the production and recognition of grammatical and … · 2020. 4. 2. · bilingual children. the...
TRANSCRIPT
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.