Citation
Oral English language assessment of first grade children in bilingual bicultural education : emphasis on phonology and syntax

Material Information

Title:
Oral English language assessment of first grade children in bilingual bicultural education : emphasis on phonology and syntax
Creator:
Bass, Bernice Marie
Publication Date:
Language:
English
Physical Description:
x, 89 leaves : ; 28cm.

Subjects

Subjects / Keywords:
Child psychology ( jstor )
English teachers ( jstor )
Ethnicity ( jstor )
Foreign language learning ( jstor )
Kindergarten ( jstor )
Language ( jstor )
Nonnative languages ( jstor )
Pronunciation ( jstor )
Siblings ( jstor )
Syntax ( jstor )
Curriculum and Instruction thesis Ph. D
Dissertations, Academic -- Curriculum and Instruction -- UF
Education, Bilingual -- Colorado ( lcsh )
English language -- Study and teaching (Elementary) ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Bibliography: leaves 61-66.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Bernice Marie Bass.

Record Information

Source Institution:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
025251242 ( ALEPH )
02719482 ( OCLC )

Downloads

This item has the following downloads:


Full Text











ORAL ENGLISH LANGUAGE ASSESSMENT OF FIRST GRADE CHILDREN IN BILINGUAL BICULTURAL EDUCATION: EMPHASIS ON PHONOLOGY AND SYNTAX





















By

BERNICE MARIE BASS


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1975

































































COPYRIGHT
1975

BERNICE MARIE BASS














ACKNOWLEDGEMENTS


"Thank you"...only two words, yet so expressive of the appreciation I give to the many persons who gave so freely and completely of themselves so that I could become the person I am today.

Great appreciation and deep regards to Dr. Ruthellen Crews my committee chairperson who spent so many precious hours with me as did Dr. Vynce A. Hines who deserves equal appreciation and regards. Many thanks to my other committee members Dr. Michael L. Hanes who served as co-chairperson and Dr. William Drummond.

Special appreciation to all of the staff at the

University of Northern Colorado in Greeley who shared their facilities and expertise which complemented my doctoral program, especialmente al doctor Jose C6rdova.. .con much'simas gracias.

Also special thanks to the many persons who encouraged me "to keep on keeping on"...the most important ones being my parents Mr. and Mrs. Arthur Bass.

Thank you to my typist Ms. Dorothy Long who so

patiently brought the final product to reality. Also great appreciation and unending thanks to the National Fellowships Fund of the Ford Foundation that made it financially


iii








possible to make my dream come true.

And last, but not least--muchlsimas gracias a los

personas y los niios de los programas bilingiies de Colorado que me daban tanta ayuda.

...siempre a todos...gracias.














TABLE OF CONTENTS
PAGE
ACKNOWLEDGEMENTS iii....................................

LIST OF TABLES ...................................... vii

ABSTRACT ............................................ viii

CHAPTERS

INTRODUCTION TO STUDY ....................... 1

Statement of the Problem ................. 2
Limitations of the Study ................ 3
Justification for the Study .............. 3
Assumptions ........................... 4
Definition of Terms ...................... 5
Hypotheses of the Study .................. 6
Procedures ............................ 6
Sample ............................. 6
Instrumentation and Data
Collection ........................ 7
Data Treatment ........................ 8
Organization of the Study ............... 8

II REVIEW OF SELECTED LITERATURE ............... 10

Introduction .......................... 10
Oral First Language Acquisition ....... 10 Second Language Acquisition ........... 16
Measuring Oral Language .................. 19
Teaching Methods ..................... .19
Observation ........................ 20
Testing ............................ 21
Conclusions ........................... 23

III PRESENTATION OF THE DATA ..................... 25

The Sample Population ..................... 25
Age ................................ 26
Language Spoken in the Home .......... 26 Ethnicity of the Child ................ 27
Number of Siblings in Child's
Family ........................... 27










TABLE OF CONTENTS (Continued)


PAGE

Kindergarten Attendance ................ 27
Sex of the Child .................... 27
Testing of Procedures .................. 28
Bass Sentence Repetition Task
(BSRT) ............................ 28
Contrastive Analysis of Various
Selected English and Spanish
Properties ........................... 28
Sounds .............................. 28
Structures .......................... 29
Pilot Testing of the BSRT ................. 30
Peabody Picture Vocabulary Test
(PPVT) .............................. 31
The Test Data .......................... 32
Bass Sentence Repetition Task (BSRT). 32
Peabody Picture Vocabulary Test
(PPVT) ............................ 33

IV STATISTICAL ANALYSES OF THE DATA RELIABILITY
OF THE BASS SENTENCE REPETITION TASK INTERRATER RELIABILITY ......................... 34

Reliability of the Bass Sentence
Repetition Task Inter-rater
Reliability .......................... 34
Relationship of Scores on the BSRT and
Selected Variables ................... 38
Data and Hypotheses Relationships ... 40
Regression Analysis .................... 43
Interpretation of the Regression
Analysis .......................... 48
Summary or Regression Analysis ...... 52

V SUMMARY, RECOMMENDATIONS AND DISCUSSION ... 56

Summary of the Study ................... 56
Recommendations ........................ 59

BIBLIOGRAPHY ......................................... 61

APPENDIX A LETTER FORMS FOR CONTACTING SCHOOLS ..... 67

APPENDIX B BASS SENTENCE REPETITION TASK ............. 70

APPENDIX C RESOURCE PERSONS: QUALIFICATION AND
INFORMATION ............................ 80

APPENDIX D PERSONS KNOWLEDGEABLE OF LANGUAGE AND BBE. 84
BIOGRAPHICAL SKETCH ................................... 88














LIST OF TABLES


Table 1 Table 2 Table 3


Table Table


Table Table


Table 8 Table 9 Table 10 Table 11 Table 12


Mean and Standard Deviation of the Age of Subjects ..........................

M-an and Standard Deviation on BSRT: Pronunciation and Structure ..............

Mean and Standard Deviation on PPVT of Subjects in BBE .......................

List of Variables ........................

Inter-rater Means and Standard Deviation ................................

Inter-rater Reliability of the Scores .... Correlation of Variables to BSRT Score Means ..............................

Correlation Matrix of Predictor Variables ................................

Means and Standard Deviations (N=78) of Prediction Variables ..................

Grouping of Predictor Variables for Regression Analysis ......................

Regression Analysis of Variables for Pronunciation ...........................

Regression Analysis of Variables for Structure ...............................


vii









Abstract of Dissertation Presented to the Graduate
Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy



ORAL ENGLISH LA GUAGE ASSESSMENT OF FIRST GPADE
CHILDREN IN BILINGUAL BICULTURAL EDUCATION:
EMPHASIS ON PHONOLOGY AND SYNTAX By

Bernice Marie Bass

December, 1975


Chairperson: Dr. Ruthellen Crews Major Division: Curriculum and Instruction


The purpose of this study was to develop a reliable instrument that could be used by teachers and staff in Colorado to assist in the assessment of oral English language, particularly the structure (syntax) and the pronunciation (phonology) of first grade children in bilingual bicultural education (BBE), and to relate the scores of the instrument to seven variables: a) intelligence, b) the child's age, c) the language spoken in the home, d) ethnicity, e) the number of siblings in the child's family, f) kindergarten attendance, and g) the sex of the child.

Seventy-eight subjects were identified for the study. These subjects were first graders who were participants in BBE programs in Colorado during the school year 1974-75. The children were tested using the researcher-developed instrument, the Bass Sentence Repetition Task (BSRT), which


viii









was developed on the basis of the contrastive analysis of English and Spanish to determine knowledge of oral English pronunciation and structure and the Peabody Picture Vocabulary Test (PPVT) for testing intelligence.

For the purpose of establishing reliability of the

test instrument, language samples were collected using the BSRT. These samples were scored by three resource persons according to the rating system developed by the researcher. The Kuder-Richardson Formula 21 was applied to estimate the reliability of each section of the BSRT. The coefficient was .71 for pronunciation and .49 for structure. The combined judgments of three raters had the effect of tripling the length of the test. Applying the Spearman-Brown Formula to estimate a new reliability, a .88 coefficient for pronunciation and a .76 for structure were obtained. These established reliabilities for the BSRT are within acceptable limits for an instrument of this type.

Scores from the BSRT were analyzed using correlation analysis as well as multiple regression analysis to determine their relationships to the predictor variables. Correlation analysis found that intelligence, language spoken in the home, and ethnicity were correlated to the BSRT pronunciation score at the .01 level and the variables of intelligence and language spoken in the home were significant at the .05 level to the BSRT structure score with the sex variable at the .01 level.










The multiple regression analysis found that intelligence as measured by the PPVT to be a contributing factor to the measure of oral English pronunciation and structure as measured by the BSRT. Also the number of siblings in the child's family is considered a contributing factor in the prediction of oral English structure as measured by the BSRT.














CHAPTER I

INTRODUCTION TO STUDY


Within -he United States live many groups of languagedifferent peoples. These are persons who speak a language other than English and/or a dialect other than standard English. In some parts of the United States, bilingual bicultural education is offered as a means cf bridging the linguistic and cultural gap of language-different children by affording the opportunity to learn school facts and concepts in the native language while learning English as a second language. Other school systems offer programs which emphasize the development of oral English language to replace the native language spoken. Still other school systems give no consideration to language differences, and all students are reauired to communicate in English from the beginning of the school experience (Robinett, 1971).

In recent years, Colorado legislators have questioned the progress of the children exposed to the forementioned programs. In order to assess the progress made by children in various programs, the Colorado Legislature designated funds for a state needs assessment (reflected by Colorado legislation, see Senate Joint Resolution 20, Colorado Department of Education, 1974). As a result of the assessment









which indicated that a considerably large number of children in Colorado public schools were in need of bilingual education, a bill requiring that public school systems provide such an education was drafted and approved by the legislature. This bill provides financial appropriations to Colorado school districts having fifty students or more than ten percent of the student population identified as being linguistically different (Earle, 1975).


Statement of the Problem


To enable teachers and staff to evaluate effectively the oral English of children participating in bilingual bicultural education (BBE) programs, the teachers of Colorado have needed reliable tools which could be used to assess the phonological as well as the syntactic abilities of their pupils. At the beginning of this study, there were no oral English assessment tools for identifying the English language differences of language-different pupils designed to take into consideration the specific regional dialect differences for use with youngsters in Colorado. Therefore, the purpose of this study was twofold: 1) to develop a reliable instrument that would enable teachers and staff to assess the phonology and syntax of the oral English of first grade children in BBE programs in Colorado and 2) to relate the results from this instrument to seven variables: a) intelligence, b) the age of the child, c) the language spoken in the home (i.e., English or Spanish), d) the ethnicity










of the child (i.e., Anglo or non-Anglo), e) the number of siblings in the family, f) kindergarten attendance, and g) the sex of the child.


Limitations of the Study


The researcher's study was bound by the following limitations:

1. The study was limited to 78 children in the
first grade in Colorado BBE programs identified by the Colorado State Department of
Education during the academic year 1974-75.

2. The researcher was only concerned with the
development of an instrument that could
help Colorado teachers and staff assess the
oral English syntax and phonology of their
students.

3. The variables considered in the study were
confined to intelligence, age, language spoken in the home, ethnicity, number of
siblings in the family, kindergarten attendance, and sex of the children participating
in the study.

The extent to which the findings in this study can be generalized to the populations in other BBE programs is limited because the sample population was taken from one specific area and gave consideration to the regional dialect spoken in Colorado.


Justification for the Study


The need for an instrument that could be used by

teachers and staff in Colorado BBE programs to help assess the oral English of first grade children has been recognized










for some time. In the needs assessment (Colorado Department of Education, 1975) mention was made of the lack of reliable instruments to measure the abilities of language-different children. The need for such instruments is becoming more widespread particularly with the growing number of programs in the state.

The information collected in this study could serve as a model for the development of additional instruments which would expand on the test items, sample other linquistic features and identify additional information about the language used as well as be of assistance to others who are interested in the measurement of the abilities of the languagedifferent for which this was developed. This information could even be helpful to those who are dealing with other types of populations, such as school programs for Vietnamese refugees, both as an example of how to devise sentence repetition task and how to use the results.


Assumptions


The basic assumptions made for this study were that the English language syntax and the phonology of first graders in BBE programs could be assessed, and that other variables extraneous to the language itself have some effects on the acquisition of the language children use in the first grade.










Definition of Terms


To assist the reader, the following terms have been

defined as used by the researcher:

Bilingual Bicultural Education (BBE) - An educational setting
in which two languages are used as the medium of instruction and the respective cultures of the languages are refiwcted and actively incorporated in the classroom setting.

Bilingual - A person who speaks and understands two languages and has displayed an understanding of the cultures of the two languages.

English-speakinq - A person who speaks English as well as
understands it. This person may speak any one of a
number of English dialects.

English-only-education Environment (EOE) - An educational
setting in which English is the only language used as
the medium of instruction.

Language-different - Those who are identified as speaking a
language other than English and/or a dialect other
than standard English.

Mexican American or Chicano - Those individuals of Spanish
surname and/or those who refer to themselves as
Chicanos and are identified as such by school records.

Oral Language Development - Defined as a function of the
measurements on the sentence repetition task.

Phonology - The study of the sound system of a language
Phonology encompasses the understanding of, and the
ability to hear and produce the sounds of a language.

Regional Dialect - A variety of Standard English (or the
standard form of any other language)as spoken by the
educated population of a given region.

Sentence Repetition Task (SRT) - The instrument designed for
use in this study to gather and assess oral language samples. This instrument used outside of this study
will be known as the BASS SRT, 1975.

Syntax - That part of grammar which studies the construction
and arrangement of words in a sentence.










Hypotheses of the Study


The present study considered the following hypotheses:

1. There is no relationship between the scores on
the BSRT and the intelligence score on the
Peabody Picture Vocabulary Test.

2. There is no relationship between the scores on
the 3SRT and the age of the child.

3. There is no relationship between the scores on
the BSRT and the language (a. Spanish, or
b. English) spoken in the home.

4. There is no relationship between the scores on
the BSRT and the ethnicity (a. non-Anglo,
or b. Anglo) of the child.

5. There is no relationship between the scores on
the BSRT and the number of siblings in the
family.

6. There is no relationship between the scores on
the BSRT and kindergarten attendance.
7. There is no relationship between the scores on
the BSRT and the sex of the child.

Each of these hypotheses reflects the review of literature and the information some authorities have found to have some effect on the oral language development of children.


Procedures


In order to test the hypotheses as previously stated, the following procedures were used:


Sample

The sample population for this study was drawn from children who attended the first grade during the academic









year 1974-75. These pupils were selected from participants attending BBE programs from five different schools in three areas of Colorado.


Instrumentation and Data Collection

The researcher developed the Bass Sentence Repetition Task (BSRT) which was recorded by a person with a regional English dialect identified as a Coloradan. The sentences constructed for the instrument reflected local English usage and the contrastive analysis of the linguistic features of English aad Spanish. This particular BSRT was designed to assess the use of the specific linguistic features of phonology and syntax.

In the BSRT, the oral English samples were divided into the categories of oral English structure and English pronunciation for the purpose of scoring. The scoring was done by a group of three resource persons who received training in the techniques of interpreting the SRT.


The Peabody Picture Vocabulary Test (PPVT) provided a measure of the subjects' verbal intelligence through measuring their listening vocabulary.

Other information relating to the subjects, i.e., the number of siblings in the family, the age, the language spoken in the home, ethnicity, kindergarten attendance, and the sex of the child was obtained from school records and through teacher interviews.









Data Treatment

All treatment samples were scored by three resource persons with expertise in the language and culture of the sample population and were trained in the use of the evaluative procedure developed for assessing the language samples. This procedure required that the corrections be made from a contrastive analysis of Spanish and English.

The scores from the instruments were compared with the .seven forementioned variables by employment of the technique of multiple regression.


Organization of the Study


The research study is presented in five chapters with a bibliography and an appendix. This first chapter has served as an introduction to the study, as well as a description of how the study was carried out.

Chapter 2 contains a review of selected literature and research that is related to the topic of the study. This review led to the formulation of the hypotheses.

Chapter 3 presents data from the study and a description of how it was collected.

Chapter 4 includes the analysis of the collected data and the conclusions drawn in relation to the stated hypotheses.

Chapter 5 summarizes the study and includes recommendations based on the findings.




9




The bibliography contains all sources and references of information and precedes the appendices. The appendices include letters, forms, and other information used in contacting school BBE programs. Also copies of the BSRT used as well as the qualifications and information on the resource persons, and other supportive information relating to this study are found in the appendices.














CHAPTER II

REVIEW OF SELECTED LITERATURE Introduction


The concern of this study was focused on developing a tool for assessing the oral English of those children in Colorado who were first graders and who were participating in bilingual bicultural education (BBE).

A consideration of the development of assessment tools for effective evaluation of the oral English of children participating in BBE programs requires a preliminary understanding of several fields of knowledge which relate to the question. First, how is oral language acquired? Next, how is a second language acquired? Third, how can oral language be measured?


Oral First Language Acquisition


All normal children master their particular first

language and acquire all the basic operations of adult grammar by the age of four regardless of native language or social settings, even though they may not verbally use these operations at all times (McNeill, 1971; Slobin, 1972; and sass, 1974). The young children extract sounds from the language that is spoken to them and with them (Chomsky, 1965;








and Katz, 1966), but they are not conscious that they are doing this and cannot say how they go about doing it (McNeill, 1971).

Research by Ervin-Tripp and Miller (1964) substantiates the theory that early language development reflects generalizations of adult operations. Language includes some degree of imitation but more important, includes the ability to generalize. For example, a young child will say "dog," "toy," or "foot," but will over-generalize in making plurals which may result in "dogs," "toys," or "foots." Similar findings were noted in the case of the regular past tense: "walk" became "walked," "sell" became "selled," "go" became "goed." The ability to generalize and to extend production of language is basic to first language learning.

First language learners acquire the ability to use

language through sequential steps (Slobin, 1972). Menyuk's (1969) research states that levels of language development can be delineated and studied; therefore, she correlates the process of language development with the process of maturation.

In the initial stages of language acquisition, the

child demonstrates an unlimited phonetic ability (SavilleTroike, 1973b). This phonetic ability soon is further developed to include single words. During the next stages the child learns noun and verb inflections leading to the acquisition of syntax. By age four or five, the child has









acquired essentially all basic language structures (Chomsky, 1969). Beyond this age level to about puberty, the child continues to refine his or her language learning (SavilleTroike, 1973b).

The learner of the first language also acquires the

ability to usp the language communication skills: listening and reading as well as speaking and writing (Zintz, 1969). Through the skills of listening and reading language is received. As the learner matures, his or her ability to decode what is received through these skills is increased. The skills of speaking and writing serve as a means of transmitting language. Likewise, as the learner gains more experience, the ability to transmit language increases (Ruddell, 1974).

Chomsky (1965) emphasized that the observation of the

child's use of the communication skills (language performance) is not the only component of language, but that usage (language competency) is also important. Language performance can be observed directly, but language competency is that which underlies observable language performance (Ruddell, 1974). Jakobovits' (1971) findings further support this theory. Understanding usually was observed to be superior to speaking, with understanding equated to competence and speaking to performance. At more advanced levels, both understanding and speaking were viewed as performance.









McNeill's (1966) observations are somewhat similar.

Phonological production was viewed as preceding grammatical comprehension which preceded grammatical production. An example of this is the child who repeats a long sentence but might not be able to understand or produce it. The same child may understand a longer sentence but might not be able to produce it. The research of Fraser, Bellugi, and Brown (1963) parallels these findings.

The child's language development is greatly influenced by two critical factors: 1) the ability to formulate and grasp generalizations about the language system, and 2) the language model or models in the environment (Ruddell, 1974). Although early home environment plays a major role in a child's language development, the teacher's language and that of other children exert an influence on language development.

Dailey and Neyman (1965) studied oral and written

facility and comprehension among children who had attended kindergarten and those who had not. The kindergarten youngsters composed the experimental group, and conclusions showed that these children did significantly better in all areas and demonstrated fewer oral language errors than the other children did.

Further studies involving the factors of kindergarten attendance and oral language include the research of Pendergast et al. (1966). In order to investigate the effect of kindergarten on oral articulation of first grade children,









over 15,255 pupils were tested. Those children who had attended kindergarten did not appear to have fewer oral English errors when they entered the first grade than those children who did not attend kindergarten, and of the twothirds of the children who did have some articulation problems and were not enrolled in speech therapy, kindergarten experience appeared to help reduce sound errors in the first grade.

Oral language skills served as a measure in Stanchfield's (1971) test of the effects of teaching prereading skills tc kindergarteners. Results indicated that girls outperformed the boys in first language learning.

Among children of average intellectual ability, it is not unusual to find one child who does not seem to be performing at an expected level. Often this child is a boy. This is attributed to cultural expectation according to Ruddell (1974). Girls are expected to play passive games which are oriented to school life while boys are expected to be active.

Kagan (1964) found that girls tended to be more

oriented to school activities and to perform better than boys at the elementary grades. On the other hand, Laney (1970) concluded that kindergarten males and females showed no differences in expressive vocabulary development in their first language.









Stanchfield's (1971) research also demonstrated that there are differences in achievement by ethnic groups. Children representing three ethnic groups were selected, all of which were attending kindergarten and represented a cross section of socioeconomic levels. Oral language skills were included as a measure in the testing and results showed that whites outperformed the other ethnic groups involved in the study.

Likewise, Laney (1970) observed the relation of expressive and comprehensive language to race and cultural background and concluded that these variables were related to the level of a child's language development.

Cohen and Klein (1968) demonstrated that language accuracy as measured by the ability to use communication skills increased with age in first language learners. In order to master the more complex elements of language, Piaget (1955) demonstrated that the normal child needs additional maturation time. Furthermore, Asher (1972) concluded after studying a population of three elementary grades to determine if task had any influence on the performance of various ages of children in oral English, that younger children's poor communication skills in the first language cannot be explained by limited explanation of the task demands.

Davis (1937) found that single children in a family

developed language facility more rapidly than children with siblings, and that chidren with siblings developed language faster than twins did. Research by Gesell et al. (1940, 1946)









found similar results. Children who come from families with many brothers and sisters tended to talk later than those children who were single children. The children who were twins also tended to talk later than single children. A single child in a family tended to develop oral English abilities earlier than those children who had siblings because of the greater opportunity and need to interact with adults. The same was found of children who were the first born in most families.


Second Language Learning


The degree of second language learning is dependent

on the time at which it is introduced as well as the manner in which it is presented (Smith, 1971). If the second language is learned at the same time the first language is learned, acquisition does not differ from that of the first. However, when the second language is acquired in another context, the learner does not go through the first language acquisition process (Saville-Troike, 1973b).

Smith (1971) observed that second language is learned by a combination of presentation, explanation, repetition, and transfer. Second language learning is an inferred process, building up rule by rule while first language learning permits "trial and error" learning. The setting for second language acquisition generally minimizes possibilities of errors, in an effort to develop the competence of a full native speaker (Cook, 1971).









Jakobovits (1970) found mastery of a second language to be based on a good grounding in the first language. He supports the age of four or five years as an optimum time to acquire a second language. At this time, an individual's cognitive development is at a later and more advanced stage, he is in possession of the grammatical structure of a language which may facilitate acquisition of another language through transfer, and he possesses concepts and meanings in the first language.

Penfield's (1960, 1965) research substantiated that

second language learning is a function of brain flexibility. Up until the age of twelve or puberty, the uncommitted part of the brain can become involved in language learning. Ervin-Tripp's (1968) work found that second language learning in children up to the age of eleven years is based on the sound of the language while adults based such learning on language sense. Furthermore, foreign accent was observed as being directly related to the age at which the second language was introduced, the accent being more pronounced after age twelve (Lenneberg, 1967).

All physiologically normal children are born with the capacity to produce any sound used in any language; however, with age, the flexibility is lost. "Foreign sounds" are not heard as such; the learner has been conditioned not to hear through previous experience with his own language.

Problems in phonology, grammar, and vocabulary are a tendency as a result of one's previous first language









experiences. These problems of perception in relation to the use of second language which reflect the first language habits are referred to as "linguistic interference." Many interference problems can be predicted through a contrastive analysis of the languages involved (Saville-Troike, 1973b).

As a result of linguistic interference, some

psychologists propose that simply learning two languages at the same time may hinder a child's intellectual development; that the impairment of intellectual functioning arises in the non-English-speaking child solely from the child's using, or being asked to use, two languages; that learning two languages at the same time may lead to use of a smaller vocabulary, simplier sentences, confused word order, misuse of negatives, misuse of idiomatic expressions, and the use of literal translations (Smith, 1949; and Darcy, 1963).

Darcy's (1963) review of the literature relating

bilingualism and intelligence showed that a child's intellectual development is often hindered by his use of two languages. Also intellectual functioning as measured by standardized tests appears to be impaired. The children who come to school speaking another language use confused word order, misuse negatives and idiomatic expressions, and use, at times, literal translations. Scores were also low on intelligence tests. The scoring on the intelligence tests tended to be correlated to their English language performances.









Young Spanish-speaking children in Arizona who were

introduced to English-only language instruction before their Spanish was well-established made a variety of oral language mistakes in Spanish (Lynn, 1945). Abraham (1960) found that children who were given insufficient English-asa-second language training performed one standard deviation below the national norm on standardized tests of mental ability.


Measuring Oral Language


According to the Encyclopedia of Educational Research (Ebel, 1969) there are three general methods of studying language: teaching (conditioning), observation, and testing. Teaching Methods

One method of teaching is behavior modification (stimulus-response-reward). This method is used by Bereiter and Englemann (1966) in their early childhood program titled Distar. The participants in this program are conditioned to respond in a specific manner according to the directions given by their teacher.

Other examples of teaching methods (Smith, 1971) are methods used in foreign language teaching. The audiolingual approach uses an aural-oral technique, "hear itsay it." Another method, total submersion, places the learner in the cultural environment of the new language generally prohibiting use of any other language except the









newone. Rote memorization is the technique of direct memorization of grammar rules with stated examples.


Observation

Observation has served as a means of language study in many research projects. Both N. Chomsky (1965) and C. Chomsky (1969) used the method of eliciting speech from youngsters by engaging children in informal conversation to observe their language and to make generalizations about the speech of their subjects.

Loban (1966) spent ten years observiing children's

language which resulted in the development of Categories for Tallying Problems in Oral Language as a means of assisting teachers in identifying crucial and frequent oral language deviations from standard English made by Negro and Caucasian subjects at pre-school, primary, intermediate,and junior high levels.

Allred (1970) elicited samples of free speech from

third grade Mexican American children with Spanish language backgrounds in Colorado and did an analysis of errors in English usage. The analysis of the language samples found misuse of negatives and of common expressions as well as the misuse of verb forms. These were defined as the effects of language differences. From these findings, Allred recommended further evaluation of the pupils' weaknesses though diagnosis followed with an instructional plan based on the specific items which are sources of difficulty.









Testing

Testing is the general method selected by most researchers in their studies of language. The results of testing, however, are of no benefit to the examiner or to the examinees unless the findings are used selectively. If not, testing becomes another means of labeling for failure. Selective use of test results warrants the use of the information to develop additional prescriptive procedures that give positive and special attention to identified problem areas.

The Gloria and David Beginning Language Tests (1958) were developed in Texas to assist in the identification of phonological and morphological elements that are difficult for primary youngsters whose speech differs from the recorded model. The instrument tests by the method of sentence repetition. After use by a considerable number of persons, a questionnaire was developed to accompany the test.

Natalicio and Williams (1972) constructed Questionnaire for Evaluations of Black Language Samples; Questionnaire for Evaluators of Mexican American Language Samples. These questionnaires serve as a means of evaluating oral language samples derived from the Gloria and David on a seven-point scale except for two binary (yes-no) items. Reliability was established by identifying experts specializing in dialect study who used the questionnaire to evaluate a number of language samples (ten Mexican American samples and ten Black samples).









In 1968 Olquin constructed a non-technical teaching aid for teachers of bilingual children titled Shuck Loves Chirley which includes a diagnostic test of sentence repetitions. This instrument has been available through the California Public Schools. The materials were developed to assist teachers in the identification of sounds which the child cannot hear.

The Linguistic Capacity Index developed in 1964 by Manning and Brengleman (Fagan, 1975) was developed as a measure of English language readiness in grouping native Spanish-speaking pupils for English language instruction at the primary level. The index includes three sections: vocabulary recognition, contrastive phonology, and contrastive grammar. Vocabulary Recognition measures recognition of verb, noun, preposition, and adjective forms. Contrastive Phonology measures ability to discriminate sounds of pairs of words. Contrastive Grammar measures understanding of English function words, word order, and inflectional constructions that differ from Spanish constructions. The test has been used by the developers with over 167 subjects. In addition, it was administered to 300 kindergarten children in U.S.O.E. Project 2821 and to more than 2400 first graders in Cooperative Research Reading Projects 2648 and 2734.

The Bilingual Syntax Measure (Burt, Dulay, and Chavez, 1975) was developed to measure children's oral proficiency









in English and/or Spanish grammatical structures by using natural speech as a basis for making judgments. This instrument is designed specifically for children ages four to nine years.

The instruments and techniques of studying language that are described here are a samoling of the types of materials that have been available for oral language assessment, some of which are specifically for BBE. The researcher has selected the method of testing to study oral language, more specifically, the testing method referred to as sentence repetition task.


Conclusions


The literature relating to first and second language acquisition indicates that many factors affect oral English language ability. The factor of intelligence is effected by the timing of and the preparation for second language learning (Darcy, 1963; Abraham, 1970). Second language learning can hinder intellectual development (Smith, 1949).

Success in second language acquisition is dependent

on age (Lynn, 1945; Penfield, 1960 and 1964; and Lenneberg, 1967), just as first language acquisition is a function of age (Piaget, 1955; Ervin-Tripp Miller, 1964; Cohen and Klein, 1968). Saville-Troike (1973b) also supports the concept that all language is learned in stages. Smith (1971) demonstrates that the manner of second language presentation affects its acquisition.









Stanchfield (1971) and Laney (1970) correlated

cultural and ethnic factors to language development. In addition, Stanchfield, Ruddell (1974), and Kagan (1964) demonstrated that young girls outperformed boys of the same age while Laney (1970) found no differences.

Davis (1937) and Gesell (1940 and 1946) reported that the number of siblings in the child's family affects oral language development.

Dailey and Neyman (1965) identified kindergarten

attendance as being positively correlated with the number of oral language errors made. On the other hand, Pendergast (1966) found the opposite to be true.

Knowledge of some of the effects of these factors led to the development of the previously stated hypotheses. The past research has either supported or questioned the extent to which there is a relationship of these factors to the acquisition of oral English language, particularly the oral English of children whose native language is not English.

The next chapter will describe the design of this

study as well as present information about the instrumentation, the subjects, and the data collected in carrying out the study.













CHAPTER III

PRESENTATION OF THE DATA


The teachers and staff of Colorado bilingual bicultural education (BBE) programs need reliable tools to assist them in the assessment of the abilities of the students they serve so that students may receive immediate assistance in acquiring oral English language skills. The purpose of this study was to develop such an instrument and to determine the relationships between scores on this instrument with seven identified variables: 1) intelligence, 2) age of the child, 3) the language spoken in the child's home, 4) the ethnicity of the child, 5) the number of siblings in the child's family, 6) kindergarten attendance, and 7) the sex of the child.


The Sample Population


The sample population in this study consisted of

seventy-eight BBE first graders who were participants during the academic year 1974-75, in three Colorado BBE programs: one in northern Colorado, one in south-central Colorado, and one just south of the Denver area. Of the programs selected, five schools were used representing populations in rural, urban, and semi-urban settings, all of which reflected the same regional dialect usage.








Permission to work with youngsters from BBE programs was gained through contacts suggested by the BBE liaison with the Colorado State Department of Education.

Information other than test scores regarding the subjects was made available to the researcher from the schools' records and through teacher interviews.

Age

The age of each child in the sample was obtained by

using the dates of birth from the school records and calculating the age of the child in months to the date of testing. The age distributions in months of the subjects who participated in the study are in Table 1. The ages ranged from 79 months to 111 months with a mean of 86.6 and a standard deviation of 5.35.


Table 1
MEAN AND STANDARD DEVIATION OF THE AGE OF SUBJECTS



Mean S. D. Range



N = 78 86.6 5.35 79-111 months




Language Spoken in the Home

The language spoken in the home was determined by the teachers of each of the subjects based on their knowledge of the child's home and the teacher's interaction with the family








of the subject. Only speakers of the languages Spanish and English were considered. Of the total population, 9 of the boys and 12 of the girls were native speakers of Spanish. Twenty-three boys and 34 girls were native speakers of English.


Ethnicity of the Child

Ethnicity was used to refer to the ethnic representation of the child. The subject was classified by the teacher of each child as being a) non-Anglo, or b) Anglo. Among the participants of this study, 19 boys were non-Anglo and 30 girls were non-Anglo. Thirteen boys and 16 girls were Anglo.


Number of Siblings in Child's Family

If school records did not carry information regarding the number of siblings in the family, the child was asked and the number was then verified by the classroom teacher. The number of siblings in the homes of the sample population ranged from zero to 12.


Kindergarten Attendance

Kindergarten attendance was determined by the fact

that a subject had attended kindergarten for any length of time prior to entering first grade. In the sample population, 75 attended kindergarten and 3 did not.


Sex of the Child

The sex of the child was noted at the testing session.









Included in the study were 32 boys and 46 girls making a total of 78 participating in the study.


The Testing Procedures


Bass Sentence Repetition Task (BSRT)

Samples of the participants' oral English language were collected on cassette tapes using the BSRT developed by the researcher (see Appendix B) and audiotape recorded by a person with a regional English dialect that the children were accustomed to hearing. The seventeen sentences constructed for the BSRT reflected the specific linguistic features identified as being significantly English as the result of the contrastive analysis of English and Spanish according to the following (Politzer and Stanback, 1961; Shane, 1967; Zintz, 1969). Persons knowledgeable of the field were also consulted and found to be helpful (see Appendix D).


Contrastive Analysis of Various Selected English
and Spanish Language Properties Sounds

Spanish English

161 palatal-affricate- 161 church
voiceless

no equivalence Isil shoe


JbI bilibial-stop-voiced _blboy, buy, berry
[6] bilibial-fricative-voiced









Sounds


no distinction from the b only one labiodentalfricative f none



none I :tt alveolar-lateralvoiced none none


English

lvI very labiodental-fricativevoiceless 1e1 thick dental-fricativevoiceless 131 rather dental-fricativevoiced none


glottal stops lI decision


Structures


Spanish

All verbs are inflected
(example: The dog run)

Adjectives usually follow the noun
(example: The man big)

Use of reflexive in many cases
(example: I put myself to
sleep)

Use of no and double negation
(example: Juan no go home;
No go:
I don't want that, not never
ever)

No auxiliaries used
(example: He works?
Works my mother?)


English

"S" as a part of present
(example: The dog runs)

Adjectives before the noun
(example: The big man)


No separation of certain verbs
(example: I go to sleep)

Seldom use of double negation
(example: Juan did not
go home. Don't go!
I don't want that)

Use of auxiliaries
(example: Does he work?
Does my mother work?)









Spanish English

Use of "to have" to express Use of "to be" English "to be" (example: I am seven
(example: I have seven years old; I am hungry;
years; I have hunger; I am thirsty)
I have thirst)

The use of gender for nouns Nouns do not have gender
(example: My car she goes) (example: My car goes)

Tendency to translate idiomatic Phrases express differently phrases (example: my watch runs;
(example: my watch he walks; I missed the bus; the
the bus leaves me; the window faces the street)
window gives to the street)

It should be understood that the BSRT is not a

normed instrument but is reflective of standard language knowledge. The BSRT is designed to aid in determining whether the child is repeating the sentences as given or whether the child is restructuring the sentences (i.e., changing to a structure that is not acceptable as equivalent to that which is presented).

This BSRT was constructed to assess use of the

specific linguistic features of phonology and syntax. For purposes of scoring, the sentences that were constructed to assess knowledge of phonology were scored as one category labeled English pronunciation and those constructed to assess English syntax were scored as another category labeled English structure. Therefore, two scores were given for this test.


Pilot Testing of the BSRT

The BSRT developed for this study was pilot tested in March, 1975, by the researcher. The testing involved a group








of 24 children representing a BBE program in the Northern Colorado area. The children were first graders and were selected randomly from the children in the BBE program.

The BSRT was administered individually to each of the examinees by the researcher as was the PPVT.

The purpose of the pilot test was twofold: 1) to

familiarize the researcher with the procedure established for the administration and scoring of the BSRT, and 2) to determine if the sentences selected for the BSRT were functional to provide data appropriate for assessing oral English phonology and syntax.

The analysis of the language samples derived from

pilot testing of the BSRT was done three times by the researcher according to the rules for grading (see ADpendix B).


Peabody Picture Vocabulary Test (PPVT)

The PPVT is a standard test designed to provide an

estimate of a subject's verbal intelligence through measuring his listening vocabulary (Dunn, 1965). It is administered individually to examinees. Each examinee is asked to select the illustration from a set of four that represents the word given orally by the examiner. According to the test manual (Dunn, 1965), correlations ranged from a low of 0.67 at the six year level to a high of 0.84 at the 17 and 18 year levels, with a median of 0.77 as the resulting coefficients of equivalence for raw scores and standard errors of measurement for IQs.









The researcher was aware of available data which

indicate that the PPVT has elements of cultural bias against bilingual children (Hickey, 1972), but for purposes of this study, this instrument seemed to be about as effective as any available.


The Test Data


Bass Sentence Repetition Task (BSRT)

All BSRT's administered for this study were done so by the researcher. Scoring of the data from the BSRT's was done by a group of three persons who met specified qualifications as developed by the researcher to provide for inter-rater reliability (see Appendix C) and in accordance with the rules set for their interpretation (see Appendix B) which reflected the contrastive analysis of English and Spanish. An average of the scores of the three resource persons who graded each part of the test was computed. Each child received a score on pronunciation (phonology) and a score on structure (syntax).


Table 2

MEAN AND STANDARD DEVIATION ON BSRT:
PRONUNCIATION AND STRUCTURE


Pronunciation Structure

Mean S. D. Range Mean S. D. Range

N = 78 16.6 2.78 8-20 8.45 1.47 1-10








The scores for pronunciation ranged from 8 to 20 for the sample with a mean of 16.6 and standard deviation of

2.78. The scores for structure ranged from 1 to 10 with a mean of 8.4 and standard deviation of 1.47 (see Table 2).


Peabody Picture Vocabulary Test (PPVT)

The PPVT was administered and scored by the researcher. Table 3 shows the deviation of intelligence scores for the subjects participating in the study.


Table 3

MEAN AND STANDARD DEVIATION ON PPVT OF SUBJECTS IN BBE



MEAN S. D. RANGE


N = 78 92.7 15.1 26-116




Note that the range of scores was from 26 to 116, more specifically, the girls ranged from 26 to 114 and the boys from 67 to 116 with a mean of 92.7 and standard deviation of 15.1.

The details of the data analyses and results will be found in Chapter 4.













CHAPTER IV

STATISTICAL ANALYSES OF THE DATA


This chapter includes the analyses of data collected

for establishing reliability of the Bass Sentence Repetition Task (BSRT) developed by the researcher and the relationship of the scores on the BSRT and the stated hypotheses. (Statistical analyses of the data collected in this study were made by the use of the computer system at the University of Northern Colorado, Greeley.1


Reliability of the Bass Sentence Repetition Task

The pupils used in this study were first graders who

were participants in Colorado bilingual bicultural education (BBE) programs during the school year 1974-75. Seventyeight subjects were selected randomly from those attending such programs.

Each subject was tested individually in an area removed from his regular classroom setting to limit external noises during the recording of language samples. The BSRT was administered to assess the oral English phonology and syntax of the pupils and the Peabody Picture Vocabulary Test (PPVT) was administered as a measure of intelligence. Other information regarding the participants in the study was







obtained from school records and conferences with teachers, i.e., the age of the child, the language spoken in the home, the ethnicity of the child, the number of siblings in the child's family, kindergarten attendance, and the sex of the child. These served as the independent variables of the study. The list of variables and their abbreviations as used in referring to them are found in Table 4.

The Kuder-Richardson Formula 21 (Thorndike and Hagen, 1955) was used to estimate the reliability of the BSRT. The formula was run for each part of the BSRT, the pronunciation (phonology) section and the structure (syntax) section. The coefficients were .71 and .49 respectively, based upon the means for three raters.

- ~~ [iI- v t4)]



Since what was finally used was the combined judgments of three raters, this had the effect of tripling the length of a test,, using the Spearman-Brown Formula (Thorndike and Hagen, 1955).





Inter-rater Reliability

Table 5 lists the inter-rater means and standard deviations which were used to derive the inter-rater reliability (see Table 6) of the scorers of the BSRT total scores for pronunciation and structure. The Pearson correlation








Table 4

LIST OF VARIABLES


Number Variable Abbreviation


1 Intelligence IQ 2 Age in months Age 3a Spanish as home language Span 3b English as home language Eng 4a Non-Anglo representative NA 4b Anglo representative A

5 Number of siblings in child's Sib family
6 Kindergarten attendance K

7 Sex of the child Sex











Table 5

INTER-RATER MEANS AND STANDARD DEVIATION


Pronunciation Structure

Scorers Mean S. D. Mean S. D.


1 16.449 3.049 8.321 1.540 2 16.603 2.866 8.628 1.594 3 17.077 2.800 8.333 1.525


Table 6

INTER-RATER RELIABILITY OF THE SCORERS


Pronunciation Structure

Scorers 1 2 3 1 2 3


1 .937 .895 .863 .757 2 .937 .890 .863 .800

3 .895 .890 .757 .800








coefficients ranged from .937 to .890 for the pronunciation scores and from .863 to .757 for the structures scores. These would fall within the 95 percent confidence intervals for the estimated reliabilities reported above. With a total subject population of 78, these coefficients are judged adequate for this study. This means that the standard errors of measurement are 0.8 and 0.6 respectively. The 95 percent confidence interval for an individual score would be the score plus or minus 1.6 for pronunciation and plus or minus 1.2 for structure.


Relationship of Scores on the BSRT and Selected Variables


In order to determine the relationship of the scores on the BSRT and the selected variables--intelligence, age in months, language spoken in the home, ethnicity of the child, number of siblings in child's family, kindergarten attendance, and sex of the child--an average of the scores of the three resource persons who graded the language samples was computed. Then, a correlation of the average score of each child was made for each section of the BSRT, pronunciation (phonology) and structure (syntax), with each variable that was identified as having some relation to oral English. Table 7 indicates the results of these correlations. With 78 subjects a correlation above .23 or below -.23 would be significant although correlations below -.30 would tend to be trivial. The correlation coefficients ranged from a










Table 7

CORRELATION OF VARIABLES TO BSRT SCORE MEANS


Pronunciation


Structure


0.675**

-0.278*

-0.549** 0.549**

-0.354** 0.354**

-0.184

0.308**

-0.143


0.473**

-0.130

-0.342** 0.342**

-0.180 0.180 0.093

0.468**

-0.243*


* Significant at .05 level with 75 degrees of freedom

** Significant at .01 level with 75 degrees of
freedom


positive 0.675 showing a fairly positive relation between the IQ variable and the pronunciation score on the BSRT to a negative correlation of -0.549 showing that when Spanish is spoken in the home there is an inverse relation to the score on English pronunciation on the BSRT. The highest correlation of the structure score correlates 0.473.


IQ Age Span Eng NA

A

Sib

K

Sex








Table 8 shows the interrelations of the predictor variables.


Data and Hypotheses Relationships

The following relates the data interpretation in terms of the stated hypotheses.

Hypothesis 1: There is no relationship between the scores on the BSRT and the IQ score on the PPVT.

The correlation analysis shows that there is a positive and significant relationship between the scores on the BSRT and the PPVT. This relationship holds true for oral English structure and English pronunciation as measured by the BSRT. Therefore, Hypothesis 1 is rejected.

Hypothesis 2: There is no relationship between the scores on the BSRT and the age of the child.

The variable of age correlates inversely with pronunciation scores on the BSRT (-0.278) at the .05 level. The correlation for structure (-.13) is not significant. Hypothesis 2 is rejected for pronunciation only.

Hypothesis 3: There is no relationship between the scores on the BSRT and the language (a. Spanish, or b. English) spoken in the home.

Language spoken in the home is related to oral English as measured by the BSRT pronunciation score. As would be expected, variable 3b Eng showed direct relationship (.549) to BSRT scores while variable 3a Span showed an inverse relationship (-0.549).






Table 8

CORRELATION MATRIX OF PREDICTOR VARIABLES


Variable with computer number

1 2 3a 3b 4a 4b 5 6 7 IQ 1

Age 2 -0.293* Span 3a -0.450** 0.102 Eng 3b 0.450** -0.102 -1.000 NA 4a -0.279* 0.251* 0.467* -0.467** A 4b 0.279* -0.251* -0.467* 0.467** -1.00 Sib 5 -0.294* 0.222 0.045 -0.045 0.114 -0.114 K 6 0.262* 0.009 -0.330** 0.330** -0.154 0.154 -0.066 Sex 7 0.010 -0.044 0.007 -0.007 -0.093 0.093 -0.149 0.234*


* Significant at .05 level

** Significant at .01 level








The BSRT score for structure showed a lower but still

significant correlation with the language spoken in the home. Variable 3b Eng showed direct relationship (.342) and 3a Span was inverse (-0.342). Hypothesis 3 is rejected for both BSRT scores.

Hypothesis 4: There is no relationship between the scores on the BSRT and the ethnicity (a. Non-Anglo, or b. Anglo) of a child.

Ethnicity did show a relationship to the scores on the BSRT. Anglo representation showed a positive correlation (.354) and non-Anglo representation showed an inverse correlation (-0.354) for BSRT score on pronunciation. Neither of the observed relationships 4a (-0.18) or 4b (.18) were significant at the .05 level for structure. Hypothesis

4 is rejected for pronunciation but not for structure.

Hypothesis 5: There is no relationship between the scores on the BSRT and the number of siblings in the family.

The correlation analysis of the BSRT score for pronunciation and the variable Sib, showed that there was no significant relationship (-0.184) between the two. Likewise, the comparison with the BSRT score for structure showed a nonsignificant relationship (.093). In this case, Hypothesis 5, as stated, was not rejected.

Hypothesis 6: There is no relationship between the scores on the BSRT and kindergarten attendance.

The variable of kindergarten attendance correlated with pronunciation (.308) and with structure (.468). Both are.








significant at the .01 level. However, since only three children had not gone to kindergarten, no conclusion will be drawn from these results.

Hypothesis 7: There is no relationship between the scores on the BSRT and the sex of the child.

The pronunciation score on the BSRT showed no significant relationsnip (-0.143) with whether the child was a boy or girl, but the structure score did (-0.243). Hypothesis 7 was not rejected for pronunciation, but it was rejected for structure.

Regression Analysis


In addition to the simple correlation analysis,

multiple linear regression (Kerlinger and Pedhazur, 1973) was used to determine the unique contribution of different sets of the variables to the prediction of English language pronunciation and oral English language structure. The contribution of a set of variables to prediction may be measured by the square of the multiple correlation coefficients (RSQ), one obtained for a regression model in which all predictors are used, called the full model (FM), and the other obtained for a regression or restricted model

(RM). The multiple correlation coefficients for the RM can never be larger than the multiple correlation coefficients for the FM. The difference between the two coefficients can be tested for statistical signficance with the variance ratio test.









The knowledge of which variables make a contribution

to the prediction of oral English pronunciation and structure would indicate which variables furnish information in determining the performance of similar first grade children in Colorado BBE programs.

Because the variables used in this study are

significantly correlated with each other, it is possible that conclusions based upon simple Pearson product moment correlations may be suspect. A more rigorous test of the relationship of these variables to the pronunciation and structure scores is through the use of multiple linear regression.

Table 9 lists the means of the predictor variables as well as gives the standard deviations of each of these variables. This information shows that the mean IQ of the test population was 92.7 with a standard deviation of 15.1 which is considered a normal deviation for the instrument. The average for the variable of Age was 86.6 months which means that the average subject was about seven years and two months old at the time of testing. Each subject had an average of 2.87 siblings. The other variables were binary coded.

If two highly correlated predictors occur in a regression system and one is deleted in order to determine its unique contribution, there is little or no drop in the square of the multiple correlation coefficient from the full model










Table 9

MEANS AND STANDARD DEVIATIONS (N=78)
OF PREDICTOR VARIABLES



Variable Mean S. D.


1 IQ 92.7 15.1 2 Age 86.6 5.35 3a Span 0.269 0.444 3b Eng 0.731 0.444 4a NA 0.628 0.483 4b A 0.372 0.483 5 Sib 2.87 2.39 6 K 0.962 0.192 7 Sex 0.423 0.494



Mean of pronunciation
scores 16.61 2.78 Mean of structure
scores 8.45 1.47








to the restricted model. This may be tested by this formula for significance of a difference between multiple Rs (Guilford, 1965).
F R2

R12 X- rX2

R = multiple R with larger number of
F independent variables

R = multiple R with one or more
R variables omitted

m = larger number of independent variables

m = smaller nuiaber of independent variables

N = number of subjects

dfI= ml-m2

df2 = (N-m1-i)


Closely related variables were grouped. These are

shown in Table 10. Where use of the above formula gives an F with a probability above .05, unnecessary information need not be collected. As a result Table 11 and Table 12 were made as a sequential guide for retaining or dropping variables.

The top blocks of Tables 11 and 12 indicate that all

variables (1-7) were used in the full model (Schmid and Reed, 1966). The next blocks indicate which predictor variables represent the sets to which multiple regression techniques were applied. The RSQ for each of these blocks could be compared with the RSQ of the FM. A large difference at or below the .05 level would indicate that the set is making






Tabl6 10

GROUPING OF PREDICTOR VARIABLES FOR REGRESSION ANALYSIS


Groups Variable Number



I Spanish Language Home (Span) 3a

English Language Home (Eng) 3b Non-Anglo Representative (NA) 4a Anglo Representative (A) 4b


II Age in Months (Age) 2 Number of Siblings in Child's Family
(Sib) 5

Sex of the Child (Sex) 7


III Intelligence (IQ) 1

Experience in Kindergarten (K) 6








an unique contribution to the prediction of the phenomena. A small or zero drop indicates that any relationship found is not significant.


Interpretation of the Regression Analysis

The regression of the criterion, oral English pronunciation and structure, on all of the variables had RSQ's of .5692 (Table 11) and .4303 (Table 12) respectively. The square root of these values is .7545 and .6560 respectively. This is higher than the best individual variable for structure, which is IQ. Each of these variables correlated at

0.675 and 0.473 respectively.

The restricted model, FM 3a, 3b, 4a, and 4b for pronunciation, (Table 11) had an RSQ of .4988. The observed drop is .5692-.4988=.0704 and is statistically significant; therefore, this difference indicates that this set of variables (Span, Eng, NA, A) makes an unique contribution to the prediction of oral English pronunciation. From this set, the restricted models FM 3a, 3b and FM 4a, 4b were noted which indicate that language (3) and ethnicity (4) as a group of variables contributed to the prediction but that one cannot look at the individual language or ethnicity as a predictor.

The second group which formed the restricted model

FM 2, 5, and 7 for pronunciation, had an RSQ of .5412. No significant drop is noted between this model containing the variables Age (2), Sib (5), and Sex (7), and the FM




Table 11


(Adapted from Schmid and Reed, 1966)




Table 12

REGRESSION ANALYSIS OF VARIABLES FOR STRUCTURE


(Adapted from Schmid andReed, 1966)








containing all of the predictor variables.

Group III, FM 1 and 6 for pronunciation, had an RSQ

of .3977. The observed drop is .5692-.3977=.1715 and this is significant at the .05 level. This indicates that Group III contributes to the prediction of oral English pronunciation. Further analysis of the model shows that IQ (1) has an RSQ of .4068 and K (6) has an RSQ of .5676. The drop observed in variable 1 is significant (.5692-.4068= .1624), while no significant drop is noted in variable 6 alone. IQ as measured by the PPVT is a contributing variable in the prediction of oral English pronunciation.

Applying the same technique to structure (Table 12), Group I FM 3a, 3b, 4a, and 4b has an RSQ of .4282 and no significant drop is noted between this RSQ and the RSQ (.4303) of the FM. Therefore, the variables Span (3a), Eng (3b), NA (4a), and A (4b) as a group are not contributing factors to the prediction of oral English structure.

Group II, FM 2, 5, and 7 for structure, had an RSQ of .3542. The observed drop, .4303-.3542=.0761, is statistically significant and further analysis was done. The restricted models, FM 2, FM 5, and FM 7, showed RSQs of .4272, .3836, and .4125 respectively, none of which show statistically significant drops. As a group, these variables Age (2), Sib (5), and Sex (7) are of use in the prediction of oral English structure. Variables Age (2) and Sex (7) do not make significant contributions, but the variable








Sib (5) is significant when the RSQ of .3836 is rounded to .38 (.4303-.38=.05).

The restricted model, FM 1,6 for structure, had an

RSQ of .1968 which gave a drop of .4303-.1968=.2335. This drop is significant and warrants further analysis. The resulting models, FM 1 and FM 6 had RSQs of .3145 and .3440. The models showed drops of .1158 and .0863, each of which is statistically significant. The conclusion is that IQ (1) as measured by the PPVT, Sib (5), and K (6) are contributing variables and are useful in the prediction of oral English structure.

It may seem that the seven predictor variables can be used to predict the English language of children in Colorado BBE programs quite well, the square of the RSQ being .5692 (pronunciation) and .4303 (structure). Furthermore some models having fewer variables than seven predictors also had a high prediction efficiency. Examples of high prediction efficiency are FM Age (2), Sib (5), Sex (7) for pronunciation and FM Span (3a), Eng (3b), NA (4a), and A (4b) for structure. Considering these findings, a prediction equation could be constructed for any group of predictor variables using the data of Tables 7 and 8. Summary of Regression Analysis

As a result of the regression analysis the following

conclusions were made in reference to the stated hypotheses:









Hypothesis 1: There is no relationship between the scores on the BSRT and intelligence as measured by the PPVT.

According to the findings of the regression analysis,

IQ as measured by the PPVT, has a significant relationship to the scores on the BSRT, thereby, giving some predictability of oral English language pronunciation and structure. Therefore, Hypothesis 1 was rejected at the .05 level of significance.

Hypothesis 2: There is no relationship between the scores on the BSRT and the age of the child.

There was an age range from 79 to 111 months and a

standard deviation of 5.35 months among the subjects. With this somewhat restricted range, regression analysis showed that no significant relationship exists between the scores on the BSRT and the age of the child. Hypothesis 2, therefore, was not rejected.

Hypothesis 3: There is no relationship between the scores on the BSRT and the language (a. Spanish, or b. English) of the home.

Language of the home as a predictor did not have any statistically significant relationship to the scores on the BSRT. However, when language of the home and ethnicity of the child were grouped as predictors using regression analysis, significant results were noted. Hypothesis 3 as stated is not rejected.








Hypothesis 4: There is no relationship between the scores on the BSRT and the ethnicity (a. non-Anglo, or b. Anglo) of the child.

Ethnicity as a separate variable had no relationship to the scores on the BSRT. Hypothesis 4 was not rejected.

Hypothesis 5: There is no relationship between the scores on the BSRT and the number of siblings in the child's family. No significant relationship was observed between the

scores on the BSRT and the number of siblings in the child's family. This variable Sib showed some predictability when giouped with the child's age and sex using regression analysis (and is a contributing factor when the RSQ value is rounded). Nevertheless, Hypothesis 5 was retained even though it was shown as indicating a relationship that was significant.

Hypothesis 6: There is no relationship between the scores on the BSRT and kindergarten attendance.

The regression analysis showed that no relationship

existed between the score for pronunciation on the BSRT and the variable of kindergarten attendance. But a significant relationship was noted between the BSRT score for structure and kindergarten attendance. In this case, Hypothesis 6 was rejected for structure and not rejected for the BSRT pronunciation score. However, as previously stated, since only three children did not attend kindergarten, no generalizations can be made as a result of this finding.










Hypothesis 7: There is no relationship between the scores on the BSRT and the sex of the child.

The regression analysis showed no significant relationship between the sex of the child and the scores on the BSRT. Therefore, Hypothesis 7 was not rejected.

The ove -ii results of the correlation analysis and the regression analysis showed a relationship between the scores on the BSRT and most of the identified oral English predictors. The variables of intelligence and language of the home had the greatest direct relationship with the pronunciation and structure scores on the BSRT according to correlation analysis. The variable of ethnicity also had a .01 level of significance in relation to the pronunciation score on the BSRT.

Multiple regression analysis identified intelligence as a significant single variable in predicting the BSRT pronunciation score while the number of siblings in the family was singly significant in its relationship to the BSRT structure score along with intelligence.

Chapter V will summarize this study and make recommendations for further research.













CHAPTER V

SUMMARY, RECOMMENDATIONS, AND DISCUSSION


Summary of the Study


The purpose of this study was to develop a reliable instrument that could be used by teachers and staff in Colorado to assist in the assessment of oral English language, particularly the structure (syntax) and the pronunciation (phonology) of first grade children in bilingual bicultural education (BBE), and to relate the scores of the instrument to seven variables: a) intelligence, b) the child's age, c) the language spoken in the home (Spanish or English), d) ethnicity (non-Anglo or Anglo), e) the number of siblings in the child's family, f) kindergarten attendance, and g) the sex of the child.

Seventy-eight subjects were identified for the study. These subjects were first graders who were participants in BBE programs in the state of Colorado during the school year 1974-75. The children were tested using the researcherdeveloDed instrument, the Bass Sentence Repetition Task (BSRT), to determine knowledge of oral English pronunciation and structure and the Peabody Picture Vocabulary Test (PPVT) for testing ingelligence.









The Kuder-Richardson Formula 21 was applied to estimate the reliability of each section of the BSRT. The coefficient was .71 for pronunciation and .49 for structure. The combined judgments of three raters had the effect of tripling the length of the test. Applying the Spearman-Brown Formula to estimate a new reliability, a .88 coefficient for pronunciation and a .76 for structure were obtained.

Correlation analysis as well as multiple regression analysis were used to determine the relationships between the BSRT scores and the predictor variables.

The correlation analysis showed that the variables intelligence, language spoken in the home, and ethnicity were related to the BSRT pronunciation score at the .01 level of significance. The variables intelligence and language spoken in the home were significantly related to the BSRT structure score at the .01 level, while the sex variable was significantly related at the .05 level. No conclusion can be drawn about the relationship of the BSRT scores with kindergarten attendance because only three pupils in the sample did not attend kindergarten.

The method of multiple regression was used as a

source of a more rigorous method of data analysis which can serve as a means of identifying the variables that significantly contribute to the prediction of certain phenomena, in this case, that of oral English pronunciation and structure.








A significant relationship was noted between the

pronunciation score on the BSRT and the variable of intelligence as measured by the PPVT. Also the structure score on the BSRT was significantly related to the intelligence variable and to the number of siblings in the family (based on the rounded multiple correlation coefficient for the variable). Even though a significant relationship of the scores on the BSRT and the variable of kindergarten attendance was observed, no conclusions can be drawn because only three of the subjects did not attend kindergarten.

Multiple regression analysis also showed a statistically significant relationship between the scores on the pronunciation portion of the BSRT and the grouping of ethnicity and language variables; however, further analysis did not determine a relationship between the BSRT scores and individual languages or ethnic representations.

As a result of the regression analysis, the intelligence variable as measured by the PPVT is considered to be a contributing factor to the measure of oral English pronunciation and structure as measured by the BSRT. Furthermore, the correlation analysis showed that this variable is positively related. The number of siblings in the child's family (when the RSQ is rounded) is considered a contributing factor to the prediction of oral English structure as measured by the BSRT.

The established reliabilities for the BSRT are within acceptable limits for an instrument of this type. The BSRT,








therefore, is a reliable tool for assessing the oral English pronunciation and structure of first grade children in Colorado BBE programs. The statistically significant variables of intelligence as measured by the PPVT and the number of siblings in the child's family are contributing factors to predicting the oral English pronunciation and structure as measured by the BSRT.

The researcher does caution that correlation analyses

do not indicate cause and effect relationships. The analyses used in this study only indicate the degree of relationship between the BSRT and the variables studied. For this reason, no conclusions about casuality were presented.


Recommendations


In view of the findings and the limitations of this study, the following recommendations are made:

1. The researcher recommends administration of the
BSRT to first grade children in Colorado BBE
programs as a quick assessment of oral English
pronunciation (phonology) and structure (syntax).
Furthermore, it could be tested for use in
Spanish-English programs in other parts of the
country.

2. The researcher recommends the development of
BBE instructional materials which give special
attention to the items in the BSRT.

3. Intelligence as measured by the PPVT when used
selectively can assist in prediction of oral
English pronunciation and structure as measured
by the BSRT.








4. The information gathered in this study can be
used as an example of how to devise a sentence
repetition task for use with other languagedifferent populations.

5. The researcher recommends the development of a
reliable and valid tool for the measurement of the mental ability of youngsters in BBE which
minimizes the factors of cultural and linguistic
biases.

6. The researcher recommends the development of a
sentence repetition task that takes into consideration the dialect differences of Spanish
spoken in a given area for use in that area.

7. Data from this study can be used in pre-service
and in-service education to prepare teachers for working in BBE programs. Also, the BSRT and instructions for administering and scoring it could
be used in such educational programs.

BBE programs are increasing in number across the United

States. With the growing concern for assuring that these

programs are indeed meeting the needs of the populations they

are designed to serve, research and development of materials

are of utmost importance. The development of the BSRT is to

be of specific service to the BBE programs of Colorado as


well as BBE programs across the nation.














BIBLIOGRAPHY


Abraham, William. "Investigation of Mental Retardation and
Pseudo Mental Retardation in Relation to Bilingual and Sub-Cultural Factors." Paper prepared for the
College of Education, Arizona State University, 1960.

Allred, Forest R. "Errors in Oral English Language Usage
of Mexican American Pupils with a Spanish Langauge Background in Grade III in the State of Colorado."
Doctoral dissertation, University of Northern Colorado,
1970.

Asher, Steven R. "Influence of Sampling and Comparison
Processes on the Development of Communication Effectiveness." Paper presented at the Eightieth Annual Convention of the American Psychological Association,
Hawaii, September, 1972. (ERIC No. ED 071 732).

Bass, Bernice. "Language Development Should Begin at Home."
Education Colorado, Colorado State Department of
Education, January 15, 1974.

Benitez, Mario. "Bilingual Education: The What, the How
and the How Far." Hispania, September, 1971, 54,
pp. 499-503.

Bereiter, Carl and Englemann, Siefried. Teaching Disadvantaged Children in the Preschool. New Jersey: Prentice Hall, Inc., 1966.

Bernstein, Basil. "A Sociolinguistic Approach to Socialization: With Some Reference to Educability." Language
and Poverty, Frederick Williams (Ed.), Illinois:
Markham Publishing Company, 1970.

Board of Education of the City of New York. Teaching English
as a New Language to Adults. Curriculum Bulletin
Number 5, 1963-1964 Series.

Brengelman, Frederick H. and Manning, John C. "Linguistic
Capacity Index." Measures for Research and Evaluation
in the English Language Arts, William T. Fagan,
Charles R. Cooper, and Julie M. Jenson (Eds.), Illinois:
ERIC Clearinghouse on Reading and Communication Skills and the National Council of Teachers of English, 1975.








Burt, Marina K.; Dulay, Heidi C.; and Chdvez, Eduardo
Hern~ndez. Bilingual Syntax Measure. New York:
Harcourt Brace Jovanovich, Inc., 1975.

Cartel: Annotated Bibliography of Bilingual Bicultural
Material, Cumulative Issue - 1974. Texas: Dissemination Center for BBE, 1974.

Chomsky, Carol. The Acquisition of Syntax in Children from
Five to Ten. Massachusetts: The M.I.T. Press, 1969.

Chomsky, Noam. Aspects of the Theory of Syntax. Massachusetts: The M.I.T. Press, 1965.

Cohen, Bertram D. and Klein, Judith F. "Referent Communication in School Age Children." Child Development,
1968, 39, pp. 597-609.

Colorado Department of Education. Colorado School Laws
Supplement, 1974. October, 1974.

Colorado Department of Education. Ethnic Group Distribution
in Colorado Public Schools, 1971-1972. March, 1973.

Colorado Department of Education. Report on Senate Joint
Resolution 20 Study: Part II Bilingual Multi-Cultural
Education Programs. January, 1975.

Cook, Vivian J. "The Analogy Between First-and-Second Language Learning." Toward a Cognitive Approach to
Second Language Acquisition, Robert C. Lugton and
Charles H. Heinle (Eds.), Pennsylvania: The Center
for Curriculum Development, Inc., 1971.

Dailey, John T. and Neyman, Clinton A. "An Evaluation of the
Language Arts Program of the District of Columbia,
Final Report." Washington, D. C.: George Washington
University, November, 1965. (ERIC No. ED 024 449).

Darcy, Natalie. "Bilingualism and the Measurement of
Intelligence: Review of a Decade of Research."
Journal of Genetic Psychology, 1963, 103, pp. 259-82.

Davis, Edith A. The Development of Linguistic Skills in
Twins, Singletons with Siblings, and Only Children from Age Five to Ten Years. Minnesota: University
of Minnesota Press, 1937.

Dunn, Lloyd M. Expanded Manual for the Peabody Picture
Vocabulary Test. Minnesota: American Guidance
Service, Inc., 1965.









Earle, Jane. "Bilingual Education Bill Is Unique in U. S."
Denver Post, June 22, 1975, 29C.

Ebel, Robert (Ed.). "Language Development." The Encyclopedia of Educational Research. Ontario: CollierMacMillan Canada Ltd., 1969, pp. 686-98.

Ervin-Tripp, Susan. "Becoming a Bilingual." Working paper
9, Language Behavior Research Laboratory, California
University, Berkeley, March, 1968 (ERIC No. ED 018 786).

Ervin-Tripp, Susan and Miller, Wick R. "The Development of
Grammar in Child Language." Monographs of the Society for Research in Child Development, 1964, 19, pp. 9-34.

Fagan, William T.; Cooper, Charles R.; and Jensen, Julies M.
(Eds.). Measures for Research and Evaluation in the
English Language Arts. Illinois: Eric Clearinghouse
on Reading and Communication Skills and National
Council of Teachers of English, 1975.

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

Fox, David. The Research Process in Education. New York:
Holt, Rinehart, and Winston, Inc., 1969.

Fraser, Colin; Bellugi, Ursula; and Brown, Roger. "Control
of Grammar in Imitation, Comprehension and Production."
Journal of Verbal Learning and Verbal Behavior, 1963,
2, pp. 121-35.

Gesell, Arnold; Ilg, Frances; Ames, Louise B.; and Bullis,
Glenna E. The Child From Five to Ten. New York:
Harper and Brothers, 1946.

Gesell, Arnold; Ilg, Frances; Ames, Louise B.; and Bullis,
Glenna E. The First Five Years of Life. New York:
Harper and Brothers, 1940.

The Gloria and David Beginning Language Tests. Texas:
Language Arts, Inc., 1958.

Guilford, Joy Paul. Fundamental Statistics in Psychology and
Education. New York: McGraw-Hill Book Company, 1965,
p. 403.

Haugen, Einar. "Problems of Bilingualism." Lingua, August,
1950, 2, pp. 271-90.

Hickey, Tom. "Bilingualism and Measurement of Intelligence
and Verbal Learning Abilities." Exceptional Children,
1972, 39, pp. 24-28.








Jakobovits, Leon A. "Implications of Recent Psycholinguistic
Developments for the Teaching of a Second Language."
Toward a Cognitive Approach to Second Language Acquisition, Robert C. Lugton and Charles H. Heinle (Eds.),
Pennsylvania: The Center for Curriculum Development,
Ind., 1971.

Jakobovits, Leon A. Foreign Language Learning, A Psycholinguistic Analysis of the Issues. Massachusetts:
Newberry House Publishers, 1970.

John, Vera and Horner, Vivian. Early Childhood Bilingual
Education. New York: Modern Language Association of
America, 1971.

Kagan, Jerome. "The Child's Sex Role Classification of School
Objects." Child Development, 1964, 35, pp. 1951-56.

Karnes, Merle B.; Hodges, Audery; and Teska, James A. "Evaluation of Two Preschool Programs for Disadvantaged
Children: A Traditional and a Highly Structured Experimental Preschool." Exceptional Children, May, 1968.

Katz, Jerald J. Philosophy of Language. New York: Harper and
Row, Publishers, 1966.

Kerlinger, Fred and Pedhazur, Elazar. Multiple Regression in
Behavioral Research. New York: Holt, Rinehart, and
Winston, Inc., 1973.

Laney, Billie J. "A Comparative Study of Expressive and Comprehensive Vocabulary Development in Male and Female Kindergarten Children." Doctoral dissertation, East
Texas State University, 1970.

Lenneberg, Eric H. Biological Foundations of Language.
New York: John Wiley and Sons, 1967.

Loban, Walter. Problems in Oral English: NCTE Research
Report No. 5. Illinois: National Council of Teachers
of English, 1966.

Lynn, Klonda. "Bilingualism in the Southwest." Journal of
Speech, April, 1945, 31, pp. 175-80.

Macnamara, John. "The Bilingual's Linguistic Performance-A Psychological Overview." Journal of Social Issues,
April, 1967, 23, pp. 58-77.

McNeill, David. "The Creation of Language by Children."
Child Language, A Book of Readings, A. Bar-Adon and
W. Leopold, (Eds.), New Jersey: Prentice-Hall, Inc.,
1971, pp. 349-59.








McNeill, David. "Developmental Psycholinguistics." The
Genesis of Language: A Pscyholinguistic Approach,
Frank Smith and G. A. Miller (Eds.), Massachusetts:
The M.I.T. Press, 1966.

Menyuk, Paula. "Children's Learning and Reproduction of
Grammatical and Nongrammatical Phonological Sequences."
Child Development, 1968, 39, 849-59.

Menyuk, Paula. Sentences Children Use. Massachusetts: The
M.I.T. Press, 1969.

Natalicio, Diana S. and Williams, Frederick. "Oral Language
Assessment." Paper presented at the Annual Meeting of
the American Educational Research Association, Illinois,
April 6, 1972 (ERIC No. ED 063 845).

Olguin, Leonard. "'Shuck Loves Chirley': A Non-Tecnical
Teaching Aid for Teachers of Bilingual Children with a
Diagnostic Test." CARTEL: Annotated Bibliography of Bilingual Bicultural Material, Cumulative Issue -1974,
Texas: Dissemination Center for BBE, 1974.

Pendergast, Kathleen; Barker, Janet; Dickey, Stanley; Soder,
Anton; Gow, Jane; and John, Selmar. "An Articulation Study of 15,255 Seattle First Grade Children with and
without Kindergarten." Exceptional Children, April,
1966, 32, 541-47.

Penfield, Wilder. "Conditioning the Uncommitted Cortex for
Language Learning." Brain, 1965, 88, 787-98.

Penfield, Wilder. "The Learning of Languages." Foreign
Language Bulletin, Spring, 1960, 7, pp. 1-17.

Piaget, Jean. The Language and Thought of the Child. Ohio:
The World Publishing Company, 1955.

Politzer, Robert and Staubach, Charles. Teaching Spanish, A
Linguistic Orientation. Massachusetts: Ginn and Company,
1961.

Robinett, Robert. "Teaching of Bilingual Children." The
Encyclopedia of Education. New York: MacMillan Company
and Free Press, 1971, pp. 466-72.

Ruddell, Robert. Reading-Language Instruction: Innovative
Practices. New Jersey: Prentice Hall, Inc., 1974.

Saville-Troike, Muriel R. "TESOL: Methods and Materials in
Early Childhood Education." Essays on Teaching English
as a Second Dialect, R. P. Fox (Ed.), Illinois: National
Council of Teachers of English, 1973a, pp. 102-06.





66



Saville-Troike, Muriel R. Bilingual Children: A Resource
Document. Virginia: Center for Applied Linguistics,
1973b.

Slobin, Dan I. and Ferguson, C. A. (Eds.). Studies of Child
Language Development. New York: Holt, Rinehart, and
Winston, Inc., 1973.

Slobin, Dan I. "Children and Language: They Learn the Same
Way All Around the World." Psychology Today, July,
1972, 6, pp. 74-77, 82.

Smith, Mary E. "Measurement of Vocabularies of Young Bilingual Children in Both of the Languages Used." Journal
of Genetic Psychology, 1949, 74, 305-10.

Smith, Philip D. Toward a Practical Theory of Second Language
Instruction. Pennsylvania: The Center for Curriculum
Development, Inc., 1971.

Stanchfield, Jo M. Research: New Hope for Kindergarten
Children. Paper Presented at the meeting of the
California Reading Association, California, November,
1971 (ERIC No. ED 058 011).

Thorndike, Robert L. and Hagen, Elizabeth. Measurement and
Evaluation in Psychology and Education. New York:
John Wiley and Sons, 1955, pp. 129, 131.

Williams, Frederick (Ed.) Language and Poverty, Perspectives
on a Theme. Illinois: Markham Publishing Company,
1970.

Zintz, Miles. Education Across Cultures. Iowa. Kendall/
Hunt Publishing Company, 1969.









































APPENDIX A

LETTER FORMS FOR CONTACTING SCHOOLS







THE UNIVERSITY OF NORTHERN COLORADO GREELEY, COLORADO 80639



February 28, 1975





Mr. Keller, Principal Letford Elementary School Johnstown, Colorado

Dear Mr. Keller:

This letter is to introduce Bernice Bass who is a doctoral student at the University of Florida working under my supervision here at the University of Northern Colorado to complete work for a degree with emphasis in bilingual education.

Ms. Bass is in the process of developing an instrument which could be of assistance in the evaluation of oral English language proficiency of young children of Mexican-American descent. In the pilot testing of this instrument, Ms. Bass has selected to work with youngsters of first grade level.

Your assistance in permitting Ms. Bass to work with youngsters in your school is greatly appreciated.

Sincerely,



Jos6 C6rdova
Assistant Professor Elementary Education









UNIVERSITY OF FLORIDA
GAINESVILLE, FLORIDA



April 6, 1975





ADDRESS



Dear

This letter is to introduce Bernice Bass who is a doctoral student at the University of Florida working under my direction to complete a degree with emphasis in bilingual bicultural education.

Ms. Bass has developed an instrument which could be of assistance in the assessment of oral English language of young children, particularly those with language differences. She has selected to work with first grade youngsters in her research study to establish the reliability for the instrument she has developed. This will require that she work with individual children for approximately 15 to 20 minutes each. Your assistance in your program would be greatly appreciated.

Ms. Bass will be contacting you within the next week.

Sincerely,



Ruthellen Crews
Professor of Education






































APPENDIX B

BASS SENTENCE REPETITION TASK










BSRT


Entering Interview:

1. Who are you? (What is your name?)
2. How many brothers and sisters do you have?
3. How old are you?


Repetitions:

1. The girl sits on a chair.
2. The girl asks the teacher a question.

(You are a good listener--for reinforcement)

3. The dog barks.
4. The berries are ripe.
5. The fruit is fresh.
6. I think the apples are ready.
7. Which apple would you choose?
8. The children choose big apples.
9. Shine my shoes.
10. The sheep were sheered yesterday.
11. The scissors are for cutting.
12. Please make the decision.
13. Choose the round puzzle.
14. I go to sleep at night.
15. I am thirsty, I'd rather have some water.
16. The teacher's window faces my house.
17. I don't want that.


Exit Interview:

Thank you, I enjoyed playing this game with you.




72






Directions for Administering the BSRT:


1. Use two tape players--one for the recorded tape
and one that will record the session.

2. Begin with the entering interview to gain the
interest of the child.
3. Record the entering interview and then play it
for the child.

4. Give the following instructions:

a. We're going to play a little repeating game.

b. I'll read something and I want you to repeat
exactly what you hear, as you hear it.
Remember, repeat what you hear exactly as
you hear it.

5. Start the tape players.

6. Turn off the players after the session and
excuse the child.

7. If the child does not seem to understand the
directions, start the directions again from
the beginning.

8. If necessary, give the directions to the child
in Spanish, but the repetitions must be made
in English.








Scoring Form for BSRT-1975


NAME OF CHILD ____________NUMBER OF SIBLINGS _________


AGE

DATE OF GRADING


PRONUNCIATION


STRUCTURE


1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

ll. 12. 13. 14. 15. 16. 17.


1.

2.

3.

4.

5.

6.

7.

8.

9.

10. 11.

12. 13. 14. 15. 16. 17.



TOTAL


TOTAL


Grader









Corrections for the BSRT--1975

Instructions:

1. Grade each part separately.

a. pronunciation

b. structure

2. Record the name and other information from the
rec,.-ding on the grading form.

a. Play first part of tape, record the
answers.

b. Make certain that all is recorded before
beginning the analysis for grading.

c. If any part of the recording is blurred
or unintelligible, mark the space UN. If
there is no response, mark the space NR.

3. Pronunciation: listen to one sentence at a time
noting the pronunciation of the words as compared to the prepared model. The word indicated on the pronunciation correction sheet must be pronounced
according to the model's sounding.

a. If correct, mark 1 for each correct response;
if incorrect, mark with "0."

b. All scores for the pronunciation section should
appear under the section so marked on the
correction form.

c. There is a total of 20 possible points.

4. Structure: listen to one sentence at a time
noting the structure indicated on the structure correction sheet. The response must be exactly
as given on the correction sheet.

a. If correct, mark "1" point for each grouping.
If the structure is not exactly as indicated,
make written notes of the variation.


b. There is a total of 10 possible points.











Corrections for the Pronunciation Section of the BSRT



1 .............. chair

2 ............... teacher .........

3.

4 ..... berries ...............

5 .......... fresh

6 ..... think .......

7. Which ................ choose?

8......children choose .............

9. Shine ...................

10 ..... sheep ...........sheered ........

11 ......... scissors ......... cutting.

12 ................... decision

13. Choose ............... puzzle.

14.

15 ............ thirsty ........ rather .......

16 ......... teacher ............

17.











Corrections for the Structure Section of the BSRT



1. girl sits 2. gill asks 3. dog barks

4. berries are ripe

5.

6.

7.

8. big apples


round puzzle go to sleep I am thirsty or window faces my house don't want or


I'm thirsty


Jo not want


or don't like


9.

10. 11.

12. 13. 14. 15. 16. 17.












Subject Information Sheet--BSRT, 1975


NAME and NUMBER Birthdate Number of sibL.ings Language of the home IQ

Test date Scores SRT

Pronunciation


Age in months


Attended Kindergarten





Structure


Comments:











Evaluation of Scores of BSRT - 1975



Pronunciation Section:

20--16 indicates that child has gained control of the given sound patterns

15--11 suggests item analysis to determine the child's needs

10--0 child should begin initial auditory program which emphasizes the sounds which are distinctly English in nature.


Structure Section:

10--7 indicates that child has gained control of the given language structures

7--5 suggests item analysis to determine the child's needs

4--0 child should begin with initial structure and pattern program.









































APPENDIX C

RESOURCE PERSONS: QUALIFICATIONS
AND INFORMATION












Qualifications for Resource Persons:


1. Has worked with Colorado BBE programs for
at least three years.

2. Is bilingual in Spanish and English as
demonstrated by continued employment in
a Spanish/English BBE program as a teacher
or staff person.

3. Participates in the training sessions conducted by the researcher.





Training Sessions:

1. There will be two training sessions.

a. The first will give background on
the process of scoring SRT.

b. The second session will allow
practice in grading SRT using selected tapes from the pilot
testing.

2. Upon completion of training, the resource
persons will begin to score the tapes
collected in the major study.










RESUME FORM


NAME Alice Carolyn Aguayo de Leal ADDRESS 1616 27th Avenue Court; Greeley, Colorado 80631 EXPERIENCES IN COLORADO BILINGUAL PROGRAMS


5 years--Johnstown Bilingual Program, Kindergarten

level bilingual teacher

1 year---Greeley Bilingual Programs, curriculum

specialist


ADDITIONAL INFORMATION


Consultant--Teaching in a Second Language

Instructor--University level courses in bilingual education

Curriculum development for oral language development Team member of Ad Hoc Team to recommend NEA position

in regard to the Child Development Associate

Consortium, Washington, D. C.









SIGNATURE--indicates intentions to participate in training
sessions








RESUME FORM


NAME Estella Anita Martinez de See ADDRESS Greeley, Colorado 80631



EXPERIENCES IN COLORADO BILINGUAL PROGRAMS


5 years--Johnstown Bilingual Program

1 year-English reading teacher, first grade bilingual classroom

2 years-bilingual classrcom teacher, third grade

1 year-bilingual classroom teacher, fifth grade

1 year-bilingual classroom teacher, fifth, sixth, and seventh grades


ADDITIONAL INFORMATION

Workshop and/or Conference Consultant
Topics--Teaching Math Bilingually
Bilingual-Bicultural Education Oral Language Curriculum for BBE Special consultant for Senate Joint Resolution 20,
Bill for BBE-Colorado State Department of
Education

Developer--K-4 curriculum for bilingual-bicultural
project



SIGNATURE--indicates intentions to participate in training
sessions











RESUME FORM


NAME Sheila Anne Towle ADDRESS 2124 6th Avenue, Greeley, Colorado



EXPERIENCES IN COLORADO BILINGUAL PRCGRAMS


4 years--Johnstown Bilingual Program

2 years-kindergarten bilingual teacher 2 years-second grade bilingual teacher



ADDITIONAL INFORMATION


Secretary--Colorado Association for Bilingual
Bicultural Education

Assistant in developing elementary bilingual
curriculum

Member of evaluation team for elementary bilingual programs




SIGNATURE--indicates intentions to participate in training
sessions






































APPENDIX D

PERSONS KNOWLEDGEABLE OF LANGUAGE AND BBE








THE UNIVERSITY OF NORTHERN COLORADO
GREELEY, COLORADO 80639



March 14, 1975





Dr. Ruthelle.. Crews University of Florida Norman Hall - 190 Gainesville, Florida 32611

Dear Dr. Crews:

This is to affirm the fact that Bernice Bass has completed an independent study under my direction in which she developed an instrument that assesses the oral English syntax and phonology of first grade children in bilingual programs.

I believe this instrument has great potential for our Colorado bilingual programs, particularly because it has been modeled to focus upon the unique linguistic differences of Colorado.

Sincerely yours,



Jose E. C6rdova Assistant Professor of Elementary Education








COLORADO DEPARTMENT OF EDUCATION DENVER, COLORADO

March 26, 1975





Dr. Ruthellen Crews University of Florida Norman Hall - 186A Gainesville, Florida 32611

Dear Dr. Crews:

Ms. Bernice Bass has developed a measure that is appropriate to Bilingual Programs in Colorado. She has worked with various school personnel and school students and has been given the support for her measure in regards to relevance.

The Sentence Repetition Task that Ms. Bass has developed has been reviewed by our office. We feel that it provides appropriate exercises in assessing students' linguistic differences in the areas of phonology and syntax. The analysis lends itself very much to giving teachers and administrators information for bringing about proper techniques in presenting concepts to children with linguistic differences.

We are hoping that we can capitalize on the basic functions of the SRT and that we can continue to use the expertise that Ms. Bass possesses.

Sincerely,



Bernard Martfnez Bilingual Bicultural Education Community Services Unit 892-2166


cc: Bernice Bass









COLORADO DEPARTMENT OF EDUCATION DENVER, COLORADO



April 12, 1975





Ms. Bernice Bass
6905 Mariposa Street Denver, Colorado

Dear Bernice:

Thank you for the opportunity to review the BASS-SRT. This instrument, designed specifically to assess the English oral language skills of primary students in Colorado Bilingual/Bicultural Programs, is extremely needed.

We, in the Colorado Department of Education, look forward to the passage of legislation this year calling for the design and implementation of Bilingual/Bicultural Programs in school districts throughout our state. A necessary aspect of local program design might be addressed through the use of your instrument in the identification of where students are with respect to English oral language skills.

Please keep me informed of any new design features of your instrument.

Sincerely,



Jennie S. Green
Consultant
Community Services Unit (303) 892-2166


JSG/1w













BIOGRAPHICAL SKETCH


Bernice Marie Bass was born an only child in

Denver, Colorado, to Mr. and Mrs. Arthur Bass. Upon graduation from Manual High School in Denver, she went to the University of Northern Colorado in Greeley where a Bachelor of Arts degree was granted in 1970, with a Spanish major and a chemistry minor. During that time she worked as a laboratory assistant and electrocardiograph technician at Weld County General Hospital and as a Head Resident in the residence halls at UNC.

A Master of Arts degree was awarded in 1972 in

elementary education with special emphases in bilingual bicultural education and reading.

The author began her experience in bilingual bicultural education working with kindergarten children in the Johnstown Bilingual Program (Johnstown, Colorado), in 1971. Other Colorado teaching experiences also include: service in the Greeley Public Schools' Home Intervention Oral Language Development Program as a teacher of youngsters aged two to five years old; bilingual kindergarten teaching in the Ft. Lupton Bilingual Program; and teaching in the Brighton Public Schools. In addition, the author served as an Early Childhood Content Specialist in the Early Childhood









Component of the Educational Technology Demonstration Project sponsored by the Education Commission of the States and affiliated with the Federation of Rocky Mountain States during 1973.

In the Fall of 1973, Bernice Marie Bass went to Gainesville, Florida, where she taught Title I reading students at Stephen Foster Elementary School and trained paraprofessionals to assist children with reading problems while at the same time completing her year's residency work at the University of Florida.

The following year, the author returned to Colorado to complete additional college courses in Mexican American Studies and bilingual bicultural education at the University of Northern Colorado, Greeley, where she also served as a teaching assistant in Center for Human Enrichment and graduate assistant to the Director of Bilingual Bicultural Education. Presently, the author serves as a bilingual educational instructor with that University's Outreach and Special Programs Educational Service.

Ms. Bass' permanent residence is Denver, Colorado, where she is involved with schools and organizations interested in the preparation of bilingual bicultural teachers as well as the further development and advancement of bilingual bicultural education programs.









I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Ruhellen Crews, Chairperson Professor of Education



I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and -irality, as a dissertation for the degree of Docor o Philo phy.



MichaelvL. Hanes, Co-Chairperson Assistant Professor of Education




I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Vnce A. Hines
Professor of Education



I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




William Drummond
Professor of Education




Full Text

PAGE 1

ORAL ENGLISH LANGUAGE ASSESSMENT OF FIRST GRADE CHILDREN IN BILINGUAL BICULTURAL EDUCATION: EMPHASIS ON PHONOLOGY AND SYNTAX By BERNICE MARIE BASS A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OP THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1975

PAGE 2

COPYRIGHT 1975 BERNICE MARIE BASS

PAGE 3

ACKNOWLEDGEMENTS "Thank you"... only two words, yet so expressive of the appreciation I give to the many persons who gave so freely and completely of themselves so that I could become the person I am today. Great appreciation and deep regards to Dr. Ruthellen Crews my committee chairperson who spent so many precious hours with me as did Dr. Vynce A. Hines who deserves equal appreciation and regards. Many thanks to my other committee members Dr. Michael L. Hanes who served as co-chairperson and Dr. William Drummond. Special appreciation to all of the staff at the University of Northern Colorado in Greeley who shared their facilities and expertise which complemented my doctoral program, especialmente al doctor Jose Cordova. . .con muchisimas gracias. Also special thanks to the many persons who encouraged me "to keep on keeping on"... the most important ones being my parents Mr. and Mrs. Arthur Bass. Thank you to my typist Ms. Dorothy Long who so patiently brought the final product to reality. Also great appreciation and unending thanks to the National Fellowships Fund of the Ford Foundation that made it financially

PAGE 4

1 possible to make my dream come true. And last, but not least — muchisimas gracias a los personas y los ninos de los programas bilingiies de Colorado que me daban tanta ayuda. ; ...siem.pre a todos ... gracias . iv

PAGE 5

TABLE OF CONTENTS PAGE ACKNOWLEDGEiMENTS iii LIST OF TABLES vii ABSTRACT viii CHAPTERS I INTRODUCTION TO STUDY 1 Statement of the Problem 2 Limitations of the Study 3 Justification for the Study 3 Assumptions 4 Definition of Terms 5 Hypotheses of the Study 6 Procedures 6 Sample 6 Instrumentation and Data Collection 7 Data Treatment 8 Organization of the Study 8 II REVIEW OF SELECTED LITERATURE 10 Introduction 10 Oral First Language Acquisition 10 Second Language Acquisition 16 Measuring Oral Language 19 Teaching Methods 19 Observation 20 Testing 21 Conclusions 23 III PRESENTATION OF THE DATA 25 The Sample Population 25 Age 26 Language Spoken in the Home 26 Ethnicity of the Child 27 Number of Siblings in Child's Family , . 27 V

PAGE 6

TABLE OF CONTENTS (Continued) PAGE Kindergarten Attendance 27 Sex of the Child 27 Testing of Procedures 28 Bass Sentence Repetition Task (BSRT) 28 Contrastive Analysis of Various Selected English and Spanish ... Properties 28 Sounds 28 Structures 29 Pilot Testing of the BSRT 30 Peabody Picture Vocabulary Test (PPVT) 31 The Test Data 32 Bass Sentence Repetition Task (BSRT) . 32 Peabody Picture Vocabulary Test (PPVT) 33 IV STATISTICAL ANALYSES OF THE DATA RELIABILITY OF THE BASS SENTENCE REPETITION TASK INTERRATER RELIABILITY 34 Reliability of the Bass Sentence Repetition Task Inter-rater Reliability 34 Relationship of Scores on the BSRT and Selected Variables 38 Data and Hypotheses Relationships ... 40 Regression Analysis 43 Interpretation of the Regression Analysis 48 Summary or Regression Analysis 52 V SUMMARY, RECOMMENDATIONS AND DISCUSSION ... 56 Summary of the Study 56 Recommendations 59 BIBLIOGRAPHY 61 APPENDIX A LETTER FORMS FOR CONTACTING SCHOOLS 67 APPENDIX B BASS SENTENCE REPETITION TASK 70 APPENDIX C RESOURCE PERSONS: QUALIFICATION AND INFORMATION 80 APPENDIX D PERSONS KNOWLEDGEABLE OF LANGUAGE AND BBE . 8 4 BIOGRAPHICAL SKETCH 88 vi

PAGE 7

LIST OF TABLES Table 1 Mean and Standard Deviation of the Age of Subjects 26 Table 2 M^an and Standard Deviation on BSRT: Pronunciation and Structure 32 Table 3 Mean and Standard Deviation on PPVT of Subjects in BEE 33 Table 4 List of Variables 36 Table 5 Inter-rater Means and Standard Deviation 37 Table 6 Inter-rater Reliability of the Scores .... 37 Table 7 Correlation of Variables to BSRT Score Means 39 Table 8 Correlation Matrix of Predictor Variables 41 Table 9 Means and Standard Deviations (N=78) of Prediction Variables 45 Table 10 Grouping of Predictor Variables for Regression Analysis 47 Table 11 Regression Analysis of Variables for Pronunciation 49 Table 12 Regression Analysis of Variables for Structure 50 vii

PAGE 8

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ORAL ENGLISH LANGUAGE ASSESSMENT OF FIRST GPvADE CHILDP^EN IN BILINGUAL BICULTURAL EDUCATION: EMPHASIS ON PHONOLOGY AND SYNTAX By Bernice Marie Bass December, 1975 Chairperson: Dr. Ruthellen Crews Major Division: Curriculum and Instruction The purpose of this study was to develop a reliable instrument that could be used by teachers and staff in Colorado to assist in the assessment of oral English language, particularly the structure (syntax) and the pronunciation (phonology) of first grade children in bilingual bicultural education (BBE) , and to relate the scores of the instrument to seven variables: a) intelligence, b) the child's age, c) the language spoken in the home, d) ethnicity, e) the number of siblings in the child's family, f) kindergarten attendance, and g) the sex of the child. Seventy-eight subjects were identified for the study. These subjects were first graders who were participants in BBE programs in Colorado during the school year 1974-75. The children were tested using the researcher-developed instrument, the Bass Sentence Repetition Task (BSRT), v/hich viii

PAGE 9

was developed on the basis of the contrastive analysis of English and Spanish to determine knowledge of oral English pronunciation and structure and the Peabody Picture Vocabulary Test (PPVT) for testing intelligence. For the purpose of establishing reliability of the test instrument, language samples were collected using the BSRT. These samples were scored by three resource persons according to the rating system developed by the researcher. The Kuder-Richardson Formula 21 was applied to estimate the reliability of each section of the BSRT. The coefficient was .71 for pronunciation and .49 for structure. The combined judgments of three raters had the effect of tripling the length of the test. Applying the Spearman-Brown Formula to estimate a new reliability, a .88 coefficient for pronunciation and a .76 for structure were obtained. These established reliabilities for the BSRT are within acceptable limits for an instrument of this type. Scores from the BSRT were analyzed using correlation analysis as well as multiple regression analysis to determine their relationships to the predictor variables. Correlation analysis found that intelligence, language spoken in the home, and ethnicity were correlated to the BSRT pronunciation score at the .01 level and the variables of intelligence and language spoken in the home were significant at the .05 level to the BSRT structure score with the sex variable at the .01 level. ix

PAGE 10

The multiple regression analysis found that intelligence as measured by the PPVT to be a contributing factor to the measure of oral English pronunciation and structure as measured by the BSRT. Also the number of siblings in the child's family is considered a contributing factor in the prediction of oral English structure as measured by the BSRT. X

PAGE 11

CHAPTER I INTRODUCTION TO STUDY Within '"he United States live many groups of languagedifferent peoples. These are persons who speak a language other than English and/or a dialect other than standard English. In some parts of the United States, bilingual bicultural education is offered as a means of bridging the linguistic and cultural gap of language-different children by affording the opportunity to learn school facts and concepts in the native language while learning English as a second language. Other school systems offer programs which emphasize the development of oral English language to replace the native language spoken. Still other school systems give no consideration to language differences, and all students are required to communicate in English from the beginning of the school experience (Robinett, 1971) . In recent years, Colorado legislators have questioned the progress of the children exposed to the f orementioned programs. In order to assess the progress made by children in various programs, the Colorado Legislature designated funds for a state needs assessment (reflected by Colorado legislation, see Senate Joint Resolution 20, Colorado Department of Education, 1974) . As a result of the assessment 1

PAGE 12

2 which indicated that a considerably large number of children in Colorado public schools were in need of bilingual education, a bill requiring that public school systems provide such an education was drafted and approved by the legislature. This bill provides financial appropriations to Colorado school districts having fifty students or more than ten percent of the student population identified as being linguistically different (Earle, 1975). Statement of the Problem To enable teachers and staff to evaluate effectively the oral English of children participating in bilingual bicultural education (BBE) programs, the teachers of Colorado have needed reliable tools which could be used to assess the phonological as well as the syntactic abilities of their pupils. At the beginning of this study, there were no oral English assessment tools for identifying the English language differences of language-different pupils designed to take into consideration the specific regional dialect differences for use with youngsters in Colorado. Therefore, the purpose of this study was twofold: 1) to develop a reliable instrument that would enable teachers and staff to assess the phonology and syntax of the oral English of first grade children in BBE programs in Colorado and 2) to relate the results from this instrument to seven variables: a) intelligence, b) the age of the child, c) the language spoken in the home (i.e., English or Spanish), d) the ethnicity

PAGE 13

3 of the child (i.e., Anglo or non-Anglo), e) the number of siblings in the family, f) kindergarten attendance, and g) the sex of the child. Limitations of the Study The researcher's study was bound by the following limitations : 1. The study was limited to 78 children in the first grade in Colorado BBE programs identified by the Colorado State Department of Education during the academic year 1974-75. 2. The researcher was only concerned with the development of an instrument that could help Colorado teachers and staff assess the oral English syntax and phonology of their students . 3. The variables considered in the study were confined to intelligence, age, language spoken in the home, ethnicity, number of siblings in the family, kindergarten attendance, and sex of the children participating in the study. The extent to which the findings in this study can be generalized to the populations in other BBE programs is limited because the sample population v/as taken from one specific area and gave consideration to the regional dialect spoken in Colorado. Justification for the Study The need for an instrument that could be used by teachers and staff in Colorado BBE programs to help assess the oral English of first grade children has been recognized

PAGE 14

4 for some time. In the needs assessment (Colorado Department of Education, 1975) mention was made of the lack of reliable instruments to measure the abilities of language-different children. The need for such instruments is becoming more widespread particularly with the growing number of programs in the state. The information collected in this study could serve as a model for the development of additional instruments which would expand on the test items, sample other linquistic features and identify additional information about the language used as well as be of assistance to others who are interested in the measurement of the abilities of the languagedifferent for which this was developed. This information could even be helpful to those who are dealing with other types of populations, such as school programs for Vietnamese refugees, both as an example of how to devise sentence repetition task and how to use the results. Assumptions The basic assumptions made for this study were that the English language syntax and the phonology of first graders in BBE programs could be assessed, and that other variables extraneous to the language itself have some effects on the acquisition of the language children use in the first grade .

PAGE 15

5 Definition of Terras To assist the reader, the following terms have been defined as used by the researcher: Bilingual Bicultural Education (BBE) An educational setting in which two languages are used as the medium of instruction and the respective cultures of the languages are ref^OiCted and actively incorporated in the classroom setting. Bilingual A person who speaks and understands two languages and has displayed an understanding of the cultures of the two languages. English-speaking A person who speaks English as well as understands it. This person may speak any one of a number of English dialects. English-only-education Environment (EOE) An educational setting in which English is the only language used as the medium of instruction. Language-di f f erent Those who are identified as speaking a language other than English and/or a dialect other than standard English. Mexican American or Chicano Those individuals of Spanish surname and/or those who refer to themselves as Chicanes and are identified as such by school records. Oral Language Development Defined as a function of the measurements on the sentence repetition task. Phonology The study of the sound system of a language Phonology encompasses the understanding of, and the ability to hear and produce the sounds of a language. Regional Dialect A variety of Standard English (or the standard form of any other language) as spoken by the educated population of a given region. Sentence Repetition Task (SRT) The instrument designed for use in this study to gather and assess oral language samples. This instrument used outside of this study will be known as the BASS SRT, 1975. Syntax That part of grammar which studies the construction and arrangement of words in a sentence.

PAGE 16

6 Hypotheses of the Study The present study considered the following hypotheses: 1. There is no relationship between the scores on the BSRT and the intelligence score on the Peabody Picture Vocabulary Test . 2. There is no relationship between the scores on the 3SRT and the age of the child. 3. There is no relationship between the scores on the BSRT and the language (a. Spanish, or . b. English) spoken in the home. 4. There is no relationship between the scores on the BSRT and the ethnicity (a. non-Anglo, or b. Anglo) of the child, 5. There is no relationship between the scores on the BSRT and the number of siblings in the family. 6. There is no relationship between the scores on the BSRT and kindergarten attendance. 7. There is no relationship between the scores on the BSRT and the sex of the child. Each of these hypotheses reflects the review of literature and. the information some authorities have found to have some effect on the oral language development of children. Procedures In order to test the hypotheses as previously stated, the following procedures were used: Sample The sample population for this study was drawn from children who attended the first grade during the academic

PAGE 17

7 year 1974-75. These pupils were selected from participants attending BBE programs from five different schools in three areas of Colorado. Instrumentation and Data Collection The researcher developed the Bass Sentence Repetition Task (BSRT) which was recorded by a person with a regional English dialect identified as a Coloradan. The sentences constructed for the instrument reflected local English usage and the contrastive analysis of the linguistic features of English and Spanish. This particular BSRT was designed to assess the use of the specific linguistic features of phonology and syntax. In the BSRT, the oral English samples were divided into the categories of oral English structure and English pronunciation for the purpose of scoring. The scoring was done by a group of three resource persons who received training in the techniques of interpreting the SRT. The Peabody Picture Vocabulary Test (PPVT) provided a measure of the subjects' verbal intelligence through measuring their listening vocabulary. Other information relating to the subjects, i.e., the number of siblings in the family, the age, the language spoken in the home, ethnicity, kindergarten attendance, and the sex of the child was obtained from school records and through teacher interviews.

PAGE 18

8 Data Treatment All treatment samples were scored by three resource persons with expertise in the language and culture of the sample population and v/ere trained in the use of the evaluative procedure developed for assessing the language samples. This procedure required that the corrections be made from a contrastive analysis of Spanish and English. The scores from the instruments were compared with the seven f orementioned variables by employment of the technique of multiple regression. Organization of the Study The research study is presented in five chapters with a bibliography and an appendix. This first chapter has served as an introduction to the study, as well as a description of how the study was carried out. Chapter 2 contains a review of selected literature and research that is related to the topic of the study. This review led to the formulation of the hypotheses. Chapter 3 presents data from the study and a description of how it was collected. Chapter 4 includes the analysis of the collected data and the conclusions drawn in relation to the stated hypotheses . Chapter 5 summarizes the study and includes recommendations based on the findings.

PAGE 19

9 The bibliography contains all sources and references of information and precedes the appendices. The appendices include letters, forms, and other information used in contacting school BBE programs. Also copies of the BSRT used as well as the qualifications and information on the resource persons, and other supportive information relating to this study are found in the appendices.

PAGE 20

CHAPTER II REVIEW OF SELECTED LITERATURE Introduction The concern of this study was focused on developing a tool for assessing the oral English of those children in Colorado v/ho v;ere first graders and who were participating in bilingual bicultaral education (BBE) . A consideration of the development of assessment tools for effective evaluation of the oral English of children participating in BBE programs requires a preliminary understanding of several fields of knowledge which relate to the question. First, how is oral language acquired? Next, how is a second language acquired? Third, how can oral language be measured? Oral First Language Acquisition All normal children master their particular first language and acquire all the basic operations of adult grammar by the age of four regardless of native language or social settings, even though they may not verbally use these operations at all times (McNeill, 1971; Slobin, 1972; and «ASs, 1974) . The young children extract sounds from the language that is spoken to them and with them (Chomsky, 1965; 10

PAGE 21

11 and Katz, 1966), but they are not conscious that they are doing this and cannot say how they go about doing it (McNeill, 1971). Research by Ervin-Tripp and Miller (1964) substantiates the theory that early language development reflects generalizations of adult operations. Language includes some degree of imitation but more important, includes the ability to generalize. For example, a young child will say "dog," "toy," or "foot," but will over-generalize in making plurals which may result in "dogs," "toys," or "foots." Similar findings v/ere noted in the case of the regular past tense: "walk" became "walked," "sell" became "selled," "go" became "goed." The ability to generalize and to extend production Of language is basic to first language learning. First language learners acquire the ability to use language through sequential steps (Slobin, 1972). Menyuk's (1969) research states that levels of language development can be delineated and studied; therefore, she correlates the process of language development with the process of maturation . In the initial stages of language acquisition, the child demonstrates an unlimited phonetic ability (SavilleTroike, 1973b) . This phonetic ability soon is further developed to include single words. During the next stages the child learns noun and verb inflections leading to the acquisition of syntax. By age four or five, the child has

PAGE 22

12 acquired essentially all basic language structures (Chomsky, 1969). Beyond this age level to about puberty, the child continues to refine his or her language learning (SavilleTroike, 1973b). The learner of the first language also acquires the ability to usp the language communication skills: listening and reading as well as speaking and writing (Zintz, 1969). Through the skills of listening and reading language is received. As the learner matures, his or her ability to decode what is received through these skills is increased. The skills of speaking and writing serve as a means of transmitting language. Likewise, as the learner gains more experience, the ability to transmit language increases (Ruddell, 1974). Chomsky (1965) emphasized that the observation of the child's use of the comjnunication skills (language performance) is not the only component of language, but that usage (language competency) is also important. Language performance can be observed directly, but language competency is that which underlies observable language performance (Ruddell, 1974). Jakobovits ' (1971) findings further support this theory. Understanding usually was observed to be superior to speaking, with understanding equated to competence and speaking to performance. At more advanced levels, both understanding and speaking were viewed as performance.

PAGE 23

13 McNeill's (1966) observations are somewhat similar. Phonological production was viewed as preceding graminatical comprehension which preceded grammatical production. An example of this is the child who repeats a long sentence but might not be able to understand or produce it. The same child may understand a longer sentence but might not be able to produce it. The research of Fraser, Bellugi, and Brown (1963) parallels these findings. The child's language development is greatly influenced by two critical factors: 1) the ability to formulate and grasp generalizations about the language system, and 2) the language model or models in the environment (Ruddell, 1974) . Although early home environment plays a major role in a child's language development, the teacher's language and that of other children exert an influence on language development. Dailey and Neyman (1965) studied oral and written facility and comprehension among children v/ho had attended kindergarten and those who had not. The kindergarten youngsters composed the experimental group, and conclusions showed that these children did significantly better in all areas and demonstrated fewer oral language errors than the other children did. Further studies involving the factors of kindergarten attendance and oral language include the research of Pendergast et al > (1966) . In order to investigate the effect of kindergarten on oral articulation of first grade children.

PAGE 24

14 over 15,255 pupils were tested. Those children who had attended kindergarten did not appear to have fewer oral English errors when they entered the first grade than those children who did not attend kindergarten, and of the twothirds of the children who did have some articulation problems and were not enrolled in speech therapy, kindergarten experience appeared to help reduce sound errors in the first grade . Oral language skills served as a measure in Stanchfield's (1971) test of the effects of teaching prereading skills to kindergarteners. Results indicated that girls outperformed the boys in first language learning. Among children of average intellectual ability, it is not unusual to find one child who does not seem to be performing at an expected level. Often this child is a boy. This is attributed to cultural expectation according to Ruddell (1974). Girls are expected to play passive games which are oriented to school life while boys are expected to be active. Kagan (1964) found that girls tended to be more oriented to school activities and to perform better than boys at the elementary grades. On the other hand, Laney (197 0) concluded that kindergarten males and females showed no differences in expressive vocabulary development in their first language.

PAGE 25

15 Stanchfield' s (1971) research also demonstrated that there are differences in achievement by ethnic groups. Children representing three ethnic groups were selected, all of which were attending kindergarten and represented a cross section of socioeconomic levels. Oral language skills were included as a measure in the testing and results showed that whites outperformed the other ethnic groups involved in the s tudy . Likewise, Laney (197 0) observed the relation of expressive and comprehensive language to race and cultural background and concluded that these variables were related to the level of a child's language development. Cohen and Klein (1968) demonstrated that language accuracy as measured by the ability to use communication skills increased with age in first language learners. In order to master the more complex elements of language, Piaget (19 55) demonstrated that the normal child needs additional maturation time. Furthermore, Asher (197 2) concluded after studying a population of three elementary grades to determine if task had any influence on the performance of various ages of children in oral English, that younger children's poor communication skills in the first language cannot be explained by limited explanation of the task demands. Davis (1937) found that single children in a family developed language facility more rapidly than children with siblings, and that chidren with siblings developed language faster than twins did. Research by Gesell etal. (1940, 1946)

PAGE 26

16 found similar results. Children who come from families with many brothers and sisters tended to talk later than those children who were single children. The children who were twins also tended to talk later than single children. A single child in a family tended to develop oral English abilities earlier than those children who had siblings because of the greater opportunity and need to interact with adults. The same was found of children who were the first born in most families. Second Language Learning The degree of second language learning is dependent on the time at which it is introduced as well as the manner in which it is presented (Smith, 1971) . If the second language is learned at the same time the first language is learned, acquisition does not differ from that of the first. However, when the second language is acquired in another context, the learner does not go through the first language acquisition process (Saville-Troike, 1973b) . Smith (1971) observed that second language is learned by a combination of presentation, explanation, repetition, and transfer. Second language learning is an inferred process, building up rule by rule while first language learning permits "trial and error" learning. The setting for second language acquisition generally minimizes possibilities of errors, in an effort to develop the competence of a full native speaker (Cook, 1971) .

PAGE 27

17 Jakobovits (1970) found mastery of a second language to be based on a good grounding in the first language. He supports the age of four or five years as an optimum time to acquire a second language. At this time, an individual's cognitive development is at a later and more advanced stage, he is in possession of the grammatical structure of a language which may facilitate acquisition of another language through transfer, and he possesses concepts and meanings in the first language. Penfield's (1960, 1965) research substantiated that second language learning is a function of brain flexibility. Up until the age of twelve or puberty, the uncommitted part of the brain can become involved in language learning. Ervin-Tripp ' s (1968) work found that second language learning in children up to the age of eleven years is based on the sound of the language while adults based such learning on language sense. Furthermore, foreign accent was observed as being directly related to the age at which the second language was introduced, the accent being more pronounced after age twelve (Lenneberg, 1967) . All physiologically normal children are born with the capacity to produce any sound used in any language; however, with age, the flexibility is lost. "Foreign sounds" are not heard as such; the learner has been conditioned not to hear through previous experience with his own language. Problems in phonology, grammar, and vocabulary are a tendency as a result of one's previous first language

PAGE 28

18 experiences. These problems of perception in relation to the use of second language which reflect the first language habits are referred to as "linguistic interference." Many interference problems can be predicted through a contrastive analysis of the languages involved (Saville-Troike , 1973b) . As a result of linguistic interference, some psychologists propose that simply learning two languages at the same time may hinder a child's intellectual development; that the impairment of intellectual functioning arises in the non-English-speaking child solely from the child's using, or being asked to use, two languages; that learning two languages at the same time may lead to use of a smaller vocabulary, simplier sentences, confused word order, misuse of negatives, misuse of idiomatic expressions, and the use of literal translations (Smith, 1949; and Darcy, 1963). Darcy's (1963) review of the literature relating bilingualism and intelligence showed that a child's intellectual development is often hindered by his use of two languages. Also intellectual functioning as measured by standardized tests appears to be impaired. The children who come to school speaking another language use confused word order, misuse negatives and idiomatic expressions, and use, at times, literal translations. Scores were also low on intelligence tests. The scoring on the intelligence tests tended to be correlated to their English language performances .

PAGE 29

19 Young Spanish-speaking children in Arizona who were introduced to English-only language instruction before their Spanish was well-established made a variety of oral language mistakes in Spanish (Lynn, 1945) . Abraham (1960) found that children who were given insufficient English-asa-second language training performed one standard deviation below the national norm on standardized tests of mental ability. Measuring Oral Language According to the Encyclopedia of Educational Research (Ebel, 1969) there are three general methods of studying language: teaching (conditioning), observation, and testing Teaching Methods One method of teaching is behavior modification (stimu lus-response-reward) . This method is used by Bereiter and Englemann (1966) in their early childhood program titled Distar. The participants in this program are conditioned to respond in a specific manner according to the directions given by their teacher. Other examples of teaching methods (Smith, 1971) are methods used in foreign language teaching. The audiolingual approach uses an aural-oral technique, "hear itsay it." Another method, total submersion, places the learner in the cultural environment of the new language generally prohibiting use of any other language except the

PAGE 30

20 new one. Rote memorization is the technique of direct memorization of grammar rules with stated examples. Observation Observation has served as a means of language study in many research projects. Both N. Chomsky (1965) and C. Chomsky (1969) used the method of eliciting speech from youngsters by engaging children in informal conversation to observe their language and to make generalizations about the speech of their subjects. Loban (196 6) spent ten years observing children's language which resulted in the development of Categories for Tallying Problems in Oral Language as a means of assisting teachers in identifying crucial and frequent oral language deviations from standard English made by Negro and Caucasian subjects at pre-school, primary, intermediate, and junior high levels. Allred (1970) elicited samples of free speech from third grade Mexican American children with Spanish language backgrounds in Colorado and did an analysis of errors in English usage. The analysis of the language samples found misuse of negatives and of common expressions as well as the misuse of verb forms. These v/ere defined as the effects of language differences. From these findings, Allred recommended further evaluation of the pupils' weaknesses though diagnosis followed with an instructional plan based on the specific items which are sources of difficulty.

PAGE 31

21 Testing Testing is the general method selected by most researchers in their studies of language. The results of testing, however, are of no benefit to the examiner or to the examinees unless the findings are used selectively. If not, testing becomes another means of labeling for failure. Selective use of test results warrants the use of the information to develop additional prescriptive procedures that give positive and special attention to identified problem areas. The Gloria and David Beginning Language Tests (1958) were developed in Texas to assist in the identification of phonological and morphological elements that are difficult for primary youngsters whose speech differs from the recorded model. The instrument tests by the method of sentence repetition. After use by a considerable number of persons, a questionnaire was developed to accompany the test. Natalicio and Williams (1972) constructed Questionnaire for Evaluations of Black Language Samples; Questionnaire for Evaluators of Mexican American Language Samples . These questionnaires serve as a means of evaluating oral language samples derived from the Gloria and David on a seven-point scale except for two binary (yes-no) items. Reliability was established by identifying experts specializing in dialect study who used the questionnaire to evaluate a number of language samples (ten Mexican American samples and ten Black samples) .

PAGE 32

22 In 1968 Olquin constructed a non-technical teaching aid for teachers of bilingual children titled Shuck Loves Chirley which includes a diagnostic test of sentence repetitions. This instrument has been available through the California Public Schools. The materials were developed to assist teachers in the identification of sounds which the child cannot hear. The Linguistic Capacity Index developed in 1964 by Manning and Brengleman (Fagan, 1975) was developed as a measure of English language readiness in grouping native Spanish-speaking pupils for English language instruction at the primary level. The index includes three sections: vocabulary recognition, contrastive phonology, and contrastive grammar. Vocabulary Recognition measures recognition of verb, noun, preposition, and adjective forms. Contrastive Phonology measures ability to discriminate sounds of pairs of words. Contrastive Grammar measures understanding of English function vrards, word order, and inflectional constructions that differ from Spanish constructions. The test has been used by the developers with over 167 subjects. In addition, it was administered to 300 kindergarten children in U.S.O.E. Project 2821 and to more than 2400 first graders in Cooperative Research Reading Projects 2648 and 2734. The Bilingual Syntax Measure (Burt, Dulay, and Chavez, 1975) was developed to measure children's oral proficiency

PAGE 33

23 in English and/or Spanish grammatical structures by using natural speech as a basis for making judgments. This instrument is designed specifically for children ages four to nine years. The instruments and techniques of studying language that are described here are a sampling of the types of materials that have been available for oral language assessment, some of which are specifically for BBE. The researcher has selected the method of testing to study oral language, more specifically, the testing method referred to as sentence repetition task. Conclusions The literature relating to first and second language acquisition indicates that many factors affect oral English language ability. The factor of intelligence is effected by the timing of and the preparation for second language learning (Darcy, 1963; Abraham, 1970). Second language learning can hinder intellectual development (Smith, 1949). Success in second language acquisition is dependent on age (Lynn, 1945; Penfield, 1960 and 1964; and Lenneberg, 1967) , just as first language acquisition is a function of age (Piaget, 1955; Ervin-Tripp Miller, 1964; Cohen and Klein, 1968) . Saville-Troike (1973b) also supports the concept that all language is learned in stages. Smith (1971) demonstrates that the manner of second language presentation affects its acquisition.

PAGE 34

24 Stanchfield (1971) and Laney (1970) correlated cultural and ethnic factors to language development. In addition, Stanchfield, Ruddell (1974), and Kagan (1964) demonstrated that young girls outperformed boys of the same age while Laney (1970) found no differences. Davis (1937) and Gesell (1940 and 1946) reported that the number of siblings in the child's family affects oral language development. Dailey and Neyman (1965) identified kindergarten attendance as being positively correlated with the number of oral language errors made. On the other hand, Pendergast (1966) found the opposite to be true. Knowledge of some of the effects of these factors led to the development of the previously stated hypotheses. The past research has either supported or questioned the extent to which there is a relationship of these factors to the acquisition of oral English language, particularly the oral English of children whose native language is not English The next chapter will describe the design of this study as well as present information about the instrumentation, the subjects, and the data collected in carrying out the study.

PAGE 35

CHAPTER III '"•1-. " " . ' —.'^ 4 PRESENTATION OF THE DATA The teachers and staff of Colorado bilingual bicultural education (BBE) programs need reliable tools to assist them in the assessment of the abilities of the students they serve so that students may receive immediate assistance in acquiring oral English language skills. The purpose of this study was to develop such an instrument and to determine the relationships between scores on this instrument with seven identified variables: 1) intelligence, 2) age of the child, 3) the language spoken in the child's home, 4) the ethnicity of the child, 5) the number of siblings in the child's family, 6) kindergarten attendance, and 7) the sex of the child. i ' . The Sample Population " * ' :' The sample population in this study consisted of seventy-eight BBE first graders who were participants during the academic year 1974-75, in three Colorado BBE programs: one in northern Colorado, one in south-central Colorado, and one just south of the Denver area. Of the programs selected, five schools were used representing populations in rural, urban, and semi-urban settings, all of which reflected the same regional dialect usage. 25

PAGE 36

26 Permission to work with youngsters from BBE programs was gained through contacts suggested by the BBE liaison with the Colorado State Department of Education. Information other than test scores regarding the subjects was made available to the researcher from the schools' records and through teacher interviews. Age The age of each child in the sample was obtained by using the dates of birth from the school records and calculating the age of the child in montns to the date of testing. The age distributions in months of the subjects who participated in the study are in Table 1. The ages ranged from 79 months to 111 months with a mean of 86.6 and a standard deviation of 5.35. Table 1 I4EAN AND STANDARD DEVIATION OF THE AGE OF SUBJECTS Mean S. D. Range N = 78 86.6 5.35 79-111 months Language Spoken in the Home The language spoken in the home was determined by the teachers of each of the subjects based on their knowledge of the child's home and the teacher's interaction with the family

PAGE 37

27 of the subject. Only speakers of the languages Spanish and English were considered. Of the total population, 9 of the boys and 12 of the girls were native speakers of Spanish. Twenty^three boys and 34 girls were native speakers of ' English. Ethnicity of the Child Ethnicity was used to refer to the ethnic representation of the child. The subject was classified by the teacher of each child as being a) non-Anglo, or b) Anglo. Among the participants of this study, 19 boys were non-Anglo and 50 girls were non-Anglo. Thirteen boys and 16 girls were Anglo. Number of Siblings in Child's Family If school records did not carry information regarding the number of siblings in the family, the child was asked and the number was then verified by the classroom teacher. The number of siblings in the homes of the sample population ranged from zero to 12. Kindergarten Attendance 'I*' Kindergarten attendance was determined by the fact that a subject had attended kindergarten for any length of time prior to entering first grade. In the sample population, 75 attended kindergarten and 3 did not. Sex of the Child The sex of the child was noted at the testing session.

PAGE 38

28 Included in the study were 32 boys and 46 girls making a total of 78 participating in the study. The Testing Procedures Bass Sentence Repetition Task (BSRT) Samples of the participants' oral English language were collected on cassette tapes using the BSRT developed by the researcher (see Appendix B) and audiotape recorded by a person with a regional English dialect that the children were accustomed to hearing. The seventeen sentences constructed for the BSRT reflected the specific linguistic features identified as being significantly English as the result of the contrastive analysis of English and Spanish according to the following (Politzer and Stanback, 1961; Shane, 1967; Zintz, 1969). Persons knowledgeable of the field were also consulted and found to be helpful (see Appendix D) . ' ' , • j v.; Contrastive Analysis of Various Selected English and Spanish Language Properties Sounds Spanish |c| palatal-affricate voiceless no equivalence lb] bilibial-stop-voiced i/6] bilibial-fricative-voiced English 1 c I church I s I shoe |b| boy, buy, berry

PAGE 39

29 Sounds no distinction from the b only one labiodentalfricative f none none |.u| alveolar-lateralvoiced none nona Structures Spanish All verbs are inflected (example: The dog run) Adjectives usually follow the noun (example: The man big) Use of reflexive in many cases (example: I put myself to sleep) Use of no and double negation (example: Juan no go home; No go: I don't want that, not never ever) No auxiliaries used (example: He works? Works my mother? ) English I V I very labiodental-fricative voiceless I e I thick dental-fricativevoiceless [3 I rather dental-fricativevoiced none glottal stops I Z I decision English "S" as a part of present (example: The dog runs) Adjectives before the noun (example: The big man) No separation of certain verbs (example: I go to sleep) Seldom use of double negation (example: Juan did not go home . Don ' t go ! I don't want that) Use of auxiliaries (example: Does he work? Does my mother work? )

PAGE 40

30 anish English Use of "to be" (example: I am seven years old; I am hungry; I am thirsty) Nouns do not have gender (example: My car goes) Phrases express differently (example: my watch runs; I missed the bus; the window faces the street) It should be understood that the BSRT is not a '' normed instrument but is reflective of standard language knowledge. The BSRT is designed to aid in determining whfether the child is repeating the sentences as given or whether the child is restructuring the sentences (i.e., changing to a structure that is not acceptable as equivalent to that which is presented) . This BSRT was constructed to assess use of the specific linguistic features of phonology and syntax. For purposes of scoring, the sentences that were constructed to assess knowledge of phonology were scored as one category labeled English pronunciation and those constructed to assess English syntax were scored as another category labeled English structure. Therefore, two scores were given for this test. Pilot Testing of the BSRT . ; The BSRT developed for this study was pilot tested in March, 1975, by the researcher. The testing involved a group Use of "to have" to express English "to be" (example: I have seven years; I have hunger; I have thirst) The use of gender for nouns (example: My car she goes) Tendency to translate idiomatic phrases (example: my watch he walks; the bus leaves me; the window gives to the street)

PAGE 41

31 of 24 children representing a BBE program in the Northern Colorado area. The children were first graders and v;ere selected randomly from the children in the BBE program. The BSRT was administered individually to each of the examinees by the researcher as was the PPVT. The purpose of the pilot test was twofold: 1) to familiarize the researcher with the procedure established for the administration and scoring of the BSRT, and 2) to determine if the sentences selected for the BSRT were functional to provide data appropriate for assessing oral English phonology and syntax. . • The analysis of the language samples derived from pilot testing of the BSRT was done three times by the researcher according to the rules for grading (see Appendix B) . Peabody Picture Vocabulary Test (PPVT) The PPVT is a standard test designed to provide an estimate of a subject's verbal intelligence through measuring his listening vocabulary (Dunn, 1S65) . It is administered individually to examinees. Each examinee is asked to select the illustration from a set of four that represents the word given orally by the examiner. According to the test manual (Dunn, 1965) , correlations ranged from . a low of 0.67 at the six year level to a high of 0.84 at the 17 and 18 year levels, with a median of 0.77 as the resulting coefficients of equivalence for raw scores and standard errors of measurement for IQs.

PAGE 42

32 The researcher was aware of available data which indicate that the PPVT has elements of cultural bias against bilingual children (Hickey, 19 72), but for purposes of this study, this instrument seemed to be about as effective as any available. The Test Data Bass Sentence Repetition Task (BSRT) All BSRT's administered for this study were done so by the researcher. Scoring of the data from the BSRT's was done by a group of three persons who met specified qualifications as developed by the researcher to provide for inter-rater reliability (see Appendix C) and in accordance with the rules set for their interpretation (see Appendix B) which reflected the contrastive analysis of English and Spanish. An average of the scores of the three resource persons who graded each part of the test was computed. Each child received a score on pronunciation (phonology) and a score on structure (syntax) . Table 2 MEAN AND STANDARD DEVIATION ON BSRT: ' = v^PRONUNCIATION AND STRUCTURE Pronunciation Structure Mean S . D. Range Mean S. D. N = 78 16.6 2.78 8-20 8.45 1.47 Range 1-10

PAGE 43

33 The scores for pronunciation ranged from 8 to 20 for the sample with a mean of 16.6 and standard deviation of 2.78. The scores for structure ranged from 1 to 10 with a mean of 8.4 and standard deviation of 1.47 (see Table 2). Peabody Picture Vocabulary Test (PPVT) The PPVT v;as administered and scored by the researcher. Table 3 shows the deviation of intelligence scores for the subjects participating in the study. Table 3 MEAN AND STANDARD DEVIATION ON PPVT , " OF SUBJECTS IN BBE iMEAN S. D. RANGE N = 78 92.7 15.1 26-116 Note that the range of scores was from 26 to 116, more specifically, the girls ranged from 26 to 114 and the boys from 67 to 116 with a mean of 92.7 and standard deviation of 15.1. The details of the data analyses and results will be found in Chapter 4 .

PAGE 44

CHAPTER IV STATISTICAL ANALYSES OF THE DATA This chapter includes the analyses of data collected for establishing reliability of the Bass Sentence Repetition Task (BSRT) developed by the researcher and the relationship of the scores on the BSRT and the stated hypotheses. (Stati tical analyses of the data collected in this study were made by the use of the computer system at the University of Northern Colorado, Greeley.) Reliability of the Bass Sentence Repetition Task The pupils used in this study were first graders who were participants in Colorado bilingual bicultural education (BBE) programs during the school year 1974-75. Seventyeight subjects were selected randomly from those attending such programs. Each subject was tested individually in an area removed from his regular classroom setting to limit external noises during the recording of language samples. The BSRT was administered to assess the oral English phonology and syntax of the pupils and the Peabody Picture Vocabulary Test (PPVT) was administered as a measure of intelligence. Other information regarding the participants in the study was 34

PAGE 45

35 obtained from school records and conferences with teachers, i.e., the age of the child, the language spoken in the home, the ethnicity of the child, the number of siblings in the child's family, kindergarten attendance, and the sex of the child. These served as the independent variables of the study. The list of variables and their abbreviations as used in referring to them are found in Table 4. The Kuder-Richardson Formula 21 (Thorndike and Hagen, 1955) was used to estimate the reliability of the BSRT. The formula was run for each part of the BSRT, the pronunciation (phonology) section and the structure (syntax) section. The coefficients were .71 and .49 respectively, based upon the means for three raters. Since what was finally used was the combined judgments of three raters, this had the effect of tripling the length of a test., using the Spearman-Brown Formula (Thorndike and Hagen, 1955). Table 5 lists the inter-rater means and standard deviations which were used to derive the inter-rater reliability (see Table 6) of the scorers of the BSRT total scores for pronunciation and structure. The Pearson correlation nn. Inter-rater Reliability

PAGE 46

36 o •H +J (0 •H > 0) )^ X> < a H C 0) (0 tn Xi X C < •H 0) < CO W z < CO Ui CO (0 •H u > H •H 0) 0) x: 01 > o (0 Id O -p c: Cn CJ1 (0 •H Id fl +J > (0 •H CO d H iH +J 0) CO ^ CO 0) c Ui W >i -P a) 0) 0 0 Cu "4-1 rH u X! tJ^ E 0 -H (« p •H x: x: cn >H g cn iH CO CO ^4 (0 *w rH -H •H •H < 0 0) M-l 0) 0 (U c iH 1 H P (U d Cr> d X a, c 0 c H < CO w 2 < 2 CO XI e z CN XJ n in

PAGE 47

37 Table 5 INTER-RATER MEANS AND STANDARD DEVIATION Scorers Pronunciation Mean S. D. Structure Mean S . D . 1. 2 3 16.449 3.049 16,603 2.866 17.077 2.800 3.321 1.540 8.628 1.594 8.333 1.525 Table 6 INTER-RATER RELIABILITY OF THE SCORERS Pronunciation Structure Scorers 12 3 12 3 1 .937 .895 .863 .757 2 .937 .890 .863 .800 3 .895 .890 ; ' .757 .800

PAGE 48

38 coefficients ranged from .937 to .890 for the pronunciation scores and from .863 to .757 for the structures scores. These would fall within the 95 percent confidence intervals for the estimated reliabilities reported above. With a total subject population of 78, these coefficients are judged adequate for this study. This means that the standard errors of measurement are 0.8 and 0.6 respectively. The 95 percent confidence interval for an individual score would be the score plus or minus 1.6 for pronunciation and plus or minus 1.2 for structure. Relationship of Scores on the BSRT and Selected Variables In order to determine the relationship of the scores on the BSRT and the selected variables — intelligence, age in months, language spoken in the home, ethnicity of the child, number of siblings in child's family, kindergarten attendance, and sex of the child — an average of the scores of the three resource persons who graded the language samples was computed. Then, a correlation of the average score of each child was made for each section of the BSRT, pronunciation (phonology) and structure (syntax) , with each variable that was identified as having some relation to oral English. Table 7 indicates the results of these correlations With 78 subjects a correlation above .23 or below -.23 would be significant although correlations below -.30 would tend to be trivial. The correlation coefficients ranged from a

PAGE 49

39 Table 7 CORRELATION OF VARIABLES TO BSRT SCORE xMEANS Pronunciation Structure 1 IQ 0.675** 0.473** 2 Age -0.278* -0.130 3a Span -0.549** -0.342** 3b Eng 0.549** 0.342** 4a NA -0.354** -0.180 4b A 0.354** 0.180 5 Sib -0.184 0.093 6 K 0.308** 0.468** 7 Sex -0.143 -0.243* * Significant at .05 level with 75 degrees of freedom ** Significant at .01 level with 75 degrees of freedom positive 0.675 showing a fairly positive relation between the IQ variable and the pronunciation score on the BSRT to a negative correlation of -0.549 showing that when Spanish is spoken in the home there is an inverse relation to the score on English pronunciation on the BSRT. The highest correlation of the structure score correlates 0.473.

PAGE 50

40 Table 8 shows the interrelations of the predictor variables. Data and Hypotheses Relationships The following relates the data interpretation in terms of the stated hypotheses. Hypothesis 1: There is no relationship between the scores on the BSRT and the IQ score on the PPVT. The correlation analysis shows that there is a positive and significant relationship between the scores on the BSRT and the PPVT. This relationship holds true for oral English structure and English pronunciation as measured by the BSRT. Therefore, Hypothesis 1 is rejected. Hypothesis 2: There is no relationship between the scores on the BSRT and the age of the child. The variable of age correlates inversely with pronunciation scores on the BSRT (-0.278) at the .05 level. The correlation for structure (-.13) is not significant. Hypothesis 2 is rejected for pronunciation only. Hypothesis 3: There is no relationship between the scores on the BSRT and the language (a. Spanish, or b. English) spoken in the home. Language spoken in the home is related to oral English as measured by the BSRT pronunciation score. As would be expected, variable 3b Eng showed direct relationship (.549) to BSRT scores while variable 3a Span showed an inverse relationship (-0.549).

PAGE 51

4 in (t3 Xi n vo o rH o o 1 m iH in as rH o • • • 1 o CD ro o rH in CTi o rH rH o • • • • rH 1 O o 1 o 1 * * * * * * rrin o r~ vo cn o ri" o m o o 1 o o 1 o o 1 « * * « o in o o vo m o o o m o o o o o o (N rH rH as rH rH O O in in 0) rH > (U

PAGE 52

42 The BSRT score for structure showed a lower but still significant correlation with the language spoken in the home. Variable 3b Eng showed direct relationship (.342) and 3a Span was inverse (-0.342). Hypothesis 3 is rejected for both BSRT scores. Hypothesis 4: There is no relationship between the scores on the BSRT and the ethnicity (a. Non-Anglo, or b. Anglo) of a child. Ethnicity did show a relationship to the scores on the BSRT. Anglo representation showed a positive correlation (.354) and non-Anglo representation showed an j.nverse correlation (-0.354) for BSRT score on pronunciation. Neither of the observed relationships 4a (-0.18) or 4b (.18) were significant at the .05 level for structure. Hypothesis 4 is rejected for pronunciation but not for structure. Hypothesis 5: There is no relationship between the scores on the BSRT and the number of siblings in the family. The correlation analysis of the BSRT score for pronunciation and the variable Sib, showed that there was no significant relationship (-0.184) between the two. Likewise, the comparison with the BSRT score for structure showed a nonsignificant relationship (.093). In this case. Hypothesis 5, as stated, was not rejected. Hypothesis 6: There is no relationship between the scores on the BSRT and kindergarten attendance. The variable of kindergarten attendance correlated with pronunciation (.308) and with structure (.468). Both are.

PAGE 53

43 significant at the .01 level. However, since only three children had not gone to kindergarten, no conclusion will be drawn from these results. Hypothesis 7: There is no relationship between the scores on the B3RT and the sex of the child . The pronunciation score on the BSRT showed no significant relationsnip (-0.143) with whether the child was a boy or girl, but the structure score did (-0.243). Hypothesis 7 was not rejected for pronunciation, but it was rejected for structure. Regression Analysis In addition to the simple correlation analysis, multiple linear regression (Kerlinger and Pedhazur, 1973) was used to determine the unique contribution of different sets of the variables to the prediction of English language pronunciation and oral English language structure. The contribution of a set of variables to prediction may be measured by the square of the multiple correlation coefficients (RSQ) , one obtained for a regression model in which all predictors are used, called the full model (FM) , and the other obtained for a regression or restricted model (RI4) . The multiple correlation coefficients for the RI^ can never be larger than the multiple correlation coefficients for the FM. The difference between the two coefficients can be tested for statistical signficance with the variance ratio test.

PAGE 54

44 The knowledge of which variables make a contribution to the prediction of oral English pronunciation and structure would indicate which variables furnish information in determining the performance of similar first grade children in Colorado BBE programs. Because the variables used in this study are significantly correlated with each other, it is possible that conclusions based upon simple Pearson product moment correlations may be suspect. A more rigorous test of the relationship of these variables to the pronunciation and structure scores is through the use of multiple linear regression. Table 9 lists the means of the predictor variables as well as gives the standard deviations of each of these variables . This information shows that the mean IQ of the test population was 92.7 with a standard deviation of 15.1 which is considered a normal deviation for the instrument. The average for the variable of Age was 86.6 months which means that the average subject was about seven years and two months old at the time of testing. Each subject had an average of 2.87 siblings. The other variables were binary coded. If two highly correlated predictors occur in a regression system and one is deleted in order to determine its unique contribution, there is little or no drop in the square of the multiple correlation coefficient from the full model 1

PAGE 55

45 Table 9 MEANS AND STANDARD DEVIATIONS (N=78) OF PREDICTOR VARIABLES Variable Mean S. D. 1 IQ 92.7 15.1 2 Age 86.6 5.35 3a Span 0.269 0.444 3b Eng 0.731 0.444 4a NA 0.628 0.483 4b A 0.372 0.483 5 Sib 2.87 2.39 6 K 0.962 0. 192 7 Sex 0.423 0.494 Mean of pronunciation scores 16.61 2.78 Mean of structure scores 8.45 1.47

PAGE 56

to the restricted model. This may be tested by this formula for significance of a difference between multiple Rs (Guilford, 1965). (I -r:)(-,--,) R = multiple R v/ith larger number of F independent variables R = multiple R with one or more ^ variables omitted m^ = larger number of independent ^ variables m^ = smaller number of independent variables N = number of subjects dfj^ = m2^-m2 df2 = (N-m^-1) Closely related variables were grouped. These are shown in Table 10. Where use of the above formula gives an F with a probability above .05, unnecessary information need not be collected. As a result Table 11 and Table 12 were made as a sequential guide for retaining or dropping variables. The top blocks of Tables 11 and 12 indicate that all variables (1-7) were used in the full model (Schmid and Reed, 1966) . The next blocks indicate which predictor variables represent the sets to which multiple regression techniques were applied. The RSQ for each of these blocks could be compared with the RSQ of the FM. A large difference at or below the .05 level would indicate that the set is making

PAGE 57

47 cn H < <: z o H W w cr; o w « o w H Pi o U H Q W o o H 04 D O (1) -H It! > '3' ID >1 H to < w c x> CP H c c H AT 5 O ar (1) : cn g e •H 0) M 0 0 -p > CO Q) ffi K C •H m 0) +> Cn 0) 0) M nJ 5 C 3 •H CP tJi Q) +J -H rH H (C (0 U C rH •H 3 Ou 0) CO X! Xi C CT> 0) en xi •H U (U •H C C « (U cn o (T3 od c H C H 1 0) Q) ITS Cn C tn 0) X -P a, C 0 C CP 13 0) c X W <: <: CO H w tn o H H H H

PAGE 58

48 an unique contribution to the prediction of the phenomena. A small or zero drop indicates that any relationship found is not significant. Interpretation of the Regression Analysis The regression of the criterion, oral English pronunciation and structure, on all of the variables had RSQ's of .5692 (Table 11) and .4303 (Table 12) respectively. The square root of these values is .7545 and .6560 respectively. This is higher than the best individual variable for structure, which is IQ. Each of these variables correlated at 0.675 and 0.473 respectively. The restricted model, FM 3a, 3b, 4a, and 4b for pronunciation, (Table 11) had an RSQ of .4988. The observed drop is . 5692. 4988=. 0704 and is statistically significant; therefore, this difference indicates that this set of variables (Span, Eng, NA, A) makes an unique contribution to the prediction of oral English pronunciation. From this set, the restricted models FM 3a, 3b and FM 4a, 4b were noted which indicate that language (3) and ethnicity (4) as a group of variables contributed to the prediction but that one cannot look at the individual language or ethnicity as a predictor. The second group which formed the restricted model FM 2, 5, and 7 for pronunciation, had an RSQ of .5412. No significant drop is noted between this model containing the variables Age (2), Sib (5), and Sex (7), and the FM

PAGE 59

49 rr\n O o» • o n o f-i * 1 1 1 1 O »H lO tn IN 1 OS X U4 (» m i-i O o • o • vn o » (N • 1 >: 1 ii V a cu V. J3 V CD X ^ -Tl CD • m .-N O 1 * t ^ 1 1 1 O oc 1 T. o CO vO in 1 O or. 1 O in 1 1 :^ a 0) O CO S O u -o 0) 4J (C VO < iH IN <^ k 1 o r«i in t 1 >:

PAGE 60

50 00 x£> ^ c C* • o «H •*t o 1 1 1 o b. a. K U o u i3 CS o u i4 ^ OJ > o o 3 in O »n n O n U 1 R 1 O (u K •d' c nt o CO g o u M-l (U 4J (U

PAGE 61

51 containing all of the predictor variables. Group III, FM 1 and 6 for pronunciation, had an RSQ of .3977. The observed drop is . 5692. 3977= . 1715 and this is significant at the .05 level. This indicates that Group III contributes to the prediction of oral English pronunciation. Further analysis of the model shows that IQ (1) has an RSQ of .4068 and K (6) has an RSQ of .5676. The drop observed in variable 1 is significant (. 5692. 4068= .1624), while no significant drop is noted in variable 6 alone. IQ as measured by the PPVT is a contributing variable in the prediction of oral English pronunciation. Applying the same technique to structure (Table 12), Group I FM 3a, 3b, 4a, and 4b has an RSQ of .4282 and no significant drop is noted between this RSQ and the RSQ (.4303) of the FM. Therefore, the variables Span (3a), Eng (3b) , NA (4a) , and A (4b) as a group are not contributing factors to the prediction of oral English structure. Group II, FM 2, 5, and 7 for structure, had an RSQ of .3542. The observed drop, . 4303. 3542= . 0761 , is statistically significant and further analysis was done. The restricted models, FM 2, FM 5, and FM 7, showed RSQs of .4272, .3836, and .4125 respectively, none of which show statistically significant drops. As a group, these variables Age (2) , Sib (5) , and Sex (7) are of use in the prediction of oral English structure. Variables Age (2) and Sex (7) do not make significant contributions, but the variable

PAGE 62

52 Sib (5) is significant when the RSQ of .3836 is rounded to .38 (.4303-.38=.05) . The restricted model, FM 1,6 for structure, had an RSQ of .1968 which gave a drop of . 4303. 1968= . 2335 . This drop is significant and warrants further analysis. The resulting models, FM 1 and FM 6 had RSQs of .3145 and .3440. The models showed drops of .1158 and .0863, each of which is statistically significant. The conclusion is that IQ (1) as measured by the PPVT, Sib (5), and K (6) are contributing variables and are useful in the prediction of oral English structure. It may seem that the seven predictor variables can be used to predict the English language of children in Colorado BBE programs quite well, the square of the RSQ being .5692 (pronunciation) and .4303 (structure). Furthermore some models having fewer variables than seven predictors also had a high prediction efficiency. Examples of high prediction efficiency are FM Age (2) , Sib (5) , Sex (7) for pronunciation and FM Span (3a) , Eng (3b) , NA (4a) , and A (4b) for structure. Considering these findings, a prediction equation could be constructed for any group of predictor variables using the data of Tables 7 and 8. Summary of Regression Analysis As a result of the regression analysis the following conclusions were made in reference to the stated hypotheses:

PAGE 63

53 Hypothesis 1: There is no relationship between the scores on the BSRT and intelligence as measured by the PPVT. According to the findings of the regression analysis, IQ as measured by the PPVT, has a significant relationship to the scores on the BSRT, thereby, giving some predictability of oral English language pronunciation and structure. Therefore, Hypothesis 1 was rejected at the .05 level of significance . Hypothesis 2: There is no relationship between the scores on the BSRT and the age of the child. There was an age range from 79 to 111 months and a standard deviation of 5.35 months among the subjects. With this somewhat restricted range, regression analysis showed that no significant relationship exists between the scores on the BSRT and the age of the child. Hypothesis 2, therefore, was not rejected. Hypothesis 3: There is no relationship between the scores on the BSRT and the language (a. Spanish, or b. English) of the home . Language of the home as a predictor did not have any statistically significant relationship to the scores on the BSRT. However, when language of the home and ethnicity of the child were grouped as predictors using regression analysis, significant results were noted. Hypothesis 3 as stated is not rejected.

PAGE 64

54 Hypothesis 4: There is no relationship between the scores on the BSRT and the ethnicity (a. non-Anglo, or b. Anglo) of the child. Ethnicity as a separate variable had no relationship to the scores on the BSRT. Hypothesis 4 was not rejected. Hypothesis 5: There is no relationship between the scores on the BSRT and the number of siblings in the child's family. No significant relationship was observed between the scores on the BSRT and the number of siblings in the child's family. This variable Sib showed some predictability v;hen grouped with the child's age and sex using regression analysis (and is a contributing factor when the RSQ value is rounded). Nevertheless, Hypothesis 5 was retained even though it was shown as indicating a relationship that was significant. Hypothesis 6: There is no relationship between the scores on the BSRT and kindergarten attendance. The regression analysis showed that no relationship existed between the score for pronunciation on the BSRT and the variable of kindergarten attendance. But a significant relationship was noted between the BSRT score for structure and kindergarten attendance. In this case. Hypothesis 6 was rejected for structure and not rejected for the BSRT pronunciation score. However, as previously stated, since only three children did not attend kindergarten, no generalizations can be made as a result of this finding.

PAGE 65

55 Hypothesis 7: There is no relationship between the scores on the BSRT and the sex of the child. The regression analysis showed no significant relationship between the sex of the child and the scores on the BSRT. Therefore, Hypothesis 7 was not rejected. The ove all results of the correlation analysis and the regression analysis showed a relationship between the scores on the BSRT and most of the identified oral English predictors. The variables of intelligence and language of the home had the greatest direct relationship with the pronunciation and structure scores on the BSRT according to correlation analysis. The variable of ethnicity also had a .01 level of significance in relation to the pronunciation score on the BSRT. Multiple regression analysis identified intelligence as a significant single variable in predicting the BSRT pronunciation score while the number of siblings in the family was singly significant in its relationship to the BSRT structure score along with intelligence. Chapter V will summarize this study and make recommendations for further research.

PAGE 66

CHAPTER V SUMMARY, RECOMMENDATIONS, AND DISCUSSION Summary of the Study The purpose of this study was to develop a reliable instrument that could be used by teachers and staff in Colorado to assist in the assessment of oral English language, particularly the structure (syntax) and the pronunciation (phonology) of first grade children in bilingual bicultural education (BBE) , and to relate the scores of the instrument to seven variables: a) intelligence, b) the child's age, c) the language spoken in the home (Spanish or English) , d) ethnicity (non-Anglo or Anglo) , e) the number of siblings in the child's family, f) kindergarten attendance, and g) the sex of the child. Seventy-eight subjects were identified for the study. These subjects were first graders who were participants in BBE programs in the state of Colorado during the school year 1974-75. The children were tested using the researcherdeveloped instrum.ent, the Bass Sentence Repetition Task (BSRT), to determine knowledge of oral English pronunciation and structure and the Peabody Picture Vocabulary Test (PPVT) for testing ingelligence. 56

PAGE 67

57 The Kuder-Richardson Formula 21 was applied to estimate the reliability of each section of the BSRT. The coefficient was .71 for pronunciation and .49 for structure. The combined judgments of three raters had the effect of tripling the length of the test. Applying the Spearman-Brown Formula to estimate a new reliability, a .88 coefficient for pronunciation and a .76 for structure were obtained. Correlation analysis as well as multiple regression analysis were used to determine the relationships between the BSRT scores and the predictor variables. The correlation analysis showed that • the variables intelligence, language spoken in the home, and ethnicity were related to the BSRT pronunciation score at the .01 level of significance. The variables intelligence and language spoken in the home were significantly related to the BSRT structure score at the .01 level, while the sex variable was significantly related at the .05 level. No conclusion can be drawn about the relationship of the BSRT scores with kindergarten attendance because only three pupils in the sample did not attend kindergarten. The method of multiple regression was used as a source of a more rigorous method of data analysis which can serve as a means of identifying the variables that significantly contribute to the prediction of certain phenomena, in this case, that of oral English pronunciation and structure .

PAGE 68

58 A significant relationship was noted between the pronunciation score on the BSRT and the variable of intelligence as measured by the PPVT. Also the structure score on the BSRT was significantly related to the intelligence variable and to the number of siblings in the family (based on the rounded multiple correlation coefficient for the variable) . Even though a significant relationship of the scores on the BSRT and the variable of kindergarten attendance was observed, no conclusions can be drawn because only three of the subjects did not attend kindergarten. Multiple regression analysis also showed a statistically significant relationship between the scores on the pronunciation portion of the BSRT and the grouping of ethnicity and language variables; however, further analysis did not determine a relationship between the BSRT scores and individual languages or ethnic representations. As a result of the regression analysis, the intelligence variable as measured by the PPVT is considered to be a contributing factor to the measure of oral English pronunciation and structure as measured by the BSRT. Furthermore, the correlation analysis showed that this variable is positively related. The number of siblings in the child's family (when the RSQ is rounded) is considered a contributing factor to the prediction of oral English structure as measured by the BSRT. The established reliabilities for the BSRT are within acceptable limits for an instrument of this type. The BSRT,

PAGE 69

59 therefore, is a reliable tool for assessing the oral English pronunciation and structure of first grade children in Colorado BBE programs. The statistically significant variables of intelligence as measured by the PPVT and the number of siblings in the child's family are contributing factors to predicting the oral English pronunciation and structure as measured by the BSRT. The researcher does caution that correlation analyses do not indicate cause and effect relationships. The analyses used in this study only indicate the degree of relationship between the BSRT and the variables studied. For this reason, no conclusions about casuality were presented. Recommendations In view of the findings and the limitations of this study, the following recommendations are made: 1. The researcher recommends administration of the BSRT to first grade children in Colorado BBE programs as a quick assessment of oral English pronunciation (phonology) and structure (syntax) . Furthermore, it could be tested for use in Spanish-English programs in other parts of the country. 2. The researcher recommends the development of BBE instructional materials which give special attention to the items in the BSRT. 3. Intelligence as measured by the PPVT when used selectively can assist in prediction of oral English pronunciation and structure as measured by the BSRT.

PAGE 70

60 4. The information gathered in this study can be used as an example of how to devise a sentence repetition task for use with other languagedifferent populations. 5. The researcher recommends the development of a reliable and valid tool for the measurement of the mental ability of youngsters in B3E which minimizes the factors of cultural and linguistic biases . 6. The researcher recommends the development of a sentence repetition task that takes into consideration the dialect differences of Spanish spoken in a given area for use in that area. 7. Data from this study can be used in pre-service and in-service education to prepare teachers for working in BBE programs. Also, the BSRT and instructions for administering and scoring it could be used in such educational programs. BBE programs are increasing in number across the United States. With the growing concern for assuring that these programs are indeed meeting the needs of the populations they are designed to serve, research and development of materials are of utmost importance. The development of the BSRT is to be of specific service to the BBE programs of Colorado as well as BBE programs across the nation.

PAGE 71

BIBLIOGRAPHY Abraham, William. "Investigation of Mental Retardation and Pseudo Mental Retardation in Relation to Bilingual and Sub-Cultural Factors." Paper prepared for the College of Education, Arizona State University, 1960. Allred, Forest R. "Errors in Oral English Language Usage of Mexican American Pupils with a Spanish Langauge Background in Grade III in the State of Colorado." Doctoral dissertation. University of Northern Colorado, 1970. Asher, Steven R. "Influence of Sampling and Comparison Processes on the Development of Communication Effectiveness." Paper presented at the Eightieth Annual Convention of the American Psychological Association, Hawaii, September, 1972. (ERIC No. ED 071 732). Bass, Bernice. "Language Development Should Begin at Home." Education Colorado , Colorado State Department of Education, January 15, 1974. Benitez, Mario. "Bilingual Education: The What, the How and the How Far." Hispania , September, 1971, 5£, pp. 499-503. Bereiter, Carl and Englemann, Siefried. Teaching Disadvantaged Children in the Preschool . New Jersey: Prentice Hall, Inc. , 1966. Bernstein, Basil. "A Sociolinguistic Approach to Socialization: VJith Some Reference to Educability . " Language and Poverty , Frederick Williams (Ed.), Illinois: Markham Publishing Company, 1970. Board of Education of the City of New York. Teaching ^ English as a New Language to Adults . Curriculum Bulletin Niimber 5, 1963-1964 Series. Brengelman, Frederick H. and Manning, John C. "Linguistic Capacity Index." Measures for Research and Evaluation in the English Language Arts , William T. Fagan, Charles R. Cooper, and Julie M. Jenson (Eds.), Illinois: ERIC Clearinghouse on Reading and Communication Skills and the National Council of Teachers of English, 1975. 61

PAGE 72

62 Burt, Marina K. ; Dulay, Heidi C; and Chavez, Eduardo Hernandez. Bilingual Syntax Measure . New York: Harcourt Brace Jovanovich, Inc., 1975. Cartel; Annotated Bibliography of Bilingual Bicultural Material, Cumulative Issue 1974 . Texas: Dissemination Center for BBE, 1974. Chomsky, Carol. The Acquisition of Syntax in Children from Five to Ten . Massachusetts: The M.I.T. Press, 1969. Chomsky, Noam. Aspects of the Theory of Syntax . Massachusetts: The M.I.T. Press, 1965. Cohen, Bertram D. and Klein, Judith F. "Referent Communication in School Age Children." Child Development , 1968, 39_, pp. 597-609. Colorado Department of Education. Colorado School Laws Supplement, 1974 . October, 19':'4. Colorado Department of Education. Ethnic Group Distribution in Colorado Public Schools . 1971-1972. March, 1973. Colorado Department of Education. Report on Senate Joint Resolution 20 Study: Part II Bilingual Multi-Cultural Education Programs . January, 1975. Cook, Vivian J. "The Analogy Between Firstand-Second Language Learning." Toward a Cognitive Approach to Second Language Acquisition , Robert C. Lugton and Charles H. Heinle (Eds.), Pennsylvania: The Center for Curriculum Development, Inc., 1971. Dailey, John T. and Neyman, Clinton A. "An Evaluation of the Language Arts Program of the District of Columbia, Final Report." Washington, D. C, : George Washington University, November, 1965. (ERIC No. ED 024 449). Darcy, Natalie. "Bilingualism and the Measurement of Intelligence: Review of a Decade of Research." Journal of Genetic Psychology , 1963, lO^, pp. 259-82. Davis, Edith A. Th e Development of Linguistic Skills in Twins, Singletons with Siblings, and Only Children from Age Five to Ten Years . Minnesota: University of Minnesota Press, 1937. Dunn, Lloyd M. Expa nded Manual for the Peabody Picture Vocabulary Test . Minnesota: American Guidance Service, Inc., 1965.

PAGE 73

63 Earle, Jane. "Bilingual Education Bill Is Unique in U. S." Denver Post , June 22, 1975, 29C. Ebel, Robert (Ed.). "Language Development." The Encyclopedia of Educational Research . Ontario: CollierMacMillan Canada Ltd., 1969, pp. 686-98. Ervin-Tripp, Susan. "Becoming a Bilingual." Working paper 9, Language Behavior Research Laboratory, California University, Berkeley, March, 1968 (ERIC No. ED 018 786). Ervin-Tripp, Susan and Miller, Wick R. "The Development of Grammar in Child Language." Monographs of the Society for Research in Child Development , 1964, 19_, pp. 9-34. Fagan, William T.; Cooper, Charles R. ; and Jensen, Julies M. (Eds.). Measures for Research and Evaluation in the English Language Arts . Illinois: Eric Clearinghouse on Reading and Communication Skills and National Council of Teachers of English, 1975. Finocchiaro, Mary. Teaching Children Foreign Languages . New York: McGraw-Hill Book Company, 1964. Fox, David. The Research Process in Education . New York: Holt, Rinehart, and Winston, Inc., 1969. Fraser, Colin; Bellugi, Ursula; and Brown, Roger. "Control of Grammar in Imitation, Comprehension and Production." Journal of Verbal Learning and Verbal Behavior , 1963, 2, pp. 121-35. Gesell, Arnold; Ilg, Frances; Ames, Louise B.; and Bullis, Glenna E. The Child From Five to Ten . New York: Harper and Brothers, 194 6. Gesell, Arnold; Ilg, Frances; Ames, Louise B.; and Bullis, Glenna E. The First Five Years of Life . New York: Harper and Brothers, 1940. The Gloria and David Beginning Language Tests . Texas : Language Arts, Inc., 1958. Guilford, Joy Paul. Fundamental Statistics in Psychology and Education . New York: McGraw-Hill Book Company, 1965, p. 403. Haugen, Einar. "Problems of Bilingualism." Lingua , August, 1950, 2, pp. 271-90. Mickey, Tom. "Bilingualism and Measurement of Intelligence and Verbal Learning Abilities." Exceptional Ch ildren, 1972, 39, pp. 24-28.

PAGE 74

64 Jakobovits, Leon A. "Implications of Recent Psycholinguistic Developments for the Teaching of a Second Language." Toward a Cognitive Approach to Second Language Acquisition , Robert C. Lugton and Charles H. Heinle (Eds.), Pennsylvania: The Center for Curriculum Development, Ind., 1971. Jakobovits, Leon A. Foreign Language Learning, A Psycholinguistic Analysis of the Issues . Massachusetts : Newberry House Publishers, 197 0. John, Vera and Horner, Vivian. Early Childhood Bilingual Education . New York: Modern Language Association of America, 1971. Kagan, Jerome. "The Child's Sex Role Classification of School Objects." Child Development , 1964, 3_5, pp. 1951-56. Karnes, Merle B.; Hodges, Audery; and Teska, James A. "Evaluation of Two Preschool Programs for Disadvantaged Children: A Traditional and a Highly Structured Experimental Preschool." Exceptional Children , May, 1968. Katz, Jerald J. Philosophy of Language . New York: Harper and Row, Publishers, 1966. Kerlinger, Fred and Pedhazur, Elazar. Multiple Regression in Behavioral Research . New York: Holt, Rinehart, and Winston, Inc., 1973. Laney, Billie J. "A Comparative Study of Expressive and Comprehensive Vocabulary Development in Male and Female Kindergarten Children." Doctoral dissertation. East Texas State University, 1970. Lenneberg, Eric H. Biological Foundations of Language . New York: John Wiley and Sons, 1967. Loban, Walter. Problems in Oral English: NCTE Research Report No . 5 . Illinois: National Council of Teachers of English, 1966. Lynn, Klonda. "Bilingualism in the Southwest." Journal of Speech , April, 1945, 31, pp. 175-80. Macnamara, John. "The Bilingual 's Linguistic Performance — A Psychological Overview. " Journal of Social Issues , April, 1967, 23, pp. 58-77. McNeill, David. "The Creation of Language by Children." Child Language , A Book of Readings, A. Bar-Adon and W. Leopold, (Eds.), New Jersey: Prentice-Hall, Inc., 1971, pp. 349-59.

PAGE 75

65 McNeill, David. "Developmental Psycholinguistics . " The Genesis of Language; A Pscyholinguistic Approach , Frank Smith and G. A. Miller (Eds.), Massachusetts: The M.I.T. Press, 1966. Menyuk, Paula. "Children's Learning and Reproduction of Grammatical and Nongrammatical Phonological Sequences." Child Development , 1968, 3^, 849-59. Menyuk, Paula. Sentences Children Use . Massachusetts: The M.I.T. Press, 1969. Natalicio, Diana S. and Williams, Frederick. "Oral Language Assessment." Paper presented at the Annual Meeting of the American Educational Research Association, Illinois, April 6, 1972 (ERIC No. ED 063 845). Olguin, Leonard. "'Shuck Loves Chirley': A Non-Tecnical Teaching Aid for Teachers of Bilingual Children with a Diagnostic Test." CARTEL; A nnotated Bi b liography of Bilingual Bicultural Material, Cumulative Issue -1974 , Texas: Dissemination Center for BBE, 1974. Pendergast, Kathleen; Barker, Janet; Dickey, Stanley; Soder, Anton; Gow, Jane; and John, Selmar. "An Articulation Study of 15,255 Seattle First Grade Children with and without Kindergarten." Exceptional Children , April, 1966, 32^, 541-47. Penfield, Wilder. "Conditioning the Uncommitted Cortex for Language Learning." Brain , 1965, £8, 787-98. Penfield, Wilder. "The Learning of Languages." Foreign ' Language Bulletin , Spring, 1960, 7, pp. 1-17. Piaget, Jean. The Language and Thought of the Child . Ohio: The World Publishing Company, 1955. Politzer, Robert and Staubach, Charles. Teaching Spanish, A Linguistic Orientation . Massachusetts: Ginn and Company , 1961. ^ Robinett, Robert. "Teaching of Bilingual Children." The Encyclopedia of Education . New York: MacMillan Company and Free Press, 1971, pp. 466-72. Ruddell, Robert. Reading-Language Instruction; Innovative Practices . New Jersey: Prentice Hall, Inc., 1974. Saville-Troike, Muriel R. "TESOL: Methods and Materials in Early Childhood Education." Essays on Teaching Englis h as a Second Dialect , R. P. Fox (Ed.), Illinois: National Council of Teachers of English, 1973a, pp. 102-06.

PAGE 76

66 Saville-Troike, Muriel R. Bilingual Children: A Resource Document . Virginia: Center for Applied Linguistics, 1973b. Slobin, Dan I. and Ferguson, C. A. (Eds.). Studies of Child Language Development . New York: Holt, Rinehart, and Winston, Inc., 1973. Slobin, Dan I. "Children and Language: They Learn the Same Way All Around the World." Psychology Today , July, 1972, 6, pp. 74-77, 82. Smith, Mary E. "Measurement of Vocabularies of Young Bilingual Children in Both of the Languages Used." Journal of Genetic Psychology , 1949, 7_4, 305-10. Smith, Philip D. Toward a Practical Theory of Second Language Instruction . Pennsylvania: The Center for Curriculum Development, Inc., 1971. Stanchfield, Jo M. Research: New Hope for Kindergarten Children. Paper Presented at the meeting of the California Reading Association, California, November, 1971 (ERIC No. ED 058 Oil). Thorndike, Robert L. and Hagen, Elizabeth. Measurement and Evaluation in Psychology and Education . New York: John Wiley and Sons, 1955, pp. 129, 131. Williams, Frederick (Ed.) Language and Poverty, Perspectives on a Theme . Illinois: Markham Publishing Company, 1970. Zintz, Miles. Education Across Cultures . Iowa. Kendall/ Hunt Publishing Company, 1969.

PAGE 77

APPENDIX A LETTER FORMS FOR CONTACTING SCHOOLS

PAGE 78

68 THE UNIVERSITY OF NORTHERN COLORADO GREELEY, COLORADO 80639 February 28, 1975 Mr. Keller, Principal Letford Elementary School Johnstown, Colorado Dear Mr. Keller: This letter is to introduce Bernice Bass who is a doctoral student at the University of Florida working under my supervision here at the University of Northern Colorado to complete work for a degree with emphasis in bilingual education. Ms. Bass is in the process of developing an instrument which could be of assistance in the evaluation of oral English language proficiency of young children of Mexican-American descent. In the pilot testing of this instrument, Ms. Bass has selected to work with youngsters of first grade level. Your assistance in permitting Ms. Bass to work with youngsters in your school is greatly appreciated. Sincerely, Jos^ Cordova Assistant Professor Elementary Education

PAGE 79

69 UNIVERSITY OF FLORIDA GAINESVILLE, FLORIDA April 6, 1975 ADDRESS Dear This letter is to introduce Bernice Bass who is a doctoral student at the University of Florida working under my direction to complete a degree with emphasis m bilingual bicultural education. Ms. Bass has developed an instrument which could be of assistance in the assessment of oral English language of young children, particularly those with language differences. She has selected to work with first grade youngsters in her research study to establish the reliability for the instrument she has developed. This will require that she work with individual children for approximately 15 to 20 minutes each. Your assistance in your program would be greatly appreciated. Ms. Bass will be contacting you within the next week. Sincerely, Ruthellen Crews Professor of Education

PAGE 80

APPENDIX B BASS SENTENCE REPETITION TASK

PAGE 81

BSRT Entering Interview: 1. Who are you? (What is your name?) 2. How many brothers and sisters do you have? 3. How old are you? Repetitions : 1. The girl sits on a chair. 2. The girl asks the teacher a question, (You are a good listener — for reinforcement) 3. The dog barks. 4. The berries are ripe. 5. The fruit is fresh. 6. I think the apples are ready. 7. Which apple would you choose? 8. The children choose big apples. 9. Shine my shoes. 10. The sheep were sheered yesterday. 11. The scissors are for cutting. 12. Please make the decision. 13. Choose the round puzzle. 14. I go to sleep at night. 15. I am thirsty, I'd rather have some water. 16. The teacher's window faces my house. 17. I don't want that. Exit Interview: Thank you, I enjoyed playing this game with you.

PAGE 82

72 Directions for Administering the BSRT: 1. Use two tape players — one for the recorded tape and one that will record the session. 2. Begin with the entering interviev; to gain the interest of the child. 3. Record the entering interview and then play it for the child. 4. Give the following instructions: a. We're going to play a little repeating game. b. I'll read something and I want you to repeat exactly what you hear, as you hear it. Remember, repeat what you hear exactly as you hear it. 5. Start the tape players. 6. Turn off the players after the session and excuse the child. 7. If the child does not seem to understand the directions, start the directions again from the beginning. 8. If necessary, give the directions to the child in Spanish, but the repetitions must be made in English.

PAGE 83

Scoring Form for BSRT-1975 Grader NA.HE OF CHILD NUMBER OF SIBLINGS AGE DATE OF GRADING PRONUNCIATION STRUCTURE 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 1, 2, 3, 4, 5, 6, 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. TOTAL TOTAL

PAGE 84

74 Corrections for the BSRT— 1975 Instructions : 1. Grade each part separately. a. pronunciation b. structure 2. Record the name and other information from the rec. rding on the grading form. a. Play first part of tape, record the answers . b. Make certain that all is recorded before beginning the analysis for grading. c. If any part of the recording j.s blurred or unintelligible, mark the space UN. If there is no response, mark the space NR. 3. Pronunciation: listen to one sentence at a time noting the pronunciation of the words as compared to the prepared model. The word indicated on the pronunciation correction sheet must be pronounced according to the model's sounding. a. If correct, mark 1 for each correct response; if incorrect, mark with "0." b. All scores for the pronunciation section should appear under the section so marked on the correction form. c. There is a total of 20 possible points. 4. Structure: listen to one sentence at a time noting the structure indicated on the structure correction sheet. The response must be exactly as given on the correction sheet. a. If correct, mark "1" point for each grouping. If the structure is not exactly as indicated, make written notes of the variation. b. There is a total of 10 possible points.

PAGE 85

75 Corrections for the Pronunciation Section of the BSRT 1 chair 2 teacher 3. 4 berries 5 fresh 6 think 7. Which choose? 8 children choose 9. Shine 10 sheep sheered 11 scissors cutting . 12 decision 13. Choose puzzle. 14. 15 thirsty rather. 16 teacher 17.

PAGE 86

76 Corrections for the Structure Section of the BSRT 1. girl sits 2. gitl asks 3. dog barks 4. berries are ripe 5. 6. 7. 8. big apples 9. 10. 11. 12. 13. round puzzle 14. go to sleep 15. I am thirsty or I'm thirsty 16. window faces my house 17. don't want or do not want or don't like

PAGE 87

77 Subject Information Sheet — BSRT, 1975 NAME and NUMBER Birthdate Number of sibMngs Language of the home IQ Test date Scores SET Pronunciation Comments : Age in months Attended Kindergarten Structure

PAGE 88

78 Evaluation of Scores of BSRT 1975 Pronunciation Section: 20 — 16 indicates that child has gained control of the given sound patterns 15 — 11 suggests item analysis to determine the child's needs 10 — 0 child should begin initial auditory program which emphasizes the sounds which are distinctly English in nature . Structure Section: 10 — 7 indicates that child has gained control of the given language structures 7 — 5 suggests item analysis to determine the child's needs 4-0 child should begin with initial structure and pattern program.

PAGE 89

APPENDIX C RESOURCE PERSONS: QUALIFICATIONS Al^lD INFORMATION

PAGE 90

Qualifications for Resource Persons: 1. Has worked with Colorado BBE programs for at least three years. 2. Is bilingual in Spanish and English as demonstrated by continued employment in a Spanish/English BBE program as a teacher or staff person. 3. Participates in the training sessions conducted by the researcher. Training Sessions: 1. There will be two training sessions. a. The first will give background on the process of scoring SRT. b. The second session will allow practice in grading SRT using selected tapes from the pilot testing. 2. Upon completion of training, the resource persons will begin to score the tapes collected in the major study.

PAGE 91

81 RESUME FORM NAME Alice Carolyn Aguayo de Leal ADDRESS 1616 27th Avenue Court; Greeley, Colorado 80631 EXPERIENCES IN COLORADO BILINGUAL PROGRAI^S 5 years — Johnstown Bilingual Program, Kindergarten level bilingual teacher 1 year Greeley Bilingual Programs, curriculum specialist ADDITIONAL INFORMATION Consultant — Teaching in a Second Language Instructor — University level courses in bilingual education Curriculum development for oral language development Team member of Ad Hoc Team to recommend NEA position in regard to the Child Development Associate Consortium, Washington, D. C. SIGNATURE — indicates intentions to participate in training sessions

PAGE 92

82 RESUME FORM NAME Estella Anita Martinez de See ADDRESS Greeley, Colorado 80631 EXPERIENCES IN COLORADO BILINGUAL PROGRAMS 5 years — Johnstown Bilingual Program 1 year-English reading teacher, first grade bilingual classroom 2 years-bilingual classroom teacher, third grade 1 year-bilingual classroom teacher, fifth grade 1 year-bilingual classroom teacher, fifth, sixth, and seventh grades ADDITIONAL INFORMATION Workshop and/or Conference Consultant Topics — Teaching Math Bilingually Bilingual-Bicultural Education Oral Language Curriculum for BBE Special consultant for Senate Joint Resolution 20, Bill for BBE-Colorado State Department of Education Developer--K-4 curriculum for bilingual-bicultural project SIGNATURE — indicates intentions to participate in training sessions

PAGE 93

83 RESUME FORM NAME Sheila Anne Towle ADDRESS 212 4 6th Avenue, Greeley, Colorado EXPERIENCES IN COLORADO BILINGUAL PROGRAMS 4 years — Johnstown Bilingual Program 2 years-kindergarten bilingual teacher 2 years-second grade bilingual teacher ADDITIONAL INFORMATION Secretary — Colorado Association for Bilingual Bicultural Education Assistant in developing elementary bilingual curriculum Member of evaluation team for elementary bilingual programs SIGNATURE — indicates intentions to participate in training sessions

PAGE 94

APPENDIX D PERSONS KNOWLEDGEABLE OF LANGUAGE AND BBE 4

PAGE 95

85 THE UNIVERSITY OF NORTHERN COLORADO GREELEY, COLORADO 80639 March 14, 1975 Dr. Ruthelle.. Crews University of Florida Norman Hall 190 Gainesville, Florida 32611 Dear Dr. Crews: This is to affirm the fact that Bernice Bass has completed an independent study under my direction in which she developed an instrument that assesses the oral English syntax and phonology of first grade children in bilingual programs . I believe this instrument has great potential for our Colorado bilingual programs, particularly because it has been modeled to focus upon the unique linguistic differences of Colorado. Sincerely yours. Jose E. Cordova Assistant Professor of Elementary Education

PAGE 96

86 COLORADO DEPARTMENT OF EDUCATION DENVER, COLORADO March 26, 1975 Dr. Ruthellen Crews University of Florida Norman Hall 186A Gainesville, Florida 32611 Dear Dr. Crews: Ms. Bernice Bass has developed a measure that is appropriate to Bilingual Programs in Colorado. She has worked with various school personnel and school students and has been given the support for her measure in regards to relevance. The Sentence Repetition Task that Ms. Bass has developed has been reviewed by our office. We feel that it provides appropriate exercises in assessing students' linguistic differences in the areas of phonology and syntax. The analysis lends itself very much to giving teachers and administrators information for bringing about proper techniques in presenting concepts to children with linguistic differences. We are hoping that we can capitalize on the basic functions of the SRT and that we can continue to use the expertise that Ms. Bass possesses. Sincerely, Bernard Martinez Bilingual Bicultural Education Community Services Unit 892-2166 cc: Bernice Bass

PAGE 97

COLORADO DEPARTKiENT OF EDUCATION DENVER, COLORADO April 12, 1975 Ms. Bernice Bass 6905 Mariposa Street Denver, Colorado Dear Bernice: Thank you for the opportunity to review the BASS-SRT. This instrument, designed specifically to assess the English oral language skills of primary studenrs in Colorado Bilingual/Bicultural Programs, is extremely needed. We, in the Colorado Department of Education, look forward to the passage of legislation this year calling for the design and implementation of Bilingual/Bicultural Programs in school districts throughout our state. A necessary aspect of local program design might be addressed through the use of your instrument in the identification of where students are with respect to English oral language skills. Please keep me informed of any new design features of your instrument. Sincerely, Jennie S. Green Consultant Community Services Unit (303) 892-2166 JSG/lw

PAGE 98

BIOGRAPHICAL SKETCH Bernice Marie Bass was born an only child in Denver, Colorado, to Mr. and Mrs. Arthur Bass. Upon graduation from Manual High School in Denver, she went to the University of Northern Colorado in Greeley where a Bachelor of Arts degree was granted in 1970, with a Spanish major and a chemistry minor. During that time she worked as a laboratory assistant and electrocardiograph technician at Weld County General Hospital and as a Head Resident in the residence halls at UNC. A Master of Arts degree was awarded in 1972 in elementary education with special emphases in bilingual bicultural education and reading. The author began her experience in bilingual bicultural education working with kindergarten children in the Johnstown Bilingual Program (Johnstown, Colorado) , in 1971. Other Colorado teaching experiences also include: service in the Greeley Public Schools' Home Intervention Oral Language Development Program as a teacher of youngsters aged two to five years old; bilingual kindergarten teaching in the Ft. Lupton Bilingual Program; and teaching in the Brighton Public Schools. In addition, the author served as an Early Childhood Content Specialist in the Early Childhood 88

PAGE 99

89 Component of the Educational Technology Demonstration Project sponsored by the Education Commission of the States and affiliated with the Federation of Rocky Mountain States during 1973. In the Fall of 1973, Bernice Marie Bass went to Gainesville, Florida, where she taught Title I reading students at Stephen Foster Elementary School and trained paraprof essionals to assist children with reading problems while at the same time completing her year's residency work at the University of Florida. The following year, the author returned to Colorado to complete additional college courses in Mexican American Studies and bilingual bicultural education at the University of Northern Colorado, Greeley, where she also served as a teaching assistant in Center for H\aman Enrichment and graduate assistant to the Director of Bilingual Bicultural Education. Presently, the author serves as a bilingual educational instructor with that University's Outreach and Special Programs Educational Service. Ms. Bass' permanent residence is Denver, Colorado, where she is involved with schools and organizations interested in the preparation of bilingual bicultural teachers as well as the further development and advancement of bilingual bicultural education programs.

PAGE 100

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ihellen Crews, Chairperson Professor of Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and ,-'5iJaaity , as a dissertation for the degree of Doctor ofPhilosophy X Midhael ii.' Hanes , Co-Chairperson Assistant Professor of Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. V/nce A. Hines I^ofessor of Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. William Dr\immond Professor of Education

PAGE 101

This dissertation was submitted to the Graduate Faculty of the College of Education and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December, 1975 Dean, College of /Education J Dean, Graduate School