vmurc,smitc11-

'4 r

DOCUMENT RESUME

ED 188 163 CS 005 533

IITIE Bilingual, Bicultural, and Bidialectal. StudiesRelattd to Reading and Communication Skills:Abstracts of Lcctoral Dibsertations Published in"Dissertation Abstracts International," Janu'arythrough June 1980 (Vol_ 40 Nos. 7 through 12)_ERIC Clearinghouse on Reading ald CommunicationSkills, Urbana, Ill.

PUB DATE 80NO1E 19p-: Pages may be marginally legible_

c

ECES PRICF MF01/PC01 PluE PoEt6qq.DESCRIPTORS American Indians: Annctated Bibliographies: Filin al

Education: *Bilingual Stud nts: Black Dialects:14'*Blacks: Chinese: *Doctora DissertatAns;

Educational Research: Elem tary Secendary EducaiAon.:

Language,Acquisition; *Lang pe Research: *Langq0e -

Skills; Language Usage: Mexic n Americans; Oral I"Language: Preschocl Children: Reading.Skills:*Spanish Speaking: writing (Composition)

tNSTITUTION

IDENTIFIERS Hawaii

ABSTRACT 94

,

This collection of.iib4stracts is part of a con uingseries providing informaticn on recent dectoraI dissertations. he 297._

Aitles deal with a variety of topics, including the4-fo11owin9::(1) ,

the language in a Spanish bilingual readingclassroom: (2) .th444ear1 y

stages.of language acquisition of tlack ohild-r heAn; (3) t de0 pdenCy -

'relation tetween oral latguage.and reading in Klingual Chinechildren: (4) language acquisition Of Hispanic third grade stOents:(5) the elect of.kindergarten claSsrccms where standard Engllsh isspoken,on -he speech of Black English speaking children; (6) -theeffect's of early and dellayed second language acguisitich on f eEnglish cospcsition skills of Spanish speaking junior high ôloo1

students; (7) the.composing procelinguistic-.ccgnitive skills- in tEnglish cemprebension scores ott'school children: (10) .vaventionsociolinguistic aspects of Engliish diversity among element*ry schoolaged students from Laguna Pueb11.(New Mexico!): and (12) th0 speech oflqwer-inceme tlack drug (GT1

ses of three black adolesQgptsp(Ek)low-income black child: 01

.ined ty Mexican-American cieAementaryn speech by children in Ha),Lwaii; (11-)

t

*' Reproductions suppflea by EDRE are the best that can be made *

from the original .documónt.4!t********************w******t*****************************0********

-ve

A

1) t or Prot l *or NT OF H AL III0t.1( Al ION *WW It ART

NATIONAL OrIIII A T.(IN

1)11% 001 LIAM NI WV, nr I N Or POO( ).V.( Y t tryt POAA

III( pf UPSON (IN ))PI.ANIJAIICINAI NI. II poINI Ill VII Ai (Itt OPINION'.\TALI 11 DO NOL Nil I \\Atilt V nr POI,1 IN.,1 01111 'At 11jI.IlIF,At ry.1111111 til

I II,, flhltIN 1.1...1111,s11,11

Bilingual, Bicultural, and Bidialectal.StudiesRelated to Reading and eommunication Skills:

AbstraCts of Doctoral Dissertations Published in DissertationAbstracts International, January through June.1980 (Vol. 40Nos. 7 through 12)

Clpiled"by the Staff of the

ERIC Clearinghouse on BlVading and Copmunication'Skills

"PERMISSION TO REPRODUCE THISMATERIAL HAS BEEN GRANTED BY

University Microfilms

International4.

TO THE EDUCATIONAL RESOIttRCESINFORMATION CENTER (ERIC)."

f.

Alkj-et

The dissertation titles contained here are published with

permissiolyof the University Microfilms- International,

publishers of Dissertation Abstracts International

(copyrights 1980 by University Microfilms International),

and may not be reproduced without their prior permission.

411.

a

4

4

1St

Thie bibliogtophy lute been cOmpiled tie putt oi 4 continuing mytieN deeignadCO make information on televant disSerlations available to imers of the ERICsystem, Monthly issues of Dissertation Abstracts Inteenational are reviewed in.ordcr h) compile abstract!; of dh;sertatfon!: oo tclated which (btu; hccomcaccessible in searches of Lli-o ERIC data base. Ordering inform! Ion loi the dis-

ortation!: themselvo!; In incInd.,d at thc ond ot Hue bibliography.

Abstracts of the following dissertations nr included in this colloctIon:

A.evedo, Carlos A.

EFFECTS OF ICTORIAL ELABORATION ONSPANISH M1NANT PILINGUAL STUDENTSFOR 1. RY TASKS IN RESPECTIVE LANG-UACP,S OF PROIQCIENCY

Arco -Torres, Edgardo

THE LANGUAGE IN A SPANISH BILINGUALREADING CLASSROOM: A DISCOURSEANALYSIS

Bowie, Geraldfne Robert Basley

A COMPARATIVE ANALYSIS OF VOCABULARYDIVERSITY, SYNTACTIC MATURITY, AND THECOMMUNICATIVE AND COGNITIVE PUNCTIONOE THE LANGUAGE OF IILACK FOUR-YyAR-OLD CHILDREN AT TWO SOCIOECONOMICLEVELS

Bragg, Willie Andrew

A STUDY QF THE EARLY STAGES OFLAN9U ACQUISITION AND MOR-PutoGçL DEVELOPMENT OF BLACKCH LDREN

Chu-Chang, Mae

THE DEPENDENCY RELATION BETWEEN ORALLANGUAGE AND READING 'IN BILINGUAL

CHIL6REN

qaronado,leopoldo Angel

THE EFFECTS OF DIFFERING DEGREESOF IFLINGUALISM ON THE COGNITIVEPERPORMANCE AND SCHOLASTICACHIEVEMENT OF SPANISH/ENGLISHBILINGUALS

Calthbertson, Beverly J.

A COMPARATIVE STUDY RELATING To ORALREADINC MISCUES AND FIRST LANCUACEINFLUMCE FOR MEXICAN-AMERICAN JUNIORHIGH SCHOOL UUDENTS

Daniels, Othella Edwards

LANGUAGE ACQUISITION OF HISPANIC.THIRD GRADE STUDENTS WITH PRIORPARTICIPATION IN THE CALIFORNI).STATE PRESCHOOL PROGRAM

t

nsato,,Arlinda Jane

A tSYCHOLINGUlSTIC ANALYSIS OF THEORAL READING MISCUES OP SELECTEDFIELD-DEPENDENT AND FIELD-INDEPENDENTNATIVE SPANISH-gPEAKING MEXICANAMERICAN FIRST-GRADc CHILDREN .

Ellsworth, Jennifer Kathleen

THE EFFECT OF KINDERGARTEN CLASSROOMS(WHERE STANDARD ENGLISH IS SPOKEN ONTHE SREECH OF BLACK ENGLISH SPEAKINGCHIT DREN

jJ

Ferris, Melvyn Roger.

THE EFFECTS OF THE EARLY AND DELAYEDSECOND LANGUAGE ACQUISITION,ON THEENGLISH COMPOSITION SKIL OF SPANISH-SPEAKING JUNIOR HIGH SCH OL STUDENTS

Fowler6 Robert James

AN ANALYSIS OF THR COMPOSING PROCESSESOF THREE BLACK ADOLESCENTS

Onnther, Vicki

4 COMPARISON OF BT4NCUAL ORAL LANG-*UAGE AND READING SKILLS AMONG LIMITEDENGLISH-SPEAKING STUDENTS FROM SPAN-ISH-SPEAKING BACKGROUNDS

Hicks, Shirley yields

. THE ASSESSMENT ()F CERTAIN LINGUISTIC-COGNITIVE SKILLS IN THE LOW-INCOMEBLACK CHILD REVISITED

Lewis, Shirley Ann'Redd

FACTORS AFFECTING THE ORAL CONftE-HENSION OF BLACK ELEMENTARY SCHOOLCHILDREN

Macias; Reynaldo Flores ,

MEXICANO/CHICANO SOCIOLINGUISTICBEHAVIOR AND LANGUAGE POLiCY INTHE UNITED STATES Li

Mandina, Genevieve Hogue

SEI1F-CONCEPT AND READING AtHIEVE-MENT IN SCHOOLS OF VARYING ETHNLCCOMPOSITIONS

McQueen, Adele Bolden

THE INFLUENCE OF INCREASED PARENT-CHILD VYRRAL INTERACTION ON THELANGUAGE FACILITY OF PRE-SCHOOLINNER-CITY BLACK CHILDREN

,Mcierhoffer-Lougoria, Lynn Vaulx

A SOCIOLINGUISTIVSTUDY OF LANGUAGE'USE AND-LANGUAGE PROFICIENCY IN

. SOUTH TEXAS.-4

Mitchell, Jacquelyn

STRATEGIES FOR ACHIEVING ONE-UPS-MANSHIP1 A DESCRIPTIVE ANALYSISOF AFRO-AMERICAN SIBLINGS IN'TWOSPEECH EVENTS

Pacheco, Richard

A STUDY OF ENCLISII COMP R EHENS I ON SCOREe;

OBTA I NED BY MEX I CAN AMER I CAN FLEmENTARYSCHOOL C1IILDRE14

-Perales, AIonso M.

THE 1;;FFV.CTS OF TEACHER-ORIENTED ANDSTUDENT-ORIENTED STRATEGIES ON SELF-CONCEPT, ENGLISH,LANGUAGE DEVELOPMENTAND.SOCIAL STUDIES ACHIEVEMENT OFFIFTH GRADE MEXICAN.AMERICAN STUDENTS

Prewitt Diaz, -Itligeph Orlando

AN ANALYSIS OF THE EFFECTS OF A BI-LINGUAL CURRICULUM ON MONOLINGUALSPANISH (MS) NINTH GRADERS AS COMPAREDWITH MONOLINGUAL ENGLISH (ME) AND BI-LINGU4e (BI) NINTH GRADERS WITH REGARDTO LANGUAGE DEVELOPMENT, ATTITUDETOWARD SCHOOL AND SELF CONCEPT

Purcell, April Komenaka

VARIATION.IN SPEECH BY CHILDREN INHAWAII

Saenz, Erasmo Andres

COGNITIVE DEVELOPMENT OF LANGUAGE;READING & MATHEMATICS IN A BILINGUAL/BICULTURAL PROGRAM

Stout, Steven Owen

SOCIOLINGUISTIC ASPATS OF ENGLISH'

DIVERSITY AMONG ELEMENTARY-AGED STUDENTSFROMJAGUNA PUEBLO I.

Taylor; Janet Marie Boijs

THE INFLUENCE OF SPECH VARIETY ONTEACHERS' EVALUATION OF READING COM-PREHENSION DURING AN ORAL READING ANDRECALL TASK

f

Tixier y Vigil, Yvonne L.

AN ANALYSIS OF ETHNIC GRARACTERISTACSAND ATTRIBUTIONS OF SUCCESS AND FAILUREON ATTITUDES TOWARD READING AND ACHIEVE-MENT MOTIVATION

()

Young, Jean Skoronski

COHESION IN SPOKEN DISCOURSE: .e

PRELIMINARY ANALYSIS OF THE SPEECHOF LOWER-INCOME BLACK DRIIG, ABUSERS

tied*

! ,),

FIFFE( -Is 01 PIC 1 MIA( 11 A Ill ill A I It r\J ( IN `,PANI`NI1

I)OMINAN I 1111.11Nt:I :Al. I 1,01..5 IN MI MOH I ASKS IN

RE.SPLIL. 1 Pi L 1 V."Ct.,1Ati flail it N( Oliki So 8011109

ACEvt00, CARL os A , 1.D I) Roston Untverstty School qf Education, 1980.

151pp iMajoi Piofes\oi Richard J I .amberski

The purpose Mali" investigation was to assess the relative achievementeffect of bilingual students foi dill erent types,of pis tolial elaboi at ioncomplementing author:wed instruction when given mernor, tasks in eitherof their rewcci lye pi On lent language's

A review of the research litelatule indicated that the usc of s isualizedmaterials has di( telsed greatly ni recent years (Dwyer, 1978), but theselecUon process for incorporating-visual stimuli into the bihngualinstrucuonal environment laeks supportive e.mpincal evidence Generally, itis believed that the use of visualired materials when used to complementoral or print instructi44 is not equally effective in facilitating acquisition orretrieval of Soot (pis I Iow evel their is little evidem r asse-ssing the additiveeffect during Irish moon of different les els of pictolial elahoiation (siropkbrie pictules. simple line pictures with wot d labels, and word tabels.only) onbilingual student achiesement nor .the related issue of which ptoficient ,

language (Fnglisli or Spanish) would be best to measure acquiredinformation This study was designed to addresS both issues.

The Enghth instfuetional materials consisted of an externally pacedaudierslide presentation on the human hean.. The test, materials consisted ofeighty items measuring four different cognitive tasks (prawing.Identification. Terminology and Comprehension) -I-lc-Anent conditionsdiffered only in type of pfeional elaboration received (Picture. Picture andWords, ot Words) arid in type of language evaluation received (English or

'Spanish) AU materials were developed from Dwyer (1972) and were

identical in text and visual fOrmat to black and white materials developed byDwyer and Lamberski (1977)

The sample consisted or filly-four undergraduate bilingual studentsfrom a Puerto Rican university. The students who volunteered for the studyhave been exposed to two languages since elementary school and arc able toconverse in both English arid Spanish, although they are native speakers ofthe Spanish language Subjects were matched according to Englishproficiency scores and randomly wigned ID one of Six treaurrent conditions(Picture-English, Pictufe/Words-English, Words/English, Picture/Spanish,View re/Wobts-Spanish, Or Words/Spanish). Subjects would view theaudio-slide aming materials under identical Instructional environmentsand would receive their respective self-paced language test materials forty-eight hours later.

Depentient variables consisted of achievement scores determined by thenumber of correct responses for each 20-item task for each subject and theamount of time spent interacting with each of the four test tasks for eachSulaject. The three factor ANOVA on the achievement scores consisting oftwo between factors (Pictorial Elaboration and Language Evaluation) andone within factor (Type or Test Tank) fesulted in one statistically significanttwo-way interaction where significance was specified as less than or equal to

Schen's poskbpc comparisoni-on the interaction of pictorialelaboration with test task indicated picture/word doinigance for thedrawing task Analysis of the time data was performed4ta a two-wayANOVA time for between subject treatment conditions combining time onthe respective test tasks. A significant main effect for language evaluationindicated longer response time necessary for subjects receiving the Spanishevaluation. While a significant main effect for type of pictortal e aborationyielded no significant post hoc comparison this appears to be q artifact ofthe present study with Picture/Words versus Words only cauafng thegreatest.time disparity.

4

my I AN(,IlAr.I. IN A SI'gN1',11 1111 IN(.1 'Al HY A NM;CLAtit,Ii0ONI A litts( tit 'WA- ()Oc1.'s:0 800086isati I mit', Los,Aluh.).1.11.1.) L'no.v 1,I. 19 /9

The main purposes of this dissertation were (1) to describe the Free-and Bound rcadimg I-xchange discourse c.itegolies in the Spanish readmgclassrooms within the context of a Bilingual I ducation prow am servinghigh entulline,lits of non 1 !iglus!) spc.ik ung students undet at hirvinglinguisn«mnonty students. arid on the basis of this dm I ipt ion. (2) 10&SUSS the (linden( cc bt't wccil Iva( Iitc dtc nicti Into 1110 I II) and 1 A.)w -

(10) piohciency groups lin the hais uf then w 010% 011 the SpanishProficiency Test, a subtest of the CFR AS Teachers TeAis farSpanish/English Bilingual Hucusion the piniciple hypothesis to bc testedis that there ate differences in the Free- and BoundTeaching Exchangediscourse categories of the teachers of the III and I 0 score groups Furtherquestions weie addreAsed which fc on the relationahips between theteachers characteristics, the discos and the students Spanich readingability scores

'The Subic( ts %vele teashm arid students in SO. oiud. thli d. and (mut)glade Spanish reading clawoorin. that were loi,ited in or Ilan/suburbansettings in the greater San I' uincisco Bay Mu, and were part of ,pecialprogram funded by both state and federal monies to serve the needs of the,non-Eng sh dominant student. The data consist of two sets El. 30-mihut evudeo'I4es of the Spanish reading classroolns Verbatim transcnpuons ofthe classloom discourse was made from audio portion of the video tapes

lbe tianscripts wei e analyzed using the Discourse Analysis System ofSinclair and Coulthard (1915), specifirdlly the Free and-fioundl cachirtgExchanges Teacher Warns reacher Elicit, Teacher Direct. l'uptl IrtforntPupll Elicit, Teacher Check Re-in:tuition I: Re-initiation 2; List andR epea t

The main firulinss of the study' indicate that teachers arc differentiated intheir amount of discourse by theij scores on the Spanish Proficiency Test.That is, the ill score group of teachers produced more discourse than theLO score group of teachers

One of the most important findings was that die 111 Proficiency scoregroup of teachers made less Spanish grammatical errors thai the 10Proficiency score group Corr dallorlal analysis also showed that-there was astrong neganve relationship between the amount of Spanish grammaticalerrors committed by teachers and student Spanish reading ability scoreaThe teachers who conmutted more Spanish grammatical errors had studentswith lower Spanish reading ability scores.

The findings of Os study indieate that languaV proficiency tests usedfor in-service and pre-service training.of Bilingual/Bicultural teachers, suchis tbe one used in this study, can be used to differentiate teachers on theirlanguage usc (discourse) and their grammar.

A COMPARATIVE ANAs?'SIS OF VOCABULARY DIVERSITY,SYNTACTIC MATURITY, ANIATHE COMMUNICATIVE ANDCOGNITIVE FUNCTION OF ,THE LANGUAGE OF BLACK FOUR-YEAR-OLD CRIWREN AT TWO 50CIOECONO4c LEVELS

Order No. 8004028BOWIE, GERALDINE ROBERT BASLEY, PH.D. SOtilhern IlliaoIs'UnlversiryarCarbondale,1979 136pp. Major Professor: Dr Nancy L. Quisenberry

The dissertation was designed to compare the vocabulary diversity,wmactic maturity, and the communicative and cognitive function OfUguage of Black four-year-olds at two sOtioeconomic levels. Mottspecifically, the research questions posed were: Does a difference existbetween the two groups in vocabitlary diversity as measured by thecorrmted typetoken'Tatio? Does a\differencelexist between the two groupsin the number of T-units elicited in the interview situation? Does adifference exist between the two groups in syntactic maturity as measuredby T-unit length? Does a difference exist between the two groups in thecommunicative function of the responses to questions in an interviewatkpation on the following variables: television, home, and bitthday? andDoes a difference exist between the two groups in the cognitive function ofthe responses to questions in an interview situation on the followingvariables: television, home, and birthday?

4

Oral language samples were elicited from twenty-thrft Head Start andtwenty-seven nursery skhool children in tape recto tied inter views I 0

determine vocabulary diversity a corrected type'tOken ratio was computedror each child. Syntactic maturity was determined by computing the totaliiumber of T-units arid the mean kngth of thc T-units for 94.-.1) child. ThecOmmunicative and cognitive ttinctions of the childrea%rftcsponses toquestions concerning three variables: tekvision. activities at home. and a

pictorial Slimy-los ofa birthday parry, were graphically presented and ".counted_ A one-way ANOVA was used to analyze and compare_thedifferenGsa-between the two groups on all three measures.

The findings of this investigation indicated that the vocabulary of thenursery vtool children was significantly (p a:, .0004) more diverse than theHead Start children. Significantly (p im .0247) lOnger T-units, as measuredby mean T-unit length, were produced by the nursery school children usedsignificantly (p .0384) more T-units than the Head Start children. Theconmiunicative function of the children's responses to questions concerningtelevision, activities at home, and the pictorial stimulus of a birthday partyshowed significance (p = 0068) favoring the nursery school children odyin second level modificitions dealing with home activilia Signifimnt

.048.1) differences favoring, the nursery school children were onlyfound in the cognitive function literal category concerning the activities ofthe horne.111ihough the differences between the two groups in the cognitiveand communicative flinction of their language were significant in only twoareas, the groups means for the nursery school children in every arra werehigher than the group means of the Head Sum children

It was concIuded that socioeconomic status may have an appreciaVeeffect on the langua development of Black preschool children. Animportant implication f the findings is the decisive role of the child's home

vlionMent _end the significance adults in that environment play in this

devflopment.A major recommendation was that a longitudinal study involving Black

children at the middle and lower socioeconomic levels using the samemeasures incorporated in the present study at levels bcyond the preschool

level might be developed. This would provide means of studymg the

consistency of the language differences across,tune and enable one togeneralivx the finding4..

A STUDY OF THE EARLY STAGES OF LANGUAGE ACQUI-

SITION AND MORPHOLOGICAL DEVELOPMENT OF BLACK

CHILDREN Order No. 8000679

BRAGG, Willie Andrew Ih.D. Indiana University, 1979, 9Opp.

A study was conducted to examine morpheme developmentof dialect ajid non-dlalect speaking children. The pnimary pur-pose wan to Invaptigate the relationship between mothers' speechpatterns and children's morpheme peroduction. Thus, the follow-ing research questions were examined:

1. What effect does the mothers' spqech patterns have onblack preschool children's production of certain inor-phemes?

2'. Is there a difference between black dialect and non-dialectspeaking children on morpheme production?

3. Are there age diffelences between black dialect and non-dialect speaking children on morpheme development?

4, Do black children melte through similar stages of mor-pheme development ae white middle-class children?

To answer the above questions, twenty-four black mothersand their children participated in the study. Also included inthe study were 12 dialect and 12 non-dialect speaking 3-, 4-, .

iced 5-year old children. The mothers were classified as either

dialect or non-dialect speakers according to their productionof four morphemes (i.e., present progressive, plural, posses-sive, third person regular). The children were classified asdialect or non-dialect speakere using the same criteria.

rwo 111,41 v(0110,104 0:pro dovoloi1.,1-and 'lewd ii elicit speechGamplen 1, ',II, (he ilwthol N INsi (IIe Chilli i co Fit tit, six activitycards with accompanying toys were constructed and used by themothers during mother-child interactions. The purpose of theactivity cards was to Olicit spontaneous speech samples fromthe mothers to determine whether they were dialect or non-dialect speakers. Second, a 24-Item sentence repetition taskwas used by the examiner (14r Ing ekansinsi child Interaction*to assess the children's production of eight morpheilles

The home visitation consisted of two parts. Part I Involvedmother-child interactions in which mothers used the activitytcards and accompanying toys while playing with their children. ......_Part II involved examlner-child intoractlons.in which the exam-iner presented the stimulus sentenees to the children. The 3-year old children were also shown picturee corresponding tothe stimulus sentences. From all transcriptions, throe scoreswere obtained for the participants, one for mother classifica-lion, oto for child morpheme production and another for deter-mining c did moan length of utterance level.

The ate wore subjected te analysis of varitace in whichlanguagi pe (dialect, non-dialect), chronological age (3, 4,and 5) and sex (male, female) we're the Independent vi6lableswhile morpheme selection was the (top:indent variable.

Contrary to expectations, no significant main effeet differ-ence was noted between dialect and non-dialect speaking chit- .

.... droll on total min-phonic development. One of the four targetmorphemes (i.e., possessive) attained significance. However,modest group means wore evidenced favoring non-diale'ct speak-ing chil iren. Subsequent analysis ehowed ettnIficant Interac-tion ween dialect and age. The greatest Increases occurredfor non-dialect speaking children between three and four yearset age while dialect.speaking children obtained maximum scoreincrement between four and five years of age.

It was concluded that black children learning language en-counter several different language models in their linguistic

A environment. Although the effects of these models are ques-tionable, children some how extract linguistic infprmation whichaids them yittring the language learning process. In addition tothe language models provided, preschool programstwith parentsend teachers an partners play a vital role In language growthand development of young children. Parents must be providedwith the necessary techniqaes and materials nurturing languageacquisition and teachers must be provided with the necessaryskills so that they can adequate! assess the strengths and weak-nesses of young blaCk children. eachers multt understand lan-guage and cultuital differences so Niat they can plan more effec-tive educational programs building on thoee skills that youngblack children bring to the learning situation..

THE DEPENDENCY RELATION BETWEEN ORAL LANGUAGE

AND READING IN BILINGUAL CHILDREN Order No. 8005239

CHU-CIIANG, MAE. Fn.D. Boston University School of Education. 1979.

116pp. Major Professor: Dr. Mana Estela Brisk

The purpose of the preseut study was to investigate the proposed'dependency relation between oral language and reading

Since the 1950's when linguistic theory was applied to reading, theAssumption has been that reading draws on the same linguistic competenceunderlying listening and speaking. Bloomfield (1942) and Rid (1962)regarded learning to read as an act of establishing the gUpheme-phonemecorrespondence. Once the transfer to a phonemic string was made, themessage was decoded. Developments in the field of linguistics have broughtcorresponding changes in the theories of reading. Goodman's (1970) modelof reading as a psycholinguistic guessing game was influenced by Chomsky's(1965) transfomiational generative grammar and the findings of Stevens and

Halle (1967) that listening' is a process of analysis-by-synthesis. Theassumption Imes that reading and listening are parallel processes.

Bilingual educators have attributed reading failure, in children who arelimited in their knowledge of English, to the incongruency between'thechild's spoken language and the language which he is taught to read. Theassumption has been that leading is dependent on an existing oral language.repertoire.

-1^ tik"leseirAv Ss- 4tt., ' .?"

In the present studs,. Cantonese speaker's learning tn read a ritienCiuncu %Tic w.r.,(1 to (eNt the alio% c assumption 1 hey 8IC Id Cal Slitirt

since in Chinese there is no correspondence between indlvidual graphemeand phoneme, and thtrefore, this factor is naturally controlled for.Fkirthermore, the Cantonese spoken dialect is different lexically, in manycases, trom the written Chinese

The three h)potheses tested were: (I) lhat accuracy in reading for themeaning of words will he higher for worrN which are congruent with thevernacular of the reader than words which are incongruent. (7) chat %%oldswhich arc incongruent with the vernaculas of the reader will require longerprOcessing time duc to the presence or an additional translation procezfrom the ckier language to the child's vernacular, and (3) That readingcomprehension scores on wnhen prose passages whichlare congruent vmhthe vernacular of the reader will be higher than passages which are

_incongruent.The quesoonsiaddressed were experimentally tested first using a aoid

comprehension paradigm And then the Clore procedure for reading-comprehension of prose passages The subjects were fourth and fifth gradeCantonese students in a Chinese bilingual program from an inner citypublic school system. The perfomiance or good readers was compared tothat of poor readers on the tests.

Results supported the first and the third hypotheses at .05 level and 005level repectively Reading of words as well as prose passages Was easier

when the written matenals were congruent with the .vernaculzu of the

reader. In addition, this was found to apply to the good readers as well as

the poor readers The hypothesis that a translation process was required for

incongruent words and passages into the vernacular was not supported.

There were no significant differences in response time dependent on type of

material. The findings were discussed in relation to a model of monolingual

vs bilingual reading.The implication for reading instruction andl-)ittngual education is that

reading should be taught in the reader's native language. Intensive oral

language training in the second language is recommended before teaching

reading in that language.

THE EFFECTS OF DIFFRING DEGREES OF BILINGUALISMON THE COGNITIVE PERFORMANCE AND SCHOLASTICACIHEVEMENT OF SPANISH/ENGLISH BILINGUALS

Order No. 8009844

CORONADO. LEOPOLDO ANGEL, Pit fl The University of Texcis at Austin.

159pP. Supervisor: Thomas D. Horn

The purOse of this study was to determine if there exists a relationshipbetween the degree of bilingualism of Mexican American students and theirabiliv to handle the eight communication concepts in the MAT-SEA-CALOral Proficiency Test: Number, Temporality, Spatial Relationship,Identification, Classification, Case Relationship, Interrogation andNegation. The study also attempted to determine the relationship betweendegree of bilingualism and school performance as Ineasured by the SRAAssessment Survey.

For this study, a sample of 271 Mexican American pupils comprised theexperimental group Tht control group consisted of 198 monolingualEnglish-speaking (ME) faupils. The total sample of 469 elementary students(grades K-6) were horn the Granger School District, Granger, Washington.

The experimental group was further subdivided according to theirlanguage proficiency scores on the MAT-SPA-CAL Oral Proficiency Test(English and Spanish versions). All Of the subjects in the experimentalgroup were assigned to one of the following language groups: low-balancedbilinguals (LB); middle-balanced bilinguals (MB): high-balanced bilinguals

.0:Ina); Spanish-dominanEpupils (SD): and, English-dominant pupils(ED).Analysis of covariance was used to compare the five language groupg

with the (ME) control group and with each other on their ability to handlethe eight communication concepts in the MAT-SEA-CAL Test (as ameasure of cognitive performance) and on their mean scores on the SRA

. Assessment Survey (as a measure of school performance). Mi data analyseswere conducted at the .01 level of significance.

With the exception of the LB group, which had problems in three of the

eight eOmmunication concepts in both languages, thus suggesting the

Ocesibility of Some type of cognitive problem, all the other groupswere able

to handle all eight, concepts In at least one Of the languages,

1.,

As for t. ho4 ci per foTili.InCe. this study les rated as have oilier slur:hes.Linn i 111,1r:1 4.1!1 tn depl ell 011 3 \LIM! 3(113'ell te'S3. vu

thc SRA Asses.sment Survey. 1( that student has an imperfect knowledge orthe language in which the zest is admimstered ibis finding is reinforced bythe poor test pet ((alliance of the Mil group and I-I) group when comparedto the (ME) control group According to the findings in this study, studentswith a high degree of bilingualism pet form as well as. or better than, theother Lingnige groims licti the t, I emu rol group :1% demonstrated by theiractual s( Ole\ on the SR A. Assessment Sinvey

A ( OMPARA I !VI' S !AIM. IZEIA.Il \c, I() OR U. RI; kI)INGNLISCULS AM) FIRS I LA.NGUAGE INFLUENCE FOR MEXICAN-AMERICAN JUNIOR I IIGII SCHOOL STUDENTS

Order No. 8004773

Ccrunrarso, Brvriu v ) Ft) n Nonhern Arhona Universiry, 1979138pp Adviser: Dr Ray Hagelberg

This study analyzed the oral reading miscues ofiwo selected groups of(unior high school Mexican-American students in relaupn to first language

fluence The grammatical and semantic aspects of the oral reading processWere also ins esugated. First langitage influence was studied in relation tothe variabks of sex. intelligence socio-economic status, and literacy. TheRMI was utilized for the analysis.

There was no significant difference in the oral reading miscues fordialect miscues, graphic miscues, gramMatical flinction miscues, or forsemantic miscues. A significant difference between the two groups was

." found in relation to the retelling score, another component of the semanticaspect

The demographic characteristics of sex showed no ddrerencein regarld tofirst language influence. The second variable of intelligence indicated a

. significant difference between the higher group and the lower group inrelation to first language influence. Socio-economic slatUS, the thirdcharacteristic, denoted no significant difference in relation to first languageinfluence. Literacy, the fourth characteriStic, indicated a significantdifference between the higher and lower group

7.,

LANGUAGE ACQUISITION OF HISPANIC THIRD GRADESTUDENTS WITH PRIOR PARTICIPATION IN THE CALIFORNIA STATE PRESCHOOL PROGRAM

DANIELS, Othella Edwards, Ed,D. University of SouthernCalifornia, 1979, Chairperson: Professor Margaret E. Smart

Problem. The problem of this stUdy was to determine ifthere were variables which, prior to classification as Non-English Speaking (NES), Limited-English Speaking (LES), orFluent-English Speaking (FES), would predict the level of En -gl ish fluency of Hispanic students In the third grade,

After a review of the literature, the following predictorvariables were selected; Gender; Age; Home language, Pre-school experience; Retention; Number of parents in the home.

To measure the relatIondhlp, the followlpg null hypothesiswas generated: None of the slx predictor variables will in-dividually or jointly eignificantly discriminate between levelsof English fluency.

Procedures. The sample was composed of 503 subjectsdrawn from the third grade population of students enrolled inseven elementary schools, each of which housed a Children'sCenter on the school Bite and had a 90% or more Hispanic pop-ulati6h. Sample selection criteria of the ex,post facto studywere limited to Hispanic studente who had been administeredthe San Diego Observation Assessment Instrument (SDOAI) inthe third ggade and whose Student Data Profile Cards containedcomplete tutor mation.

The SDOAI measured the ntudent'n levet (If EnKiinii vertiAIcoinnuinicuthin at three language levels- I.evel I - lithel ohjeets a.nd, plaures-Nun-English Speaking (NEZi), Level Iitell about the picture using sentences andlor phases-Limited-English Speaking (LES); Level 111 - expand convernationallYfrom the picture-Fluent-English Speaking (FES).

The stepwise discriminant analysis technique war; iised todetermine which of the independent variables indivhtually orjointly are the best predittern of gi,,f nip ineinhernhip in "NES,LES, or FES categories.

Tor discussion perposes of the findings, the chi-squarefUnction was used ae a measure of association to determine ifthere were significant associations between the independentvariables studied and English fluency levels.

Findi s. (1) Gender and hoMe language were significantpredictors of levels of English fluency (1) -" .01). (2) Classifi-cation of the levels of English fluency by the two significantvariables was correct In 20% of the cases. (3) Gender andhome language correctly identified NES students but dld notproperly classify LES or FES students.

Conclusions. (1) Gender and home language are good pre-dictors of English fluency levels for Hispanic students fn thethird grade. (2) Age, preschool experience, retention, andnumber of parents in the horns are poor predictors of Englishfluency levels for Hispanic etudente in the third grade. (3)Gen-der and.home language do not correctly classify levels of En..glish fluency for all Hispanic students in the third grade.

Recommendations. Based upon the findings of this study,reco-mmendatIons-for further research are offered as follows:(1) ivuntlyze other combinations of variables which may relatemore closely to home language and gender to make them betterpredictors of levels of English fluency (2) study differencesbetween boys and girls which may explain why Hispanic girlsare more likely to be classified as NovEnglieh Speaking andLimited-English Speaking than Hispanic boys, (3).4nalyze dif-ferences among preschool programs which may affect the levelof English fluency of students In tho primary grades, (4) con-duct a study of the construct validity of SDOAI, (5) study thelong-term effects of preschool programs, (6) collect norma-tive developmental data on children In preschool, kindergarten,and primary grades, and (7) develop a standard methodologyfor data analysis.

A P5YCHOLINGU1STIC ANALYSIS OF THE ORAL REAbINGMISCUES OF SELECTED FIELD-DEPENDENT AND FIELD-INDEPENDENT NATIVE SPANISH-SPEAKING MEXICANAMERICAN FIRST-GRADE CHILDREN Order No. 7926279

EATON, A rlinda Jane, Ph.D. The University of Texas atAustin, 1979. 359Pp. Co-Supervisors: Drs. Thomas D. Horn,Rudolph F. Martin

The present study was designed to analyze in depth anddescribe the oral reading behavior of native Spanish-speakingMexican American child renwho were identif led by the Children's

inbedded Figares Test (Within; 1950) as being field-dependent(Thy, d-dependent- independent (FD/I), and field-Independent(FI) in cognitive style orientation, as they read in their nativelanguage, Spanish, and in their second language, English. Thesample included 19 Mexican American first-grade childrenattending two different schools within the same district of asoutheastern Texas border community. All subjects receivedtheir initial fermal reading instruction In the first grade.Balt of the children received instruction hi Spanish whilelearning English symbol to sound correspondences for the

, first semester of the year, transferring to:English reading

inntruction at the onset of the ;444* mut twnientei. HeadingmSpanish occurred an time peritted during the latter half f

the yeal. 1he lernaining unbjectn I ecoved Ii,hlI.ti ieading In-struction in English with Spanish reading instruction added atthe midyear.

The oncopt of the reading process, construed as a psycho-linguistic prooess which involves nampling, predicting, testing-and confirming, was Investigated by analyzing the oral readingmiscues generated by the subjects while reading in their tarolanguages. The analysis was conducted as detailed by_Y. Good-man and Burke (1972) in the Rectin Miscue Inventoiy; sub-jects were listened to as they read sereal;;;s In English andSpanish, during separate sessions,-until they had generated25 miscues in each language. After reading each seleetion,subjects attempted to retell as much as possible of what theycould remember having read.

In light of the limitations of the studygeneralizability,classification of miscues, experimenter variable, and designa-tion uf cognitive style groupsthe following tent.Olve conyu-*ions are offered for consideration:

1. Native Spanish-speaking Mexican American bilingualreaders, identified and grouped as FD and Fl in cognitive

8 style orientation, looked like very distinct and separate groupswhen they read in both languages. The manner in which theyprocessed printed material looked very much the same acrossL'inguages. For the most part, the FIs employed the variouscue systems to a greater extent than did the FD6 In 17 of the21 categories analyzed in each language.

1 The reading strategiee employed by the bilingual sub-jects of the present analysis did not vary according to thelanguage in which they read; the same strategies were ap-plied across languages. These data support Previous researchfindings which also Indicate that there Is ono reading process.

subjects drew on the same linguistic cue systemsin their attempts to reconstruct meanbig from print whilereading In both their first and second languages. The differ-ence between the two extreme groups Iles in the extent towhiA they employed the cue systems.

9. The Fl readers comprehended more during the readingact than did the FDa in both Spanish.and English. The FIs ap-peared to understand substantially more while retading In theirfirst language, whereas the FDa understood slightly morewhile reading In their second language.

5. Retelling figures paralleled each other across lan-guages; the FIs retold a greater amount of their reading thandid the FDs. Both were able to relate a considerably greateramount of their readthg in Spanish than of their reading hiEnglish.

11-1E EFFECT OF KINDERGAR1EN CLASSROOMS WHERE.STANDARD ENGLISH IS SPOKEN ON TH.E SPEECH OFBLACK ENGUSII SPEAKING CHILDREN Order No.8008811aLSWORT1I, JENNIFER KATHLEEN, Pit D The University al-Wisconsin -Madison, 1979. 225pp. Supervisor: Professor Herbert M. Kliebard

This study examined the effect of kindergarten classrooms wherestandard English is spoken on the speech of Black English speakingchildren The research analyzed samples of language used by twenty-twoBlack English speaking children in seven kindergarten classrooms *threeschools during one academic year. Speech samples from each ofstudents were analyzed to identify the phonological and syntactic featucharacterizing the speech in September, February and May. Tht followingquestions were addressed In the ttudy: (1) What phonological and syntagis,_features constituted Black English in the speech sample0 (2) To whatextent did the Black English speaking students increase their use of standardEnglish phonological and syntactic forms during the year? (3) To whatextent did the Black -English speaking students retain Black Englishphonological and syntactic featuresduring the year? (4) What language-climate factorsin the classroom, including such things as the presence ofStandard speaking peers and teacher knowledge of Black English, contributetO the acquisition or non-acquisition of standard forms? This studydocumented linguistic change due to what has been called the."naturalinformal authority of the speech community."

This lese,o,h was tondo( led wrilun die theorem al frameii k of the-.difference- as opposed to -deficit- theory model of language lin former.based on desi i rimy(' linguism s and mole rei (nth.- 011 fIr F(.11('!,01%e

transformational model of language. posits that non mainstream varieties ofEnglish are rule-governed, and differ in systematic ways (Torn standardEnglish They are in no way -deficient

The study used methods and techntques established by dialectgeogiaphers. soi iolinguisis, and educators concetned with dialect diversityin the classroom I ape-recorded samples of the sordent s' cper Ii weretranscribed using the International Phonetic Alphabet and analyzed for hvcgrammatical featu es (Lei u copula. Invariant be, Suffual z, MultipleNegation. Pronominal Apposition). and four phonological features (Ward-final consonant ('luster simplification. th soupds. Syllable-final -d, antrorostvocalic r). Building on Interview techniques established by socinfinguists.the investigator elicited speech samples by asking the students to watch aten-minute animated film with the sound track turned oft and to re-tell thestory in then own winds Students were 'writes viewed in pairs to minuilizefeelings of intimidation In addition. some quesinnis from Shuy,and Riley's Deo ort Dialect Study were posed

in At:Wit-Inn to the tape retor ded sathples of the students' speech, thedata sources for this study included notes made during direct observation ofone session in each of the seven classrooms, and interviews with the fivekindergarten teachers to determine their knowledge of Black. Fngfish and toestablish the family backgrounds of the students.

Analysis of the data indicated that the Black Vnglish speaking studentsas a group did thcrease their use of standard Frighsh forms of thephonological and syntactic features considered A hierarchy of degree ofchange in use of features was established rhe findings suggested thefollowing conclusions- (1) The features showing the most change in usewere those identified in other studies as being the most stigmauted andsocially diagnostic; (2) The students showing the most change in the use offeatures were those who began the school year using Black Friglish formswith-greater frequency than the 'group as a whole; (3) Factors contnbuung

to the lingtnstic change appeared to include both linguistic constraints andextralinguistic constraints such as student awareness or dialect differences.peer influence, and teacher knowledge of Black Frighsh; (4) The hierarchyof degree of phange in use of features was suggested as a practical guide forteachers in determining which features to address systfmancally in the.cla.ssroom.

Further resurch sampling the Informants' speech in the second, fourth,and sixth grades would probably shed light on both substanbve change orlack of change, and the roles of the various factors cited above,

THE EFFECTS OF THE EARLY AND DELAYED S)CtINDLANGUAGE ACQUISITION ON THE ENGLISH COMPOSITIONSKILLS OF SPANISH-SPEAKING JUNIOR H1611 SCHOOLSTUDENTS Order No. 8001915

FERRIS, Melvyn Roger, Ph.D. Stanford University, 1W19.11l4pp.

Reeehtly, in the field of bilingual education, there has beenconsiderable debate over the use of a child'e ler or 2nd lan-guage as mediums of classroom instruction (Von Maititz, 1975;U.S. Commission on Civil Rights, 1975; Garcia, 1978). Thepresent research project endeavored to provide further insightinto this issue by contraeting the effects of the early and de-layed second language acquisltkon in English of two groups ofSpanish-Speaking students: those born and schooled in Mexico(k-2,3) (Group A), and those/born and schooled in the UnitedStates (k-7,8) (Group B). At the date of the study, both groupswere between the ages of 13 and 14 years old, were from sim-liar socio-economic backgrounds, and were relatively balanced ;

with regard to sex. Students in Group A were literate in SpanishAat least to the-3rd grade level while those in Group B were not.To assist with the interpretation of results, other variables /measured were; bilingual dominance in Spanish, grades earnqdin English, foreign travel to Mexico, mobility rate in the UnitedStates, cchool attendance in the U,S. and Mexico, and varioui./attitudes and values about school.

It was hypothenived that students who i(co.ived cal iv nnothertongue instructlon In Mexico would not be sIgnificantly different(p:. W) in fenglish c.,mpofit I,m skills fl.:In sundial Spanishspeaking students born and nhooled entirely In tho UnitedStates after the former groop had received three or four yearsof U.S. public ochool instructiun.

A. random selection of 30 students per group was made from2 pools of 50 and 58 students respectfully. Qualitication fureach po(4 was determined by a I higui st I.. IrAckertnind 1111VatiOnnalr e and the lnteramerican test of Spaninh reading, level .

The collection of data for the etudy required students towrite an essay based on a 15 minute flim topic selected for itslevel of interest and relevance to Spaqish -speaking students.Then, essays from both groups were evaluated by 3 separatemethods of componition analysis: a holistic analysis for para-graph development, sentence boundaries and verb inflhctions,a frequency count analysis for various grammar errors (thefused sentence, the period fault, verb tense, pronoun agreeMent,article agreement, and the possessive), and a T -unit analysisfor structural complexity by the number of words per T-unit,the number of clauses per T -unit, and the average clause length.

The statistical procedures selected for the evaluation ofdata wore the t-teet, chi-square, and the Spearman non-para-metric correlation.

From the analysis of data, it was found that a significantdifference (p: .05) was tenaWe only for the holistic variableverb inflections And the frequency count v.Lriable verb tense. -

.14 h e remaining ten variables were not significantly different forthe two groups.

giaFrom a consideration of the research ftndings, it was con-

cluded that E.:My schooling in Mexico in the first language,Spanish, was, at least, not a detriment to the later acquisitionof English by Group A students.

The results of the study may be more fully understood byreviewing the outcome of variables which were significantlydifferent between the 2 groups. Though it was determined HaatGroup A students were significantly lower (p .05) in socio-economic stiLtus, they were, on the other hand, significantlyhigher (p 4.M, for grades earned in English, language domi-nance En Spanish, forelgn travel to Mexico, student reports ofa desire for higher education, motivation to get good grades,student-teacher discusaion about school work, and student..teacher discussion about goals and interest&

It was thus ultimately concluded that the achievement inEnglish composition by Group A students was most likely acombination of both linguistic variables and-various attitudesand values which students received from their p-irents andothers in theiis family background.

AN ANALYSIS OF TIIE COMPOSING PR FSSES OF TIIREEBLACK ADOLFZCENTS Order No. 8004807

FowLER, ROBERT AMES, PH.D. University ofPit h, 1979. 243pp.

The purpose of this case study investigation was t -fold: (1) to providest descriptive profile of three Black adblescents' composing processes 1.5

Manifested in transactional, expressive, and poetic compositions andinformation ascertained from personal interviews and observations; and(2) to see if information contained in these descriptive profiles variedaccording to the individual's classification as a "low," "moderate," or4'his,h"nonstandard dialect user.

The research procedures hivolved the following steps: (1) A groUp Of 45Black adolescents enrolled in an alternative high schkol was givensdiagnOstic ln-class composition to write. (2) The compostitions were reedand examined for the frequency of occurrence of specific predeterminedBlack dialect featires in order to determine the "degree" of dialect presentin the compositions. (3) From the 45 studentS, three were chosen as subjectsfor the study: one "low" nonstandard dialect user, one "moderate"nonstandard dialect User; and one "high" nonstandard dialect user. (4) Thesubjects were assigned six compositions to write: They were requested Wuse the talk/write and the non-talk/write method.s in each of three modes ofdlscourse--transactional, poetic, and expressive. ,

'Me students' writing processes were analyted in two ways ( I) hyotgedrving them dining the NOV% of 1111111A111M/((11(

p oc:sest:in, and revision; and (2) according to their use of thr intellectualpr of change, contrast, classification. physical context, focus, andsequence.

The study revealed that. generally, the low nonstanaard dialect userneeded the irast amount of time to write tin ioost (actual number of woids),and the high nonstandard dialect user nohred the most am-ount of time to

write the least The low and moderate nonstandard dialect users generallythought of mental outlines prior to wining, but the lugh nointandald dialectuser was more concerned with putting her thoughts on paper the "right"way. Also, the low nonstandard dialecsuser made the leaskamount ofpauses during the writing process, and the high nonstandard dialect usermad. the most amount 'All three Of the students' Writing patterns werecontinuously progressive and additive in nature The student who wasclassified as the high nonstandat d dialect user made the greatest number oftranslations/front Black dialect to standard English doting the writingprocess, while thc low nonstandard dialect user made the least number ofthesp-rrCnslations Also, all three of the students placed a great deal of

t--eniphasis upon grammatical ruse revisions. Thc three students revealed nosignificant differences in their Zse of the intellectual processes of change.physical context, focus, classification. contrast, and sequence; and they wereable to use all of these processms.

Some major conclusions drawn from this study were as follows: (1) Thedegree to which the *dent was a nonstandard dialect user appeared toinfluence the degree of time spent on a writing assignment. as well as thenumber of words produced. (2) Wriung fluency appeared to be influencedby the extent to which the student was a nonstandard dialect user (3) Theextent to which the student was a nonstandard dialect user appeared toinfluence the extent of her translations from Black dialect to standardEnglish during the writing process. (4) ThertalkSwrite process took moretime to produce less written words than the non-talk/write process

A COMPARISON OF BILINGUAL ORAL LANGUAGE ANDREADING SKILLS AMONG LIMITED ENGLISI-I-SPEAKINGSTUDENTS FROM SPANISI-I-SPEAKING 13ACKGROUNDS

Order No. 8007379

GUNTUR, VICKI, PH.D. Northwestern University. 1919. 255pss Adviser:Norman D Bowers

This study was an attempt to provide practitioners with data upon ss7higyto base instructional decisions in the area of bilingual education Twohypotheses were generated regarding the acquisition of oral language andreading skills among limited English-spe4ers. The first related todifferences an-iong instructional groups in performance on oral language endreading measures. The groups were based on the following bilingual readingapproaches: (1) the native language approach with initial readinginstruction in Spanish; (2) the concurrent approach with reading -

instruction in Spanish and English; (3) the direct method with leadinginstruction exclusively in English.

The second hypothesis related to the identification of sociological,instructional, and linguistic variables associated with the acquisition ofskills. Variables included sex, ethnic background. socio-ecOnomic status,years enrolled in a bilingual program, fluency in LI and L2, ESLinstruction, teacher training, and teacher attitudes

The sample consisted of 306 Spanish background students of limitedEnglish-speaking proficiency, ages 6, 8. and 10, who were enrolled in eightpublic and five nonpnblic schools in Chicago. Students were pretested-inOctober 1977 and posttested in May 1978, in Spanish and English, on ses enoral language and reading measures. Additionally, data were collected onstudent and teacher background informatiOn as well as teacher attitudes.

Data were analyzed separately for the three age groups and furthersubdivided by'skill area, i.e.. oral English. English reading, and Spanishreading. Multivariate analyses of covariance were used for the firsthypothesis relating to differences among reading instructional groups. TheAndings were as follows: (1) in oral English proficiency, 6 aDj.kyear oldsreceiving reading instruction exclusively in English perfoVed significantlybetter than students receiving bilingual instruction; groups of 10 year oldSperformed MI a comparable basis regardless of instructional approach;(2) in English reading, 6 year olds receiving instruction exclusively in.English performed significantly better than students receiving bilingual'instsoetionSgroups Of13 and 30 year -tilds performed on a comparable basisreprdless of instructional approachs(3) in Spanish reading, 6, 8, and 10Year Olds receiving reading instruction in Spanish and English performedsignificantly better than students receiving reading instruction ixclusivelY inEnglish

WrIli referent e ro I no tloottlic ihe funding. Nhoo hi be interpreted with4.4ution onse the Luniii;aus e benefits of bthugual uituut.uuu Jo nut alstaysurface in cross-Sectional designs It should be noted. however, that gainswere evident for the bilingual program participants

Regression techniques were used for the set ond hypothesis relating tothe independent variables aqsociated with the acquisition of oral languageand reading skills The findings were as follows (1) among stains children,1.1 flue vv-is a good prrtbr tor of I ? (,') for ;In Aye groups, 1 1oral skills weie consistent pi edit-tots of 1 1 reading sk ( for youngerchildren. LSI:instruction, teacher training, and teacher attitudes werepositively associated with the acquisition of skills, (4) no definitive trendswere observed with respect to other vartables.such as sex, ethnicbackground, and socio-economic status.

Findings from this study suggest sesetal implications for praciitionersConsideration should be given to implementing different instructionalapploaches tot dilletent student populations Additionally, considerationshould be given to designing plogr anis which pros ide a film bast in the LIprior to um oducing reading and content area instruction m the 1.2. Finally,provisioas should be made for staff development activities which fosterpositive attitudes and prepare teachers in ESL and bilingual methodologies.

Recommendauons for future research include longitudinal studiesmessing specific educational approaches for particular student populations;studies on the relationchip between the 1.1 and 1.2. including transfer ofskills: and, studies identifying panic-1'31er teaching techniques whichcontribute to the academic success of second language learners in all areasattic curriculum

vrs

Tlfr ASSESSMENT OF CERTAIN LINGUISTIC-COGNITIVESKILLS IN THE _l_Q_NV INCOME BLACK CHILD REVISITED

Order No. 8001937

HICKS, Shi.rley Fields,IPPh.D. Stanford University, 1979.114pp.

As a result of the lyw income black child's poor perfor-mance on the Illinois Test of Psycholinguistic Abilities (ITPA)iii general, and lower sco (is on the auditory subtests than onthe visual counterparts, e ucators perceive this child as onewho is deficient in lingo! ic -cognitive ukiilri. The linguistic-cognitive deficit, as projected by the 1TPA, supports theories.p( black intellectual inferiority as set forth by Terman, Jensen,and Bereiter and Engelmann. 9

Three factors that have contributed to the low-Income blackchild's depressed performance on the ITPA are I) his exclu-sion front the standardization population; 2) the middle-classspecificity of (he content and vocabulary of the 1TPA; and 3) the

A middle-class bias In administration procedures and test milieu.The 1TPA consists of five auditory subtests and five visual

counterparts, plus two supplementary subtests. This study fo-cused on standardized and modified versions of three subtests:Auditory Association, Verbal Expression, and Grammatic Clo-sure. The research hypotheses are as ((knows:

41111111pothesis I: The performance of the low-income blackchild will Improve significantly on (he Auditory Associa-tion, Verbal Expression, and Grammatic Closure Subtestswhen the content or the administration frocedures are Mod-ified to render them more compatible with the backgroundof tht subject.

Hypothesis II: The Scores on the motility subtestsshow a significiintly higher correlation with a relativelyculture-free perceptive-cognitive -motor rating than thestandardized analogs.

Criteria for inclusion in the subject population were 1) 1ov/-income as required for Follow-Through placement; 2) ver-nacular black English (VBE) dominance or VBE/standardEng lish (SE) balance as determined by a production test; 3) en--rollment in a regular third or fourth gra4 class; and 4) chro-nological age of 10-1 eve below.

2

.14

let deo

The final sample wan maitc tu and

10 boys A epea ed. 'nett enicid !ft-41w, ,..,trt iMll s;o11

bt:t.s givt,t lxx4.h it,tidtltIt:litts, CALL WIWI:LI A;t1:1! at Arcontrol. The three standardized subtests and the modifiedcounterparts constituted the control and experimental condittone, respectively. The independent variables were the Productlon Tee( and the Pet ceplive rngnitive Motio Ilming TheProduction Tent consists of two parallel fox ins A (V11E) a fl d

(8E). Die i'v-lieptive Cognitive iUiigfifteen relatively cultut e flee items

A.t-test analysis shows a higher mean performance on eachof the throe modified subtento than on the parallel standardizedforges at the .001 level of significance. Chi -square cross tabu-lations indicate that a significant number of subfeets who scoredabove the deficit range on the modified Auditory Assiwiationand Verbal Exprestiton Subtest ate, hatted on their 913ndardized suhtest scores, potential candidates Go learning disabiltties ethsses Scores 29 constitute the deficit area.

A t -test analysis of significance of difference between twocorrelation coefficients for correlated san151es shows none ofthe correlational differences between the Perceptive -CognitiveMotor Rating and parallel versions of the three auditory sub-tests to be significant. An item analysis shows a -lack ofcorrespondence between item oildering On the standardizedAuditory Association Subtest and item difficulty.

The data suggest that the three ITPA IsliPtenta under investlgation are biased against the low -dricome black child..

FACTORS AFFECTING THE ORAL COMPREHENSION OFBLACK ELEMENTARY SCHOOL CIULDREN Order No. 8006376

SHIRI.EY ANN Ram, Pat D. Stanford University, 1979. 148PP

This study concerned the effect of language dominance and cultural andlanguage attitudes on the listening comprehension of Black elementary

, sclaool children. An investigation was carried out to determine whetherdominince in Black Vernacular English (IWE) or.Standard English (SE) orhigh or low attitudes toward Black culture and Black language had thegreater effect on subjeas' perfonnance on comprehension tests treated withBVE and SE language and with Black or Neutral to Blatk story content

Sateen Black subjects and eight White subjects were selected from eachof grades two, four and six according to high and low ability, sex andlanguage dominance resulting in a total of seventy-two subjects_ Thelanguage dominance of Black subjects was determined by a Black Englishdominance test whicif identified Black Vernacular English (BVE) and BlackStandard English (BSE) language dominance. White subjects were selectedwho were Standard.English speakefg(WSE) All subjects' attitudes towardBlack culture and Black language were measured by Black cultural andBlack language attitude tests.

The criterion measureconsisted of four tape-recorded storiesmanipulated for language and content as follows: BVE - Black content:BYE Neutral! content: SE Black content; and SE - Neptral content.Subjects lisened to each story and then answered eight (apc-recorded,multiple choice questions accompanying each story.

,Analysis of variance, correlational analysis and t-tests were used todetermine whether Black subjects perfotmed differently on BVE and SEtreatinents according to language dominance and language attitudes,whether spbjects perfonned differently on Black and Neutral ciantenttreatments according to culturAl attitudes and whether the WSE controlgreup per(ormed in a pattern different frbm the BVE and BSE groups.

KO significant difference Iss found in the performance Of the EVE andSSFEonps on the BVE treatm t, while BSE subjects perforinedsignifIcantly betterthan BYE svbjèt.con the- SE treatment -WSE subjectsscored higher than both Black groups. Black subjects with positive attitudestoward Black language performed significantly better on the BVE treatmentthan subjects with low Black language attitudes.

An analysis of subjects'performance on the Black and Neutral storycontent treatments showed that Black subjects with high attittides towardOteek ctilture performed significantly better than Black subjects with lowMed Cultural attitudes On the Black content treatment as hypothesized, odthey also performed significantly betteion till of the other treatnients.

. .

1;

An intc%riF..iiiiin t...1% t.iIlleii iiiit in drienione itit cfie,t of a! mutie onthe pet formance of the I1V stallevis I he investiparion chowird Ilxt Rs+niter, !itt: .. ::"C 1!!.xt ,inguageperformed differently from EVE subjects with low attitudes The highattitude BVE group performed significantla better than the low attitudeBVI poop on all except the Sl- treatment !his subgroup performed as wellss the total BSE and WSE groups on all treatments except on the neutralcontent treatment some white subjects perImmed aignificantly higher

-11" reculis of this !ouch, showed that ihe use of IIVI material% does notkid to minim ed si hoot performance by BVI speakinp children whileatu(odc toward Bid( 1. (awe and Mai. k Language is a ver) unpuitant factorin language learning for these children Eunher research is nteded, forcultural and language atutudes for Black children may interact withachievement in contexts other than those measused in this study.

TILE 11\1 E OE EN( REASED PAlU. I -( HIED VERBALINTERACtIQN ON HIE LANGUAGE EA(11.I tY OF PRE.-SellOOL INNElt-Cl FY BI.ACK CHILDREN Order No. t1008351

McQtrrrN. Arw r Rcit rwN, Pu fl Me Catholic University of Anterlea,1979 lftipp

BacAgrnund Black children front low socio economic families 7,7, a gioupwore below as ci age on ineasures of tribal competence at sc.hool entranceages, and each year the gap widens For many ,.ears morrey and time havebeen spent in attempting to solve their problems. with little or no resultsEducators feel that one reason for the dilemma is the children's lack ofverbal competence (I evenstein, 1969: Eorst and Kissinger. 1976. Muller1940). The major reason for this, as suggested by some researchers, is theabsence of positive verbal/interaction with adults in the early stages of life(Cazden 1975: 1 t4nstein. 1969)

Purpme The purpose of this studY was to examine the effects of varyinsdegrees of mother child language stimulation activities on three groups oflow-income Black children and their mothers.

Procedure The sample included three gioups of twenty Pre-school .

children each and the parents-of two or the groups The parents were all on .,

welfare, from the same neighborhood and their ages ranged from 14-19. Thechildren and their parents in Group I attended a school program daily. Theparents received a series of activities including instructions in ChildDevelopment, standard English and conversational skills Patrents along withtheir children received stimulating instructions including toy making andbook reading activities Vhddren in Group II attended pre-school Their

parents wsre given books to read to them at home and they.laped thef-readings and conversation about the books The children hi Group III alsoattended-the same preschool and their parents worked outside the school.The children and their parents received no treatment. The languagestimulation for the children and their mothers in Groups I and II continuedfor three months. The Peabody Picture Vocabulary standardized teak°designed to measure the individual's verbal intelligence by measuring hisreceptive vocabulary, was administered, before and after treatment, to all

designed by the researcher an administerenf establish the mean length of ,

the clnldren and Groups I II Parents. The Peabody Picture VocabularYtest required no spoken Ian e, thereforktheSultural Linguage test was

utterance and to measure the Subjects' use of standard English._ Meanlength of utterance and standard English measures werealso taken frontconversation during a play activity between The mother and her child.

Design and Analysis. Twelve kpotheses were tested utilizing:a one-weyanalysis of covariance. The hypotheses suggested that the children in GroupI would score significantly higher following the treatment on the Peabodyand the Cultural Language testS than children in,Group II and III, and thatchildren in Group H would score higher following the treatment, than

a children in Group III on the sante measures. In addition, children in GroupI would score higher following treatment than Group II and III childrenumeasured by informal taped conversations for standard English and MeanLength of Utterance. Group H will score highea, after treatment, thanGrout:011 on the sante measures. The parents in Gitup I will score higherafter treatment on the Peabody test and the standard English measure* Oenparents hi Group II. Chilren of the older m'othcrs *tau ld score higher aftertreatment than children of younger mothers on eachmeasure.

The tool for idemifYing group differences was analysis of covaiiance ofpoat test groups using pre-test scores as covariable. T-tests were used tocompare groups pre and post test,

Results. Children and mothers in Group I scored significantly higher on,t.

all test (p ( .001).Conclusion:The study's finding indicated children of mothers that

receive training speak more fluently and have a better command on English.

Nit.X1( AN.)4 Ill( AN() SO(. IN(.111,-111( IH 11AVH)14 ANDIAN(.I kt.1- 1111 It \ 1.51111) ',TA II"-,

Order No. 8004697MACIAS, RFYNAl IX) FLOM'S, PH I) Georgelown University. 1979 171pp

This study aescnbes the language resources ahditiec and. to comeextent, the language usc of a national (gilt& gioup ( lucanos I heliterature which forms the immerieue hack pround to this .tudy Can best hedesimbed as scarce and fragmented the data to desrolie on a hi oad nasis,the current language abilities and use of the national population in theUnitctl Sttç. has not been available until recently in 1975, the U SBureau of e Census undertook a nauonal sample survey that included 15l.njuqeybihty and use questions (The Survey of Income and Fducetion).It is this riala base which was explored in this study of the Mexican Originpopulation The analysis of this survey data was informed by the availabkliterature in linguisubs, Chicano Studies, education, ,1 l .aw, Itan_ Les( [Liesand analyzes (111Cario speech behavior and Its social context languagepolicy at *institutional level -public..servict fed political institutions andthe nature and Source or language "nights" wtre examined as well

lakanOS. and the other Spanish speaking in the United States. tireunique among non-English language groups. The number of Spanishspeakers has grown steadily thia-centiViielfc in indigenous group itrepresents the northernmost portion of over 200 million Spanish speakers inthe America& Chicanos art also overwhelmingly bilingual. native born, andyoung Institutional language discrimination is still very evident in hospitals,fire and police seAlcct, telephone services, administration of government,vctins, and the administration ofjustice. A recognition of the long histoncalpresenct of the Spanish language in North America, present United SteteAtreaty obligations, theliuman rigbts poncerns throughout the world, and thehettleLD nails of Mexicanos in the United States, all call for a reevaluation ofdornestk language "rights." their source. narure,.and enforcement

SEIZTCONCEPT ANb READING ACHIEVEMENT LN SCHOOLSOF VARYING F.:THNIC COMPOSITIONS Order No. 8000010

MANDINA, Genevieve Hogue, Ed.D. Fast Texas State Univer-ity, 1979. 121pp. Supervisor: Dr. M. D. Nelson

Purpose 9f the Study: The major purpose of this study wasto determine whether there were statistically significant &Her-mon in the self-concepts and reading achievement of blackseventh-grade students and the self-concepts and readingachievement of 'white seventh-grade students in schools withvarying percentages of biack enrollment. A .picondary purposewas to determine if Mere were statisticall significant differ-ences in the se1I-concepts and reading achievement of the malestudents and the self.- concepts and reading achievement of thefemale students. 0

Procedure: _The School Attitude Inventory piu.), which wandeveiloped by the author, was administered to five hundred stu-dents in three Northeast Texas middle or junior high schoolshaving black populations of 22 percent, 45 percent, and 92 per-cent. Reading achievement of the sample population wae deter-mined by scores.on standardized achlev meet tests. The datawere analyzed through independent t ests d Pearson r tests.

._.

1

Each hypothesis was rejected at the 05 levFindln a: The major findings qf is stu ere as follows:

. As measured by scores cal ill dardiz chlevementNets, the white students had I stall icaJly s lficantly higherreading achievement than did the bl ck etude

2_ As meaeured by scores on tandard ed achievementtests, the female students had a s ist,icall significantly higherreading achievement then did the ale ents.

3_ As measured by a self-me re of academic self-con-cept, the black students had it statts cally significantly higheracademic self-concept than did the white students.

4. As measured by a self -meastwe of academic self -con-cept, the female student(' had a statistically signIlicantly hIgherjieadannic self-concept than did the male students.4 5. As measured by scores on etandardlzed achievement

14

Ir.-I,. III, t.i.%,10. ,0,:d,,,,,f .. to . , h,,,.! ,.. th -, 1.1.i, 1, 1,put.iioni i;.4', ; , , ;i h L.; i ... %il..!, .111., .ici,i,. :,i I. ;;ilii, ! I I' 1.1ing

achievement than dui the black students in a school with a blackpopulation of 22 percent and the black students In a school witha black population (4'02 percent.

ft As indicated by scores on Nlandardired achievementlusts, thy white students in a school with A black population 01

:".' pi. t , crvi .m1,1 Ow white uhidcitir. In .% fictl.:.:1 with a 1,Iack pop. ulat ion of 45 percent had a statistically significantli higher

reeding achievement than did the white students in a school witha black population of 02 percent . .

7. AR indicated by scores on a self measure of academicself convept, the black studente in a school with a black popu-hition of 92 percent had a etatistrcally eignificantly htgher selfconcept than did the black fitlident ft III a Reilaill with n black pop-ulat ion of 5 percent

0 A ti Minimally signif want relationship botwevn academic. soli enik(el t and reading achievement existed for ail four

. groups- b tAcke, whites, males, and females.Conclusions: Dased upon the findings of this study, the fol-

lowing conclesTons seem warranted:1. The percentage of black classmates does not appear to

aifect the achievement of black student.n.2. It is possible that the achievement of whiteS is deprefised

in schoolS with a largo majority of black students.3, While this study did show the sell concepts of title white

students to be los,es.k...than the self-concepts of the black stu-dents, this research did not indicate that a white studears self-concept is affected by the number of black classmates.

4. Altho'ugh the mean self-concept of the black students inthe school with a large majority of black students was signifi-cant ly higher than that of the black students in the sehool witha black population of 45 percent, there was nO statistically sig-nificant difference in the self -concepts of the black students inthe large majority black school and the black students in thesmall minority black' school. Thus, no definitive conclusion.can be made regarding the effects the number of black class-mates has on black self-concept.

al

A SOCIOLINGUISTIC STUDY OF LANGUAGE USE ANDLANGUAGE PROFICIENCY IN SOUTH TEXAS

Order No. 8003351MEIEJUIOFFER-LONGORlit, LYNN VAOLX, ED.D. Texas A & I Unlversioo,1928, 528pp. Adviser: David L ZUfell

Purpose. The purpose of this study was to conduct a comparativeanalysis of the degree of bilingualism, as measured by relative proficiencyIncluse across four sociolinguistic domains, of selected Mexican Americanchildree in grades one, three, and five enrolled in bilingual educationprograms in selected districts in South Texas.

.S'ummary. In March and April of 1977, a sample of 90 MexicanAmerican children, thirty each in grades one, three, and five, were chfrom three school districts in South Texas. All the children were enrolled ina bilingual education program and were from a similar socio-economicbacks! ound.

The children were interviewed individually using two instruments. TheSpanish usage rating scale required the children to report on-their uie ofSpanish and English with vanOus bilingual interlocutors in four domains:(1) school, (2) church, (3) neighborhood, and (4)' home. In the wordnaming task, the children were asked to name all the objects they couldthink of which woul5I be found In each of the same four domains. A timelimit of one minute per domain was set. This instruhlent was administered-in Englishior the four domains and then in Spanish for the four domains.

Normative data,were obtained from the Mean weighted valuescomputed for the-domains total and subgroup responses on the Spanishusage rating scale. Shident's t-tsstS of significant mean difTerences Werecomputed. In addition, the Pearson product-moment coefficients ofcorrefatiOn were Calculated for the purpose of establishing a formula for'predicting total Spanfsh usage from the English word naming and Spanishword naming scores.

.41i

For the word naming task, the total number of words in 83(h languageand the total number Of words in each domain were determined Normativedam ware calculated for these word naming scores and t-tests of significantram differences were figured A ko A SpaniNh dornmarn e wa

ascertained for each child in each domain using the English and Spanish'word naming soores.

Conclusions Based upon the analyzed data, the following malorconclusions soem tenabk: (1) Robstown children of all ages reportedsignificantly more usage of Spanish, named more words in Spanish in alldomains, and obtained high Spanish dominance score!, than the subjects inthe other school distncts (2) Almost all children reponed using the mostSpanish with older interlocutors, slightly less with their peers, and the ICaStwith children younger than they. (3) The word naming scores indicated adevelopmental trend in acquisition of both languages. The older childrentended to name more words than those subjects in the two lower grades.(4) A comparison of the total word naming scores showed that, in general.more words werogiven in Enghsh than in Spanish (5) On the average, firstgrade children obtained higher Spanish dominance scores than their oldercounterparts (6) Thc Spanish usage and word naming scores correlatedfrequently. indicating that the Spanish usage scores could possibly bepredicted from the subject's word naming scores

STRATECIFS FOR At1 V1NG ONE.UPSMANSHIP: ADESCIUT lIVE ANALYSIS OF AFRO-AMERICAN SIBLINGS INtiVO SPEECH EVENTS Order No. 8012834

MUCHELL, JACQUELYN. ED.O. Harvard University, 1979. 374pp,

This thesis is a d riptive analysis of sibling speech. as manifested byThe aional i tegics sustained and treated by an Afro-Americanbryther ans si r The children's one-upsmnnship strategies were tape-recorded over a period of one year. Two naturally occurring speech eventswere analyzed The model of analysis has as philosophical underpinnings in

phenomcnological perspective which maintains that the meaning of anevent is not predetermined by the arbitrary assumptions of the researcher ol-by a theoretical model. The researcher is not pursuing the ultimate meaning01 a phenomenon but, rather, tnultipk meanings within arid across events

The descriptive portion of the analysis relie.s on the approach developedby Dell Hymes for analyzing speech. which he terms "the ethnography ofcommunication." It focuses primarily on seven comminents: addressee,addressor. channel, codes, setting, message form, and topic Integration ofthese features results in a close description of the.jnteractions.

Strategic issues in the interactional events analyzed involved themicropolitics of sibling rivalry. The domain of strategies the siblings use isexplored and examined, and an expanded view of sociolinguistic 0competence is discussed Selected aspects of children't sociolinguisticrepenories are analyzed-teasing, fighting; insisting. punning, mocking, andmaking up. The two speech events were both r&orded while the childrenwere cleaning the kitchen after being instructed to do so. TheirsOciolinguistic repertories are extlibited as they delay, begin, stray from, andfollow through their tasks. The mother/researcher structured the tasksduring the first event. Chores were not assigned in the second event,however, and Uie siblings had to negotiate tor the tasks they preferred_

Each speech event was segmented into phases, qr chunks of time, unifiedby comitron contextual properties. To extend the analysis, categories weresuperirnixised. on each phise, The categories emerged during theinteractions and were not predetermined. They are: "bickering." or phasesin which voices were loud, tempo was-accelerated, and tone was sarcastic;"ordinary talk," phases marked by periods of calm; and "siblingnegotiation," phases in which the siblings define self and other throughcontinual negotiation of status and role (this involves defining rights,privileges, and obligations). Calm interchanges were interspersed with angryoutbursts.

Etch 'phase type was compared within the events anl across the twoeyents to determine how linguistic features differ in style and function in asingle episode and livers period of one year. Significant differences werefound within the two events. The frequency and percentage of directiveswere significantly different within phase types and across events. Moredirectives were Issued during bickering phases in both events than duringother phases*. imperatives dominated' the second event, the siutrititnwhich

tiblivigs had to vie for preferable duties. Significant.differences were alsoAd in t ie frequencies of simultaneous talk and topic shifting within and

Wircoss events. Chi-sivare tests of significance revealed that deliberateftfierruptions in conversations were more _apt to otcur during the secondevent

Ihe li(1n\ xi. (-le examined in kcxcial toniext. .%pecchnia). ?nth cri:c7Fcll wixed 3 4110) Offunction\ (-wailed PI 'if mi-.0.,c,F men.speekii pia) icfkald clic :414111ig.s korikcpu. ol rat- nut) And 4c-1i Analysis ofthe prosodic fe.atures of the siblings utterances revealed that rhytfunappeared to be a central element of the childim's speech Certain types ofmessages possessed an accentuated-rhythmic regularity in Order tO inferdomination. cmphasue sArracin. or :Kt cnitiate Limit, Change.. In Cernpooften ix ion led xx.ht ii a ',owe! ,11111 %%as at hie% ed of atleinlqtd

e -t Ant h Oic (1(10, of :lbIlltr;,.!IStICS. thefamily. Anil ethit Afton

A slimy oF ENGLISH CONIPRFIIENSION SCORES OBTAINEDBY Mt XICAN AMIRICAN Ill .MF.N.I Ala SCHOOL CHILDREN

Order No. 8009123

1AcifECO, RICIIARD, PH D. University of Illinois of Urbana-Champaign,1979. 164pp.

This was an experimental study of English comprehension scoresobtained by fourth, fifth and sixth grade Mexican Amencan elementaryschool children For analybc purposes the students were separated into twogroups, those enrolled id bill gual classrooms and those enrolled intraditional all English ctuncu im classrooms. 'Ibi students enrolled in thebilingual classrooms were. furth r separated into two subgroups, thosestudents who had a high *vel o panish language (L1) proficiency andthose students who had 4 low deg et of Spanish language proficiency. Thedegree of Spanish language profici nes,. high or low, was determined by theLanguage Assessment Battery (LA11). The stu,sifms who scored at the 40thpercentile and above on the LAB wek selected as the high proficiencygroup. Although the non-Mexican Athcrican population was not studied,their scores were used as the norm for P.nglish comprehension for.companson purposes-

Analysis of the data, English compreiCension seOres, showed that therewere statistically significiInt differences in favor ofthe students enrolled inthe bilingual program. Within the bilingullgroup there were also significantdifferences in favor of the high Spanish cfroUeiencS, group. Arr MexicanAmerican groups, except the high proficiency'group, scored significantlylower than the non-Mexican American group. \

The findings indicate that a Mexican Amerioan child needs to reach athreshold level of 1..1 profiCiency in order for a "ceiling" effect not to occurin the acquisition of the English language. Research must be conducted todetermine whether there is a specific level of L1 proficiency a child mustreach before being introduced to the English curriculm.

THE EFFECTS OF TEACHER-ORIENTED AND' SrIDENT-ORIENTED STRATEGIES ON SELF-CONCEPT, ENGLISHLANGUAGE DEVELOPMENT AND SOCIAL STUDIESACHIEVEMENT OF FIFTH GRADE MEXICAN AMERICANSTUDENTS Order No, 7928338

PERALES, Alonso M. Ph.D. The University of Texas_at Aus-tin, 1079. 151pp. Supervlsor: Professor Jo Ann Cutler Sweeney

,.

This study comfrared the effects of teacher-oriented and student-oriented teaching strategies on the self-concept, Englishlanguage development and social studies achievenient of fifthgrade Mexican American students. The study population wascomnriaed of 104 fifth grade student9..at an Inner.eity elernan-

-etktary s ool in San Antonio, Texas, it'our classes were used inthe ex?ol iment which covered a period qf nlpe moths. Socialstudies lessons were eiveloped which focused on the enhance-mentiof the student's self-concept through student orientadclasstoom aetivities drawing upon the' student's cultural hack-ground. .The lessons required teaching strgtegies aimed atmaximizing the use of newly acquired Enlish language struc-tures for oral and written classroom co\ mmunication. The trait-

pt classes received the lessons. The control teachers usedThe lessons ae;reqtlired in the regular curriculum.

v

144

"rhree of l,tl tlyvothemek eilated Diet e welt, no rtivntfltAnt1,etween I gct'ot ami 4 4141i r cL4.....4,4 to Non' ci.

cept., English language development and eoclai .14 1114'1;t4

meet Th, vie! e It1 ,44 .00 :4' ,.! tTLanguage Indicator Test and the Metropolitan Ac hievement I tlut(oetal fitudiee) were mood to toot the hypothenen.

The Welt- denign used to evaluate the effectiveneos of thetreatment was the Pretest-Posttent Control group Derilsrn. Theanalyses of the ((tilt data Wt'I 0 cond.nyted toting (he (nmpotec 1)10gram Statistical Parkago foi the Sot IA Sctenc.Lti (SPSS) wringPearson (.orrelation and ANtS t iv A (Sub routinen) the loot)ability levels of All obtained F ratios were epor (ed and forthe purpose of rejecting null hypothenen, prohabilitien of 001were accepted au significant.

The reeulte Indicated that the claases receiving the treat-ment guieed compared to ttw control group in the over-all tentscoriae for ;ill three major scales. Caution ia recommended Ininterpreting the resulta of tho study eince nome euhtent ecoresfor both the Plers-Itarrls Self Concept Scale and the Shutt Pri-mal), Languago indicator Test showed no nignificant differencesbetween the control cinnnen and the clasnes receiving the treat-ment. With regard to correlations, the results of the Plera-Harris Seale suggest development of self-depreciatory self-concept In the control group. A revereal Is indicated for eachof the six nubteete.

The otudent-oriented lessons uned no treatment appeared tobe a nallent meann of enhancing the self-concept and developingEnglish language nkills of the 5th grade Mexican American stu-dean in the study.

V

AN ANALYSIS OF THE EFFECTS OF A BILINGUALCURRICULUM ON MONOLINGUAL SPANISH (MS) NINTHGRADERS AS COMPARED WITH MONOLINGUAL ENGLISH(ME) AND BILINGUAL (B) NINTH GRADERS WITH REGARD

TO LANGUAGE DEVELOPMENT, ATTITUDE TOWARDSCHOOL AND ALF CONCEPT Order No. 800375?

PWEV/IXT DA7.. JOSEPH ORLANDO, PH.D. The a\niversity of Connecticut.

1979. 81pp.

The purpose of this study was to evaluate whether a bilingualcurriculum affects monolingual Spanish (MS) students as compared tomonolingual English (ME) and Bilingual (131) students exposed to amainstream curriculum with regard to language development, self conceptand tOtude toward school This research used OQC treatment group (MS),

and two comparison groups 04E and BO. A limitation which largely affectsthis study is that it is not an eiperiment and the findings arc not universallyapplicable. The treatment was the bilingual curriculum.

A group of ninth grade students of a high school in Hartford.Connecticut was divided into three groups (MS. ME, and B1) andadministered the following instruments: Comprehensive Test of BasicSkills, Test 1 (Spelling) and Test 3 (Reading Vocabulary) (CBT/McGraw-Hill, 1973); Iriteramerican Test of Reading, Level 4 in Spanish (GTA.Manuel. 1967); Self Esteem Inventory (Coopersmith. 1967); Self Appraisal IInventory (Frith and Nakimura; 1972): and School Sentitnent Inventory(Frith and Nakimura, 1972) as a pre and post test at the beginning and endof a semester (17 weeks). The final pre-post test sample was (N =138).

The change between pre- and post test administration was assessed byng an analysis of covariance for each dependent measure. The pretest f

e the covariatt and the respective post te'st the dependent variable.pbell and Stanley, 1963)..e'resultS of the analytit of Covariance (ANOCOVA) did not produce a

signi t difference in the dependent Measure Reading CTBS and SpellingOBS A significant difference was recorded in the dependent measuresReadi (Interamerican) (Ff± 11_129; p < .002) and Vocabulary(lute erican) (F= 16.443; p < .001). The SEI (Fst 21.447; p < .001) andthe SA I F2.3.367; p < .038) yielded a significant difference between thegroups -self concept, while the SS1 (F=3.255; p .< ,042) was statisticallysignifi t

The need for further reseal% in the field of secondary bilingualeducation, as well as the need for replicaton of the study with otherpepuladons is discussed. The Importance of longitudirial inidies indetermining the effects of secatdary school bilingual education On ,,addeseents is alsojotcd.

VARIA I ltiN Nil lt II II\ 1 bill luLl t 11A kliCitilei Nu 800.1289

IPA;.( 1., ,, '.;

The purpose Of this study has been to char acter ire some of the ways inwhlch children respond to and use systematic differences in speech TheSettog is a single neighborhood in I !await, a speech comniunity in the latestages of decreoliration Data were gathered from foorteen chtidren agedfive through rwelvC. of different ethnic arid so locconomic backgrounds...while they were engaged in spontaneon e. con, e! ;it `riit'i fit hi ether Invarious settings. activities . and self selected groupings

!systematic speech variation was described 1 rom several perspectives.First, subjects were ranked by relative frequencies of nine ilawaiian Englishkxicogrammauail variants in their speech- A Hawaiian English-GeneralEnglish (HE-GE) continuum was found. paralleIhng ethno-sociocconornicgroupings Rank correlation values were highly significant It appears thatthe young subjects have strong grounds for associating recurring differencesin speech with particular persons arid social gi carps At the same nine,ranking inconsistencies among the eight midscale speakers stiguest that eachchild may he more usefully described as Operating within certain ranges offrequencies for ust of different variants, than as 'has Mg one or severalcodes or lects

Second, contingency tables for relative-frequency distributions of HEand GE variants vs. contextual features showed high correlauon betweenvariation and both linguistic and macrosociological features of context, butlow correlation between variation and rnicrosaaological features Similarinvestig.auon of structural variants of directives. instructions, andevaluations by three subjects of different backgrounds showed highcorrelation between variation and microsociological features but low..correlation between variation and macrosociological features In short, whilesubjects of different social backgrounds differed significantly in terms of anumber of phonological, lexicogrammatical and intonational features, theydid not differ significantly in terms of choices of stnictures to accornplisbcertain social purposes.

Finally, style-shifting or code-shifting between I IE and GE by threesubjects of different backgrounds were examined, using a combination ofdiscursive microanalysis and a linear-diagram technique on severalconversauons in which the subjects participated. Thcre appear to be twomajor sitylistic strategiesaccommodation' ro the speech of the interlocutorover a whole conversation, and marking bf brief passages of talk foraffective, aesthetic, ahd discourseistructuring purposeswith one or theother strategy consistently preferred by each child.

The linear diagram technique,.anadiptatidn of the continuous-recording methods Used tosecord lift signs, atmospheric changes, andvolcanic activity, appears to be an effective way of handling many of thespecial characteristics of coder or style-shifting, Shifting between HE andGE fluctuates in intensity (number of }1F or GE variants per clause),duration (whether a shift is maintained tlithln or acrosS clauses. within Oracross speaker turns at talk), and choice f variants. It is IS =Kb COnHattiveas substantive: contrast between the sp ker'simtnediately precedingspeech and her current speech, betweetj her speech In most conversationswith her speech in the current conversation, between her speectrand that ofher interlocutor. Shifting involves a convergence of a ntimber of observablephenomena, conversational and linguistic, so that discussion.of shifting andits possible motivations requires methods sensitive to many factors and tocontrasts that develop ever time, both withal- conversations and betweenconversations.

In sum, this description of variation in speelli by children inHawall hasemployed quantitative and microanalytic techniques, and is based onlinguistic, social, and parbcular facts about the speech and speakers understudy.

COGNiTIVE DEVELOPMENT OF LANGUAGE, READING &MATHEMATICS IN A BILINGUAL/BICULTURAL PROGRAM

Order No. 8003352SAENZ, ERASMO ANDRES, ED.D Tesas A cf I University, 1978. 84pp.Adviser: David L. Zufelt

Purpose of the Study. The purpose of this study was to assess inter-groupgtowth in the cognitive development of language, reading and mathematicsutilizing two basic groups Of children participating in the BishopConsolidated Independent School Districes`bilinguallbicultural educationprogram for a period of five consecutive years

Summary thlingual/hicultutal education ptogiant doer:toll are 'Nuttedto annually evaluate their invtrwtional progtams I ongumtinal anal\ sic ofcognitive growth in bilingual/bicultural education prograrnC ran pro. ide a .model tot analycis that enhances the credibility of diem' plogiatie, I

Study Involved the selection of a bilingual/bieultwal program in a Texaspublic school that had been implemented and maintained prior tomandatory bilingual/bicultural education

Tbe Bislup (ISD program originated during the 1971-71 school yearwhen the chstnct applied fot and was funded thi ()ugh 1 ale \Ill source%Fneurang on the cognmve development of-language. rending andmathematics in a bilingual/bicultural environment. enabled this wnier toextend the study for five consecutive years. Yeai to yeal data genciatedkim the verbal, non-verbal, numerical and total subtests of the InterAmerican rests of Geperel Ability were collected, computed and analyzedfor the five yeas penod. Complementary data were derived from thevocabulary and comprehension components of the Gates MacGinifieReading Thsis. Additional data were drawn from the reading aniimathematic subtests of the Cal(fof nu) Achievement Testi The data fromInter- Amencon Tests of General Ability ser ved as the basic foi the stuth

Conclusions The following conclusions can he drawn (loin the finalanalysis and applied to cognitive development in selected Content areas(1 ) The Bishop Consolidated Independent School District continues to

oally assess inter-group growth of all children in the bilingual/biculturale t tion program (2) The educational alternative specifically designed tomeet the unique individual instructional needs of limited English speakingability (USA) chiidren appears to be feasible (3) The bilingual/biculturalprogram children were able to maintain comparable or grade levelachievement with children in thejegular school program. (4) Continuedsystematic data collection and Subsequent analyses ShOuld aSSistbilingual/bicultural program directors to better understand theinterrelatedness of English and Spanish when used as mediums ofinstruction. (5) The cognitive development of language, reading andmathematics Can be qualitatively measured when the tools of analysis are4dequately applied_

SOCIOLINGUISTIC ASPECTS OF ENGLISH DIVERSITYAMONG ELEMETARY AGED STUDENTS FROM LAGUNA

).PUEBLO Order No. 8000704

STOUT, Steven Owen, Ph.D. he American Unlversity, 1079.217pp.k

This dissertation documents the sociolinguistic patterningof English usage among fourth, fifth, and sixth grade ArforIcanIndian students from_LAuna Pueblo, Laguna,-New ;c1co-a, Theoverall nature of Enetsh usage among these spea repre-.

sents a very standard range of English expressictomparedto the expectations developed in the literature for varieties ofAmerican Indian English.

Three linguifitic variables are examined:

Neittive constrirtionSu ect-verb concordDistributive be

Negative construction generally reflects principles of vari-ability identified in other vernacular speech communitiee inthe United States, Distributive be is argued not to be an as-pect of the vernacular English grammar at Laguna Pueblo.Subject-verb concord, like negative construction, reflects-vernacular usage,. However, the patterning of vernacular usagedoes not parallel the patterns found in other vernacular speecheemillunities studied to date,

The structural contrasts between Keres and English gram-'. matloal detaile,which could give rise, respectively, to these

variables are explored. These contrasts, unlike the situationfound In other vernacular English codes, do not directly appearto condition the occurrence of the variables as found in LagunaEnglish aentences. Reference to a larger configuration of'eausal factors" seems more appropriate. In fact, two differ-ent configurations of socially based variants are Identified inthe analysis and then used to predict the occurrence of ensuing;taisdard vs. nonstandard usage patterns

Irbe most stigmatized English usage patterlis appear tohart Important eiinsequeneett at Laguna Pueblo. Students with

Inrithnce ni,14.tit,try Icnd iii Ci tittli,-, k.ii cs. Ltilotagv, tI evidence

Ket eri Lt411115 t P 3"45.ti'dubagc ot .titandar d hirgnalt vontitruct h mit MU /NV grunter atisocla -Mil and control with Keret; language usage. Of particular in-terefil In the fact that exper lenr e off the reservation may DOtbe an productive to ntandard English development as educatorsoften claim, off i eservation i esidence, in fact, appears to In-crease the likelihood of stigmatized English tinage patternswitido thc LagunA speech cimicimnity Conversely, traditionallyburied 00tpertenre1110 reanmi the probAbility of less stigmatizeduttage. The central pueblo value of prescriptivism or con-servatism and the acculturational strategy of compartmental-ization arc cited as partial explanations for the paracular pat-torning social (Retort' bring to details of lingutstic usage.

TIIE INFLUENCE OF SPEECH VARIETY ON TEACHERS'EVALUATION OF REUIING COMPREHENSION DURCG ANORAL READING AND RCALL TASK Order No. 8007518

TAV1 OR. JANET MAIM' Botts, Pit D The Florida State (Jri:versify. 1979.15Spp Major Professor: Nihcy l Douglas

The reading failure of a high proportion of black students is a majoreducational concern. Although a number of causes have been suggested andtested, no conclusive findings pave yet emerged. Sociolinguistic theArysuggests tbat political and cultural differences between teachers andstudents create conflict in the classroom. This conflict is thought to be amajor cause of the reading failure of man). black students. An experimentalstudy v, as conducted to examine the influence of sociolinguistic (actors onteachers' evaluations of reading comprehension and on the expected nsthey form concerning the capabilities of the reader for future reading'achievement. These expectations include the teachers' perceptions ofstudent charactenstics suth as confidence and eagerness. student beMviorsaich as work habits and indtistry, and student aptitude for continuedprogr'ess and future achievement. Th e. purpose of the study WAS tOinvestigate how tWo difl'erent splefh varieties, standard EDglish Ind BlackEnglish, used during an oral reidfrigrd recall task influenced the teathers'ec aluations of reading comprehensionind the expettancies formed, andhow teacher attitude toward Black Eoglish related to those.evahlaiions andexpectancies.

Iwo audio-tapes of actual student performance were used as thetreatment materials. The experimental tape presented the oral roading andrecall performance of a Black English speaker and the control tapepresented the oral-reading and recall performance of a gandard Englishreader Both 'readers were male, and the tapes were judged functionallyequivalent in reading comprehension.

Sex enty-two teacheas volunteered to participate in the studyAnd were; randomly assigned to evaluate one of the two treatment tapes. They

recorded their evaluations and expectancies on two Likert-type scalesgenerated for the study.After the evaluation, the teachers were asked torespond to an attitude scale designed to measure attitude toward BlackEnglish.

The dhti were analyzed-using a multiple linear regression aptitude-treatment interaction approach. This analysis yielded a significantinteraction (p < .05) between the speech variety used and the teacherattitude toward Black English, for the evaluation of reading comprehensionduring the oral reading and recall task. Significant contrasts between the.

)evaluati of the two tapes were found for teachers who have negative'luta es toward Black English. The tape of the Black English speaker waseval ated significantly lower than-that of the equivalent standard English'speaker, There '.'ras no significant contrast between the two tapes forteachers with a neinral or Positive attitude toward Black English. Althoughthis finding supportS Stociolinguistic theory, the results should be viewed.with caution. The small sample size used in the study did not alloW to. aprecise estimate Of the population parameters. No significance was found inthe model used to examine the influence of speech variety on teacherexpectations. The results indickte that young Black English speakers arecited lower in reading comprehension than equivalent standard Englishspeakers when their teicher holds a negative attitude toward Black English.Since reading is a daily instructional activity, the compounding effect oflower evaluations may not only Impede the actual reading progress of theBlack English speaker, but also may contribute to a lowering of the learner'S-self-esteem.

41,

AN ANALYSIS 01. LIILNIC t...11ARAt,,ILRISIIGS ANDATTRIBUTIONS ,OF SUCCTSS AND FAILURE ON ATI [LUDO'TOWARD READING AND A(11IEN'EMFAT MO.nvAlioN

Order No. 8004397

\'itat YvoNNu 1 . Pir D ih tIniversm of A lolu)r-no. 1979

10fipp. Majut l'ioteSNOI l it hat d P W illiams

The purpose of this study was to identify whether ethnicity, attributionsmade in a successful reading situation, and attributions made in a failurereading situation affect attitudes toward reading and achievementmotivation_ Eig,hty.-fiverthird grade males from six elementary schools, nroIn Oklahoma City knd four in Lawton, Oklahoma. comprised the

ulation for this stlicly. The ohnic.distribution of students was fortyack students, thirty White students, and fifteen Chicano students

The Neathington Auitude Scale was used to obtain students' attitudestoward organized reading in the classroom and reading in general Studentsread passages from the Sland.ard Reading Inventory to obtain theirindependent and faistration reading level. The independent reading levelrepresented 1 flICCC.SS reading situation, and the frustrati4n reading levelrepresented a failure reading situation. In each of the two reading situatioStudents attributed their success and failure in reading to either ability oeffort Teachers identified their students as either high or low inachievement motivation

Twelve hypotheses were forMulated for this study. A 3 x 2 x 2 fael'orialanalysis of variance and a multivariate analysis of variance was employed totest the hypotheses. An analysis of the data revealed no significanymaineffect for ethnicity, attributions made in a successfUr reading siwation, andattributions made in SI failure reading situation. Analysts also rrealed noSignificant interaction effect.

I 4 I

7

COITESION LN SPOKEN DISCOURSE: A PRELINDNARYANALY,SIS OF TITE SPEECH OF LOWER-INCOME BLACKDRUG ABUSERS Order No. 80036t9YOUNG, JEAN SKORONSKI, Pu D.- Illinoislnstitute of Taljnblogy, 1979.29dpp. Adviser: Dr. Mackie Blanton . .

The purpose of this study was to determine if there was or was not acohesion pattern in the speech of lower-income Black drug abusers. To adialectal outsider, their speech seems to-ke disjunctive and, conscquentby.incoherent The procedure used to analyze thik" pr2blem was that proposedby M A. K. Halliday and RunaiyA Hasa') in Cohesion in Englishi- in theiranalysis of written work_ Since all language is linear and systemic. Halliday'and Ilasan's cohesion tic analysis system was applied to spoken discourse.The results of this study showed that the language of loaierincome-Blackdrug abusers is coherent and has its own distinctive pattern that differsdecisively from that of thc general dialect_ In conclusion, Black languagestyle was foonditrbe as significant as lansuage content in cohesion analysiS.Four jcategories of styles of speech was found for lower-income Black,chug llsers. ii

4

iNif.A.K.lialliday and Ruqaiya Ilasan, Cohesion in English. LongmanGroup Limited, London, England, 1976.

:184

*.

4.

Copies of the dlssertations may be obtained ly addressingyour request to:

University Microfilms International300 North Zeeb RoadAnn Arbor, Michigan 48106

or by telephoning (toll-free) 1-800-5h-3042

'1

4

19