dele t i o n 5. The boy the g i r l h i t f e l l down. A The g i r l whom the boy h i t f e l l down. yes no .500 .700 .714 .563 .567 .895 14. The horse which was brown stood behind the cow which was black. A The brown horse stood behind the black cow. yes no .917 .900 1.00 .875 1.00 .947 23. The old t i g e r chased the young l i o n . The t i g e r which was young chased the l i o n which was o l d . A yes no .917 1.00 1.00 .938 .900 .842 32. The man whom the teacher saw ran down the steps. A The man the teacher saw ran down the steps. yes . no .833 .800 1.00 .438 .500 .579 Subtask F: Double transformation ( R e l a t i v i z a t i o n + Passivization) 6. The f a t boy kicked the t i n g i r l . A The g i r l who i s thin was kicked by the boy who i s f a t . yes no 1.00 1.00 1.00 1.00 .967 1.00 15. The f a t boy kicked the thin g i r l . A The boy who i s f a t was kicked by the g i r l who i s t h i n . yes no .917 .500 1.00 .938 .867 .789 151 24. The f a t boy kicked the thi n g i r l . The g i r l who i s f a t was kicked by the boy .who i s thi n . yes no .917 1.00 .952 .938 .967 1.00 33. The f a t boy kicked the thi n g i r l . The boy who i s thin was kicked by the g i r l who i s f a t . yes no .917 1.00 1.00 1.00 1.00 1.00 Subtask G: Ask (query) contrasted with T e l l 7. The boy asked the g i r l what to do. The boy asked the g i r l what she should do. yes no , -o.' .667 1.00 1.00 .813 .700 .789 16. The woman t o l d the man what to do. The woman t o l d the man what she should do. yes no .833 .900 1.00 .750 .700 .842 25. The woman asked the man what to do. The woman asked the man what she should do. yes no .667 .800 .762 .813 .633 .737 34. The boy t o l d the g i r l what to do. The boy t o l d the g i r l what she should do. yes no .750 .800 .810 .625 .767 .842 Subtask H: Easy to see 8. To see the g i r l i s easy. The g i r l i s easy to see. .917 1.00 .857 1.00 .800 .842 17. I t i s easy to see the g i r l . The g i r l i s easy to see. 1.00 1.00 1.00 .875 .767 1.00 26. The g i r l i s easy to see. The g i r l sees e a s i l y . .917 1.00 1.00 .563 .500 .684 35. The g i r l i s easy to see. The g i r l i s e a s i l y seen. 1.00 1.00 .952 .750 .400 .789 Subtask J : Promise contrasted w i t h T e l l 9. Mary t o l d Jack to come here today. Mary t o l d Jack that he should come here today. .917 .800 .762 .750 .867 .895 18. Mary promised Jack to come here today. Mary promised Jack that he would come here today. .500 .900 .857 .563 .600 .789 153 27. Mary t o l d Jack to come here today. Mary t o l d Jack that she should come here today. yyes no .917 .900 .952 .750 .900 .789 36. Mary promised Jack to come here today. Mary promised Jack, that .she would come here today. yes no .417 .700 .762 .563 .667 .842 154 Test No. 2 Direct i o n s : Read the f i r s t sentence c a r e f u l l y . Then read the second sentence and decide i f i t i s true or f a l s e . Sample A: i f : Ann and Helen walk to school together. does i t mean: Ann and Helen walk to school at the same time. (yes) no Sample B: i f : Mother said , "You must come home earl y . does i t mean: Mother must come home earl y . yes Q i o ) Subtask K: Indirect speech i f : Kathy said to her brother, "I want your skates." does i t mean: Kathy to l d her brother that he wanted her skates, yes .no .833 1.00 .952 .750 .867 .789 2. i f : Jack said to h i s s i s t e r , "Throw your b a l l to me." does i t mean: Jack t o l d h i s s i s t e r to throw her b a l l to him. yes no .833 1.00 .952 .938 1.00 1.00 i f : Susan said to B i l l y , "Do not come i n . " * does i t mean: Susan t o l d B i l l y that he should not come i n . yes no 1.00 .900 1.00 .875 .900 .895 i f : Peter's brother said to him, "Do not t a l k i n church. does i t mean: Peter's brother did not t e l l him to t a l k i n church. yes no .750 .800 .810 .750 .867 .789 155 Subtask L: Ask (request) contrasted with T e l l 5. i f : Mike asks Bob to go f i r s t . does i t mean: Mike wants to go f i r s t . yes no .917 .900 .905 .875 .767 .947 6. i f : Mike t e l l s Bob to go f i r s t . does i t mean: Mike wants to go f i r s t . yes no 1.00 1.00 1.00 .875 .833 .947 7. i f : Mike asks Bob to go f i r s t . does i t mean: Mike wants Bob to go f i r s t . yes no .917 .900 .905 .938 .767 .947 8. i f : Mike t e l l s Bob to go f i r s t . does i t mean: Mike wants Bob to go f i r s t . yes no .917 1.00 .905 1.00 .733 .895 Subtask M: Pseudoimperatives 9. i f I say: S i t down and I w i l l scream. does i t mean: I w i l l scream i f you don't s i t down. yes no .833 1.00 .952 .875 .933 .842 156 10. i f I say: Telephone me on Friday or I w i l l not come. * does i t mean: I w i l l not come i f you telephone me on Friday. yes no 1.00 .900 1.00 1.00 .633 .895 11. i f I say: Turn on the radio or I w i l l leave. does. it.,mean: ..If you don't turn on the radio, I w i l l leave. yes no .833 .800 .952 .688 .667 .947 12. i f I say: Come here and I w i l l t e l l you. * does i t mean: If you come here, I w i l l t e l l you. yes no .917 1.00 .952 .938 1.00.. .947 Subtask N: Agentless p a s s i v i z a t i o n (N. B. Only item d i f f i c u l t i e s for the en t i r e sample are reported here as t h i s subtask was excluded from further analyses.) 13. i f : The meat was eaten. does i t mean: Someone ate the meat. yes no .083 14. i f : The money was stolen. does i t mean: Someone sto l e the money. yes no .750 15. i f : The b a l l was thrown across the room. does i t mean: Someone threw the b a l l across the room. .917 yes no 157 16. i f : The chicken was k i l l e d . does i t mean: Someone k i l l e d the chicken. yes no .167 158 Test No. 3 Direc t i o n s : (given o r a l l y ) Sample: I had an apple for lunch. It. was good. I t apple  Subtask P: Pronominal reference The dance of the bees t e l l s which.lway to f l y . _It t e l l s how f a r to go. 1. I t (dance)  .333 .900 .952 .500 .433 .737 They picked up the bats and b a l l s and put them away. 2. them (bats and b a l l s ) .833 1.00 1.00 .938 .900 .947 Joe picked up the bat. He i s a good h i t t e r . 6. He (Joe)  .417 .900 .952 .813 .967 .737 159 A farmer knows that a hen w i l l not lay an egg i f it; i s shot. 11. I t (hen)  .333 .800 1.00 .813 .900 .842 Bees take the nectar that they l i k e and make jtt into honey. 14. they (bees)  .417 .800 .952 .875 .900 .842 15. i t (nectar)  .500 .800 .952 .750 .867 .789 To make butter, the cream i s taken from the milk and set aside to sour, It i s then churned u n t i l b i t s of f a t come together. 16. I t (cream)  .333 .500 .952 .250 .300 .526 Subtask R: Pronominal reference (Minimal Distance P r i n c i p l e ) Jack said, "When he was s i x years o l d , Jim learned how to read." 4. he (Jim)  .000 .600 .762 .375 .600 .158 Mary knew that Anne wanted her to pick up the toys. 8. her (Mary)  .417 .900 .952 .688 .900 .737 Helen t o l d her mother that she was t i r e d . 10. she (Helen)  .417 .800 1.00 .813 .933 .684 When Tom found out that Mike won the race, he was very happy. 13. he (Tom)  .333 .900 1.00 .813 .867 .737 Peter asked h i s father i f he was hungry. 20. he (father)  .417 .900 .952 .438 .600 .789 Subtask S: Nominal s u b s t i t u t i o n Some dogs have c o l l a r s with b e l l s . Others do not. 3. Others (dogs)  .417 .900 .857 .563 .933 .684 161 Although the season for cherry blossoms i s a short one, people can eat the f r u i t of the cherry tree a l l summer long. 12. one (season)  .333 .900 1.00 .563 .567 .579 For Chinese New Year, the windows which are made of thick r i c e paper are torn down and new ones are put up. 17. ones (windows) .250 .800 .952 ,188 .433 .421 In many c i t i e s there are buildings made of wood and others made of stone. Some are very o l d . 18. others (buildings) .417 .800 1.00 .750 .800 .632 19. Some (buildings) .417 .800 .857 .375 .800 .684 Subtask T: Clausal s u b s t i t u t i o n The boys played b a l l very hard. This i s what won the game. 5. This (playing very hard) . .250 .600 .762 .250 .467 .526 Jim might come and play. The team hopes so. 7. so (Jim comes and plays) .167 .500 .714 .438 .633 .842 B i l l hurt h i s hand. This worried the team. 9. This ( B i l l ' s hurting h i s hand) .500 .600 .810 .813 .967 .947 163 Test No. 4 (The following passage was presented on a separate page which was was detached from the test booklet.) (1) Long, long ago, people did not know how to b u i l d houses. (2) They had to l i v e i n caves on the sides of h i l l s . (3) ..They::could'. keep '.themselves -dry.: arid :warm:..in:".there. (4) These people hunted wild animals for food. (5) The skins of the animals were used f o r clothes. (6) The cave men did not know how to write but they could draw. (7) So they t o l d many s t o r i e s i n p i c t u r e s . (8) They drew t h e i r pictures on the stone walls of the caves. (9) In the l a s t few years, some of these caves have been found. (10) The p i c t u r e s t o r i e s are s t i l l there. (11) They show animals and people of those early times long ago. (12) They t e l l us things about cave men that we never knew before. Di r e c t i o n s : Read the story on the short paper f i r s t . Read each question sentence below. In the parentheses ( ) write the number of the sentence, or sentences, i n the story that t e l l s you the answer. If the question sentence i s true, c i r c l e "T". If the question sentence i s f a l s e , c i r c l e "F". If none of the sentences i n the story t e l l you the answer to the question, put }C i n the parentheses ( ) and c i r c l e "?". The cave men did not use the skins of animals. ( 5 ) ,.'T They l i v e d i n caves on the sides of h i l l s where they could keep dry and warm. (2+3 ) (T) F Sample A: Sample B: 164 Sample C: The cave men b u i l t f i r e s i n front of t h e i r homes. ( > C) T F Q? 1. Long, long ago, no one knew how to b u i l d houses. Task 3 1.00 1.00 .950 1.00 .967 1.00 Task 4 .833 .800 1.00 .938 .967 1.00 Task 5 .833 1.00 .950 .938 .931 1.00 2. People kept warm by l i v i n g i n caves on the sides of h i l l s . (2+3)^T) F Task 3 .417. .857 .850 .438 .793 .889 Task 4 .833 .800 .900 .875 .900 1.00 Task 5 .417 .857 .850 .563 .828 .889 3. These people hunted food f o r wild animals. Task 3 .750 ..714 .850 .813 1.00 .889 Task 4 .583 .800 .800 .563 .667 .789 Task 5 .500 .714 .800 .438 .690 .778 ( 1 ) Q F ? ( A ) T 0 ? 165 The p i c t u r e s t o r i e s about cave men t e l l us new things. (1Q+12)(T) F Task 3 .083 .571 .800 .688 .517 .500 Task 4 .500 .900 .900 .625 .500 .579 Task 5 .083 .571 .750 .563 .448 .444 The cave men could raw because they did not know how to write. ( 6 ) T (T?) Task 3 .833 .571 .800 .875 .897 1.00 Task 4 .417 .500 .250 .188 .167 .000 Task 5 .250 .143 .250 .188 .172 .000 In early times, a l l animals we] Task 3 .500 .429 .700 1.00 .897 .833 Task 4 .500 .400 .650 1.00 .900 .842 Task 5 .500 .429 .650 1.00 .897 .833 (>>) T F Q The picture s t o r i e s are there s t i l l . ( 10)(?) F ? Task 3 .417 .714 .950 1.00 .931 1.00 166 Task 4.... •. .750 .600 1.00 .938 .900 1.00 Task 5 .417 .714 .950 .938 .897 1.00 These people fed the wild animals. Task 3 .500 .429 .550 .813 .655 .833 Task 4 .417 .400 .550 .813 .633 .737 Task 5 .417 .429 .550 .813 .655 .722 9. Without drawing, these people used w r i t i n g to t e l l s t o r i e s . (6+7) T (F) ? Task 3 .083 .000 .400 .063 .172 .222 Task 4 .750 .600 .800 .313 .633 .526 Task 5 .083 .000 .400 .063 .138 .167 Long, long ago, everybody had to hunt Task 3 .333 .286 .350 .688 .621 .722 Task 4 .333 .400 .350 .688 .600 .579 Task 5 .333 .286 .350 .688 .621 .6(11 for h i s own food. (X_) T F (?) 167 11. The cave men wore the skins of wild animals. ( 5 )(T) F ? Task 3 .667 1.00 1.00 .625 .793 .833 Task 4 .750 .900 1.00 .563 .300 .789 Task 5 .667 1.00 1.00 .625 .759 .778 They hunted wild animals to eat. Task 3 .583 1.00 1.00 .813 .931 .944 Task 4 .667 .900 1.00 .750 .867 .947 Task 5 .583 1.00 1.00 .813 .862 .889 13. The picture s t o r i e s which show animals and people of early times are no longer there. (10+11) T ('F) ? Task 3 .000 .000 .250 .063 .138 .111 Task 4 .500 .700 .950 .375 .567 .737 Task 5 .000 .143 .250 .000 .069 .111 14. The pi c t u r e s t o r i e s about cave men give us new information. (10+12)0 F Task 3 .583 .857 .900 .188 .517 .556 168 Task 4 .667 .800 .950 .250 .533 .684 Task 5 .583 .857 .900 .188 .414 .500 15. There were wild animals i n the caves. ( ^ ) T F (T^ n the Task 3 .583 .714 .700 1.00 .897 .778 Task 4 .583 .600 .700 1.00 .900 .684 Task 5 .583 .714 .700 1.00 .897 .722 16. Pictures were one of the means of t e l l i n g s t o r i e s . ( 7 ) ( T) F Task 3 .250 .429 .900 .313 .448 .778 Task 4 .833 .800 .900 .313 .733 .789 Task 5 .250 .429 .850 .313 .448 .722 17. How to b u i l d houses was known long, long ago. ( 1 ) T Cfj ? Task 3 .500 .571 .900 .688 .690 .722 Task 4 .583 .700 .800 .688 .667 .684 Task 5 .417 .571 .800 .625 .655 .611 169 18. These people hunted wild animals f o r food. ( 4 )(?) f ? Task 3 .750 1.00 1.00 .938 .966 .944 Task 4 .833 .900 1.00 1.00 .967 .895 Task 5 .750 1.00 1.00 .938 .966 .889 People sometimes drew pictures Task 3 .667 .286 .600 .688 .448 .722 Task 4 .667 .300 .600 .688 .467 .632 Task 5 .667 .286 .550 .688 .448 .611 20. Although they did.not know how to could draw. Task 3 .583 1.00 .950 .813 .931 1.00 Task 4 .917 1.00 1.00 .750 .900 1.00 Task 5 .583 1.00 .950 .750 .862 1.00 write, the cave men ( _ 6 _ ) 0 F ? 21. Hunted by these people, the wild animals were eaten. ( 4 )(?) F ? Task 3 .583 .714 .900 .375 .724 .833 Task 4 .667 .700 .850 .313 .600 .842 170 Task 5 .583 .714 .850 .313 .621 .833 22. We can learn new things by looking s t o r i e s about cave men. Task 3 .583 .714 .850 .375 .621 .833 Task 4 .667 .700 .950 .563 .600 .789 Task 5 .583 .714 .850 .438 .552 .333 at the picture (10+12)0) F ? 23. Some of the picture s t o r i e s w i l l Task 3 .750 .857 .850 .625 .655 .722 Task 4 .750 .800 .750 .625 .633 .579 Task 5 .750 .357 .750 .625 .655 .611 be found. ( X ) T F (?) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0094464/manifest

Comment

Related Items