THE PRODUCTION AND RECOGNITION OFGRAMMATICAL AND UNGRAMMATICAL ENGLISH

WORD SEQUENCES BY BILINGUAL CHILDREN

Item Type text; Dissertation-Reproduction (electronic)

Authors Pialorsi, Frank

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 22/02/2021 07:53:54

Link to Item http://hdl.handle.net/10150/288084

INFORMATION TO USERS

This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted.

The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction.

1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame.

3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete.

4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received.

Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 4B10S

73-28,787

PIALORSI, Frank Paul, 1937-THE PRODUCTION AND RECOGNITION OF GRAMMATICAL

AND UNGRAMMATICAL ENGLISH WORD SEQUENCES BY BILINGUAL CHILDREN.

The University of Arizona, Ph.D., 1973 Education, general

University Microfilms, A XEROX Company, Ann Arbor, Michigan

THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED.

THE PRODUCTION AND RECOGNITION OF GRAMMATICAL

AND UNGRAMMATICAL ENGLISH WORD SEQUENCES

BY BILINGUAL CHILDREN

by

Frank Paul Pialorsi

A Dissertation Submitted to the Faculty of the

COLLEGE OF EDUCATION

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

19 7 3

THE UNIVERSITY OF ARIZONA

GRADUATE COLLEGE

I hereby recommend that this dissertation prepared under my

direction by Frank Paxil Pialorsi

entitled JHE PRDTTFJCTTON AND RECOGNITION OF GRAMMATICAL AND

UNGRAMMATICAL ENGLISH WORD SEQUENCES BY BTLTMiTTAT. flHTTiTIRBM

be accepted as fulfilling the dissertation requirement of the

degree of DOCTOR OF PHILOSOPHY

Dissertation tion Director Date /

After inspection of the final copy of the dissertation, the

follovring members of the Final Examination Committee concur in

its approval and recommend its acceptance:""

Q>cJ? . —.

c L S,

r/n/73 S-/6- 73

This approval and acceptance is contingent on the candidate's adequate performance and defense of this dissertation at the final oral examination. The inclusion of this sheet bound into the library copy of the dissertation is evidence of satisfactory performance at the final examination.

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfill­ment of the requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under the rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quota­tion from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judgment the proposed use of the material is in the interests of scholarship. In all other in­stances, however, permission must be obtained from the author.

ACKNOWLEDGMENTS

The author wishes to express his appreciation to the

members of his committee for their assistance in the planning

and completion of this study.

A special word of gratitude is due the pupils and teach­

ers of the six schools in which the study was conducted and the

three school districts which granted permission and offered sug­

gestions for the improvement of this project.

Lastly, to my wife, Leyla, for her loving support and

good humor, I dedicate this work.

iii

TABLE OF CONTENTS

Page

LIST OF TABLES vi

ABSTRACT viii

1. INTRODUCTION 1

Statement of the Problem 5 The Testing Model 7 Need for the Study 8 Hypotheses Tested 9 Definitions of Terms Used 10 Assumptions 14 Limitations of the Study 14 Summary 15

2. REVIEW OF THE LITERATURE 16

The Effect of Bilingual Education on Learners ... 16 How Bilingualism Has Been Described 18

Categories 18 Dichotomies 18 Scales 19

Teaching English Sentence Patterns 22 Teaching Bilingual Children 26 Problems of Language Interference 28 Summary 30

3. PROCEDURES 32

Selection of Subjects 32 Description of the Instrument 33 Data Collection 34 Analysis of Data 35 Construction of the Test 36 Translation and Testing 37 Summary 38

iv

V

TABLE OF CONTENTS—Continued

Page

4. RESULTS OF THE STUDY 40

Reliability 40 Validity 41 Summary 58

5. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 60

Summary 60 Conclusions 63 Recommendations 64

APPENDIX A: EXPLANATION AND INSTRUCTION SHEET .... 68

APPENDIX B: PERSONAL DATA SHEET 70

APPENDIX C: TEST DESIGNS 71

APPENDIX D: TEST OF THE PRODUCTION AND RECOGNITION OF ENGLISH GRAMMATICAL SEQUENCES PART I: RECOGNITION 73

APPENDIX E: TEST OF THE PRODUCTION AND RECOGNITION OF ENGLISH GRAMMATICAL SEQUENCES PART II: PRODUCTION 77

APPENDIX F: TEST DATA: GROUP 1 NATIVE ENGLISH SPEAKERS (NES) 80

APPENDIX G: TEST DATA: GROUP 2 BILINGUALS/ BILINGUAL PROGRAM (URBAN) 83

APPENDIX H: TEST DATA: GROUP 3 BILINGUALS/ BILINGUAL PROGRAM (URBAN) 87

APPENDIX I: TEST DATA: GROUP 4 BILINGUALS/ NON-BILINGUAL PROGRAM (URBAN) 91

APPENDIX J: TEST DATA: GROUP 5 BILINGUALS/ BILINGUAL PROGRAM (BORDER) 95

APPENDIX K: TEST DATA: GROUP 6 BILINGUALS/ NON-BILINGUAL PROGRAM (BORDER) 99

SELECTED BIBLIOGRAPHY 103

LIST OF TABLES

Table Page

1. Division of Groups for Test of Production and Recognition of Grammatical and Ungrammatical English Word Sequences 8

2. Means, Standard Deviations, and Alphas for Groups 2-6 Combined 40

3. Classification of Items According to Grammatical Problems 42

4. Means, Standard Deviations, and Internal Con­sistency of the Test of Recognition for Groups 1-6 43

5. A Comparison of Choice Distribution Percentages Between Group 1 (NES) and Groups 2-6 (NSS) ... 44

6. ANOVA: Recognition Results—Groups 1-6 48

7. Mean Differences: Results of Tukey Post Hoc Test 49

8. ANOVA: Production Results—Groups 2-6 50

9. ANOVA: Total Test Results—Groups 2-6 51

10. ANOVA: Recognition Results--Bilingual Program Subjects (Group 1) vs. Non-Bilingual Pro­gram Subjects (Group 2) 52

11. ANOVA: Production Results—Bilingual Program Subjects (Group 1) vs. Non-Bilingual Pro­gram Subjects (Group 2) 53

12. ANOVA: Recognition Results—Urban Groups (2, 3, 4) vs. Border Groups (5, 6) 54

13. ANOVA: Production Results—Urban Groups (2, 3, 4) vs. Border Groups (5, 6) 55

14. ANOVA: Recognition Results—Male vs. Female .... 56

vi

LIST OF TABLES—Continued

Table Page

15. ANOVA: Production Results—Male vs. Female ..... .57

16. Intercorrelation Analysis—Recognition and Production Groups 2-6 58

ABSTRACT

The purposes of this study were to measure the variance

between the recognition and production of bilingual fourth grad­

ers, to what extent the first language (Spanish) interfered with

the second (English), and which patterns might yet be unlearned

by both native English speakers and non-native English speakers.

Comparisons between the performances of bilingual program and

non-bilingual program groups were made to determine whether or

not there would be significant differences in their abilities to

recognize and produce correct English grammatical sequences.

Since no reliable test of this type existed for bilingual

children, the study first involved the constructing of a test of

basic English sentence types for bilingual children. Initial

pilot testing of the test items took place in a Tucson, Arizona,

elementary school with thirty subjects in groups of ten. These

groups were not included in the final testing program.

Procedures

Test items were adapted from a diagnostic test for stu­

dents of English as a second language by A. L. Davis. The subjects

were ninety native Spanish-speaking fourth grade pupils who had

participated three or more years in any of three bilingual pro­

grams or any of two non-bilingual programs in southern Arizona

viii

schools. The groups were further classified as urban (Tucson,

Arizona) and border (Nogales, Arizona). In addition, Part I of

the test was given to twenty native English speakers at the

fourth grade level and of a socio-economic background similar to

that of the five native Spanish-speaking groups . This was done

to determine whether or not the native English speakers are sig­

nificantly different from the non-native English speakers in

their ability to recognize correct English grammatical sequences.

Part I of the test was made up of thirty-four multiple

choice items based on a constrastive analysis of English and

Spanish grammar (Recognition). Part II consisted of the correct

items of Part I in their Spanish equivalents and required the

subjects to write the correct English equivalents (Production).

There was at least a one-week interval between the administration

of Part I and Part II of the test.

Results

The results of the study showed that:

1. The native English-speaking groups differed from the five

Spanish-speaking groups in their Recognition performances.

2. There were no significant differences among the five

Spanish-speaking groups in their Production performances.

3. There were no significant differences among the five

Spanish-speaking groups in their total test performances.

4. There were significant differences between the three bi­

lingual program groups and the two non-bilingual groups

X

in their Recognition performances, It appears that the

non-bilingual program subjects performed higher on the

Recognition scale,

5, There were no significant differences between the urban

groups and the border groups ,

6, There were no significant differences between sexes,

7» There was a positive correlation between the subjects'

recognition and production performances,

Conclusions

On the basis of this study, the following conclusions

•were made;

1, Problems with basic grammatical English sequences among

bilingual children can be identified,

2, Native English-speaking children had less difficulty in

identifying the correct basic sentence types presented

than the bilingual groups,

3, The five native Spanish-speaking groups, whether in

structured bilingual programs or not, showed no signifi­

cant differences in their Production performances.

4, At this point in their English language development, the

non-bilingual program groups in the study seem to have

an advantage over the bilingual groups in their abilities

to identify correct English grammatical sequences,

CHAPTER 1

INTRODUCTION

This chapter begins with an overview of the more basic

issues vital to the education of Spanish-English bilingual chil­

dren. The problems of the study are defined in the form of the

research model, followed by an explanation of the need for the

study. Ten hypotheses are stated, the definitions of terms used

are identified, followed by the assumptions and limitations of

the study.

With the passage of the federal ESEA Title VII bill (The

Bilingual Education Act) calling for bilingual education, the

problems of bilingualism and English as a second language have

been of concern to many schools with bilingual and non-English

speaking populations. The Bilingual Education Act defines bilin­

gual education as "the use of two languages, one of which is

English, as mediums of instruction (Saville and Troike 1970,

p. 2)." This broad definition has prompted some schools to ap­

proach the ideal of producing students who can function effec­

tively in both English and their native language.

The passage of the Bilingual Education Act, then, and the various programs instituted in recent years represent a change in the philosophy of American education, from a rejection of other languages to an acceptance of them as valuable national resources and as valid and even neces­sary mediums for instruction (Saville and Troike, p. 3).

1

2

Concerned educators have become increasingly aware of the

need to define and measure bilingualism and solve some of the

theoretical and procedural problems besetting them. Prior to the

passage of Title VII, among the more basic issues vital to the

education of Spanish-English bilingual children in the Southwest

were the following, as stated by the National Advisory Committee

on Mexican-American Education (1968):

1. The inadequacy of existing educational programs for non-

native English speaking children.

2. The lack of suitable materials for measuring the intelli­

gence and achievement potential of these children.

3. The small percentage of minority students who qualified

for college who actually enrolled.

4. Legal restrictions in various states which discouraged

instruction in any language other than English.

5. The high dropout rate in secondary schools.

6. The lack of recognition on the part of society of the

need and benefit of a multilingual, multicultural educa­

tional environment.

At present, there are many federally supported programs

to develop bilingual education in various school districts across

the country. "Bilingual education" should be understood as edu­

cation in two languages, with the pupil's mother tongue being

used initially for all or the major part of instruction and with

continued use depending upon the bilingual educational model

3

adopted (Valencia 1969, p. 4). Aside from the demoralizing ef­

fect of forbidding a child to use his native language in the

school environment, many linguists agree that a child who does

not speak the national language learns to read it more efficient­

ly if he first learns to read in his mother tongue (Modiano

1970).

Researchers, concerned with the equal development of both

languages, especially in Spanish-English programs, are examining

each of the many dimensions of bilingual education. Among these

are:

1. The methods of measurement of the development of the lan­

guage processes in Spanish and English.

2. The effect of bilingualism, if any, on cognitive skills.

3. The development of the minority child's positive self-

concept in relation to his native culture and his role

in the dominant culture.

4. An understanding of the dominant and native cultures .

5. The involvement of parents, teachers, and the community

in school programs.

This study was concerned with dimension one above and further

limited to English, the second language. Children with little or

no English are generally not able to handle the tasks formulated

by a middle-class curriculum. Upon entering school, many have

been forced to abandon their native language and use only English.

4

This results in a restricted language, both native and second.

The problem of a restricted language has been summarized by

Bernstein (1961, p. 31):

1. Short, gramatically simple, often unfinished sentences.

2. Repetitive use of conjunctions.

3. Little use of subordinate clauses to modify the dominant

subject.

4. An inability to hold a formal subject through a speech

sequence.

5. Rigid and limited use of adjectives and adverbs.

6. Frequent use of the personal pronoun, and little use of

the self-reference pronoun.

7. Frequent statements that confound the reason and the con­

clusion to produce a categorical statement.

8. Considerable use of phrases that signal a requirement for

the previous speech sequence to be reinforced.

9. Above all, it is a language of implicit rather than ex­

plicit meaning.

Many researchers (Bernstein 1961, flusubel 1965) generally

agree that the child, lower-class or non-native speaking, who has

difficulty shifting from the concrete to the symbolic of lan­

guage, gets farther and farther behind in language skills as he

progresses from grade to grade. For the non-native English

speaking child, the problem is multiplied by his having to shift

initially from one language code to another. Considering this

5

difficulty, the investigator was concerned with developing a re­

liable instrument to measure the production and recognition of

grammatical and ungrammatical English word sequences by bilingual

children, and with this instrument, compare the performances of

Title VII and non-Title VII groups after three years in bilingual

and non-bilingual programs.

Statement of the Problem

Phase I of the problem involved the construction of a

reliable test of basic syntactic structures of English for bilin­

gual children. No reliable test of this type existed (Saville

and Troike 1970, p. 6). The need for the periodic testing of

bilinguals in English syntactic structures was confirmed by an

analysis of compositions written by native Spanish-speakers in

college freshman English classes. Basic word order, faulty use

of prepositions, and the formation of verb phrases were recurring

errors not common to native English-speakers, even though many of

the samples were written by students who had attended English-

speaking schools for twelve years . In bilingual programs in

which the goal is to have students become proficient in two lan­

guages , tests of grammatical sequences should be given periodi­

cally to determine deficiencies and methods of correction. These

tests should vary from basic sentence types as used in this study

to sophisticated forms used in college reading and writing.

Initial pilot testing of the test items took place in a

Tucson, Arizona, elementary school with thirty bilingual subjects

6

in groups of ten at the fourth and sixth grade levels. For a

description of the test construction, see Chapter 4.

Part I of the test (Appendix D) consisted of thirty-four

multiple choice items in English; the subjects were asked to

choose the correct form. Preliminary investigation indicated

that fourth graders who are native speakers of standard English

have little or no trouble in choosing the correct forms. (Exam­

ple: In the winter it is cold as opposed to In the winter it

makes cold.)

Part II of the test (Appendix E) required the subjects to

translate the thirty-four correct items, in the form of their

Spanish equivalents, back into English. There was a one-week

lapse between the administering of Part I and Part II of the

test.

Phase II of the problem involved the analysis of the test

data to measure the variance between the competence and perform­

ance of bilingual fourth graders, to what extent the first lan­

guage (Spanish) interfered with the second (English), and which

patterns might yet be unlearned by both native English speakers

and non-native English speakers. (See Appendix C for test design

model.) Comparisons between the performances of bilingual pro­

gram and non-bilingual program groups were made to determine

whether or not there are significant differences in their abili­

ties to recognize and produce correct English grammatical se­

quences . Part I of the test was also given to twenty native

7

English-speaking fourth graders to determine whether or not they

would have difficulty with the test items for competence.

The Testing Model

The testing model for this study included six groups of

fourth graders: three in Title VII bilingual programs whose home

and native language was Spanish, two in non-Title VII programs

whose home and native language was Spanish, and one with native

English speakers only. The groups were of similar, (lower)

socio-economic backgrounds. Initially, there were twenty sub­

jects, ten males, ten females, in each group. Several subjects

were later dropped due to continued absence or home language var­

iables . ft Personal Data Sheet (Appendix B) was used to obtain

language background information.

Table 1 illustrates the population distribution and com­

position of the six groups used in this study.

The variables which were controlled in the study included

the following:

1. The school programs were either bilingual or non-bilingual

as determined by Title VII outlines.

2. The subjects1 native language was Spanish or English in

the case of the monolingual group.

3. The subjects' home languages were Spanish or English in

the case of the native English-speaking group.

4. The subjects' length of time in the school programs was

three years or more.

8

5. All schools represented a similar socio-economic environ­

ment .

Table 1. Division of Groups for Test of Production and Recogni­tion of Grammatical and Ungrammatical English Word Sequences

Monolingual Program Native

English Speakers

Test Schools Bilingual Programs

Native Spanish Speakers

Non-Bilingual Programs Native

Spanish Speakers

Group 1

N-10 N-10

Urbc

Group 2

N-10 N-7

in

Group 4

N-9 N-6

Group 1

N-10 N-10

Group 3

N-10 N-9

in

Group 4

N-9 N-6

Group 1

N-10 N-10

Bord

Group 5

N-9 N-10

er

Group 6

N-10 N-10

Need for the Study

Among the more than ten thousand entries on educating the

minority child in the Bibliography: Education of the Minority

Child (Weinberg 1970), many attest to the concern of educators

regarding the description and measurement of bilingualism and

9

bilingual education. International seminars on bilingualism in

Canada in 1969 and Wales in 1965 indicate a world awareness of

the educational problems involved in countries with bilingual

populations. Most important is the fact that English-speaking

countries have been forced to study the school problems of bi­

lingualism in an objective and scientific way (Mackey 1967,

p. 3). At present, we are finding that the United States, which

was officially one of the most unilingual nations in the world

a short time ago, is in practice one of the most bilingual

(Mackey 1967, p. 29). To provide better instruction in bilingual

schools, it is necessary to establish valid measures for each of

the many dimensions of bilingualism (Kelly 1969, p. 6). There­

fore, the development of a reliable instrument to measure the

production and recognition of basic grammatical and ungrammatical

English word sequences by bilingual children and compare results

is prerequisite and most conducive to the establishment of these

measures.

Hypotheses Tested

The hypotheses which gave direction to this study were

that:

Hj: There will be no differences between the native English-

speaking group and the five Spanish-speaking groups in

their Recognition performances.

H2: There will be no differences among the.five Spanish-

speaking groups in their Production performances.

10

H3: There will be no differences among the five native

Spanish-speaking groups in their total test performances.

H4: There will be no differences between the three bilingual

program groups and the two non-bilingual program groups

in their Recognition performances.

H5: There will be no differences between the three bilingual

program groups and the two non-bilingual program groups

in their Production performances.

H6: There will be no differences between the urban groups and

the border groups in their Recognition performances.

H7: There will be no differences between the urban groups and

the border groups in their Production performances.

Hg: There will be no differences due between males and fe­

males in their Recognition performances.

Hg: There will be no differences between males and females in

their Production performances.

H10: There will be no positive correlation between the sub­

jects' Recognition and Production performances.

Definitions of Terms Used

The following definitions applied throughout this study:

1. flnglo: pertains to members of the dominant culture and

population in the United States.

2. Bicultural: pertains to behavior typical of two distinct

cultures.

Bilingual: a user of a second language to varying de­

grees in addition to his native language.

a. Balanced Bilingual: proficient in two languages,

with no interference of the native language with

the second language. An individual who is equal­

ly skilled in the use of two languages (Saville

and Troike 1970, p. 6).

b. Compound Bilingual: an individual who translates

from one language to another, does not keep lan­

guages separate, and experiences considerable in­

terference between them (Saville and Troike 1970,

P. 6).

c. Coordinate Bilingual: an individual who has two

separate language systems, usually learned under

different conditions, which cause minimal inter­

ference with each other.

Bilingual Science: systematized knowledge of the using

and users of two languages, especially with the fluency

of a native speaker.

Binary Phenomenon: a structured, simultaneous use of two

languages at the same time.

Culture: the totality of learned, socially transmitted

behavior of a specific group (Keesing 1958, p. 30).

Dialect: the variety of language spoken by members of a

single speech community, either geographical or social.

12

8. English as a Second Language: the planned instruction of

Standard English in varying degrees to non-native speak­

ers, usually in classroom situations.

9. Ethnic Group: "A group of people who have a distinct

culture, or racial heredity, or both (Zadrozny 1959,

p. 111)."

10. Language Distance: the range of difference between two

languages in phonology, morphology, and syntax.

a. Phonology: the sound system of a language.

b. Morphology: the word structures of a language.

c. Syntax (grammatical): the orderly system of word

arrangement of a language.

11. Minority Group: a group of people not considered members

•of the dominant culture of a country. In this study,

Mexican-Americans are considered a minority.

12 . Multicultural: pertains to behavior and interchange typ­

ical among three or more distinct cultures.

13. Polyglot: a user of several languages to whom the mis­

nomer of linguist is often applied.

14. Production (performance): the ability to produce correct

grammatical sequences,

15. Recognition (competence): the ability to recognize cor­

rect grammatical sequences.

16. Sense of Urgency: a feeling of wanting to acquire a sec­

ond language mainly for the sake of communicating with

13

one or more of its speakers without the feeling of social

insecurity or dissatisfaction.

17. Sequence (pattern): a recurring design or arrangement of

sounds, morphemes, words, phrases or sentences. Each

language has its own characteristic patterns at each of

these levels. The sentence patterns of English are de­

signs or arrangements into which different words can be

put to fit the lexical situation. The sentence patterns

of Spanish, French, etc., are different from those of

English not only in the words but in the essential ele­

ments of the design as well (Lado 1961, p. 219).

18. Single Linguistic Repertoire: the theoretical view of

treating two languages as part of the same repertoire in­

stead of as two distinct entities (Tyler 1970, p. 438).

19. Sociolinguistics: an area of study concerned with the

relationships of language to society.

20. Standard: designates a language or dialect conforming to

the usage of educated speakers of a speech community.

21. Structural Linguistics: "Linguistic study in which each

language is viewed as a coherent, homogeneous entity

(Pei and Gaynor 1954, p. 205)."

22. Transfer: the extension of a native language habit into

the target language with or without the awareness of the

learner. When transferred habit is acceptable in the

target language, we have facilitation. When the

transferred habit is unacceptable in the target language,

we have interference, and an extra learning burden is

assumed (Pei and Gaynor 1954, p. 222).

Assumptions

For the purposes of the study it was assumed that:

English grammatical sequences are of varying difficulty

for bilingual pupils.

The English competence of bilingual pupils exceeds their

performance in basic English grammatical sequences.

For an accurate description and measurement of bilingual-

ism, each of its many dimensions must be observed and

analyzed.

Limitations of the Study

The following limitations of the study were anticipated:

The two-part test was limited to ninety fourth grade,

native Spanish-speaking pupils who had attended any of

five subject schools for three or more years and whose

home language is Spanish. In addition, Part I of the

test was given to one native English-speaking group of

twenty subj ects.

The two-part test was limited to the subjects' abilities

to recognize and, by means of translation, produce English

grammatical sequences.

15

3. The study had no regard for sibling relationships among

the subjects.

4. The study included subjects with socio-economic similari­

ties which comply with the Title VII socio-economic

guidelines.

5. The study was not concerned with classroom methodology.

6. The study was not concerned with classroom teachers'

ethnic background or experience.

7. The study had no regard for the subjects' performances on

any other aptitude scale. All six groups were selected

from heterogeneous fourth-grade classrooms.

Summary

The Bilingual Education Act (ESEA Title VII) has prompted

some schools to develop programs to enable their students to func­

tion effectively in two languages. Among the many dimensions of

bilingual education, development of correct grammatical forms in

the second language is a major concern and the subject of this

study.

The researcher has designed and administered a reliable

test of selected basic English sentence types for bilingual chil­

dren in Title VII and non-Title VII programs. This was done to

determine and compare their abilities to recognize and produce

these structures.

This study resulted from the need to examine one of the

many dimensions of bilingualism and bilingual education.

CHAPTER 2

REVIEW OF THE LITERATURE

A review of literature concerned with current views held

by sociolinguists on bilingual education is presented in this

chapter. Material is also cited on the description and measure­

ment of bilingualism by categories, dichotomies, and scales.

Special attention is given to the language acquisition of chil­

dren, controversies in the approaches to the teaching and testing

of grammatical structures and problems of language interference

among bilinguals .

The Effect of Bilingual Education on Learners

In the preface to the report on the Description and Mea­

surement of Bilingualism; An International Seminar, Bartlett (in

Kelly 1969) stated that as early as the Renaissance, studies of

bilingual children and works on foreign language teaching were

reported. In the first part of the twentieth century, observers

of language behavior linked bilingualism with human behavior in

that "... some of them were convinced that facility in a second

language had an adverse effect on the development of intelligence

(Bartlett in Kelly 1969, p. vi)." In some countries, governments

16

17

were concerned with the incidence of bilingualism in order to

develop a strong educational policy to stamp it out. In his dis­

cussion of the effect of bilingual education on learners, Haugen

(1964, p. 86) stated:

We may safely dismiss those studies which have tried to demonstrate that bilingualism is an intellectual handicap. These results have invariably been attained by some means of verbal intelligence tests . They have generally been administered to pupils who have learned one language at home and are being taught in a school system which uses a different language.... Even if there is some verbal handicap, the level attained may still be perfectly adequate, and the learner has the total satisfaction of being acceptable in both cultures: that of his home and that of his school. Lambert has clearly shown that when the true bilinguals are sepa­rated out from the partial, their performance ... is far superior to that of mono-linguals even on the verbal tests.

Similar views have been stated more recently by Mackey (1972) and

Lambert and Tucker (1972). Studies at the local level, such as

the Dade County Report (Inclan 1970 reach the same conclusions.

Recently, researchers have started to examine the phenom­

enon of bilingualism for its own sake. In the past twenty years,

studies have attempted to settle the arguments concerning models

for the description and measurement of bilingualism (Mackey 1967,

Anderson and Boyer 1970). The many conflicting theories among

linguists, sociologists, and psychologists have made it fruitless

to argue about what bilingualism really means. Discussions among

these groups have been limited mostly to description and measure­

ment. Mackey (in Kelly 1969, p. 5) stated that progress in the

analysis of bilingualism has been in recognizing bilingualism

". . . as a relative, complex, and multidimensional phenomenon,

best described through quantitative methods." It is important to

note the basic assumption that bilingualism can be measured.

How Bilingualism Has Been Described

Mackey's (in Kelly 1969) representative models for the

description of bilingualism can be summarized as follows:

Categories

Categorically, bilingualism has been described according

to proficiency and function. Proficiency categories are "com­

plete bilingual" (equi-lingual) and partial bilingual. As for

function, there are terms such as "home bilingualism," "school

bilingualism," and "street bilingualism." The disadvantage of

this categorical description is the impossibility to delimit.

Mackey (in Kelly 1969, p. 6) went on to ask; "When does a person

become a complete bilingual?" and "Doesn't bilingualism of the

street penetrate the home?"

Dichotomies

In the dichotomy model, there are coordinated versus

compound bilingualism; individual versus national bilingualism,

stable versus unstable bilingualism; balanced versus unbalanced

bilingualism; pure versus mixed bilingualism; general versus

specific bilingualism; simultaneous versus sequential bilingual­

ism; comprehensive versus limited bilingualism; organized versus

19

incidental bilingualism; and regressive versus progressive bilin-

gualisra, to name a few. The failure of the dichotomy description

is that such mutual exclusion takes no account of individual

cases of bilingualism which are "... rarely an either/or prop­

osition (Mackey in Kelly 1969, p. 6)."

Scales

A scale description of bilingualism presupposes standard

units of measure and valid procedure for their delimitation.

Such units do not actually exist. Furthermore, such units pre­

suppose an understanding of the nature of what is measured. Of

this type of description Mackey (in Kelly 1969, p. 6) stated;

Although the creation and standardization of units create a practical difficulty, the basic research re­quired constitutes a theoretical advantage, since it forces us to establish valid measures for each of the many dimensions of bilingualism. It is also this type of description which is prerequisite and most conducive to the establishment of measures.

The aspects which Mackey (in Kelly 1969, pp. 7-8) felt are most

important in the study of bilingualism are the following:

1. Developmental Aspects. What does the developing bilin­

gual hear, say, and read?

2. Psycholinguistic Aspects. To what extent are language

tests indices of bilingual proficiency or capacity?

3. Linguistic Aspects. How can we improve our measurements

of interference and language distance in the bilingual's

speech?

20

4. Sociolinguistic Aspects. Beside studying how well the

bilingual knows his languages, how can we best determine

what he does with them?

5. Socio-cultural Aspects. From the individual we pass to

bilingual group behavior. How can we describe and mea­

sure the behavior of bilingual groups?

6. Demographic Aspects. How can countries keep more accu­

rate language statistics?

Concerning Aspects (2) and (3), to which this study is related,

Macnamara (in Kelly 1969, p. 81) stated that bilingual study must

consider the division of skills which have long been traditional:

listening and reading, which are decoding skills; and speaking

and writing, which are encoding skills. In each of these, four

aspects can be distinguished: the semantic, the lexical, the

syntactic (grammatical), and the phonological. Discussing these

four, Hasselmo (in Kelly 1969, p. 129) stated that for grammati­

cal interference no methods of measurement are known.

Menyuk (1969, pp. 152-153) listed the hypotheses concern­

ing the linguistic processes that occur in the acquisition of

grammar by a native speaker. Some of these are the following:

1. The child can determine the linguistically significant

generalizations of categories in his acoustic environment.

For example, he can determine what a "sentence" is and

what a "speech sound" is.

21

2. The child can store in memory the features of the above

categories. That is, he can recognize a "sentence."

3. The child can store in memory the functional relation­

ships between falling and rising intonational contours.

4. The child can determine the fit and structure of se­

quences he produces and hears.

5. The child can both expand and alter his structural de­

scriptions of these categories, classes, and properties

to more closely«approximate the structural descriptions

in the grammar of his language as he matures. He can

generate utterances from these altered descriptions.

Carol Chomsky (1969, pp. 6-7) explained the difficulty of

syntactic complexity for the native-speaking child by the pres­

ence of the following four conditions:

1. The true grammatical relations which hold among the words

in a IsentenceJ are not expressed directly in its surface

structure.

2. The syntactic structure associated with a particular word

is at variance with a general pattern in the language.

3. A conflict exists between two of the potential syntactic

structures associated with a particular verb.

4. Restrictions on a grammatical operation apply under cer­

tain limited conditions only.

At age ten, a child's command of the grammar is found to

approach that of adults. The stages found in intervening years

reveal an orderly picture of gradual acquisition (Carol Chomsky

1969, p. 1).

Lado defined grammatical sequence as the patterns or ar­

rangements of words that have a meaning over and above the sepa­

rate meanings of the words that constitute them. For a language,

Lado (1961, p. 142) said, "the patterns of arrangement of words

in sentences and the patterns of arrangement parts of words into

words are its grammatical structure." To test grammatical con­

trol of the second language, Lado assumed that learning the

grammatical problems is learning the second language and by test­

ing the problems we are testing the language proficiency of the

learner. Like most linguists, Lado (1961, p. 150) makes a divi­

sion between production and recognition similar to Saussure's

langue et parole and Noam Chomsky's performance versus competence

It is a good practice to indicate the production and the recognition problems for each {grammatical] pattern ... use both labels in each pattern: "production prob­lems," "recognition problems." If a pattern is not a recognition problem, it can be so stated after the label.

Teaching English Sentence Patterns

Transformational linguists are currently voicing their

objections to the prevailing structural approach to second lan­

guage learning, an approach more concerned with pattern habits

than with language in context (Lester 1970). Textbooks and audio

lingual programs organize their materials in the interest of a

particular view of the principles governing linguistic form,

23

thereby isolating these forms from natural context. Newmark (in

Lester 1970, pp. 212-213) represented the views of various lin­

guists such as Teeter, Gunter, Temmerly, Rankin and Kane, who

prefer a transformational approach rather than a structural:

When structural linguists first faced the problem of developing methods to teach exotic languages, and later languages like English, they maintained a natural empha­sis on teaching concrete uses of language.

. . . a s s t r u c t u r a l i s t s g r e w m o r e a n d m o r e c o n f i d e n t about the "scientific" analysis of language, they modi­fied their teaching programs more and more to reflect these analyses: phonemic drills and structural pattern drills were increasingly elevated from the minor role they played in the early Army language courses to the major role they play in, say Michigan English Language Institute textbooks or in recent Foreign Service Insti­tute books. This increase in pattern drill is an index of the return from "natural" material to grammatical-illustration material.

Aside from pattern drill, differences between the two ap­

proaches exist in their approaches to the phonological rules,

which the structuralists state should be emphasized from the very

beginning. Again Newmark (in Lester 1970, pp. 215-216) summa­

rized the transformational view:

The fact that the detailed phonological rules come late in the grammar suggests that attention to the de­tails of pronunciation might be left until relatively late in a foreign language teaching program. Note that such delay in teaching "a good accent" is at sharp var­iance with the attitudes of most applied linguistics today, but is in good agreement with our common sense feeling that it is more important to be able to speak a language fluently and to say a lot of things in it than to have marvelous pronunciation but not know what to say. The relative lateness of phonological rules in a transformational grammar helps account for the fact that we can often understand a £ non-nativeJ speaker even when he lacks most of the phonological habits of English;

24

if we attempt to follow the order of grammatical rules in teaching simple before complex sentences, by the same token we should teach meaningful sentences before we worry much about teaching their proper pronunciation.

A basic tenet of both approaches, however, is that no

analysis or theory should detract the teacher from teaching his

pupils to use the second language effectively. Language learning

must never be inhibited by grammatical theory. The major empha­

sis should always be on the natural use of language rather than

the synthetic composition of sentences.

Summarizing the language problems of Mexican-American

pupils, Johnson stated that those who don't speak Spanish as a

first language might still have Spanish interferences due to

their cultural environment where English is influenced by Span­

ish. Their specific language problem is "that they impose the

sound and grammar system of Spanish on English." Johnson (1970,

pp. 170-171) stated further that this nonstandard language handi­

caps them in the same way that nonstandard English handicaps

Blacks and Appalachian whites: "It is important for teachers to

understand that the interferences are from Spanish; thus an un­

derstanding of Spanish—or at least an understanding of the in­

terference phenomenon—will help teachers to understand this

nonstandard language system." Johnson's (1970, p. 173) summary

of the more common nonstandard pronunciation and grammatical

features typical of Mexican-American pupils whose English is in­

fluenced by Spanish is as follows:

1. Difficulty with English sounds contained in these words;

miss, brother, share, very, cap, rice.

2. Accent on wrong syllables: perfectly, office.

3. Final consonant clusters dropped: the last consonant

sound in a consonant cluster at the end of words is

dropped: strength.

4. Syllable added for preterit ending represented by the

letters -ed in writing: jumped—jumpted; talk—talk-ed.

5 . Use of double negative.

6. Reversal of past-tense form of verbs and part-participle

form of irregular verbs: he should have gone—he should

have went.

7. End agreement sound not pronounced for third person sin­

gular present tense verbs: he runs all the time—he run

all the time.

8. Double subject: that man lives next door—that men he

lives next door.

9. Addition of the sound represented in writing by the let­

ter £ to irregular plurals: men—mens; children—chil-

drens .

10. Combining English and Spanish: market—marketo; watch—

watcho.

In the structuralist approach to the teaching of standard

English to bilinguals, it is evident that phonology, morphology,

and syntax, if they are to be correctly taught, cannot be

26

separated. In order to write grammatical sequences competently,

the pupil must first learn to pronounce them, with special atten­

tion given to those sounds and sound clusters which are not a

part of Spanish phonology and morphology, thus causing fewer

problems in syntax. The problem, then, is to create a second

complete set of language habits:

Given the capabilities of human beings as we know them, we can understand that the use of a language is made possible by reducing much of the operation of these .[language! categories to sets of complex habits. The user of a language does not verbalize or define these distinctions before he uses them. He does not identify them by technical names, and he probably does not iden­tify them as separate units at various levels of struc­ture. He must, however, react to their functions with precision and speed through highly developed habits (Finocchiaro 1964, p. 146).

Testing Bilingual Children

In his concluding remarks at the 1967 International Semi­

nar on Bilingualism in Montreal, Mackey (in Kelly 1969, pp. 362-

363) summarized the many variables involved in the testing of

bilingual children (the following is a translation by Linda

Haughton and the writer, Frank Pialorsi, of The University of

Arizona, from the original French):

We have emphasized that tests which utilize images of objects are not valid in the case of bilingual chil­dren who have not yet reached the age of mental general­ization. The researchers, in their testing, have been asked to take into account the relative use of the two languages in the environment of the person being tested, to remember to test attitudes toward the language and personality changes which occur when the person ex­presses himself in the second language.

We have been asked to describe and to test the types and degrees of bilingualism necessary for a whole range of individual functions and to elaborate tests which will permit us to predict the linguistic progress of a monolingual who may find himself placed in an environ­ment where only the other language is used.

We have been asked to test physical behavior, ges­tures and expressive movements of the bilinguals, their intelligibility, their repertory and varieties of bilin­gualism and at the same time the social prestige of the different types of bilingualism. Also we have been asked to study the number of roles which the bilingual can play in each language. The members of the colloquy, all of whom have, for a long time, thought about these questions concerning the testing of bilingual children have not hesitated to point out problems.

In the area of language acquisition, we ask ourselves how it is possible to determine competence of a language by testing the linguistic performance of the person. And, in testing performance, we wonder if it is justifi­able to test such automatisms as grammatical and lexical systems separately, as they always function simultaneous­ly in the person apeaking. We have also questioned the value of tests of a single automatism, as such tests pre­suppose habits not yet formed—for example, the tests of expression assume a certain degree of comprehension of the language.

We have also criticized the tests which presuppose an analytical knowledge of the language, for they are like trying to measure the output of a worker by asking him to describe the functions of his tools.

Aside from the complexities above, the purpose or func­

tion of language tests for bilinguals must be established. The

following list summarizes the chief objectives of language test

ing:

1. To determine readiness for instructional programs.

2. To distinguish degrees of proficiency.

3. To diagnose individual strengths and weaknesses .

28

4. To measure the degree of student achievement of the in­

structional goals.

5. To evaluate the effectiveness of instruction (Harris

1969, pp. 2-3).

The categories above are by no means mutually exclusive.

Problems of Language Interference

Language interference is difficulty in learning a sound,

word, or construction in a second language as a result of differ­

ences with the habits of the native language (Lado 1964, p. 217).

Where two languages are spoken in a given community, there might

be examples of uncontrolled grammatical fusion. Garza (1971,

p. 90) gave an example of a native Spanish speaker in a Spanish-

English border area:

Un fenomeno muy interesante que surgio por la amal-gama resultante de multiples factores en todas estas personas que vivieron o viven en los Estados Unidos fue el de la distorsion del lenguaje. . . . Pondre un ejem-plo ...

El Tijuana

Mi menda es chicano, he pedaleado por los estaires bute de anales. Soy solano, no tengo cuev , ni waifa [Angli­cisms are underlined3, pero me laican las gavachas, las husas y las tintas. Mi canton esta en Tijuana, ese es mi terre, allcf estan mi rucaila, mis carnales y los batos de la raza .... Era medio malero le laicaban tu moch las cebadas .... Se pasonderf alii gud taiiii porque medio apano a un bolillo pa'tumbarle su jando.

After this example we can look at the work of bilinguals such as

Alberto Alurista whose poetry has been described as a "binary

phenomenon" by Ortega (1969, p. 258).

29

Mis Ojos Hinchados

Mis ojos hinchados flooded with lagrimas

de bronce melting on the cheek bones

of my concern razgos indigenas the scars of history on my face and the veins of my body

that aches vomito sangre

y lloro libertad I do not ask for freedom

I am freedom (Alurista 1969, p. 172).

Jakobovits (in Kelly 1969), in his discussion of the po­

tential linguistic and psycholinguistic advantages of being bi­

lingual, counteracts the implied negative evaluation in the

universally used term of "interferences Linguistic interfer­

ences, he stated, stem from the psychological concept of trans­

fer, and transfer effects may be either negative or positive.

Although linguistic interferences have been emphasized in experi­

mental reports, the positive effects have not (Jakobovits in

Kelly 1969, p. 101):

An examination of the literature on the problem of the effects of bilingualism on educational and intellec­tual development . . . reveals an exclusive interest on the part of researchers in this area in the negative effects of bilingualism. The supporters of bilingualism are apparently quite happy in demonstrating that there is no evidence for negative effects of bilingualism. The Peal and Lambert study which has demonstrated the superiority of a bilingual group of children in the Montreal setting is a notable exception to this trend, and these authors have permitted themselves to at least examine the possibility that early bilinguality may con­tribute to greater cognitive flexibility in the child.

30

In structured bilingual programs, the objective is to

enable the bilingual child to develop the practice of alternately

using two languages. As a classroom discipline, this is very

much in the experimental stage. Gumperz (in Tyler 1970, p. 448),

in his discussion of treating two languages as "single linguistic

repertoire," noted that the speech systems of the Spanish-English

bilingual are distinct in every component and that language dis­

tance is measured as a function of the number of nonshared rules.

Keeping the rules separate is an important problem in Spanish-

English code switching. When we consider intralanguage varia­

tion, which is a major aspect of bilingual behavior, it is

important to note that among both educated and uneducated bilin-

guals, especially in informal speech, intralanguage rules are

often violated. Considerably more research is necessary in the

area of language interference. That language is primarily a

function of the mind is enough to justify further inquiry, re­

assessment, and experiment.

Summary

The attempts to describe and measure bilingualism and to

determine its psychological effects have been many. The conclu­

sion today is that coordinate bilingualism is no longer consid­

ered a handicap. Linguists, however, are still in disagreement

over the methods used to describe and measure bilingual behavior.

Most agree that the creation and standardization of units for

31

description and measurement create a practical difficulty, and

that research to establish valid measures must continue.

Research on native language acquisition has proven useful

in the teaching of a second language and has led to the formation

of two prominent schools of theory and approach in the structural

and the transformational. Structural methodology, however, seems

to be predominant due to scarcity of applied transformational ma­

terial .

Tests to evaluate bilingual performance should be de­

signed with the most complete knowledge possible of the social

and linguistic complexities of the bilingual child, especially

in the areas of first language interference.

CHAPTER 3

PROCEDURES

The procedures for 'the study are discussed in this chap­

ter. It begins with a description of the subjects chosen for the

testing, followed by a description of the testing instrument,

methods used for data collections, and the statistical analyses

of the data. The format of the test is explained by a summary of

Spanish and English grammatical systems, and, since Part II of

the test requires translation from Spanish to English, a discus­

sion of the advantages and disadvantages of translation in the

testing of a second language.

Selection of Subjects

The subjects for the study were ninety fourth-graders

whose first and home language is Spanish, and who had partici­

pated three or more years in any of the three selected bilingual

programs or any of two non-bilingual programs. In addition,

twenty native English-speaking fourth-graders were given Part I

of the test to determine whether or not English speakers would

have difficulty with the test items for competence. The study

included pupils from four Tucson, Arizona, elementary schools

and two Nogales, Arizona, elementary schools. The two areas are

32

33

classified as urban and border respectively. An equal male-

female ratio was sought, but was not possible for all of the six

heterogeneous groups. All groups were of similar socio-economic

background, ranging from upper-lower to lower-middle.

Socio-economic factors were considered since many studies

support the notion that environmental influences preclude ade­

quate language development. Bernstein (1961) characterized Brit­

ish working class children as having a restricted language code

in comparison with middle-class children. Deutsch (1960), in his

study of first and fifth grade children, found those of low socio­

economic status to be noticeably deficient in their use of cor­

rect structural patterns. In relation to bilingual children, the

testing of a child's language ability is based more and more on

his own language system. Serious attention is also being given

to the advisability of developing reading series using nonstan­

dard dialects, as well as the native language (Cheyney 1967,

pp. 82-83).

Description of the Instrument

The researcher designed his own test of grammatical and

ungrammatical English word sequences for elementary school chil­

dren. The item design is based on the Diagnostic Test for Stu­

dents of English as a Second Language developed by David (1953)

and published and distributed by Educational Services, Washington,

D.C. (see Zintz 1966, pp. 125-127). The test items are simple,

34

grammatical structures inherent in the speech of native speakers

of standard English.

Part I of the test was given to random samples of foreign

students in The University of Arizona's Center for English as a

Second Language. Native languages in the sampling included Span­

ish, French, Chinese, Greek, Arabic, and Portuguese. The sub­

jects, although varying in the years of study of English in their

native countries, were placed at the lower-intermediate level in

the CESL program. Results reaffirmed that basic English sentence

structures are of varying difficulty for non-native speakers.

Pilot testing was also done in a Tucson, Arizona, elementary

school. This involved thirty bilingual subjects in groups of ten

at the fourth and sixth grade levels at three different times.

The results were checked for reliability before administering the

test in its present form to the groups chosen for the study. The

subjects used in the pilot program were not included in the final

testing.

Data Collection

The testing was conducted by the researcher along with

the subjects' teachers. The research had previously provided the

teachers with an orientation to the test and its purposes (Appen­

dix A). The directions for the test were given by the teacher.

Both the teacher and the researcher answered any questions con­

cerning vocabulary. Thirty minutes were allowed for Part I of

the test. Forty-five minutes were allowed for Part II of the

35

test. Part I of the test was given at least one week prior to

Part II. The total period of testing was from January 5 to

February 12, 1973.

Analysis of Data

Statistical measurements to determine the test reliabil­

ity and to make inter-groups comparisons included the following:

1. To measure the internal consistency or obtain a reliabil­

ity estimate of both the Spanish and English parts of the

test, the Kuder-Richardson formula K-R was used. This

formula, as explained by Mehrens and Lehman (1969, pp. 38-

39), represents the average correlation obtained from all

possible split-half reliability estimates:

where rxx = the estimated reliability of the test.

N = the number of test items.

p = the percent of subjects who answered the item correctly (if, for example, 20 of 100 sub­jects answered the item correctly, p. 20/100 or .20).

q = the percent of people who answered the item incorrectly.

Sx2 = the variance of the total test.

2. Means and Standard Deviations were primarily used as

major statistical indices to describe the trend of the

data and for making inter-group and inter-regional com­

parisons .

z pq

~SP

3. One-way Analyses of Variance were used to determine the

group differences at the .05 level of confidence.

4. The Tukey Test was used if the F-ratio was significant

to determine which pairs of means differed significantly.

Construction of the Test

The test items (Appendix D) were based on a contrastive

analysis of Spanish and English grammatical systems. Since all

responses were in English, special attention was given the fol­

lowing in the development of test items:

1. Spanish makes use of two genders, masculine and feminine.

Both nouns and adjectives are inflected to show number

and gender.

2. While English has relatively few inflections, Spanish

uses variations in the endings of nouns and adjectives

and numerous inflections of the verb.

3. Singular and plural forms are not consistent between the

two languages.

4. Word order differs between the two languages. In Spanish

adjectives, for example, are usually placed after the

noun; in English they almost always precede nouns.

5. English auxiliaries such as do have no equivalents in

Spanish.

6. The many forms of the English negative are difficult for

Spanish speakers and require a careful presentation.

37

7. Question tags, such as He isn't here today, is he? and

He's here today, isn't he? are numerous in English,

whereas in Spanish there are only two: jverdad? and

<Lno?

The grammatical systems in English and Spanish sentences

are an important manifestation of a fundamental difference in the

thought sequences required by the two languages. As a consequence

of this difference, analogous sentence structures cannot always

be used in moving from Spanish to English or vice versa (Alien

1965, p. 149).

Translation and Testing

Part II of the test was a translation exercise. Formerly

one of the more common teaching and testing devices, translation

is still used in many parts of the world (Harris 1969). With the

structural approach to language learning in the United States,

translation has lost its appeal.

Translation is in reality a very specialized and highly sophisticated activity, and one which neither de­velops nor demonstrates the basic skills of listening, speaking, reading, and writing. Indeed, the habit of translating is now felt to impede the proper learning of a £secondJ language, for one of the first objectives in modern language instruction is to free the learner from native-language interference--to teach him to react in the target language without recourse to his mother tongue (Harris 1969, pp. 4-5).

Lado (1961, p. 261), however, stated that translation tests have

a self-evident validity: "... they are performance tests."

38

In evaluating a translation exercise, Lado (1961, p. 262) defined

"goodness of translation" as the degree to which it reproduces

the original material:

This essential simplicity, however, develops into at least five dimensions which complicate the job of produc­ing good translations and assigning scores or grades to translations. These dimensions are goodness of fit to (1) the letter and patterns of the original, (2) the meaning of the original at the sentence level, (3) the connotations of the original for its readers applied now to the readers of the translation, (4) the original as the readers understood it plus the flavor of the original language and culture for the readers of the translation who know they are reading a translation, (5) . . . the original in artistic effect rather than in detail.

In spite of the objections cited above, for bilingual

programs, especially in teaching the grammar of the second lan­

guage, translation of a grammatical sequence from the native to

the second language is useful in determining to what extent the

subjects are keeping the two language codes separate, what pat­

terns are incorrect due to first language interference, and what

patterns are still unlearned. The advantage of using basic sen­

tence types is that they allow little variation in the responses.

Summary

This study included ninety bilingual fourth graders from

three Title VII schools and two non-Title VII schools in order to

compare their ability to recognize and produce basic English

grammatical sequences. In addition, one group of native English

speakers of similar socio-economic level was tested to compare

their recognition ability with that of the bilingual groups .

The thirty-four items were based on a contrastive analy

sis of English and Spanish grammatical systems. Appropriate

statistical measures were used to determine the reliability of

test, as well as the significant differences among the six

groups.

CHAPTER 4

RESULTS OF THE STUDY

This chapter presents the results of the study. These

results, presented in descriptive terms and numerical tables,

precede the discussion relating to test reliability and each

hypothesis as stated in Chapter 1.

Reliability

Since the two-part test was designed by the researcher,

its reliability was an important consideration in the testing of

Groups 2-6. Table 2 presents the means, standard deviations, and

reliability figures (alphas) for Groups 2-6 combined.

Table 2. Means, Standard Deviations, and Alphas for Groups 2-6 Combined

Part I Part II Total Test

Means 23.7889 18.74444 42.5333

Standard Deviations 5.0033 6.0710 9.8140

Alphas .8708 .8430 .8804

Total test reliability coefficients among each of the five

schools, however, vary from .7332 (Group 2) to .9188 (Group 3)

(see Appendices F-K).

40

41

In judging reliability, Mehrens and Lehman (1969, pp. 40-

41) stated that the degree of reliability depends upon the pur­

pose for which the test is used: "Although there is no universal

agreement, it is generally accepted that tests used to assist in

making decisions about individuals should have reliability coef­

ficients of at least .85. For group decisions, a reliability

coefficient of about .65 may suffice."

Validity

The primary basis for the selection of the items was

established by a contrastive analysis of English and Spanish

grammatical systems and literature concerning the language ac­

quisition of children. Additional content validity for the total

test was determined by examining the choice distributions of each

item after pilot testing. Omitted items were usually concerned

with usage, rather than syntactic structure. Following pilot

testing, discussions of the test and recurring syntactic errors

in the subjects' writing with the subjects' teachers proved

helpful in the evaluation and final selection of the thirty-four

items.

Test test items may be classified by the following head­

ings (Table 3), according to what problems they were designed

to test.

Table 3. Classification of Items According to Grammatical Problems

Item Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 IS 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

F i r s t L a n g u a g e I n t e r f e r e n c e X X X X X X X X

M o d a l s / A u x i l i a r i e s X X X X

S u b j e c t - V e r b A g r e e m e n t X X

T a g Q u e s t i o n s X X X X

W o r d O r d e r X X X X X

U s e o f A d j e c t i v e s X X X

U s e o f P r e p o s i t i o n s X X . X X

U s e o f P r o n o u n s X X X X

ro

43

Table 4 shows the results of the Kuder-Richardson test

for internal consistency of Part I (Recognition) for Groups 1-6.

Table 4. Means, Standard Deviations, and Internal Consistency of the Test of Recognition for Groups 1-6

Group 1 NES*

Group 2 BBP(U)

Group 3 BBP(U )

Group 4 BNBP(U)

Group 5 BBP(U)

Group 6 BNBP(B)

M 28.35 24.2941 22.1053 24.1333 22 .3158 26.1000

SD 2.0803 4.3490 5.7756 4.0639 5.2219 4.0237

.1774 .7227 .8272 .7120 .7913 .7252

*NES Native English Speakers BBP Bilinguals in Bilingual Programs BNBP Bilinguals in Non-Bilingual Programs (U) Urban (B) Border

The results in Table 4 indicate that the test of Recogni­

tion had no reliability for Group 1 (Native English Speakers).

These results indicate further validity in that the native

English-speaking group had little difficulty with the Part I test

of Recognition in comparison with any one of the native Spanish-

speaking Groups 2-6. Although the internal consistency for

Groups 2-6 ranges from .7120 to .8272, a respectable realiability

coefficient, several items should be revised. A discussion of

these revisions follows.

The following table shows the Recognition choice distrib­

ution percentages of Group 1 compared with Groups 2-6 combined.

44

The item number is followed by the correct keyed answer and the

multiple choice distribution (see Appendix D, Part I).

Table 5. A Comparison of Choice Distribution Percentages Between Group 1 (NES)* and Groups 2-6 (NSS)**

Group 1 N=20 Groups 2-6 Combined N-90 Item Key 1 2 4 Item Key 1 2 3

1 3 0 10 90 1 3 12 17 71 2 2 0 100 0 2 2 6 94 0 3 1 90 0 10 3 1 71 10 19 4 1 60 15 25 . 4 1 67 10 23 5 2 0 100 0 5 2 7 87 7 6 2 0 90 10 6 2 7 77 17 7 2 0 100 0 7 2 4 89 7 8 1 100 0 0 8 1 93 1 6 9 1 70 5 25 9 1 60 10 30 10 1 100 0 0 10 1 91 7 2 11 2 0 100 0 11 2 6 76 19 12 1 65 25 10 12 1 41 20 39 13 2 0 100 0 13 2 0 88 3 14 3 5 20 75 14 3 1 28 71 15 2 0 90 10 15 2 3 89 8 16 1 100 0 0 16 1 93 1 6 17 2 0 95 5 17 2 6 84 10 18 1 90 5 5 18 1 61 19 20 19 3 0 55 45 19 3 3 40 57 . 20 2 15 80 5 20 2 31 58 11 21 1 100 0 0 21 1 81 3 16 22 3 0 5 95 ' 22 3 8 12 80 23 3 5 45 50 23 3 11 49 40 24 2 0 45 55 24 2 10 33 57 25 3 0 45 55 25 3 10 40 50 26 2 5 75 20 26 2 13 52 34 27 1 100 0 0 i 27 1 87 9 4 28 2 0 45 55 I 28 2 11 41 48 29 1 80 10 10 ! 29 1 36 33 31 30 3 0 10 90 30 3 14 23 62 31 2 0 100 0 31 2 12 78 10 32 2 10 80 10 32 2 19 73 8 33 2 15 85 0 33 2 26 67 8 34 1 95 5 0 34 1 81 9 10

*NES Native English Speakers **NSN Native Spanish Speakers

45

Group 1 was tested to establish whether or not native

English speakers are significantly different from non-native

speakers in their ability to recognize correct basic English

sentence patterns.

All twenty subjects in Group 1 responded correctly to

items 2, 5, 7, 8, 10, 11, 13, 16, 21, 27, and 31. Further exami­

nation of Table 5 indicates that Groups 2-6 scored consistently

high on the same items. This first test-run indicates that the

following items should be revised.

Item 2; a. Maria bought this book yesterday; it is his. b. Maria bought this book yesterday; it is hers. c. Maria bought this book yesterday; it is its.

Item 2 was to test pronoun gender. In the pilot testing there

was some indication that first language interference might cause

a problem as to pronoun choice. Subjects were choosing state­

ments such as "She was a good film," following the Spanish femi-t

nine for la pelicula. Omit choice (c)--substitute: Maria

bought this book yesterday; he is hers.

Item 5: a. Is near the school the church? b. Is the school near the church? c. Is the school the church near?

Item 5 was used to test word order. Near is the equivalent of

Spanish cerca de. Omit choice (c)—substitute: Is the school

near of'the church?

Item 7: a. Can you going with us tomorrow? b. Can you go with us tomorrow? c. Can you to go with us tomorrow?

Item 7 was used to test the Auxiliary plus or minus the to mark­

er. Omit all three choices.

Substitute: a. I arrive to school in the morning. b. I arrive s chool in the morning. c. I arrive at school in the morning.

Item 16: a. How are you feeling today? b. How are feeling today you? c. How today are you feeling?

Item 16 was used to test word order. Since the sequence is prob­

ably heard daily by the subjects, it is not useful in testing the

word order of questions. Omit all three choices .

Substitute: a. What means this word? b. What does this word mean? c. What this word means?

Items on which there were consistently low scores by all

subjects may be due to first language interference or their being

unlearned patterns. A possibility of determining which would be

a comparison with the native English-speaking group, as well as

the Production response in Part II. Item 29, for example, shows

Group 1 with a distribution of 80, 10, 10; Groups 2-6 with 36,

33, 31.

Item 29; a. There is not much news in the paper today. b. There are not many news in the paper today. c. There is not much of news in the paper today.

It should be noted that in their Production responses from Span­

ish to English for Item 29, 61 percent of Groups 2-6 combined

wrote "many news." It is possible that the problem is due to

first language interference (in Spanish—muchas noticias).

Items 23, 26, 28, and 30 presented various forms of the

tag question, with a comparatively mixed choice distribution in

the combined Groups 2-6, as well as in Group 1. Menyuk (1969,

47

p. 89) gave brief attention to tag questions in her study of the

development of transformation rules in English-speaking children

at kindergarten age:

In some instances the tag question was added, "She didn't spill on the floor, did she?" "He won't get dead, will he?" The usual form was negative statement-tag, rather than the inverse .... The frequent oc­currence of negative statement-tag question, indicates that although the structure of the sentence is that of a question, children are, at this stage, frequently telling you rather than asking you.

Earlier, Menyuk (1969, p. 70) noted the use of tag questions as a

form of the imperative:

There are no developmental changes that can be ob­served in the structure of imperative sentences per se. What does appear in the negative form [ "Don't touch my crayons'."] and, to a much lesser extent, with tag ques­tions [Put that there, will you'."]. Even with the tag question the utterance seems to be used to command rather than request.

The wide choice distribution of the tag question items indicates

that among the subjects of this study, the tag question remains

as yet, an unlearned pattern, causing difficulty among the non-

native English speakers. As indicated by Table 5, the choice

distribution of Item 24 also indicates an unlearned pattern for

both native and non-native speakers .

Item 24: a. We live in 1824 South Fifth Street. b. We live at 1824 South Fifth Street. c. We live on 1824 South Fifth Street.

Forty-five percent of Group 1 (Native English Speakers) chose

answer (b). The same is also true of the patterns in Item 25.

Item 25: a. Doesn't he have nothing for you? b. Doesn't he have something for you? c. Doesn't he have anything for you?

Forty-five percent of Group 1 (Native English Speakers) chose

answer (b).

Table 6 shows the results of the Analysis of Variance to

determine whether or not there are any differences among the six

groups in their ability to recognize correct grammatical se­

quences .

Table 6. MOVA: Recognition Results—Groups 1-6*

Source MS DF F-Ratio

Total 24.5868 109

Groups 109.6912 5 5.352

Error 20.4953 104

G Mean 1 2 3 4 5 6 G Mean 28.3500 24.2941 22.1053 24.1333 22.3158 26.1000

*Group 1 NES, 20 subjects Group 2 BBP(U), 17 subjects Group 3 BBP(U), 19 subjects Group 4 BNBP(U) , 15 subjects Group 5 BBP(B), 19 subjects Group 6 BNBP(B) , 20 subjects

Hypothesis 1 stated: There will be no differences be­

tween the native English-speaking group and the five Spanish-

speaking groups in their Recognition performances. Since the

obtained F in Table 6 is greater than the veilue of F required to

reject Hi, it is rejected. The data indicate that there are

significant differences among the six groups.

Table 7. Mean Differences: Results of Tukey Post Hoc Test

Group 3 22.1053

Group 5 22.3158

Group 4 24.1333

Group 2 24.2941

Group 6 26.1000

Group 1 28.3500

Group 3: BBP(U) X .2105 2.028 2.1888 3.9947 6 .2447*

Group 5: BBP(B) X 1.8175 1.9783 3.7842 6.0342*

Group 4: BNBP(U) X .1608 1.9667 4.2167

Group 2: BBP(U) X 1.8059 4.0554

Group 6: BNBP(B) X 2 .2500

Group 1: NES X

*p_$.05

50

ft post hoc Tukey Test was used to make multiple compari­

sons among the six means. This was done to determine which among

the six groups were significantly different from one another.

The results are shown in Table 7.

Table 7 indicates that the highest Mean of Group 1 dif­

fers significantly from those of Groups 3 and 5. It borders on

differing significantly from Groups 2 and 4.

Table 8 shows the results of the Analysis of Variance to

determine whether or not there are any differences among the five

native Spanish-speaking groups in their ability to produce cor­

rect English grammatical sequences .

Table 8. ANOVA: Production Results—Groups 2-6*

Source MS DF F-Ratio

Total 37.2710 89

Groups 62.7047 4 1.738

Error 36.0742 85

G Mean 2 3 4 5 6

G Mean 18.2353 20.0526 15.4000 18.9474 20.2500

*Group 2: 17 subjects Group 3: 19 subjects Group 4: 15 subjects Group 5: 19 subjects Group 6: 20 subjects

51

Hypothesis 2 stated: There will be no differences among

the five Spanish-speaking groups in their Production performances.

Since the obtained F as indicated in Table 8 is less than the

value of F required to reject H2j it is not rejected. The data

indicate that there are no significant differences among the five

bilingual groups in their abilities to produce grammatical se­

quences from Spanish to English.

Table 9 shows the results of the Analysis of Variance to

determine whether or not there are any differences among the five

native Spanish-speaking groups in their total test results.

Table 9. MOVA: Total Test Results—Groups 2-6

Source MS DF F-Ratio

Total 102.0938 89

Groups 135.3754 4 1.346

Error 100.5458 85

G Mean 2

41.5882 3

41.4737 4

38.5333 5 6

39.1000 45.3500

Hypothesis 3 stated: There will be no differences among

the five native Spanish-speaking groups in their total test per­

formances . Since the obtained F as indicated in Table 9 is less

than the value of F required to reject H3j it is not rejected.

The data indicate that there are no significant differences among

the five bilingual groups in their total test performances.

52

Table 10 shows the results of the Analysis of Variance to

determine whether or not there are any significant differences

between the Bilingual Program subjects and the Non-Bilingual

Program subjects in their ability to recognize correct grammati­

cal sequences.

Table 10. ANOVA: Recognition Results—Bilingual Program Sub­jects (Group 1)" vs. Non-Bilingual Program Subjects (Group 2)**

Source MS DF F-Ratio

Total 25.3145 89

Groups 123.4668 1 5.102

Error 24.1991 88

G Mean 1

22.8545 2

25.2571

*Group 1: 55 subjects, Schools 2-3-5 **Group 2: 35 subjects, Schools 4-6

Hypothesis 4 stated: There will be no differences be­

tween the three Bilingual Program groups and the two Non-Bilingual

Program groups in their Recognition performances. Since the ob­

tained F is greater than the value of F required to reject H4, it

is rejected. The data indicate that the Non-Bilingual Program

subjects performed higher on the Competence/Recognition scale.

Table 11 shows the results of the Analysis of Variance to

determine whether or not there are any significant differences

53

between the Bilingual Program subjects and the Non-Bilingual Pro­

gram subjects in their ability to produce correct grammatical

sequences.

Table 11. ANOVA; Production Results—Bilingual Program Subjects (Group 1) vs. Non-Bilingual Program Subjects (Group 2)

Source MS DF F-Ratio

Total 37.2710 89

Groups 18.8053 1 .502

Error 37.4809 88

G Mean 1*

19.1091 2**

18.1714

*Group 1: 55 subjects, Schools 2-3-5 **Group 2: 35 subjects, Schools 4-6

Hypothesis 5 stated: There will be no differences be­

tween the three Bilingual Program groups and the two Non-

Bilingual Program groups in their Production performances. Since

the obtained F is less than the value of F required to reject H5,

it is not rejected. The data indicate that there are no signifi­

cant differences between the Bilingual Program groups and the

Non-Bilingual Program groups in their abilities to produce gram­

matical sequences from Spanish to English.

Table 12 shows the results of the Analysis of Variance

to determine whether or not there are any significant differences

54

between the Urban Groups and the Border Groups in their ability

to recognize correct grammatical sequences.

Table 12. MOVA; Recognition Results—Urban Groups (2, 3, 4) vs. Border Groups (5, 6)

Source MS DF F-Ratio

Total 25.3145 89

Groups 15.0432 1 .592

Error 25.4312 88

G Mean 1*

23.4314 2**

24.2564

*Group 1: Urban, 51 subjects **Group 2: Border, 39 subjects

Hypothesis 6 stated: There will be no differences be­

tween the Urban Groups and the Border Groups in their Recognition

performances. Since the obtained F is less than the value of F

required to reject Hg, it is not rejected. The data indicate

that there are no significant differences between the three Urban

schools and the two Border schools in their abilities to recog­

nize English grammatical sequences .

Table 13 shows the results of the Analysis of Variance to

determine whether or not there are any significant differences

between the Urban Groups and the Border Groups in their ability

to produce correct grammatical sequences .

55

Table 13. ANOVA: Production Results—Urban Groups (2, 3, 4) vs. Border Groups (5, 6)

Source MS DF F-Ratio

Total 37.2710 89

Groups 52.2052 1 1.407

Error 37.1013 88

G Mean 1*

18.0784 2**

19.6154

*Group 1: Urban, 51 subjects **Group 2: Border, 39 subjects

Hypothesis 7 stated: There will be no differences be­

tween the Urban Groups and the Border Groups in their Production

performances. Since the obtained F is less than the value of F

required to reject H75 it is not rejected. The data indicate

that there are no significant differences between the three Urban

schools and the two Border schools in their abilities to produce

grammatical sequences from Spanish to English.

Table 14 shows the results of the Analysis of Variance to

determine whether or not there are any significant differences

between males and females in their ability to recognize correct

grammatical sequences.

56

Table 14. Recognition Results—Male vs. Female

Source MS DF F-Ratio

Total 25.3145 89

Groups .7627 1 .030

Error 25.8750 88

G Mean 1* 23.8750

2** 23.6905

*Group 1: Male, 48 subjects **Group 2: Female, 42 subjects

Hypothesis 8 stated: There will be no differences be­

tween males and females in their Recognition performances. Since

the obtained F is less than the value of F required to reject Hg>

it is not rejected. The data indicate that there are no signifi­

cant differences between males and females in their abilities to

recognize correct English grammatical sequences .

Table 15 shows the results of the Analysis of Variance to

determine whether or not there are any significant differences

between males and females in their ability to produce correct

grammatical sequences. '

Hypothesis 9 stated: There will be no differences be­

tween males and females in their Production performances. Since

the obtained F is less than the value of F required to reject Hg,

it is not rejected. The data indicate that there are no

57

significant differences between males and females in their abili­

ties to produce grammatical sequences from Spanish to English.

Table 15 . MOV A: Production Results—Male vs . Female

Source MS DF F-Ratio

Total 37.2710 89

Groups 101.7175 1 2.784

Error 36.5387 88

G Mean 1*

17.7500 2**

19.8810

*Group 1: Male, 48 subjects **Group 2: Female, 42 subjects

Table 16 shows the results of the intercorrelation analy­

sis for Groups 2-6 to determine whether or not there is a posi­

tive correlation between the subjects' ability to recognize

correct grammatical sequences and their ability to produce cor­

rect grammatical sequences.

Hypothesis 10 stated: There will, be no correlation be­

tween the subjects' Recognition and Production performances. The

obtained r of .7567 is significant at the .05 level of confi­

dence . H10 is rejected .

58

Table 16. Intercorrelation Analysis—Recognition and Production Groups 2-6

1: Recognition 2: Production

Means 23.7889 18.7444

Sigmas 5.0033 6.0710

r 1 1.0000 .7567

2 .7567 1.0000

Summary

The results of this study showed the following:

1. There were no significant differences between the native

English-speaking group and the five Spanish-speaking

groups in their Recognition performances.

2. There were no significant differences among the five

bilingual groups in their abilities to produce grammati­

cal sequences from Spanish to English.

3. There were no significant differences among the five bi­

lingual groups in their total test performances.

4. Non-Bilingual Program subjects performed higher on the

Recognition scale.

5. There were no significant differences between the Bilin­

gual Program groups and the Non-Bilingual Program groups

in their abilities to produce grammatical sequences from

Spanish to English.

There were no significant differences between the three

Urban schools and the two Border schools in their abili­

ties to recognize English grammatical sequences.

There were no significant differences between the Urban

groups and the Border groups in their abilities to pro­

duce grammatical sequences from Spanish to English.

There were no significant differences between males and

females in their abilities to recognize correct English

grammatical sequences .

There were no significant differences between males and

females in their abilities to produce grammatical se­

quences from Spanish to English.

There was a positive correlation between the subjects'

Recognition and Production performances.

CHAPTER 5

SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS

The purpose of this chapter is to summarize the findings

of this study, to draw conclusions based on the findings, and to

make recommendations pertinent to the problem of bilingual pro­

grams, especially the current means of evaluating them.

Summary

This project was undertaken to provide an effective in­

strument to measure the Production and Recognition of grammatical

and ungrammatical English word sequences by bilingual Spanish-

English children. The initial task was to construct a reliable

test of basic syntactic structures with which mature native

speakers of English have little, if any, trouble, and which would

determine the variance among Spanish-speaking fourth graders in

bilingual schools between the production and recognition of these

basic sentence types at this stage of their language development.

Results indicated that native English speakers perform better in

their recognition of correct sequences. Differences between

groups of children in bilingual programs and non-bilingual pro­

grams whose native and home language is Spanish were minimal.

Patterns such as the tag question house number and street address ,

60

subject + p-group + verb proved difficult for native speakers, as

well as bilinguals. These may be classified as unlearned pat­

terns, and, in the case of subject/verb agreement (one of the

boys is rather than one of the boys are), as a matter of usage

rather than syntax. This is an error with which instructors of

college freshmen are still confronted.

Part II of the test, Production, required the ninety

subjects to translate the thirty-four test items from Spanish to

English. This was the first time in their school experience

they were given a task of this type. After receiving instruc­

tions from the researcher and the classroom teacher, they pro­

ceeded with a minimum of questions which concerned unfamiliar

vocabulary items. (For example, there were questions from every

group about the meaning of helado [ ice cream] and divertimos

Iwe enjoy ourselves3.)

Two basic criteria were used to evaluate the translations

correct word order and tense and plural markers. Many of the re­

sponses were comparatively free in their translations:

Cue: En el verano hace calor en Arizona, i verdad? Response: Arizona's hot in the summer, don't you think?

Cue: Yo no quiero mas leche; ya tengo. Response: I don't want any more milk; I've had enough.

t Cue: iComo le gusto el cuento? Response: What did you think of the story?

Cue: Nos divertimos en la fiesta. Response: We had a lot of fun at the party.

Frequent error-types are demonstrated by the following:

Cue: Esta resfriada ella? Response: Does she has a cold?

Cue: Favor de leer esta carta. Response: Please to read this letter.

Cue: No puede el correr rapido? Response: Does he can't run fast?

Comparisons of all groups between Recognition and Pro­

duction performances revealed a positive correlation of .7567

and that some error would be made in prediction.

Statistically verified results showed:

1. The native English-speaking groups differed from the

five Spanish-speaking groups in their Recognition per­

formances .

2. There were no significant differences among the five

Spanish-speaking groups in their Production performances.

3. There were no significant differences among the five

Spanish-speaking groups in their total test performances.

4. There were significant differences between the three

bilingual program groups and the two non-bilingual groups

in their Recognition performances at .05 level of confi­

dence .

5. There were no significant differences between the urban

groups and the border groups in their Production perform­

ances .

63

6. There were no significant differences between the urban

groups and the border groups in their Recognition per­

formances .

7. There were no significant differences between sexes.

8. There was a positive correlation between the subjects'

Recognition and Production performances .

Conclusions

The major conclusions from this study are the following:

1. The instrument developed to measure the English Produc­

tion and Recognition performances had acceptable relia­

bility and validity; however, revisions such as those

suggested in Chapter 4 should be made.

2. Native English-speaking children had less difficulty in

identifying the correct basic sentence types presented

then the bilingual groups who showed some evidence of

first language interference.

3. The five native Spanish-speaking groups, whether in

structured bilingual programs or not, showed no signifi­

cant differences in their Production performances.

4. The scores of the non-bilingual program groups indicated

an advantage over the bilingual groups in the ability to

identify correct English grammatical sequences.

5. In considering the causes for the above, the extensive

use of Spanish initially and the relatively late formal

64

introduction to the English language in the bilingual

programs are important factors.

Recommendations

In view of the limitations of this study, which was pri­

marily concerned with English proficiency among bilingual chil­

dren and not with the equal development of two languages, the

following are recommended:

1. The instrument developed in this study should be used

with other bilingual groups to test and possibly increase

its reliability and validity.

2. Since problems with basic grammatical English sequences

among bilingual children can be identified, similar tests

should be developed, expanded, and administered periodi­

cally to determine competence in basic English sentence

types. This data will be useful in the continuing effort

to improve the methodology in the teaching of English as

a Second Language.

3. If balanced bilingualism is the goal of bilingual pro­

grams, educators in bilingual schools should consistently

compare the language development of their pupils with

that of native speakers. In so doing, they can develop

a realistic scale of measuring native-like proficiency.

Fluency in two or more languages is a major factor in

enabling a person to live and function in more than one culture.

65

It can help to free him from limitations imposed upon him by be­

longing to a single cultural and linguistic group. Making two

languages part of tha educational process of children involves

many complexities; therefore, further recommendations stated here

must go beyond the limited scope of this study.

Linguistic habits take time to develop. One of the

strong arguments for bilingual education in elementary schools

is that a series of progressive skills can be organized according

to the best available methodology which should lead to competence

in the two languages (Stern 1967, p. 9). However, three impor­

tant factors must be considered to determine in which direction

individual bilingual programs can and should go:

1. Since conditions, aims, and needs vary greatly in differ­

ent communities, it is impossible to derive from any one

bilingual program all of the procedures which would be

universally or nationally applicable.

2. ail teachers in bilingual.schools should have the oppor­

tunity to participate in preservice and in-service pro­

grams which provide training to meet the following

requirements (Saville and Troike 1970, p. 26):

a. a willingness to participate in an innovative pro­

gram.

b. a knowledge of the structures of both languages of

instruction.

66

c. a general understanding of the nature of language,

including the acceptability and inevitability of

dialect variations.

d. a specific understanding of one's own dialect and

the dialect of the area in which he teaches.

e. a knowledge of methods for teaching a second language.

f. an understanding and acceptance of all cultures rep­

resented in the community.

g. a knowledge of the growth and development patterns

of children from different cultures.

h. the competence to provide a good linguistic model,

preferably in both languages.

3. Current methods of evaluation such as that attempted in

this study should be supported by follow-up studies or

by more evidence from a sufficient amount of adequately

controlled research. The need for further investigations

of all kinds is recognized (Stern 1967, p. 79).

The above supports the view that in research on the dif­

ferences between languages there is a need to develop indices

for all aspects of bilingual education to determine the absolute

and relative difficulties of the native and second languages.

Further measurements of Production and Recognition in both lan­

guages should be made for all children at different ages and

different degrees of language aptitude. Concerning the area of

language testing, Saville and Troike (1970, p. 64) stated:

67

A great deal of work still remains to be done in the construction and validation of tests of bilingual educa­tion programs. As such tests are being developed and made available, necessary checks on the progress of chil­dren and programs are going to come from the combined ob­servation of experienced and knowledgeable teachers, coordinators, administrators, and linguists.

Such studies would involve a great deal of effort; nevertheless,

it is only through these studies that satisfactory knowledge can

be gained concerning bilingual education and the amount and type

of investment it requires. More research needs to be made in in­

nate language acquisition and its relationship to nonstandard

language patterns. Robinson (in Shane, Walden, and Green 1971,

p. 12) explained further that:

. . . clearer distinctions need to be drawn between language competence and performance. In the sequence of language development ... we need to ascertain the significance of the variability with which usage skills are acquired and also to explore more fully the influ­ence of the child's membership in a given subculture in relation to his language development.

The schools' future posture with respect to bilingual programs

needs to be determined and altered as greater agreement is

reached on the virtue of developing two languages . In addition,

careful consideration should be given to the changes in proce­

dures and teaching and testing techniques such a bilingual policy

would require.

APPENDIX A

EXPLANATION AND INSTRUCTION SHEET

Fourth Grade Teachers

Frank Pialorsi

English Word Sequences Test

The Test

Part I of this test is designed to test the subjects', ability to recognize correct English word sequences . The test items are based on a contrastive analysis of English and Spanish grammatical sequences.

Part II requires the subjects to shift from their native language (Spanish) to their second language (English). The cor­rect English answers should closely resemble the thirty-four correct answers to Part I.

Test Procedures

Part I

1. Be sure each subject has filled out the Personal Data Sheet.

2. Read each example and emphasize the correct answer.

3. Allow thirty minutes for Part I of the test.

4. If there are any questions about vocabulary items, we may read the item to the subjects.

Part II

1. There will be at least a one-week time lapse between Part I and Part II of the test.

68

TO:

FROM:

RE:

2. Read each example and allow time for the subjects to consider the English answer.

3. Allow forty-five minutes for Part II of the test.

4. Since this is a structure test, we may answer any questions about Spanish vocabulary.

I will be present at the time of testing. However, I would like to maintain normal classroom procedures as much as possible and simply observe.

APPENDIX B

PERSONAL DATA SHEET

Name of Student ID.#

Name of School

I have been a pupil at this school for years.

Sex

Date of Birth Place of Birth

Father's Place of Birth

Mother's Place of Birth

Home Language: Check the statements which are true.

1. I speak mostly Spanish away from school.

2. I speak mostly Spanish with my father.

3. I speak mostly Spanish with my mother.

4. I speak mostly Spanish with my brothers and sisters.

5. I speak mostly Spanish with my neighbors .

6. Both of my parents speak to me both in Spanish and English.

7. One parent speaks to me in Spanish and the other speaks to me in English.

8. I speak mostly English away from school.

70

APPENDIX C

TEST DESIGNS

Part I

\l/

LANGUAGE B No Grammatical Interference

LANGUAGE A Grammatical.. Interference

RESPONSE: LANGUAGE B

UNLEARNED PATTERN

CUE: LANGUAGE B

LANGUAGE A: Spanish LANGUAGE B: English

71

72

Part II

CUE: LANGUAGE A

COMPOUNDED BALANCED UNLEARNED PATTERNS

RESPONSE: LANGUAGE B

APPENDIX D

TEST OF THE PRODUCTION MD RECOGNITION OF ENGLISH GRAMMATICAL SEQUENCES

PART I: RECOGNITION

English Practice

Directions:

Example 1:

Example 2:

Place a checkmark next to the sentence that you think is correct.

/ He is my father. She is my father. It is my father.

She like her teacher. She likes her teacher. She liking her teacher.

1. Tom has a bicycle but wants other. Tom has a bicycle but wants one. Tom has a bicycle but wants another.

2. Maria bought this book yesterday; it is his . Maria bought this book yesterday: it is hers. Maria bought this book yesterday; it is its.

3. This pencil is my sister's. This pencil is of my sister. This pencil is to my sister.

4. Mr. Sanchez comes to school in a bus. Mr. Sanchez comes to school by a bus. Mr. Sanchez come to school on bus.

5.

6.

Is near the school the church? Is the school near the church? Is""the school the church near?

After to study I play with my friends. After studying I play with my friends . After the study I play with my friends,

73

Can you going with us tomorrow? Can you go with us tomorrow? Can you to go with us tomorrow?

The dog wants something to eat. The dog wants something for to eat. The dog wants something for eat.

The girl I saw was happy. The girl what I saw was happy. The girl which I saw was happy.

We had a good time at the party. We made a good time at the party. We did a good time at the party.

Mary eats much of ice cream. Mary eats a lot of ice cream. Mary eats many ice cream.

She wishes she were pretty. She wishes she is pretty. She wishes being pretty.

Please to read this letter. Please read this letter. Please reading this letter.

How liked he the story? How he liked the story? How did he like the story?

Does he can't run fast? Can't he run fast? Does he can run fast?

How are you feeling today? How are feeling today you? How today are you feeling?

I very well speak English. I speak English very well. I speak very well English.

We arrived there late yesterday. We arrived late yesterday there. We arrived yesterday there late.

One of the boys am not at school today. One of the boys are not at school today. One of the boys is not at school today.

20

21

22

23

24

25

26

27

28

29,

30,

31.

75

Columbus has discovered America in 1492. Columbus discovered America in-1492. Columbus has been discovering America in 1492.

Does she have a cold? Is she have a cold? Has she have a cold?

Robert is a gooder swimmer than James. Robert is a more good swimmer than James. Robert is a better swimmer than James.

Arizona is hot in the summer, not true? Arizona is hot in the summer, is it? Arizona is hot in the summer, isn't it?

We live in 1824 South Fifth Street. We live at 1824 South Fifth Street. We live on 1824 South Fifth Street.

Doesn't he have nothing for you? Doesn't have he something for you? Doesn't he have anything for you?

The school nurse comes tomorrow, won't she? The school nurse comes tomorrow, doesn't she? The school nurse comes tomorrow, isn't she?

I don't want more milk; I have some. I don't want more milk; I have any. I don't want more milk; I have other.

Mr. Gonzalez isn't at school today, does he? Mr. Gonzalez isn't at school today, is he? Mr. Gonzalez isn't at school today, isn't he?

There is not much news in the paper today. There is not many news in the paper today. There is not much of news in the paper today.

She hasn't been sick, did she? She hasn't been sick, was she? She hasn't been sick, has she?

He talked to each other. They talked to each other. He talked to other.

Dolores likes school and Juanita likes too. Dolores likes school and Juanita does too, Dolores likes school and Juanita is too.

We went to school during a year. We went to school for a year. We went to school while a year.

I wish that I had a new bicycle. I want that I had a new bicycle. I like that I had a new bicycle.

APPENDIX E

TEST OF THE PRODUCTION AND RECOGNITION OF ENGLISH GRAMMATICAL SEQUENCES

PART II: PRODUCTION

Say It In English

Directions: You are helping someone who doesn't understand Spanish. How would you say the following sen­tences in English?

Example 1: Me llamo Pancho. My name is Pancho.

Example 2: Les hablo en ingles. I speak to them in English.

Example 3: Juan y el son mexicanos, verdad? (DO NOT USE TRUE) Juan and he are Mexicans, aren't they?

1. Tomas tiene una bicicleta pero el quiere otra.

2. Maria compro este libro ayer; es suyo.

3. Este lapiz es de mi hermana.

4. El senor Sanchez viene a la escuela en el bus.

5. £ Esta la escuela cerca de la iglesia?

6. Despues de estudiar, juego con mis amigos.

7. ' iPuedes ir con nosotros manana?

77

8 . El perro quiere algo para comer.

78

9. La muchacha que yo vi estaba contenta.

10. Nos divertimos en la fiesta.

11. Maria come muchohelado (ice cream).

12. Ella quisiera ser bonita.

13. Favor de leer esta carta.

14. £Como le gusto el cuento (story)?

15. £No puede el correr rapido?

16. oComo te sientes hoy?

17. Yo hablo muy bien el ingles.

18. Ayer nosotros llegamos alii tarde.

19. Uno de los muchachos no esta en la escuela hoy.

20. Colon descubrio America en 1492.

21. iEsta resfriada ella?

22. Roberto es mejor nadador (swimmer) que Jaime.

23. En el verano hace calor en Arizona, verdad?

24. Vivimos en la calle Douglas sur, numero 1824.

25

26

27

28

29

30

31

32

33

34

79

iNo tiene el nada para ti?

La enfermera de la escuela viene manana, iverdad?

Yo no quiero mas leche: ya tengo.

El senor Gonzalez no esta en la escuela hoy, iverdad?

No hay muchas noticias en el periodico hoy.

Ella no ha estado enferma, iverdad?

Ellos se hablaron.

A Dolores le gusta la escuela y a Juanita tambien.

Nosotros fuimos a la escuela por un ano.

Ojala que yo tuviera una bicicleta nueva.

APPENDIX F

TEST DATA: GROUP 1 NATIVE ENGLISH SPEAKERS (NES)

Reliability Coefficient

RECOGNITION

Item N 34.0000

Mean 28.3500

Sigma 2.0803

Alpha .1774

Item Analysis

ITEM MEAN SIGMA R( TOTAL)

1 .90 .300 .2964 2 1.00 0.000 I 3 .90 .300 .0561 4 .60 .490 .2355 5 1.00 0.000 I 6 .90 .300 .1362 7 1.00 0.000 I 8 1.00 0.000 I 9 .70 .458 .4773 10 1.00 0.000 I 11 1.00 0.000 I 12 .65 .477 .4762-13 1.00 0.000 I 14 .75 .433 .5967 15 .90 .300 -.2644 16 1.00 0.000 I 17 .95 .218 .3694 18 .90 .300 .2964 19 .45 .497 .1860 20 .80 .400 .3245

80

ITEM MEM SIGMA R( TOTAL)

21 1.00 0.000 I 22 .95 .218 .3694 23 .50 .500 .0240 24 .45 .497 -.0072 25 .55 .497 .0556 26 .75 .433 -.2359 27 1.00 0.000 I 28 .45 .497 .4276 29 .80 .400 .3846 30 .90 .300 -.1042 31 1.00 0.000 I 32 .80 .400 .5047 33 .85 .357 .5419 34 .95 .218 .2592

Choice Distribution (Percentages)

ITEM KEY 1 2 3

1 3 0 10 90 2 2 0 100 0 3 1 90 0 10 4 1 60 15 25 5 2 0 100 0 6 2 0 90 10 7 2 0 100 0 8 1 100 0 0 9 1 70 5 25 10 1 100 0 0 11 2 0 100 0 12 1 65 25 10 13 2 0 100 0 14 3 5 20 75 15 2 0 90 10 16 1 100 0 0 17 2 0 95 5 18 1 90 5 5 19 3 0 55 45 20 2 15 80 5 21 1 100 0 0 22 3 0 5 95 23 3 5 45 50 24 2 0 45 55 25 3 0 45 55 26 2 5 75 20

ITEM KEY 1 2 3

27 1 100 0 0 28 2 0 45 55 29 1 80 10 10 30 3 0 10 90 31 2 0 100 0 32 2 10 80 10 33 2 15 85 0 34 1 95 5 0

APPENDIX G

TEST DATA: GROUP 2 BILINGUALS/BILINGUAL PROGRAM (URBAN)

Reliability Coefficient

RECOGNITION PRODUCTION TOTAL TEST

Item N 34.0000 34.0000 68.0000

Mean 24.2941 18.2353 42.5294

Sigma 4.3490 4.4791 6.5898

Alpha .7227 .7012 .7332

Item Analysis: Part : I

ITEM SCALE MEAN SIGMA R( TOTAL) R( SCALE)

1 1 .53 .499 .2546 .3889 2 1 1.00 0.000 I I 3 1 .76 .424 - .0396 .1969 4 1 .76 .424 -.3342 -.0900-5 1 .94 .235 .1718 -.0981 6 1 .76 .424 .4865 .1969 7 1 .82 .381 .0606 .3152 8 1 .94 .235 .3995 .4193 9 1 .59 .492 .1035 .2764 10 1 1.00 0.000 I I 11 1 .88 .322 .3064 .2766 12 1 .65 .478 .0593 .2481 13 1 .88 .322 .2787 .402 5 14 1 .76 .424 .3392 .3883 15 1 1.00 0.000 I I 16 1 .88 .322 .2233 .4445 17 1 .76 .424 .1077 .2288 18 1 .47 .499 .4965 .3969 19 1 .59 .492 .2486 .3589 20 1 .53 .499 .2188 .0909 21 1 .88 .322 .5557 .6964

83

22 23 24 25 26 27 28 29 30 31 32 33 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

84

SCALE MEM SIGMA R( TOTAL) R( SCALE)

1 .82 .381 .3650 .5635 1 .53 .499 .4692 .4431 1 .41 .492 .2230 .2732 1 .35 .478 .3890 .4029 1 .47 .499 -.2546 -.2535 1 .88 .322 .5557 .6964 1 .41 .492 .1504 .4106 1 .35 .478 .0154 .5444 1 .65 .478 .5637 .5311 1 .82 .381 .4821 .5635 1 .71 .456 .0519 .3702 1 .59 .492 .4662 .2490 1 .88 .322 .5557 .6964

Item Analysis: Part II

2 .82 .381 .4353 .4377 2 .53 .499 .2725 .4968 2 .59 .492 - .1142 - .1962 2 .76 .424 .2971 .4007 2 .65 .478 -.2395 - .0986 2 , .59 .492 .2305 .4976 2 .76 .424 .7180 .6483 2 .94 .235 .4753 .4038 2 .82 .381 .0372 .2655 2 .53 .499 .5586 .7073 2 .88 .322 -.0538 - .1439 2 .59 .492 .4662 .5243 2 .82 .381 .0606 .2999 2 .59 ..492 .3393 -.0094 2 .35 .478 .2582 .1536 2 1.00 0.000 I I 2 .41 .492 .3137 .3029 2 .65 .478 .4703 .5335 2 .35 .478 .1648 .2085 2 .65 .478 .8439 .6983 2 .41 .492 .1142 - .0173 2 .47 .499 .6217 .5030 2 .53 .499 .3619 .4442 2 .18 .381 .2672 .2857 2 0.00 0.000 I I 2 .24 .424 .1448 .2495 2 .41 .492 -.0854 .0895 2 .24 .424 .2921 .4043 2 .47 .499 .2998 .6083 2 .18 .381 -.0138 .0446

85

ITEM SCALE MEM SIGMA R( TOTAL) R(SCALE)

31 2 .41 .492 -.1035 .3563 32 2 .06 .235 .2834 .3776 33 2 .59 .492 -.1504 .2841 34 2 .76 .424 .0867 .2768

Choice Distribution (Percentages); Part I

ITEM KEY 1 2 3

1 3 18 29 53 2 2 0 100 0 3 1 76 12 12 4 1 76 6 18 5 2 6 94 0 6 2 12 76 12 7 2 0 82 18 8 1 94 6 0 9 1 59 6 35 10 1 100 0 0 11 2 6 88 6 12 1 65 24 12 13 2 12 88 0 14 3 0 24 76 15 2 0 100 0 16 1 88 6 6 17 2 12 76 12 18 1 47 29 24 19 3 0 41 59 20 2 24 53 24 21 1 88 0 12 22 3 6 12 82 23 3 12 35 53 24 2 12 41 47 25 3 6 59 35 26 2 18 47 35 27 1 88 6 6 28 2 6 41 53 29 1 35 35 29 30 3 24 12 65 31 2 6 82 12 32 2 18 71 12 33 2 41 59 0 34 1 88 6 6

Choice Distribution (Percentages); Part II

86

ITEM

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

1 2

82 18 53 47 59 41 76 24 65 35 59 41 76 24 94 6 82 IP 53 4, 88 12 59 41 82 18 59 41 35 65 100 0 41 59 65 35 35 65 65 35 41 59 47 53 53 47 18 82 0 100 24 76 41 59 24 76 47 53 18 82 41 59 6 94 59 41 76 24

KEY

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

APPENDIX H

TEST DATA: GROUP 3 BILINGUALS/BILINGUAL PROGRAM (URBAN)

Reliability Coefficient

RECOGNITION PRODUCTION TOTAL TEST

Item N 34.0000 34.0000 68.0000

Mean 22 .1053 20.0526 42.1579

Sigma 5.7756 6.2784 11.4030

Alpha .8272 .8568 .9118

Item Analysis: Part I

ITEM SCALE MEAN SIGMA R( TOTAL) R( SCALE)

1 1 .84 .365 -.2092 - .1920 2 1 .89 .307 - .0103 .0953 3 1 .74 .440 .3122 .2799 4 1 .53 .499 .0963 .0355 5 1 .74 .440 .4590 .4248 6 1 .79 .408 .7657 .8141 7 1 .74 .440 .5324 .4869 8 1 .84 .365 .6136 .6077 9 1 .47 .499 -.0316 .0010 10 1 .84 .365 .5250 .5327 11 1 .79 .408 .1430 .2776 12 1 .32 .465 .2090 .2229 13 1 .84 .365 .6389 .6826 14 1 .68 .465 .2874 .3065 15 1 .74 .440 .6791 .6731 16 1 .95 .223 - .1621 -.1589 17 1 .84 .365 .6136 .6077 18 1 .53 .499 .6325 .6561 19 1 .63 .482 .3455 .4106 20 1 .68 .465 .1881 .1888 21 1 .79 .408 .6412 .6576

87

88

ITEM SCALE MEM SIGMA R( TOTAL) R( SCALE)

22 1 .68 .465 .7541 .8358 23 1 .37 .482 .3243 .4395 24 1 .32 .465 .1793 .1445 25 1 .26 .440 -.0292 -.0523 26 1 .63 .482 -.0181 .0895 27 1 .74 .440 .6372 .6524 28 1 .42 .494 .6145 .5013 29 1 .42 .494 .0162 - .0709 30 1 .53 .499 .5123 .6196 31 1 .63 .482 .6517 .7696 32 1 .84 .365 .2718 .3078 33 1 .42 .494 .6052 .6120 34 1 .63 .482 .4412 .5618

Item Analysis: Part II

1 2 .74 .440 .0921 .1002 2 2 .53 .499 .4661 .5620 3 2 .63 .482 .5943 .5451 4 2 .79 .408 .2675 .1483 5 2 .74 .440 .3018 .3477 6 2 .74 .440 .2074 .2525 7 2 .95 .223 - .0174 .1897 8 2 .95 .223 -.0174 .1897 9 2 .74 .440 .5428 .6142 10 2 .58 .494 .3951 .4826 11 2 .79 .408 .6072 .5595 12 2 .95 .223 .3960 .3399 13 2 .89 .307 -.3111 - .4069 14 2 .47 .499 .5138 .5629 15 2 .16 .365 - .2592 -.3715 16 2 .79 .408 .7770 .6212 17 2 .63 .482 .4125 .4061 18 2 .63 .482 .4029 .4235 19 2 .74 .440 .7315 .6523 20 2 .58 .494 .4044 .3467 21 2 .37 .482 .3243 .4281 22 2 .47 .499 .6987 .6804 23 2 .37 .482 .3626 .4454 24 2 .16 .365 .4244 .4562 25 2 .32 .465 .0601 .1386 26 2 .42 .494 .4182 .4343 27 2 .53 .499 .4106 .3941 28 2 .42 .494 .6800 .7739 29 2 .58 .494 .5821 .6184 30 2 .21 .408 .5816 .6948

31 32 33 34

ITE1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

89

SCALE MEM SIGMA R( TOTAL) R( SCALE)

2 2 2 2

.53

.32 .68 .68

.499

.465

.465

.465

.5863

.6956 .2080 .3569

.6292

.7698

.2221

.4025

Choice Distribution (Percentages): Part I

KEY 1 2 3

3 0 16 84 2 11 89 0 1 74 5 21 1 53 11 37 2 11 74 16 2 5 79 16 2 16 74 11 1 84 0 16 1 47 16 37 1 84 11 5 2 5 79 16 1 32 32 37 2 5 84 11 3 5 26 68 2 5 74 21 1 95 0 5 2 0 84 16 1 53 11 37 3 5 32 63 2 26 68 5 1 79 5 16 3 5 26 68 3 21 42 37 2 5 32 63 3 16 58 26 2 11 63 26 1 74 21 5 2 21 42 37 1 42 21 37 3 11 37 53 2 21 63 16 2 5 84 11 2 26 42 32 1 63 21 16

Choice Distribution (Percentages): Part II

90

ITEM KEY 1

1 1 74 2 1 53 3 1 63 4 1 79 5 1 74 6 1 74 7 1 95 8 1 95 9 1 74 10 1 58 11 1 79 12 1 95 13 1 89 14 1 47 15 1 16 16 1 79 17 1 63 18 1 63 19 1 74 20 1 58 21 1 37 22 1 47 23 1 37 24 1 16 25 1 32 26 1 42 27 1 53 28 1 42 29 1 58 30 1 21 31 1 53 32 1 32 33 1 68 34 1 68

2

26 47 37 21 26 26 5 5

26 42 21 5 11 53 84 21 37 37 26 42 63 53 63 84 68 58 47 58 42 79 47 68 32 32

APPENDIX I

TEST DATA: GROUP 4 BILINGUALS/NON-BILINGUAL PROGRAM (URBAN)

Reliability Coefficient

RECOGNITION PRODUCTION TOTAL TEST

Item N 34.0000 34.0000 68.0000

Mean 24.1333 15.4000 39.5333

Sigma 4.0639 4.5578 6.8592

Alpha .7120 .7473 .7818

Item Analysis: Part : I

ITEM SCALE MEAN SIGMA R( TOTAL) R(SCALE)

1 1 .80 .400 .3548 .3035 2 1 .93 .249 -.0571 -.1885 3 1 .80 .400 .6220 .7956 4 1 .73 .442 .2667 .4649 5 1 1.00 0.000 I I 6 1 .73 .442 .0908 .3907 7 1 1.00 0.000 I I 8 1 1.00 0.000 I I 9 1 .73 .442 .5084 .6133 10 1 .93 .249 .6442 .6664 11 1 .80 .400 .0632 .3855 12 1 .20 .400 .3013 .2707 13 1 .87 .340 .4594 .3989 14 1 .73 .442 .0689 .2053 15 1 .93 .249 .4883 .6006 16 1 .87 .340 .8311 .9298 17 1 1.00 0.000 I I 18 1 .60 .490 - .0556 .1607 19 1 .40 .490 .1746 -.0938 20 1 .27 .442 .0190 - .0569 21 1 .73 .442 .3326 .7246

91

92

ITEM SCALE MEM SIGMA R( TOTAL) R( SCALE)

22 1 .87 .340 - .1411 - .2767 23 1 .47 .499 .4533 .1337 24 1 .13 .340 -.0591 .3732 25 1 .67 .471 .0756 .1624 26 1 .60 .490 .3016 .3951 27 1 1.00 0.000 I I 28 1 .40 .490 -.2222 .1406 29 1 .33 .471 .1512 .3944 30 1 .73 .442 .4645 .5020 31 1 .80 .400 .2576 .3855 32 1 .73 .442 .5744 .5020 33 1 .60 .490 .1627 .1272 34 1 .73 .442 .6843 .6504

Item Analysis: Part II

1 2 .60 .490 .7777 .7285 2 2 .47 .499 .5897 .5629 3 2 .33 .471 .0687 .3103 4 2 .33 .471 .5842 .7447 5 2 .47 .499 -.0727 -.1701 6 2 .67 .471 .6529 .5275 7 2 1.00 0.000 I I 8 2 .93 .249 .0987 .3753 9 2 .93 .249 .1377 .2580 10 2 .40 .490 -.0238 - .0418 11 2 .73 .442 .7722 .6814 12 2 .73 .442 .3985 .4829 13 2 .80 .400 .3548 .4096 14 2 .67 .471 .1375 .1862 15 2 .13 .340 .1411 .0086 16 2 .87 .340 -.1411 - .0947 17 2 .47 .499 .4533 .4163 18 2 .53 .499 .3650 .5512 19 2 .53 .499 -.3169 -.0352 20 2 .33 .471 .1924 .4034 21 2 .27 .442 .4366 .5425 22 2 .40 .490 .6904 .8838 23 2 .20 .400 .3742 .4681 24 2 0.00 0.000 I I 25 2 .07 .249 -.1377 -.2580 26 2 .07 .249 .2130 .2111 27 2 .40 .490 -.4603 -.2209 28 2 .07 .249 .0961 -.0235 29 2 .33 .471 .5429 .6826 30 2 .07 .249 -.1377 -.2580

93

ITEM SCALE MEAN SIGMA R( TOTAL) R(SCALE)

31 2 .20 .400 .3985 .4681 32 2 .07 .249 .2520 .4457 33 2 .60 .490 .6190 .5494 34 2 .73 .442 .2887 .3837

Choice Distribution (Percentages): Part I

ITEM KEY 1 2 3

1 3 13 7 80 2 2 7 93 0 3 1 80 7 13 4 1 73 13 13 5 2 0 100 0 6 2 7 73 20 7 2 0 100 0 8 1 100 0 0 9 1 73 13 13 10 1 93 0 7 11 2 0 80 20 12 1 20 13 67 13 2 7 87 7 14 3 0 27 73 15 2 7 93 0 16 1 87 0 13 17 2 0 100 0 18 1 60 13 27 19 3 7 53 40 20 2 53 27 20 21 1 73 0 27 22 3 13 0 87 23 3 0 53 47 24 2 13 13 73 25 3 7 27 67 26 2 13 60 27 27 1 100 0 0 28 2 13 40 47 29 1 33 33 33 30 3 7 20 73 31 2 13 80 7 32 2 27 73 0 33 2 33 60 7 34 1 73 13 13

Choice Distribution (Percentages): Part II

ITEM KEY 1 2

1 1 60 40 2 1 47 53 3 1 33 67 4 1 33 67 5 1 47 53 6 1 67 33 7 1 100 0 8 1 93 7 9 1 93 7 10 1 40 60 11 1 73 27 12 1 73 27 13 1 80 20 14 1 67 33 15 1 13 87 16 1 87 13 17 1 47 53 18 1 53 47 19 1 53 47 20 1 33 67 21 1 27 73 22 1 40 60 23 1 20 80 24 1 0 100 25 1 7 93 26 1 7 93 27 1 40 60 28 1 7 93 29 1 33 67 30 1 7 93 31 1 20 80 32 1 7 93 33 1 60 40 34 1 73 27

APPENDIX J

TEST DATA: GROUP 5 BILINGUALS/BILINGUAL PROGRAM (BORDER)

Reliability Coefficient

RECOGNITION PRODUCTION TOTAL TEST

Item N 34.0000 34.0000 68.0000

Mean 22.3158 18.9474 41.2632

Sigma 5 .2219 6.6765 11.1678

Alpha .7913 .8919 .9147

Item Analysis: Part I

ITEM SCALE MEAN SIGMA R( TOTAL) R( SCALE)

1 1 .58 .494 .5260 .6232 2 1 .89 .307 .4227 .3163 3 1 .63 .482 .4870 .5477 4 1 .58 .494 .4783 .5415 5 1 .68 .465 .2391 .2362 6 1 .74 .440 .3566 .3795 7 1 .89 .307 .4074 .4805 8 1 .89 .307 .3152 .4477 9 1 .63 .482 .3404 .4641 10 1 .89 .307 .2231 .2506 11 1 .58 .494 .3828 .4190 12 1 .58 .494 .4210 .5007 13 1 .84 .365 .4755 .4684 14 1 .79 .408 -.2537 - .0924 15 1 .84 .365 .5143 .4408 16 1 1.00 0.000 I I 17 1 .79 .408 .4399 .4515 18 1 .63 .482 .3893 .4641 19 1 .58 .494 .1728 .0924 20 1 .63 .482 .5553 .6730 21 1 .74 .440 .3780 .5168

95

ITEM .SCALE MEM SIGMA R( TOTAL) R(SCALE)

22 1 .68 .465 .6142 .7566 23 1 .26 .440 .0394 .0554 24 1 .26 .440 .1572 .0783 25 1 .53 .499 .1167 .2189 26 1 .32 .465 .3997 .2408 27 1 .79 .408 .6133 .6246 28 1 .37 .482 .0504 .0165 29 1 .21 .408 .2422 .1913 30 1 .53 .499 .3433 .1785 31 1 .79 .408 .3705 .4762 32 1 .53 .499 .3244 .2794 33 1 .84 .365 - .1061 .0538 34 1 .79 .408 .6133 .6246

Item Analysis: Part II

1 2 .63 .482 .4870 .5496 2 2 .32 .465 .5112 .5480 3 2 .84 .365 .5143 .5371 4 2 .68 .465 .6649 .6291 5 2 .68 .465 .4925 .4695 6 2 .68 .465 .6041 .6221 7 2 .84 .365 .5530 .5803 8 2 .89 .307 .2999 .3312 9 2 .89 .307 .2538 .2285 10 2 .32 .465 .3591 .4802 11 2 .79 .408 .2549 .4020 12 2 .89 .307 .1770 .3055 13 2 .74 .440 .3245 .4607 14 2 .63 .482 .4967 .4842 15 2 .26 .440 .2321 .2553 16 2 .95 .223 .3855 .4571 17 2 .63 .482 .3599 .4189 18 2 .37 .482 .5780 .5617 19 2 .68 .465 .4824 .4695 20 2 .89 .307 .4688 .4853 21 2 .42 .494 .7817 .8210 22 2 .42 .494 .3426 .3261 23 2 .11 .307 .5601 .5421 24 2 .05 .223 .2055 .1784 25 2 .11 . .307 .1301 .2339 26 2 .37 .482 .5096 .5780 27 2 .47 .499 .7894 .7495 28 2 .32 .465 .6227 .6328 29 2 .32 .465 .4605 .5820 30 2 .32 .465 .2973 .2089

97

ITEM SCALE MEAN SIGMA " R( TOTAL) R( SCALE)

31 2 .53 .499 .4377 .4188 32 2 .21 .408 .3924 .3135 33 2 .74 .440 .6027 .6577 34 2 .95 .223 .2377 .0687

Choice Distribution (Percentages); Part I

ITEM KEY 1 2 3

1 3 32 11 58 2 2 11 89 0 3 1 63 16 21 4 1 58 16 26 5 2 16 68 16 6 2 0 74 26 7 2 5 89 5 8 1 89 0 11 9 1 63 5 32 10 1 89 11 0 11 2 5 58 37 12 1 58 21 21 13 2 16 84 0 14 3 0 21 79 15 2 5 84 11 16 1 100 0 0 17 2 11 79 11 18 1 63 32 5 19 3 5 37 58 20 2 26 63 11 21 1 74 11 16 22 3 11 21 68 23 3 11 63 26 24 2 11 26 63 25 3 11 37 53 26 2 11 32 58 27 1 79 11 11 28 2 11 37 53 29 1 21 47 32 30 3 16 32 53 31 2 11 79 11 32 2 32 53 16 33 2 16 84 0 34 1 79 5 16

Choice Distribution (Percentages): Part II

ITEM KEY 1 2

1 1 63 37 2 1 32 68 3 1 84 16 4 1 68 32 5 1 68 32 6 1 68 32 7 1 84 16 8 1 89 11 9 1 89 11 10 1 32 68 11 1 79 21 12 1 89 11 13 1 74 26 14 1 63 37 15 1 26 74 16 1 95 5 17 1 63 37 18 1 37 63 19 1 68 32 20 1 89 11 21 1 42 58 22 1 42 58 23 1 11 89 24 1 5 95 25 1 11 89 26 1 37 63 27 1 47 53 28 1 32 68 29 1 32 68 30 1 32 68 31 1 53 47 32 1 21 79 33 1 74 21 34 1 95 5

APPENDIX K

TEST DATA: GROUP 6 BILINGUALS/NON-BILINGUAL PROGRAM (BORDER)

Reliability Coefficient

RECOGNITION PRODUCTION TOTAL TEST

Item N 34.0000 34.0000 68.0000

Means 26.1000 20.2500 46.3500

Sigma 4.0237 6.3943 9.8096

Alpha - .7252 .8691 .8969

Item Analysis: Part • I

ITEM SCALE MEAN SIGMA R( TOTAL) R( SCALE)

1 1 .80 .400 .3874 .3542 2 1 1.00 0.000 I I 3 1 .65 .477 .4216 .4090 4 1 .75 .433 .3973 .5596'. 5 1 1.00 0.000 I I 6 1 .80 .400 .6805 .7891 7 1 1.00 0.000 I I 8 1 1.00 0.000 I I 9 1 .60 .490 - .1165 .0203 10 1 .90 .300 .4876 .5053 11 1 .75 .433 .1265 .2152 12 1 .30 .458 .4772 .5803 13 1 .95 .218 .6396 .6329 14 1 .60 .490 .3517 .3500 15 1 .95 .218 - .3192 -.2794 16 1 .95 .218 .3590 .5188 17 1 .85 .357 .1435 .2192 18 1 .80 .400 .2600 .3542 19 1 .60 .490 .4245 .2739 20 1 .70 .458 .4349 .4501 21 1 .90 .300 .2158 .4639

99

22 23 24 25 26 27 28 29 30 31 32 33 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

100

SCALE MEM SIGMA R(TOTAL) R(SCALE)

1 .95 .218 .0550 -.0513 1 .40 .490 .2830 .3095 1 .50 .500 .4944 .5468 1 .70 .458 .3125 .4501 1 .60 .490 .1540 .0710 1 .95 .218 .6396 .6329 1 .45 .497 - .1040 -.0724 1 .45 .497 .4083 .4271 1 .70 .458 .4905 .4773 1 .85 .357 .1863 .2192 1 .85 .357 .4861 .3933 1 .85 .357 .4575 .3933 1 1.00 0.000 I I

Item Analysis: Part II

2 .45 .497 .5005 .4833 2 .45 .497 .5517 .6248 2 .50 .500 .5760 .5239 2 .75 .433 .6209 .5643 2 .65 .477 .2933 .3238 2 .70 .458 .4794 .5204 2 .85 .357 .2719 .3449 2 .90 .300 .5046 .5083 2 .90 .300 .6915 .6907 2 .60 .490 .6222 .5586 2 .75 .433 .5268 .6004 2 .70 .458 .4905 .5546 2 .80 .400 .4256 .5278 2 .75 .433 .3031 .3296 2 .60 .490 .4037 .4150 2 .90 .300 .2667 .2997 2 .65 .477 .1544 .1271 2 .70 .458 .2569 .2474 2 .60 .490 .5181 .4948 2 .80 .400 .6167 .6060 2 .70 .458 .4127 .2645 2 .60 .490 .3829 .5267 2 .30 .458 .4994 .5204 2 0.00 0.000 I I 2 .45 .497 .3058 .2790 2 .25 .433 .3325 .4289 2 .60 .490 .3517 .4310 2 .20 .400 .3390 .3519 2 .25 .433 .5444 .5553 2 .30 .458 .3548 .4692

101

ITEM SCALE MEAN SIGMA R(TOTAL) R(SCALE)

31 2 .75 .433 .3149 .3657 32 2 .25 .433 .3678 .3566 33 2 .90 .300 .4027 .3258 34 2 .70 .458 .4683 .4522

Choice Distribution (Percentages): Part I

ITEM KEY 1 2 3

1 3 0 20 80 2 2 0 100 0 3 1 65 10 25 4 1 75 5 20 5 2 0 100 0 6 2 10 80 10 7 2 0 100 0 8 1 100 0 0

' 9 1 60 10 30 10 1 90 10 0 11 2 10 75 15 12 1 30 10 60 13 2 5 95 0 14 3 0 40 60 15 _ 2 0 95 5 16 1 95 0 5 17 2 5 85 10 18 1 80 10 10 19 3 0 40 60 20 2 30 70 0 21 1 90 0 10 22 3 5 0 95 23 3 10 50 40 24 2 10 50 40 25 3 10 20 70 26 2 15 60 25 27 1 95 5 0 28 2 5 45 50 29 1 45 30 25 30 3 15 15 70 31 2 10 85 5 32 2 15 85 0 33 2 15 85 0 34 1 100 0 0

102

Choice Distribution (Percentages): Part II

ITEM KEY 1 2

1 1 45 55 2 1 45 55 3 1 50 50 4 1 75 25 5 1 65 35 6 1 70 30 7 1 85 15 8 1 90 10 9 1 90 10 10 1 60 40 11 1 75 25 12 1 70 30 13 1 80 20 14 1 75 25 15 1 60 40 16 1 90 10 17 1 65 35 18 1 70 30 19 1 60 40 20 1 80 20 21 1 70 30 22 1 60 40 23 1 30 70 24 1 0 100 25 1 45 55 26 1 25 75 27 1 60 40 28 1 20 80 29 1 25 75 30 1 30 70 31 1 75 25 32 1 25 75 33 1 90 10 34 1 70 30

SELECTED BIBLIOGRAPHY

Allen, Harold B. Readings in Applied English Linguistics . New York; Appleton-Century-Crofts, 1964.

. Teaching English as a Second Language. New York: McGraw-Hill Book Co., 1965.

Alurista. "Mis Ojos Hinchados," El Espejo, ed. 0. Romano. Berkeley: Quinto Sol Publications, Inc., 1969.

Anderson, Theodore, and Mildred Boyer. Bilingual Schooling in the United States. Austin: Southwest Educational Develop-ment Laboratory, 1970.

Ausubel, D. P. "How Reversible Are the Cognitive and Motivation­al Effects of Cultural Deprivation? Implications for Teaching the Culturally Deprived Child." Paper read at a conference on the teaching of the culturally deprived child, Buffalo, New York, March 1965.

Bar-Adon, Aaron, and Werner F. Leopold. Child Language: A Book of Readings. Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1971.

Bernstein, Basil. "Social Class and Linguistic Development: A Theory of Social Learning," Education, Economy., and So­ciety, eds. A. H. Halsey, J. Florid, and A. Anderson. New York: Free Press, 1961.

Bolinger, Dwight. Aspects of Language. New York: Harcourt, Brace, and World, 1968.

Burling, Robbins. Man's Many Voices Language in Its Cultural Context. New York: Holt, Rinehart and Winston, Inc., 1970.

Burma, John H. Spanish-Speaking Groups in the United States . Durham: Duke University Press, 1954.

Carter, Thomas P. Mexican-Americans in School: A History of Edu­cational Neglect. New York: College Entrance Examina-tion Board, 1970.

103

104

Cheyney, Arnold B. Teaching Culturally Disadvantaged in the Ele­mentary School. Columbus, Ohio: Charles E. Merrill Books, Inc., 1967.

Chomsky, Carol. The Requisition of Syntax in Children from Five to Ten. Cambridge, Mass.: MIT Press, 1969.

Chomsky, Noam. Aspects of the Theory of Syntax. Cambridge, Mass.: MIT Press, 1965.

Davis, Alva L. Diagnostic Test for Students of English as a Second Language. Washington, D.C.: Educational Services, 1953.

Department of Education and Science. International Seminar: Bi-lingualism in Education. London: Her Majesty's Station­ary Office, 1965.

de Saussure, Ferdinand. Course in General Linguistics . New York: Philosophical Library, 1959.

Deutsch, Martin. "Minority Group and Class Status as Related to Social and Personality Factors in Scholastic Achievement." Monograph 2. Ithaca, N.Y.: Society for Applied Anthro­pology, 1960.

Erwin, Susan M. "Learning and Recall in Bilinguals," American Journal of Psychology, Vol. 74 (1961a), pp. 446-451.

. "Semantic Shift in Bilingualism," American Journal of Psychology, Vol. 74 (1961b), pp. 233-241.

Finocchiaro, Mary. Teaching Children Foreign Languages . New York: McGraw-Hill Book Co., 1964.

Fishman, Joshua. "Bilingualism, Intelligence, and Language Learning," Modern Language Journal, Vol. 49 (1965) pp. 227-2371.

. Language Loyalty in the United States. The Hague: Mouton Press, 1966.

Gallarza, Ernesto, Herman Gallegos, and Julian Samora. Mexican-Americans in the Southwest. Santa Barbara: McNally and Loftin, 1970.

Garza, Hernan Solis. Lbs Mexicanos del Norte. Mexico City: Editorial Nuestro Tiempo, 1971.

105

Gleason, H. A. An Introduction to Descriptive Linguistics . New York: Holt, Rinehart and Winston, Inc., 1961.

. Linguistics and English Grammar. New York: Holt, Rinehart and Winston, Inc., 1965.

Gumperz, John J. "The Linguistic Markers of Bilingualism," Journal of Social Issues, Vol. 23 (1967), pp. 48-57.

Halsey, A. H., J. Florid, and A. Anderson (eds.). Education, Economy, and Society. New York: Free Press, 1961.

Harris, David P. Testing English as a Second Language. New York: McGraw-Hill Book Co., 1969.

Haugen, Einar. "Bilingualism as a Goal," On Teaching English to Speakers of Other Languages, ed. Virginia French Allen. Champaign, 111.: National Council of Teachers of English, 1964.

Heller, Celia. Mexican-American Youth: Forgotten Youth at the Crossroads. New York: Random House, 1966.

Helm, June (ed.). Spanish-Speaking People in the United States. Proceedings of the 1968 Annual Spring Meeting of the American Ethnological Society. Seattle: University of Washington Press, 1970.

Inclan, Rosa. A Reevaluation of the Dade County Bilingual Pro­gram. Miami, Fla. Dade County Board of Education, 1970.

Johnson, Kenneth. Teaching the Culturally Disadvantaged a Ra-tional Approach. Palo Alto: Science Research Associates, 1970.

Keesing, Felix M. Cultural Anthropology: The Science of Custom. New York: Holt, Rinehart and Winston, Inc., 1958.

Kelley, Victor H. "An Experimental Study of Certain Techniques for Testing Word Meanings," University of Iowa Studies in Education, Vol. 9 (1934), pp. 53-94.

Kelly, L. G. (ed.). Description and Measurement of Bilingualism: An International Seminar. Toronto: University of Toron-to Press, 1969.

King, Paul. Bilingual Readiness in Primary Grades, Report 107, HEW. Bethesda, Md.: ERIC Document Reproduction Service (Document Number ED 033 248), 1969.

106

Krohn, Robert. English Sentence Structure. Ann Arbor: Univer­sity of Michigan Press, 1970.

Lado, Robert. Linguistics Across Cultures. Ann Arbor: Univer­sity of Michigan Press, 1957.

. Language Testing. New York: McGraw-Hill Book Co., 1961.

. Language Teaching A Scientific Approach. New York: McGraw-Hill Book Co., 1964.

and Charles Fries . English Pattern Practices . Ann Arbor: University of Michigan Press, 1969.

Lambert, W. E.» and G. R. Tucker. Bilingual Education of Chil­dren, The St. Lambert Experiment. Rowley, Mass.: New­bury House Publishers, 1972.

Lennenberg, E. H. "On Explaining Language," Vol. 164 (1969), pp. 635-643.

Lester, Mark (ed.). Readings in Applied Transformational Gram­mar. New YorKl Holt, Rinehart and Winston, Inc., 1970.

.Lorentin, Joseph, and Shelley Umans. Teaching the Disadvantaged. New York: Teachers College Press, 1966.

Mackey, William F. Language Teaching Analysis . Bloomington: Indiana University Press, 1965.

. Bilingualism as a World Problem. Montreal: Harvest House, 1967.

________. Bilingual Education in a Binational School. Rowley, Mass.: Newbury House Publishers, 1972.

Manuel, Hershel. Spanish-Speaking Children of the Southwest--Their Education and the Public Welfare. Austin: Univer­sity of Texas Press, 1965.

Mehrens > William A., and Irvin J. Lehman. Standardized Tests in Education. New York: Holt, Rinehart and Winston, Inc., 1969.

Menyuk, Paula. Sentences Children Use. Cambridge, Mass.: MIT Press, 1969.

.(

107

Modiano, Nancy. "National or Mother Language in Beginning Read­ing," Research in the Teaching of English, Vol. 2 (1970), pp. 32-43.

National Advisory Committee on Mexican-American Education. The Mexican-American Quest for Equality. Washington, D.C.: United States Office of Education, 1968.

National Consortia for Bilingual Education. Tests in Use in Title VII Bilingual Projects. Fort Worth: National Con-sortia for Bilingual Education, 1971.

Ohannessian, Sirapi (ed.). Reference List of Materials for English as a Second Language Part 1; Texts, Readers, Dictionaries, Tests . Washington, D.C.: Center for Ap­plied Linguistics, 1964.

Ortega, Phillip. "Mexican-American Literature," Nation, Vol. 209 (1969), pp. 258-259.

Pei, Mario A., and Frank Gaynor (eds.). A Dictionary of Linguis­tics . New York: Wisdom Library, 1954.

Regional Educational Agencies Project. A Resource Manual for Im­plementing Bilingual Education Programs. Austin: Texas Education Agency, n.d.

Roberts, Paul. Patterns of English. New York: Harcourt, Brace, and Javanovich, 1956.

Romano, 0. (ed.). El Espejo. Berkeley: Quinto Sol Publica­tions, Inc., 1969.

Saville, Muriel, and Rudolph C. Troike. A Handbook of Bilingual Education. Washington, D.C.: Clearinghouse for Linguis­tics, 1970.

Shane, Harold G., and June G. Mulry. Improving Language Arts Instruction Through Research. Washington, D.C.: Asso­ciation for Supervision andCurriculum Development, National Education Association, 1963.

James Walden, and Ronald Green (eds.). Interpreting Language Arts Research for the Teacher. Washington, D.C.: Association for Supervision and Curriculum Devel­opment, National Education Association, 1971.

Stern, H. H. Foreign Languages in Primary Education. London: Oxford University Press, 1967.

108

Stockwell, Robert P., and J. Donald Bowen. The Grammatical Structures of English and Spanish. Chicago: University of Chicago Press, 1965.

Taylor, Marie E. fin Overview of Research on Bilingualism. Sac­ramento: California State Department of Education, 1970.

Tyler, Stephan A. (ed.). Cognitive Anthropology. New York: Holt, Rinehart and Winston, Inc., 1970.

Ulibarri, Horacio. Interpretive Studies on Bilingual Education. Albuquerque: University of New Mexico Press, 1969.

Valencia, Atilano A. Bilingual-Bicultural Education: A Perspec­tive Model in Multicultural America. Albuquerque: Southwestern Cooperative Educational Laboratory, Inc., 1970.

Weinberg, Meyer. Bibliography: Education of the Minority Child. Chicago: Integrated Education Association, 1970.

Whorf, Benjamin L. Language, Thought, and Reality: Selected Writings. Cambridge, Mass.: Technology Press, 1956.

Zadrozny, J. T. (ed.). Dictionary of Social Science. Washing­ton, D.C.: Public Affairs Press, 1959.

Zintz, Miles V. Education Across Cultures . Dubuque: William C. Brown Co., 1963 .

. Corrective Reading. Dubuque: William C. Brown Co., 1966.