Citation
Construct validation of an instrument to measure teacher attitude toward the use of international content in the K-12 social studies curriculum

Material Information

Title:
Construct validation of an instrument to measure teacher attitude toward the use of international content in the K-12 social studies curriculum
Creator:
Erb, Thomas Owen, 1945-
Copyright Date:
1977
Language:
English
Physical Description:
xi, 106 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
African studies ( jstor )
Art teachers ( jstor )
Construct validity ( jstor )
Educational research ( jstor )
International education ( jstor )
Ions ( jstor )
Political attitudes ( jstor )
Students ( jstor )
Teacher attitudes ( jstor )
Teachers ( jstor )
Curriculum and Instruction thesis Ph. D
Dissertations, Academic -- Curriculum and Instruction -- UF
Social sciences -- Study and teaching ( lcsh )
Teachers -- Attitudes -- Testing ( lcsh )
City of Tampa ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Bibliography: leaves 76-86.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Thomas Owen Erb.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
026311260 ( AlephBibNum )
04055265 ( OCLC )
AAX3769 ( NOTIS )

Downloads

This item has the following downloads:


Full Text












CONSTRUCT VALIDATION OF AN INS'TRUMENE T TOl MEIASURFI TEAlCHER AT~lT'ITUDIE
TOWARD) THE IISE OF: T'.I~NTERNATION CO)N'TENT IN TH~E K-12
SOCIAL. STUDIES CUIRRICULUMI[E









By5

THOMPAS OWIENJ EKIl


Ai DISSERT'ATION PRE\SENTED:J TO T~HE3 G:RADUATE~ COU1NCT. of:
TTHE UNIVERSITY OF FtLRIDA
IN P'ARTIAL FUI~'LF. IILLEN OF1 TliF REQU'IIREMIEN1S FR( THE
DEGREE O)F OUCT~OR OF PHIILOSOPHY













UNIVERSITY OF FLORIDA



































Copyright 1977

by

Thomas Owen Erb








































To Karonl, Christopher, aind Grtapo~ry















ACKNOWTLEDGMENTS


Special thanks are due to my chairperson Dr. Arthur J. Lewis and my

cochairperson Dr. Linda MI. Crocker for the great support they have offered

throughout this project. Thanks are due also to D~r. Cordon D. Lawrence

and Dr. Rene Lemarchand for their editorial assistance in preparing the

final manuscript.

Special appreciation is due the following people in international

and African studies who have supported this project: Drs. Haig Der-

Hloussikian and R. Hunt Davis at the University of Florida, Dr. Abdelwahab

Hechiche at the University of South Florida, and Dr. Wilfred Owen at the

University of Illinois.

I m indebted to 31 other professors at thle University of Florida,

both in the College of Education and the Center for African Studies, who

assisted this project either by serving on a panel of experts or by

allowing me class time to gain access to the 166 students who participated

as subjects. In the latter regard, I wish to thank Dr. Gus Jimenez for

supporting this research by arranging for 65 Hlillsborough County teachers

to participate in the pilot phase of the project.

Soime assistance oin computer programming was received from the Center

for InsLructi o)na l and ReSeaRTch Computling AZctivities at the University of

Florich~. Compu~lting) wasll rlrlone at l i; ingf thle facilities of the Northeast

Regiondi Data Cenlter of thle State Uiniversity System of Florida located on

the enmpuls of thle Unliversity ,F F~lorida in Gainesville.
















TABLE OF CONTENTS


iv

vii

viii

ix


ACKNOWLEDGMENTS. ...... .

LIST OF TABLES ......

LIST OF; FICURES. .......

ABSTRACT .. .

CHAPTER

ONE RESEARCH PROBLEM.

The Problem .
Need .
Definitions
Construct Validation.
Hypotheses.
Significance.

TWOr REVIEW OF LITERATURE.

International Education .
Teacher Attjitudes And Their
Summary of Research

THREEI METHOD.

Attitude Scale Construction
Procedures for Establj shing

FOURX RESULTS.

Reliabilities.
Sample D~istribultions.


Measurement







Reliability


Validity.


Tests of Construct Validity

F:IVE- CONCLUISIONIS AZND RECOMME1I[NDATI'ONS

Introduce lo rn.
Reliability Comlparisoans.
Discussio-n of FTCAS Validlity.


Mulltiple Asplects of U~nidimensional Scales.
A~dequacy of INTTrOT anld E0RTOT as Attitude Scales..












Further Research . .. .. . . .. .. .. 69
Applications ... .. . . ... .. 73
Conclusion .. .. .. .. .. .. ... .. 75


REFERENCES. . .. .. . . .. ... 76


APPENDICES


A INSTRUMENT USED TO CATEGORIZE 90 OBJECTIVES AS NATIONAL-
TSTIC, INTERNATIONAL,0R( ETT~HER/OR, INCLUDING, THE TALLIY
OF RESULTS . . .. .. . . .. 87


B FAKS ITEM ANALYSIS ... .. .. .. .. .. .. . .. 93


C FLORIDA INTERNATIONAL CURRICULUM ASSESSMENT SCALES
SHOWING SUBSCALE STRUCTURE . .. .. .. .. 94


D FLORIDA AFRICAN KNOWLEDGE SCAL. . . .. .. .. .. 98


E WORLDMINDEDNESS SCALE. . .. .. .... .. .. .. 103


BIOGRAPHICAL SKETCH . .. . . .. .. . 106
















LTST OF TABLES


Table Pg

1 Response Distribution for Open-Ended Questionnaire on
African Content in the Curriculum. . .. .. . .. 26

2 Reliabilities and Standard F~rrors for the Florida Inter-
national Curriculum Assessment Scales and the Florida
A~frican Knowledg~e Scale based on the Pilot Sample
(N.= 131). . . .. ... .. . .. .. 48

3 Raw Means and Standard Deviations for Hypotliesis One
(N = 33) ... .. ... .... .. . 52

4 Raw Means and Standard Deviations of All Samples and
Variables Entered into the Analyses of Hypotheses Two
Through Five ... .. .. .. ... .. .. .. .. 54

5 Inltercorrelations of Cognitive and Affective Subscales
on FICAS .. .... . .. . . . 66

6 lhP~ Biserial Correlations Between the 30 Items Selected
for Inclusion on the FAKS and the FAKS Total Score .. .. 93





LIST OF FIGURES


Figure Pg

1 Subscale Structure of the Florrida International Curriculum
Assessment Scales. .. . . . . 30

2 Item Assiginment to Subscales on the Floridn International
Curriculum Assessment Scales .. .. .. .. .. . . 31

3 Correlation and Item Mean Difference Patterns for NATTOT,
EORTOT, and INTTOTT .. .. .. .. . 67


4 Hypothesized Relationship Between Knowledge of Africa and
Attitude Toward African Content. .. .. .. ... . 70


V111





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


CONSTRUCT VALIDATION OF AN INSTRUMENT TO MEASURE TEACHER ATTITUDE
TOWARD THE USE OF INTERNATlIONAL CONTENT IN THE K-12 SOCIAL,
STUDIES CURRICULUM

By

Thomas Owen Erb

December 1977

Chairperson: Arthur J. Lewis
Cochairperson: Linda M. Crocker
Major Department: Curriculum and Instruction

Over the past decade educational researchers have established the

importance of teacher attitudes for the success of educational programs.

However, little wans known specifically about teacher attitudes toward

international content in the curriculum. Consequently, it was necessary

to learn more about these attitudes. Since no reliable or valid scales

existed to measure such attitudes, this study sought to develop and

validate the needed instrumentation, which has b~een titled the Florida

International Curriculum Assessment Scales (FICAS).

Using Cronbach's alpha as the measure of internal consistency, this

study employed three established techniques employed for testing construct

validity: Correlates, group differences, and study of change over

occasions. Three subscales of thle FLICAS were tested for validity.

After da:ta were collected fromn 131 experienced teachers, scores on the

three FIrCAS subsea.lle were rorrelated with the Worldmindedness Scale

(WJ-Scale) total score and the Floridan African Knowledge Scale (FAKS) total

score. After this procedur-e was completed, the scores on the FICAS sub-

scales of these 13L teachers who represented a cross section of the








general teacher poplarl~tion were compared to those of 49 teachers from the

Southiiast and Midwest who elected to participate in summer institutes on

African1 studies. Finally, the pretest scores of these institute par-

ticipants were compared to their posttest scores to see if the subscales

of interest could detect change in attitude as a result of institute

parti:i pat ion.

Thec Florida International Cuirriculum Assessment Scales proved to

have high reliability and adequate validity to justify their use in fur-

ther studies. Estimates of coefficient alpha ranged from .87 to .93 for

the three subscales under study. Both the International (INTTOT) and

Either/or (EORTOT) subscales showed the expected correlation with the

W-Scale. The Nationalistic (NATTOT) subscale had a nonsignificant cor-

relation with the WJ-Scale. None of the subscales produced the predicted

correlatlion with th~e FAKS. Al~l three subscales were successful in

measulrtug a difference between the cross section of teachers and those

self-selected for institute participation. The results were inconclusive

from th~e study of change over ocensions. A multivariate F test sug-

gested that the subscales did not detect a significant change over

occasions. However, EORTOT's significant univariate F test suggested

that further validation work needed to be done on EORTOT.

Icenuse of their positive correlations with the W-Scale and because

of their ability to distinguish between groups which should be different

on the attribute being measured, both INTTIOT and EORTOT are recommended

for fu~thlEr researchl. Validated for African conitent, these two subscales

need to be validal-ted for mother international studies areas. Research to

establish their relationships, to other constructs such as beliefs systems

or level of cognit ive moral development could prove fruitful. They also








have potential for basic research into the formation of attitudes as

well as into the consequences of holding certain attitudes. These scales

may also prove to be useful devaluation devices in either pre or inservice

teacher education where an internationalization of perspectives is an

intended result of instruction. Finally, the social studies objectives

comprising the items on these scales could be used to help teachers

develop teaching strategiies for using international content.
















CHAPTER ONE
RESEARCH PROHI.l?



TIhe Prob~lem


Ignorane of the world beyondr our bordecrs is a Insuiry that Americans

cannot afford. Schools must play a positive role in inicreasingi knowledge~

of the non-Western world while working: to diminish Pthlnocentric attituders

toward these areas. Hlowever-, neither preservice Inor inservice teacher

education gives muich, if any, emphansis to drevelopingf thle k~nowledge and

attitudes consistent with techcling Fro~m a cross-raltln t rn1 r global per-

spective. Recoginizinji thiis deficiency the Uinitedl Statns Olfficer of

Ellucation has funded 80 14ralnguae and arena studnies Icnters (including

oight African studies Ccnters) at ulniversities thrroulghout~ the Counltry

(Easton, 19)77). As a condition ofE findings, thofsr r ntrrs are to engagee

in outreach work aimedl at cr-eatingi greater inlternlt ional uinderstanding

among Amefrican public school pupils, teachers, andr the community at

large. These funded centers have fuill disr~rttlon in developingi oultrenchh

programs, and several of them, including: those at the Ulniversity of

Florida and the University of Illiniois, have chos~en to emphasize pre-

parinfi public school leacherss to teach African co~ntent to their stuldrnts.

Have these ouitreach e~ffo~rts bee~n effective in al teriing teacher attituldes

toward the uisefulness of African content to meet arrepted socin1 Ltudics

objectives? Prior to thiis study there hans beren no valid meacnls to assess

RIuch channel inl teCnher attitudes.PR










Consequntlyy this studyl was desiginedl to devempil .l relia~le andi valir

method of assesising teachers' attituides toward frrironr conltenlt in the

curriculum. The resulting instrument was intended~i to be- onel that coulld

be used to measure change~ in teacher attituldes as a1 resullt of preservice

or inservice experiences designeldl to broadecn tearchrs' perceptions of th~e

relevance of African conte~nt. In addition, the instrulment was intended

to be useful for more basic researchi on attitudefs, their forma~tion, a7nd

conseq ueic es .

This studry wans confinedl to validanting on instrumento specificallyp for-

African content rat-her- than international conntent IgeneraivniI. Thlis decision

was made because! the concepF~t Of international cntent was rather vagueC

since it subsu~med many geograp'hical areas or1 divigio~n of thei worilld nhbout

which individual teachers wrre likely to have d ivois oEIpininns: LaJt in

Amierica, Europef, Canada, Asin, th~e nonl-Western world., theI I[hird Worrld.

the Communist world, andl so for~th. For g~reater c(lcertual clarity it

seemed necessary to define a1 spccific intecrna~tionnl region for- this

validation study. Africa was lchosenl bercause of thel rcqearcher's fam~ili-

arity withi programs almed at improving~ Afric~an studies in ther public

schools through teacher inservice edulcation. Howerver, any o~ther- inter-

national region might have served equally well.




Need



As never before Americals need to have an unldt.rstandj~i ngf huiman

experience that transcends I:rochial perspectlives. Dentsions male~ in

centers orf power in or ne rea of the world canir effect peop"le` onr all con-

tinents. Edwin 0. Reischauorr (1973) states thle case for international

edurcation in this way:










Before long, humalnity will face malny diffirtllties thant
can only be solvedl on a iglobal. Rr:le~. ForI this therE
must he a muich h irgher deglree of uinderst ndtinc arlnd afar
greater capacity forr cooperation beltweenl di p;larate
peoples and natjions than exist nowl. dctnhwvr
as it is presently conducted in this country -- and in
every other coulntryy in toe w'ord, for that matter -- is
not moving rapiidly enloulgh inl the right direction to
produce the knowledge about thle oultside world1~ andl the
attitudes toward other peoples that may b~e essential for
humann survival wi thin a1 generationl or tw. Trh is, I
feet, is a much Igrenter international prob~lerr thain thle
military balance o~f power that absorbs so, murch of our
attention to~day. (p. 4)


Americans, who~ inh~abit thle mnost powerful national entity, cannot afford

to remain ignorant of conrd itions elsewhere in the wolrldl. Afr-icai. Asia,

and Latin America have a pLarce in our lives beyondl a mere interest in the

exotic.

In agreeiing with the need for better internati~ionn studies. educaltors

have recognized the need for bothl cnrriculalr reform (no~w. 1976) and

positive tenclerr attitudles (Billinlgs, 1971; Rich, 197h). Yet so little

is done in collegefs and tecnrher training institultions toi prepare trachprs

to adequately handle interrlntional studies in their teaching (A11plgren &

Gustafsson, 1974; Bidwell., 1964; Carver, cited in Klassen & Moore, 1968;

Intercultuiral Understandityg, 1972; Phillips, 1963; Taylor, 1968/1969;

Wise, 1975a, 1975b). Often those teachers who do nttempyt to handle

international topics leave their- Stuldents with more nlegative andr stereo-

typed images than they hadt when they bfegn their study (Almaron &

Gustafsson, 1374; Beyer & Hlicksi, 1968; Eicher, 1975). However, where

teachers were well prepared and held positive attit~des about inter-

national edulcation, students' knowledge and attitudes~ imrproved as w~ll

(Arlmgren & Gustafsson, 19741; Frech, 1973, 1975; Klnsson & Mloor-e, 1968).










Because of inadequate preservice~ training, inse~rvice colmponents are

needed to help Leachers gajin the competencies necessary to trnchl shoiit

foreign cultures. Outreach p~rograms such al~s those offered byr ther Arican

Studies Centers at the Universities of Florida, 11linois, and Wisennsin

giive teachers the opportunity to participate in workshops and institutes

aimed at increasingg their substantive knowlrlede of Afrr-ica ndM helpinG

them expand their perceptions~ of the va~luef il ntelerating Africann content

into the social studies curriculum. Teachers are exposed to films,

lectures, and discussions on African topics: p~artic ipate in modifiedl

African cultuiral experiences; meet African sc~holars~ ilnd students; and

are introduced to curricularn manterialss and teachiing strategies which aIre

designed to improve their te~ach~ing of Africa.

Natulrally, the people whino aminiister African s~tudiies oultr-ench pro-

grams are inrterested in teachers' knowledge rrf, andl attjitudes toward,

African content in the curriculum. A seairch of thet litrrnture revealed

no studlies of teachers' Attitudes towardl teac-hing about AfriCal. A pr.iva~ntP

communication with Prof. Mnr-i~n .[. Rice oF thle Anlthropololn! Curricullum

Project at the Ulniversity of Georgia has confirmedl that, aside from

content-specific instruments of questionable validity, there was no

reliable. valid measure of teachers' attitudes toward Africa in the

curriculurm. At the University oE Lllinois workshop, inl 1974, trainsactionali

analysis was used to evaluate the interaction amlong: worksholp participants

and the attitudes of the wor-kshonp participants torwnrri thle workishop format

and its suibject matter content (Schmidt, 1917',). Ilrowever, thec instro-

mentation failed to produce danta thalt wouilld nllowlr an ojectivp evaluation

of the teacher attitudes involved. In fact, the teacher attituldes

studied in thle Illinols project were not directly related to the curricular










issue of using African material in the classroom, bult focusedr on thp

internal workings of the workshop i tse~l 'Thus there was a needl to

design, develop, and vallidate a1 reliable attitude s:cale thait would be

suitable for assessinf the formation of attitudes, the influence of

attitudeFs, and change in attitudes toward inlcluding African material in

the curriculum. Withonut suc-h instrumenntati n no researrch or hiecjtive

evaluation could Lake place to assess the impact of inse~rvice education

on teacher attitudes.




Deff i ni tions


Attitude



Attitude is the psychological constructl wh~iir refers to the comp~osite

of all feelings about, and ~redispostitins for bhavhi~inr r towaird, somne

object. Although Attituldes are covert, they findl rvert expression in

the form of verbalized o~pinio~ns thu~s making: them meiasurrable. Though

based on coginitive processes, attitudes are primairily affective in natur-e.

As measured by attitude scale~s, an attitude is rep~resenrted by a score

which corresponds to the point along, an ulnderlying continualn which der-

Fines the degree of positive evaluation attached to a specific referent

by an individual (see Fishbein, 1967, pp. 257-260; Opprrnheim, 1966,

pp. 105-112; Osgood, 1967, p,. 112; Shaw & rightill. 1967, pp,. 2-7; Trhurston~..

1967, PP. 77-79).



African Content



The attitude referent in this study is thle inclus~ion of African

content in the established K-12 curriculum. The uise of Afr-ican content










is not restricted to nn are.a stuidies approach where stuldents focus inl

turn on various culture arease of thle world. Africanr colntrnt refers to

any materials gamese, recordlings, readings, calSe'-stu~i('S, maps. films,

biographies, Activity pacts, learning renters, sp~eakers, field trips,

kits, etc.) which are used w~ith students to achieved srome currjcicula

objective regalrdless of thie ogannizing plrinlCiple unlder whtich thle objee-

Live fits. A teacher's aIttitulde toward inlcluding Afrie~nn content is of

interest regardless of whiether thle fralework is worldl jingiraphy, world

cultures, peace studies, future studies, de~velomienit studies, global.

studies (see Recker, 197?), or any other conceiptual sicheme under which

African content could be subsumed.




African Studies Oultrench Institutes


For the Pulrpo~ses of this study AFrican studiies outlrrech ins~titutes

refer to one two-week sulmmer~ inslittute and (one fo'lr-week~ Blummi' institute

aimed at expanding teachc r aIwarencss of Africa andi de~veloping new

approaches to teaching about Africa. Ther inrstitute~s includlE thr~ee com-

ponents. T`he first uses formal presentations to dissemijnate Enformation

about AfrijcA. A second involves the teachers in enltlural Activities

including African music, films, art, and cuisine. Thle thirdl compo~nent

involves thle teachers in developing materials andi teaching strategiies

for tEaching about Africa..




Internal Consis tency


Reliability is the accuiraiy with which:I a test mo~nsirres thiat which

it measures. A reliability coefficient expresses the squraredl correlationl









between subjects' observed scores on an instrument andi their truie scores

on thle trAit heing measured. WheIn individual tpes items hanve higih cor-

relations with the total test scorrE relativi to their item variances, ther

test of whichi they are a part is said to possess hligh Enternnl consis-

tency which is a measure of reliability if ther test is intenlded to be

bomageneorus~as in an attitude scalP desijined to meansure only one dimensions,

(see Cronhach, 1967, PP. 132-167; Knider & Richardsonl 19?7, pp. 151-160;

Lord & Novick, 1968, pp. GI, 87-95, 179-140. 211-216, 331 ; Remmers. Gnge,

& Rummel, 1965, pp. 129-130).




Construct Validity


Construcrt vaidity is thle degree tor which n1 tst mneasulres the con-

struct or attribute that it is designed to neasure. Thep construct or-

attribute is no~t operationnlly definedl. Thei1frefor the problem is for

the researcher in construct validation to ide~ntify thc iionstriretor-

constructs which account for thie variance on a teist (noe Campbefll &

Fiske, 1959, p. 100; Cronhach & Meehl, 1967, pp. 263--270: Lord & Novick,

1968, pp. 261. 278-279).




Construct Validation


Cronbach and Mleebl (1967) describe a variety of Lochniques that are

used to establish the construct valid-ity of 1 test. Th1ey conclude thant

the problem is not to determined that a test is valid for mensulring a

construct buit to state as definitely as puonible thr degree of~ validity

the test is presumled to havel (p. 255). A mai ior test o~f construlct

Validity that was used in thie current study is thp croup differences










method (p. 251). Constr-uct validity can he de~monstrated if groups which

should differ in the constrlctt do in fact hnve different mIean scores o~n

the instr-ument. In addition, to establish conistruc~t validity it must bi

determined that the construct beingi measured loateally relates to other

similar constructs (p. 252). Consequently, the snciies onl the attiturde

scale should correlate with1 measures of other cornstructs relalted to

attitude toward Africa~n content in th~e curriculum. As a final test of

construct validity a studyl ofT change over occasio~ns (p. 253) wais con-

diited to see if the attitude scale could detect channge in attitude

toward using Aflrican contenlt in the curriculum as .a result of partici-

pation in summer instituites nimned at inc~reasinji teacher competence in

the area of A\fricaln studies

In order to test the construct validity of a new inst-rument it was

necessary to state specific hyp'otheses conlcerningi thle oultrnmes of thle

proposed analvfses. If thle anailyses confirmedd thec hy~poltheseS, thlen con1-

struct validity would be established for thr instrument. In t he hypothesess

one had to state the expected outcomes and the crite~ri: by whiich juldg-

ments would be made about whether the hypotheses were confirmed or not.

Listed in the next section are the set of hiypotheses which if con-

firmed by statistical analyses would establish thle construct validity of

the Florida International Curriculuim Assessmnent Scales (F1ICAS), the name

chosen for the attitude instrument developed in thi is study. The cOverall

Hypothesis provided a Framework within which specific tests of constructt

validity could be designed. Oper-at ional HypothellSis Onei dic1I(' nt di reCtly

dleal withi establishing
as to whether or not teachers' ratings of thle relevancel of African content

would be affected by the natulre of the test format. Spectfically, if











teachers were asked to~ rate thle reclevanrce ofI Africa.n Iontlent Along with

the relevance of Latinl Amerlean or Eurropean co`il~ntent woulld the~ir TraineS

of Africa be arffctted? Consequelntiv, three for1me of the F~1CAS were

developed -- rone asking teclrhers to rate Africanl co~ntenti only: a seennd,

African and Latin American content; and a third, Africain and E5uro~pean

content .

Hypo7theses Twro through F~ive were designed on, test the construct

validity of thle variousi FTCA~S subhscales. liypotheses Two and Three stAte

expected correlations between thle FICAS subsrcale; andl relatedl constructs:

worldmindedness and knowledge ofi Africa as mea:su~red by the Flo~rida Africani

Knowledge Scale (rAKS). Hyplothiesis Foulr i s biedl an thle irolp, d i fferenres

technique to test whether the instrument courlld derlret an expect d dlif-

ference between a1 pilot group consisting of a1 cross section of teachers

and thle institute groups consiistin ng oE teachers who11 were self-selected~

to participate inl summer institutee s on Afrtcanlr studl(ies. Hypothesi~i S FIve

was desigined to test wjhether or not thef FI('AS could de~ltect a channge inl

attitude that was expected to occur as a resullt of plr-ticipation in th1e

summer in~stitit~es.




Hyoteses


Overa ll Hypothiesis



There exists in teachers an attitulde to~waird telling abioutt Africa

that can b~e measured by hlavinga teachers evainatetr the impolrrtance of sorcin

studies Fbjectives andl the 1elevance of African content to achiieving:

these objectives.










Operational Hypotheses


One. Ihere will1 be no si gIn i icant d ifi Iereces~r (p .25) ?1mong1J thet

mean scores of subjects in the pi lot samplr assiclnarl to thIlree dl ffe rent

Forms of the Florida Inte~rnational Curriculumn A~ssersnment Scales.

Two. There will be a significant poiit-ive enrrplation (p < .05)

between scores on the Floridan Enternational Cul~rrienllum A~sCSement Sr-aleS

and scores on thle Worldminrledness Scale.

Three. There will be a positive significantly (p .05~) correclatonn

between scorPs orn the Floridan International Ciirriculawin Assesssmen Scalcs

and scores o'n the Flor-ida Aifrican1 Knowldicag Scale.

Four. Sulbjects who~ elect to p~artiripalte in a r iriculuml wonrkshiop

oni Africa will have a significanntiv higher menc, scorec (p i .05) o~n thle

Florida International Curriculumm Assessment Scanles thn~nl I comprlrisonl

group of a c-ross section of ten:cher-s.

Five. Thef meanr postterst scortes of a group' of tflncherrs enlrolled in

an African, studies center smmerliC inlstitulte will ber Signlifil.anltly


(P_ < .05) higher than their mean score on a protest of thle Florida

Internationally Curriculum Asfsessment Scales.




Significance


The vaidation of an attitude scale to~ mrasulre teachers' attitudeis

toward A2frican material in the crlrriculumli is expctedr to open the door

to a variety of research, appllications. TheI1 instruments, couple~d withi a

knowledge~ test, can, be used hy Afriienn stmllion naltienchl programs throuih-

out the United States to assess the impact of their efforts. T`he

attitude andl knowledge Enstruiments can be use~d as p're anld p)slt ~tes






-11-


to evaluate c-hange as a Iresult of wo~rkshonps, semlinuls, ori cu~irse treat-

ments.

In addition to programs evallenti on, the newly Inntruelfll edtl inlStrulml'n

canl be used to measure attit-ude as either an intlepre1~~ndet r dependent

variable in more basic research stuldies. rthswhardsiin

outreach and/or inservice components for teachers, it may~ he well to know

what factors relate to teachers' attitudecs about Afiric: in the clrr-iculllm.

Knowing what factors mos~t influence the forrmntion If Ilese attitudes,,

designiers of teacher reduction programs can Laike this information into

Account. Other research~r- s may be interested inl using ther scale to rellacte

alttituldes and other variabless to pupil perfo~rmanr le.r The hiypoth-)es sized

importance of teacher attituldes for pupil growthll could~l be empiri really

tested.

The design of the attitude scale in thle cir-renrt project coul~ld csily

be adapted to measure attituder toward mother contrlnt Iwcns as wfll. Thle

adapted instrument couldl assess ther perceived relowa~ncef a~ .ny drsignatet d

content domanin for achi~iving: social studlies obhject i ve or even general

educate ion obje ct ives. For :Ilthiough this study is focursinS on social

studies o~bjectives, both Engile (1965) atd ME1tcalf (I(191) maintain that

social studies objectives are often indlistingulishnble fro~m thlose for

gen~eral education .





CHIAPTER IT4'0
REIVTEW OF: LTITERATURE




Thiis review of literature demornstrates the Rl hollarl concern for

international eduicatio~n in its many manifestations. Horwe~ver, in spite

of that concern, very littli is being done in ourl publl iC schOols or

teacher training, instituitionis to promote inlternaitjlinll education. The

second theme developed Es thaot teacher attitudel s towanrd eduelntiolnal

processes as well. as toward stuldents themselves hanve b~reen rlmo~nstrated

to play a major role in thle success or failurer of mllnin ntional prog~iirms.

Finally, thle methodolog~ical literature exsplicating issules in attitude

measurement is reviewed. 1hie uinifying frctorr is thle imlportanre of

teacher attitude for the snercess of internaio~lina ~l duration programs.




International Edu~caltion


Thie Theore ti cal Need


The occurrence of ;1 second world war provokled somec sociaL scientists

into conceptualizing a "moral equivalent to war" (I:. Nu~rphly, 1945).

Central to their prescriptions for a new world order was the conviction

that education neerdedl to be freed from natio~nalism to become waorldminded

in orientation (pp. 240-2421. By this thle sorcial 'Ici(.tistL: meant thatt

education should focus~ on precparinE studel~nt s for intelligent world

citizenship basedl upon demoirracy at home (p. 242,). Ju~st fourr years later

Flhria Montessori pub~ilishd hier Educazione e Paoce (19493/1972) ralling for










edulcationr to enlcouragn e the piritua~l andi~ mra~l l derveInlpment of indtividainl s

in anr effort to promoted a plnrcful worrldi.

Growinp out of the bhreakdown of world olrder ini thei 1940'9 were thre

intellectual threads calinl: for greater intrernati~n ional udersFtandingl)

which were to be picked up byv eduicatos in th~e 1960's. A~t tha~t time

thlEre arose a tide of critic ism among, socin studics rechtentorr- s ann:inst

the othnlOCEntric. orientationi of Americanl scho~oling.. lhe as i r orga:ni zn-

tion of social stuldies culrriculln had not cha~nged sincer thec 1916 Natioinal

Educational Associationi Colmmittre on Social Studies o~f thle Conmmission onI

t-he Reoirganizationo of Secrondalry Edulcation, Ircommende~d n1 pradrs Rrven to

twelve scope aInd sequenle whlich rentererl on Ur.S. andli Faropleanl hiStolry

and civics. relaltively newJ social science~ suchi as scio~logy anid

an thropolg wnire re excl udeld f rom the rec ommienda thu asr e was any! mrn ti oni

of non-Wrestern history or reniture (Massialns & Cox,z 1066). :Insfoi (19F61)

in hlis study of schools accredtied by th~e Northll Crntral A~ssciation of

Colleges and Secondary Schools conflirmedl tha~t high, schools conrtinllarl to

iignore non-Werstern arras Of the world right intor the~ 19h0's. Sepai;,rte

courses dominated by A~mericaln and world (Europea~;n?) hlisto`ry characterized

the curriculum (Nlassinlns & C:ox, 1966; No~reland, 1"h2).

By the early 1960's several wr~it~rs w'Ere cailing [nc efforts to

internationaltze the social studies curriculumm. Hairlandi Clevpland (196,0)

suggested that an ulndergraduatef edlcationr in foreign ;I~ntrs Fhoulld

defvelop cultural empathy or still in uinderstaindinr ther innler logiic of

other ways of, life as we~ll aIs theF predisposit inni to~ relrnin fr~om e~nn-

demning these ways of life be~nllSe they arr differently If promoting

world peace and international understanlding were to brcome edulcational

goals. Stearns (1966) assertedl that teacher education muist





identify and~ clarify thc inlformntion, anttitudes,: ilr an kills thant are

related to worldminldelness.. He urged the es:tabllishrment of an indepenldentt

international curriculum lab~oraltory, worldi college:~ flnters,, and a

foundation for international edrrention. In 19h6, thec 11.S. oongress

gave impetus to the movement by passing thc laternat ional ~dulcatio~n Act.

Thiis act called for developingi knowledge of other couniitries to promotee

muitual unders~ntandin anid to strengthen re Lat ions between th~e UnitEd

States a7nd mother countries (quoted in Klassrn & Mo~ore, 1968,H p. I).

Edwin 0. Reischauerr (1973) added his statulre to thei rlebate by criticizing

the past orientation oif emphansis on Western Eulrope. 110 Irged that thec

various culltulral dimensions of the non-Western world1~ bec given more

emiphasis.

C~onernedd sholar-s andl educators have attemnpted to, dsc~ribe inter-

national ctdwaetion and non-Wecstern studlies as they woilrl rPply to a

school curriculum. Robannnn (19)73) defined interenitluranl reduction "nis

a structurrini of loomning cvxperinces thatId will helpII hath Studen~ltS aInd

teachers undterstand and useP conICets for u1nderstandringl anld working~ toward

solutions of indlividual. and intergroup problems -- loical, international.

and world wide -- that~ arise fromn cuituiral diVersity" (p. 31). 110

offered a conceptual framework for interculturanl edulcation buijlt on the

view Lthat culture is two-tilered. TJhe "macr-oculturc" is hosed on power

and is large-scale, shared, and worldwide. "Mlicrocullture is based on

love and trust or sperin1 interests andr is smanll-side~ and family and

community oriented (p. 19). Hle argued thant ednenitllrs needed to dis-

tingiiish thie two culturess and~ rebuild schools arenldingly in rrder to,

(a) enable everyone to nlegotiate the staire bertweeni malcrocullturre and

micracultuir e equally wat f n b~othl directlions and (hI) Icuch peo~plee ablout










thle natulre of the marllny mter-lcultures! C in orde t -derfarofoe o

thle next (p. 20). 11e presentedl three key values whichl shiouli lunder-lie

e~ducatio ni in th~e two-tiered frameworrk. At thei rnrvlo~ le~vel thet principal

value is "equlality of opportunity for all." Her Iluggested that "live aind

let live" should govern a Ifai rs wi thin and between mi croc ut ltres.

Thirdly, e~quln access to the stairs between micro andnr marocrllturr e shouldl

be guaranteed. Bohannan insis ted that hIi s aIpproach-l is ne either ont i-

patriotic nor culturally relativistic sentimentalism.

in the Parly 1970's gilrobal stiatles b-ecame the fansl of several

educators interested in international ediuentionl. narilon (1971) proposed

peace stud~ies which concentrate on the issues of huminii aggression andl

conflict resolutionr from thec interpersoinal t the interlnational level.

In a monograph Global Develr lament Studies (19)7 ) ther Manage~r mcnt Institulte

for Nationn] Development ouitlinll a senior hiigh eniirirfrulum thant gioes

heyond the trannsfer of know~ledge to thle chanllingl ni thle stuidenits' per~-

ceptions of reallity. Consc:iousneiss raisingr andi uninesr examination in

relationship to in~ternationall dievelopment issues conrstitute major

objectives inl this program. Becker (1973, 19)7/) advo.cated a cuirriculumi

which helps students to seec the world as others son it. to become aware

of and adept at using alternative sources of information and evaluation.

and to develop a willinginess to consider comlpeting views ofr reality.

Bover (1975) added a futures comrponent to the procless of developingi a

world view among~ students. W)l~hlberg (1974) and~r UNESCO(:( (The UNESCO

Associated Schools. 19)75) conlceptualx~ ie n1 clnhn1 studlils aPPT'rox h bSed

on the relationships between peop le aInd the ir envi ronments. concern-

trating on the theme o~f interdlependence Almgren a nd (:ustafssonl (1974)

suiggestedl thiat international education is aI reqluisite for hulman survival.









A majority of Amerlean public school tearher-s shanre a concern for

greater international underst-~lnd ding. Ciphity iPercent who respoinded toi n

recent National Education Association questionnaire said better rcoopern-

t-ion among nations was a key iSSule inl promoltIn: ai PeCnlEful world

community (McCarron & McCune, 1974).

Within the broadl framerork k of inter-national eduelntioln several recent

books hiave been written to Guide teachers and crirriculumi plannners in

using Africaln content. E~.J. Murphy and Stein (1973) suggcstted six broad

reasons for including Aifrical in the social stuioris curriculum. Under

thle headings of superficiality and rthnocentricity. Hattl (19)77)

has categorized 30 problems~ which currently exist withi the treatment of

Africa in Almerican textbooks. The Afrianl-Amefri cni I nsti tute (Co lliins,

1970), Boyer (19)69), and W~illmer (1975) pro~videdl pract~tiinners with hanth

theoretical considerations and coincrete suggecstions for including an

African perspective in an social studies p~rogral..




Existing Programs


Over the past two, decad(es, intellectual leadership hans OfFFfere n

variety of conceptual frameworks for incorp~orating international studies

into th~e curriculum. A key concern is to what extent these scholarly

appraisals and recommendations have found their way into the r~urriculai

of American schools and into~ the inistructio~nal strategies of its teachers.

John Lee (1974) offered a1 somewhat hopeful view. Heo detected several

trends in the late 196,0's anid earlyy 1970)'s which sugge sRte<1 thait cnrrirulal

were being inte national izedl: Jnternational reduction wais beginning

Earlier in the child's schiooling, the concept of thle woirld had been

broadened to include the non-west, and there~ was a1 shiift to the study









af the world's people as a society (a gilihni view as ropposedl toi the study

of one nation ?fter- another) (p. 149). Tog oeecuaigsit

may be occurring, Rose Haydecn (1976,) said thle responses: of Amerricain

public education to thle challenge of preparing citizens to respond

intelligently to global problems was inadequalte. She was palrticullarly

critical of state education naencies whiichi she claiimed dlid not perceive

the need to provide leadecrship for gilohol izine public fdulcation. In

spite of some g~loball curricullum develocpment, some inservice training on

international topics, and increasedl stress on international training: in

teacher education, too little has been dlone to internationalizef public

instruction or to evaluate its results ag~ainstt plobal Roals andi objectives.

Thle A~spen Institltc e has formedl a naitio~nal commnission on coping with~

interdependence (M~orehouse, 1975). Th1e institutee is concFr-ned with two

questions: First, to what extent do American institlt ions,: percelve thie

predienment of interdependence and its implication;; second, what new

attitudes and arrangements many be requIreTd Ln enhance(. thel rnpacity of

Americans to cope withi it? Thle findings revealedl tiat edulcatilnall i~n-

stitutions are ill-equlipped to, deal with the issue o~f interdlependelnce

because it is perceived as a national concern while schiools are run as

a state and local responsibiity. For this and other reasons. it was

concluded that much of the burden for creating a civic literacy on

interdependence issues wouliid have to be ho~rn b~y instttitutios other thann

formal educational ones.

Nonetheless, some schnols anti other sulpportiny agencies f have made

attempts to change curriculum and rearienlt inservice to implement

international education. Several school districts have dervelopedl

international studies programs of one natuire or another (Frepman, 1974;










A~frican Studies Handbook, 1971; Worl~d 1 story Serlie: Afrt-ica. 1972)

State departments of education (Jonles, 1970)! andl other agrencies (Oswald,

1974) have contributed t-heir efforts to RssistinG publliic schantl ls to

improve international stuldies progprams. Thle Afriennl-Americ ien Institutee

(Taching; About Africa, 1972) and the African Stuidies P'rogram at thec

University of Tllin~ois (Schinide,. 1975) harve mnade Ionlgoing efforts to

improve African studies through teacher inservice eduiration.



Teacher Art titudes and Their Mleasurement


A crucidl issue in assessing the role teachlern plaiy in the instruic-

tion of children is toi what extent teacher attitudleN influence the

outcomes exhiibited in children. In conjulnction with tha~t issue, is the

problem of measuring teachers' attitudes anrd their relatiionlship to other

variables. W~iseman (1971) believedl that thle sing~le moist significant

outcome of educational resonrr ch in the last deade~l wasn the rentlizattion

of the power' of teacher Itt ituldes and teacherr~~~r expr tions.

Much of the research into teacher attituldes hais foculsed on their

relationship to various educational issues anid pra t ices In anl experi-

mental study, Homme (1968) reportedly a failure to imprrove teacher attitudes

toward programmed instruction, apparently because o~f lack of commulnica-

tion between experimenters aInd teachers and lack of opltimulm classrooo

conditions. However, Ma~lini (1969) foulnd that nrw "Spanish for Co~mmuni-

cation" materials designed for programmed mlantery of instructional

objectives Emnproved teacher attitude towardi Ilanguagel in~struiction, as

well as changed their teachingl methodologies~ and student performance.

A third study deaing withl tracher attitudesR towedri so~me aspect of

educational practice was conductedl by Walmon, Rell, andi Ramsever (1971)










who hoped to show that teacher randiidates trn inedl in the microplanninfi

technique~ would score higherl onl tests of teancher elffectiveness buit

remain unchanged in attitudel towrdmt students ;Ind teaching:. Thef study

found that teaching teancher-s to contrn1l cognitive struictuire variables

increased their students' reasoning ability buit hadl no effecct on recall

not did it affect any mneasulre of teache r aIt t itde. T'his last finding waos

considered a1 positive one because the experimenters; were aIttempting to

avoid change to undesirable attituides on the part nf tnacherT cnd~idates.

A study by Mlax-in (19)74) showed how negative teacherr attitudles could black

progress in educational practice. He found that trnChers had little

knowledge of, and negantive attitudes toward, edurlnifountl resealrch.

Consequenltly, they shied awayl from its fin~dingis.

Okey (1973) studied thc reflects of learntni: Bloom's mastery teachingf

strategy on teacher candlidate's attitudes~ to~ward tests, clrades, and

diagnostic teaching,. Thle experimentltn group shoedp i sjignificanltly poSi--

tive Attitude Rains in this studyv. Inl a ne~-shot ryxperiment in which

teacher attituldes toward ~lth effectiveness; of diFfe~rent math programs

were supposedly manipulated by havingi technlers read~ bi-ised statements

about the programs, Brager (1970) found no sig~nificant difference in

students related to teacher attitude. A moreH sophisticated longiitudinall

study by Beauchamp and Conran (1975) was set uip to measure the e~ffets

of the operation of a cuirriculum eng~ineering sy~stem. A\ cauisal model

employing path Analysis (se~e Blalack, 19hi4, 1971; licise, 19)75: Van de

Geor, 1971) was; being: usedl I to xamine the rclationships between leader-

ship, currliculum engiinpering,, teacher attitudesR teanchr r performances,

and student performance. Frey (1973) in a survey study of 406 elementary,

middle, and high school. teachers at-tendinfi a summer session at Northeirn










Illinoi~s University found thatl tenchers knew much nbout~ behanviornl

objectives buit held rather neiitral attiutudes about their cffectlyoness

in pupil performance. 13ogarz (1970) found a po~sitive relationship betweencr

teachers' attitudes toward the effectivenress of instrulctional materials

and actual pupil achievement.

Research on teacher attitudes toward snhlject matterc had b~een confinled

mostly to the subject of science. Oshiman (1966). in stuidyingi the dif-

ference between lectulrcle-downstr ation and individunt investi~tientn ars

approaches to teachingi science methods courses to Ilrmentalry school

teacher candidates, measured attitudes toward scIenICe a~s wiell as con-

fidence toward teachingf science, achievement in sc~irener, an d student

teAchingi behaviors in science. Only' the confidnce va riable shOwefd

signi ficanlt d difference between expe~rimentanl and~ ontroll- groupsI). Srge

(1974) studied attitude toward science and its relat ionship to know~ledge

of science. Hie found a low positive corrplatiio n f .?5 hetween measulres

of scientific knowledge and attitudl e toward scince.c S;inceF the coef-

ficient of determination was .06, he questioned the relationship between

knowledge and attitude. However, Bl1eh and Zakha~rindefs (1975), finding

a .25 correlation between science attitudes scoresi and g~rades in science

taken as a mensurie of learning, concluded tha;t ktnowledge of science and

general exposure to science hadl a positive influernle on attituides toward

science. Thle previously mentioned Almgren andl Gusltfsson (1.974) study

looked at the relationship orf teacher ac~ttitmle to~ward teaching about

int-ernational questions and studeints' RttitudelS. TheyC~ found a positive

relationship.

An important set of research rln teacher attituides has focused on the

relationship between teachers' attitudew toward their students and the





-21-


pupils' subsequent achievement. Arno (1975), Clarki (1975), anrd Cecil

(1.971) all found a positive relationship between techellr attitude towrdirr

students and pupil achie~vement.

TIhese studies have made u~se of one or several of the attitude

measurement techniques developed and reftned since thle 1920's. TIhurstonet

(1927, 1929, 1931, 1967) was the first to take attitude measulrenpnt ouit

of the suirvey stage and malke it more rirgorous by intlrodlucjin the equial-

appearing interval technignor. At the time thant Senshore and Ileyner

(1933) were modifyinR Thulrstole'ss techniques. Likert (1~97, 1970) pro-

posed the summated ratinfi method for measuring: attitudesr. Subhsequiently.

G~uttman (1944, 1947), and O:;good (1967; casGood~ & Suci. 195i: nsgoord,

Suci, & Tannenbaun, 1957, Snider &r Osgood,, 19h9) devisedr niternative

systems for attitude mensurlrment.t Edwardls (1957), alppnheim (19h6),

and Shaw and Wrigiht (1967) have each madle attempts tor Rynthesize the

previous work in attitude mealsurement anrd providi rsidelines or devrelop-

ing attitude scales according to thle different techlniqucs avani~lable

Two studies have attempltetl systemati c enmparisons~ of likiert and

Thurstone techniques of scale construction and scoring. Providing a

critical review to previous research, Seiler and Hough (1970) concluded

tha~t the Likert scoring methiod was superior to the Thulrstone method with

regard to the value of the reliability coeflicient obtained for a given

number- of items. Using L~ikert construction and scoring techniques, one

could usulal~v develop a scale with a .90 re tab i lity coerffiicint w~ith

only 20 to 25 items, It would take twice as many Thulrston~e connstructed~r

and scored items to obtain ther same rel aini Ity (p. 171). .ItRecrd,

Weber, and Lundmark~ (1975) used a mul~titrai t-mu~ltimethord matrix (see

Campbell & Fisko, 1959)) to colmpare the test-rftest r~liabil~ities and










divergent and convergecnt validitiEs: of fouir meth~ods of atl-itude measulre-

ment: Likert, Thurstone, nsg~ood, andi Guilford's self-rating. Threy fouindl

little diElerence in the companrati ve rel iahj it les, thr mletho~d variancEs,

or in the divergent and convergient validities of eachl method. W'hate ver

the merits anid demerits of each method may be, these ilifferences cancelledl

out when put to the empirient test. Thurs any of the above techniques

might. be applied to assessiing teacher attitudies towardl social studies

cuirricula, or specifiently African content.




S ummary ~ of eRser~chI


The chanengin nature of the modern world has ranisedi concern amngnfi

educators that curriculal and teaching practices neecd to become more

internationaized. Some senttered eEfacts at curi-rien~la r-eform have b~e-n

made. In addition, scatterted attempts have hern made~l to prepanre tecrhers

to use international content~ in their teach~inS. yet little reserrch has

been dlone into the effects of teacher ttiitudesR on sublsequentt students

learning oE either concepts or alttituldeS relilatin to ilterlnational

issues.

However, for fifty years educatio7nal psychollogists have heen de-

veloping and refining a nuimber of techniqises for mcasulring attitudes.

These techniques have been used in recent years to study teacher attitudeis

toward stuldents, varioulS sulbject matterss, and different educational

practices. Whllat is needed now is a mneans of measuringC teacher attitudesi

toward teaching about fo~reirgn cultuires anr interlnationall1 problems.

Making use of accepted techniiques oE a~ttitulde measuremnent, thris study has

uindertaken to devise instrumentation that could validlyv measure these

teacher attitudes.















CHAPTER TH~lREE
MElTHlOD



In establishing the cronstruct vnlidity of theF new scale to measure

teachers' attitudecs toward African content in thef rurriculum,~ three

approaches to construct validation were used. First, corrreltions betw~een

scores on the F~orida International. Curriculumn Assessmeint Scailes and other

measures of similar construlcts were examined. A p~rEViously dePveloped

knowledge test of African content wa7s refined for uise as 1 cr-iterion in

this study. In addit ton another attitude scl l mieaisuring at ti tudesn

similar to, though not idential~ w~ith, attituldes torinld tonehiingi abour

Africa in the classroom was nlso used. The total fiscal sores from these

two instrumnents were correlated with thle niewly deveclopecd FTCAS sublscales

to establish the first type of~ construct vallidity. Se cond, nrollps whiich

should have differed in scores o~n the FlCAS\: subscalrs were .Ilso tested

and compared to see if expected dlifferences had berrn detected. The me~an

scores from the pilot snnmple were compared to the two summer institute

samples to see if the newly developed sub~scalesi could dletect aIn expected

difference inl attitudes between the pilot and mother two groups. Finally,

the pretest scores of the institute samples were co~mpared to th~e post-

test scores to see if any sublscales could meansure a dlifferencef in attitude

expected as a1 result of part iciLpation ini thle institutes.~

This chapter includes the details of the development of thle FTCAS.

A discussion of the procedures usedl to establishI its reliability follows

a description orf its developmentl~. Finally, measures taiken to confirm









the construct validity of thle FTCA1S are lielin~nated along withl a discussion

of the other instruments used( inl thle validaltion procteSs.




Attitude Scale Construction



Purpose


The purpose of this study was to develop n reliable instrumllent with

sufficient validity to allow it to be used in attitude research andl in

curriculum and instruction evaluation. Ano instrulment was desired to yield

scores on teacher attitude toward African content whlich coulld be analyzed

to learn mnore about the relintionship of thiis attitude to other cons~r rnets

and observable behavior. It was expectFed that thle new instruments would

serve a uiseful purpose ini evaluating p"re and inservicie efforts to~ instill

an international perspective in teachers with reFgardi to their own

teaching.




Scale Dimnensions



In constructing an inst-rument to measures attitude~s towanrd Africani

content in the curricullum several steps were tak~en. The first concern

was to clearly define the dimensionls of thle const rllt under considerat ion

(G~ardner, 19)75). The identified construct hand to- be unridimensional so

that all scale items could be summed to produce a meainingiful total. scale

score (G;ardner, 1975; TVhurstone, 1967). Suchi a scale score wediid hep

uninterpretable if A single linear continuumn did no~t u~nderlie th~e attitude

being measured. Attitudies ar1e assumed to viry- in qualnity anrd intpnsity

along a continuumn from p~osit ive through neitranl to negative (Shaow &*

Idright, 1967, p. 7). If, ini fact, somne scale items measure dimensions









indepenident of the major underlying: conistirut the reslltnin attilude

score will he meaningless wi~th regardl to the attitude that1 was jintended

to be measured.

While pains were taken to insure that1 the attitudef to be mneasured,,

attitude toward thle inCl1sionl of African content inr school culrriculumm

was not confouinded with otherr attitudes suchl as attitudec towir-d AFr~o-

American studies or perhaRps attituides to~warl trradi tional versuls inqui ry

social studies methodologiecs, care was also` exercised to include all

aspects of the uinderlyin6 dlimEnsion (Oppenholm. 1966. p7. 117).A

teacher's attitude toward teanchingi about Africa in the curricultun might

be made up of feelings ab~outl priorities inl thle elr-riculltma feelings about

the school's rol~e vis-a-vis cross-clltluran l reductini l feelings about

Africa's relevance to the canriculumo l or any number of other aspEcts.

These various aspects had to, be identified and mleasulrted~ if a valid scale

were to result from this stu~dy. An analysis of rtesponses to open-endedd

statements about African studies in the curr-iiculum was uised to determine

the r-e~evant undpr-lying dimensions .

A sentence completionl questionnaires withr threer itrms on it was uised

to elicite teachers' views onr African co~ntent. Thei seconnr d item on the

questionnaire was thle most p~ertinent for formullatngi the uinderlying

aspects of the construct which the reseaircher was nttemlpting~ to measur-e.

This item began "Africani content in thle pubtle school enrrriculumn

should. . ." After- completing thiis sentence, teachers wEre ~aked to~

respond to the second part of the iteml whlich said simply ". . heicause

SThul~s in addiction to open-endedl juldfnmnts s on African content,

reasons for these judgments were also solicited. A~lSo inclu(1de in the

questionnnire were two orthetr Ltemis: Thle first of these was simply "Africal

and the second was "Clobal cosiuses









'This qurestion~naire wasR admini:;terl to 100 pri andi inrservicr

teachers. The responses were categoirizedl in an effot~i t d1etermine thec

dimensions underlying attitude toward Africai in thec curirirurlumr (seef

Table 1). It appeared from analyzing the~se responses thait dile relation-

ship of African content to rognitive chieetives is ther major uinderlyingi

dimension of attitulde toward African cointent. lidhen combined with responses

related to affective obljectives, over- 78 percent of thie responses relate

to curriculum objectives.




'Table 1

Response Distributioo n for Open-Enldedl Ilnrstiolnnaire
on Africaln C~ontent in thie Cuirriculum?



Ca tego ry Pe rrc lnt of Responses



Cognitive Ob~jectives 1,7.0

Affective Objectives 11,T

Relative Importance or
Time Considerations 9.4

Intrinsic Interest 5.7

All1 Other Respionses 6.h


Note. Number of subjects equailled LOO, buit thr nulmber of responses
equalled 106 because OE multiple respo~nses.




Source of i tems



Trhe next step was to de~velopi a pool ofr items whlichl coni~ld he pilotedl

in order to construct. an adcequatef scale. These items had to differentintp

between those teachers who hold positive andl those Who~ horld negative

attitudes toward the object in qulestion (L~ikert, 1967. p. 92). Trhe .items










needed to be statements in which respondents could recognize real view-

points and feel strongly about them one wayV or the o~therr (Oppenheim,

1966, p. 114). These items~ had to cover ther ranne OEC asp~cts underlyinin

the attitude and, in a tradlitional scale, the range~ of intensity from

negative to pos itive.

For the attitude that this study attemptedi to mensuire, us~ina standard

Likert-type statements presented a potential problem of interpretation.

Since relationship to curricurlar objectives seemedl to he thle major

underlying dimension of attitude toward Aflrican content in the curricullum,

most items in the L~ikert-typef instrument woulld containi statemernts which

related African content to objectives. ?tro examples would be the

following: "The study of Africn is nercssary to pr-omotee international

understanding," or "The study o~f Africa needs to7 be~ inlud~l~ed in the study

of world geog~Raphy.l If an alttitude isl a rredispositition to art in a

certain WayI toward some objiect, then positive rrsponlsess to thcfie items

might not be measulring~ a Lrancher's inclinati onl to~ Iie A~frican content.

For it is possible thiat a teacher couldl agree that Afr~ienn content was~

important to achieve an objective but believe at thle saime time that thle

objective is not too important. Such a situaltton could~ proutlce a miis-

leading item response.

To overcome this potential problem, the current situdy used statements

of objectives as the items in the instrument. Each teaicher was asked to

rate the importance of the objective as well as ralts the relevance of

African content to achieivins it. TIhe objectives were selected to repre-

sent three content categarie~s covering.. a broad range o~f social studies

objectives. Some objectives are naltiona~list ically adiented in that

Amlerican content would he riliregnlr to achieve the c~hjecitivr (see Appendix









A). Examples include thefsi: "Studelnts should kinl'w ablout the impact of

technology on modern l i f in America," "Studeint s shoruldl hav~e a ktrnowledg~e

of the strict ure of Ame~rican g~overnment," aInd "Studentiis Shoulld~ sIPPOrt

efforts to protect America's environment." Othecr objectives are inter-

national ini orientation because they require content from outside the

U.S. in order- to develop thecm. Such objectives woulld hie the following:

"Students should acquire wor-ldmindedne~ss," "Studelnts shouldl comprehenid

the gulf between rich and poor countries," aind "Studrenits Shoulld understand

different ways of life on different continents." Still a third groulp o~f

objectives are those whose achievement does not dictante the nattennd

or international focus of contenlt in order to necomplisish themn. This

third group of oh jectives enalld be devrlopedl w~ith rather Amecrican or-

international content or both. Examples o~f this third type include

"Students should ulnderstand the balance of annture," "Students shou~ldi

understand thle exchrange orf Goods andl services," an11 "Stuidentl s shoulldd

appreciate a artistic ex~p rcss;ion."

In selecting thie objectives for incluston in th( instrulment, the

first step was to examine thle rspo~nse~s to the oPenI-ended d ques~tiOnnaire

given to pre and inservice teachers and then to review social studies

mnethods texts and school district object-ives statements. h olwn

sources were checked Ear examples of both cognitive and affective ob-

jectives: L.F. Anderson (1968); D.L.. B~rubakear (1973); L,. Ehman. 11.

Mlehlinger, and J. Patr-ick (19741); S.11. Encle (196,5); I.R. Le~e (1974);

B.G:. Massialas and~ C.n. Cox (1966); E..J. Muriphyv and II. Strin (1971);

D.Wd. Oliver and J1.P. Shaver (1966); E.o. Reischanllr (1973); .A.~ Scott

(1972); and M1.D. Waniman, D.D. Bell, and C..C. Ramsever (1971).









Based on thefse sources, a 11st of 54 cognitive chiectives andi 36

affective objectives was drawn up to be subhmitted to a panel of weight

professors of social studies edu~cation and cuirrienium Tispaelrae

the objectives as to the type of content thlat would be required to

accomplish the objective (see Appendix A). The coGn i t ie obj ectives

were written to include on equal number from e7c~h ofI six social studies

disciplines: history, political ficien~ce, anthropology, sociology,

g~eography, and economics. The affective objectives were not so easily

classified accordjIng to disc.ipline.

The FICAS could hiave been analyzed not only as:- a total score, but

according to several subscalecs as well. The sublscale structure of the

FICAS allows for 11 different subscal~es to be examninecd for conlstructt

validity. NJot all scales are independent of ealrh other, butI there exist

three sets Of independent subscales (see Figure~ 1). Thie 48 objectives

on thle scale can he divided accordling to cogrnitivc-n;ffecti ve criteria

(Bloom & Krnthlwohl, 1956; K~Rth~WOhl, Bloom, &r Navlia, 1964)!. inltO two

subscales wi th 28 An7d 20 i tcmsS r-espectively. Alte~rnaltively the scale

can be divided into subhscalrs of nationalistic, international, and

either/or objectives. Pl~acrlemen on1 these scales wals m.ade by a panel of

eight professors who were afiked to judge objectives on~ whether or not

American content or international content was requlirred to achijeve the

stated objective. The third possibility was that th~e objective could

be devel~pell with either type of content.









Naion al~,n1i stic




Nat ionlllis tic i
COGNAT (8)


INTTOT~ (13)


Co nir t ive
Inte rnatiolnnL
COUlNT (8)


E~i
I-O


C:
Ei
COl


thler/or






fective
th~er/or
FEroR (10)



bscale .
the text.


Cogn~itive

COGTOT (28)


Affective Affective Affective Af
Nalt onal1is tic Tnlte rnn tionall Ei
AZFFTOT (20) AFFNATI (5i) AF1FINIT (5) AF;



Note. Numbers in parentheses~ are nulmber of itens onr each sui
Words in all capitals are abrbreviations of subscalonn used in




SubscAle Structure of ther Florian Iiici~tntrn iona
Currinoluim A~ssessmient Scales


At lenst 75 percent agreement among professor wais required for an

objective to be fiirther considered. Of others objectives Rubmitted to thie

panel of professors, 43 of the 54 cognitive objectives andl 2 of the i6

af fective objectives met th is cri terion (see Aplllndi x A). Fulrthcr paRri ng

of obhjectives was done wijth an eye to maintanininfi anl equal rinumer of

nationalistic and internationally objectives on eachi scnle. Similalr

objectives were entlecd by p~icking thle one whiich had1~ the greatest percent

of agreement. The final 28-objective cognitive scale consiconal of 9

items which had 75 percent agFreement, ] 3 wi th 87.5 per:ent, and 6 with

100 percent agreem~ent. On th1E 20--objective lffective scale the~ numbers

were 7, 11, anrd 2 respectively,




Scoring


The Florida International Curriculum Scales wiere scored aIccordingi

to the summated rating process (see Edwlards, 1957, pp. 149-170). sin










data points were repinced with the roundledl group mrran foIr rnld1 itemi hasrrl

on the pilot sample of 131.. There were 80 missingf datal points suplplictl

in this manner out1 o1f 6,288 sucrh points. Trhe groupl meauns~ bnsed on the

pilot sample were uised wheni necessary for gieneratingi the scores requiired

to test Hlypotheses One, Tlwo, andl Three. Ilowever, for Hlypothleses Four anld

Five, the group means sulbstituted for the missing aintn points wsere based

on the 222 subjects which comprised the pilot samlelll as well as the pre

and post-test samples fromn the summer institultes. 1This base was choseni

as the most conservative procedure in gienerating total scores to be

entered into an analysis of variance, to thle ]atter en:se 105 data points

were supplied by subhstitutcing the group mean ouit of 10.656 total data

points processed. After respondents ratedl rnch iteml onl a Scanle fromn 0

to 5 along a1 continuum measuiring the celevanlce of African content ranging

from useless to essential, the total scale score wais coImpultedl b summingu

the item scores. The suibscale scores were generatedl in the same manner

by summingi the items on thle relevant subscales (see FiGure 2 and A2ppen-

dix C). All1 nationalistic items can be ident~iifid by reading down the

first column of cells in Fig~ure 2. For international items read down

the second column of cells. The complete either/or sublscale comprises

thle thirdl colmann of cells.


Either/or


12, 15,
12, 33, 41


38. 45


Nationaistic


Internationn1


Cogniti


Affecti


ve 1, 20, 2 1, 28, 7, 11, 19, 27, ,5 ,1,
30, 35, 19, 43 40, 42, 44, 48 21, 25, 26,

ve 4, 18, ?222,1 1 8 9 7 4
46, 47 16. 71 3h, 36, 17,


FigureT 2

Ltem Assignmefnt to Subhsenlls onl the Floridn International
Curriculum Assessment Scales










Alternative Scoring Methtlods Considered andi Discarded


Another option for scoring also presenitedl itsplf. As mentiorned

earlier, teachers could rate African content hiighly relevant for an

objective which they considered to be relatively ulnimportanlt. Forr ex-

ample affective objective number 27, which renlds "Studelnts should unlder-

stand different ways of life on d iffrerent rantinentls." had an atti tude

scale rating which was well above thie averdge for a F1C(AS Ltem. Ilow-

ever, its importance rating was only average. Cons equen~ltly, nlthlough

many teachers could see the relevance of African content to suich an

objective, they were not inclined to teach thle objective herouse they

perceived it as unimportant.

Therefore, in scoring the FICAS two special weighlting: options; were

tested to see if greater vaIlidity coniid be obtainal.l Since it was thoughti~l

that the importance a teacher attached to an ob~jective would he importa~nt

information in computing attitudef t~ward Afriennl rententl, tochlersi were

asked to rate the imlportancce of eachl abjectiv e whilh irade up the FTCAS.

First, all objectives with a rating of 5 in importance wer-e sroredl as

marked on the FICAS and all other items were weightedr 0. Second, a

rating of either 4 or 5 was usecd to solvct thie items to be scored, all.

Other items being weighted 0. In both cases i tem means were ursed as the~

scale scores inst-ead of summated ratings because of thle different

number of items receiving higih ratings fro~m diffe~rent subjec-ts. Whnen

these weightedt scoringi methods were used, thle values o~f coefficient alpha

remainerl virtually unChannged~ from those bhserTved w~ithl c.onve~tntion)

scoring procedures. Whe:n items with an importanicp ratinE of 5 we~re

Included an alpha of .94 was- observed, .95 fo~r items with impo~rtance

ratings of 4 or 5.









AIlthough these weigihtedl scoring methiods prodluced relitabilities

comparable to the conventional sulmmated rating techiniquie, theyv did~ niot

produce improved validity coefficients as predlicPted. Tefrt tr

native scoring procedure pro~duced an item mean thatl correlated .14

(p_ > .05) with the Worldmindednesss Scale and -.21 (p 05) with the

Florida Afrrican Knowledgef Scale. The second methodl resulted in an item

mean correlatingi .16 (p < .05) with the W-Srale anid -.18 (p < .05) withi

the FAKS. It appeared that using the scor-ingi alternatives of rnlculaingnf

item means for only those objectives which received a high1 importance

rating did not measurably inlcrease their vanlidity roefficients. Con-

sequently, further analysis was limited to testing the validity of the

subscales which were scorerl with the conventiornal liker-t pro~edulres (sci

Chapter Four).



Procedures for Establish ingReli h nlIitv nulll Val1i d l ey


Reliabilit


Coefficient alpha (Cronbach, 1967) was uised as a measure of relia-

bility in this study. The suibscales of the Florida Inte~rnationall

Curricullum Alssessment Scales were designed to be uinidimensiond.ll

Consequently, coefficient alpha which is a mleasuree of internal con-

sistency of the items on a unidimensional scale was an appropriate

device to use.

For thle reliability stuidy thle instriiments we~re naministered in

regular College of Edulcation class sessions meetings at thle University

of Florida during spring quarter 1977. This was trule for these courses:










ED) 600, The School Curricululm (tw~o sectionsl); ED)E h'0. Social Studiesi

Education -- Eleme~ntary Sch~ool; EDE)l 670, Lanlguag): e Alrts in thle Elempentanry

School (off-campus in Marionl Counity, Florida);) tiDF h20 Soclinecronomic

Foundations of Eduication; EDFi 666, Seminar on Research, on Effective

Teaching;; EDF 450, Mleasuirement anid Evalua~tionl in Edrucation: and EDFI 768,

Evaluation of Educational Projects and Systems. Thp students in EDS

635, Suipervisiion of Preservice Teanchers (off-campusll in A\nlachna Coulnty,.

Florida). were given instrulctions for talkine th~e tests and then allorwed

to complete th~e scales at home and return them at thle next class meetinfi.

The elementary teachers participating: in the study fro~m Hi~lsharoulgh

County were given the instruments duiirng ; n meetinl: nE um inservice class1

in social studies meth-ods in the spring of 1977. 'lec Ii~llthborugh~ Couinty

middle school teachers wrre administered theI scniOn duIjl:ng meeting~ Of

a county-widel social studlies curriculumi committees. Ife hligh rhoo~l

teachers in Ilillsborougfh Coulnty were ap~proalched thirrought their reprt~rmente

chairpersons w~ho Ieach~ requested three ofi their dpartment memblers to

respond to thle instruments o~n their own tjlime.

Thie instrument battery used to determine the reclinh~ility and

validity of FICAS contained 12 pages of material. ThIe first pag~e asked

for demogiraphlic data rrlatedl to teaching rxperiencer. ThPe three questions

asked concerned level, subhjrct area, andl years of R-12 teachingif ex-

perience. Tlhe second pagef contained thle inistruciioin s for completing the

importance ratings and thle I:ICAS items. Thisi requiiredl a two-step process:

First, for ench of the 48 enlrrienlar obijectives, thet prubiocts were naked

to rate their importance for the school Culricullum onl a scale from 0 to

5; second, for each objective the subjects were to rate the relevancie

of A~frican content for achievingi the ch~ject ives on n 0 to 5 scale. Thef









next three pages mande up the Flo1riia TInter-nationa~ (l Crrirltum Assessment

Scales. Page~s 6 through 8 containedl the W'orldminidcdnnss Scali (Sampsorn

& Smith, 1957; Shaw & Wirigiht, 1967) entitled "So~cial Attituides cquestio~n-

naire." The last fours pages consisted of the Florida African Knowledgje

Scale which is a 30-item revision of Projec~t Africai's Africa South of thle

Saharn test (19)68). MIost respondents too(k between 30 and 415 minutes to

complete the total battery.

Data uised to estimate coefficient alphar wer~e icollectedl from

131 persons with K-12 teaching Experience drawn Trromi iradulate courses at

the University of Florida Collegie of Edllation and from the ranks of

practicing social studies teachers in thle Htllhboroughil Counlty, Flor~ida,

Public Schools. A total of 1693 people responded to the instruments.

Thirteen subjects wzere eliminated from the sample bconrse thley failed

to complete one or mor-e of the scnlesL or herouIse theCY wiere foreign

nationals without American, publlic schanl cxp~erie~nce.. An additiornal 25

respondents in the graduante coir-ses were set asiC becauISe theyV did no't

have K-12 te~achini: exper-iencre. Consequ~ently, thle pilot~ sample consistedl

of 131 subjects all of whom were either- practicing, teachers or hadr had

previous K-12 teachling experience.

Sixty-six sub~jects were enrot~cd in giradualte coureses at thle Univer-

sity of Florida, including 19 teachers enirolled in an off-cnmpus course

in M~arian Coulnty,, Florida (pop. 69,030). T'he other h5 surbicts w~ere

p~racticingf teachers in Hli~llsoroughl County, Flonridal (Pop. 190(,265), which

includes Tampa and surroulndinC. area. Thle total pilotl samlel' consistedl

of 39 people with only elementary experience, 20 wiithl only middlef school/

junior high experience, and 25 with only highi schnotl experience. Eleveni

teachers had experience at both elementary anid mladlee school,, 18 had





both middle school and high school experience. Eiilhtee~in other subjects

had experience at all levels. With rega~rd to years of experiences, 12

teachers were in their first year of teachiing; 28 hadt 2or- 1 years

experience; 23, 4 or 5 years; 34,, 6 to 10: 18, 10) to 20 years; and 12,

more than 20) years experience. TIhe subjectl matter breakdocwn was weighted

in favor of social studies teachers -- who were xps[ected to comprise

the majority of par~ticipannts in the suimmer insta nltCs. Thirty-five

subjects only had experienrr as social studies Leachellrs and an additional

16 ha~d experience teachinF social studies as wf]l as some other subhject

TwJenty-eigiht taught the basic siubjects (Languagne :IL~s, manthemaltics,

science, social studies) in elementary schools. Seventee~n hadl basic

subject experience as well as e~xperience inl somer iohir aIrea. Ninle Slb--

jects were humanities teachers in that they were nnlyl Clexeienlced in

language arts, art, music, or foreign Innguages. TIwenity-six teachers

only had experience in other areas: math. science, media, phy~sical

education, or vocational education.




Validity


Correlates. To establish the construcct validity of the florida

International Curriculum Assessment Scales, several hypo~theses were

generated to test predlictedi relationships invonlvinR the~ construct of

interest, attitude toward thre ulse of Afrie~nn contentl in thle enirricu~lum.

If a predicted pattern of corrrelations enluld he cmp i r i rn Iy veri fied,

then the construct valid ty of a test would receivr rnfi rmat ion. Th~is

study sought to verify positive correlations beptween Envnrableness

toward African content and two related construlcts. worldmindedness and

knowledge of Africa. Neithecr of the criterion attributes wars identical









with thle construct under studyl. Ilnwever, it wasrf tllo~rized thiat tearches

higih o~n worldmindedness anrd know~ledge~ of A~frical wounlld nlso tendir to be

hrigh in attitude toward the usfe of African1 Conlti'nt.

The sample, sometimes reffrerre to as the p~l~itl qsample, ulsed to test

these correlationail IIypOthesesC was the same one descrihod in the section

on reliability. The methodsc of data coll tionti~r wrlre also those rescribed

above .

The first of the instruments uised to establish the ronstr-uct validlity

of the FTCALS was the F~loridnl African Knlowlledgec Scale (FA`KS), developed

by thiis p~rojctl. Examination~ of the tent Africa Srouth of the Sahara

(1968), which is a 60-item test orig~inliv dclve-lopo1, for highi school

students as a part of "Prolject Afrien" (Rover & Ililks, 1968), revealed

a solid base upon which to build an up-to,-dat e knowkllge~# instilument.

Beyer's instrument contatinral six subhsenies whtchi werer re~tained in an

effort tor broadly measures knowledger of Africal. tlhese six subhscales were

(a) physical geographyv, (b) histo~ry bcfor-e IEruropennrl penetratioln,

(c) history of Euiropenns in Africa. (d) indigenouls fiociety, fe) rconominc

development, and (F) cur-rent affairs. Thliis orinilnni Iocumrent was used

by the University of Tllinois Program in A~frican Studies in a pre-post

design to test knowledge gaiin at their innrytrvi e woriknlhop for teachers

in 1974 (Schmidt, 1974, p. 17). Beyer uisedl a 15-member panel to sift

through 120 multiple choicr itrims and enime uip with 70 itemns to~ be itemr

analyzed. O~f thle 70, 60 werre chosen for thle final instrulment. Using anl

add-even spl it-hal f technique,, Boyer foulndl a .60 rerl inh~il i t coefficient

for his seventh grade sample and a .80 conrfricintt for- the twelfth girade

sample (Reyer & Hicks. 1968, p. 22). Althou~lGh IFyer alnd Ilicks (1968)

did not specifically report the number o~f sub~jects inivolvedl in the










reliability study, thley reportedly clsewherr in the~ir pre.Selntatlio thatl

845 seventh grade~rs and 794 twelfthi gradiers wer-r slrvleyed in their

study (p. 5).

Three facts suggested that the Beyer and tlicks test neededl to be

modifiedl for- use withi techellrss in thiis stuidy. The fir-st was that the .60

reliability coefficienlt with thle seventh Crader sample was too low to be

acceptable. Second, since thle test wans developrd in 196,8. events bothi in

Africa and in the field of African scholaarshipl have rrndrb edi several

items outdated. Thirdl, since~ this instrumnn~lt woullld he usled in co~njlne-

tion with an attitude inst ruimprt to evalulate oit reach ] workShop)s, a

shorter test was desired.

Consequently, the original 60-item instrument~t wa:s sublmitted to a

panel of 13 A~friennist s aIsso~ciatedl with ther (c-!nter for Al rican S~~tuies

at the Ulniversity of F-lorira i, evalrnto the colnte~nt validity of the

items. TheseF scholarTS representedtr thle following: disciplines: Engylish,,

geography-, bEhanvioral studirs, cllrriculum an~l d ins~truc~tion, solciology,

art history, French~, anthlropollogy, food ;Ind resoul~r econoimics, history,

linguistics, and comparative educational. All professors we~re asked to

circle the letter of the correct answer as they snw it from thle per-

spective of their ex~perti~se on Africa and then to comment on~ questions

which they thought were poorrly wo~rdedl, deltn with trivia, were ouitdated,

or in any way needed adapting before beinG uisedl wi th teachers. On the

basis of thle responses recrtvell fromn thsr IT experts on Africa. 28 of

the items in the original test were disca:r~lded rtri o

elimination of items included d thle foll~owing: rQuestions basedl on no

longer accepted assumptions (i.e., the correctness of W~estern-style

economic development as a modlel for Aflrica); qurestions calling for value









judgments; qulestionss containing: negative ~i;nses towairds Alfric:a; questions

which were confusing, simplistic, o~r container d imnplnnible-h diqtractor r

(incorrect alternatives): or quecstionsR which were daltedl and Ino longer

valid. Under these criteria the subscales on economic development and

current affairs were especially hard hiit as 60) parents of the former andl

71 percent of thle latter were eliminated from the oirginal tcst. Elevenl

other questonns from the original 60 were slightly mo~dified to bring,

thiem up to date or to chanGe a single implaunlsible~ <1isractor. The outl--

line map of Africa used toi .Inswer the first six items in thle origiinal

test was modified by addling: the four major rivers ofr Afric.7 -- Nile,

Niger, Congao, and Zambezi -- to, the continenital oultline.. Eleven new

items were written For the revised knrowledglle test, Iresultinc in a 43-itrim

instrument to be pilot tested.

The pitat test was administered to 1410 undtergi-.~r lluates at the Ulniver--

sity of Florida. Eigihty-fourl students enrolled ini the Coillegr ni

Education took thle test dlurinR fall quartrer 19761. Anir idditionail 56

students took the test wintter qluarter 1977 dluringi thle first class meetings

of two different introductory African studies courses. These 140 students

were dividend alternativelyp intor two girolpss of ln tor ni low an item

analysis and a double across validation to be performedc'. Thie biserial

correlations between item sncre and total test score were uised as the

indices of item discrimination (L~ordl & Novick, 1968, Chapter 15). The

35 items with the hiighest hiseriail correlations based on submanmple one

were cross validated on subhsample two. A~t the snme t ime the 15 biest

items revealed by an item analysis oni sulbsample two wer-e crossR validatedl

on subsample one. The 30 items which had thle highest aInd most stable

biserial correlations across the two samples were sele~.ctd for thle Einal










knowledgec test. The onnly exception to th~is strict pswholcmetric criterann

was that one item pertaining: to Zaire was diroplped becanuse of thle larLun

number of items alreadyq referring to Zaire and ;no iteml which~ had a toweri,

but acceptable, cross validat~ion biscrial correlation wtith thle total score~

( r= .31) replaced it to keep the origiinil subsolee percentagies tht~n

had been established by Be~yer and flicks (1968, p. 23) Basedl on a new

sample of 131 subjects described ab~ove in the section on reliability, th~e

biserial correlations for the Final 30 items rangedl from .1A La .76 (see

A~ppendix B).

The survivingi 30 itcims we~re given a final ichecrk forr rlntant validity

and proper wording. Thlree items were alte~rcld at th~is p~oint. Thle re.-

sulting instrument consists of 11 ifems takenI unaltered (except for-

changes in the map ulsed to answer some physically genernphy questions)

fromn the original 60-item trst. Teni other items T(lreprSent slight modi-

fications of original quefstions and nine itemis are lomrpletely new ones

wIdtten for the revised test. Consequlently, the rev~ised test consists

of approximately one-third completely new items, one-third ulnchanged

items, and one-third modify led items (see Appendix D)). To overcome any

effects of a response set, the correct answers were arpportio~ned as evenly

as possible among the Eour niternatives. The total scale scor-e coin-

sisted of the sum of correct answers.

The second instruiment uised in thle validantion prncess wars aIn already

constructed L~ikert-type scale known as the Wrorldminded~cness Scale (W-Seale)

(see Appepdix E). This scale was Ilevelopei as p.art oF a studyv to lernr-

what types of intercultuiral e~perienCes haive ai Spec~ial Empallct on

attitudes, the relationship between personallity differences and world-

mindedness, and what relationship exists between attitude prior to an









intercultural expecrience anld reactionl to that expcr ienice (Smnithi, 1935).

The Worldmindedness Scalle hans eightL subscales thant mear~sure attiltudes on

the following: dimensions: religion. immiyr-.tion. !:overnment, econnomics,

patriotism, race, education, andi war (Shaw &r Wright. 1967. p. 203). In

validating the scale Smiith (1955, p. 470) re~porterd thrse correlations

between the Worldmindedlness Srale and thie following: construdts: Ethlno-

centrism -.71, Facism -.46, and Polittenl andI Eco~nnmii n Co~nservatism -.51.

Smith~ found thaRt unstutrurcture heterogecneour s intercultural experifiene hnd

little impact on entering worl~lmindedness. Th1e tendeniicy was fo~r people

on both ends of the Wdorldminldedne~ss Scale to have their attitudces

strengthenedl by initerculturall experiences. fe oprngWrdidd

ness scores to results on factors mankingi up Guii lford anrd Guliltood's

(1934. 1936) int rovers ionr-pt reversion scaleSmlj th (1'155) c~onci ndedl thnt

the "hfihl~y worrldminded person is less mnasculnfine ad lS~cendantt more

impuilsive and emotionally decpendernt, and mo~re ine~l ined toward intro-

spection andl internalizaltion of impulses thnn the \-rry nationallistic

individual" (P. 476). Salmpsron and Smith (105r7) decfinn worldmindedness

as "a frame of reference, oi value orientation, faivoringi a world-vieww of

the problems of hulmanity, with mankinld, raither thain rhi natio~nals of a1

particular country, as thle primary reference granlp"~ (p. 105). Thel~y

emphasize that worldmindledness as they conceive o~f it dlesignaites a value

orientation, or trame of reference, apart from knowslelpr about, or

interest in, international r-elations. SamTPson and ISmith, report a

corrected split-half reliability of .91 and a tcst-rlellst reliability

also equal to .93. Allmnii (1961) uised theF Worldminir ed~ness scale to

compare differences in thle stud~enit bodies of two schooills. Ga rrison

(1961) used thle scale to compare diifferences based on re~ligion, region









of the country lived in, famiily backgrondnd sex, anid yeiar in college.

Based on hlis findings at the University of Georgin, a male frrshman

Baptist from an agricultuiral backgrounds in the Soulthea~stt woulld h~e the

most likely ennididate to be low on wo~rldminldedness. Newman and Wanre

(1976) found a low positive relationship, (r- = .35) hetween worldmindedne~s

and aesthetic perception.

The Worrldmindedness Scale was chosen for uise in th~is study for

several reasons. Not only dioes it hiave solid credrntials for rettability

and va lidity but i t also has a timely quanl i t not shaired with other scal es

aimed at measuiring, internationaolism. Manny scales dcveloped in the

1940's and 1950's contain itemrs which are noi longer- unlicl becalsee theyv

refer to personalities or specific~ situntions which would niot he Eamilialr

to subjects in the 1970's. Tlhe WorlIdmin ded ness Sc; e. how~ever, con ta7ins

items which refer to more gerneric situations which an audirience in 19)77

could r-elate to as easily as one in 1957 whe~n the s:ale was developed.

In addition, thre Sampson and Smitth (1957) scal~e attempts to measure a

broad range of dimensions related to worldmindednr4s. Mlany other scales

are limited to a narrower range of attitudle objects -- fo'r example, war,

communism, patriotism, or traide policy.

In scoring the Wonrldmindedness Scal.c, thef proc-edures outlined in

Shaw and Wrighit (L967, pp. 2031-204) were followed. However, sincp some

item responises were mission ai preliminary step was taken. Foir each

item a group mean was gefnerntte hnsed on the~ p~ilot sample oif Ill teachers.

In order to compute totanl anrd subsorilE series, these item means were

rounded to the nearest integer and substituted for the b~lanks on the

score cardsi. The Statistical Panckage for the Social Sciences (Nie, Hulll,

Jenkins, Steinbhrenner, and Bent, 1975, pp. 181-191) sublprogiram






-43-


"Ccndescriptive" was usedi to comp''te thle ite~m means.11~ The mliSSinlg dalta

problem was a very minor one in this case as only iR datn points were

missing out of a total of 4,192. Total sco~rrs werl- g~lernerate for each

subject by addring the value oIf thle item respo~nsess whichr rang~ed from 6

for strong agireement to 0 for strongly disag~rr eement

Discrimninant validation. The second step in e~stablishingi thle con-

struct validity was to state a hiypothesis about thr relationship hotweeni

scores on the FICAS suibscales of two grouips whose scores should be

different. It was assumed thaRt teachers who~ olectedl t attend sulmmcr

institutes on African studies would have morep positive attitudres toward

the use of A~frican content thann wouldl a crOSs section1 Of te nilerS.

Therefore, it was predicted that thle FTCAS suibscalr scores of thle

institute participants wou~ldi be siginificilntiv hi~iighe thn those of a

cross sec-tion (or- pilot) g~ro'up

After preliminary testing to determine the colrrelations between

the FICA\S subscales, the W'orldmindledness Scalle, anlll the FAKS, the in-

struments were administered to 49 tendhers aittendijn) sulmmer institutes on

African studies at either thle University of Illinois, Champaign, or the

University of South Florida, Tampa. Twenty-eight people toolk the thrEe-

test battery on Juine 20, 1977, in Tampa. ITwenty-o~nr took the FICAS and

FAKS in Champaiign on Jlune 9. These protests, given at thie beginning orf

the respective summer institutes, providledl Janl on the Africa attitude

scale to be used in the group differences construct validation procedures.

Bloth i ns ti tut es were s imi lar i n the ir gioaIs Eanch inistituIte sought

to increase teachers' subhstaintive knowledge of Africa aInd Africans, La

introduce teachers to new sorclres of information aboullt Africai, and to

assist teachers In develo~pingi cuirricuilum plans and teachinfi skills









appropriate to integrating: the studry of Africa in~to the~ public school

curriculum. The Champ~aign institute spolnsord hv thle University of

Illinois African Studies Pro-gram under a granlt from11 thl Naitionlal

Endowment for the Hiumanities lasted fouir weeks from Julne 6 to Jully 1,

1977. The Tampa institute~ wais sponsored jointly hv thle Departments of

International Studies and Afro-American Stuldies andii ther (`ntor folr

Economiic Educ~ation at the Uniiversity of Souith F~loridR andl the Center for

African Studies at thle University of Florida.. It lasted for two weeks

from June 20 to July 1, 1977. Each~ workshop Provi~ded participants with

a variety of resoulrces on Africa: Amongi the~se w~ere films, lecture/

diiscuss ions, meet ings w ith ~f ricans s i de-l ec tuIres, anid r~nd i ngs. Ln

addition, the participannts in each workshop were to~ pr-oduce cllrricllarr

materials which were to be implemented in thle schnotr~ veni 1977-78. At

the conclusion of each workshop the FTCAS andn FARS inst~ruments were re-

administered to allow the measuringj of change o~n thle two scales as a

result of workshop participation.

In testing the hypothlesis on giroupl diffcrences~ th~ FICAS subhscale

scores of thie pilot sample were comparerl to those of thie participants

in the two summer institute on African studieS. Attenldingi thle inStitulte

in Tampa were 28 teachers (.!6 fro~m Floridan, o~ne fro~m Lourrisiann. and o~ne

from Georg~in). Twenty-two wcre. white and six were ~lalck. NearlTy, halfl

the teachers, 12 of them, ha~s a combinedl middle schlroo-high srhan1l

background. Five more were exclusively high~ school treachers and four

were excl1us ivelIy middle school~ I/j un ior h1i gh ten;chers. only one teachers

had a comlbinedd middle schan1-elemeenntar backg~rround. Six other teachers

hiad taught at all levels. Mo~re than twice as many macnhers (10) had

taught from 6 to 10 years as had taught any other length of time. One





-4ri-


was in his first year of tealchingi: 5 had t-augh~t 2 ior i years: 5, ii or- 5

years; 3, 10 to 20 year-s; andi 2, 20 years or- molri. AsR hnrl been prediictpd

the majority o~f institute palrticipants were snr-ial studries teachers; 17

exclusively sor and seven more taught social studies and some other

subject. Two teacherss were art tenichers in the humultllliieS entagOry and

only one hiad experience ouitside of social studlies or humannities.e

Thre sample attending thle Champaign wo~rkshop consisted of 23 Nissandr

teachers. Of thiese, data exist on only 21. As oppoised to thle Tampa

group, the Mlissouiri group was predomirinately blacrk. 13~ to eight. This

group also contained a Ilrgepr percentage of elemecntary school teachers:

Four were exclusively sol andi one hald a middle~ srlhool-elementary backgjrolnrl.

Two were exclusively juinior h~igh/middll e school techrrs and six were

exclusively so, for high schant~. F~ivr hand n combinled mnichl10 shool-high

school background. Thle thlrre others had tilughtl at :Ill levelS. TIhe

Missourii teachers also tendred to be more expcriencdc~ thanir their souithern

counterports. Nine of the 21 hiadl 1-aught from 10 to~ 20 Yeairs; 'r, 2 or 3

years; 3. 4 or 5 years; 4, i) to 10 years; andN 1, more thanll 20 YeTrs.

'The Missouri group were for the most part also social studies teachers:

12 fit this description, five of whiom had hand experiernre in another area

as well. T~wo had taugfht only basic elementary Sub~jncts. Three had

taught elementary basic siubjects plus something rlse. TIhree were

humanities teachers and the Inst participant hadl experience ini an aren

outside of social stuidies or humanities.

Change over occannilsios Yet a thirdl methodl~ to- establhish constr-uct

validity is to experienltallya)\ manipulate conditions inl suich a waly that

a change in the construct undel~r studry would occur, thien measure that

change. In thiis study thle "experimental manipulantions"' were summer






-46-



institutes dlesignedl to increase teachers' knowledevi~ andl understalnding: of

Africa as wjell as to assist them in incorporlntin A1Frironn conltentt into

their own teachingi. If arttitude re flec ts ai pred isposi fitio to acit toward

an object in a positive or negative manner, then thp instituter experience

should improve the attitude by increasing teachers' desires to include

African content in their tec~ihingi. If the FTCAS Rlubscales couldl measure

the predicted significant rise in attituide preF to po~st as a result of

attending the summer instituites, then another confirmation of construct

validity would have been found. To this end the FICA~S was readministered

to the institute participants at the conclusion of the workshops so that

posttest scores could be comipar-ed to pr-etest senrrs in order to analyze

the ability to measure channge.

















CHAPTER FOUR
RESULTS




The Florida International Curriculum Assessmenlt Scnle~s (FICAS) were

validated using two other instrumments and two separate sam~ples of siub-

jects testedl on three different occasions. Staitistical estimates of

the relinh~ilities of these instruments and their sublsenles are presented

inl this chapter followed by a discussionn of thle sample distributions.

Final~y, results of tests of the experimental hypot-heses designedl to

establ~ish thle construct validity of the FICAS aIre Iceported..



Reliabilities


In1 this study Crolnhachl's (1967) nip~hn, a memos~re `f inlternal con-

sistency, was used to estimate reliability. According to Crollnbach.

"alpha is . an estimate of thle correlation expected between two tests

drawn at random From a pool. of items lIke the items in this test" (p.

141). Alpha is the average of all the possible split-half coefficients

for a given test (p. 11~5). The formula uisedl to estimate alphan in this

study is a generalization of the Kuder-Richardson Fonrmu~la 20 (Veldman,

1967, p. 173).

Thle internal consistency coefficients for thle FA\K; andl FICAS are

displayed in Table 2. For F-AKS, only the total scaile score had an alphaR

of greater than .70 based on the pitat samplre ofT 111 subhjects. CoEffi-

cient alpha was .83 fo~r this group. Thlis speci~fler re]liabillty (L~ord &





Table1 2

Reliabilities and Standrdnr Erro~rs for the Florida~ Internation al
Curriculum A~ssessment Scales anid thei Floriian Afrienll
Knowledge Scnle basedl on the Pilot Sample (N = ~11)


No. of
Items

F1C:AS


Cofffircient Standalrd
A2 lpha Er rors


Scale


Cognitive Nationalistic

Cognitive International

Cognitive Either/or

Affective Nationalistic

Affective International

Affective Either/or

Cognitive Total

Affective Total

Nationalistic T`otal

International Total

Either/or Total

Total Scale





Physical Geography

History before European
Penetration

History of Europenns
in Africa

Indigenous Society

Economic Development

Current Affairs

Total Scale


CO;NAT

COI)( NT




AF'NAT

AF[FINT

AFFE[OR

CO(:TOT

A\FF FOT '




TfINTOT

EORTO 1'

AFRTTOT


.81

.58

.96




.42

.83

1,97

1.58

1.35

.93

1.75

3.46


FAKS


PH~YSGE:O 8



PREUlRHST


EUROIllST

INDSOC

ECO~NDEVL.

C:URAFFA~

KNOWITOT1









Novick, 1968. Chapter 91 re~mainedl hiigh whcn the teci was adlministere~d to

the 49 subjects attending the summeifr instiltuts. Thei r pre tests y ie lded

an alpha of .77 and their posttests, .78. The FICAS hais hiigh reliabiity

as estimated by the alphan coefficients. Thei coeffricir ent alpha values of1

the 11 subscales ranged from .78 to .93 (sec Tablee 2). 'The internal

consistency estimaLes of thle NATTOT, INTTOT, and EORIKIT subscales were

.92, .87, andl .93 respectively. Slightly hlihelr coecfficients wre~r

observed as a result of administering these scales to the institltce

samples, ranging from .91 to .94 on the protest annd fr~m .92 to .94 on

the posttest.

A~n alpha of .85 was observed for thle to-tal WorldmI~nindednent ss Scale

based on the pilot sample. For- the 28 participants inl the Taimpa Inisti-

tute the coefficient of .83 was computed. As with thc FnkS, thle W'-Srale

subscale reliabilities were ton low for further annnivses Since thie

reliabilities of' the subscales were all below .7i (see Danvis, 1964,

P. 24; Tinkelman, 1971, p. 71; Fox, 1969, p1. 162), o~nly the tiota IJ-SIcale

and the total FAKS scores weire used in attempts to valida7te the FTCAS.

The results of this reliability study revealedl that eachi sub~scale on

the FTCAS was reliable enoughl to be subh jectedl to valIidiLy checks. How-

ever, because of lack of inidependlence between some ofT the 11 sub~scales,

some scales being composites of others, and high mullticaol ineairityl` among

some scales which were thleoreticlly indepenldent, Ionly three sublscales,

NATTOT, INTTOT, and EORTOT, were selected for thle investiGation of

construcrt validity.




Nlulticallinearity refers to hiigh correlations among independent variab~les
entered into mul~tiple regression (see Blalack, 1964; Gordon. 1968;
Kerlinger & Pedhaz.ur. 1973)






-50-


s~nm~pl Disfiltributioq


Because the hypotheses set forth in thiis studyv rlledl for the uise

of significance tests used withi the calcuSlatin of both Penrson r's andl

analyses of variance, thle distributions to be entered into each analysis

were checked for normality and homogeneity. Alth~oughl muchl literature

claims that the Pearso~n r (Carrol, 1961; Havliceb & IPeterson, 1977) and

analysis of variance F tests (Kirk, 19683; Linlquist, 1953; Pertson, 1931)

are robust with respect toi violations of the assumlptinns of no~rmality of

distributions and homo~geneity of unriance. except vlen sample size also

varies greatly in conjunctionn with discrcpancies in va'riance (Bonieau,

1971), some distributions in thlis study appeared to~ rep~resent extreme

violations of these assumptions. Thereforealll scresrr on the FI;CAS as

well as the W-Scale and FAKS total scores were norma,;lized befr~ie pro-

coeding with the parametric statistical tests of Ilypothe~ses.




TIests of Construct Validity


In the first phiase of construct validation thle three selected FICAZS

subscales were correlated with measures of attitude anld knowledge which

should have had some relationship to attitulde tOWarl inclu1Cding African

content in the curriculum. The tests of Hyprotheses Tiwor ad Thlrefe con-

stitute this phase of validantion. In the second phanse of construc~t

validation the differential validity was inlvrstigatedi to determine if

the scales could be used to distinguiish herween Groups'T w~ho should have

differed in attitude towardl Africa in the enrrrirenklm (Aee Hyponthesis

Foulr). Finally, the FICAS subhscales were tested to see if they could

be used to detect a channge In ~tltitude asH a re~sult of attendingf summer





institutes on African studlies aimed at increasing rincheir compnetency in

uising African content in clalssroomn teaching: (see lypo~theesis Five).



Hypothesis One


There will be no significant differences (p .2'1) amo~nF: the mean

scores of subljeCts in the pilot sample assiigned tor take threec different

forms of the Florida Internaitional Cuirriculuim Assessme~nt Scales.

Because the possibility ex~isted that thle rating: of the relevance

of African content might he affected b~y thle co~ntpxt within which the

rating occurred, three forms of thle FICAS were tested. On F'orm A respon-

dan7ts were asked to rate the relevance of studyir ing Eur-openn peoples andl

places in Raddition to rating the relevance of studyling: Africa~n peoples

and places. Form B included In~tin Amnerica along wit~h AFricai. On orm~n

C respondents were askerl to raite the relevancr of Afrriien conltent only.

Thirty-three subljects were assigned at randolm toi oneir of the thiree test

conditions. These subjects were the first 33 to rtpondt to the instrll-

ment from the pilot sample described inl Chapter Thr~ee

Because thle researcher expected thle null hypothesis -- M : X =

X2 = X3 -- to be confirmed, a liberal alpha level of .2!5 was chosen

increasing thle chances of a Type T error, and thus redulcing: the chances

of a Type IL error. The observed difference in meanis wa~s tested against

the expectation that it woulld occuri by chance~l at lest 25 times in 100

before the researcher would accept thle nlll hypothes~ris A o~newny analysis

of va~riance supported the nall h1Ypolhesis. T'he obseilrve F' ratio, .71,

indicated that the differences in mneais woulld occurl 50 limps in 100 by

chance aloner (p = .50)










~Table 3

Raw Meanrs andl Stancllrrl Datavitionsl
for- fypothesis One (N = 11)


Treatment M~ean Standard Deviation


Form A 145.45 33.16
Form B 144.00 35.58
Form C 128.72 78.50





H ~thsis Two


There will be a significant positive correlatiorn (p < .05) between

scores on the F~lorida Interna~tional Curricullum Assessnolnt Scales and

scores on the Wiorldlmindedness Scale.

11ypothlesis Two was tested uising th~e responsesi of thle pilot Rroup of

131 experienced K-12 teac~hrrs. This group consistled I a crossi-section

of teachers representing all1 grade levels andi sub jlact areas. Becausef

multicallinenrity can enuse problems of interpretation in multiple regres

sion analysis,2 the six indF~epdende subscales making uip the total FLCAS

were reduced to three composites. Because there were three subscales of

interest, multiple regression was uised to provide an o~verall significance

test of the relationship betweenl the three FICAS~ sub~cales and thle

criterion of worldmindedness. NATTOTr, INTTO3T,, and ORoTOT had a multiple





For a djSIiscussin o the problems of multicrl linenrity,. see Blalack.
1964; Dalrlingiton, 196;8; Carrdon,, 1968; Johnsosn, 1q72; Nie et al1., 1975.
Tlhe researcher opted to, crrnte compositesi rather than eliminate some of
the highly correlatedl subscales. COGEOII andl AFFFOR (rxy = .84) were
co~mbinedl to form EORTOT; COG:NAT and AF1FNATr (rxy = .8 4) wrre combined
to form NATTOTI; and COGINT andi AFFINT (r,, = .72) wrFIe combined to form
INTTOT.









correlation, R, with thie total Worldlmindedlness Scaile oif .28 (withi


F3,127 = 3.47, tl < .05). Since thp overall F ratiio was significant, theI
indi vidtual Pearson-producle t moment correlit ions~ between the :T CAS subh-

scales and the Wi-Scale to~tal were examined to decterminle which ones

differed signiificantly from zero in the expected direcrtion. As hypothe-

sized, both, INTTOT and EORTO(T hand signi ficant posi tivc correlations withi

worldmindedness. Their corrections were .26 and .21 (p_ < .DS) respec-

tively. However, thef correlation between NATTO(T and the W-Scnle total

was not sigrnifieantly greater thann Zero. Conslequenrtly, for Hypothesis

Two there is some evidence nf construct validity for TNiTTOT Hnd EORTOT

but not for NATTAT.




Hypothesi~s Three


There will he a sifnificant positive correlation (p .05) hetween

scores on thec Florida Interinational Curricuilum Assessmeinnt Scales andi

scores on thle F~lorida African K~nowledge Scalr.

Hlypo~thesis Three was tested using the same pilot samnple of 131

teachers that was used to test Hypothesis Two. An R wais calculated

between NATTOT, INTTOT, and EORTOT on the one hand and the FAKS total

scale score on the other. T'he observed value was R = .2'3 (F = 2.17,
-3,12;
not significant). No significant relationship was found between thle

FICAS subscales anid the test of African knowledgeF.



Fly~potheisisFour


Subjects who elect to palrtjicipate in a cuirriculumn w~orkshopi on Africa

will have a significantly higher mean senre (p < .05) o~n the Florida













RRaW M.eans nd Standard Deviations ofalll Simples anld Variables
Entered into the Anailyses o~f Hypothe1SSe inr thlroulGh F'ive


Variables

E:ORTIOT

aInd llThree


NATT`OT TNTITOT

Hypothe~ses T~wo


281.14 41.79
15.43 10.59


Group




Pilot X
(n =131) Sx


IWORLDITO'T KNOWJTOT'


h'1.36 92.91
20.05 22.54


13.94
5.837


Hypothesis Four


Pilot
(n =131) 5

TampR Prpb r
(n =28) E

Lllinois Pre )
(n =21) E





Tampa P'reb X
(n =22) S

Tampa Postb x
(n =22) S

Illinois Pre X
(n =20) S

Illinois Post X
(n =20) S


ab~ssing values

Missing values
mean (n = 222).


28.18
15.42

35.82
I6.11

39.52
L3.21


43:.83
10.59

51 .00
11. .54

5_5.24
8.10


64.43
20.06

76.93
21.78i

83.00
15.76


Hypothesis Five


S36.16 51.05 i5.71
x16.61 12.60 21.69

S37.68 53.91 83:i.50]
15.49 11.06 16.66

S39.75 55.05 R2.9)0
13.51 8.26 16.16

S38.65 55.35 84.65
16l.30 h.07 12.L3


were supplied based on pilot jiroup mieans (n


were


131) .


sulpplied honsed on pilot and inlStitultE giroups grand









International Curriculum ARSsessment Scales th31n ai riomlpurti o group OF a

cross section of teachers.

The next step in establishing the constrruct vilidityv of the F'ICA\S

was to demonstrate that a significant dirfcferece existedd between the mean

scores of two or more groupsn whio theorceically shoullld hav1e differed on

the scale. TIhe two groups selectedl in this study wecre (a) the pilot

group and (b) the combined institute giroupS. TrhE pilot grollp was a croSS

section of teachers whio happened to be available for ttesting. Thecy wtere

assembled to take coursrwork whiich in no wayv was related to internationniil

education or cross-cultulral studies. The IillilsboroughI Coulnty suibsample

was composed of social studiies teac~hers who were selected hernllse their

availability was unrelated to internatioionu ediiention issues. On thec

other hand, the institute samples were composed of tow;lhers whlo were

self-selected to participate in summer insatuttes o~n Afriian studiies.

They could be presumed to have a mo~re posjtivl nt itlde to~ward teachingi

about Africa than the cross section of teachers in the~ ptilnt group.

Th~erefore, if a significant dlifference- in mean scores between these two

groups could be detected by) the FTCAS sublscales, 'lhen terce won~d be some

evidence for the validity of these attitude mealsurles.. Because the two

institute samples came from diffe~rent areas of the conirtry, they were

treated as separate groups in the analysis. AZ discrimlinant analysis was

performed with the three relevant FICAS subsentell s b~einX uIsed to diS-

criminate between the pilot~ groulp and thie two inlstitulte groups. A post

hoc orthogonal planned comparison (Kirk. 1968, pp. h9-78) wa~s used to

test whether the means of the institute giroups lirffered From each other

and whether their combined means differedi from thef pilot groupl mean.

With three groups, only twoi (k 1) orthiogonail pallanned comparisonss were









possible (Kirk. L968, p. 72). However, withi thpse two, complarisons the

researcher was able to chtniin all thle infoirmationn nee~ded to test

Hypothesis Four. The comparisonr s ulsed can he symbli; 1:. l in thiis wany

(Kirk, 1968, pp. 69-70):



i1 = X2 X3


xX?




The first of the two compar-isons testedl whether aIny significant di~ffer-

ences existedi between the inlstitulte groups on thp r-elevant FICAS sulbscales.

The second comparison tested whether thle pilot giroupl rrain was siginificantl y~

different from the average institute g~roupi mean.

Wheni NATTOT. INTTOT, anrd EORHTOT were entered inrto a direct dlis-

criminantn analysis one significant discriminant furnctirion was produced.

The standardized discrimiinantt function coeficiients for this function

were .76 for LN1TTT, .28 for NArTTO)T, andl .09 for- 0I~nTlrl.

After determining that these thlree suibscales do discriminate betweenPI

the three groups in th~e analysis, post hoc analyqiis of variance employing

planned comparisons was usedl with each variable sepairately to test first

if there was a significant difference betwpen ther means of thet institute

groups and second if there was a significant difference between the mean

of the pilot group and thle combined institrlte groups. Thel overall F

ratio for each variable was siginificant at belo~w the .05, rriterion alpha

Icvel. In addlitio~n, on nio vairiable were th~e means of thle institutee

groups found to be significantly di fferent. However, o~n echct variable

the mean of the pilot group was found to bre signiftcantly flower than thie

average of the two Institute groups means. Hyp~othesRis Four confirmed









that INTTIOT, NATTOT, and I:ORT`OT cachl measlredl the~ expected differences

between the piilot group and the instituted C1roIPs. Ytt the~ two institute

groups, each of which consisted of educators seflf-selecteed to participated

in an African studies program, did not differ froim achcl other-.




~tepotess Five


The. mean posttest score of a giroupl of teachers riinrlled in ann Afr-ienn

studies center summer inlstitulte will be signlificantly (p .05) higher

than their mean score onl a pretest of thle Floridal International Cuirriculumm

Assessment Scales.

Hypothesis Five was designed to see if ther ne~w attitudelF scales could

detect attitude change as a result of summer inst i tute partic~ipAtion.

A multivnriate analysis of variance rep~eated measures IIesignI controlling

for site was developed to test if a changije in attitude1 couldll be measuredt

by the instrument. This anailysis was followed hv post Iir split-plot

analysis of variance withi unequall nI (Kirk, 1968. pip. 2761-279).

When all three vnriables wrer entered into a muiltivariate analysis

of variance an F approximation of thie Pillat-Bartlctt trace statistic

was not significant at the .05 alphan level (see ol son. 1976). Conse-

quently, it must be concluded that the tlree subs~fcales did not distin-

guish between pre and ponsttest scores of inlstitulte pairticipants.

Thefrefore, [Iypothesis Five was not supported.i















C:HA\P~ITE FIVE~
CONCLU(STO)NS AND) RECOMMEIINDAlTION S




Int reduction


Two su~bscales of thle Floridn International Cur~riculunm Assessment

Sales have been shown to possess su~fficient reliability and validity to

be employed in further researchI andl evaluation studies. Bothi TNTTOT and

EORTOT demonstrated the Predicted relationship to wa,~rllminll~dredss and

distinguishied successfully between~ the pilot andi institute groups.

NATTOT has served a usefull purpose in this va3lidantion study buit aIppears

to be of no further use in measuring tEa1Cher'S attitudelS tolWard uIsing

international content, Aflriinan or otherwise. In thiis c-haptrer the

psychometric quality of the FICASS is disculssed,. Srlfiestions for fulrthefr

research and applications ar-e also presented.




Reliab-ility Comlparisonss


Before focusing on th~e FTCAZS, the reliability daita observed for the

Floridla African Knowledge Srale and the W~orldmnindedr~ness Snl~e will be

compared to thiat reported elsewhere in thie literature. Foir thle pilot

sample of 131 subjects nn alpha of .83 w~as observedl for thre FA
subjects attendling the suimmler institutes vieldled anr alp~ha of .77 for the

prerest and .78 for the poisttest. These figuires compared favorably with

the reliabilities reportedly by Boyer and 114cks (1968X. p. 22). 'They used

an odld-even split-half techiniquie to compute a1 cocff icintit of .60 for









their seventh grade sample and .80 for their twel Eth g:rade samnple.

However, reliability is a fuinction of test length anld the Beyer andl

Hicks instrument was twice ns Long as the on~e refjinal in the current

study. Later Beyer and Hlicks (.1970) used their own 48-item revision

of the or-ig-inal inst rumen t i n a s tudy wh i h p rod ucedr Kcller-R Kichardson

reliabilities of .73 for the protest and .88 fo~r the posttest.t

As a1 by-pro~duct of vAlidating the FICAS, the 10-i'cm FAKS has been

produced. Its reliabilities are comparable to,, if not sulper~rir. to on::pr

versions of African knowledge tests producedl by earllier stuidies. The

content validity of the FAK~S was assured hy submittinG the itemis poten-

tially to be included in ther instrument to a panell of IT~ Africanists

associated with the Center Ear African Studies atC the Unriversity of

Florida. Items which were danted, poorly wourledl, hiasell, or otherwise

vulnerable were deleted from thle item annalysis aInd cross validation whlich

produced the finial 30-itcnm scalee.

For the W~orldminidedlness Scale, the reclinlhility- enefficie~nts obsiervedl

in this study were slightly lower than those reported b~y Sampson and

Smith (1957). They reportedly both split-hllnl and test-retest reliabilittes

to be .93. For the pilot saimple in this studiiy ain nlphni of .85 was ob-

served. For the 28 participaints in the Tamnril Tnsti~ltut thle coefficient

of .83 was computed. The eight four-item sublscnles on the Wr-Scale hadl

relatively low reliabilities among the pilot sample ranriingi from a .31

for thle "Education" suibscale to .62 for thr "Patriotisrn" sublscale.









Discusionjos ofT FICAS Validity


INIT`OT



Among the 11 subscales of FICAS with potential fo~r use inl measuring

teacher attitudes toward including Africanr content in thle curriculum,

only INTTOT and EORTOT emerge as reliable, valid. and independent men-

sures. N"INTTT consis ts o~f IT entrrii curlr orjer t lives requiiri ni i nterna tional

content to achieve (i.e., "Sutudents should decvelop a sense of helongingf

to a world community" and "Students should ulnderstanld dliffrernt ways of

life o-n dlifferent contingents"). Teacher's INTT~lOT crores co~rrelated .26

with their Worldmindedness totAl scale scor-es. Thiis enrrclation was not

only sta~itistilly significant buit of an necepltablle magnitulde in spite

of its mlalest size. The W-Scale wais designed t, mrasurel nationailistil:-

internationalistie attituldes. Through it was theorrized thant teacherles wiho

were positively disposedl to uisinig internatiornal rcontenit (sp~ecifielivll

Af rican con tent) in thei r t each ing woulld hF hIighelr -i n wo rldmini dctnrss

than those not so disposAEd, a highl correlation beftween INrTTOT and the

W-Scale would not be expected because the two scales~ wre~ intendecd to

measure dliffEerent construct-s. A\ general worldmindeldness is dlistinct from

an inclination to uise international content in one's traching.

Althiough no relationship b~etween TNTTOTrn scores and FAK;S to~tal scalle

scores was observed for the pilot group, there is prl-cedencrce for suchi a

finding. Shrigiley (1974) found n low correlation between knlowledge of

science andi attitude townrd srcience among pr-rserviice tceachers. 110

questio-ned the existence o~f a positive relatlionshiip beftween thle two

constructs. Perhanps the culrrrllnt researcehr was expcting. too much to

expect a positive rel~latonshilp to exist betwrsrn knowlrledge of Africa anrd





-6L-


attitude toward African contentll in thle cur-rienlum.r n Anolthler possibility

also exis~ts. TIhere may' be .1 nonlinearr relationships bet~ween~ knoledge1L: of

Africa anid attitude to~ward African contrnt. Th is Ioss ibil ity wi ll he

taken up later when fuirther- research is dliscused.

Of the tested subtstales, INTTOT'I had the~ highest standardized dis-

crimiinant fulnction coefficient on the function dliscriminati ngi between a

cross section of teachers ilnd a grouIP whiChl w~s sel--lf-slcted to attend

African studies summer institutes. Theorrticallv teaichIEIS who( Would

choose to attend a summer inlstitulte inl African studlies should be more

inclined to use African conltent to achitne international obljectives thann

a more general population of teachers. INTTOT' clearly measures this

expected difference.

That INTTOT failed to measure a pre-past lifferencel~ inl attitude

among summer instituted part~iripants might be explained in neverni wiys;.

It is quiite possible that no, change occurired in the alttitudef measulred by

this scale. Strength is added to this explalnalti On1 b the PrevIous

finding that institute tachclers are relatively h~irgher in their INT'TOT

scores compared to the pilot sample thani they are o~n thle other twio scalles

being tested. Their initial higih scores may have left Little room for

additional change. in addition, the phenomenon of r-rgrission toward the

mean could have plnyed al major role in FINIT~OT's nout mea:surina a changer

in attitudde. It is also possible that INTTUI'T failed to detect a change

in attitude thlat didl occuir amongrn thle inst~itute techellrs. Ilowreverr

INTTOT's power in detecting a difference bet~weeni institutep techellrs anid

a cross section of teachiers makes the explanation of no real change mo-re

p~lausible than that of an unldetetedl real change.









EORTOT



The EORTOT subsente consists of 22 obilortives whiich can he devel~lopd

with either nationalistic or international content, oir both (i..e.,

"Students should understand the balance of rnature," "Studtents should

know about the role of government," or "Stuldent s should accept chiangef

as a natural feature of the human condition"). Siichi obeCtivesS are not

specifically international in scope, as are INTTOT(,'s, but one would

suspect that a worldmindjed teacher would be mo~re inlrlinled to use inter-

national content to develop either/or objectives th~n u~nal~l n teacher

with a more othlnocentric viewpoint. Consequently, o~ne would expect to

find EORT~OT scores to be positively related to thoRe noi thle W-Snent but

not qulite as strongly as INfTOr'ss scores. Sulch wals in factL the case as

the correlation between EORTOT and thle WJ-Scale equa.lld .21 (p < .05).

Althougfh no significant re~lationship was found bectwee~n EO(RT10f and the

FAIKS total scores, furithcr r-esearch maiy show this findingF to resullt froml

a nonlinear relationship, existing between the two' (.nnstruCtS.

EORTOT also distincluishledl between the piilot cross sortion sample

and the institute samp~le. however, because of i ts h i[gh enrrelation with

TNTTOT (r = .72), it adds little information wheln TNITTOT is; also entered

into a dliscriminanti analysis. Conse~quently, its discriminant fuinction~

coefficient is rather 10w when compared to 1NTTOT. Nonellcthelss,. when

used separately in its ow~n right, EORTOT'1 is anl rffective dlicriminnator

between pilot andr institute grollps.

The EO~RTT scale was expectedl to menaur-e the t;tlitudel l variableL mosnt

amenable to change as a result of attending sulmmer institute, A

teachers learned more about Africa aInr developed culrriculair applici~ations

for African content, it was expected that they would see more and more








uses for- Afri can content aImong thle ei ther/or object~i vs Because of :1

nonsiginificant multivalrintc F ratio, the researchers r dlirl rit examine ther

univariate (ratios for NAT10~lT, INTfTO'T,; ll and EORTOT fo ir thle pulrpose~s of

drawing statistical conclusions. However, it should be noted that in

spite of thep mulltivaridte F'Is being nonsiginificant, EoRTOI'sR univnriate

I' was significant (p < .0)5). Suich a result, which con neculr when several

weak dependent variables; are included inl a mat~tivnr-iat anarlysis (see

Uack, 1975, p. 155), suggetss that the issue of EnrKTOT~'s ability y to

detect pre-post differences as a result of summer institute attendance

is still unresolved. TIhough1 thiis stu~dy has not beelln ble to establish

the abi~lityr of EORTOT to detect a pre-post differececc neither has Lt

Inid to rest the possibility of EORTOT's ability to~ do so,.



NATTOT


NATTOT has served a ulseful purpose in thle vail id~t ion process by

being distinct from INTT0T alnd E01RTOT'1. Ilowever, it no~s not appear to

be a valid measurer of teacher inclination to use internationally co~ntent.

This result was expected. NAT'ITOT consisted of 13 c~urricular objectives

which required American content to achieve (i.e., "Studeints should be

proud of our national accomplishments" and "Student-~s should have a

knowledge of social change inl America"). Although somr individuals who

like to teach fromr a comiarattve perspective might~ filnd international

content relevant to achievingR natiornalistic objectives, it was not ex-

pected that a cross section of American school tealrlchern nr teven the

institute participan-ts, wouldl perceive Afriro; n Icontent to b~e relevant

for teaching to national tstic objiectives.









NATTOT scores proved to, have no significant relationship to IJ-Scale

scores. A correlation of .04 was observed!. Neither did NATTOnT sco~res

shrow the predicted relationship to FAKS scores. A-2 p<.5 a

observed where a positive relationship was hlypothlesized.l It appears that

as teachers knew more about Africa they were less EnclinedI to ulse this

knowledge to achieve nationalistic object i ves. Tholrse fi ndings fulrthrc

reinforce the conclusion that NATTOTr) 7 is indeedl measulrinc a traiit distinctt

from that meAsuredl by INTTnT and EO(RTrOT.

White it is true thiat NATTOT was useful in distinguiishing between

pilot and institute groups, its complete resistancr La rhangei as a result

of in1stitutE treatment has added weight to thc argulmenlt that it is d~is-

tinct from TNTTOT and EOnRTOcT and useless as la ensure ofi teacher attitude

toward thle use of Africain !rnntent. Among thec tortal Inumber of institute

participants there was virtually no changefi in NATrl` laf il mean sors pre-post.

In fact, amonga the Illilnois institute sublsample threnr wais a no~nsignifi-

cant decline in Group meanr. Thlis palttcrn of findli ings leads to thle coni-

clusion that NATrTOT is not nI use~ul subscale to meaisure teacher attitudes

toward includling African content in the curriculum.

However, NATTOT has yinyedl a uiseful role in thle vallidatio~n process

precisely because it hacs shorwnn up as distinct from TNTTOTT andl EORTO)T,

One precaution that mo~st Likert-type attitude scanles take to correct for

response set is to word one half of the items negati~~rvel and one half

positively. The negative items are theni refletctd in scoring:. However,

since FTCAS items consist ofT curricular o~jectives it wasB not possible

to word one half of them nrCuntively. COnISequently,~ it wais realssuringl to

find that the nationalistic i tems scattered througfhoutt tbo FTCAS were

responded to differently by the subjects. Ro~th subscalles, NATITOTL and









INTTOT,~ hiad a potential rannge of zero to 65. Among thl pilot groulp the

raw mean for NAT'TOT wals 28. 1 while thle raw mnean for INTITO'T was 43.8.

This difference w~as significant at thle .001 nipha~ lo~Vrl. 1hlis finding

lends support to the argument thlat FICAS and its subhscales were measuiring

muich more than response set.




Mlultip~le Asp~ects of Unidimenlsionall Scales


Thee hlih murlticallincari ty foindl betwe n several sets of FICAS suib-

scales has confirmed the uinderlyingi structure of It ti tude toward African

content in the curriculum.1 In1 alnalyZinfi the open-ienderle qlestionnnares

during thle early stages of the test develop~ment, the~ re~searchefr depterminll~r

that the two major undlerlyingf cimensionss of thie aItetitrile undefr considerni-

tion were the relationshiip ofl Africann content- to ronin Live ob~jectives on

the one hlandl and to affective objectives on the( mother (see Tab;lelt 1).

Consequently, bot, conjintive and affective objectives w ere includedc oni

thle FTCAS under each of the three entejio~riie: nationa~listic, interna-

tional., and either/or. TIhe intercorrel.ations amony: the vertonsi cournttive

andi affective sulbscales indicates that they belong togthier on their

respective scales. Tnb le 5 displays these cog:ni tive-n ffectivr initer-

correlations which show tha:t on the two Risubseale ofE host validity,

INTTOT and EORTOTr, the cogn~itive-af~fctive compone~ts haln .72 and .84

correlations respectively.










fablel 5

In terrr c~torrltos of Coign t ivr ind
Alffctive Subscales Iin FICTAS


Cognitive A2ffective SubhsI.ales
Subscales AFFNAT AFFINT AIFEO)R AFFTOTI


CoGNATI 0.83
CoGINT 0.72
COGEOR I,.84
COGTOT 0.88





Adeq~uacy of INTTOT and~1 EORTOT as AttitudIre caley~


INTTOT and EORTOTr have been shown to possess highl reliahtility and

various degrees of validfity. Buit how do thecy mpasuire up whtEn checrked

argainlst theoretical criterin for evainat~inga goold atltitudelt inst rumecnts?

Gardner (1975) has cri tiqued~ recent research inl :Ittitllli measurementn

pa'rticularly in the area of at tituides to~ward sc icns.r 11 points out

three major defects whiichi he has found andl thant sholdtll he avoidIEd:


(1) scales which lack any discernible uinderlyingf
theoretical c~onstruct;; (2) scales in whichl va~riouss
thieoretical constructs are confouindedl tloge~ther, i.e.,
scales which attempt to reduce multi-dimen~ nsikunil1
attribu~tes to sinlerJ scores; (3) experimnta~rli trc:t-
ments in which ~lthre is little dliscernible relationi-
ship between the experimental treatment applied and
the scale used to measures its oultcomles.. (p. 101)


Both INTTOT` and EORTOT have unlderlyingi thenretien1 constructs. INTTOTl's

items as judge~d by aI panel ofT eight social stuidies and cllrriculumr pro-

fes~sors at the University of Flor-ida relate to teachiit~: ngaout foreign

peoples and places; thley are international content curricullar objectives.

The LatnSI of hIypothesesP coupled( With thet finding thatI iNT'lO1T's mean scrre~










wals s ign tfica~n tly di f lfeen t from NA F`TOT's anld thart the~ I NTTOT()~ NATTOT1

intercorrelaition is: low (r = .35) confirms thant INrnnT is indeed mEasulr-

ing teachers' perceptions oif the relevance of using: African contpn t n

achieve international curricular objectives.

EORTOT's items were jldyedl by the smle eiight-member panel to be

cuirricular objectives specifically requirina neither national. nor inter-

national content but amenable to usingf either or haoth. Figuire- 3 shows

that th~e pattern of intercorrelations amongi NATTOTT EO)RTOT,, andl NTTOTI

reinforced by the raw mean differences between sub~senies, supplorts thec

theoreticnl basis ulnd~r~lying the subscales. NA~TTOT and INTTOrT which

reflect very different subhjectt matter domains arc shownl to be far apart

by both a Inw correlat ion (r .35) andl 1 lInrre rau i tem mieani di fference

of 1.20. EORTOT which reflects a suibject matter doma~in thalt is somewhere

between NTAT'TOT and INIT(~al SinceF it Shollld theonreticallly represent a

combination of the two, is lolndd to fall. between the~ NATTnT and TNTTOT

extremes on two COunits, correlation coeFfiCients andl rnlw itemI meanl dil-

ferences. INTTOTI, EORTOT, and NATT~OT doi thenr a1pponr to represent

distinct uinderlyingf theoretirnl constructs.




Correlations:-


SuibscaleS : NAFTOT ORTT INTT`1OT


Raw item means: 2.17 2.1 .37


H~ean dilffrences: 20.?



Fi gure

Correlation andr Item Hefan Diffe~rence P'Atterns
for NATT`OT, EORTOT, and INTT~OT










Begiinning this validat ion study by examininfi all siz independent

subhscales, thet researclerr hans met Cnrdlner's second crite~ricn thant various

Lbearetical constructs not h~e confounded. Where mullticallinearity sup-

ported theoretical expectations as in thle case of coGnlitivee and affectivec

subscales, scales were combined to include all aSp)ectS of a single

unidimnens ional construlct. In this process NTJ~TOT, I NTTOT, and~ EORTIIT

emerged as scales measuringr separate attributes. As a consequence no

attempt has been made to analyze the FICArS total sco~res. Indeed, thle

FICAS total score, rep~resenting a combinations of NAiTTOT,. INfTTOT, and

EORTOT, is mlultidimensionall anrd as suchi is difficult to interpret. Con-

sequently, INTTOT andi EORTOT() are heing reco~nmmended as separate uinidimien-

sional scales to be used in fuTthler research and e~valuaion.n

The response to Ganrdlner's thiirdl criterion vni les fo~r IN'TTOT1 and

EORTOT in terms of the current study. Clearly IN1TOT is not suiited to

be used as a measure of thef effects of institutes treatments where the

selE-selected particip~ants .Ire already highr on thip scall. Thie case is

not closed on EORTOT. Theonretically EORTOT should be :Imenable to change

as a result of institute treatment. It contains objectives for which

teachers have probably not thou~htt of using inlternational conitent until

they engagfed in an intensive experience to increase their knowledge of

foreign areas and to help th~em write curriculumi projects to incorporate

international content. There would seemn to be a dliscernable relationship

between the institute treatment and the EORTOT,1 subsca~le. it remains to

be shown, however, that EORfoT does indee etl rc~t the texpcrtedl change in

attitude. Gardlner's criteriai for de~veloping I:ood attitudelF 8rcles appear

to have been met by the INTTOl)T and EORnTOT sanles.









Further Researchi



New Test of EOR~TOT's Validity


The first order of business will be to test E0(RTOf's Abil~ity to

discriminate pre-post attitude changes as a result of institute partici-

pation. LL wais necessary for thiis studyv to use a multlivarintr F: test ini

order to simultaneously evaluate thre ability of echi~l ii thle thiree scales

of interest to detect thle experctd changej in tende~lr aittitudte. Tlhe three

scales in combination dil Inot detect a change;; howuever, EnRTOT1 by itself

did. Yet thls latter resullt must be viewedl with sispi ien beenuII se of

the nonsignificant multivariate I' test. TIhei currently stuldy hans eliminated

NATTOT and INTITOT from cons Ede-ration to detect. pre-post8 diff~erenres.

However, it is necessary to test EORTOT, using a ulnivnrinte F Lest. on a1

new institute sample indecpendent of thant used in thiis studyl. If a test

based on a large enough sample is: condu~cted, a nonsi~iicnifcnt result

would indicate that EnRTOT, dloesi not dlistingulis h prec-postt changes. A

Positive result would confi rm expectations thant it does dertect real

channges when they occur.




P~ossibility of aNon-Linea Re~lationship Bc` t weenr ~INETTOTI~ anld FAZKS


The failure to findl a sig~nificant linear relationship between the

otherwise valid FICAS sulbscales. INTTOT anid EORTOT. andl kn~owlrede of

Africa raises questions about the nature ofi thle relationship between

knowledge of Africa and inclination, to Iise At ricaln content in tecnhing.

A nonlinear relationship may exist (see Figulre 4) h~etwteen the two

attributrs. It is quite possible that the teacher w~ithi a little bit of

knowledge might be less inclinedl to use Africani co~ntentl than thle teAchefr





-70-


who knows virtually nothing. 'This phenenolnoln cou11 ll occur bo(rnullSe thle

little bit ofI knowledge a typical tahrmgtnqiewudpoal

come froml Tarzan movies, Idi A2min news repor-ts, or fromn mythological

relics of the colonial era. Such sources of information aIre hardly

likely to, inspire a teacher's uising: African coltlent for iurricuilar pir--

poses. However, as knowledge of Africa increasesR in both breandthl and1

depthi so that teacher s can uinderstand ACr-iro;n soictietie as mankingy a c~n-

tribution to human understaindingi, these teach~rrs wourldl he more and more

inclined to use African content. A preliminaryv Ilok at the danta in this

study lenids support to this theory with respecct to IN'TTOT. For thle pilot

sample whose knowledgec level was signi fienntly Iower thann thle institute

pre sample, there was observed a negative correltion h~etween INTTOT and

FAZKS. For the institute prl Iroulp there was a1 mo<1oral;~ te pcs itIi ve relat ion-

ship. For the institute post giroup whose know~ledG- level wa~s significantly

higher thian the institutep pre sample, an7 even strongei~r positive reJlatjin-

shiip between [NTTOT and FAKhS was ob~ser-ved. A studei~ is neededl to ader-

quately test the possibility of a nonlinear relationship beftween knowledge

of Africai and teachers' inclinatio ns to uise Afrien:n conltent to accomplish

international objectives.





Attitutde Toward









Kniowledige of Africa

Figuire 4

ilypothiesized Relatlionship Between Knlowledgee of Aifrica
and Attituide Toward African Content









Validity Studiies for O~the(r international Areas



In another line of research the Florida Internatnnaln Cir-riculuim

Assessment Scales need to be validated for- international co~ntent areas

other than Africa. The items on the scales, clnrril cullar- objectives,

should be as appropriate for measuringi teacher attitude~ toward teaching

about other areas of thle world as they hanve bee~n sho~wn to be for

measuiring teacher attitulde toward African co~ntent. Sincer n~one of thle

items needs to be changed, thre FTCAS ontly needs to hanve two wolrds chaniged

in its inistructions (Appenldtx C) to alter it to test teacher attituide

toward Latin American, Euiropean, or Asian content. More generic terms

such as non-Western. foreiign, or non-Amertenln could~ also b~e tested.

Although this study focusedt on validating, thc Afrienin content because of

thre availability of data from two African studies summer institutes for

teachers, there is every rcasorn to suspect thatL thle in~strument would hie

useful for other international co~ntent areas -- henirce thle worrd

"International" in its title.




Namological N'etwork


Additional validlity studriies shoulrl be carrird outl to relate scores

on TNTTOT and EORTOT to other constructs and to tecicher performance.

According to Cronbach and Mleebt (1967), to bre scientifical? ly missile

a construct miist occuir in a nomological not, at Irngt same of whose l~aws

involve observables. They define a nomnological network as an interlocking:

system of lInws which constitute a theory (p. 25'7). Thc laiws in a namo-

logical net may involve thlree types of relationshlips. First, observable

properties may be related to each othcr; se~ond,, thefolraticn] construlct s





-72-


many be related to abservablels; and third, different thieoreticall con-

structs may be relatedl to one another (P. 255). Ihis studyl hans beguln

the thiirdl part by demonstrating a molerate re~lationshipl of ther constructs

measured by INTTOT and EORTOT to tha~t menusred by thle worldminldedness

Scale.

Yet more needs to be donne in this thirdl area to more completely

describe the relationshiip between the construlcts measuredl by INTTnT and

EORTOT andr other constructs. One potential fruitfuil rcintions~hip to

investiqate would be that between attitude tolward inclunding international

content and bElief systems as defined by 0..J. Ha~rvry (196,7; )Irvey, Hunt,

& Schiroder, 1961). He hias dlescribed fouir systems of helief alonfi a

concreteness-abstr actness co`ntinuium. One would~ expect~ to, findl a positive

relationship between thle ablstrnetness withi which n techejlr thinks andi

willingness to use interinatiirina content. nelief systems have beeni

related to attitude changes (Hanrvey, 19617), parent chibi relations

(Harvey & Felknor, 1970), classrooms atmnospherte andl stlndent performance

(Hlarvey, P'rather, Wihite, & linffmofister, 1968), anrd alongi withi sub~jert

matter and sex of teacher to students' Igrade achievement and student

perception of teachers (Ilarvey. Wells, Schimidt, & IGrimnm, 19171). Con-

sequently, if a relationships between attitulde towardt international content

and belief systems can be shown, a big step in deve~loping thle nomologient

net around the new construct will have been taken.

Anoither potentially reproductive area o~f research fo~r expandling, the

nomological network around thle construlct m~easured by~ INTTOTr and EORTUT)' 1

would be to investigate thec relationship between level of teacher cogni-

tive moral development (Kohlbherg, 1964, 1969, 1970, 1971) and attitude

toward uising International content. Erb (1976,) has arguled that










international or cross-rlltltral studlies are ess.ntinI for arbievingi the

higher levels of moral dleve~lopmenit. However, it re-mainis to be demon-

strated that n1 positive reintionshhip does in fact exYis It betwoo n atitude~

toward international conlten~t and higher levels of oran~l development.

Cronibachi and Mleebil's (1967) second category for the laws in a

nomologiical net is the relationshi p between thleoreticall conistruct s andr

observables. T'o further vnltd;ate thle INTT1OT and I-[,inO suibsentes one

would have to observe the classroom tecA~hing of those who achieve dliT-

ffrent levels on the two scalles to see if dlifferncel~s in thie uise of

international content did exist. Althougfh atltitudr is an incinaation to,

behave in a certain way and not a b~ehavior ini its ownl rightI (Finlhbein.

1967, p. 2h0; Shaw & Wrifiht, 19)67, p,. 6; Trhurstonr,, 19~7, p. 78), niti-

tudes must predict behaviors to make them wo~rth worrying about.. Suich

variab les as the t ime teachers spendl on int-erna tirnal:~ topi cs the variety

of ways they introduce international content, nr then numbehcr of timnes

they necept stulen~t initiatives to introduced inlter-national ideras inlto

lessons are some of the teacher behnviors which mighlt he~ looked at and

related to scores on EORTOTr and 1NTTOT.




Applicationsl"


Basic Research



There are two General aIreas in which thr FTICAS col~lld be put to goodl

use. One aren is in haiste researchICI into' atjtitude frma~n~tionl. As: thel

nomoloaical net is built upy nroundl atituide toward 'Ising international

content. in the curriculum, it will be possible to test hypotheses about

whiat vlriablels contribute to, the formation of this atltitude. A causal









model coiid be developed andi testedl by path1 analysiR (Blalack, 1964,

1971; Eleise, 1969, 1975; Wrifiht, 1921, 1934). Both11 mr.sures of enn-

struicts and observable roaid be enteredl intor analysis. nnckgroulnd

variables such as previous exposure to internationally situations such as

foreign~ travel or meeting foreign visitor- s or~ formal study Ruch as course

work or study abroad mnight form one cluister of variablel s relatedl to,

attitude formation. Personality factors mirght forl ancother cluster of

traits which could be shown to have a h~earingi on aIttitudes toward usinn

international content. Environmental factors suich as school r~lmate,

type of community, or aspirartions of students might contribute yet a

third set o~f factors affect ing teachers' attituldes toward international

content. The possiibilitie appear endless for invisticating th~e factors

related to the formationi of thle attitudes meansurred hv INT~cT;T ad EORTOTr.




Evalua tion andi Teac~hiri


The instruments developed by thiis study, IN'rraf,r FnRTOUT, andl FKS

total score, could be used in evaluation stuldies. PendijnJ further

validation studies on EORTOT', that scale could be uised to, evaluate the

results of teacher inservice programs. Intenlsive institutes and year-

long progcramns could be candidates Ear evnaluatio usinlg EORTOTI. If the

teacher participants waere no~t self-selected so that- their init tal INT10T'l

scores were very hjigh, INTTOIT might also, be ulsed tn evalluate suchl

iniservice efforts.

TNTTOT and EORTOT could be used in preservice teacher educational to

measure where teacher candicl3Les were in their thinkil6nh abot interlnationall

content in the curriculum. INTTOT and E()RTIOTI used with preservice

teachers might be employed rjther in more baisic rpeserch or evaluation









of efforts to make teacher randidates miore inclinedl to use international

content .

Finally ENTITOT and E~ORTOT might be Iised ini lonju~nctio n withi cur-

ricululm research. Efforts to modify cuirricula often necessitate changless

in instruction as well, AnI effort to inter~natiolnnalive thp culrricululm in

a given school~ district could he supported by5 knorwledgye of where the

teachers stood in their attituldes toward includlingl international content

in the curriculum. These inistruiment s could be uised in teaching as we~l

as research devices in an effort to changef curriculla. The obljectives

which make up the items o~f FICAS could he used to Fcensiitiz techerlfs to

thle applications of international content to bo~th inte~rnatiional and

either/or objectives. or to~ natiolnalistic obrjectives on a enmpal~rative

basis. However, using thle scales for teachingr precluldes their uise with

the same group of teachers for research purproses.




Conclusion



In this study two attitude scales and a kniowlj ge scale were

developed that can he used to further research andi evaluation in the

area of international cuirricuilum and instruction. Vallidated for lf rican

content, INTTIOT and EORTOTr stand ready for vailidationl and application in

a wide range of international studies arena. These scales have potentint

for basic research into the formation of attitudels arnd the co~nsequences

of certain attituldes. liey can serve n uiseful pulrpose in cvaluaitio~n,

measuring change as a result of inlservice o~r preslervice treatmennt. ia

thley could he used for teachiing devices by having: teachers focus on thle

individual items (i.e., curricular objectives) inl n effort to develop

strategies for using international content for achiieving the objectives.















REFERENCES


Africa Southl of th~e Sahara: An_ Objective Test fo~r Seondary Schanls.
Pittsb~urgh1: Carnegie-Mellon University, 1968. (ERIC: Document
Reprodutlioni Service No. ED 070 010)

African Studies Handlbook for: ElementIary and S~econdarily Shooll Tea~chers:
Part I. Amherst: University of Mlassachulsetts and Wol~rcester Teacher
Corps, 1971.

Allman, R.iJ. A study ofi the~ social attitudel s of college~ studecnts.
Jolr-nal of Social Ps~ychologfy. 1961, 53, 33-51

Almgiren E. & GuistaFssnn, T. World Citizen Rieslionsibility: As~sssment,
Technirppes, Develop~me~nt Stu~dies, Mater-ial Constr-uct an, anrd
Experimental Tenrl~ing. April, 1.974. (ERITC Docum~llent Rep roduclrt ion
Service No. ED, 110 173)

Anderson, L..F. An examination of thie structure and~ chjectives of inter--
nationn] education. So~cial Education. 1968, 12, 639-667.

Aron, R. Effects of Teq;clietr Expectyncies: My~th olr nrlity1 Pape~r pre-
sented at- the annontn. rreting oif the Americani Per-sonnel anirt Guidanc~e
Association, N'ew York, Ma~rchl 1975. (ERI1C thrulment Reproduction
Service Noe. ED 115 722)

Beauchanmp, G;.A. & Conran, R.C. L~ongtituinal Studly in Curricu~lum Eng6ineer
ing. Paper presented ait thle annual meetings o~f thle American Edu~ca-
tional Research Associaition, Washington, 11.C., Aprril 1975. (ERKIC
Document Reproduction Service No. EDU LO2 670)

Becker, J. Education for a Globail Society. Bloominigfton, Indiana: Phii
Delta Ka1ppa Edulcatiolnal Fouindation, 1973.

Becker, J. Perspectives on globdl education. Social Eduention. 1976,
38, 678-682.

Beyer, B.K. Africa Soulthr of the Sahanra: A Resout~re and Cu~rriculumm
Guide. New York: Randrom louise, 1969.

Beyer, B.K. 8 Hicks, E-.P. mesoAfi:A prtnWhtArca
Se!condaryy School Student~s Know anid nelieve Aboult Africal Soulth ofr
the Sahara. Pitishu~rgh,: Carnegie-MElllr n Uniiversity, 191,8. (ERIC
Doculment Reproduction Service No,. ED 023 691)


-76-









Beyer-, B.K. & Hicks. E.P'. A Soc-iil Studies Curri-cuilulm Projctl to Develop
and Terst Instructionn} Manterials, Teaching: (hides,. and Co~ntent Uniits
on the Hlistory andi Cllturle of Sub-Saharann Afrirn for use a7t Selected
Grade Levels in Selonda~ryI Selools. Project Afrien~l: Fnl eot
Pittsbulrghl: CarnefgifPe-Melon Ulnivers i ty 1')70. (RCDcmn
Reprodluction Service Nn. ED 042 673)

Bidwell, P.W. Foreign affairs in the colleges: Strcnjctheningl interna-
tionalR edulcrt-ion. :Joumnal of HFi_ghe Ir [duc;tingr. 194, 35, 4 26-4 33.

Billeh, V.Y. & Zakharinies, G~.A. The developments anld aplplicaltion of a
scale For measulring scientific attituldrs. Scie nce Edlucation. 1975,
59, 155-165.

Billings;, C.E. The challenrge of A1frica in thle ciirricilum. Social
Education. 1971, 15, 139-146 & 153.

Blalock, H.M., Jr. Causnl Inferences in Nonexperimental Research. Chapel
Hill: University of North Car~olinia Precss, 1966.

Blalock, fl.M., Jr. (Edl.). canusal Models in thle Social Sciences. Chicagpo:
Aldine-Atherton, 1971..

Bloom, B.S. & Kr3thwolWOII D.R~. TaIx~onomy of E~duentliona~l Objectives:
Handbook I, The Cognitive Domfg'yg. Newi Yor-k: Davlid Mclrlay, 1956.

Bock, R.D. tlultivarintr Statiti~ical Mlethods; in liehavior Resealrch.
New York: Mc~raw-fllill, 1975.

Bogatz, B.F. An InvestiEat ion o~f Tleacher E:xpeLnctacls of Instructional
aElriarjls: Reseairtl Report #12. Engenel(' (Iregan': Northwest Regiionall
Special Eduicatio, n inlstrulction al. Mnter-ials (Center, 1970. (ERIC
Document Reproduction Service No. ED 075 3451

BOhlannan, P. A\ Prelininalry Review of thec Intercultura~l Dimiensions in
Interpat ional /In tercul tural Edluc~at ion, Grades h%- 14' Final Report .
Publication No. 156. Boulder, Colofrdll: So~cial service Edulcation
Consortium, 1973. (ERIC Document Reproduction, Service No. ED 100
786)

Boneau, C.A. The effects of violation of assulmtiotlns underlyingR the t
test. In n. Lieberman (Edl.). Contempolrary Problems in Statistics:
A Book of Readingis for the B~ehavioral Sciences. New York: ox.ford
University Press, 1971.

Boyer, !u. World order rdura~tion: Whatn is it? PhIi Declta Kapp~an.
1975, 56, 524-527.

Brager, G.1,. Teacher exportancy and~ mothematics alichievement. Journal
of Research in Ma~themat ics Edulcationr 170;, I. X8-9i.

Brubalker, D~.L. Secondary S~cia~l Stu"dies; for ly 70's: P~lnlapnig for
Instru~ction. New York: Thomas Y. Crowell C:o., 1973.









Campbell, D,.T. & Fiske, D.IW. Conlvergent nn~l dis~ rimiinant validationi by
thle multitrait-mul timethlod mntr-ix.. P)sycl~(gchlogi ni Blletin. 1959,
56, 81-105.

Carroll, .J.B. Thle nature of thle danta, or howr toi < home a correlationr
coefficient. Psychometrika. 1961, 26, 347-37'.

Cecil, C..E. A Summlary of th~e Four-Yea: Research Pr'ogram Condulcted aIs a
Project of the fluntsvilIle-Mandison Counrty Education Improvement
Program. Cumulative Research Bulletiin. Atlantn~: Southern Associ-
ation of Colleges anid Schools, 1971. (ERIC Docuiment Reproduc ti on
Service No. ED 058 371)

Clark, R.M1. A\ Study_ of Taclcher Behavior and Attit~des in Ejlementary
Schools with 11(gh aind L~ow Puipil Achfervement. Paper presented at
the arninul meetings of thle American E~ducational lrRfserch Association,
wdashrington D.C., Aplril 1975. (ERIC D~culment Reprioduction Service
No. ED 109 025)

CLevelanid, H. The Overseas Americans. New Yo~rk: Mlc~raw-llill, 1960.

Collins, il.Tr. Are You G:oingE to Teach About Afrirna Nlew York: African-
Amnerican Institute, 1970. (ERIC Dnenimenti Repiroductionl Servire No.
ED) 044 324)

Cronbalch, L.J. Coefficiie((nt llpha and the internall mar-ncture nf testfi.
In W.AZ. Mehrens & R1.L. Ebel (Eds.), PrinCiplesi of Educat.ionn i and
Psyhological Measu-re~net: A Book_ of Srflected~ Ralndglngs. Chlirng,~ :
Rand Mlcrally, 19h7.

Cronhach, L.J.I & Nlechl, P.F. Conistruct validity in psychollogirn1 tests.
In1 W.A~. Mlehrens & R~.L. Ebe~l (Elds.), Prin~~l~ciple ofI Educational and~
Psychological Nlensulrement. Chicago: Rand Mc~`!lly, 1967. (R~e-
printted from _1y~chlogical Oulletin 1955, 52.)

Dahiberg, K.Ai. Environmental Studies: Some~ Probilemis and Prtarities.
Paper presented at the annual meeting of ther inte~rnatio nal Stuidies
A2ssciaition, M-arch 19741. (ERIC Document Reproduciiti on Service No.
ED 103 283)

Darlingiton, R.B. Multiple regression in PSycholl ogicai l rESe~rch and
practice. Psycholrggicill Bulletin. 19h8, 69, 161-182.

Davis, F.B. Educationii MElcsuirements anld Thleir Intlerp~retiions. Belmont,
California: Wadswo~rth Publishing, 1964.

Dow, P.11. MAICOS : Social studlies inl crisis. E~duc.tio~nal Leadership.
1976, 34, 35-19.

Easton. S.C. NDEA Centers fnr Internatir~nl 1 nd Lar~~nguge ang( Aren
Studies -- 1976-78. Wa~shiing~ton, D.C.. UI.S. Derp-'rtment of Hlealth,
Eduction~,~ and elfare, Office of Educantion, .Tuine 1977. (Mimeo-
giralph Report)









Edwar-ds. A.L,. Techniqes_ of Attitllde Scale (Iranstruction.. New Yrork:
Appleton-Centulry-Croffs, 1957.

Ehman, L.., Mlehlinger, fl., & Patrick, JT. Towarld E-ffeelive ~Instrlction in
Secondary Social Studies;. Boston: Illoughtotn Mifflin,, 1974.

Eicher, C.E:. An Investigaltion of Eileme~ntary Chlildrenl's Pe~rceptions of
Selecc~te Countries of thle World: A Techlnical~ Report. Vermillion,
Sounthh Dakato: Southi Dakota University E3durntionn: l Iresearch and
Service Center, 1975. (ER1C Documient Reprodulction Servie No,
ED 123 149)

Engile, S.II. Objectives of the social situdies. I ..Msi~s&FR
Smith (Eds.), Newj Chlalleng~es in t-he Soc~ial Studli('s: Emp~licationss
of Resear-ch for ~Techin. Belmont, Caifornin:l Wntdsworth
Futblishing, 1965.

Erb, T.O. The Deve_4olopmen o~f Noral Juidgern L W~ith n D~iscurssion of thle
R~ole olf Cross-CuIltura! and Internatio~natiS~l Suifs Un1pbl1lishied
manuscript, University of F~loridla, 197fi.

Fishbein. )i. A consideration of beliefs and~ their rn le in attitude
measulrement. In MI. F~ishbein (Ed.), Rca!iings i~n Attit~ude 'Theory and
Measurement. New York: Johln Wiley & So~ns. 1967.

Fox, D.JT. The Research Process in Education. Niew Yoik: Horlt, Rinehart
and W'inston, 1969.

Frech, V.P., Jr. An annnivis orf thle effect of the: Anthlroplognly Culrriculuml
Project material. Thc C:oncept of Culttrue,~ (n the( Cthnlocenttrc
attitudres of Fourthi Fprale students (Donctoral l isseration. Ulnivrrsity
of Geo~rgia, 1973). Disserttatin Abstracts Interna~tionaal, 1973,
34, 38?0A. (University Mlicrofilms No. 73-31, 881)

Frech, W.P.., Jr. The effc~ts of cognitive tra~iningC ini ntroporrloiy on ethno~--
centric attitldes;. I'sycholgg~ in the Slloorls. 1975, 12 36,4-370.

Freeman.. R.E:. Curriculum Materin s E7valuation as -a Process for 1..~;,
Educating:~ Work of the D~iabrlo Va~lley Eguca~tion Proj~ect. Or-inda,
Cali fo rn ia : Diable Valley Elducation Pro~ject. 19)7i. (ERIC D~ocumecnt
Reproduction Servic~e Nn. ED 099 283)

Frey, S. Teachers and! Behnyior Objecrtivces. 1973. (FRJlC Doc~ument
Reproduction Service No,. E:D 088 849)

Gardner. F.L. Attitudre measuremilent: A2 c'ritiqule o~f some recent research.
Edwaltiona~l Research. 1975, 17, 101-109.

Garrifion, K.,C. Worldmrinded attituldes of colllege stiidenrts in a soutihern
university. Journal ofl Scia~lsy~ l Psychc logy. 191 54, 167-151.

Global Development Studirs: A Mondel Curricululm forl anil Aademic Year
Colrse in Clobal Systems and Human Development~ aL tlle Secondary and
Undergraduate Levels of Genieral Eduication. New Yo'irk: Manangement
Tnstitute for National Development. 1973.





-80-


Gordorn, R.A. Issues in multliple rlTegrSSion. American.;ll .Jurlnal of
Sc~ciology;. 1968, il, 592-616h.

Guifon, JP.& Gilfrd.R.Anl analvasj of thr I'll.ors in a typical
t es t of int rove rsio n- est rove rs ionl. Junlo homlSca
I'sychology. 193'1, 28. 377-399.

Guilford, J.P. & Guilford, R. Pers-onality factors S, T:, & M1 and their1
measulrEment. Jollrnal Iof Psychlology. 96 ,1917

Gllttman, Li. A basis f~or sca.lingf qualitativ e daita. Amelric in Srcirlog(enl
Review. 1944. 9. 199-150.

Guttman, L. The Corn~l~l Te'chniqule for Rcale :Ind intensity aInalysis.
Educational and P'sycpho~l~ogcl Measurempt.nt 1')47. 7, 247-280.

11311, S..A. Africa in Ul.S. Edulrantiona lMte-r-ials: lhir-ty P'roble~ms and~
Responses. New York: The African-Amerjrrica Ilstitlltc School
Services Division, 197.

Harve~y, ('.J. Coneptal Rystems~ and attitrlde change:~. In M1. Sheriff &
C.WJ. Sherif (Eds.), Att~itulde-, Ego-In~lvolvement, anl Chnn:e Nlew
Yorkr: J.ohn Wiley & Sons, ]1967.

Harvey. 0... & Felknor, C:. Iarent ciildl relations ais an nnlteredent to
conrlepelnal functiouninC. ..Mlo .C iml(~h1 al
Experience and the ProlCesses OF Soc'ialiationl. NewV Ynrkr: ACademic
PressR, 19 70 .

Harvey, ('..[., Hulnt, D.E., & Sch~rodlor,, I.M. Colciqpf ug11 Systems ni d
Personality. New Yo~rl: .Tohn WJiley & S~ons, l')hl

Ha rvey, ()..., Prather, M., W~hi t e. I.J., & lin f fmr ls t er-, ..LE,: In M1.I.
M~ilrs. Wi.W. Ch~arter. Ar., & N.L. Clfe (E-ds.), Readn~ings ini thle So~cial
Psyc hology of Edulcationn Roston: A~llyni & Raironr 1970. (Reprinted
from Americanl Educational Researchi Journa).l, 198, 5.)

Harvey, O..T., liells, I;., Sllhmidt, C., & G~rimmll, C. Efllects of Subject
Matter. thle Beliefs System annd Se,. of 'rnliadr: and Stuidents upon
Students' Gradle A~chievcnment and Percepction of fenchler. Urnpub lishecd
manuiscript, University of Colorado, 1973.

Havlicek, 1,.1 & Peterson,. N.L.. EFfects of the~ vio~latio~n of assimptio~ns
uupon significance levels of thle PeIrSon r. yhooca ulin
1977, 84, 373-377.

Hayden, R.L,. Internationall izing rubliC E~duca.tionl: whalt thet States AreO

and International Educat~io~n So~ciety, Tulranto,, February 1"76. (ERIC
Document Reprodurtion Service No. ED 124 462)

Heise, D.R. Problems in pathl analysis aInd (iisail inferrencer. In1 E.F.
Bo~rgatta (Ed.), Socig~gical Methodolol gy 19(19. San F~ralnesco:
Jossey-Bass, 1969.





Hleise, D.R. Causal Analysis. New York: Joh~ln IWiler! & Sons, 1975.

Hioamm, L..E. A Denlonstration of the Ulse of Self-ln~struc~tinal anll Othler'
Teaching Techniques for Rcmedial Instrucltionl of I.aw-Achlievin ng
Adol escents in Read~nin and Mathemat~icy:s Tecigiegl~ Progr-ess ReFport
No. 2. Albllquerq ue, Neww Mextrao: TPI Instiiiute, 196R. (ERIC
Document Reprodulction Service No. ED 018 111)

Interculltural Unlderstanding: Projet Finall Reporrt. P'ittsburgih: All.e-
gheny County Schools, 1972. (ERIC Do~cumnent Rcpro~ductionl Service
No. ED 081 705)

Jaccard, JI., Weher, JI., & Lanndmlark.. .1. A mulltitralit- nall time~thodl anarlysis
of four attitude assessment procedures. Jounalml of E~xper-imenltal
Social EPsychology. 1975, 11, 149-154,

Johlnston, .l Econometri ~Ftlotods (2nd ed.). Nrvw York: M~a-il
1972.

Jones, E. An Evaluatioiq Re or~t on the Regiongll Eduypt~ionn1 Agencies
Project (Alabama, L~ouisiana, Tennessee, Texas) ini Internlt ional
Education. Austtn: ~lexas Elducation A1:ency, office o~f InternoLionnil
and Bilinguint Eduication,. 1970. (ERIC Documient RIeprorlrction Service
No. ED 048 086)

Kerlinger, F~.N~. & Pedhiazuir, K..T. Nhl tiple r. .inn inl Behalvior
Research. New York: 1101[, Rinehart and Winstnn. 1977.

Kirk, R.T,. Expefrimentala De-sign: Procedur!es for ther liehavior Sciences.
Belmlont,, California: Brooks/Cole~ Publ~ ishinge En., 1968.

Klassen, F.11. & M~oore, R,.S. De-velo~pment of qql Instrumefnt or th~ie Study
of International E~ducytfion in Teache-r IEducatilly IProgram of U.S.
Collge and Universities: Final Repor-t. Wal:Shing/ton, D.C.:
American Association of COLleges for Tea:herr Fdulcation, 1968.
(ERIC Document Reproduction Service No. ED 032 010)

Kohlberg, L. Development ofI morall character- aind irelog~~ly. nM
Hioffman and L.W. Hoffman (Eds.), Review of Child Developm~ent
Research. New York: Russell Sage Fou~lndation, 196,4.

Kohlberg, L. Stage and sequenlcce: 'The colgnitiv-lve-developmental approach
to srcialization. In D). Cosli n (Ed~.), Ilandil, lokl of Socialization
Theory. Chicagn: Rand MlcNally, 1969.

Kohlbergi, L,. Education For- justice: A moderrn statelment of the Platonic
viewr. Inl T. Sizer (Ed.), Noral Edulrcatio~n. Cambrhidg~e: Ilarvard
University Press, 1970.

Kohlberg, LI. The claim ofr meanl adequnry of a1 highest stnge of moral
judgment. Th~e Jurnal ofl Philosophy. 1973, 70, h10-646.

Kra thwohl D 1.R., Bloom, H.S. & Mad~a, I!.R. Jaxol~nany of Educational
Objectives: Handbook II, The A~ifective Domain. New Yolrk: Davidl
MlcKay, 1964.










Kuder, (:.F. & Richardson. MI.W. Thet thleory of the est imot ion o~f test
reliablility. Psychometrika 13, 5-1

Lefe, JI.R. Teaching~ Srcal Stldies in the E~liemallpry Schpol~~,. Nerw Yo~rk:
The Free Press, 1974.

Likert, R. The method ofr consltructinf an attitudl e scale. In n. Fishbein
(Ed.). Readings 'gn Attlii lde Theor~y anld NEIgsujlt'lrlment New York: lohln
Wiley &~ Sons, 1967. (Excrpte~d from thle appe.nrlin of A technique for
the measurement of attituldES, Arclives Of Psveh~lology,~ 1932, 22.)

Likert, R. A techniquef for ther menlsuremen t of nt i tudcs. In G;.F. Sulmmcrs
(Ed.), Attitudle Measurem~nent. Chiicag~o: Ralnd Mclnally, 1970.
(Excerpted from A~rchive~s ofr Psy~chol~rlogy, 1912, 22.)

Lindquist, E:.F. Decsigi.n d nrul Aslysis of I':4per~irpelqs in Psycholog~y and
Education. Boston: lollghto~n Mifflin, 195 1.

Lord. T.MI. & Novick, Mi.R. SLatistical Th~eor-ies of] Mental TecsL Scores.
Reali n g, th ssa chuIseItts : Andd~son-Wesley, 1968 .

Maicias, C.C. Progirammnedl lyprningi as Ulsedl in tite f;roma Public Schiiols.
Includedl in th~e proceedlingi s of the~ 20thi Anniin1 Pac~ific Lnorthwest
Confernce o~n Foreiign .Languagers, Portlanrd,, clroun, April 1969O.
(ERIC Document Reprodulction Se~rviee No. 1:D 1.25 33

Nlnrion, K.:l. Kappon initerview: Krvinl Michacl Marllian talks about pea~ce
studies. Phi Delta Kalppan. 19173 55. 187-180,.


riculumln in Nor-th Cclntral Assoi Ention Schols.,li Th1e Northi Central
Association Qularterly. 1963, 38S, 205-213.

Hlassialas, B.(:. & Cox, C:.C. Inqryi ScplSyle. York:
MlcGraw-Hlill, 1966.

Maxim, TG.W. TIhe role of reseanrch~ in competenryl hoinied towhrll ednelntion.
Education. 1974, 95, 94-96.

McCarron, J. &, McCunle, S. A Tefachr's' Pe~rsppetiyeg I six keta~l iss~los.
Washingtonl, D.C.: Naitional Educatio~n Asscc~iatiion, 1974. (ERIC:
Document Reproduction LService No. ED1 120 112)

Nletcalf, L.E. Resenrih onl teaiching, thec Fsorji stuies. n NL. age
(E<1.). Hanndbook of Rcsea;rchi on T'eachLngl. Chicngn: Rand M~cNilly,
1963~.

Mlontessol i, NI. LEducationn anrd Powe l (.R.lnet n ) hian
Henry Reg~nery, 1972. (Originally pulblishecd, 1949.,)

Mlorehouse,, W. A New Civic ~iteracy: Angeriyno F~drlrgti~ng gd I:loba~l
Intelrdeppn~dencE. New Yorlk: A\SPEn InStitultr fr~ ((umanilStiC StudrieS,
1975. (ERIC Docuiment Replrodulction Serv~ices Nor. EDI 120 084)









Moreland, W.D). Currientumi trends in thef snrin1l studiins. Social Education.
1962. 26, 73-76.

Murphy, E. I. & Stei~n, 11. TechigS Africa 19d tly: 41 linlll~rulboo for fearhers
and CuriTTcuum1 Plannlers. New York: Citaltionl Pre1.Ss, )197.

Murphy, G. (Ed.). Human Naturlre and Elndurilgg Pencr. Boson~il: Houghton--
M~if flin~, 1945 .

Newman, A\..l. & Ware, Wi. Acsteh~tic perceptions anr dnl wIldmindftednes -- A
research note. Thle NewJ Era. 1976, 57, 45-116; 'IR.

Nie, N.1l., Lnt C.FI., Ienkins, J.i. ,. Stei inbrcn ner. K., h listl, D.11.
Statistical Packangi l Ear' the S~cial Sciegres (2ndl edt.). New~ York:
M~cGraw-Hlill, 1975.

Okey, J .Rl. The EFffects of Mrsltery TachiggF Strqr~egy r! ~legacher Attitudes
and Pupclil chvem ent. Paper pre~sented aIt the' annon~i. meeting of
the National A~sso~ciajion for Researchl in Scien~ce rfeaching,~ Detroit,
March 1973. (ERIC Doculment Reprodu~ction Service No. ED, 080 311)

OlIVEr, D).W. & Shaverln J.P. Tching; Publ_1~ ISSleS inl the IHiph Schoorl.
Boston: lioughi~ton-Hifflin, 19h6.

Olsoni, C..L. On choosing a rest statistic in mullltive.rinte analysis of
variance. Psycholo~gien1 Bulletin. 1976, 8 1, 57'1-586.

OIppenhleim, A.N. Questionna~iire Design and Atti~tude leas1ur-ement. News
York: Basic Books, 19h6.

Osgiood, C.Er. Cross-ciltltral compara~bility in afttitude~ monSurfment via
mul t ilingual seman, t ic dIi ffe rent i ls. Inl M. F-ishhlnlc (Edl.), Readingis
in Attitude Trheory~mt and easur-ement. Ne~w Yo~rk: .Ilnhn Wdiley &~ Sons,
1967.

Osgoord, C.E5. & SuJci, 6.J1. Factor analysis of mleaningf. Journal of~
'I|xperimlental Psychlggy.~I 1935, 50, 325-138.

Osgood, C.E., Suci, G~.J.. & Tannonbaum, P.11. T)(e Metnsurfemeq of Meanling.
Urbana, Illinois: Uniiversiity of Illinnis Preass 1957.

Olshima, E.A. Changes ini attituldeS toward scienCe and1( Confidellce in
teaching% science of pro~spective flelemntar techeclrr s (Doctordl
dissertation, Oklahoma State Ulniversity, 1966). D~issrt~ation
A~bstracts Internatiolnal, 196r7, 27, 31981A-41701. (University
MlicrofTlms No. 67-017, 268) (ERIC Documrnts Rerrodulrctionl l Ser-vice
No. ED) 014 4124)

O)swald, J.M.. _Ilt~erculltural Social Stuldies Plrojllt:L Appendcl ees 8, C, D,
G, 11, and 1. Hano~ver, NIew Ilamplshire: Amelrie:nn Fuiversit les Field
Staff, 1974. (ERIC Do~n Iment Reprodulction, Servic~e NIo. EDI 099 286)

Pearson. E.S. The analysis of varinacen in casers ('f nonl-tlrnormal vrtatiol.
Biometrika. 1931, 23, 114-111.









Phillip~s, C.S., JIr. The p~rrsent wor ld chl lenger t o h i ghr edulcat ion.
The Educational Recorrd. 19h3, 44. 266-276.~

Reischauer,, E.O. Toward ~th~ Twefnty--First (centur-y: detnfoa
Changlrlin World. Newi Yolr-k: Vintage Hoorks, 1971.

Remmers, II.H1., CGe~, N.L., & Rummel, JT.F. A Practicail Introduction to
Mleasulrement and Evalatlion. N~ew York: Hlarper & Ronw. 1965.

Rich, Ei.J. Good news andl bard news: Afrienn stud~ies ini Americanl school0S.
1955-1975. Issue: A\ quart~erly Jourrna of 33-410.

Sampson. D).L. & Smniti, II.P. A scale to mealsure w~r-ldlmin~deal artituides,
JournTal of _Socil Isy~clloloy 95,4, 916

Schmidt, N.J1. African Oultrenc~h Workshopl 19174. thbannrl:, Illinois:
Universi ty of Illinnis, A1Frican Studlies Progiram, 1975. (ERIC
Document Reproduction Service No,. ED 107 64h)

Sott, J.A. Teachin for a Change. NFW YOrk: UninL;1 Rook;S, 1972,

Seasbore, R.H. & ftevncr, K.4. A time-saving deviar foi th~e ennstrulctio n
of attitude scalES. Jlour~na~l Socinl 1sychology. 1931, ii, 3h6-372.

Seiler, L,.1. & Rlough,~ R.L,. Empiricl3 compar-isons ofr the rhurllstone and
Likecrt techn~iquefs.. in G.F. Suimmers (Ed.), A~ttiitude Measulcmrement
Chicagio: Rand McrNolly, 1970.

Shaw, M.E. & Wrightl, .M. Scales for the Neasulrement of Attitudels. New
York: Mlc(raw-fill, 19r67.

Shrigley, R.L. T'he cnrr-elation onf sciene aIttitudelr and science knowjledge
of preservice elementary teachers. Science Edlucation. 1976, 58,
143-151.

Smith, H.P. Do interculltmal l experiences alffet lt itritdes? Journal of
'110111 Social P'sycl(lwty_. 1955, 51, 469-h77.

Snider, J. & Osgood, C.E.. Semantic Differentin le~chnique:: A Source
Book. Chicagor: Aldine Publishing! Company, 1969.

Stearns, T. Ideas forT thle Developmen~lt of I'.rograpisR Rellting _to the
International Shcene andi its KRole in thle Scho~ols. Lnig ihgn
Mlichiigan Sltat Roardi of E~duration. 1966. (F.RIC Ilocument Reproduc-
tion Service No. EDl 021 644)

Taylor, 11. The Worrld as Teaicher. Cronle11ini:Sthn
Illinn~is University Prress, 1969. As ulse ne h il
T'he W'orld and the American Tealcher. Was-hingitci D.C.: American
Association of Collegies for Teacher Eduication, 1968.

T~eachigg about Africa inl thec Sccial Studies C:urriclrulum. Ralleighl, North
Caroling: State Department of Public Instruiction, Division of
Research, 1972.









Thulrsto ne .L. Ps~ychorphysicnl analysis. Americanl Jou~rrnlc of P'sych-ology..
1927, 38, 368-389.

Th~urstone, L.LI. Theor-y of attitulde measulrr mntPscolgrp Blet
1924, 36, 222-241.

Thurstone, L,.L. The measurement of social attituides. .Journal of A2bnormal
Social Psych~ology. 19 11, 26, 249-2h9.

Trhurstonel, L.L. Attituldes can he measulred. In MI. F~islhbin (Ed.),
Readings in Attitude Theorly and Meansyrr~ement New Yo~rk: Jlohl WIjley
& SIons, !.967. (Rep'rintedl from Ame ri enn Jouirnal of~ Socirology, 192L8,
33.)

Tinlketman, S.N. rlanningj thle objectives test. I ..Tonie(d)
Educational Mecasurerment (2nd ed.). Washingttonl. Ir.C.: Amerie~nn
Council on Educatio~n, 1971.

The UNErSCO associated schlools project in reduction fo~r internationlll
cooperation and peace. Edlucational Do~cumentation and Informatioio n
Bulletin of the International Burfani of Eduerntion. 1975, 49, 29-31.

Van de Geor, J..P Introduct iont to M-ultivariatt Analysis for the Social
Sciences. San Francisco: Freemlan, 1971.

Veldman, D.J1. Fortran Progr.4mi~ng for- to Behaviorntl Sc~inceLs. New York:
Hlolt, Rinohart and W~inston, 1967.

Waimon, M1.D., Bell, D.D., & Ramlseycr, (.C. 1eEht fKolcg
AboutC Subj~Lect~ Maitter~l ir thc Pcr~for~mance and A!t itudescr of Pggyspertiy;
Teachecrs. Norma ,. Illino~is: 11Ti no i s St ate 'n iversri ty, 1 71.
(ERIC Document Reprolduction Service? No. ED, 052 1 In)

Willmer, J.E. (Ed.). Afric,: To~ny)~ in Prslpiectif and Apprqclies.
Tualatin, Oregon: G~eographic and Area Study Publlications.. 1975.

Wise, .J.11. hn Apppg~lnin Isgnorane f !, ye World: lile I'rte r7 f Affluence
Pape~r presented at Lhr annuiral meeting: of the Institute of Auistralian
Geographers, WollongonC, Necw Sout~h Wales, Australia, Aufiust 1975.
(ERIC Document KRproduction Service No. IED Ill 745) (a)

Wise, J.H. Student-- deficienicy in haste world kno~wledrte. Journal of
Geogirapliv. 1975, 74, 477-488. (b)

Wisemann S. TIhe ctdnen~tional obstacle race: Factor! thait hinder pupil
progiress. Educational Research~. 1973, 15, 87-9 3.

World Ilistory Ser-ies: African. Rockville, Nolry ilnnd : Thar- i ofT E~ducaton
of Mo~ntgomery Couinty, 1972.





-86-



iright, S. Correlation aInd causatio~n Jagrnal of AXT icultuir;l Research.
1921, 20, 557-585.

Wright, S;. Thie methodl of pathl enefficients. Ann.!ls of MaIlfthematiCal
Statistics. 1934, 5, 161-215.















APPENDIX A
INSTROUMENT USED TO0 CATEGI~ORIZE 90 OBJEC(T~IVES AS NATIONAITS1|C,,
INTERNATIONAL, OR EITI`IER/OR, INCLUDINGh'( THEII TA'l~lY OF RE~SULTIS




Attached is a list of 5/4 cognitive anil 36 effective objectives

which could be selected in rlevelopingi a social stud~ies ciirrjicultun. Some

objectives focus on A~merica and can beSt, pe~rhaps only, be pulrsuled usingC

content derived from the Ameirican experience. Other objecttives are more

international, or global, or jiaterculturalnl in focusR so, that thery coulld

only be achievedl by using conitent whiich is deprived fromr outeir cultures,

nations, and peoples. A third set of objecttvesR aIre harder to- classify

as to their American or internationally fol'us. That:1 is they could be

achieved by usingR American conntent or international contentn o~r a com-

hination of bo~th.

After reading each obljective, plel nso mrk thef ent-gKory into which

youi think it should be placed.

-- If an objective would~ require mainly Amerlean cntent~~ o c~hieve

place a check in Columnln I.

-- If an objective wouldc require internlatiional ot- global content to

achieve, place a chEck in Coluimn Ll.

-- If an objective dores not specificallly requiire other American or

international content but conuldl be developed d withi either or both,

place' a check in Column Ill.


-87-





EXAMPLETS :

Co I umn

I If flI StudeFnts shou~rld

X L. ulnderstand thle U.S. Consiit it I tion

X 2. hanve a ktnow~ledge of Chtincs! dynasties

X 3. know the role of natuirnl rresources in
production




CCGNITiIVE OBJIECTIVES

Column

I II ZII Studrents shoulld.

6 2 1. halve a knowledge. of unrl<11 history

6 2 2. comprehernd the CI1lf b~l~etwee rich\ and poor
countries

8 3. understand thle consumer'-s role~ in the
American economy

8 4. understand social conf iit

8 5. comprehend Lhe role ofT Inhaor in produlction

6 2 6. nodterstaind difCferent wayvs or li fe on
li ffere~nt c~ont inents

7 1 7. have a kno~wledge of social chnane in America

7 1 8. comprehend the p~roiness IF nging in oulr so~ciety

8 9. understand the cau~ses aIndl Iffects of
hLstorical events

7 I 10. know about the imlpact of technology on
modern life in America

7 1 1.1. haJve a knowledlge of thec relationship betw~ieen
environmental and wasv of ma~king, a Ilving round
thlE world

7 2 1.2. understand the reintionship between coloniid-
ism and nationalism

8 13. Hiave a knowledgei of social interaction

8 14t. know about ~the role of rrovcrnment





-89-


Coluimn

IIf III Studelnts should.

6 2 15. hanve a knowl'`thi~r of woiilrld Gl'ongrPhyY

8 16. have a chlrounlog~iral I~nnwledger of major
Americaln historical events

7 1 17. hanve a knowledge OE thec structurer of American
gove rnoen t

5 3 18. have a knowledge' of d ifferent cul tures

7 1 1.9. comprehend the principlesi of American
democracy

1 7 20. understand~ the princriplels of coo~peration
among people

7 1. 21. know the locattion of t he world's major
resources

6 2 22. understand thle Americ:,i economic systemr

1 7 23. understand intrtaction amnngi variousf social
groups

1 1 6 24. understand urba~nizatio~n

6 2 25. haive a knowlEd!: e ofI vaiirous American ethnic
groups

2 6 26. understand thle rnuses of wair

5 37 27. comprehend dirffrent systems of government

5 3 28. comiprehend the pinranlistic nature of Ame~rican
society

8 29. kniow the major produicts of the student's
home state

33 3a a30. comprehend~ Ameri can forei rn polio cy

6 2 31. ulnderstmulr l d iff erent~ Iitterns o f family
orga~nization arolndd thle wotrldl

3a 4; 3n 3:2. comprehendl Ameri c~a 's ris to a po~sition o F
world power

3 8 33. rnmp~rehen crult rlr l cli ffus irtn

1 7 34. unlderstand thle e~xch-angF nf goods and
services




Full Text

PAGE 1

CONSTRUCT VALIDATION OF AN INSTRUMENT TO MKASURI" TEACHUR ATTITUDE TOWARD THE USE OF INTERNATIONAL C0NTI:NT IN THE K1 2 SOCIAL STUDIES CURRICULUM By [IIOMAS OWEN ERR A DISSERTATION PRESENTED TO THE ORADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE RE()U I REMENI S FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1977

PAGE 2

Copyright 1977 by ThomaK Owen Erb

PAGE 3

To Karen, Christopher, and Grcyniy

PAGE 4

ACKNOWLEDGMENTS Special thanks are due to my chairperson Dr. Arthur J. Lewis and my cocha i rperson Dr. Linda M. Crocker for the great support they have offered throufihout this project. Thaiilcs are due also to Dr. Gordon D. Lawrence and Dr. Rene Lemarchand for their editorial assistance in preparing the final manuscript. Special appreciation is due the following people in international and African studies who have supported this project: Drs. Haig DerHoussikian and R. Hunt Davis at the University of Florida, Dr. Abdelwahab Hechiche at the University of South Florida, and Dr. Wilfred Owen at the University of Illinois. 1 am indebted to 3.1 other professors at the University of Florida, both in the College of Education and the Center for African Studies, who assisted this project either by serving on a panel of experts or by allowing me class time to gain access to the 166 students who participated as subjects. In the latter regard, I wish to thank Dr. Gus Jimenez for supporting this research by arranging for 65 Hillsborough County teachers to participate in the pilot phase oE the project. Some assistance on computer programming was received from the Center for f list rue tionnJ .uid Researcli Coiii])uting Activities at the University of Florid.i. Coniput i nj', was dene Mtili:iiig the facilities of the Northeast Region! 1 Data Center of the Stat<^ University System of Florida located on the campus of the University of I'lorida in (Jainesvi lie .

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMENTS iv LLST OF TABLES vii LIST OF FIGURES viii ABSTRACT ix CHAPTER ONE RESEARCH PROBLEM 1 The Problem 1 Need 2 Definitions 5 Construct Validation 7 Hypotheses 9 Significance 10 TWO REVIEW OF LITERATURE 12 International Education 12 Teacher Attitudes and Their Measurement 18 Summary of Research 22 THREi: MEIHOD 23 Attitude Scale Construction 24 Procedures for Establishing Reliability and Validity. . 33 FOUR RESULTS 47 Reliabilities 47 Sample Distributions 50 Tests of Construct Validity 50 FIVI' CONCLUSIONS AND RECOMMI'NDATIONS 58 Introduction 58 Reliabiiitv Conipa r Lsons 58 Discussion of FICAS Validity 60 Multiple As[iects of Un i d imensional Scales 65 Adequacy of INTTOT and LORTOT as Attitude Scales. ... 66

PAGE 6

Page Further Research 69 Applications 73 Conclusion 75 REFERENCES 76 APP EOT) ICES A INSTRUMENT USED TO CATEGORIZE 90 OBJECTIVES AS NATIONALISTIC, INTERNATIONAL, OR EITHER/OR, INCLUDING THE TALLY OF RESULTS 87 B FAKS ITEM ANALYSIS 93 C FLORIDA INTERNATIONAL CURRICULUM ASSESSMENT SCALES SHOWING SUBSCALE STRUCTURE 94 D FLORIDA AFRICAN KNOWLEDGE SCALE .' 98 E WORLDMINDEDNESS SCALE 103 BIOGRAPHICAL SKETCH 106

PAGE 7

LIST OF TABLES Table Page 1 Response Distribution for Open-Ended Questionnaire on African Content in the Curriculum 26 2 Reliabilities and Standard f'rrors for the Florida International Curriculum Assessment Scales and the Florida African Knowledge Scale based on the Pilot Sample (N. = 131). . . 48 3 Raw Means and Standard Deviations for Hypothesis One (N = 33) 52 4 Raw Means and Standard Deviations of All Samples and Variables Entered into the Analyses of Hypotheses Two Through Five 54 5 I n tercorrelat ions of Cognitive and Affective Subscales on PICAS 66 6 The Biserial Correlations Between the 30 Items Selected for Inclusion on the FAKS and the FAKS Total Score 93

PAGE 8

LIST OF FIGURES Figur e Page 1 Subscale Structure of the Florida International Curriculum Assessment Scales 30 2 Item Assignrrent to Subscales on the Florida International Curriculum Assessment Scales 31 3 Correlation and Item Mean Difference Patterns for NATTOT, EORTOT, and INTTOT 67 4 Hypothesized Relationship Between Knowledge of Africa and Attitude Toward African Content 70

PAGE 9

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in P.irtial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CONSIRUCT VALIDATION OF AN INSTRUMENT TO MEASURE TEACHER ATTITUDE TOWARD THE USE OF INTERNATIONAL CONTENT IN THE K-12 SOCIAL STUDIES CURRICUI.UM By Thomas Owen Erb December 1977 Chairperson: Arthur J. Lewis Cochairperson : Linda M. Crocker > Major Department: Curriculum and Instruction Over the past decade educational researchers have established the importance of teacher attitudes for the success of educational programs. However, little was known specifically about teacher attitudes toward international content in the curriculum. Consequently, it v;as necessary to learn more about these attitudes. Since no reliable or valid scales existed to measure such attitudes, this study sought to develop and validate the needed instrumentation, which has been titled the Florida International Curriculum Assessment Scales (PICAS) . Using Cronbach's alpfia as the measure of internal consistency, this study employed three established techniques employed for testing construct validity: Correlates, group differences, and study of change over occasions. Three subscales of the PICAS were tested for validity. After 'fita were collected from 131 experienced teachers, scores on the three PICAS subs'-ale,'^ were correlated with the Wor I dmindedness Scale (W-Scnie) total score and the Florida African Knowledge Scale (FAKS) total score. After this procedure was completed, the scores on the PICAS subscales of these 131 teachers who represented a cross section of the

PAGE 10

gcner.i 1 teacher pcipulat-ion ^^7ere compared to those of 49 teachers from the Southeast and M.id\^est who elected to participate in summer institutes on African studies. Finally, the pretest scores of these institute participants were compared to their posttest scores to see if the subscales of interest could detect change in attitude as a result of institute partiii pation . Ihe Florida International Curriculum Assessment Scales proved to have high reliability and adequate validity to justify their use in further studies. Estimates of coefficient alpha ranged from .87 to .93 for the three subscales under study. Both the International (INTTOT) and Either/or (EORTOT) subscales showed the expected correlation with the W-Scale. The Nationalistic (NATTOT) subscale had a nonsignificant correlation with the W-Scale. None of the subscales produced the predicted correlation with tfie FAKS . All three subscales were successful in measuring a difference between the cross section of teachers and those self-seJ.ected for institute participation. The results were inconclusive from the study of change over occasions. A multivariate F^ test suggested tliat the subscales did not detect a significant change over occasions. However, EORTOT' s significant univariate F test suggested thai: further validation work needed to be done on EORTOT. H
PAGE 11

have potential for basic research into the formation of attitudes as well as into the consequences of holding certain attitudes. These scales may also prove to be useful evaluation devices in either pre or inservice teacher education where an Internationalization of perspectives is an intended result of instruction. Finally, the social studies objectives comprising the items on these scales could be used to help teachers develop teaching strategies for using international content.

PAGE 12

CHAPTER ONE RESEARCH PROBEEM The Probler Ignoranr-e of the world beyond our borders is .1 Inxury that Americans cannot afford. Schools must play a positive role in increasing knowled(;e of the non-Western world while working to (iiminish p thnorent r i < attitudes toward these areas. However, neither preservice nor inservice teaclier education gives much, if anv, empliasis to developing the knowledge and attitudes consistent with teaching from a cross-cul tur.i I or global perspective. Recognizing ttiis deficiency the United Stat<^s (ifficn of Education has funded 80 lang.uage and area studies (cntm-s (iiv,hiding eight African studies centers) at universities throughout tlie country (Easton, 1*^77). As a condit. ion of fvmding, thf'se f enters are to eng.age in outreach work aimed at creating greater interna t ioTial understanding among American public school pupils, teachers, and the community at large. These funded centers have full discretion in developing outreach programs, and several of them, including those at the University of Florida and the University of Illinois, have chosen to emphasize preparing public school teachers to teach African content to their students. Have these outreach efforts been effective in altering, teacher attitudes toward the usefulness of African content to meet accepted social studies objectives? Prior to this study there has been no valid means to assess such chnn^e in teacher attitudp.q.

PAGE 13

Consequently, Lhis stndv was tlesigiKMl I (> dovlop ,i r"liablp and v:\lU} method of assessing teachers' attitndes to\7ard Afrir.-in content" in tlic^ curriculum. The resulting instrument was intended t" he one that could be used to measure change in teacher attitudes as a result of preservice or inservice experiences ciesignefl to broaden teachers' percejitions of the relevance of African content. Tn addition, tlie instrument was intended to be useful for more basic research on attitudes, their formation, and conseq uences . This study was confine'd to validating an instrnnient specifically for African content ratht^r than international, content reiierallv. This decision was made because the concept of international contfnt was rattier vague since it subsumed many geographical areas or divisions ci f the world about which individual teachers wi-re likely to have diverse opinions: I,atin America, Europe, Canada, Asia, the non-Western wo t 1 d , the I'hird V.'orld, the Communist world, and so forth. I'or gr(Mt;er cfMueptual (^laritv it seemed necessary to define a specific international re,i',ion for this validation study. Africa was chosen becaus(^ of the researcher's familiarity with programs aimed at improving African studies in the public schools through teacher inservice education. However, any other international region might have served equally well. Need As never before Americans need to have an nude rsl and i nj:, of human experience th.it transcends parochial perspectives. Do'-jsions made in centers of power in one area of the wcirld can effei I people on all continents. Edwin 0. Reischau^'r (1973) states the case for international education in this way:

PAGE 14

-3Refnre long, liiiniaiiity will fa(.:e ninnv d i f f i t-nl t ies lli'it c;3n only be solvfl en n r.lobaL scale. I'
PAGE 15

Because of inndequafe preservice training',, in:;"rvic.e riMn]>oncn Ls are needed to help teachers p;alii the ccnipetenc i es necessarv to teach ahont foreign cultures. Outreach proprains sucii as ( (lose offored hv the African Studies Centers at the Universities of Florida, [llin(vis, .ind Wisconsin give teachers the opportunity to [lartic i pa tc in workshops and institutes aimed at increasing their substantive knowledge of Africa and tielpi.ng, them expand their perceptions of the value of intej-.rating African content into the social studies curi'lculum. Teachers are exposed to films, lectures, and discussions (mi African topics; [)aT-t i < ipa tc in modified African cultural experiences; meet African scholars and students; and are introduced to curricular materials and teaching sti'ategies which are designed to improve their teaching of Africa. Naturally, the people who administer African f;tudies outreacli programs are interested in teacliers' knowledge of, and at<;itufies tciward, African content in thc^ curriculum. A scare 'i of thf literature revealed no studies of teachers' attitudes toward teaching ahouf: Africa. A private communication with Prof. Marion d. Rice of the An t hropci logy ('urriculum Project at the University of Georgia has confirmed that, aside from content-specific instruments of questionable vallditv, there was nri reliable, valid measure of teachers' attitudes toward Africa in the curriculum. At the University of Illinois workshop in 1974, transactional analysis was used to evaluate the interaction among, worl^shop participants and the attitudes of the workshop participants toward the workshop f(.)rmat and its subject matter content (Schmidt, 197'3). ihn./ever, the instrumentation failed to produce data that would a 1 1 ov; an objective evalaiation of the teacher attitucies involved. In fact, the tta'-h(^r attitudes studied in tlie Illinois project were not directly related to tiie currlcul.ar

PAGE 16

issue of us iiig African material in tho classroom, iMit focusod on the Internal, workings of tho workshop itself. Ilius tliern was a need to design, develop, and validate a rel iahle attitude scale that would he suitable for assessing the formation of attitudes, the influence of attitudes, and change in attitudes toward including African material in the curriculum. Without such instrumentation no research or objective evaluation could take place to assess the impact of inservi.ce education on teacher attitudes. I)ef ijii_tJ_ons Atj:j.tjKi_e Attitude is the psvcho 1 og i ca 1 construct which refers to the composite of ail feelings about, and predispositions for bidiivior toward, some object. Although attitudes are covert, tliev find evert expression in the form of verbalized opiiiifms thus making them measurable. Thoii,B;h based on cognitive processes, attitudes are primarilv affective in nature. As measured by attitude scales, an attitude is represented by a score which corresponds to the point along an underlying conl. inuTim which defines the degree of positive evaluation attached to a specific referent by an individual (see Fishbein, 1967, pp. 257-260; Oppenheim, 1966, pp. 105-112; Osgood, 1967, p. 112; Shaw & Wright, 1967, pp. 2-7; Thurstone, 1967, pp. 77-79) . African Content The attitvide. referent in this study is the inclusion of African content in the established K-12 curriculum. Ihe use of African content

PAGE 17

is not restricted to an arcn studies approndi v/herc students Foeus in turn on various culture areas of the world. Airiciu confent refers to any materials (games, recorriings, readings, case-studies, maps, films, biographies, activity pacts, learning centc>rs, speakers, fielii trips, kits, etc.) whicii are used with students to achieve some curricular objective regardless of th(> organizing principle under which the objective fits. A teacher's attitude tciward including African ccuitent is of interest regardless of whetlier the franieworl-. is wc)rld geograpliy, world cultures, peace studit\s, future studies, development studies, global studies (see flecker, 1971), or any other conceptuai scheme under which African content could be sulisuiiied . Africa n S ^u d i t^s_ On tre ; i_c_h Inst i tjut^e s For the purposes of this studv African studies outreach institutes refer to one two-week summer institute and (>ne four -week summer institute aimed at expanding teacher awareness of Africa and dc"veleping, new approaches to teaching, about Africa. I'lie institutes include three components. The first uses formal presentatlcuis to disseminate Information about Africa. A second involv(\s the teachers in cultural activities including African music, films, art, and cuisine. I'he third ciimponent Involves the teachers in developing materials and leardiing sti'ategies for teaching about Africa. Inter nal Cons is tency Reliability is the accuraiy with which a test measures that which it measures. A reliability coefficient expresses the squared correlation

PAGE 18

-7between subjects' obso>rved scores on an instrument: and their true scores on the trait beinj^ measured. l^Jlien individual test items liave h if;h correlations with the total test score relative to their item variances, the test of which they are a part is said to possess h i r,h internal consistency which is a measure of reliability if liie test is intended to he homogeneous, as in an attitude scale designed to measure onlv one dimension (see Cronbach, 1967, pp. 112-L67; Kuder & Richardsnn, ]'^^7, pp. 151-160; Lord & Novick, 1968, pp. 61, 87-95, 139-140, 211-21.'*, 331; Remmers, Cape, & Rummel, 1965, pp. 129-130). C onstr uct _Va_lldiJ:_^ Construct validity is the degree to whicli a test measures the construct or attribute that it is designed to measure. Ihe construct or attribute is not operationally defined. llif re f ore , tlm problem is for the researcher in construct validation to identifv the construct or constructs which account for the variance on a test (see ('ampbell & Fiske, 1959, p. 100; Cronbach & Meehl , 1967, pp. 243-270; Lord & Novick, 1968, pp. 261. 278-279) . C o n s_t_ru c t Val id a^t j. o n Cronbach and Meehl (1967) describe a variety of techniques that are used to establish the construct validity of a test . Tliey conclude that the problem is not to determine that a test is valid for measuring a construct but to state as definitely as possible the degree of validity the test is presumed to have (p. 255). A major test: of construct validity that was used in the current sttidv is tlie croup differences

PAGE 19

method (p. 251). Constrnct vnliditv can he cicnicms t r;i tod if proiifs which sliould differ in the construct do in fact have different nieaTi scores on the Instiuraent. In addition, to establish constrmt validitv it must be determined that the ccinstruct heinp measured lopic.iilv relates to other similar constructs (p. 252). Consequently, the scores on the attitude scale should correlate v/ith measures of otlier constructs related to attitude toward African content in the curriculum. As a final tc'st of construct validity a study of change over occasiotis (p. 253) was conducted to see if the attitude scale could d(^tect chaiiRc in attitude toward usinp; African content in the curriculum as a result of participation in summer institutes aimed at increasinp, teacher competence in the area of African studies. In order to test the construct validity of n uov instrument it was necessary to state specific hyj'otheses concf^rni nj', the nutconies of the proposed analyses. If the analyses confirnK^l the hypottieses , then construct validity would be established for the instnimi-nt. In the hypottiesci one had to state the expected outcomes and the criteria by which judgments would be made about whether the hypotheses were confirmed or not. Listed in the next section are the set cvf hypotheses which if confirmed bv statistical analyses would establish the construct validity of the Florida International Curriculum Assessment Scales (PICAS), the name chosen for the attitude instrument developed in this study. The Overall Hypothesis provided a framework within which spec i I ic tests o[ construct validity could he designed. Operational Hypothesis (iiie did not directly deal with establishing (f)nstruct validity but with resolx'ine a question as to whether or not teachers' ratings of the relevance of African content would be affected by the nature of the test format. Sfiec I f i cal ly , if

PAGE 20

teachers were asked tc rate the relevance nf Arric.ni i-nntent alnnp with the relevance of hatiri American or (•'.uropean c(Mi(cnt, ^-.'oiild tiie i r ratin;,',s of Africa be affected? Consequently, three foT-nis of the FICAS were developed — one asking teachers to rate African oiMitMit oiilv; a second, African and Latin American content; and a tli i rd , African and f.uropean content . Hypotheses Tv.'o throiis^h Five were designed to test the construct validity of the varions FK'AS subscales. Ilvpothesf^s Two and Three state expected correlations betwc^'n the FICAS subscales and related constructs: worldmindednoss and knov^ledge of Africa as measured bv the Florida African Knowledge Scale (FAKS) . Hypothesis Fonr is based on the gmnp differences technique to test whether the instrument coiild d(>trrt ,iii expected difference between a pilot group consisting of a cross section of teachers and the institute groups consisting of teaclu'rs wIim \.;ere self-selected to participate in summer institutes on African studies. Hvpothesis Five was designed to test whether or not the FH'AS c(mi1 J detect a cliange in attitude that was expected to occur as a result of pari fcipaticni in the simimer institutes. Hypo the ses O verall H y p^o 1 1 l es i s There exists in teachets an attitude toward tea'diing, about Africa that can be measured bv having teachers evaluate tlv i iiipor t ancf^ of social studies objectives and the relevance of African content to actiieving these objectives.

PAGE 21

-10Operationa] Hypotheses One. There will he ne Rij;nificant: d i f f itpiups (p • .23) Tiiii'ni', the mean scores of suhjects in tiie p.ilcit sample assip.nid t i> three different forms of the Florida International Currirulnn Asse.^^snicnt Scales. Two. There will be a significant positive cnrrplation (p ' .05) lietween scores on the Florida International r:\irriciiliirn Assessment Scales and scores on the Wor 1 dmi ndedness Scale. Threja . There will be a positive significant ' p .05) correlation between scores on the Florida Fnte rna t iona 1 Cnrricnlnii' Assessnent Scales and scores on the Florirla African Knowledge Scale. Foujr. Subjects who elect to participate in a • niriciilurn wcirkshop on Africa will have a s i gn i Fi c.ni 1 1 v higher mean sc(iie (p .0"')) on thc Florida International Curriculum AssessmcMit Scales thin a comparison group of a cross section of teachers. Five. The mean [losttest scor<-s of a grouji of lea(diers enrol led in an African studies center summer institute will be s i gji i f i cant 1 y (p < .05) higher than their mean score on a pretest of the Ffc^rida international Curriculum Assessment. Scales. Si^n i fi_9J^ 11?. The validation of an attitude scale to measure t(achers' attitudes ttward African material in the curi iculum is expected to open the door to a variety (if research appl ic.atiiins. I'he instrumetM , coupled with a knowledge test, can be used by African studies outieatli programs throughout the United States to asspss the impact of tfieir efforts. The attitude and knowledge i^ns t ruiiietit s can be used as j^re and post tests

PAGE 22

-11to evaJunte change as a result of workshops, sfMiiiivits. oi' coiitse tronLments . In addition to procrani evaluation, the newlv . rins I rue t ed i ns t Dinn'ii t can be used to measure attitude as either an indcpi'iident or dependent variatjle in more basic research studies. I'nr ttiosi^ v.'h'i are d(\sipnin^ outreach nnd/or inservice components for teachers, it may be well to know what factors relate to teachers' attitudes about Africa in tht^ f:urriculum. Knowing wtiat factors most influence the formatioii of I hese attitudes, designers of teacher education prop,rams can take this information JTito account. Other researchers may be interested in u?:inf tlio scale to relatr attitudes and other variahlr.s to pupil per f r>rmance . ihe hvpo t lies i zed importance of teacher attitudes for pupil growth could be empirically tested . The design of the attitude scale in the currfuit p^'o'icct could easily be adapted to measure attitude toward other (-ontent areas .is well. The adapted instrument could assess the perceiv'MJ relcance of anv designatc^l content domain for ach ii'vini', social studies (di jec t i ves or even general education objectives. For althougli this study is focusinc on social studies objectives, both Engje (19fi5) and Metcalf (I'h'^'^) maintain that social studios objectives are often ind is t i ngui sh.'b 1 e from those for gene ra 1 educa t i on .

PAGE 23

CHAPTER TOO Ri;vrra%' of LiiHRATaRE This review of literature deintms tr.i tes tlie sduilnrly i-oncern for internnt iona.l edncatien in its many man i t'es tat i ens . llewevcr, in spite of that coneern, very little is beiriR done in (^ir pub] ic seliools or teacher training institutions to promote international edncation. The second theme developed is that teacher attitudes tovard echicational processes as well as toward students themselves ha\'e been demonstrated to plav a major role; in the success or failure of 'duca t i una 1 prc>granis. Finally, the methodological literature explicatlnp issues, in attitude measurement is reviewed. ilie unifying factor is tlie inportance (d teacher attitude for the success of international cdii'-ntion programs. 1 n t ( ^ ni aj^ i^o n ai. E d u c a t i o n The Theor etica l Need The occiirrence of a sei^ond world war provoked some social scientists into conceptvia 1 izing a "moral equivalent to war" (('. Murjiliy, 1945). Central to their prescriptions for a new world ordct was the couviction that educatiiin needed to be freed from n.i t venal i sm to become worldminded in orientation (pp. 2'(0-242). By this the social scientists meant that education should focus on pieparing student;s for intelligent v.'orld citizenship based upon demo'-racy at home (p. 242). .lust four years later Maria Montessori published her Educazione e Piice (1949/1972) calling for -12-

PAGE 24

1 1cduoatioii to enrournpo the spiritn.Tl aivl nmral (i.'v 1 npmen t nf 1 lul i v i cltia 1 f; in an effort to promote a [nacefnl world. Grf^wine out of the breakdown of world order in Itie 19A0's were the intelleetnal threads rallin.r, for greater ui tenia t i ona 1 understand i nj.; which were to be pick<_^(! up bv edueators in tiie lOfiO's. At that time there arose a tide of criticism amon^ Rot:ial studies (vhieators a,i;ainst the ethnocentric orientation of American schoo 1 i nj',. The basic organization of social studies cairricula had not chanjied sinc.:> tdie 1 '^ 1 6 National Education Association Conimi'tee on Social Studii^s of the (loinni i ssi on on the Reorganization of Secf)ndary i'ducation leconiiiiended a grades si^ven to twelve scope and sec|uence which r-entered on M.S. and bnropean history and civii^.s. Relatively new social sciences such as sc^^'iology and anthropologv v.'ere excluded from the recommendations as was any menticMi of non-Western history or ("Iture (MassiaJ/is S* f>iy,. I'^'iT-i). Masia (!9fi3) in his studv ci f schools accredtied by the Morth Central Assoriati(in of Colleges and Secondarv Schonls confirmed that high schools continued to ignore ncin-Western areas '^ f the world right into the I'^fSO's. Separate courses dominated by American and world (F'jiropean? ) histciry (diarac tor i zed the curriculum (Massialas f< Cox, L966; MoTelanrl, ]'''(')?). By the early 196()'s several writers were calling for cffc^rts to internationalize tlie social studies curriculum. Harland Cleveland (lOCiO) suggested that an undergraduate^ education in foreign affairs should develop cultural empathy or skill in understanding tlie intier logic of other ways of life as well as the predisposition tc refrain from condemning these ways of life because thev are different. If promoting world peace and internation.il undcrs tatid ing were to become educational goals, Stearns (1966) asserted that teacher educatioti must

PAGE 25

identify nnd rlnrify tho information, nt: L i t ndcs , niul skills that are related to w(irl dniindednosf; . tie ur;;ed tho (.s tah I i .sbni'Mit of an indopendont international curriciilnri lai'oratorv, world '-ollr!;!^ leuteis, and a foundation for international education. in ]9C)fi, tlie U.S. Congress gave impetus to the movement by passing tlir International Education Act. This act called for developing knowledge of other 'imntries to promote mutual understanding and to strengthen relations hi'twooTi the United States and other countries (quoted in Klassen fv Mtiore, l')f)8, p. 3). Edwin 0. Relschauer (1973) added his stature to tlic debate by criticizing the past orientation of enifibasis on Western Eairope. Me urged that the various cultural dimensions of the non-Western world be given more emphasi s . C^onceriied scholars and educators have ,1 1 teinp( (>(1 to describe international education and non-U'estern studies as they vjould apply to a school curriculum. Bohannan (1973) defined i nt e rcu I t mt a 1 education "as a structuring of learning experiences that '.\i I 1 1 liclfi bnth students and teachers unrlerstand and use concepts for undo rs t a.ivl i ng arid working toward solutions of individual and intergroup problems -local, international, and world wide — that arise from cultural dlv(M'sily" (p. 31). He offered a conceptual framework for i ntercu! tural education built on the view that culture is two-ti(>red. The "mac roc ul t ure" is based on power and is large-scale, shared, and worldwide. "Mic rooul t ure" is based on love and trust or special interests and is small-siale and family and community oriented (p. 19). He argued that educaturs needed to distinguish the two cultures and rebuild schools accordingly in order to (a) enable everyone to Tieg^tiate the stairs between macroculture and mlc roc 111 I ure equally well In both tlirectlons and (b ) teach people about

PAGE 26

15the nature of the manv m [r rocu 1 lures in otdrt" to it'Iik c fear of (utc for tlie next (p. 20). He presented three key '.-. lines whifh should uiulerlip education In the two-tiered framework. At the maci-o love! the principal value is "equality of opportuTiity for all." He sup.r^ested that "live and let live" shturld govern affairs within and hetween mi' I'ocu 1 tures . Thirdly, equal access to tlu^ stairs between micro and macroculture should be guaranteed. Bohannan insisted that his approach is neither antipatriotic nor eultiirallv relativistic sent i men tal ism. In the early 1970's global studies became the focus of several educators interested in international education. f'arion (l'?71) proposed peace studies which concentrate on the Issues of humm aggression and conflict resolution from thc^ in teryiersona 1 to (he i uternat i oni 1 level. In a monograph Glc)_bal Development Studies {\^7\) the Manag.ement Institute^ for National Development outlined a senior h i eh cuTri< ulum that g.oes beyond the transfer of' kno\.'ledge to the chaneinr, oi the students' perceptions of reality. Consr: i (nisness raising. <'ind values examitrition in relationship to interna t Ion 1 1 development issi.ies ceusi itute major objectives in this program. Becker (1973, l^y^) ad'/oc-a ted a rurrlculum which helps students to see the v^orld as oLhei-s see it, to beeome aware of and adept at using alternative sources of i n fi')rn'at ion and evaluation, and to develop a willingness to consider competing views of reality. Boyer (1975) added a futures component to tln^ prociss of dc've 1 op ing a world view among students. Dahlherg (1974) and UNl'SCd (The UNI^CO Associated Schools, 1975) conceptualized a global studies approach based on the rel -a t ionships between people and tlieir en\' i t oninent s . ('oncentrating on the theme of Interdependence Almg,ren au'l Custafsson (1974) suggested that Intern.a t iona I education is a re't|uisit(^ for human survival.

PAGE 27

-16A majority (if Anioricaii public rcIiooI Lrnchers share a cnticern for greater international undet s i and ing . F i gh t v jjorccnl: \vho responded to a recent National Education Association questionnaire said iu^tter cociperation among nations was a key issue in promoting a peaceful world commtmity (McCarron & McCune, J.974). Within the broad framework of international eiucation several recent books have been written to s',uide teachers and curriculum |ilanucrs in using African c::ontent. E..I. Murphy and Stein (197?) sugj',ested six broad reasons for including Africa in the social studies cui ricul um. Under the headings of superficiality and e thnocentr ici tv , i^laU (1977) has categorized .30 problems which currently exist wit.b the treatment of Africa in American textbooks. The Af r i cnu-Amer i can Institute (Collins, 1970), Beyer (1969), and Willmer (1975) provided practitioners with both theoretical considera tic^ns and concrete suggestions f(;>r iticluding an African perspective in a social studies prop.ram. Existing Fr og rjims Over the past two tlecades, intellectual leadership has offered a variety of conceptual frameworks for incorporating international studies into the curriculum. A key concern is to what extent these S'^holarly ajipraisals and recommendations have found their vkw into the curricula of American schools and into the instructional strategies of its teachers. John Lee (197A) offered a somewhat hopeful view. M" detected several trends in the late 1960's and early 1970's which sug.gisted that curricula were being internatiornal izeil : International education was beginning earlier in the child's schooling, the concept: of the world had been broadened to include the non-West, and there was a shift to the study

PAGE 28

-17of the X'/nrld's people as a society (a global view as opposccl to the study of one nation after another) (p. 1A9). Tlu-'URh siinii> en'Mmragi ng shifts may be occurring, Rose Hayden (197fi) said the resiionse of American pviblic education to the challenge of pre[iaring citizens to respond intelligently to global problems was inadequate. She was particularly critical of state education agencies which she claimed did not perceive tlie need to provide leadership for globalizing public (nlucation. Tn spite of some global curriculum development, some inscrvicc training on international topics, and increased stress on international training in teacher education, too little has been done to internationalize public instruction or to evaluate its results against glcibal goals and objective;^ The Aspen Institute has formed a national coiiinlssion on (oping with interdependence (Morehouse, 1975). Die institute is concerned v/ith two questions: First, to what ','xtent do American i nst i t ut ions perceive the predicament of interdependence and its implications; second, what new attitudes and arrangements may be required to enhance the capacity t^ f Americans to cope with it? Ihe findings revealed that e(bicational institutions are ill-equipped to deal with the issue of interdependence because it is perceived as a national concern while schools are run as a state and local responsibility. For this and cither reasons, it was concluded that much of the burden for creating a civic literacy on Interdependence issues would have to be born by institutions other than formal educational ones. Nonetheless, some schools and other supporting ar.oncies liave made attemi^ts to change curriculuni and reorient inservicc to implement international education. Several school districts have developed Internat i oiial studies programs of one nntunor ancithei' (Freeman, 1.974;

PAGE 29

-L8A fri can Studies Handbook, 1971; World History Scrios: Africa, 1972). State departments of education (Jones, 1970) and ntiier ajiencicjs (Oswald, 1974) liave contributed their efforts to assisting public schools to improve international studies programs. ilie Af r ican-Amc-rican institute (Teaching About Africa , 1972) and the African Studies I'r(\p,ram at the University of Illinois (Schmidt, 1975) have made iMij'.oing efforts to improve African studies throuf;li teacher inservice education. Tea_c_he r A 1 1 i t iides^ and^ _'ni^e_ir Me^asurement A crucial issue in assessing the role teacliers play in the instruction of children is to what extent teacher attitudi's influence the outcomes exhibited in cliildren. In conjunction with that issue, is the problem of measuring teachers' attitudes and their re 1 a t i(Mish ip to otiier variables. Wiseman (19 7 3) believed that the slnjj;li' iiK^st significant outcome of educational research in the last decade was tlie realization of the power of teacher attitudes and t(>acbcr exper tat i ons . Much of the research Into teacher attitudes li.is focused on their relationsliip to various educational issues and practices. In an experimental study, Homme (1968) reported a failure to improve teactier attitudes toward programmed instruction, apparently because of lack of coTi)Tiiunication between experimenters and teachers and lack of optinium classroom conditions. However, Macias (1969) fovmd that m.'W "Spanish for Conimunlcation" materials designed for programmed mastery (.> f instructional objectives im(iroved teacher attitude tov;ard languaje instruction, as well as changed their teaching methodologies and student performance. A third study dealing with teacher attitiides toward some aspect f ) f educational practice was conducted by Waimon, Bell, and Ramseyer (1971)

PAGE 30

-19v\rho hoped to show that teacher r.-Hid Ida tos trained i ti the mierciplanning technique would score hi^;her on tests of teacht?r f^ I f ec t i\'eness but remain unchanged in attitude toward students and KMchin;;. The stvidy found that teaching teaetiers to contrtil cogjiitive structure variables increased their students' reasoning ability hut fiad no effect on recall not did it affect any measLire of teacher attitude. lin's last finding was considered a positive one because the experimenters, were attempting to avoid change to undesirable attitudes on the part '>f teacher candidates. A study by Maxin (1974) showed how negative tt^Tchcr attitudes could block progress in educational practice. He ftnmd that toaciiers had little knowledge of, and negative attitudes toward, educational research. Consequent I V , tliey shied away from its finding,s. Okey (1973) studied the effects of learning UNm.ihi's mastery teaciiing strategy on teacher candidate's attitudes toward t'-sts, r,radcs, and diagnostic teaching. lite experimental group shou'ed significantly positive attitude gains in this study. In a one-siiot experiment in which teacher attit\ides toward the effectiveness of different matli programs were supposedly manipulated by having teachers read biased statements about the programs, Brager (1970) fijund no significant difference in students related to teacher attitude. A more sopli i s t i ca ted longitudinal study by Beauchamp and Conran (1975) was set up to measure the effects of the operation of a curriculum engineeritig system. A causal model employing patii analysis (see Blalock, I9fi''4, 1971; Ileise, 1975; Van de Goer, 1971) was being used to examine the r c I a t ionsh i jis iMrHween leadership, curriculum engineering, teacher attitudes, teacher per f (irm.ince , and student performance. F'rey (1973) in a survey study of 406 elementary, middle, and high school tencliers attending a summer session at Northern

PAGE 31

-20Tllinois University fmnui t:h;U Icichers knc\.-i much nbmit hetinvi ora 1. objectives but held rather neutral attitufles about tbiir e f fee t i veness in pupil performance. Bogatz (1970) fecund a positive re.l a t ionstii p between teachers' attitudes toward the effectiveness of instructional materials and actual pupil achievemonl . Researcli on teacher attitudes toward subject matter iiad been confined mostly to the subject of science. Oshima (l^fif)), in studvinp, the difference between lecture-demnnstra ti on and individu.i] investigation as approaches to teaciiin^; science methods courses to elementary school teacher candidates, measured attitudes toward science as well as confidence toward teacliiiAR science, achievemc;nt in sc'enc', and student teaching behaviors in science. Only tlie confidence Viiriable siiovved significant difference hetv;een experimental and conVrnl croups. Sliripjey (197A) studied attitude toward science and its relationship to knowlerige of science. tie found a low positive correlation of .,'.5 between measures of scientific knowledge and attitude toward science. Since the coefficient of determination was .06, he questioned the relationship betwofni knowledge and attitude. However, Billeh and Zakhatiades (1975), finding a .25 correlation between science attitvjdes scores and grades in science taken as a measure of learning, concluded that Icnnv.' I edge of science and general (exposure to science iiad a prisitive influence on attitudes toward science. The previously mentioned Almgren and Gustafsson (197/4) study looked at the relationsliip of teacher attitudes toward teaching about international questions and students' attitudes. llu'v found a positive relationship . An important set of research on teacdier attitudes has focused on the relationship between teachers' attitudes toward their students and the

PAGE 32

21pupils' subsequent achi e vemcMit . Aron (197S), Clark m75), and Cecil (1971) all found a positive relationship between te.ieber attitude tow/ird students and pupil achi evenn^nt . These studies have made use of one or several () f the attitude measurement techniques developed and refined since tlu^ 1920' s. Thurstone (1927, 1929, 1931, 19(S7) was the first to take attitude measurement out of the survey stage and makt^ it more rie.orous by intmducinp tfie eqnalappearinsj interval technique. At the time that Seashore and llevner (1933) were modifying; Thurstone's techniques, I,ikett ('19ft7, i970) proposed the summated rating method for measurinp attitudes. Subsequently, Guttman (1944, 1947), and Osg(iod (1967; dsgood & Suci, 19SS; Osgood, Suci, 6i I'annenbaum, 1957; Snider & Osgood, 19b9) devised nli:ernative systems for attitude measur'-ment . Edwarcks (1957), Op[ionheim (19b6), and Shaw and Wright (1967) have each made attempts to synthesize the previous work in attituche measurement and provide r.nidelines for developing attitude scales according to the different tccliniques available. Two studies have attem|:>ted systematic comparisiMis of Likert and Thurstone techniques of scale construction and scoring,. Troviding a critical review to previous research, Seller and flcngb (1^70) concluded that the lAkert scoririg method w£is superior to the Thurstone method with regard to the value of the reliability coefficient obtained for a given number of items. Using l,ikert construction and scoring technicjues, one could usually develop a scale with a .90 reliabilitv coefficient with only 20 to 25 items. It would takce twice as many Thurstone constructed and scored items to obtain the same reliability [p. Wl). Jaccard, Weber, and Lundmark (1975) used a multi I ra i t-mul t imct hod matrix (see Campbell & Tlskc, 1959) to comi^are the tost-retest reliabilities and

PAGE 33

divergent rtud convergent validities of four nipthod,'^'. ni attitude Fieasnrement: Likert, Thurstont^, Osgood, and Cni1 ford's selfrating. Tliey found little difference iti the coniparat i ve re I ialn' 1 i t ios , the method variances, or in the divergent and convergent WTlidities of each method. Whatever the merits and demerits of each method may he, these differences cancelleiJ out when put to the empirical test. Thus any of the .ihove techniques might he applied to assessing teaclier attitudes towartl social studies curricula, or specifically African content. S Liinma r y _o f Re^s e a r c 1 1 The changing nature of tiie modern world has raised concern among educators tliat curricula and teaching practices need to hecome more internationalized. Some scattered efforts at curricMl.ir reform have been made. In addition, scattered attempts have beeti made to prepare teachers to use international content in their teachin.t;. Yet little research has been done into the effects of teacher attitudes cm sul^seciuent student learning of either concepts or attitudes relating to international issues . However, for fifty years educational psycholcn;! st s have heen developing and refining a number of techniques for measuring attitudes. These techniques have been used 1n recent years to studv teacher attitudes toward students, varicuis subject niatters, and diff'rcnt eriucational practices. What is needed now is a means of measuring', teacher attitudes toward teaching about fiireig.n cultures and i ii t(M;nat i onnl prciblcuiis. Making use of accepted techniques of attitude measurement, this study has undertaken to devise instrumentation that could va I i d 1 \' measure these teacher attitudes.

PAGE 34

CHAPTER THREI MI'THOn In establisliing the coiistriirt vnlidiLv of Iho new scale to rneasiire teachers' attitudes toward African content in the rurriculuin, three approaclies to construct validation were used. Fi.rst, correlations between scores on the Florida Internationa] Curricnhini Assessment Scales and other measures of similar constructs were examined. A previously developed knowledge test of African content was refined for use as a criterion in this study. In addition another attitude scale mcisuring attitudes similar to, though not identical with, attitudes towaid tcaohing about Africa in the classroom was also used. The total scale scores from these two instruments were correlated with the newly developed FICAS subscales to establish the first type of construct viliditv. Second, croups which should have differed in scores on the FICAS subscales were also tested and compared to see if expected differences had been detected. The mean scores from the pilot sample were compared to the two summer institute samples to see if the newly developed subscales cc^uld detect an expected difference in attitudes betv^/een the pilot and (ither two groups. Finally, the pretest scores of the institute samples were compared to the posttest scores to see if any subscales could measure a difference in attitude expected as a result of pa rV i c ipa t ioii in tho instilutt-s. This chapter includes the details of the development of the FICAS. A discussion of the procedures used to establish its reliability follows a description of its development. Finally, measures taken to confirm -23-

PAGE 35

the construct validity of the FTCAS are dolineatecl aloiiji witli n discussion of the other instruments usocl In the validation [irccess. A ttit ude Scale Construction Purp ose The purpose nf tliis study was to devcl(ip a reliable instrument with sufficient validity to allow it to be used in attitude research and in curriculum and instruction evaluation. An instrument was desired to yield scores on teacher attitude toward African content v/hich could be analyzed to learn more about the relationship cif this attitude to other constructs and observable behavior. II was expected that the new instrument would serve a useful purpose in evaluating pre and inser\'ice efforts to instill an international perspective in teacliers wi tti regard to their own teaching . Scale Dimensions In constructing an instrument to measure attitudes towaril African content in the curriculum several steps were taken. The first concern was to clearly define the dimensions of the construct under consideration (Gardner, 1975). The identified construct had tn be unidimensional so that all scale items cc>uld be summed to prothice a mcNiningful total scale score ((Gardner, 1975; Thurstone, 1967). Sucli a scale score would be uninterpret able if a single linear ccintiiuuim did not underlie the attitude being measured. Attitudes are assumed to vary in quality and intensity along a continuum from positive tlirough neutral to negative (Shaw f< Wright, 1967, p. 7). If, in fact, some scale items measvire dimensions

PAGE 36

Lndepen(]ent of the m.ijor uivlcr 1 y i iii; cons trur t , t\\v rpsMiltinp; nttitude score will he meaningless with regard to the attitude th.it was intended to be measured. While pains were taken to insure tliat the attitude to be measured, attitude toward the inclusion of African content in sctiool curriculum, was not confounded with other attitudes such as ati itude toward AfroAmerican studies or perhaps attitudes toward traditional versus inquiry social studies methodologies, care was also exercised to include all aspects of the underlying dimension (Oppcnh'7' i m. ]9h(^ . [> • 117)A teacher's attitude toward tivaching about Africa in the curriculum miglit be made up of feelings about priorities in the curriculum, feelings about the school's role vis-a-vis cross-cultural educaticn, feelings about Africa's relevance to the curriculum, or anv number oJ other aspects. These various aspects had to be identified and measured if a val id scale were to result from this study. An analysis (^f responses to open-ended statements about African studies in the curriculum was used to determine the relevant underlying dimensions. A sentence completion c|uestionnai re with three items on it was used t:o elicite teachers' views on African content. The sc>cond item on the questionnaire was the most pertinent for formulating the underlying aspects of the construct which the researcher was attempting to measure. This item began "African content in tlie public school curricirlum should. . , ." After complecting this sentence, teachers wt^re asked to respond to the second part of the item whicdi said simply ". . . because . . . ." Thus in addition to open-ended jud^tnients on Afiican content, reasons for tliese judgments were also solicited. Also included in the questionnaire were two other Items: The first of tliese was simply "Afrlc . . . ," and the second was "Clobal consciousness. ..."

PAGE 37

-26This questioTmalre wan admini.s t er'.'d Ip 100 ptp and insorvice teachers. The responses were catPRori /ed in an ofrorl to dpt(-nnine the dimensions underJyinp attitude toward Africa in thicu rr i mi 1 nm (see Table 1). It appeared from analyzing these respons(\s that the relationship of African content to (Cognitive object ives is the major undt^rlying dimension of attitude toward African content. \\'hen combined with responses related to affective objectives, over 78 percent of thf' responses relate to curriculum objectives. Table 1 Response I3i s t»-j button for Open-F.nded Ones t i onnaire on African Content in tfic Curriculum Category Cognitive Objectives Affective Objectives Relative Importance or Time Considerations Intrinsic Interest All Other Responses Percent of Responses r. 7 . 1, 1 . T 3.7 6.6 No_t j2 . Number of subjects eqtralled 100, but the nnmf>cr of responses equalled 106 because of multiple responses. Source of Items The next steji was to d(>vclop a ptM) 1 of items which ci^uld be piloted in order to construct an adequate scale. These items had to differentiate between those teachers who hold positive and those who hold negative attitudes toward the tibject in question (Tikert, 1967, p. 92). The items

PAGE 38

-27needed to be statements in which respondents could roco^^ni^.e real viewpoints and feel strongly about them one wav or ibi' othip (Op|nMihetm, 1966, p. 11-t). These items had to cover the ranpe of aspects imderlvinK the attitude and, in a traditional scale, the range of intensity from negative to positive. For the attitude tliat this study attempted tci measure, using standard Likert-type statements presented a potential problovii of interpretation. Since relationship to curricular objectives seemed to be the major underlying dimension of attitude toward African content in the curriculum, most items in the hikert-type instrument would contain st:atements x-^hich related African content to objectives. Two examples would be the following: "The study of Africa is nec(2ssary to promote international understanding," or "The study of Africa ne(>ds t (i bo included in the study of world geography." If an attitude is a predisposition to act in a certain way toward some object, then positive rrf;pcnses to those items might not be measuring a teacher's inclination to use African content. For it is possible that a teacher coulrl agree that Afiican content was important to achieve an objective but believe at the same time that the objective is not too Important. Such a situation could produce a misleading item response. To overcome this potential problem, the current studv used statements of objectives as the items in the instrument. Each teaciicr was asked to rate the importance of the objective as well as ratethe relevance of African content to achievin
PAGE 39

-28A) . Examplt^s jncLude these: "Sfu'lentR stionld know .iljout: ttie impact, of tecfinology on modern 1 i f o in America," "Strudcnts .';liniil,l liavo a knowie'lj^e of the structure of American governmf>n t , " and "Slnrlents sh'MiM support efforts to protect America's environment." Other ohi(H:tives are international in orientation because they require (ontent from outside the U.S. in order to develop them. Such objectives would he the following: "Students should acquire wcu;] dmindodness , " "Students sliouitl comiirelieml the gulf between rich and poor countries," and "Students should undorslanc different ways of life on different continents." Still n third group of objectives are those whose achievement does not dii'tate the [rational or international focus of content in order to accomplish them. This third group of objectives could be developed with nither American or international content or both. Examples of this third type include "Students should understand the balance of nature," "Students should understand the exchange of goods and services," and "Students should appreciate artistic expression." In selecting the objectives for iiK;lusion in t.lie instrument, the first step was to examine the responses to the open-ended questionnaire given to pre and inservice teachers and then to review social studios methods texts and school district objectives statements. The following sources were checked for examples of both cognitiveand affective objectives: L.F. Anderson (l^bS); D . I, . Rrubaker (1973); L. Ehman, H. Mehllnger, and J. Patr'ick (1474); S.H. Kn^Jc (1963); d.R. Eee (1974); B.G. Massialas and C.R. Cox (1966); K.J. Murphv and 11. St(M'n (1973); D.W. Oliver and J. P. Shaver (1966); E.O. Rcischaner (19/3); ,I.A. Scott (1972); and M.D. Waimon , D.n. Bell, and ('..C. Ramsever (1971).

PAGE 40

-29Rased on these sources, a list: of 34 (ognitiv" objectives and 36 affective objectives was drawn up to bo submitted to a panel of eij;hf professors of social studies education and curr i i:u 1 um. This |)anel rated the objectives as to tVie tvpe of content tliat would be requi.red to accomplish the objective (see Appendix A). The cognitive objectives were written to include an (>qua] number from each of six social studies disciplines: history, political science, an thropo 1 fH:;v , sociology, geography, and economics. The affective (>bjectivos were not so easily classified according to discipline. The PICAS could have been analyzed not only .r^ a total score, but according to several stibsciles as well. I'he subscale structure of the FICAS allows for 11 different subscales to be examined for construct validity. TJot all scales are independent of each other, but there exist three sets of independent subscales (see Figure ]). Ihe 48 (ibjectix'es on the scai.e can be divided according to cos;ii i t i vea f f ec t i ve r:riteria (Bloom & Krathwohl, 1956; Krathwohl, Bloom, 6< Mas! a, 1964). into two subscales with 28 and 20 itiuiis respectively. Al te rn.i t i ve 1 y the sr:ale can be divided into subscales of nationalistic, i n te rnat icma 1 , and either/or objectives. Placement on these scales was made by a panel of eight professors who were asked to judge objective's on whether or not American content or international content was required to achieve the stated objective. The third possibility was that the objective could be developed v/ith either type of content.

PAGE 41

30Cognitive COGTOT (28) Af fee tive AFFTOT (20) M.'i I i onn 1 i s ti < NATTOT (13) C o p n 1 t: i vc Nat iona lis tic COGNAT (8) Affect i vG Nat idna 1 is tic AFFNAT (5) 1 n t (.'rna t i imii 1 iNTTOT (13) Cogn i t i ve 1 lUorna t icma I COG INT (8) Affective ] n terna t iona 1 AFFINT (5) VA tlier/or FOR TOT (22) Gopn it i ve Fi the r /or COG FOR (12) Affect ivc Fither/or AFFFOR (in) Note . Numbers in parentheses are number of items nn i^ach subscale. Words in all capitals are abbreviations of subscales used in the text. Figure 1 Subscale Structure of the Florida in t(^ rna ti ona 1 Curriculum Assessment Scales At least 75 percent agreement among professors was required for an objective to be further considered. Of these objertives submitted to the panel of professors, ss (see Edwards, 1957, pp. 149-170). Missing

PAGE 42

-31dat.,-! points wiM-e replaced w i Lb the feiinded p,i-onp mr'ni for each iJc-ni liased on the pilot sanipl(> of 131, 'I'here were 80 nvissinp data points snpplied in tills iTianner out of 6,288 siich points. 'Ihe p,rnnp nu-ms based on the pilot sample were used when necessary for generating Ihe sf;ores required to test Hypotheses One, i'wo , and Three. However, for Hypotheses Four and Five, the group means substituted for the missing flat.i points were based on the 222 sub i ects which comprised the pilnf samph^ as well as the pre and posttest samples from the summer Institutes. Tin's base was chosen as the most conservative procedure in generating total scores to be entered Into an analysis of variance. In the latt( r case 10'5 data points were supplied by substituting the group mean oiit of 10,656 total data points processed. After respondents rated each i t(Mii rni a scale from to 5 along a continuum measuring the relev.ince iif African content ranging from useless to essentia], the total scale score was ( ompnted bv summing the item scores. The subscale scores were generatf^l in the same manner by summing the items on the relevant subsi^ales (see Figure 2 .and Appendix C) . All nationalistic items can be idc'ntified hv reading down the first column of cells in Figure 2. For international items rrvad down the second column of cells. The complete either/or subscale comprises the third column of cells.

PAGE 43

32Alteriiative^ Scoring Methods Considored and Discarded Another option for scorinji also presented i(se1f. As mentioned earlier, teachers could rate African content hi^hlv relevant for an objective which they considered to he relatively un inifiortant . For example affective objective number 27, wliich reachs "Students should understand different ways of life on different continents," liad an attitiide scale rating which was v;ell above the avGraj;e for a flCAS item. However, its importance rating was only average. Cnnse(|uen 1 1 y , although many teachers could see tiie relevance of African content to such an objective, they were not inclined to teach the objective becai.ise they perceived it as unimportant. Tiierefore, in scoring the FICAS two special weighting optifnis were tested to see if greater \'alidi.ty couJd be obtaineil. Since it was th(iugtit that the ImjKjrtance a teacdier attached to an objerl ive wouid lie important information in t:omputing attitude toward African ciMitcnt, tea(!iers were asked to rate the importance of each objective wtii(h made up the FTC'AS. First, ail objectives with a rating of S in importance were scored as marked on the FICAS and alj other items were weight(^d 0. Sec(^nd, a rating of either U or S wns used to seiect the iteiTis to be sci'^red, all other items being weighted 0. in both cases item means were vised as tlie scale scores instead of summated ratings because of the different number of items receiving high ratings from different subjects. When these weighted scoring methods were used, the vaiu(\s of coefficient alpha remained virtually unchanged from those obsi rvetl witli inm-en t i c^na 1 scoring procedures. Wlu.'n items with an imnortance rating of 'i were included an alpha of .9^4 wa ^ observed, .93 for items with importance ratings of h or 5.

PAGE 44

-33A] though these weif;hte'l scni-iTig nipthixls produced rel Labi 1 i ties comparable to the conventional snmmated rating tecfmique, thev did not produce improved validity coefficients as predicted. 'flie first niternative scoring procedure produced an item mean that correlated .14 (p > .05) xvitli the Worldmindedness Scale and -.21 (p .O'l) with the Florida African Knowledge Scale. The second method resulted in an item mean correlating .16 (p '' .05) with the W-Scale and -.18 (p < .05) with the FAKS . it appeared that using the scoring alternatives of calculating item means for only those objectives which received a high importance rating did not measurably increase their validity coefficients. Consequently, further analysis was limited to testing the validity of the subscales which were scored with the convcmtional fikert procedures (sec> Chapter Four) . JLr ocedures for Est ab^ ishi ng_ R el 1 a b i_l_i^t y a n d_ V a 1 1 d i t y Reliab ili ty Coefficient alpha (Cronbacli, 1967) was used as a measure of reliability in this study. The subscales of the Florida International Curriculum Assessment Scales ^^7ere designed to be un idimensi (jnal . Consequently, coefficient alpha which is a measure of internal consistency of the items on a unidlmenslonal scale was an appropriate device to use. For tlie reliability study ttie instruments were adini n i ste rcnl in regular Collej^e of Education class sessions meeting at the University of Florida during spring quarter 19 77. This xvas true for these courses:

PAGE 45

34ED 600, The Schciol Curriculum f twn st-rtitins); F.IH: << 10 , Sorin] Studies Education -!'"lem'Titarv Sclmo] ; EHE 670, l.anr.uaj'.e Art'^; in the Illementary School (off-campus in Marion County, Florida); 1,1)1' ()J() Stu' i oe'-onomi c Foundations of Education; EflF 666, Seminar on Research on Effective Teaching; F"iiF 450, Measurenn'Ut and Evaluation in F.ducation; and EOF 768, Evaluation of Educational Projects and Systems. The students in EDS 635, Supervision of Preservice Teachers (off-campus in Alachua Cf)nnty, Florida), were given instructions for tak inj', the tests and then allowed to complete the scales at home and return them at the next class meetinj'!;. The elementary teachers participating in the study from Hillshorough County were given the instruments during a meeting o[ an inservice class in social studies methods in the spring of 1977. The Hillshorough County middle sclmol teachers were administered the scales during a meeting of a county-wirle social studies curriculum ccjiiuiii ttce . fiie higji school teachers in Hillsborough County were approached thrtMich their (h>par t men t chairpersons vho each requested three of their d'|iar tnent members to respond to tlie instruments on their own time. The instrument battery used to determine the reliability and validity of PICAS contained 12 pages of material. The first page asked for demographic data related to teaching experience. The thrc>e quest iotis asked concerned Im'el, subject area, and years of K-12 teaching experience. I'he second page ccintained the i ns t rue t i c-ns for comfileting the importance ratings and the i'lCAS items. Tin's required a t\-;o-step process: First, for each of the 48 curricular objiM-tives, tlic subjects were asked to rate their importance for tiie school cui-iiculum on a scale from to 5; second, for each objective the subjects were to rate the relevance of African content for achioving the objectives on a to 5 scale. The

PAGE 46

-13next three pages made up the Florid.i Tn ternn t i out I Ciii-r icn ] vim Assessment Scales. Pa;-^es 6 through 8 rniitaiiied tlie Wor 1 dmi udcdness Sr.il.'> (Sampson & Smith, L957; Shaw & Wright, 1967) entitJcd "Social Attitudes Questionnaire." The last four pages consisted of the Florida African Knowledge Scale which is a iO-item revision of f'rojoct Africa's Africa South of the Sa hara test (1968). Most respondents took between 30 and 43 ininutes to complete the total battery. Data used t(i estimate I'oefflciont alpha were 'ollected from 131 persons with K-12 teaching experleTice drawn from graduate courses at the University of Florida Clollege of Education and from the ranks of practicing social studies teachers in the Hi 1_ 1 sbo t-ough Countv, Florida, Public Schools, A total of 169 people responded to the instruments. Thirteen subjects were eliminated from the sample liecause they failed to complete one or more of the scales or hec;uise tliev were foreign nationals withoijt Aniericaii public school exi'erience. An additional 25 respondents in the graduate courses were set aside bet ausc thev did not have K-12 teaching, experience. Consequently, the pilot, sample consisted of 131 subjects all of whom were either practicing teachers or had had previous K-12 teaching experience. Sixty-six subjects were enrolled in graduate courses at the University of Florida, including 19 teachers enrolled in an off-campus course in Marion Ccninty, Florida (pop. 69,030). The cither 65 subjects were f)racticing teachers in fli 1 Isborough C'ounty, Florida (iiop. 490,265), which includes Tampa and surrounding area. The total pilot sample consisterl of 39 people with only eleiiK^ntary experic-nce, 20 with onlv middle school/ junior high experience, and 25 with only high school experience. Eleven teachers had experience at both elementary and midflle scIkhvI , 18 had

PAGE 47

-36both middlo school and hij^h schon] cxperieticr. Kij',ht"on other stihiects had experience at all levels. With regard t(^ year'; of experience, 12 teachers were in their first year of teachj'nj^; 28 had 2 or 3 years experience; 23, 4 or 5 years; 34, 6 to 10; 18, 10 to 20 vears; and 12, more than 20 years experienre. The sub jei-t matter breakdown v-;as weighted in favor of social studies teachers — who were exficcted to comprise the majority of participants in the summer institutes. ihirty-five subjects onlv had experience as social studies teachers and an addition.il 16 had experience teachinp social studies as well as some other subject, [wenty-eight taught the basic subjects (language arts, itiathemat ics , science, social studies) in elementary scliools. S('\'cnteen had basic subject experience as well as experience in some etfier area. Nine subjects were humanities tcMchers in that they were oulv experienced in language arts, art, music, or foreign languages. Twenty-six teacTiers only had experience in other areas: math, scienoe^, media, phvsical education, or vocational education. Validity C orre lates . To establish the construct validity of the Florida International Curriculum Assessment Scales, several hvp(>theses were generated to test predicted relationships involving the construct of interest, attitude toward the use of African content in the curriculum. If a predicted pattern of correlations could be empirically verified, then the construct validity of a test would receive icn f i I'ma t ion . lliis study sought to verify posifivc> cot rela t Iimts between favorabl eness toward African content and two related constructs, wor 1 dmi ndedness and knowledge of Africa. Neither of the criterion attributes was Identical

PAGE 48

-Uwith the ronstruct under studv. llmvcver, it v;,is tlunrized that teachers high en wor 1 dmindedness and knowledf;p ot Afrir.i wmild alsc^ t(Mid td be hiph in attitude toward the use ef African (-(Mitont . The sample, sometimes referred to as tlie pilnf sample, used to test these correlational hvpothe'^es was the same one described in the section on reliability. The methods of data collecticMi were also those described above . The first of the instTumeiits userl to establish the construct validity of the FICAS was the Floritla African Knowled<;p Scale fTAKS), develciped by this project. Examination of the test Afrii'a SiMith of the Sahara (1968), which is a 60-item test oripinallv rlpvelcip<'d fc^^r hiph school students as a part of "Project Africa" (liever (5( Hi(ld six subscales whicdi x-;er<^ retained in an effort to broadlv measure knowledge of Africa. ihcse six subscales were (a) physical peographv, (b) history before i'.uropean penetration, (c) history of Europeans in Africa, (d) indiponous society, (e) economic development, and ff) current affairs. This tiripinal document was used by the University of Tllinois Program in African Studies in a pre-post design to test knowlecige gain at their inservice workshop for teachers in 1974 (Schinidt, 1974, p. 13). Beyer used a IS-member panel to sift through 120 miiltiple choice items and c.ime up with 70 items to be item analyzed. nf the 70, 60 were chosen for the final i ns t rumr'nt . Using an odd-even split-half technique, Beyt-r found a .60 rcliabilitv coefficient for his seventh grade sampK' and a .80 co(ifri( ient foj' the twelfth grade sample (Beyer & Hicks. 1968. p. 22). Although Beyer and Hicks (1968) did not specifically report the number of sulijects involved in the

PAGE 49

-38rpl iah 11 i l:y study, tlicy rc[M)rtc'(i elsewhcn' in t:fir>ir pt f'scn tat idn that 8A5 seventh j:raders and 7^54 twelfth graders were surveyed in tdielr study (p . 5) . Three facts suggested that the Beyer and Hicks test needed to he modified for use wl tli teachers in this study. The first vv'as that the .60 reliability coefficient with the seventh gradc> sample was too ] ciw to he acceptahle. Second, since the test was develepi'd in 1968, events both in Africa and in the field of African schri 1 a rsli ip iiave rendered several items outdated. Third, sinctness of Western-style economic development as a mndel for Africa); questions calling for value

PAGE 50

V-)iudgments; questions con ta i n Lnp; n(^)',ativo Iiiases towards Afi'ifa; questions which were confusing, simplistic, or containocl i mp I aus i b 1 e distractors (incorrect alternatives); or questions whicti were dated and no longer valid. Under these criteria the suhscales on economic development and current affairs were especially hard li i t as 60 percent of the former and 71 percent of the latter were eliminated from the c^ripinai to^st. Eleven other questions from the original 60 were slightlv modified to bring them up to date or to chanr.e a singl(> implausible distractor. Tiie outline map of Africa used to .answer the first six items in th(> original test was modified by adding the four major rivers of Afrii.-a -Nile, Niger, Congo, and Zambezi --to the cnntincMital outline. Eleven new items were written for the levised knowledg,e test, resulting in a 43-it('m instrument to be pilot tested. The pilot test was administered to 1 -lO undergt iduates at the University of Florida. Eighty-four students enrolled in tluCollegt^ of Education toiik tlie test during fall quarter 1^5/6. An additi(^nal 36 students took the test winter quarter 1977 rluring, the first cl.iss meeting of two different introductory African studies courses. These 140 students were divided alternatively into two groups of 70 to allow an item analysis and a double cross validation to be performed. The biserial correlations between item scores and total test score were used as the indices of item discrimination (Lord fv Novii/k, 1968, Chapter 15). The 35 items with the highest biserial correlations based on subsample one were cross validated on subsample two. At the same t i me the 15 l^est items reveali-d by an item analysis on subsample two v^'cre cross validated on svibsample one. The 30 items which had the high(>st and most stable biserial correlations across the two sarii]-iles were selected for the final

PAGE 51

-''.0knowledpp test. The only exception to this strirt r'.'-\ ihomet r Ic criterion was that one item pertaining, to Zaire was dropped luu-auso of tiie lariN.> number of items already referring to Zaire and an i i em which had a lower, but acceptable, cross validation biseria] correlation with the total score {r_ .31) replaced it to keep the orij;inal snbscale percentages that had been established by Beyer and Hicks (10C)8, p. ?']). Based on a new sample of 131 subjects described above in tiusect )(mi -an reliability, tlie biserial correlations for the final 30 items ranged from .18 to .76 (see Appendix B) . The surviving 30 items were given a final (diet 1: for content validity and proper wording. Three items were altered at this iioint. fhe resulting instrument consists of 11 items taken unaltered (except for changes in the map used to answer some physical goi'i'.ra]iliy questions) from the original 60-item test. Ten other items represent slight modifications of original questions and nine items an^ comi'letely nev.> ones written for the revised test. Consequently, the revised test consists of approximately one-third completely new items, one-third unchanged items, and one-third modified items (see Appendix 0). To overcome any effects of a response set, the correct answers wer(> apportioned as evenly as possible among the four alternatives. Ihe total scile score consisted of the sum of correct answers. The second instrument used in the validation process was an already constructed l.ikert-type scale known as the Wor 1 dmi ndedness Scale (W-Scale) (see Appendix E) . This scale was developed as part of a studv to learn what types of i ntercul tura 1 experiences have a special impact on attitudes, the relationship between personality differences and worldmlndedness, and what relationship exists between attitude prior to an

PAGE 52

-/,lintercul turn ] experience aiul rearlion to t\\nt (>xpprienre (Smith, [9S5). The Worldmindedness Scale has eight stibscalos that mctmne attituries nn the fo 1 1 owiiij:; dimensions: rnllj^lon. inimir, ra t i on , syi ve ninien t , economics, patriotism, race, education, and war (Shaw ^ Wripjit. 1967, p. 203). In validating the scale Smith {\93'), p. A70) reported thi'se correlations between the Worldmindedness Scale and the followins', constructs: F.thnocentrism -.71, Facism -.46, and Political and F,con"iiiir Conservatism -.SK Smith found that una t rue t ur(vl , heterogeneous i nterc ul t ural ex]ieri euces had little impact on altering WMrldmi ndedness . The tentleri'-y was for people on both ends of the Worldmindedness Scale to have their attitudes strengthened by Intercul tura 1 experiences. After (Diiipnring Worldmindedness scores to results on factors making up (Guilford and Cui Herd's (1934, 1936) int rovers ion-(\'';t rovers ion scale, Smith (T'S'i) concluded that the "highly worldminded pers(-in is less masculine and ascendant, more impulsive and emotionally dependent, and more iri'-l ined toward introspecti(in and i nte rnal i za t i (mi of impulses th/ui the \erv nationalistic individual" (p. 476). Sampsc^n and Smith (r'S7) defin'worldmindedness as "a frame of reference, oi' value orientation, ra\-oring a world-view of the problems of humanity, with mankind, rather than the nalirMials of a particular country, as tlie primary reference group" ([>. 105). They emphasize that worldmindedness as they conceive (^ f it designates a value orientation, or frame of reference, apart from knowledi'c abcmt. or interest in, international relations. Sam[ison and Sniith ri^poit a corrected split-half reliability of .93 and a test-ietest )-eliabilitv also equal to .93. Allman (1961) used the Worldmindedness scale to compare differences in the student bodies of two scliools. Garrison (1961) used the scale to compare differences based on religion, region

PAGE 53

-/4 2of the country lived in, fnnii ly bnckp,r(iuii(i , srx, -nid \-i-ar in (-ol.lef^e. Based on his findings at thf^ University of Ceorgia, a male frpshman Baptist from an agricuLtural background in the Southeast would he the most likely candidate to be low on wor 1 dmi ndedness . Newman and Ware (1976) found a lov; positive relationship (r = .33) between wor 1 dmindedness and aesthetic perception. The Worldmindedness Scale was chosen for use in this study fcir several reasons. Not only does it Iiave solid credrntials for reliability and validity but it also has a timely qualitv not shared with other scales aimed at measuring internationalism. Manv scales developed in the 19A0's and 19S0's contain items which are no longer valid because they refer to personalities or spocific sittiaticuis which would not he familiar to subjects in the 1970's. Ihe Wor Idmi ndc-dness Scale, however, contains items which refer to more generic situations which an audience in .1977 could relate to as easily as one in 1957 when the scale v/as developed. In addition, the Sampson and SmLtii (195/) srale atteiiipts to measure a broad range of dimensiorrs related to worldmindedness. Many other scales are limited to a narrower range of attitude objef^ts -for example, war, communism, patriotism, or trade policy. in scoring the Worldmindedness Scale, tlie prot endures outlined in Shaw and Wright (1967, pp. 203-204) were followed. However, since some item responses were missing, a preliminary step was taken. For each item a group mean was gener.ited based on the piiot sampie of ] 31 teachers. In order to compute total and subscale scores, ttiesn item means were rounded to the nearest intev,er and substituted for the blanks on the score cards. The Statistical I'ackage for the Social Sciences (Nie, Hull, .Jenkins, Stei iib renner , and Bent, 1975, pp. 181-193) subprogram

PAGE 54

-/4 3"Condescript ive" was used to fompnlo the item riiMiis. The missing data problem was a very minor oik^ in Ihis casp n=; only 28 data poiiits were missing out of a total of 4,192. Total scores wcri' pruernted for each subject by adding the value of (Tie item rcspcMises whi'h ranged from 6 for strong agreement to for strong disagreement. lH^(;riminari_t^ Yfl-^A'^'A'^ ^' "'^ • '^"^'^ second step in establishing the construct validity was to state a hypothesis almut the relationship between scores on the PICAS subscales of two gr(Tups vdiose scores should be different. It was assumed that teachers who elected to attend summer institutes on African studies would havt^ more positive attitutles toward the use oi African content than woulrl a cross set ion of teachers. Therefore, it was prediiUed that the FTCAS subscale scnres of the institute participants woulrl be significantly higher than those of a cross section (or pilot) group. After preliminary testing to determine the correlations between the PICAS subscales, the Wor 1 dmindedness Scale, auT the PARS, the instruments were administered to 49 teachers attending summer institutes on African studies at either the Universitv of Illinois, Champaign, or the University of South Florida, Tampa. Twenty-eight piM'file took the threetest battery on June 20, 19 77, in Tampa. Twenty-one took tlie FI(;AS and PARS in Champaign on Tune 9. These pretests, given at the beginning of the respective summer institutes, provided data on the Africa attitude scale to be used in the group differences construct validation procedure. Both institutes were similar in their goals. P.ach institute sought to increase teachers' substantive knowledf'.e of Africa and Africans, to introduce teachers to new sources of information about Africa, and to assist teachers in developing curriculum plans and teacdiing skills

PAGE 55

appropriate to inteprat i nj!, Ihc snuly of Africa iiito tfi-^ public scliool curriculum. The Chnmpaisui institute sponsored bv tlic University of Illinois African Studies ProRram under a K.rant from tb'' NaticMial Endowment for the Humanities lasted four weeks from .June 6 to .luly 1. 1977. The lampa institute was sponsored j(Tintly bv the bepartraents of rnternational Studies and Afro-American Studies and the ('enter for Economic Education at the University of South tHorida and tiie Center for African Studies at the University of inorid/i. It lasted for two weeks from June 20 to July ], 1977. Eacli workshop providc'd part ic i pants with a varietv of resources on Africa: Amon^ tliese were films, lectvirc/ discussions, meetings witii Africans, slide-lectures, and readings. in addition, the participants in each wcirkshop were I" produce curricular materials which were to he implemented in tlie scli'iol V'^ar 1977-78. At the conclusion of each uorks,hop the FICA:^ and FAKS instruments wei'c readministered to allow the measuring of change on tlu> two scales as a result of workshop participation. In testing the hvpothesis on group differences th--^ 1 K'.AS subsc:ale scores of the pilot sample were compared to tliose <'f the pa rt i cijiants in the two summer institutes on African studies. Attending tlie institute in Tampa were 28 teachers (26 from Florida, nne from Louisiana, and one from Georgia). Twenty-two were wiiite and six were black. Nearly, half the teachers, 12 of them, has a combined middle schodl-high school background. Five mcire were exclusively li i gh scbiiol teachers and four were exclusively middle S(Ti(hi 1 / j un ior higii te.ichers. duly one teacher had a combined middle sclioo I -elementary iTackground. Six other teachers had taught at all levels. More than twic(> as many teachers (10) had taught from 6 to 10 years as had taught any other length of time. One

PAGE 56

was in his first year o[ toMchinf:; S ti.id t.!iip,ht ? or '• years; S, A or 5 years; 3, 10 to 20 years; and 2, 20 years or ni')r'>. As had been predieti^d the majority of institute part ic: ipants were S(H-ial "Studies tr'nehers; 17 exclusively so and seven moi'e taught social studies and some other subject. Tvv'o teachers were art teachers in the hnnianities category an(i onl)' one had experience outside of social studies or humanities, Tlie sampie attending the C'liaiiipaign workshop ccMisisted of 23 Missouri teachers. Of these, data exist on only 21. As opposed to tlu^ Tampa group, the Missouri group was predominante 1 v black. 13 to eight. This group also contained a larger percentage of elementary school teachers: Four were exclusively so and one had a middle schon 1 -e 1 omenta ry hackgrciunc Two were exi^Jusively juTiior high/middle school tca(diers and six were exclusively so for high schnol. IM ve had a combined mirjdle school-high school background. The throe others had t/night at all levels. ITie Missouri teachers als(i tended tci be more exper i euci^d tha'i their southern counterparts. Nine of the 21 had tauglit fr(Mn 10 (c 20 years; '» , 2 or 3 years; 3, A or 3 years; ''i , ''1 to 10 years; and 1, nmre tlian 20 years. T'he Missouri group were for the most part also social studies teachers: 12 fit this description, five of whom had had expo?' fence in another area as well. Two had taught only basir elementary subjects. Three had taught elementary basic subjects plus something else. Three were humanities teachers and the last participant had e:>;pe rience in an area outside of social studies or humanities. Change over occ:asions. Yet a third iiK'thod to eslalilish construct validity is to exper inienta 1 1 v manipulate conditions in such a way that a change in the construct under study would ocour, then measure that change. In this study the "e>xper 1 men ta 1 manipulations" wore summer

PAGE 57

-46institutes
PAGE 58

CHAPTER FOUR RESULTS The Florida Interna t ioiial Curriculum Asscssmeul Scales fFTCAS) wore validated using twf) other instruments and two separate saniplos of subjects tested on three different occasions. Statistical estimates of the reliabilities of these instruments and their subscales are presented in this chapter followed by a discussion of the sample distributions. Finally, results of tests of the experimental hypolhest^s desi)',ned to establish the construct validity of the FH'.AS are reported. Re 1 ijib i li^t i_e^s In this study Cronbach's (1967) aliilia, a moasnre nf internal consistency, was used to estimate reliability. According, to CrcMibatli, "alpha is . . . an estimate of the correlation expected between two tests dra^.^m at random from a pool of items like the items in this test" (p. 141). Alpha is the average of all the possible split-half coefficients for a given test (p. 1 i5) . The formula used to estimate alpha in this study is a generalization of the Kuder-Richnrdson F'lrmula 20 (Veldman, 1967, p. 173). The internal consistency coefficients for the FAKS and FICAS are displayed in Table 2. For FAKS, only the total scale score had an alpha of greater than .70 based on the pilot sami^le (^f I 'I subjf^cts. Coefficient nlpha was .83 for this group. This specific reliability (Lord f* -47-

PAGE 59

T;i!.1g 2 Reliabil i l ies and Stand-irci I'rrors for ttio Florid,! International Curriculum A.sscHsmcnt Scales and the Florida African Knowledge Scale based on tihe Pilot Sample f N 131) No. of Coef f i
PAGE 60

-49Novick, 19ft8, Chapter 9) remained lii,n;h when the test was admi nis tered lo the <49 snbjects attending the summer i nst i tutt-s . Hie i r pretests yielded an alpha ef .77 and t Ium' r pusttests, .78. Ihe FICAS his high reliability as estimated by the alpha ettef f icients . The eoeffirii-nt alpha values of the 11 subsca]es ranged from .78 to .9"^ (see Table 2). The internal consistency estimates of the NAl'TOT, INTl'OT, and EORTdT subsrales were .92, .87, and .93 respectively. Slightly higher coefficients were observed as a result of administering these scales to the institute samples, ranging from .91 to .94 on the pretest and from .92 to .94 on the pos ttest . An alpha of .85 was observed for the total Woi I d''ii ndedness Scale based on the pilot sample. For the 28 participants in the lampa Institute the coefficient of .83 was computed. As with the FAKS, the U'-Scale subscale reliabilities were too low for further analvs'\^s. Since the reliabilities of the subscales were all below .7S (see Davis, \9(t^, p. 24; Tinkelman, ]97], p. 71; Fox, 1969, p. 362), onlv the total W-Scale and the total FAKS scores were used in attempts to validate the FiCAS. The results of this reliability study revealed that each subscale on the FTCAS was reliable enougli to be subjected to validitv cher:ks . However, because of lack of independence between some of tlie 11 subscales, some scales being composites of othei^s, and high mu 1 t i col 1 ine.arity among some scales which were theoretically independent, enlv three subscales, NATTOT, INTTOT, and EORIOT. were selected for the investigation of construct validity. Mul tlcollinearlty refers to high correlations among independent variables entered Into multiple regri'sslon (see Blalock, 1964; Gordon, 1968; Kerlinger & Pedhazur, 1973).

PAGE 61

-50S_a mpj-_e D i s trjAi uti o n s Because the hypotheses set forth in this studv c-ille'l for the use of significance tests used wltli ttie calculation of both Pearson r's and analyses of variance, the distributions to be enteifd into each analysis were checked for normality and homop,enei ty . Althouj'ji much literature claims that the Pearson r (Carrol, 196.1; havlicek fv Peterson, 1977) and analysis of variance F tests (Kirk, 1968; hinqulst, 1953; Pearson, 193]) are robust with respect to vif^lations of the assumpLions of ncirinality of distributions and homogeneity of variance, except when sample size also varies greatly in conjunction with discrepancies in variance (F^orieaii, 1971), some dis trlbut ioris in ttils study appeared t (^ roijrescnt extreme violations of these assumptions. Iherefore, nil s( -ires on tlie FTCAS as well as the W-Scale and FAK"^ total scores were nonnalized befc.ire proceeding with the parametric statistical tests (~i f h\-not heses . Tc-st;s of Construct Validi^ty In the first phase of construct validation the three selected FICAS subscales were correlated with measures of attitude and knowledge which should have had some relationship to attitude toward including, African content in the curriculum. The tests of Hyptitheses Two and Three constitute tills phase of validation. In the second phase of construct validation the differential validity was investigated to determine if the scales could be used to disting\iish between groups who shoultl have differed in attitude toward Africa in the eurricMlum (see llypcithesis Four). Finally, the FlCAS su!)scales were tested to scm:^ if they could be used to detuct a clmnge In attitude aw a result of attending summer

PAGE 62

institutGS on African sLndiPS anmed at increasinp, tcnrher rompetenry in using African content in classroom toachin;^ (soe llvpoiliesis Five). Hy pothesis O ne There will be no significant differences (p .2''-') anionp the mean scores of subjects in the pilot sample assij-ned to take three different forms of the Florida international Curriculum Assessment Scales. Because the possibility existed that the ratine (if the rr-levance of African content mij^ht be affected by the context witliin which tlie rating occurred, three forms of the F I CAS were tested. On Form A respondants were asked to rate the relevance of studviri): Furriyiean pf^oples and places in addition to rating the relevance of st>idvin<'. African peoples and places. Form B included Latin America ahmp v.'ith Africa. Op Form C respondants were askerl to rate the relevance^ of Alriran content onlv. Thirty-three subjects were assigned at random to nv nt the three test conditions. These subjects were the first 33 to respond to the instrument from the pilot sample described in Chapter Three. Because the researcher expected the null hypothesis — H : ^^ = X^ = X -to be confirnied, a liberal alpha level of .2S was chosen increasing the chances of a Type 1 error, and thus reducing the chances of a Type II error. The observed difference in means was tested against the expectation that it would occur bv chant^e at least 25 times in TOO before the researcher would accept the null hv|iothisi s . A oneway analysis of varianf;e supported the null hypothesis. The observi'd F rat:io, .71, Indicated that the difference in means would occur 50 times in 100 by chance alone (£ " .50).

PAGE 63

-52Tnhle 3 Raw Means and Standard lu-v iat" [(hv^, for Hypothesis One (N \'V) Treatment Mean Standard Deviation Form A 17.5. "^5 33.16 Form B V4A.OO 33.58 Form C 128.72 38.50 Hypo thes is JIwo There will, be a sip,nificant positive correlation (p -' .05) hetween scores on the Florida International Currienltim Ass'-ssnient Scales and scores on the Worl dmindedness Scale. Hypothesis Two was tested using the rtv-^poiisf^s of tiie pilot group of 131 experienced K-12 teachers. Tliis group consisted nf a cross-section of teachers representing all grade levels and suhjoft areas. Because mul t i coll inearity can cause problems of interpretation in multiple regression analysis, the six independent suhscales makin;; up tiie total FICAS were reduced to three cc>mposites. Because there were three su()scales of interest, multiple regression was used to provide an overall significance test of the rela tionsliip between the tliree FICAS suhscales and the criterion of worldmindedness . NATTOT, INl'1'0 1 , and F.OKTOT had a multiple 2 For a discussion of the problems of mul t i co 11 i nca ri t \, see P.lalock, 1964; Darlington, 1968; (audon, 1968; lohns-Mi, 1972; Nic et al.. 1975. The researcher opted to create composites rather than eliminate some of the higlily correlated suhscales. COCfiOK and AlFFnr-! ( r^^ = .84) were combined to form EORTOT; COONAT and AFFNAT (r^^ = .8 1) were combined to form NATTOT; and CO(;iNT and AFFINT (r^^, .72) wer(> combined to form INTTOT.

PAGE 64

-53correlation, R, with the total Wor 1 'imiiideclncss Scale of .28 (vn th F = 3.A7, p < .OS). Since the overall F ratip w,ip, significant, the Individual Pearson-product moment correlations hotveeii the FICAS snhscales and the W-Scale total were examined to determine which ones differed significantly from zero in the expected direction. As hypothesized, both IMTTOT and EORKIT had significant i)ositive correlations with worldmindedness. Tlieir correJations were .26 and .21 (p < .05) respectively. However, the correlation between NAiiOf and the W-Scale total was not significantly greater than zero. CcMisequen t 1 y , for Hypothesis Two there is some evidence of construct validity for INTTOT and FORTOT but not for NATTOT. Hypothesis Thr ee There will be a significant positive correlation fp .05) between scores on the Florida International Curriculum Assessment Scales and scores on the Florida African Knowledge Scal(>. Hypothesis Three was tested using the same pilot sample of 131 teachers that was used to test Hypothesis Two. An R x^7as calculated between NATTOT, INTTOT, and EORTOT on the one hand and the FAKS total scale score on the other. Tiie observed value was R = .23 (F =2 37 3,127 ' not significant). No significant relationship was found between the FICAS subscales and the test of African knowledge. \ lypotl ) e^s i s Fo_u r Subjects who elect to participate in a curriculum workshop on Africa will have a significantly higher mean score (p < .05) (ui the Florida

PAGE 65

•54T;)hle 4 Raw Means and Standard Deviations of all Siiiip1(>s and Variables Entered into the Analyses of Hypotheses Two throiipji live Va r i ab 1 e s Group NATTOT INi'TOT FORTOT WORLDTOT KNOWTOT 92.91 13.9A 22.54 5.8 3

PAGE 66

-55Internat iona] Curriculum Api^essmPiU Scales than a conipar LsiMi j;ronp of a cross section of teacliers. The next step in es tab 1 isliiiig the construct validity of the FICAS was to demonstrate that a sigiilficant difference existed between the mean scores of two or more proups who theoreticallv sliouLd have differed on the scale. The two groups selected in this study were (a) the pilot group and (b) the combined institute groups. The pilot group was a cross section of teachers who happened to be availafile for testing. They were assembled to take coursework which in no way was mlated to international education or cross-cultural studies. The Hillsborough County subsample was composed of social studies teachers who were selected because their availability was unrelated to international education issues. On the other hand, the institute samples were composed of teachers who wei-c> self-selected to participate in summer institutes t^n African studies. They could be presumed to have a more positiv<^ att itude toward teaching about Africa than the cross section of teachers in the pilot group. Therefore, if a significant difference in mean scores lietv;een these two groups could be detected by the FICAS subscales, :hen tliere would be some evidence for the validity of these attitude measures. Because the two institute samples came from different areas of the country, they were treated as separate groups in the analysis. A discriminant analysis was performed with the three relevant FICAS subscales being used to discriminate between the pilot group and the two institute groups. A post hoc orthogonal planned comparison (Kirk, lOfiR, pp. 60-78) was used to test whether the means of the institute groups differed from each other and whether their combined means differed from the pilot group mean. With three groups, only two (k 1) orthogonal platuied comparisons were

PAGE 67

-55posslble (Kirk, 1968, p. T?). Hnwevor, with t;tiet;p twd comparisons the researcher was able to nlit,rin nil the information iie(=(le(i to test Hypotiiesis Four. The comp.irisons nsed can he symbo 1 i ,-,erl in this way (Kirk, 1968, pp. 69-70): "'l = ^2 S "'2 = \ \ + s The first of tlie two comiiar i sons tested whetlu^r any siRniflcant differences existed between the institute groups on tlie relevant I'lCAS subscales. The second comparison tested whether the pilot p.roup mean was si gnif icant 1 \ different from the average institute grou[i mean. When NATTOT, INTTOT, and KORTdT were entered into a direct discriminant analysis one significant discriminant function was produced. The standardized discriminant function coefficients for this function were .76 for LNTTOT, .28 for NATTOT, and .09 for fORixn . After determining that these three subscales do discriminate between the three groups in the analysis, post hoc analysis of variance employing planned comparisons was used with each variable separately to test first if there was a significant difference between the means c)f the institute groups and second if there was a significant difference between the mean of the pilot group and the combined institute grotips. Ihe overall F ratio for each variable was significant at below the .0'? criterion alpha level. In addition, on no variable were Ihe means of the institute groups found to be significantly different. However, on each variable the mean of the pilot group was found to be s igni f ic£Ui t ly lower than the average of the two institute grotips means. Hvpothcsis Four confirmed

PAGE 68

-57that INTTOT, NATTOT, ,md ECRFOT fvich mcasurod tl).' .'-pr'tfri d i f fe rences between the pilot group and the inst.itnte j^ioups. Yet the two institnto groups, each of which consisted of educators self-selected to participate in an African studies program, did not differ from each other. Hy pothes is Fi^e The mean posttest score of a group of teachers enrolled in an African studies center summer institute will be s iij;n i f 1 can 1 1 v (p •' .OS) higher than their mean score on a [iretest of the Florida In 1 t-rna t i ona 1 CMir r i ciil um Assessment Scales. Hypothesis Five w,is designed to see if the new attitude scales could detect attitude change as a result of summer institute pa rt it i pat ion . A multivariate analysis of variance repeated measuies desip.n controlling for site was developed to test i i" a change in att i t ufle c'ould be measured by the instrument. This analysis was follov; Clson, 1976). C(msequently, it must be concluded that the three subscales did not distinguish between pre and posttest scores of institutiparticipants. Therefore, Hvi^othesis Five was not supported.

PAGE 69

CHAPTER FrVE CONCLUSIONS AND RECOMMKNDATTONS Int roduct ion Two subscales of the Florida International C'nt r ir nl iim Assessment Scales have been shown to possess suffji-lent reliahility and validity to be employed in further research and evaluation studies. Both [NTTOT and EORTOT demonstrated tlie predicted relationship to K'or 1 dmindedness and distinguished successfully between the pilot and institute groups. NATTOT has served a useful purpose in this validation study hut appears to be of no further use in measuring teacher's attitudes toward usinp, international content, African or otherwise. In this ihapter the psychometric quality of the PICAS is discussed, SuRKostions for further research and applications are also presented. Rel iabi^lity _Conipar^isons Before focusing on the PICAS, the reliabilitv dat.i observed for the Florida African Knowledge Scale and the Wor 1 dmindedness Scale will be compared to that reported elsev'/here in the literature. For the pilot sample of 131 subjects an alpha of .83 was observed for the FAKS . The 49 subjects attending the summer institutes yielded an alpha nf .77 for the pretest and .78 for the posttest. These figures cfunpare favorably with the reliabilities reported by Beyer and Hicks (19^i.S, p. 22). They used an odd-even split-half teclinique to compute a coe f I icicnt of .60 for -58-

PAGE 70

-59their seventh parade s;imple and .80 for ttieir twelftfi crado Paniple. However, reliability is a f-inrtion of test IcTigtli and i he F.oyer and Hifks instrument was twice as long as the one refinod in the ctirrent study. i.ater Beyer and tlicks (1970) used their own .'iH-item revision of the orii^rinal instrument in a study which prochiced Knder-R ichardson reliabilities of .73 for the pretest and .88 f(>r the posttest. As a bv-prodnct of validating the FICAS, the ^0-item FAKS has been produced. Its reiiab i 1 i t i es are comparable to, if not superior to, lon;',er versions c^f African knowledge tests produc(.'d by <'ar-Mer studies. The content \'alidity of the FAKS was assured by submit I in?; the items potentially to be included in the instrument to a panel of 1^ Afi'icanists associated with the Center for African Studies at the finiversity of Florida. I terns which were datt^d, j)oorly worded, biased, or otherwise vulnerable were deleted from the item analysis and cross validation wfiich produced the final 30-iteni scale. For the Worldmiiidedness Scale, the reliabililN coo f f i r ients observed in this study were slightly lower than those reported by Sampson and Smith (1957). They reported both split-half and test-retest reliabilities to be .93. For the pilot sample in this study an alpha of .85 was observed. For the 28 participants in the Tampa Institute the coefficient of .83 was computed. The eiglit four-item subscales on the W-Scale had relatively low reliabilities among the pilot sample ranging from a .33 for the "Fducation" suliscale to .62 for the "Patriotism" subscale.

PAGE 71

-60DiscuRsions of FICAS Valic]it^ TNTTOT Among the 11 subscales of FICAS witli potential for use in measuring teacher attitudes toward including African content in the cnrriculum, only INTTOT and EORTOT emerge as reliable, valic], arvi independent measures. TNTTOT consists of 11 curricular objectives requiring i nterna t i
PAGE 72

-61nttlt.ude toward African conlt-nt: in tlie nir ri ciil uiii. Anotiii.T pcss ili i 1 i t y also exists. There may ho .1 nonlinoar ro 1 a t ion'^ii i [> hrtwoen knowlecip.o of Africa and attitude t<->ward Atrican rontent. Tliis possibifity will be taken up later when further research is discussed. Of the tested subscales, INTTO'f had th(> iii,f',hest st.anda rd i 7,ed discriminant function coefficient on t lie function discriminating between a cross section of teachers and a group which was st'l f-se J ec ted to attend African studies summer institutes. Theoretically teachers who would choose to attend a summer institute in African studies siinuld be more inclined to use African content to achieve international objectives than a more general population of teachers. INl'TOT clearly meas\ires this expected difference. That INTTO'l failed to measure a pre-post difference in attitude among summer institute participants might be explaincl in several ways. It is quite possible that no change occurred in I be attitude iTieasured by this scale. Strength is added to this (explanation bv the previf)us finding that institute teachers are relatively higher in their INTTOl' scores compared to the pilcit sample than they are on the other two scales being tested. Their initial high scores may h/ive left little room for additional ch.mge. In addition, the phentTmenon (^f regross ion toward the mean could have played a major role in INIl'OT's not measuring a change in attitude. It is also possible that INTlOi failed to detect a change in attitude that did occur among the institute teachers. However, INTTOT's jiower in detecting a difference between institute tC'achers and a cross section of teachers makes the explanation of no real change more plausible than that of an undetected real change.

PAGE 73

-r,2EORTOl' The EORTOT suhscnle consists cif 22 obi'^rtivns vh i cli cnn bo dovelopoH with eitlier nationalistic or Internationa! contont, or both (i.e., "Students should understand the balance of nattire," "Students should know about the role of government," or "Students sb(uild accept chanf^e as a natural feature of the human t:ondi t ion") . Such objectivcvs are not specifically international in scope, as are TNl'Tdl's, but one would suspect that a worldminded teacher would be more inclined t(^ use international content to develop either/or objectives thin would a teacher with a more ethnocentric vi'/wpoint . Conse(]uen 1 1 y , one would expect to find EOR'iOT scores to be pi^sitively related to th-i';e on tlie IJ-Scale but not quite as strongly as INircvr's scores. Such was in far-t the case as the correlation between I-nRTOT and the W-Scale equalled .21 fp <.OS). Although no significant relationship was found lu'tween EORi'OT' nnd the FAKS total scores, further r-esearch may show this (inding to result from a nonlinear relationship existing between the t\^o constructs. EORTOT also distinguished between the pilot cross section sample and the institute sample. However, because of its higli correlation with INTTOT (r = .72), it adds little information when INTT'dT is also entered into a discriminant analysis. Consequently, its d i srr iminaiit function coefficient is rather low when compared to INllOF. N(ine t he! ess , when used separately in its own right, EORTOl is an efft^ctive tl i sc r imi na tor between pilot and institute groups. The EClRfOT scale was e>;pected to measure the attitu'le variable most amenable to change as a result of attending summer institutes. As teachers learned more about Africa and developed curricular applications for African content, it was expec:ted that they would see more and more

PAGE 74

-6> uses for African content niimii^; the cither/or objec Lives. I'oc.'Uise of a nonsignificant, nui] tivar ia t (> F ratio, the re,c;ear<-hi r did not examine the univariate F ratios for NA'l TOT , INTTOT, and EOin'OT f(^r the purposes of drawing statistical; conclusions. However, it shouJd b(^ noted that in spite of the multivariate F's being nonsignificant, ECRrOl's univariate F was significant (p •' .05). Such a result, wh i tdi can occur when several weak dependent variables are included in a imil t i v.i I'late analysis (see Bock, 1975, p. 155), suggests that the issue of KORl'OT's ability to detect {ire-post differences as a result of summer institute attendance is still unresolved. Thovigli this study has not been al)le to establish the ability of EORTOT to detect a pre-post difference, neither has it laid to rest the possibility of EORTOT's abililv to do so. MTTOT NATTOT has served a useful purpose in the validatii>n process by being distinct from INTTOT and EORTOT. However, it dof\s not appear to be a valid measure of teacher inclination to use international content. This result was expected. NATTOT consisted of 13 curricular objectives which required American content to achieve (i.e., "Students should be proud of our national accomplishments" and "Students should have a knowledge of social change in America"). Although some individuals who like to teach from a comparative perspective miglit find international content relevant to achieving nationalistic objectives, it was not expected that a cross section of American school t (•aclu/rs , or even the institute participants, would perceive African ccnitcuit to be relevant for teaching to nationa] is t i i? objectives.

PAGE 75

-64NATTOT scores proverl to havn im si j^n i f i cnn t re 1 n t ionshi p to W-Scale scores. A correlation of .04 was observed. Ni.-ithcr did NAT'ini' scores stiow the predicted relationship to FAKS sc(ires. A -.21 (p .05) was observed where a positive relationship was hypothesized. It appears that as teachers knew more about Africa tliey were less inclined to use this knowledge to achieve nationalistic objectives. These findings further reinforce the conclusion that NAl'ITVI' is indeed measuritiK a tr.iit distinct from that measured by INTlOl' and EORTOT. l-Zliile it is true that NATTOT was useful in distinguishing between pilot and institute groups, its complete resistance t" change as a result of institute treatment has .idded weight to tlie argument that: it is distinct from INL'TOT and F.OIMd'C and useless as a iiu^asure of ten(!u-r attitude toward the use of African centent. Among the total nunhcr of institute participants there was virtually no change in NATJivr moan scores pre-post. In fact, among the Uliiieis institute subsample tliere was a nonsignificant decline in group mean. iliis pattern of findings leads to the conclusion that NATTOT is not a useful suhscale to measure teacher attitudes toward including African content in the curriculum. However, NATTOT has played a useful role in the validation process precisely because it has sltown up as distinct from INTITiT and KORTOT. One precaution that mcvst Likert-type attitude scales take to correct for response set is to v;ord one lial f of the items negatively and one half positively. The negative items are then reflected in scoring. However, since F1'"AS items consist of curricular tib j ec t i vc^s it was not possible to word one half of them negatively. Consequently, it I'jas reass\triiig to find that the nationalistic items scattered thrcmghout the FTCAS were respofided to differently by the subjects. Both subscales, NATTOT and

PAGE 76

-65INTTOT, had a potential ranpe of ;'.(ro to (i^ . AmiMi;;', the pilot i',rouii the raw mean for NATTOT was 28.1 while the raw mean for INITOT was ^^ 3 . 8 . This difference \-7as significant at the .001 ;i 1 plia l.-vel. This findins: lends support to the argument that PICAS and its suhscales were nieasnriiig much more tlian response set. Multiple Aspec^t^ of UiiidjuTicMisjona 1 Scales The high multicol linea r i ty found bctv.'een several sets of FTCAS subscales has confirmed tlie underlying structure of attitude toward African content in the curriculum. In analyzing tlu^ open-ended questionnaires during the early stages of the test development, ttun^searchr r determined that the two major underlving dimensions cif the attiturie under consideration were the relationship of Afrif:an content to ((ignitive objectives i^n tlie one hand and to affective objectives on the other (see I'ab 1 e 1). Consequently, botli cognitive and affective objectives wore includet' on the PICAS under each of the three categoric>s: nationalistic, international, and either/or. The intercorre 1 a t i ons amonr. tin;various cognitive and affective subscales indicates that they belong togi-ther on their respective scales. Table 5 displays these cognitive-affective intercorrelations which show that on the two sul'scales >if best validity, INTTOT and EORTOT, the cognitive-affective components had .72 and .84 correlati ons respect ively .

PAGE 77

-66lahle 5 In ttM-cor rel a t i (ins cif Cnrji i t L viim: Affect i.ve Subscales on FICAS Cognitive Affective Subsi-ales Subscales AFFNAT AFFINT AITF.OR AFFTO'l COGNAT 0.83 COGINT 0.7 2 GOGEOR n.8''. COGTOT 0.88 Adequacy o f I NTTOT^ and ECJ^RJOT as j\t ti t ude^ Seal es INTTOT and EORTOT liave be€>n shown to posstvss li i f fi reliability and various degrees of valicUty. But bow do they ini^asur(> up when cliecked against theoretical crit(>ria for evaluating ',:ond al.titurle i ns t ri.iments? Gardner (1975) has criliqutd recent researcli in /tttitudf nioasurenien t , particularly in the area of attitudes toward scKu'-o. He points miL three majoi' de'fects which he has found and that sliould be avoideri: (1) scales which lack any discernible underlying theoretical construct; (2) scales in wh i i h various theoretical constructs are confiuiuded torit.hc'r , i.e., scales ^^7hich attempt to reduce mul t i-di mens icuia 1 attributes to singje scores; (3.) exper i hkmi ta 1 treatments in which there is little discernible relationship between the experimental treatment applied and the scale used to measure its ovitcomes. (p. 101) Both INTTOT and EORTOl have underlying theoretical c(uistructs. TNTTOl'Vs items as judged by a panel of eight social studies and curriculum professors at the Universitv nf Florida relate to teaihiu;; about foreign peoples and places; tliev nrv international '-'uitent curricular objectives. The tests of hypothe.ses coupled with the finding that INTTdT's mean score

PAGE 78

-67was siRniflcnntly diffore-nt fnmi NATinT's nnd that Ihf TN ITOT-NATTOT interrorrela t ion is low (r = .35) confirms that INTKvr is indrcd nipasuring teacliers' perceptions (if ttie relevance of using African content to acliieve international curriciilar olijectives. EORTOT's items were judged hv the same eight-member panel to be curricular objectives specifically requirinii neither national nor intc>rnatlonal content hut anienahle to using either or both. Figure 3 shows that tlie pattern (if in tercorrelnt ions among NAflOT, f'ORTOT, and INTTnr reinforced by the raw mean differences between suliscaJes, suinK^rts tlie theoretical basis und(U-IyinK tlie subscales. NATiny and INITOT whicii reflect very different subject matter domains are shown to be far a[iart by both a low correlation (r = .33) and a large rav: item nu>an difference of 1.20. EORTOT which reflects a subject matter domain that is somewhere between NATTOT and INTTCT since it should theore t i i a 11 .' represent a combination of the two, is i Ound to fall between th
PAGE 79

-68Beginning this val i da t ion sLiulv by oxnmininf; all six independent subscales, tlio researcher has met dardner's second crii.ci'inn that variiMis tlieoreticai constructs not he confcinnded. Where mnlticell inearitv supported theoretical expectations as Ln the case of roiniitive and affective subscales, scales were coTiibinod to inclucie all aspects of a single unidimens ional construct. In this process NATI'O'I', rNTTOl, and FOr
PAGE 80

-69[•Yi^rt h e [ J^_es_e a r c] i New Test oL E0R7X)T'^_Va.l idi ty The first order of business will be to test f'CRTOT's ability to discriminate pre-post attitude changes as a result of institute participation. It was necessary for this studv to use a multivariate F test in order to sininl taneousl y evaluate the ability of (^nch o^ 1]-^^ three scales of interest to detect the expected change in teacher altilude. The three scales in combination did not detect a clianfr,e; hnv;ever, I'ORTOT by itself did. Yet this latter result must be viewed with suspi^ ion because of the nonsignificant multivariate [ test. The current study has eliminated NATTOT and INITOT from consideration to deter-l: pre-post differences, llc^wever, it is necessary to test EORTOT, usinj.', a univariate F test, on a new institute sample independent of that used in tin's ^tudy. If a test based on a large enous;h sample is conducted, a nonsignificant result would indicate that EORTOl does not distiru;uish iire-post changes. A positive result would confirm expectations that it does detect real changes when they occur. Possib ility of a Non-Linear Relat ionsh^ip Between 1 NTTOT and FAKS I'he failure to find a significant linear I'ela t ionshi p between the otherwise valid FICAS subscales, INTTOT and FORirrr, and knowledge of Africa raises questions abotit the nature of the relationship between knowledge of Africa and inclination to use Alrican con'iMit in teni-hinf;. A nonlinear relationship mav exist (see Figure 4') between the two attributes. It is quite possible that the teacher with a little bit of knowledge might be less inclined to use African conten': than the teacher

PAGE 81

-70who knows virtually ncitli i nj: . This phenoiDcnon cimiI d '>rrnr hec-.iupp tlie little bit of knowledjL^p ;i tvpical I encher mif.'Jit arquiti' would proliablv come from Tarzan movies, Tdi Amiu news reports, or f rrMii mv tlio I ori cal relics of the colonial era. Such sources o I" informatitni are hardly likely to inspire a teacher's usinp, African content for curricular purposes. However, as knowlpd;',e of Africa inrreases in both breadth and depth so that teachers can understand African sivicties as makins; a <"oritribution to human understatul i n^;; , these teachers would be more and more inclined to use African content. A preliminary look ;it the tlata in this studv lends support In this theory with respect to INTfOT. For the pilcit sample whose knowledge level was significantly lowfr than the institute pre sample, there was observed a negative correlation lietween INTfdT anrl FAKS . For the institute pri' g,roup there was a medcrate positive relationship. For the institute post group whose knowledge li^vel was significantly higher than the institute pre sample, an even str'Mig.er pi^sitive relationship between INTTOT an(.\ FAKS w.is observed. A studv is needed to adequately test the possibility of a nonlinear re la t i "nsh i ]> between knowledge of Africa and teachers' inclinations to use African content to accomplish International objectives. 4 Attitude Toward A f r i c a n Content Knowledge of Africa Figure A Mypotlieslzed Relationship Between Knowledge of Africa and Attitude I'oward African Content

PAGE 82

-71Valid lty Studies for OjiIut [nte^rna t:ion_al_ Areas In another line of research the Floridn 1 n t r rnat i Cronbach and Meehl (1967), to be sc i cuit i f i (^al ly admissible a construct must occur in a nomologlcal net, at leac:t some of whose laws involve observables. They define a nomologlcal network as an interlocking system of laws which constitute a theory (p. 25"^). Ihe 1 ,iv;s in a nomologlcal net may involve thiee types of relationships. First, observable properties may be related to each other; second, theoretical con.struc-ts

PAGE 83

-72mav be related to observnbles; and third, different theoretical ronstriicts may be related to one another (p. 255) . Ihi.s stndv has beRiin the third part by demons tra tinj;; a moderate re 1 a t icMish I p of th(> consti"urLs measured by INTTOT and EOR'IOT to that meausred by the Worl dm: ndedness Seal e . Yet more needs to be rlnne in this third area to more completely describe the re lationsti ip between the construc-ts pleasured by INTTOT and EORTOT and other constructs. One potentially fruitfiil roiationship to investigate would be that between attitude toward including international content and beiief systems as defined by 0..I. Ilarvcy (1967; Harvey, Hunt, & Schroder, 1961). He has described four systems of belief a 1 onp, a concreteness-abs trac tness continuum. One would expect to find a positive relationship between the abstractness with wliich a teacher thinics and willingness to use international content. Relief ';yst(>ms have been related to attitude chani:!:es (Harvey, 1967), parent child relations (Harvey & Felknor, 1970), classroom atmosphere and student performance (Harvey, Prather, White, ^ Hoffmeister, 1968), anfl alonj^ with subjeet matter and sex of teacher to students' j;rade achievement and student perception of teachers (Harvey, Wells, Sciimidt, & Orimm, 1973). Consequently, if a relationship between attitude toward international content and belief systems can be shown, a big step in developing the nomological net around the new construct will have been taken. Another potentially productive area of research for expanding the nomological network around the construct measured by INTIOI' and l^ORTOT would be to Investigate the rel a t ionsh Iji hetu'eeii Ic'vel of tea<-her cognitive moral development (Kohlberg, 1964, 1969, 1970, 19 7 3) and attitude toward using international content. Erb (19 76) has argued that

PAGE 84

-7:3internat ioTia I or cross-ciil (:nral studies are essential for achieving the higher levels of moral devi.' 1 opment . However, It remaitis to he flenonstrated that a positive relationship does in fact (.-xist between atlitnde toward Internationa], content and higher levels of rmral development. Cronb.ach and Meeh.l's (1967) second category for the laws in a noniol ogica.l net is the relationship between theoretical ccMislructs and observables. To furthervalidate the INTTOl' and MORld f siihscales one would have to observe the r'lassroom teaching of those who achieve different Ic^vels on the two scales to see if differences in the use of international content did exist. Although attitude is an inclination to behave Iti a certain way and not a behavior in its own right (Fishbein, 1967, p. 260; Shaw & Wright, ]967, p. 6; Thurstone, 1967, p. 78), attitudes must predict behaviors to make them worth w(itrvin^ about. Su(di variables as the time teaclvrs spend on international topics, the variety of ways they introduce i n ter nat it^nal content, or tli-:^ number of times they acce]5t student initiatives to introduce international ideas into lessons are some of the teacher behaviors which migjit be looked at and related to scores on EORTOT and INTTOT. Ajjpl lea t ions B asic Res earch There arc two gener"al areas in which the PICAS co\ild be put to good use. One area is in basic r-esearch into attitude formation. As the nomologic:al net is built irp aroirnd attitude toward using International content in the curriculum, i t wi 1 1 be possible to test hypotheses abtnit what variables contribute to the formation of this attitude. A causal

PAGE 85

-74model could be developed and tested by p;?Lb .-nialvsis (Hlalock, 196A, 1971; Heise, 1969, 1975; Wripht, 1921, 19Ti). ?,n[h nio.isures of constructs and observables could be entered into analvsls. Rackf.round variables such as previous exposure to interna t i (^nal situations such as foreign travel or meeting foreign visitors or formal study su'-h as course work or study abroad might form one cluster of variables related to attitude formation. Personality factors might forrn another cluster of traits which could be shown to have a b(\aring on attitudes toward using International content. Environmental factors sucli as school climate, type of community, or aspirations of students might contribute yet a third set of factors affecting teachers' attitudes tov.;ard international content. The possibilities appear endless for i n vest i g,a t i ng the factors related to the formation of the attitudes moasurerl by iNTTOT and F-:ORTOT. E valua t io n and Teac h i n_g The instruments developed by this study, INTIOI, ''ORTOr, nnri FAKS total score, could be used in evaluation studies. Pending, further validation studies on EORT(Vr, that scale could be used to evaluate the results of teacher inservic*^ programs. Intr'iisive institutes and yearlong programs could be candidates for evaluation using, EORTOT. If the teacher participants were not self-selected so that their initial INTTOT scores were very high, INTK^T might also be used to evaluate such inservice efforts. TNTIOT and EORTOT could be used in preservice toacher education to m(_vasure where teacher candidates were in their thinking about i nterna t iona ' content in the curriculum. INTTOT and F.ORICI' used with preservice teachers might be employed either in more basic research or evaluation

PAGE 86

-75of efforts to make tenclier f-nnd i dn t es more incliiicil to use in I or'nnt ional content . Fin;)lly INTTOT and loORTOT mij^tit be used in rrui ] luic t i on with eurriculutn research. F^fforts to imidify curricula oftr-n necessitate chanpes in instruction as well. An effort to i nt erna tiona 1 i ;'e the curriculum in a given school district could be supported by knowledge c^ f where tiie teachers stood in their attitudes toward i nc 1 \id i ns', international content in the curriculum. These instruments could be used in teaching as well as research devices in an effort to change curricula. The objec^tives which make up the items of I'lCAS could be used to ^cuisitize teachers to the applications of international content to both international and either/or objectives, or to nationalistic obiectiv''s on a comjiarative basis. However, using the scales for teaclilui; pre(~ludcs their use with the same group of teachers tor research purposes, Co ac;^! us i o n In this study two attitude scales and a knowledge scale were developed that can be used to further research and evalua t i<-in in the area of international curriculum and ins true: t i on . Validated for African content, INTTOT and EORTOT stand ready for validatioti and ap[ilication in a wide range of international studies areas. These scales have potential for basic research into the formation of atti fides and the consequences of certain attitudes. Ihey can serve a useful puriKise in evaluation, measuring change as a r(^sult of ins(?rvii'e or preservicr treatinent. Finally they could he used for teaching devices by having, teachers fo'Uis on the individual items (i.e., curricular objective=;) in an effort to develop strategies for using international content for achieving the objectives.

PAGE 87

REFERENCES Afric a Stnith c^ Uie Sa|ia_ra: An Objective Tost f(ir S e < cvnd a r j Schools. Pittsburgh: Carnegie-Mellon University, 1968. (ERIC Document Reproduction Service No. ED 030 OiO) African St udie s HaiKfb<25lk J;or Elementary and Secondary S(Jio()] Teachers: Part I. Amherst: University of Massachusetts and Worcester Teacher CorpsV 19 71. Allman, R.W. A study of th»> social attitudes of col lege students. Jo^urnaJ^ o_f Soc ial Psychology. 1961, 51, 33-') 1. Almgren F. Si Gustafssnn. F. World Cit^izen Respons ibi 1 i tv : Assessment, Tec hniques, Deye^l opmen t ,U Stuclies^, Material Construction, and Experimental Teaching. April, 1974. (ERIC Document Reproduction Service No. ED 110 l7 3) Anderson, 1,.F. An examination of the structure ami objectives of international education. Soci^al Educa^t ion . 1968, 32, 639-647. Aron, R. 'If/ects^ o_f Teacher Expectan cie s: MyUi or Ri'ajity? I'apet" presented at the annual n-iM'tinj; of the Amerii^au ['(rs('nne1 and Cuidance Association, Ne'W York, March 1975. (ERlC; Document Reproduction Service No. ED 115 722) Beauchamp, C.A. & Conrau, R.C. ''Ongijzj.id Inal Study in Curriculum Engineering. Paper presented at the annual meeting of the American Educational Research Association, Washington, D.C:., April 1975. (ERIC Document Reproduction Service No. ED 102 670) Becker, d. Education for a Global Society. Bloomington, Indiana: Phi Delta Kappa Educational Foundation, 1973. Becker, J. Perspectives on global education. Social Education. 1974, 38, 678-682. Beyer, B.K. Afj^i^ca Soutji of the Sahara: A Resource and Curriculum Guide. New York: Ranrlom House, 1969. Beyer, R.K. & Hicks, F.P. Images of Africa: A Report on What American Sec_ondary ScliooJ^ Students Know and BelieviAbout Afric.i South of the Sah ara. Pittsburgh: Carnepie-^ie 1 Wui University, 1968. "(ERIC Document Reproduction Service No. ED 02 3 69'') -76-

PAGE 88

-77Beyer, B.K. & Hicks, 1^ . P . A Socinl S_tudi>s Cnrr i cu 1 um Project to Develop and Jest Instruc t iona] Materials, T^eachinj; r.uides, and Content Units on tlve History an^d C/ulture oj Siib-Saharjjn Alrlca for use at Selected G rad e Levels in Sci-ondary Schools, Project Africa: Final Report. Pittsburgh: Carnej-, io-Mel 1 on University, 1/170. (I'.RIC l")oiunient Reproduction Service No. ED 042 673) Bidwell, P.W. Foreign affairs in the colleges: St rpngtheni ng international education. Jiuirnal of Higher Fchicatiini. 1964, 35, A26-433. Rilleh, V.Y. & Zakhariades, C.A. 7he devclopnient and application of a scale for measuring, scientific attitudes. Science Education. 197S, 59, 15 5-165. Billings, C . F, . The challenge of Africa in the curriculum. Social Edluca^tion. 19 71, 35, 139-146 & 15 3. Blalock, H.M., Jr. Causal Infer^n^ces in Nonexpejrirnental Research. Chapel Hill: I'niversitv of North Carolina Press, 1964. Blalock, H.M., Jr. (Ed.). '-'liisai Mock'ls Jn the SociaJ Sc^le'ii-^t-^s . Cbic:ago: Aldine-Atherton, 1971. Bloom, B.S. & Krathwohl , D.R. Taxonomy of Fflucational Objectives: Handbook I, The Cognitive Domai^ii. New York: David McKay, 1956. Bock, R.D. Mult ivar iate Statistical Met^hods in Behavior Research. New York:" "McGraw-Hill , 1975'. ~" ' " Bogatz, B.E. An Inwst^lgat ion of Teacher Expectancies of Instructional Materials : Rese^vrch Repor^ //2. Eugene, Mr(.';;on: Northwest Regional Special Education Instructional Materials ('enter, 1970. (ERIC Document Reproduction Service No. I'll) 75 14 5) Bohannan, P. A Preliminary Review of the Tntercul turn 1 Dlmensiojis in Inte rnat L onal/ Inter cul tural Educati on, (grades K-14. F 1 na 1 Report . Publ i cation No. 156 . Boulder, Colorado: S
PAGE 89

78Campbell , [).T. f« Flsko, D.V,'. Convergent .inrl d i s( riniinnnl va 1 i da t icii by the nnil t i trait-nui] t imr thod matrix.. I'svchol ngi ca 1 Rullotin. 1*^59, 56, 81-105. Carroll, J.R. The nature of the data, or how to i Imr^se a rorrelatlon coefficient. Psychomotr ika . 1961, 26, 347-372. Cecil, C.E. A Summary of tlie Four-Year Rese^irch I'rogram Conducted as a Project of the Huntsvi li e-Mad ison Cmi£ty Education l^mproveTnent Program. Cumulati ve R esea rch Bulletin. Atlanta: Southern Association of Colleges and Schools, 1971. (ERIC Document Reproduction Service No. ED 058 371 ) Clark, R.M. A Stj.idy of Teache^r Beha^viojr ajid At^titudes in Elementary ScluMxls^ wl^th IHjh and l,o_w PupiJ^ A^'liievement . Paper presented at the annual meeting of the American I'>lura t i ona 1 R'-^earch Association, Washington D.C., April 1975. (ERIC Document Reproduction Service No. ED 109 025) Cleveland, H. T^he Ovcnrseas Am_erlcans. New York: ^k''.raw-Hi 1 1 , 1960. Collins, 11.1'. Are You Coirig to Teach About Africa? Mew York: AfricanAmerican Institute, 19 70. (ERIC Document Reprorlnc t i on S(.>rvice No. ED 04 '4 32 4) Cronbac:h, L..I. Coefficient alpha and the internal ^^ti'ucturf^ of lests. In W.A. Mehrens & R.I,. F.bcl (Eds.), Principles o *' Educational and Psycho logical Measurement : A F5ook o_f Selected Readings. Chicago: Rand McNallyT l^fT? . '" ~ "' "' Cronbach, E.J. & Meehl, P.F. Construct validitv in ps vcdio 1 og i ca 1 tests. In W.A. Mehrens 8, R.I,. Ebel (Eds.), Pj: i n_c ip 1 e^'. n\ Educaticuu)! anri Psychol o gica l Measurement. Chicago: Rand Mc'^lally, 1967. (Reprinted from PsycliologicaJ^ H_ul_let^in, 1955, 52.) Dahlberg, K.A. Environmental Studies : Some Problems and Priorities . Paper presented at the annual meeting of the International Studies Association, March 197-i. (ERIC Document Refiroduct ion Service No. ED 103 283) Darlington, R.B. Multiple regression in psychol(igi ca 1 research and practice. Psycliological Bulleti_n. 1968, 69, 161-182. Davis, F.B. Education a I Me.isuremonts and Their Interpretations. 15elmont, California: Wadsworth Publishing, 196^1. Dow, P.B. MACOS : Social studies in crisis. Educational Leadership. 1976, 34, 35-39. "" Easton, S.C. NDEA Centers f(ir International and LanK'iage and Area Stu.l ies -1976j:^7_8. W.tshingto'n, D.C.'r U . S . Dcp ir tment of ileal th. Education, and Welfare, Office of Education, Tune 1977. (Mimeograph Report)

PAGE 90

-79Edwards, A.L. Tedirnquos of Aj^titude S_cale Construction. New York: Appl eton-Century-Crof t s , 195 7. Ehman, L., Mehlinger, II., ^. Patrick, .1. Tow.ird F,ffcc(ive Instruction in Sec onda ry So cia l Studies. Boston: ll(inj;liton Mifflin, 107'^. Eicher, C.E. An Investlgat ion of Elementary Children's PercejUlons of Selec ted Co unt ries o^ the W orld : A Technical Report. Vermillion, South Dakota: South Dakota University Educational I^esearch and Service Center, 1975. (ERIC Document Reproduction Service No. ED 12-3 149) Engle, S.Il. Objectives of the social studi(>s. In }].('. Massialas & F.R. Smith (Eds.), New Challen^gcs jln the Social Studies: Implications of^ Research for Teacjj^lnj^. Belmont, California: Wadsworth Publishing," 1965. Erb, T.O. The Development fif Moral Judgment: Witli a Discussion of the I^A*^' ".£ Cros s-Cultural and International Shindies. Unpublished manuscript. University of Florida, 1976. Fishbein, M. A consideration of beliefs and their r<^le in attitude measurement. In M. Fishbein (Ed.), Readings in Attjtude Theory and Measurement . New York: John Wiley ^< Sons, 1967. Fox, D.J. Ihe Re sea rch Process i n Ecki^c a t_i o n . Ni'w Yotk: Holt, Rinehart and Winston, 1969. Freeh, W.P., Jr. An analvsis of the effect of the Ant h rojiolog.y Curriculum Project material. The Concept of Culture, on Ihe ethnocentric: attitudes of fourth grade students (Do<'toral disseration, Universitv of Ceorjua, 1973). D i sse r ta tion Ahs t rac ts 1 n ternatjona 1 , 1 <5 7 3 , 34, 3830A. (University Microfiims 'No." 73-31, 883)' Freeh, W.P.,Ji-. Theeffectsof cognitive training in ,iti th r(ipol cigy on ethnocentric attitudes. I^'sychology J^n th^ Scliools. 1975, 12, 364-370. Freeman, R.E. Currj.cinum Matej-fals EY'T l_"''i_'^ ' 'l" ''^^ ' Process fn : Work of the Di^l^^lo Valley F^ducation Project. Orinda, California: Diablo Valley Education Project, 1.974. i^ERlC Documr^nt Reproduction Service No. ED 099 283) Frey, S. Tea chers ami Behavior Object Ives . 1973. (FRIC Document Reproduction Service No." ED'08~8'"'84 9) Gardner, P.L. Attitude measurement: A critir|ue ol some recent researrJi. Ed_u(-ational Research. 1975, 17, 101-1(19. Garrison, f . C . Worldminded attitudes of dlU.^f. stiuhtits in a southern university. Journa 1 of Social Psychology. 1961, 54, 14 7-153. Global Development Stud ies : AModel Curriculum for an Academic Year C our se in Global Systems and HuTnan Development at the Secondary and Undergraduate Le^v^ls of Gejne£cil Education. New York: Management Institute for National Development, 1973.

PAGE 91

-80Gordcin, I'. A. Issues in mull iplc trj^ress iou . AnuM'i.can .Journal of Sociology. 1968, 7], 592-616. Guilford, .I.P. h Guilfot'l, R. An nnnlysis of t Itfu-rors in a typical test o[ i iitro vers i(iTi-c :-;t rovers ion . .lournal of Abnormal Social Psychology. 19 3A, 28. 3 7 7-390. Guilford, J. P. & Guilford, R. Personality factors S, K , & M and their measurement. .Journal (>f I'sj/^cliology . 19 36, 2, 1"'^-I27. Guttman, J.. A basis for sc.iliTig qualitative data. American ^"^^i ologij^aj Review. 194/4, 9, 139-150. ' '" '" ' "" Guttman, 1.. The Gomel] Technique for scale and inlensltv analysis. Educa^tionsLl aiid I'syclu^l ogical Measurement. 19A7, 7, 247-280. Hall, S. '. Africa in U.S. f'duca ticuiaj. Materials: Ihirty Probjems and Responses. New York: The African-American Institute, Scliool Services Division, 1977. Harvey, ('..I. Gonceptual sv^-tenis and attitude chaniM-. Tn M. Sherif fi G.l.'. Sherif (Eds.), Atlitude, F,goInvo 1 vement , and Ghanj',e. New York: Tohn Wiley & Sons,' 1967. Harvey, 0.1. & Felknor, G. Par'ent citi'd ri^lations as an antecedent to conceptual functioninjN In G.A. Milt(Mi & I".G. S i mnie 1 (f!ds.), Ka^rlv Experience and the Processes of Site ia I i za t i (ui . New Yor(<: Academic Press7 19^70. -• Harvey, 0..)., Hunt, D.E., .S Sclirodi-r, H.M. Gonceptual Systems and Personality. New Yorb : .lobn Wiley {« Sons, 1''6 1. Harvey, (KT., Prather, M. , White, P.. J., & Ho f fmei s to r , .I.K. In M.B. Miles, W.W. Charter, dr., & N.T. Ga^e (I'.ds.), Readings in the So^cjal Psychology o^f Echication. I'K^ston: All\'n & J'acon, 1970. (Reprinted from American Ed^u<:atioiiai Researcti Journal. 1968, 5.) Harvey, 0.,T., Wells, K., Schmidt, G., & Grimm, f" . Inflects of Subject Matter^ the Beliefs System and Sex of Tca(her'^ and .Studentjs upon Studeiits'_ Grade Achievement ami £^''t^t'pt. ion of leacher. Unpublished manuscript. University of Colorado, 19 73. Flavlicek, [^.1,. ^ Peterson, N.b. Effects of the '.M o 1 at i oil of assumptions upon significance levels (if the Pearson r. Psvi-bo logixa 1 Rj.!_lletin. 1977, 84, 373-377. Hayden, R.l,. lnt;ernat i(>na 1 i z ing Public Edui-.ition: Wliat. tli^i? Stjatcvs Are DoJ^ng. Paper presented at tlu^ annual meeting, ol the Comparative and International Education S(ieiety, loronto, February 1976. (ERTC Document Reproduction f^ervice No. ED 12 4 442) Heise, D.R. Problems in path .inalysis and '-niisal inf(M-<-nce. In E.F. Borj'.atta (Ed.), Soc io 1 ogj_cai^ Methodology 1969. San Francisco: Jossey-Bass, 1969.

PAGE 92

-81lleise, [i.R. Causal Analysis. Now York: John Wiley /v Sons, 1975. Homme, 1,.E. A Demonstration of tlie Use of S(^ 1 fI nst rnc t iotia 1 and Other Tcachiiig Techniques for Reme^dlal Instruction of fciw-Ach ieving Adolesc ents ijl Reacling and Mathematics : iechnicril Progresj; Repo^rt No. 2. Albuquerque, New Mexico: TMI Institute, 1968. fi:R!C Document Reproduction Service No. ED 018 111) Interc ul t ura I Unders t.ind i ng Project: Final Report. Pittsburgh: Allegheny County Schools, 1972. (ERIC Document Rcprcnluc tion Service No. ED 081 705) Jaccard, .1., Weber, .1., <^ Iiinrimnrk. .1. A iiiu 1 t i 1 1 a i tn'ul t inn^t tuid analysis of four attitude assessment procc'lures. .Journal of Experimental S9_cAii Psy chology. 19 75, 11_, 1 49-1 54 . Johnston, J. E conometr ic Method_s (2nd erl.). Nev.' York: Mcfiraw-Hi] 1 , 19 72. Jones, K. An Ewiluation Report on the Regional Etlura t iona I Agenc ies Project (Al abama , Louisiana, Tennessej?, Te>^Ts) in Inf ern.i t ioiml Education . Austin: Texas Education Agency, Office of International and Bilingual Education, 19 70. (ERIC Document Reproriuc t ion Service No. ED 048 086) Kerlinger, F.N. & Pedliaziir, E.J. Multiple Rc\e,ression in IkJiavixir Research. New York: Holt, Rlneliart anrl Winston, 1973. Kirk, R.E. Experimental Design: Procedures for the 15ehaviojr Sciences. Belmont, California: Brooks/Cole Pub 1 i sh inf. On., 1968. Klassen, F.H. & Moore, R.S. Development of an Instrument £or the Study ^A I nternat ional Educatum i_n Teaclif!r Ixlucation Progr-ams of U.S. Col leges an d ffoJ_ve^rs^ijL ies : Final Report. Washington, D.C.: American Association of Colleges for i'eaiher Education, 1968. (ERIC Document Reproduction Service N(}. ED 032 010) Kohlberg, 1.. Development c^ f moral character and idcr^lcTf^y. In M.I,. Hoffman and L.W. Hoffman (Eds.), Review of C!li i 1 d Development Researc h . New York: Russell Sage Foundation, 1964. Kohlberg, I, . Stage and s(M.]Mence: The cog.n i t i ve-devel opmen t a 1 approach to socialization. In D. (kislin (Fd.), llandbo(ik of^ S^cjcia 1 iz;i tlon^ Theor y. Chicago: Rand McNally, 1969. ^ ' Kohlberg, L. Education for justice: A modern stat'MTient of the Platonic view. In T. Sizer (Ed.), Moral Education. ("nmbt i dge : Harvard University Press. 19 70. Kohlberg, 1. The claim of moral adequacy of a hipli(>st stage of moral judgment. The Journal of P_h i 1 o s^pi by . 19 7 3, 7 (J, 6 30-646. Krathv.'ohl , D.R., Bloom, B.S., 8i Mas la, B.R. iaxo^iv'm)' of FjJ^ucci ti_onal Objecti ves : Handbook 11, The Affective Doma in. New York: David McKay, "1964.

PAGE 93

-82Kuder, C.F. & Richardson, M.W. Thi> ttuMiry of tJic Ci-timation of tost reJinhility. PsyclToim't rika . 1937,-2, ISl-H.n. Lee, ,J . R . Teachinjj Soj^inl Studies In the !•', 1 etiKMit ,it y S(ho .^^cale. In M. Fishbein (Ed.), Readings^ In Attitude Theory and ^k^lsurelnen t . Mow York: .lohn Wiley & Sons, 1967. (i'.xcerpted from the ap|iendi" of A technique for the measurement of attitudes, A_rchl_ve_s of Psychology, 1932, 22.) Llkert, R. A technique for t ht> measurement of ntiitudes. In C.V. Sunim'M/s (Ed.), Att it_u_d_e MeasutojiiejU. Cliicajio: Rami McNalJy, 1170. (Excerpted from Archives of Psychology, l'^12, 22.) Lindquist, E.F. Design and An/ilvsi.s of Fx[ie r imeiits i ii Psychology aiul EducatJ^o_n. Boston: IIonght(in Mifflin, 1953. Lord. F.M. & Novick, M.R. Statistical Theories of Mental Test S_ec>_res . Reading, Massachusetts: Add ison-Weslev , 1968. Hacias, C.C. Programmed Learning as Used in the Tafoma Public ScTio_ols. Included in the proceedings of the 20th Annual Pacific tjorthwest Conference on Foreign l:ang,uages, Portland, drfiion, April 1969. (ERIC Document Reproduction Service No. ED V, 3 ri) Marion, K.M. Kappan interview: Kevin Michael ''laiii-n lalks about peace studies. Phi Deltf) Kappan. 1973, S^, 187-18". Masia, R.A. I'rofile of the current secondary school sni;ia1 studies curriculum in Nortti Central Assoc iat ion Schnols. 'ihe Not^tih Central Association Qua_rtejly. 1963, 38, 20 S213. Massialas, B.C. & Cox, C.G. Inquiry'; in Social Studies. New York: McCraw-Hill, 1966. Maxim, C-.W. The role of re:u'arch in competencv bara^d teacTier education. Educa^tion. 1974. 93, 94-96. McCarron, T.
PAGE 94

-83MorelnnH, W.D. Curri r iil nm trends in the sorinl studipf?. Social Education 1962, 2fi, 73-76. Murphy, F,.,I. & Stein, H. Teaching Africa lod ly : A Ilaudhrxik for Teacfiers an_d Curr iculj.im Planners. New York: ("itatinn I'riss, 197 i. Murphy, G. (Ed.). H^L"!'^'! Nature ajui Endurinj_; Peace. Boston: HoughtonMifflin, 1945. ' " '"' '" Newman, A..1. & Ware, W . Acstehtic perceptions and wor 1 dmindedness — A research note. The New Era. 1976, S7, 't5-^i(^: ^'< . Nie, N.ll., [lull, C.H., Jenkins, .I.e., S te i nh rf>nner . K., f. P.pnl , D.H. Statisti cal Package for the^ S^icial Sciences (;'nd ed.). New York: McGraw-Hill, 19T^V' Okey, J.R. Tli£ E ffec ts of Mastery^ Teaching Strategy on I'cacher A 1 1 i t ud e s and Pup_i^l Achi eve ment . Paper presented at the annual meeting of the National Association for l^esearch in Science reacliing, Detroit, March 1973. (ERIC nocument Reproduction Service No. ED ORO 311) Oliver, D.W. & Shaver, J . I\ Teaching Public Issues in the 11 i f.h School. F5oston: Houghton-Mifflin, 1966. Olson, C.L. On choosinj; a 'est statistic in mu 1 t i v,i r i ;i te analysis of variance. Pjjycho I og ical Bi_iJ_Ictin. 1*^76, Pi\. '37^-586. Oppenheim, A.N. Questionnaire Design and Attitude Measurement. New York: Basicl5ooks , 1 9f^6T Osgood, C.E. Crc^ss-cu 1 I ura 1 compa lahi 1 i ty in attitude nuMSuremc^nt via mul t i 1 injuial seman) ic differentials. In H. Eishhcin (E(i.), Readings in Attitude ITie ory ajid Measurement . New York: doim Wiley & Sons, 1967. " Osgood, C.E. & Suci, 0. I. factor analysis of moaning. Journal of E_xp e r_i_ine n_t a 1 P_s y cdi oj o g y . 1 9 5 '^ , 50, 325-3 3 8 . Osgood, O.E., Suci, .J., f, lannenfiaum, F.H. Tljo Measurc^ment o_[_ Me a n i n g . Urbana, Illinois: University of lllineis Pre.ss, 1957. Oshima, E.A. Changes in attitudes toward sci(mic<> and confidence in teaching science of priispective element ary teacdn-rs (Doctoral dissertation, Oklahoma State University, 1966). Dissertation Abst racts I_nt ermitional , 1967, 27, 3981A-A37^A. CUniverslty Microfilms No. 67-07, 268) (ERIC DocuiiKMits Rcji todnc t i v; Hampshire: American Universities Field Staff, 1974. (ERIC Dommeiit Reproduction Service No. ED 099 286) Pearscin. E.S. The analysf-^ of vnrlntire In cases of nen-norm;il varlatl(5n. Biometrika. 1931, 23, 114-135.

PAGE 95

-8^Phillips, C.S., .Ir. The present: W'^rld rhnl1ciir,P tk'as_uremcnt of Attitudes. New York: McC;raw-H i I 1 , 1967.' " " " ----Shrigley, R.L. The cf^rrelation of science attitudo nm] science knowledge of preservice elementary teachers. Science Education. 1974, 58, 143-151. Smith, H.P. Do intercul tin a 1 experiences affect attitudes? Journal of Abno rmal Social 1'sychci logy . 1955, 51, 469-47 7. Snider, J. & Osgood, C.F. ScMTiantic Di^f f crcMi t ial 'IcclKiique^ A Source Book. Chicago: Aldine Publishing Company, 1''69. Stearns, T. I deas fi^ir t_h_G Dev£l.oj5^ment of I'rograms Relating to the Interna tio nal Scene a nd H_s Role in the Schools. Eansing, Michigan: Michigan State Board of Education, 1966. (F.RH.: liocument Reproduction Service No. ED 02 3 6A4) Taylor, 11. The World as Teacjior. ('a rbonda I (^ , Illinois: Soulhern Illinois University I'rtvss, 1969. Also [uiblisbed under the title The Wor ld and the American TeachcM:. W.ishingtoti , D.C: American Association of Colleges for Teacher Education. \^6H. Teac h tiig about Afric a In the Societal Studies Curricul uir. Raleigh, North Carolina: State Department of Public Instruction, Division of Research, 1972.

PAGE 96

-HSTliiirs toni-' , L.T,. Psychophysical nnn lysis. American Ic'iriial of Psychology. 1927, 38, 368-389. ' ~ "" TJiurstone, I,.L. liietiry of attitude measiu ( mfn t . Psyciio i nj^i ca 1 Bulletin. 1929, 36, 222-241. ' " Thurstonc, L.L. The measurement of social attitudes. Journal of Abnormal Social P sycho logy. 19 31, 26, 2A9-269. Thurstone, I,.L. Attitudes can be measured. Tn H. Fishbein fF.d.), Readings in Attitude Theory and Me.a^snremen t . New York: John Wiley & S(^ns, 1.967. (Reprinted from American Journal ('f Sociology, 1928, Tinkelman, S.N. I'lanning the objective test. Tn I^.T.. Tliorndike (I'aI.), Edu catio nal M eas urement (2nd ed.). Washington, 1) . C; . : American Council on Education, 1971. The UNESCO associated sdiools project in echicat ion lor international cooj^eration and peaces. Educational Documentation and \ nlormatioTi Bui letin of the 1 ntei^rna tjoruil Bureau of Education. J97S, A 9, 29-3 3. Van de Gccr, .T.P. Introduction to MultivarKite Analysis^ fr^r the Scm: i^aj_ Sciences. San Francisco: Freeman, 1971. Veldman, D.J. Fortran Programming for the Behavixnal Sciences. New York: Holt, Rinehart and Winston, 1967. Waimon, M.D., Bell, D.D., & Ramseyer, C . C . ll:e I'f loots <.f Knowledge About S ubje ct Matter mi the Performance and Altitudes o^f ProsjMic^tive Teachers. Normal, Illinois: Illinois State I'n i v(.' rs i ty , 1971. TerVc Document Reprodu'tion Service No. ED 0S2 1 Ul) Willmcr, J.E. (F,d.). Africa: Teaching Perspectives and Approaches . Tualatin, Oregon: Geog,ra[ihic and Area Study Publications, 1975. Wise, J.H. An Appalling Ignorance of the World: The I'rjce ol; Aff luence ? Paper presented at the annual meeting of the Institute of Australian Gecigraphers , Wollongong,, Nev; South Wales, Australia, August 1975. (ERIC Document Reproduction Service No. ED 111 7/(5) (a) Wise, J.H. Student deficiency in basic world knowledge. Jo^urnal o^ Ge_ogrjip_hv. 1975, 74. 4 77-488. (b) Wiseman, S. The erlucational obstacle race: Factors that hinder pupil pri.igress. Educational Research. 1973, 15, 87-9?. World History Scries: Africa. Rockville, Maryland: I'oard o\' Education of Montgomery County, 19 72.

PAGE 97

-8fiWright, S. Ccirrel.i t i(>n nnd cnusntion. Joiininl of Agi i c:ii! t iiral Research. 19Z1, 20, 557-585. .' " Wriglit, S. The method of pith roe f f Icien t s . Ann.ils of Mn tliciiiat i ca_.l. Statistics. 1934, 5, 161-215. ' '

PAGE 98

APrr.NDix A INSTRUMliNT USKD TO CATIvmRIZF 90 ORJKC r[ VF.S AS MATIONAI.IST 1 f. , INTERNATIONAL, OR F,[ri!ER/OR, INCLUDINO THE TAM.Y OF RKSUf/IS Attached is a list of 'i4 cognitive and 3fi affective objectives which could be selected in develop i n;,', a social studies cnrriculum. Some objectives focus on America and can best, perhaps cmt 1 v , b^ pursued using content derived from the American experience. Otiier objectives are more international, or global, or i ntercul tura 1, in focus so that Miey could only be achieved by using content whicli i-S derived from otiier cultures, nations, and peoples. A tli i rd set of objectives are harder to classify as to their American or i nti-rna t ional focus. That is. tliey could be achieved by using American content or i n tern/f t i imm 1 ccMitent or a combination of both. After reading eacfi idijictive, please mark tiio cat^gorv into whi(h you think it should be placed. -1 f an objective would retjuire mainly American ccmtient to achieve, place a check in Colunm 1. -If an objective would require international or global cMTntei]_t to achieve, place a check in Co^lumn II. — If an objective does not specificallv r(H|nire iM'ther American or international conteTit but could be develoiied vitii e;i t^lie_r cijboth, place a check in ('o 1 umn III. -87-

PAGE 99

EXAMPLES: Co 1 uiim I I J ni SLudpiits sluMild . . . X '1. untie rstnnd thf l) . S . ('tni.st i I ut ion X 2. have a knowledge n( CUi i iicf^':^ dynasties X 3. know the role of natural rr'sources in product ion COGNTTIVF, OB.H'CTlVrS Co_'-__H?lIL I I! TIT Students should . . . 6 2 1. h.ave a knowledge oT world history 6 2 2. {'.oinprehend the p,nlf hetvecn rich and poor count r i OS 8 3. understand the consumoT's role in the American economy 8 A. understand social coni lict 8 5. comprehend the role of labor in production 6 2 6. understand difiCrent w.ivs (^f life on different continents 7 1 7. have a knowledr,p of social change in America 7 1 8. comprehend the process of at!.inp, in our society 8 9. understand the causes and (effects of ti is tori c a 1 e ve n t s 7 1 10. know ab'Tut the inip;ict "f technology on modern life in America 7 1 11. have a knowleds^e of the n.' 1 a t i oiish ip between environment and wavs of making a living around the wor 1 d 7 2 12. understand the relationship between colonialism and nationalism 8 13. have a knowledge of social interaction 8 14. know about the role of government

PAGE 100

-89C o I umn I T r III Sfrii'lonts sliotild '. . . 6 2 IS. hr'ive a know] ''ilr.c' nf v.-.M-lrl j',rnp.rnphy 8 16. have a chnmn 1 1»}', i I'a 1 1now 1 (Mi;;e (>! iiiajor American historiial evenis 7 1 17. have a knowledge of tlie strurture of American government 5 3 18. have a knowledfie of diffcrcMit cultures 7 1 19. compreh(Mid the principles of American democ racy 1 7 20. understaml the principles of cooperation among people 7 1 21. know the location of I he world's major resources 6 2 22. understand the American ccom^mic system 1 7 23. understand int(Maetion amonp. various social groups II 6 24. understand urh.ni iza t i "u 6 2 25. have a knowledge of v.irious American ethnic groups 2 6 26. understand the causes of war 5 3 27. comprehend different svstems of government 5 3 28. comprehend the pluralistic nature of American Scheie ty 8 29. know the major products of the student's home state 3' 3 3' 30. comprehend American foreign policy 6 2 31. understand difiercnt pitierns of family organization around the world 3' 4' 3' 32. comprehend America's rise to a position of world power 3 8 33. rnmprehenri cultural fliffusicMi 1 7 34. understand the exchange of goods and services

PAGE 101

-90_ ___Cill ' 'M! T n TTl Sturlents should . . . 6 2 Vi . know of di f fcrfii ( pi t t orii'-'; of conRunip t i on throuj'juMit tho world 7 1 3ft. oomprt'liend thoir roles as members of their (iwn coiTimuii i t ies 1 7 37. understand the balance of nature 1 7 38. iinderstand tlie re 1 a t i (-nshi p between continuity and change 7 1 39. know about the ;;r{iwfh of the American nation 7 1 AO. have a knowledt;e of the im[iact of Anglo culture on American Indiaii groups 5 3 41. understand trade relations between nations 1 7 42. know the role of eronomic specialization in development 1 7 43. understand the role of leadership 3 5 44. comprehend nationalism 7 1 45. know the elements Tuvcssarv for ecoTiomic development in the (^m'-rj'j'ng nations 8 46. understand the rr-l a I i'-nsh i p of westward expansion to America's development 1 7 47. apply iiistorical knowledge to the understanding of ci'>ntempora.ry problems 5 l' 3' 48. understand domestic racial conflict 6 2 49. utidc^rstand thiproblem of the world's population grovv/th 8 50. recognize various landforms 1 7 51. understand tlie meaning, "f racism 6 2 52. know about the impart of technology on d eve J. o p i n g n a t ions 8 53. know the locatinn of n'ajniU.S. cities 11 6 54. comprehend the iTieaning of justice

PAGE 102

-91AFFECTIVK OB.lF.CTIVrS C<.i 1 unm I II III St iiHonts should . . . 7 11. ,'ippreciate tlie p j ^',n i f i < .inn^ cf our shrlnkin!; worJ d 3 5 2. transcend ethnorentr i sni 2 6 3. hav^e a desire to fonserve natural resnurres 2 6 A. make intelligent adjustments to rliange 1 6 5. arcept the respons ih i 1 i t v to function well as a t^roup member 6. develop a ci^mniilment to etadicating injustice 7. appreciate ttic values of other cultures 8. appreciate thi> con t r i luit i otis of .-i 1 1 ethnic groups to AmericaTi devi' 1 opmen t 9. become a happy familv men'ber 10. exj-ierience multif)1e l(.'\-altieR 11. accept the limitations of national power 12. appreciate artistio expression 13. be proud of our national accomplishments lA. appreciate the aesthetic in our present c i villzatlon 3 5 15. become a judicious consumr-r 8 16. respect the ideas and opini(Tns of others 7 1 17. develop a global [)erspcc t i ve 8 18. promote social progress 5 3 19. accept thc^ moral implications of mankind's growing into rdcpendence 7 1 20. develop a senst^ (-f !>.• 1 ong i h);, to a world communi ty 5 2 21. feel pride in being an American 1 7 22. accept change as a natural feature of the human condition 1

PAGE 103

-92CoJuTim I ir I] I Si.n.l(>nLs sliciihl . . . 1 7 2!. hecdini' cununi t I fd to i .1 1. ii iiin 1 .let ion t ') solve snv in ] proh] cms 7 1 2/4. support efforts to proteri Aniofifn's envi ronment A 4 25. feel prirle in all huin-!ii ,-Mconip 1 i shnion ts 4 4 26. demonstrate pjvul c i t i ;'ensli ip 7 1 27. npprec late the ripjits and responsibilities of American c i. t i zen'^h i p 11 6 28. a[)preeiate contributions to society of tlmse who represent different social backRrounds 5 3 29. be committee] to human welfare on a worldwirle basis 2 1 '5 30. demonstrate a tolerau'o for sfirial differenci 6 2 ,il . become committed to assuming' civic responsibility in one's own (onmunitv 1 7 ')2 . tolerate arabijuions and conflicting informat ion 4 4 31. act in accord with demociatic principles and values 7 1 34. develop a respect for foriMpners as human equa Is 7 1 I'i . acquire wo 1 rdmi udedness 5 3 36, be aware of their misconceptions about other people "Multiple responses were recorded by some respondents. Some res-poiisi^s were omittrvl bv resjiondoni s .

PAGE 104

APl'ENDIX P, AKS TTF.M ANALYSIS Tabl The Blserial Corre I 1 1 ions Between the '30 Item?^ Seleeted for Inclusion on the FAKS and the FAKS Total Score Item

PAGE 105

APPENDIX C FLORIDA INTERNATIONAL CUKRl CLn.UM ASSLSSMLNT SCALES SHOWING SUBSCALE STRUCTURE Inst rur, tilons for Completing the Scales TASK 1 Beginning on the next page is n list of -'18 objrctive.s ivhirh could be selected in developing a K1 2 scIuhiI curricnlum. '.'irious teachers and Pupervisc>rs would place different emphasis on tlio iniportame of choosing these several objectives. Now you have the oi)porlunity to rate o,u Ti cih j ec t i ve according to tiow important: you think it is to Inclnrle it in the K-i;^ curriculum. Record your judgements in Column A by circling, the numlicr that corresponds t(i your view. Use the following as a guide tn marking your responses in Column A: C'luals "insignificant" au'l S equals "absolutely essential;" Tiumbers 1 throuj'ji A represent increasing degrees of importance between the two endpoints. TASK 2 After evaluating the imiiortance of eaoh objective, you are asked to make a second judgement concerning each objective. Curriculum objectives usually can be achieved in several different ways, depending on available materials, teacher and student interest, etc. In column B please record your judgements about the relevance of the study of AFRICAN peoples and places for achieving each objective. Regardless of what you know about spei.'ific materials that may be currently available, rate the usefulness of employing, AFRICAN content for reaching the stated objective. Use the following as a guide for marking your answers in Column B: equals "useless" and 5 equals "i ndispens ib le ; ' numbers I through 4 represent Increasing des^rees of usefulness between the two endpoints . -94-

PAGE 106

-'}5AN EXAMPrE A Students should ... B 0(1)2 3 '1 5 1. comjireticnd American fnreijrn policv 1 ^ \}) ^ 5 12 3(4)5 2. demonstrate a tolerance for social 1(2)3 4 5 differences A Students should ... B 1 2 3 A S 1. know ahout t:he impact of tech1 2 3 A 5 NATTOT nologv on modern life in America 12 3 4 5 2. develop a sense of belongin.c to 12 3 ,4 5 INTTOT a world community 12 3 4 5 3. understand the exchange of Roods 12 3 4 5 EORTOf and servifcs 12 3 4 5 4. become committed to assuminjj civic 12 3 4 5 NA li'dT responsibility in one's own commun i t y 012345 5. understand social conflict 01 2345 EORTfvr 12 3 4 5 6. applv historical Inowled)-/-to the I 2 3 4 5 EORTOT understanding of contemporary prob.l ems 12 3 4 5 7. comprehend the gulf between rich 12 3 4 5 INTTOT and poor r-ountrles 12 3 4 5 8. become committed to rational action 12 3 4 5 EORTOT to solve social problems 12 3 4 5 9. tolerate ambiguous and conflicting 12 3 4 5 EORTOT information 12 3 4 5 10. understand the balance of nature 1 2 3 4 5 EORTOT 12 3 4 5 11. understan'l the problem
PAGE 107

-96A Students sliould . . . B 12 3 4 5 16. apprrrinte tlie sij;nlficanco of our ] 2 3 4 5 INITOT shrink liip, world 1 2 3 A 5 17. promote S'lclal progress 1 2 3 A 5 F.ORTOT 12 3 4 5 18. be pmud of our national arcom12 3 4 5 NATTOT pi i shnient s 12 3 4 5 19. have a knowledge of world Keopraphy 12 3 4 5 FNTTOT 12 3 4 5 20. understand the relationship of 12 3 4 5 NATTOT westward expansion to America's devel opnient 12 3 4 5 21. recognize various landforms J 2 3 4 5 F/IRTOT 12 3 4 5 22. feel pride in bein^ an American 1 2 3 A 5 NATTOT 12 3 4 5 23. have a knowledge of the structure^ 12 3 4 5 NATTOT of American Kovernnient 12 3 4 5 24. have a desire to conserve natural 12 3 4 5 roinof resources 12 3 4 S 25. understand interaction among var1 2 3 4 5 i'.OKlOl' ious soc i 1 1 groups 12 3 4 5 26. understand the re 1 a t i .uish i p be12 3 4 5 EORTOT tween continuity and chanr.e 12 3 4 5 27. understand different wavs of 1 i f1 2 3 4 5 INTTOT on different continents 12 3 4 5 28. have a knowledge of the impact of 12 3 4 5 NATTOT Anglo culture on American Indian groups 12 3 4 5 29. accept the limitations of national 12 3 4 5 EORTOT power 12 3 4 5 30. have a chronological knowledge of 12 3 4 5 NATTOt major American historical events 12 3 4 5 31. develop a global perspective 12 3 4 5 TNfTOT 12 3 4 5 32. understan.l the causes and effe.-ts 12 3 4 5 EORTOT of Ills tori cal events 12 3 4 5 33. understand urbanization 12 3 4 5 EORTOT 1 2 3 4 5 34. appreclaf(> contributions to society 1 2 3 4 5 EORTOT of those who represent different social backgrounds

PAGE 108

-971 2 3 A 5 12 3 4 5 1 2 3 't 5 12 3 4 5 12 3 4 5 12 3 4 5 12 3 4 5 ]_ 2 3 -4 5 12 3 4 5 J. 2 3 4 5 12 3 4 5 12 3 4 5 12 3 4 5 12 3 4 5 Students should . . . 35. have n kn"wledp,e of soc.L-il chanpr' in Amf^rir.i 36. accept rhanj^e as a natural feature of the human condition 37. appreciate artistic expression 38. experience multiple loyalties 39. have a knowledj^e of various American ethnic groujis 40. have a knowledge of the relationship hetveen environment and vavs of making a living around the world 41. understand the meaning of racism 42. have a knowledge of world histor\' 43. understanil the consumer's role in the American econcimy 44. understand the relationship between colonialism and nationalism 12 3 4 5 NAT 10 T 12 3 4 5 EORTOi'

PAGE 109

ArrENDIX D FLORIDA AF'RTCAN KNOWLEDflF, SCALF, Questions 1-4 reler to the map of Africi shown below. For each question choose the letter from the map which best completes the statement or answers the question and circie the corresponsi np letter beneath each question. 1. The equator passes through : A. R R. S C. W D. Z 2. Of tlie following, the area with the higliest elcvafinn: A. T R . W C. U D. Z

PAGE 110

.()93. Victoria Fails on tlie Z.imhezi \^ closest to: A. i' P, . 11 C . VJ n. Y A. The country of Nigeria is closest to: A. I' r> . U C. X I). Y Oviestions 5-30 are nn3l t i |i] e-choico questions. For each (juestion. select the word or phrase whicli best completes the statement or answers tlie question and circle the letter corrospondi ni; t" the correct answer beneath eacli question. 5. The largest lake in Africa is Lake: A. ('had B . Victor ia C . '['angativ 1 ka D. Malawi 6. The highest mountain in Africa, Mount Kilimanjaro, is located in: A. Tanzania B. Kenya C. Zaire D. Republic of Soutii Africa 7. The distance from the northern tip of Africa to the soutliern tip is approximately: A. Z'SO miles B. 5,000 miles C. 14.000 miles D. 25.000 mi les 8. The west coast of Ai;rica touches the: A. Pacific Ocean Fi . Arc I ic Ocean C. Indian Ocean D. Atlantic Ocean 9. In tlie 1600's Africans: A. had many strong kingdoms B. lack(^d towns or cities C. began worshiping Furopoan eyplor'-r'D. developed forms of political organ i za I i i"i for the f i t s t t imc 10. Which of the following existed in Africi: A. Sonjdiai Fmjiire H. Monjrol Kmpi re C: . fJupt a F.mpi re D. Aztec Empire

PAGE 111

-lon11. Wliii-h f>f the following has hocn most inriiitiirtd hv Araliir culture? A . N i ^e r i a B. Zanzibar C . El. li Lopia D. rihana 12. The greatest numher el coiuitries in Africa south ef the Sahara became independent: A. before 194 5 P.. between 1945 and 1960 C. betwfHMi 1960 and 19 70 D. after 1970 13. Munfo Park, Sir Riihard Burton, and Henrv Stanley: A. were explorers B. served as Christian mi ss i oTiar i es C. dir^d in Africa I). came from the Uniti-d States 14. The Berlin Confer ence of 1884-85 resulted in: A. the independence of many African countries B. the establishment of fluropean claims to most of Africa C. the division of Cerm.in colonies betwecMi Fns'laiid and f'rance D. a policy of training.; Africans for eventual self-government 15. Which of the following is a Bantu language? A. Khoisan r. . Zulu C. A-nhara n . A f r i k a a n s 16. Of the following countties, which has the larr.i'St white population? A. Nigeria B. Liberia C . Za i i-e D. Republic of South Africa 17. Traditional religions ^^f Africa south of the Sahara stress a belief: A. in Heaven as reward and Hell as punishmi-nt B. in a Supreme Force or Being who created the universe C. in the Ten Commandments D. in the necessity of human sacrifice to |->lease tlie gods when they are angry 18. The econcimy of the Masai people of Fast Africa is based on: A. Industry B. Agriculture C . Mining D. Herding 19. Cab'^ra Bassa Dam, potentially one nf the world's largest hydroelectric projects Is located in: A. Nigeria B. Mozambique C. Republic of South Africa n . Za 1 re

PAGE 112

-]01 20. Wln'ch of the followinj; pairs matches a prodiut with t!ie area where it is produced? A. Silk Chaiia R. PetfoleuT'i Tanzania (; . R nil iter Liberia D. Tea Niper 21. A chief product of Zaire is: A. peainiLs B . pet ro 1 cum C. copper I). wool 22. Most people in Africa south of the Sahara earn their living, workinj; as : A . farmers Pi . hunters C. factory workers n. fishermen 23. Whi'-h European cnuntry retained its colonies until after 1970 only to lose them in wars of independence? A. France r. . I'.ritain (' . Relf, iiim n. Portugal 24. Zimbabwe is the African name lor: A. Cnnfji R. Repulilic of South Africa C. Rhodesia n. Southw(>st Africa 25. Which person served for 4''i years as the Frnjiernr of Ethiopia? A. Sekou Toure P> . Hai 1 e Solas si C . Jul i us Nye re re D. Eeopold Senghor 26. The man that unified Zaire after a post-independence civil war was: A. Yakubu Gowan B. Sekou Toure C. Kenneth Kaunda D. Mobuto Sese Seko 27. The leader of independent Tanzania has been: A. ,T(m'ii> Kcnyatta R. Mt>lMitn S(\sr. Sekd C . dull us Nye re re 11. S a mora Machel 28. The capital of Ghana is: A. Kinshasa B. Dakar G. Nairobi D. Accra

PAGE 113

10229. The Mau-Mau rebellion Look plnce in; A. K(-nva n. Chnun C. Rcpiihli'of South Africa n. Liberia 30. In terms of dollar value, the most lirpc^rtant ' xports of Afriea south of the Sahara are: A. mineral products B. agricultural products C. manufactured goofls D. services

PAGE 114

APPFNDIX E WORLDMI NDE DNF SS SCALE Please read each iteTn rarefnllv and circle the clioice tliat: most close corresponds to your position. Thank you for your cooperation. Choice Codes: SA = Strongly Agree; A = Agree; MA -^ Mildly Agree; MD ^ Mildly Disagree; D = Disagree; SD = Strongly [)i sagree .''' 1. Our country should have the right to prohihit certain rac-ial and religious groups from entering it to live. 2. Immigrants sliould not he permitted to come into our country if they compete with our own workers. 3. It would be a dangerous procedure if every pers'in in the wr^rld had equal rights which were guaranteed by an international charter. 4. All prices for exported food and manufactured '.'.nods should hi' set bv an international trade committee. 5. Our country is probably no liettor than nianv fathers. 6. Race prejudice may be a good thing for us because it keeps many undesirable foreigners from coming into this comitrv. 7. It would be a mistake for us to encourage certain raci'il rroufis to become well educated because they might use their kno\>'l ed;',e against us 8. We should be willing to fight for our country without questioning whether it is right or wrong. 9. Foreigners are particularly obnoxious because of their religious bel iefs . 10. Immigration stiould be controlled by an international organization rather than by each country on its ov^;n . 11. We ought tci have a world government to guarantee the welfare of all nations irrespective of the rights of anyone. 12. Our country should not ctu'pcrate in anv intt'rnat ional trade agreements which nttcmpt to better world economic conditions at our expense. -10 3-

PAGE 115

10413. It would be better to be a citizen of the world than of any [lart icul.-ir country. lA. Our responsibility to people of other races (Minht to be as great a=; our responsibility to people of our own race. 15. An international committee on education ';hen.ld have full control over what is taught in all countries about history and politics. 16, Our country should refuse to cooperate in a total disarmament program even if some other nations agreed to it. 17. It would be dangercius for our country to make internatinnal agreements with nations whose religious beliefs are an t af;nii i s t ic to ours. 18. Any healthv individual, regardless of i ace or religion, should be allowed to live wherevf.ir he wants to in the world. 19. Our country should not participate in any international organization which requires that we give up any of cnir national rights or freedom of action. 20. If necessary, we ought to be willing to lower our statul.ird of living to cooperate with other countries in getting an equal standard for every person in the world. 21 We sluuild strive f
PAGE 116

10531. We should teach our chLlclreii Lo uphold Uu? wc 1 f .:i re of all people every\.;here even though it may be ag
PAGE 117

BinCRAPlllCAL SKETCH Born in Fort Wayne, Indiana, on Fohruary 18, 19A3, Thomas (Xien Erb was educatod in Fort Wayne's public schools boforo attcndinR neFauw University from whicli ho was graduated Phi Beta Kappa in ]967. The following, vear Mr. Erb received his M.A.T. des^ree from Northwestern University, joined Phi Delta Kappa, and bepan his teachinp career. After four years of toachinR social studies in Wilmette. Illinois, Mr. Erb taught for one year at the University of Ch i_ca£;o Iab(^ratorv Sfhool. In J971, he married Karen Simmons, and a year later t hf^v both bcRan teachinp, at the international school in Luanda, Ani'.ola, Africa. In 1974, Mr. Erb resumed his graduate education in curriculum and instruction at the University of Florida, where he has served as a !;t iduate assistant in both middle school education and curriculum evaluation. He is currently coordinator of outreach services for the public siTiools at thr^ Center for African Studies at the University of Florida. Mr. Erb has published in the areas of middle school education, alcohol education, and international education. He is a member of the Association for Supervision and Curriculum Development, the American Educational Research Association, and the African Studies Association. He and Mrs. Erb have tw(^ sons. 106-

PAGE 118

I certify that I have read this study and that in ray opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. (a^-M-ca-^ K (r^^^-^%~ Arthur J. Lewis, CHa/rperson Professor of Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. V . ?)i u'Y-" '' '" ^ / ^ Linda M. Crocker, Cochairperson Associate Professor of Education I certify that 1 have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Gordon D. Lawrence Associate Professor of Education

PAGE 119

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Rene Lemarchand Professor of Political Science This dissertation was submitted to the Graduate Faculty of the Division of Curriculum and Instruction in the College of Education and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1977 Dean, Graduate School

PAGE 120

UNIVERSITY OF FLORIDA 3 1262 08552 9526