THEStS

7' lllllllllllllllllllllllllllllllllll'llllllll‘llllllllllll

31293 01712 7444

This is to certify that the

thesis entitled

Sub r0013 D'FFéRéUCEE W. Cogumve Madly
TEBT £RFORanC£=TH£ :NCR MENTAL
CONTRIBUTVDMQ OF SELF’EFFICan

presented by

DANIELLE A. JEMMFMBS

has been accepted towards fulﬁllment
of the requirements for

M. A. demeinEncxowja
LﬂW

Major rrop oefssor

 

Dates” H ’q 8

MSUisanAﬂiram n'ctveA onu/Eq uundOpport nittylm ititu

 

 

 

 

 

LIBRARY

Michigan State
Unlverslty

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

[ DATE DUE DATE DUE

DATE DUE

 

WAR 2 7n 2002

"i u 2) . Z”-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 alClRC/DabDuopGS—p.“

SUBGROUP DIFFERENCES IN COGNITIVE ABILITY TEST PERFORMANCE: THE
IN CREMENT AL CONTRIBUTIONS OF SELF-EFFICACY

By

Danielle A. Jennings

A MASTER’S THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology
1998

 

ABSTRACT

SUBGROUP DIFFERENCES IN COGNITIVE ABILITY TEST PERFORMANCE: THE
INCREMENTAL CONTRIBUTIONS OF SELF-EFFICACY

By

Danielle A. Jennings

The purpose of this study was to investigate whether self-efﬁcacy for cognitive
ability test performance was related to subgroup status. One’s race, individualocollective
orientation and beliefs about his or her group were investigated to determine potential
inﬂuences on one’s level of self-efﬁcacy. This study also attempted to discover whether
the relationship between self-efﬁcacy and performance was moderated by race. Overall,
this study assessed an individual’s performance on cognitive ability tests, and attempted to
determine the role that self-efﬁcacy and race play in determining cognitive ability test
scores. Moderated regression analyses were performed to determine the results of the
hypotheses generated. Results suggested that self-efﬁcacy did not directly account for
difference in cognitive ability test performance between African Americans and Euro-
Americans. Contrary to predictions, African Americans possessed a higher sense of self-
efficacy for cognitive ability tests than Euro-Americans, but were outperformed by their

Euro-American counterparts.

ACKNOWLEDGMENTS

First giving honor to God, I know that without Him none of this would have been
possible. “Bless the Lord, 0 my soul: and all that is within me, bless his name.” Psalm
103:1.

Next, I would like to thank my chair, Neal Schmitt, PhD. He is an amazing
advisor who is willing to kick you in the pants or hold your hand whichever is necessary
at the time. Without his guidance and direction I know that this thesis would not have
. been accomplished.

Additionally, thank you to the remaining members of my committee, Dan Ilgen,
PhD and Rick DeShon, PhD. Your support was instrumental in the formulation of
thoughts and the resolution of problems. Thank you for challenging me and the way in
which I thought about the research covered in this thesis.

I would also like to thank my cohort-the incoming class of 1995. Thank you for
being there when I needed a shoulder to cry on or an ear to listen. Your support and love
are irreplaceable.

To Christine Streets, Brenda Mitchner and Patricia Green, words cannot express
the depth of my gratitude for all of your love and support throughout this thesis. I thank
God every day for blessing me with incredible friends like you. You will always have a

special place in my heart. I love you!

iii

To my family who has been there for me for the last 25 years of my life, I want to
say “thank you.” Your prayers and commitment and your undying and unwavering love
have helped me through this storm. Thank you for pushing and for holding me up when I
thought I could no longer stand.

Last, but deﬁnitely not least, to my love, Marshaun, “thank you.” You have been
there for me since the beginning and have seen this thesis through to the end. I want to
thank you for your support and undying love. When I did not think that I could ﬁnish or
face another day, you were there with a smile and a hugll Thank you for your loving

arms and for being such an important part of my life!! I LOVE YOU”

iv

TABLE OF CONTENTS

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

LIST OF FIGURES .................................................... viii

INTRODUCTION ....................................................... 1

CHAPTER 1

SUBGROUP DIFFERENCES IN COGNITIVE ABILITY TESTING ............... 7

CHAPTER 2

SELF-EFFICACY and the DEVELOPMENT OF SUBGROUP DIFFERENCES ..... ll
Self-Efficacy ..................................................... 11
Self-Efficacy, Previous Performance, and Collectivism ................... 13
Collectivism and Subgroup Status .................................... 16
Subgroup Differences in Self-Efﬁcacy ................................ 20

CHAPTER 3

HOW SELF-EFFICACY INFLUENCES PERFORMANCE and MODERATING

FACTORS ............................................................. 26

CHAPTER 4

METHOD ............................................................. 36
Subjects ......................................................... 36
Measures ........................................................ 36
Procedure ........................................................ 39

CHAPTER 5

DATA ANALYSES ..................................................... 41

CHAPTER 6

RESULTS ............................................................. 4-4

V

CHAPTER 7

DISCUSSION .......................................................... 67
Conclusions ..................................................... 74
Limitations ...................................................... 76
Future Research .................................................. 78

APPENDICES .........................................................
Appendix A ..................................................... 80
Appendix B ...................................................... 82
Appendix C ...................................................... 83
Appendix D ..................................................... 85
Appendix E ..................................................... 86
Appendix F ...................................................... 87
Appendix G ...................................................... 88
Appendix H ...................................................... 90
Appendix I ...................................................... 91
Appendix J ...................................................... 92
Appendix K ...................................................... 94

REFERENCES ........................................................ 95

vi

 

LIST OF TABLES

TABLE 1 - Scale Reliabilities and Factor Intercorrelations of Individualism-Collectivism

Measure .............................................................. 40

TABLE 2 - Means, Standard Deviations and Scale Intercorrelations for Entire Sample . . .
...................................................................... 45
TABLE 3 - Means, Standard Deviations for African Americans and Euro-Americans . . .

...................................................................... 46
TABLE 4 - Scale Intercorrelations by Race .................................. 47
TABLE 5 - Regression Analyses: Math Knowledge Test Self-Efﬁcacy .............. 50
TABLE 6 - Regression Analyses: Verbal Analogies Test Self-Efficacy ............. 52
TABLE 7 - Regression Analyses: Math Knowledge Test Self-Efﬁcacy ............. 55
TABLE 8 - Regression Analyses: Verbal Analogies Test Self-Efﬁcacy ............. 60
TABLE 9 - Regression Analyses: Math Knowledge Test Performance ............. 65

TABLE 10 - Regression Analyses: Verbal Analogies Test Performance ............ 66

vii

LIST OF FIGURES

FIGURE 1 -Interaction between Race and Individualism-Collectivism ............. 23

FIGURE 2 - Interaction between Race, Individualism-Collectivism and Awareness and
Beliefs about African Americans

FIGURE 2a - Interaction between Individualism-Collectivism and Awareness and

Beliefs about African Americans held by African Americans ............... 24
FIGURE 2b - Interaction between Individualism-Collectivism and Awareness and
Beliefs about African Americans held by Euro-Americans ................. 25
FIGURE 3 - Interaction between Race and Level of Self-Efﬁcacy ................. 33

FIGURE 4 - Model of Theory ............................................. 34

viii

INTRODUCTION

Today’s employers have many pre-employment testing measures to assess the
potential of job applicants. The type of measures utilized range from cognitive ability
tests to assessment centers to structured interviews. Although these measures are used to
select the most qualiﬁed applicants for the position, some of the most popular and valid
measures utilized elicit a disparity in performance scores between subgroups, such as
African Americans and Euro-Americans. For example, the cognitive ability measure
administered in a paper-and-pencil fashion has been shown to be one of the most valid of
selection measures and one of the most predictive; however, it also reﬂects the greatest
degree of subgroup difference in performance scores with Euro-Americans scoring higher
than African Americans. This disparity in performance scores between racial groups is
reﬂected in the disproportionate number of African Americans relative to Euro-
Americans selected for employment. This disparity in hiring ratios has been called
adverse impact.

Adverse impact has resulted in legal action, such as charges of discrimination and
reverse discrimination, pursued by those negatively affected by the selection process,
testing in particular. The courts and various guidelines have also indicated that adverse
impact creates a burden for the employer to show that the selection test is job-related and

not biased. In addition, Executive Order 11246 (1967) required federal contractors to

 

2

ensure that individuals were employed and "treated while employed, without regard to
their race, color, religion, sex or national origin" (Turner, Pratkanis, & Hardaway, 1991,
p.798). Afﬁrmative Action plans, which set goals and deﬁne procedures designed to
increase opportunities for minority groups, have become commonplace in the
employment arena. As a result, companies have attempted to increase the representation
of minority groups, such as, African Americans and women, in their applicant pool. This
attempt, however, is hampered by the selection or testing process in which subgroup
differences in performance on the selection tests imply lower ability of one group in
comparison to another group. Consequently, the hiring ratio often is greater for the Euro-
American majority than for minorities.

Title VII of the Civil Rights Act of 1964 prohibited discrimination in employment
due to an applicant’s or employee’s race, religion, sex, or national origin. However, if a
selection measure was not intentionally developed for a discriminatory purpose, and if the
employer could show that the test was needed, it could be used regardless of the potential
exclusionary impact on minorities and women (Seymour, 1988). As a result of the Civil
Rights Acts and other legal cases challenging the selection process utilized by employers.
employers began "to play the numbers game." They were made aware of the fact that as
long as the hiring ratio was acceptable, the company would not be a target for legal
action. In the case of legal action, the courts would ﬁrst focus on the number of
applicants hired from each respective group, and if a disparity existed, then the hiring
process would be further investigated. However, if a disparity did not exist, then no

further investigation was warranted.

3

In the ﬁrst Supreme Court case concerned with discrimination in the workplace--
Griggs v. Duke Power Company, the Supreme Court ruled that employers were not only
required to state that selection measures were job-related, but that they had to prove them
to be (Seymour, 1988). According to Bersoff (1981), the case of Griggs v Duke Power
Company "introduced the concept of job-relatedness’ into the law of employment testing"
(p. 1050). Consequently, employers have been required to prove the job-relatedness of
selection tests. Job-related tests as deﬁned by the Equal Employment Opportunity
Commission (EEOC, 1966) Guidelines on Employment Testing are measures that fairly
assess the requisite knowledge or skills required to perform a job (Bersoff, 1981

For decades, it has been believed that African Americans are inferior to Euro-
Americans in cognitive ability. In 1942, about 60% of Euro-Americans believed this to
be true, and in 1964, that number dropped to about 25% (Bobo, 1988). Still to date,
society continuously reminds African Americans that they are inferior in cognitive ability
to Euro-Americans. Repetitively, research has shown that Euro-Americans and African
Americans differ in regards to paper-and-pencil cognitive ability test scores by
approximately one standard deviation (Jensen, 1980). Summaries of such ﬁndings are
included in The Bell Qurve by R. J. Herrnstein and C. Murray and are commonly reported
on standardized cognitive ability tests such as the Scholastic Aptitude Test, the American
College Test or the Graduate Record Examination (Jensen, 1984).

The consistent stereotype of inferiority reinforced by these reports may impact the
way an African American views himself/herself. When developing one’s identity, an
African American must engage in a dual task. The African American attempts to

assemble a positive self image while disavowing the negative stereotypes attributed to

4
African Americans (Oyserman, Gant, & Ager, 1995). My belief is that this negative

stereotype often becomes part of the African American self-identity and that it serves to
identify African Americans in such a way that their test performance is affected.

Another way that society may negatively inﬂuence the sense of self-efﬁcacy held
by African Americans is the unintended effect of the implementation of Afﬁrmative
Action and preferential selection programs that elicit the belief that recipients are inferior.
With the introduction of Afﬁrmative Action programs and preferential selection,
beneﬁciaries of such programs may hold beliefs that they do not possess the requirements
necessary for the position-discounting their knowledge, skills, and abilities (Turner, et al,
1991). Likewise, the targeted populations are viewed by others as incompetent and
observers believe that inferior applicants have been selected (Blaine, Crocker, & Major,
1995). The recipients and fellow employees are aware that characteristics other than
those relevant to the job (e.g., gender, race, or age) played an important role in the
selection of the applicant (Pettigrew & Martin, 1987). All of this information may impact
an individual’s belief about his or her ability to perform. Consequently, if the group in
which he or she is a member is consistently regarded as inferior and tests reﬂect such, an
applicant may begin to internalize this information which, in turn, may impact his or her
sense of self-efﬁcacy. In addition, if programs are implemented that further suggest
inferiority, then the applicant may not believe that he/she has what it takes to successfully
complete a given task or perform well on the job.

Whatever the origin, this thesis investigates the proposition that differences in
self-efﬁcacy between subgroups (e.g., African Americans and Euro-Americans) account

for some of the variance in scores between subgroups on cognitive ability tests. In

5

Graham’s (1994) opinion, many African American students do not perform well in school
and do not pursue future goals, "not because they lack basic intellectual competencies or
speciﬁc learning skills, but because they have low expectations, feel hopeless, deny the
importance of individual effort, or give up in the face of failure" (p.55). As a result, an
African American’s sense of competence is diminished due to the belief that, for him or
her, success is unlikely. (Oyserman, et al., 1995). Findings from a study conducted by
Eden and Kinnar (1991) provided evidence that those who lacked a belief in their ability
to perform well were less likely to even attempt the necessary behavior and thus, fulﬁll a
prophecy of incompetence and lack of ability. According to Bandura (1982), people give
up when attempting to accomplish something or simply do not attempt to accomplish a
given task because they believe that their efforts will go unnoticed or unrecognized.
Therefore, an African American’s lack of belief about his or her ability to perform could
explain some of the difference in performance scores on cognitive ability tests between
African Americans and Euro-Americans.

If self-efﬁcacy is found to be signiﬁcant in regards to the amount of adverse
impact displayed in selection measures, this ﬁnding opens the door of opportunity to
increase the self-efﬁcacy of those that are lacking, in hopes of enhancing minority group
performance on such tests. Self-efﬁcacy is ﬂuid and amenable (Gist & Mitchell, 1992);
whereas, an individual’s general mental ability is rather static and less likely to be
increased. With this understanding, an individual may increase his or her performance
score on a selection measure, cognitive ability test in particular, by increasing his or her
belief that he or she possesses what is necessary to successfully execute the task before

him or her. According to the Galatea effect, increasing one’s expectations about his or her

 

6

performance, results in an increase in performance (Eden & Kinnar, 1991). The Galatea
effect is a type of self-fulﬁlling prophecy. If an individual’s belief about his or her ability
to perform is enhanced resulting in increased expectations of success, then better
performances should be observed.

The purpose of this study is to investigate another potential explanation to adverse
impact in selection testing besides that of subgroup differences in cognitive ability. In
particular, this study will attempt to determine whether self-efﬁcacy is related to
subgroup status and to identify and validate the relationship between self-efﬁcacy and
performance on cognitive ability tests. In addition, it will attempt to discover whether the
relationship between self-efﬁcacy and performance is moderated by race. This study will
also assess an individual’s performance on cognitive ability tests, and attempt to
determine the role that self-efﬁcacy and race may play in determining test scores.

First, we will examine previous research investigating subgroup differences in
performance on cognitive ability tests. Next, we will focus on self-efﬁcacy, which is
proposed as one of the primary reasons for differences in cognitive ability test scores
between African Americans and Euro-Americans. Following, if the predicted difference
between African Americans is found, we will suggest potential reasons for the disparity in
self-efﬁcacy between African Americans and Euro-Americans (e.g., collectivism-
individualism and awareness of disparity in performance scores). Then, we will review
literature addressing the relationship between self-efﬁcacy and performance and what
factors may inﬂuence such a relationship. Hypotheses will be stated at the end of these
sections and an overall model of the hypothesized relationships will be presented at the

conclusion of the literature review (See Figure 4).

Chapter 1

SUBGROUP DIFFERENCES IN COGNITIVE ABILITY TESTING

As previously stated, research has repeatedly found a one-standard deviation
difference between African Americans and Euro-Americans on cognitive ability tests,
with that of Euro-Americans being higher (Hunter & Hunter, 1984; Schmidt, Greenthal,
Hunter, Bemer, & Seaton, 1977). Several theories, which encompass the nature vs
nurture controversy, have suggested reasons for the difference in cognitive ability: the
genetic or heredity theory, the environmentalist theory, the culturalist perspective (a
speciﬁc focus of the environmentalist theory), and the interactionist theory (genetic x
environment). Those favoring the heredity theory (internal) focus on the cognitive ability
test score as a true measure of one’s intellectual capacity, regardless of demographic
issues, such as quality of education or socioeconomic status. Starting with Jensen’s
(1969) monograph, mean cores on intelligence tests were found to differ between African
American and EurooAmerican groups (Loehlin, Lindzey, & Spuhler, 1975). According to
Loehlin, et al (1975), research has speciﬁcally found that differences in intelligence
scores within the Euro-American population can be attributed to genetic differences. In a

table constructed by Loehlin, et al (1975) based on studies conducted between 191 1

 

8
1962, ﬁgures suggest that seventy-ﬁve percent (range of 60 to 80%) of the variance in

intelligence test scores can be attributed to genetic factors.

Researchers advocating the attribution of subgroup differences in cognitive ability
to genetic factors do recognize the potential inﬂuence of environmental factors, such as
Jensen (1969b) who suggested that the environment was a "threshold variable" (Bemal,
1984). The environment may limit the potential to which one may aspire. According to
Loehlin, et al (1975), performance on an intelligence test can be a function of many
factors, not just cognitive ability. The environmental perspective explains African
American-Euro—American differences by highlighting external factors, such as oppression
or deprivation. Further, in regards to racial differences in cognitive ability, Jensen (1969)
acknowledges that the environment, or its interaction with genetic factors, may inﬂuence
the differences found in test scores between African Americans and Euro-Americans
(Loehlin, et al, 1975). When examining studies conducted between 1911 and 1962,
Loehlin, et a1 (1975) found that twenty-ﬁve percent of the variance in intelligence test
scores could be attributed to environmental factors. Additionally, Grubb and Ollendick
(1986) reported research that has shown gender, socioeconomic status, father’s occupation
and one’s residence decreases the difference in intelligence test scores between African
Americans and Euro-Americans to about one-half standard deviation. In particular,
McGurk (1978) found that an improvement in the socioeconomic status of African
Americans did signiﬁcantly increase the intelligence scores of African Americans
(Eysenck, 1984). Likewise, Crane (1994) cites Humphreys (1985) who states that racial
differences in intelligence tests were substantially accounted for by socioeconomic status.

Research has also looked at what else is being tapped by cognitive ability

9

measures; for example, motivation and knowledge of the majority culture. The culturalist
perspective has argued that items developed and the manner in which an intelligence test
is administered reﬂect a bias in favor of the dominant culture (Euro-Americans), resulting
in subgroup differences in performance (Helms, 1992). Perhaps a deﬁnition of
intelligence that could be embraced by the advocates of this perspective is that proposed
by Humphreys (1971): "the entire repertoire of acquired skills, knowledge, learning sets
and generalization tendencies considered intellectual in nature that are available at any
one period of time" (Grubb & Ollendick, 1986, p.399). Acquired information may be
different for majority and minority cultures. Members of the Euro-American culture
gather information about the Euro-American culture; whereas, members of the African
American community must gain knowledge of not only their own culture, but also that of
the Euro-American community. Consequently, differences between groups in
intelligence test scores may be observed when the content of the intelligence test reﬂects
only that of the Euro-American culture. As suggested by the culturalist perspective,
adverse impact is inherent in intelligence tests which suggests a defect in the test
construction and administration (Wollack, 1994).

According to Loehlin, et al. (1975), any psychology test of cognitive ability can
not be "culture-free" or "culture-fair" when administered to other cultures because
development of the measure will reﬂect the cultural biases of the developer and the
environment. Furthermore, in an attempt to demonstrate cultural bias in cognitive ability
tests, Williams (1972) developed the Black Intelligence Test of Cultural Heterogeneity
(BTTCH Test; Harrington, 1984). This intelligence test was purposely biased in favor of

African Americans since the vocabulary test was based on words and phrases distinctive

10

of the African American culture. Consequently, African Americans, on the average,
scored higher than Euro-Americans on this culturally-biased test. However, the issue of
cultural bias in intelligence tests, in which the content of the tests is thought to reﬂect
information inherent primarily in the majority population, has been challenged by the fact
that the scores of Asian test-takers generally equal or exceed that of Euro-Americans
(Jensen, 1984). Additionally, in a study conducted by Quay (1971), the Stanford-Binet
intelligence test was translated into "Black English" and administered to African
American school children. When compared, performance on the translated version of the
test was not signiﬁcantly different than performance on the standard version of the
Stanford-Binet test (Loehlin, et al., 1975). As evident, research concerning cultural bias
in cognitive ability test construction has been mixed.

Finally, the interactionist theory suggests that heredity and the environment are
both causes for differences in cognitive ability test scores between groups (Eysenck,
1984). According to the interactionist theory, individual differences are a product of
heredity and their surroundings.

All of these theories, the biological, the environmental, and the interactionist
theory, suggest that African Americans have a deﬁcit in intellectual capacity in
comparison to others, regardless of how the deﬁcit was created or developed, whether
biological or environmental or a combination of the two. An alternative to the above
theories or perspectives, the underlying assumption of this study contends that differences
in cognitive ability test scores can be explained partly by an individual ’3 belief about his

or her ability to perform well on a cognitive ability testuhis or her sense of self-efﬁcacy.

Chapter 2

SELF -EFFICACY AND THE DEVELOPMENT OF SUBGROUP DIFFERENCES

Self-Efﬁcacy,
Self-efﬁcacious individuals believe that they possess the necessary knowledge,

 

skill, and ability to successfully complete a given task (Gist & Mitchell, 1992). Self-
efﬁcacy is an evaluative belief about oneself which has motivational implications for
behaviors and cognitions (Gist & Mitchell, 1992). Self-efﬁcacy differs from self-esteem
in that self-efﬁccy focuses on the potential to perform a speciﬁc behavior; whereas, self-
esteem deals with one’s general assessment or evaluation of the self. Self-efﬁcacy has
been suggested as a subcomponent of self-esteem; however, research has also identiﬁed
that the two constructs are not the same. Thus, although a relationship may exist between
self-esteem and self-efﬁcacy, the focus of this investigation is with the task-speciﬁc sense
of self-efﬁcacy.

When determining level of self-efﬁcacy, pe0ple assess their ability to execute a
speciﬁc behavior. Individuals consider the difﬁculties and demands of the given situation
and their own skills and abilities when evaluating their level of self-efﬁcacy (Cervone,

1989). Ultimately, one's sense of self-efﬁcacy inﬂuences thought and behavior, i.e.,

12
effort expended, persistence, and goal setting (Bandura & Adams, 1977; Gist & Mitchell,

1992).

Assessment of one’s ability and potential to perform involves the individual ’5
perceived performance capability, the task to be completed and others in the environment.
There are four categories of experiences suggested as being used in the development of
self-efﬁcacy: modeling (vicarious learning), verbal persuasion, autonomic arousal and
prior performance (Bandura, 1982; Gist & Mitchell, 1992). First, according to Gist and
Mitchell (1992) and Bandura (1993), people partly judge what they are capable of doing
or their ability by watching and comparing themselves to models-others like themselves,
in some manner-performing the intended behavior or task and who also teach
appropriate strategies to execute. If the models successfully accomplish the given task,
then the observer’s sense of self-efﬁcacy may be raised. He or she should believe that he
or she can accomplish the task as well as the model did. On the other hand, if the models
fail, then the observer’s sense of self-efﬁcacy may be lowered and the observer will
expect to fail in the attempt to perform the task (Bandura, 1982). Second, positive
feedback concerning one’s ability or potential can raise one’s sense of efﬁcacy; however,
failure may negate such a reaction. Verbal persuasion can enhance one’s sense of self-
efﬁcacy, if it is realistic in the potential accomplishment (Bandura, 1982). Third,
physiological states can inﬂuence self-efﬁcacy and hamper performance if the signs
indicate a "vulnerability to dysfunction" (Bandura, 1982). "Gut feelings" can be
indicators of one’s potential to perform. Finally, prior performance is the most inﬂuential
developmental experience of one’s sense of self-efﬁcacy because it is based on an actual

experience of the intended behavior (Bandura, 1982).

13

A relationship between performance and self-efﬁcacy can be derived from the
notion that prior performance is one of the determinants or ways in which self-efﬁcacy is
developed. Schunk (1991) suggested that the most reliable factor when judging the level
of self-efﬁcacy was one’s own previous performances. How an individual performed on a
similar, prior task will inﬂuence his or her belief about his or her ability to successfully
perform the focal task. Successful completion raises one’s sense of self-efﬁcacy and
failures diminish it. Furthermore, although Locke, Frederick, Lee, & Bobko (1984) found
self-efﬁcacy to be more highly correlated with past performance than future performance,
self-efﬁcacy was signiﬁcantly related to future performance when past performance was
partialled out. The relationship between self-efficacy and previous performance suggests
that subgroup differences in self-efﬁcacy on cognitive ability tests should be substantiated
given the history of African American experiences alluded to above. Consequently,
because an individual ’3 belief about what he or she can do affects most aspects of his or
her life, self-efﬁcacy judgments should remain predictive of future performance when

prior performance and ability are held constant (Pajares & Kranzler, 1995). Further, the
partial correlation between self-efﬁcacy and future performance suggests one’s belief
about what he or she is capable of doing predicts later performance independent of their
actual previous performance.
Self-Efﬁcacy, Previous PerfgrmanCe, and Collectivism.

One potential explanation of subgroup differences in self-efﬁcacy may be the fact
that, according to Bandura (1986), "self-efﬁcacy is, in part, socially constructed and that
such construction may differ as a function of national culture" (Earley, 1994, p.90).

Furthermore, Bandura ( 1986) suggested that one’s belief about his or her ability to

l4

perform a given task is inﬂuenced by "maturation and socialization experiences" (Earley,
1994, p.90). Maturation and socialization experiences may differ depending on one ’5
culture and individual- or collective-orientation. Various research studies have led to the
notion that those individuals with a collective-orientation may differ in the development
of their self-efﬁcacy from those individuals with an individualistic orientation. An
individual or collective-orientation represents the feelings or behaviors generated by the
degree of self-interest or concern for others that may be exhibited in different magnitudes
and in different situations depending on the target persons; the identiﬁcation of oneself
with the group may also depend on the type of group in question (Hui, 1988; Hui &
Triandis, 1986). More speciﬁcally, individualists "deﬁne the self independently of
groups . . . believe that they can stand or fall on their own and survive on their own" (Hui,
1988, p.18). On the other hand, collectivists "see the self as an aspect of a group, and
value interdependence" (Hui,l988, p.18). Triandis (1989) stated that Tajifel’s (1978)
social identity is included in one’s collective self. One’s social identity refers to that part
of one’s self image which develops from his (or her) awareness of his (her) membership
in a social group (or groups), in addition to the values and emotional signiﬁcance attached
to that membership.

Due to differing orientations, collectivists and individualists sample their
environments differently (T riandis, 1989). Research has found that individualists refer to
personal information, such as their own performance, when judging their level of self-
efﬁcacy; whereas, collectivists refer to in-group information (e.g., the in-group’s
performance) when determining their ability to perform well on a given task (T riandis,

1989; Erez & Barley, 1993; Earley, 1994). More speciﬁcally, results of a study conducted

15

by Barley ( 1994) showed that "a collectivist’s self-efﬁcacy is based on information that he
or she gets about a (work) group, whereas, an individualist’s self-efﬁcacy comes from
self-referenced cues" (p.115)

In general, as previously stated, people who embrace a collective-orientation
deﬁne themselves in regards to some reference group; meanwhile individualists deﬁne
the self independent of salient reference groups. Therefore, prior performances or
modeling by members of the referent group may impact the level of self-efﬁcacy for those
who are collectively-oriented, but may not affect the level of self-efﬁcacy for those who
are individually-oriented. When suggesting differences in level of self-efﬁcacy between
African Americans and Euro-Americans, one may focus on the type of developmental
experiences of modeling and prior performance. For instance, in regards to cognitive
ability tests, the prior performances of African Americans have reportedly been inferior in
comparison to Euro-Americans. Awareness of this information in addition to one’s
collective- or individualistic-orientation may differently inﬂuence the level of self-
efﬁcacy for African Americans and Euro-Americans. African Americans who are aware
of the difference in performance scores on cognitive ability tests between African
Americans and Euro-Americans, with Euro-Americans outperforming African
Americans, may identify with other African Americans who have reportedly done poorly
on cognitive ability tests in comparison to Euro-Americans and this identiﬁcation with
the African American population may dampen their sense of self-efﬁcacy. On the other
hand, Euro-Americans who are collectively-oriented and are aware of the superior
performance of fellow Euro-Americans on cognitive ability tests in comparison to

African Americans may have their level of self-efﬁcacy enhanced due to this knowledge.

 

l6

Contrary to the impact that the referent group has on the development of a collectivist’s
level of self-efﬁcacy, an individualist’s level of self-efﬁcacy will not be affected by the
superior or inferior prior performance of a speciﬁc reference group because he or she
views the self independent of his or her race, age group, or class. While collectivists and
individualists may develop their sense of self-efﬁcacy in different ways, these two
orientations are also differently embraced by the African American community and the
Euro-American culture.

1 tiv's Sub u Statu

In general, individuals develop their impression of themselves through social

relationships and social group memberships. Social comparison theory proposes that
people compare themselves with others in order to understand and evaluate their own
behavior. In particular, individuals engage in self-categorization by which a person
identiﬁes with or intemalizes the social category of which he or she is a member. For
example, Gecas and Schwalbe (1983) state that individuals see themselves in their
reference groups and signiﬁcant others. A referent group is a particular group of people,
similar to ourselves in certain key ways (i.e., gender, race, or age), used as a standard in
social comparisons. A person deﬁnes, describes and evaluates the self in regards to this
social category or label (Hogg & Turner, 1987). As a result, perception of one’s self is
biased and is more closely aligned with the stereotypic characteristics (i.e., "African
Americans are not as intelligent as Euro-Americans") and norms of the group in which
he or she belongs. Research has shown that "group members and nonmembers expect

individuals to adhere to corresponding group identity and norms" (Arroyo, & Zigler,

 

17

1995, p.903). In particular, as previously stated, collectivists deﬁne themselves in
relation to the group of which they are members.

According to Gaines (1994), the African American community embraced the
cultural norm of collectivism--a concern for the entire group; whereas, the Euro-
American community was oriented toward the concern for oneselfuindividualism.
Oyserrnan, et a1 (1995) suggested that African Americans possess a higher sense of
collectivism and ethnic identity than Euro-Americans. Additionally, Phinney (1996)
suggested that Western industrialized countries stress the importance of the individual
over the group and view the self as "independent, autonomous, and self-contained"
(Phinney, 1996). On the other hand, a number of non-Western countries, such as Africa,
stress the priority of the group over the individual and view the self as "interdependent
and connected." In particular, members of the Asian, African, and Latin American
communities are assumed to reﬂect an emphasis on dependence on one another and the
importance of the group as opposed to the idea of independence and survival of the ﬁttest
(Phinney, 1996). Furthermore, according to Waters (1990), most Euro-Americans do not
readily identify themselves according to their ethnic background due to the lack of
salience of one’s culture (Phinney, 1996) as opposed to how it is for members of other
ethnic groups where ethnicity is obvious, simply by the color of one’s skin, for example
(Phinney, 1996). According to research the more salient one’s ethnic identity, the greater
its inﬂuence on the development of one’s identity and how he or she views himself or
herself (Phinney, 1996). Also, Gaines and Reed (1995) determined that there are
psychological consequences of "belonging to a group that has been oppressed or exploited

throughout an historical period" (Phinney, 1996, p. 97). One of the consequences may be

18

the emphasis on the importance of the group over the importance of the individual--
showing a preference for collective values (Phinney, 1996). For African Americans, this
oppression came in the form of slavery, segregation and discrimination. In sum, African
Americans are thought to be members of a collective community, their ethnic identity is
salient and have been oppressed historically fostering a collective sense of self-efﬁcacy.
On the other hand, Euro-Americans are predisposed to individualistic values, in addition
to the absence of salient features of ethnicity and the lack of historical oppression and
exploitation. Therefore, if African Americans are more collectively-oriented, then the
prior performances of African Americans as a whole and the observation of fellow
African Americans may inﬂuence the self-efﬁcacy of the individual African American
applicant due to his or her identiﬁcation with the members of his race.

Research by Fordham and Ogbu (1986) revealed that poor academic achievement
by African Americans represented their need to maintain a sense of unity with their own
cultural community (Arroyo & Zigler, 1995). Findings suggested that African American
students who strive for academic achievement engage in behaviors and possess attitudes
and values reﬂective of the majority Euro-American community as opposed to their own
cultural community. These behaviors and experiences of academically achieving African
American students have been termed racelessness (Fordham & Ogbu, 1986). In testing
the concept of racelessness, Arroyo and Zigler (1995) found that African American
students possessing a sense of racelessness were more likely to accept derogatory
stereotypes about their group, such as "In general, Blacks are to blame for their negative
image among Whites" (p. 905). For African Americans who believed that public

perception of their ethnic group was negative, they too "lowered their evaluations of and

l9
identiﬁcation with their racial group" (Arroyo & Zigler, 1995, p.91 1). Additionally, a

negative relationship was also observed between stereotyped beliefs about academic
performance of African Americans and their support of commonly held stereotypes and
an African American’s identiﬁcation with their ethnic group (Arroyo & Zigler, 1995).
Based on study results, the greater the agreement with negative stereotypes of African
Americans by African Americans, the less they identiﬁed with the African American

community and the less important one’s group was to the deﬁnition of oneself.

 

Consequently, the stereotypes, such as "African Americans are cognitively inferior to
Euro—Americans," may be related to an African American’s attempt to deﬁne the self
independent of the African American community. Conversely, African Americans who
responded that one’s social group membership was important to how one viewed and
deﬁned the self also expressed a higher level of conﬁdence in their strengths and abilities-
-self-efﬁcacy (Arroyo & Zigler, 1995).

According to Steele (1988), African Americans who indicate ethnicity on a
standardized test of cognitive ability were reminded of their ethnic identity (Oyserman, et
al, 1995). This identiﬁcation heightened "connectedness" and the obstacles faced by
African Americans, such as public perception, became more salient to the individual
African American, thereby discouraging the exertion of effort. Cervone (1989) found that
individuals who concentrated on potential obstacles and difﬁculties of a cognitive task
reported lower beliefs of performance capabilities than those who focused on aspects of
themselves or the task that could improve performance or those who did not consider
either. Consequently, subgroup differences in self-efﬁcacy may be a function of one’s

cultural identity towards collectivism or individualism and how these orientations, in

20

conjunction with information about the group, impact one’s development of his or her
belief about his or her ability to perform successfully. Furthermore, this identiﬁcation
with the group by African Americans may hinder the development of a high level of self-
efﬁcacy for performance on cognitive ability tests due to the awareness of differences in
cognitive ability test scores and poorer performances.

Subggoup Differences in Self-Efﬁcacy

To date, limited research has investigated subgroup differences in self-efﬁcacy;
however, researchers have found African American-Euro-American differences in self-
efﬁcacy. For instance, Oyserman, et al., (1995) stated that the level of general self-
efﬁcacy was lower among African Americans than Euro-Americans. In particular,
Pajares and Kranzler ( 1995) found a signiﬁcant relationship between math self-efﬁcacy
and race. More speciﬁcally, a signiﬁcant mean difference in math self-efﬁcacy between
African Americans and Euro-Americans was observed with African Americans reporting
a lower sense of math self-efﬁcacy.

Due to the limited existing research suggesting that African Americans are more
collective and Euro-Americans are more individualistic and the potential inﬂuence of
negative stereotypical beliefs about African Americans’ performance on cognitive ability
tests, the resulting level of self-efﬁcacy between African Americans and Euro-Americans
may show a greater belief by an Euro-American for his or her ability to perform well on
cognitive ability tests than an African American. Using this summary of research on self-
efﬁcacy, individualism-collectivism, and the experiences of African Americans and Euro-

Americans, I have formulated the following hypotheses:

21
Hypothesis 1: Differences in self-efﬁcacy on cognitive ability tests will exist

between African Americans and Euro-Americans, with that of Euro-Americans being
higher.

Hypothesis 2: The relationship between race and self-efﬁcacy is moderated by
one’s individual-collective orientation (See Figure 1), such that African Americans and
Euro-Americans will not differ signiﬁcantly if they have an individualistic-orientation,
but will differ signiﬁcantly as orientation towards others or the group increases.

thesis a: African Americans who are collectively oriented and are aware of
the difference in performance scores on cognitive ability tests between African Americans
and Euro-Americans will possess a lower sense of self-efﬁcacy than those African
Americans who are collectively oriented and are not aware of the differences in
performance scores. On the other hand, African Americans who are individually-oriented
and are aware of the difference in performance scores between African Americans and
Euro-Americans will not possess a signiﬁcantly different level of self-efﬁcacy than those
African Americans who are individually-oriented and are unaware of the difference in
performance scores between African Americans and Euro-Americans (See Figure 2a).

Hymthesis 3b: Euro-Americans who are collectively-oriented and are aware of
the difference in performance scores on cognitive ability tests between African Americans
and Euro-Americans will possess a higher sense of self-efﬁcacy than those Euro-
Americans who are collectively-oriented and unaware of the difference in performance
scores. On the other hand, Euro-Americans who are individually-oriented and are
unaware of the difference in performance scores between African Americans and Euro-

Americans will not possess a signiﬁcantly different level of self-efﬁcacy than those Euro-

22

Americans who are individually-oriented and are aware of the difference in performance

scores (See Figure 2b).

23

Em_>:oc=oU.Em=§E>€£ new 35% 5253 55825 - ﬂ Semi

 

 

Eﬂéoczev 8326365
cactus}.
:85
acme—hm
.hom
going
.eSm

 

24

88:32 SSE< 3 Bo: mgotcE/w 5054‘ Sena ﬂow—om a.
305322 a. .EmguoB—ouéﬁ3:232: .coam 5250; 530225 - mm Semi

 

.8302 oﬁﬁmoz

 

caeeoeeo

zone—mum
Axum

 

8338365

 

25

ESEETeSm 3 Eu: 285:2 5332 Bone whom—om a.
328522 6% .Em_>_.oo=eU-Em=m=E>€E 603— c838: 5383.: .. em 2sz

35:02 03832

 

oumsasEzeE

 

2.23388 sesam—
cam

 

Chapter 3

HOW SELF-EFFICACY INFLUENCES PERFORMANCE
AND MODERATING FACTORS

Self-E rcac and Performance

In order to successfully execute behaviors to accomplish a given task, an
individual must believe that they possess and can utilize the necessary skills (Eyring,
Steele Johnson, & Francis, 1993). According to Bandura (1983), there is a difference
between possessing the required knowledge and skills to perform a given task and
actually being able to utilize those skills. Furthermore, Bandura (1982) suggested that "if
self-efﬁcacy is lacking, people tend to behave ineffectually even though they know what
to do." Findings from a study conducted by Bouffard-Bouchard (1989) showed, despite
the fact that the individual possessed the necessary skills to perform the task, "perceived
self-efﬁcacy operates partially independently of those skills" (p.361). In studies
conducted by Collins (1982; 1984), the relationship between self-efﬁcacy and ability and
its impact on performance were investigated (Bandura, 1982; Bouffard-Bouchard, 1989).
The ﬁndings (Collins, 1982) demonstrated that self-efﬁcacy predicts motivation and

achievement across levels of ability (Bandura, 1982). Regardless of ability level, those

26

27

students with a high sense of self-efﬁcacy performed better than those students with a low
sense of self-efﬁcacy (Bandura, 1982). Additionally, Collins (1984) found that although
mathematical ability contributed to performance at each level, those possessing a high
sense of self-efﬁcacy outperformed those with a low sense of self-efﬁcacy (Bouffard-
Bouchard, 1989). With cognitive ability being held constant, self-efﬁcacy had a direct
effect on performance. Consequently, in order to achieve, one must not only possess the
knowledge, skills and ability necessary to perform, but also believe that they have what it
takes to perform the task; therefore, applicants possessing the same KSAs may vary in
performance due to their individual beliefs about their capacity to perform.

Many researchers concur that performance is a function of ability and motivation
(Ruvolo & Markus, 1992; Chan, Schmitt, Deshon, Clause, & Delbridge, 1996). Further,
Eden and Kinnar (1991) state that self-efﬁcacy is a key determinant of work motivation.
It contributes to motivation in the goals that people set, the effort expended and their
persistence despite obstacles. According to Bandura (1988b), "self-efficacy mediates the
translation of knowledge and abilities into skilled performance" (Gist & Mitchell, 1992,
p.202). Self-efﬁcacy indirectly effects performance by inﬂuencing initiation, the amount
of effort expended, goals set, and persistence (Bandura & Adams, 1977; Bandura, 1982;
Locke, et al, 1984; Gist & Mitchell, 1992; Ruvolo & Markus, 1992). If an individual
believes that he or she possesses the necessary skills and ability to perform, then he or she
will initiate the behavior, set higher goals, exert energy to accomplish the task, and
continue in the behavior despite obstacles. Bouffard-Bouchard (1989) and Cervone
(1989) found that perceived self-efﬁcacy was signiﬁcantly related to task persistence. In

particular, Multon, Brown, and Lent (1991) found one’s sense of self-efﬁcacy accounted

28

for approximately 12% of the variance in academic persistence. This ﬁnding supports
Bandura’s ( 1982) contention that one’s belief in his or her ability to perform contains a
motivational component that inﬂuences initiation and persistence in order to achieve a
desired outcome.

Researchers have also found that self-efﬁcacy demonstrates a direct effect on
performance. Regardless of the type of developmental experience, whether vicarious
learning, prior experience or extinction of anxiety, one’s belief in his or her ability to
perform predicts his or her eventual performance (Bandura & Adams, 1977).
Furthermore, according to Bandura (1982), performance varies as a function of one’s
belief in his or her ability to execute the necessary behavior. As one’s sense of self-
efﬁcacy increases, his or her performance increases as well. Likewise, at the group level,
those groups with the higher level of self-efﬁcacy outperformed those with a lower level
of self-efﬁcacy (Bandura, 1982). In a study conducted by Multon, et al (1991), self-
efﬁcacy beliefs accounted for approximately 14% of the variance in student’s academic
performance. Zimmerman, Bandura, and Martinez-Pons ( 1992) found that perceived
self-efﬁcacy was signiﬁcantly predictive of ﬁnal course grade. Further, Pintrich and
DeGroot (1990) found that self-efﬁcacy was signiﬁcantly related to student performance
on seatwork, exams/quizzes, and essays/reports. The more a student believed that he or
she had what it took to successfully complete a task, the better the student performed.
Research by Weinberg, Gould, and Jackson ( 1979) further supported prior research that
self-efﬁcacy beliefs directly inﬂuence performance. Study results showed that high self-
efﬁcacy individuals outperformed those with a low sense of self-efﬁcacy. Likewise,

Tuckman and Sexton (1990) found a signiﬁcant relationship between what peOple

29

believed they could do and performance. High believers outperformed their personal
expectations; whereas, low believers did not perform as well as they had anticipated.

Cooper and Robinson (1991) and Pajares and Kranzler (1995) found, more
speciﬁcally, a correlation between math self-efﬁcacy and math performance. In addition,
Pajares and Kranzler (1995) found a signiﬁcant direct effect of math self-efﬁcacy on math
performance. As a result, Pajares and Kranzler (1995) contended that one’s sense of self-
efﬁcacy in regards to math capability strongly predicted his or her later academic
performance in that area. In combination, the ﬁndings regarding the relationship between
self-efﬁcacy and performance and the fact that self-efﬁcacy beliefs account for variance
in performance when KSAs are held constant can partially explain varying performances
between individuals who possess similar KSAs. In particular, these ﬁndings may
potentially explain some of the variance in performance scores on various selection
measures between African American and Euro-Americans. If there is a signiﬁcant
difference in level of self-efﬁcacy between African Americans and Euro-Americans,
performance differences may result.

For some positions or jobs, cognitive ability has been identiﬁed as necessary to
execute the requisite behaviors. Consequently, some of the selection measures utilized
assess cognitive ability. The greatest difference in performance scores between African
Americans and Euro-Americans has been observed on cognitive ability measures. As
stated, repetitively, intelligence tests have shown that, on the average, the scores of
African Americans are one standard deviation, or 15 IQ points, lower than that of Euro-
Americans (Manning & Jackson, 1984; Hunter & Hunter, 1986; Gottfredson, 1988;

Wollack, 1994). Cognitive ability represents an underlying and broad capacity for

3O

reasoning, abstract thinking, problem solving, and related skills (Snyderman & Rothman,
1986). According to Gottfredson (1988), the predictability of cognitive ability tests is
better than that of other predictors primarily because of its relation to general mental
ability. Some of the predictors used as pre-employment selection measures are more
related to general mental ability than others; the more related a measure is to general
mental ability, the greater the adverse impact between African Americans and Euro-
Americans (Lynn & Owens, 1993). Jensen (1985) also supported this contention in that
he found a signiﬁcant relationship between an ability test’s g loading and the degree of
difference between African Americans and Euro-Americans (Lynn & Owens, 1993). As
already stated, African Americans have reportedly scored lower on cognitive ability tests
than Euro-Americans; therefore, if African Americans are confronted with a selection
measure that looks like it assesses cognitive ability, this perception may inﬂuence one’s
sense of self-efﬁcacy, which eventually may negatively impact their performance on the
measure more so than Euro-Americans.

In a series of studies conducted by Steele and Aronson (1995), it was
hypothesized that African Americans would do poorly on a standardized cognitive ability
test if the measure was presented as assessing a facet of one’s intellectual capacity or,
simply if African Americans perceived the test to assess cognitive ability. The resultant
behavior would be the consequence of what the authors referred to as "stereotype threat."
According to Steele and Aronson (1995), a stereotype threat is "being at risk of
conﬁrming, as a self-characteristic, a negative stereotype about one’s group" (p. 797). For
African Americans, in particular, this stereotype threat is manifested by the notion that

"whenever African American students perform an explicitly scholastic or intellectual task,

31

they face the threat of conﬁrming or being judged by a negative social stereotype-a
suspicion-about their group’s intellectual ability and competence" (Steele & Aronson,
1995, p. 797). As a result, this threat was predicted to hamper African Americans’
performance on the cognitive ability test. Results of the studies showed that African
Americans do perform poorly on cognitive ability tests in comparison to Euro-Americans
when the measure is expressed as an assessment of intellect (Steele & Aronson, 1995).
Additionally, Steele and Aronson (1995) found that African Americans who were
anticipating a measure to assess their cognitive ability showed "greater cognitive
activation of stereotypes about African Americans, and greater activation of concerns
about their ability" than African Americans who were not expecting such a test (p. 805).
Findings also showed that heightening awareness of racial group membership hampered
African Americans performance on a verbal test although the test was not presented as
assessing cognitive ability. This series of studies highlights the differences in
performance on cognitive ability tests between African Americans and Euro-Americans
when African Americans are aware of the stereotype that “African Americans are
cognitively-inferior in comparison to their Euro-American counterparts” and perceive the
measure as assessing cognitive ability. Extending this line of research, this study will
focus on whether the knowledge of and belief in such a stereotype differentially impacts
African Americans who are individually-oriented in comparison to African Americans
who are collectively-oriented. Likewise, it will address how such a stereotype may

impact the level of self-efﬁcacy for African Americans.

32
Hypothesis 4: A positive relationship exists between performance on cognitive

ability tests and self-efﬁcacy when previous performance on cognitive ability tests is held
constant.

Hymthesis 5: The interaction between race and level of self-efﬁcacy will
inﬂuence performance on the cognitive ability test. The mean difference in performance
scores between African Americans and Euro-Americans will be smaller at a higher level
of self-efﬁcacy than the mean difference between the two subgroups observed at lower
levels of self-efﬁcacy (See Figure 3).

In sum, a potential explanation to adverse impact in selection testing--subgroup
differences in self-efﬁcacy—-will be tested. The model in Figure 4 represents the various
hypotheses detailed above. It is hypothesized that race directly inﬂuences one’s level of
self-efﬁcacy with respect to performance on cognitive ability tests (Hypothesis 1).
African Americans will possess a lower sense of self-efﬁcacy than Euro-Americans. The
relationship between race and self-efﬁcacy is also moderated by one’s individual-
collective orientation (Hypothesis 2). African Americans are more collectively-oriented
in general and this collective orientation leads to a lowered sense of self-efﬁcacy
compared to those possessing an individualistic orientation. The effect of the
collectivism orientation of African Americans will result in further losses in self-efﬁcacy

when African Americans are aware of the disparity in performance scores on cognitive

33

Enoch—ﬂow mo 32.1 95 33— 52.33 53835 - m oeswﬁ

58553
E

 

 

mcﬂomu08<
can?

meaotoaxx
65m

 

oocmscetom

>825. mo Bee—2 .. v 83.?—

 

 

Enema

 

ooeagotom A

3*

atom

235:2 56514.33“
amazom aw mmoeoeaaxs.

 

 

F

 

 

g:
I mm III. comm
mm
Emmzuoozeu

-aaesaﬁs

35

ability tests between African Americans and Euro-Americans, with Euro-Americans
outperforming African Americans. (Hypothesis 3). For Euro-Americans, those who are
collectively-oriented will possess a signiﬁcantly higher level of self-efﬁcacy than those
who are individually-oriented, if a signiﬁcant difference is observed. As before, this
relationship is moderated by the awareness of the difference in cognitive ability test
scores with Euro-Americans outperforming their African American counterparts.
Therefore, a race x collective- or individualist-orientation x beliefs about group
interaction is predicted. The nature of the interaction is depicted in Figure 2a and Figure
2b. Next, as supported in the literature, a positive relationship between self-efﬁcacy and
performance is hypothesized (Hypothesis 4). Those who believe that they possess the
ability to perform well on a cognitive ability test will outperform those who possess a
lower sense of self-efﬁcacy in comparison. Finally, the hypothesized relationship
between self-efﬁcacy and performance will be moderated by race (Hypothesis 5). The
performance of an African American on a cognitive ability test will be signiﬁcantly
affected by his or her level of self-efﬁcacy. Those African Americans with a high sense
of self-efﬁcacy for the cognitive ability test will outperform those African Americans
with a lowered sense of self-efﬁcacy. Likewise, Euro-Americans who possess a high
sense of self-efﬁcacy for the cognitive ability test will score higher. on the cognitive
ability test than those Euro-Americans who have a lower belief in their ability to perform
well. However, the magnitude of the difference in African American versus Euro-
American performance scores on cognitive ability tests will be greater when the level of

self-efﬁcacy is low than when self-efﬁcacy is high (See Figure 3).

Chapter 4

Method

Subjects.

See Appendix A for the power analysis for estimation of the necessary sample
size. A total of 110 African American and 101 Euro-American undergraduate students
from a midwestem, predominantly Euro-American university and a midwestem minority
research conference participated in this study. Sixty-two percent of the examinees were
female and thirty-six percent were male. Ages ranged from 17 to 39.

Measures.

The measures utilized to assess cognitive ability were subtests of the Air Force
Officer Qualifying Test (AFOQT) Form 0 (Skinner & Rec, 1987). The AFOQT Form 0
is a paper-and-pencil measure with multiple-choice questions constructed for group
administration (Skinner & Ree, 1987, p. 1). The two subtests of the AFOQT Form 0
chosen were the verbal analogies subtest and the math knowledge subtest. According to
the publishers, these two subtests were correlated, r = .55.

The AFOQT verbal analogies subtest taps the verbal factor (Skinner & Ree,

1987). It measures "ability to reason and recognize relationships between words"

36

37

(Skinner & Ree, 1987, p.24). The subtest contains 25 questions. {he reported internal
consistency/reliability using coefficient alpha = .80. Discrimination values (item-total

correlation; Ibis) for this subtest ranged from .43 to .78. See Skinner & Ree (1987) for a

reproduction of the Verbal Analogies subtest.

The AFOQT math knowledge subtest taps the quantitative factor (Skinner & Ree,
1987). It measures "ability to use learned formulas, mathematical terms, and
relationships" (Skinner & Ree, 1987, p.24). The subtest consists of 25 questions. The
reported internal consistency/reliability using coefﬁcient alpha = .88. Discrimination
values (item-total correlation, gbis) ranged from .43 to .84. See Skinner & Ree (1987) for
a reproduction of the Math Knowledge subtest.

The individualism-collectivism measure was adapted from the Wagner (1995)
Individualism-Collectivism Measure. This measure assesses ﬁve speciﬁc aspects of
individualism-collectivism (Wagner, 1995, p. 161): (a) ﬁve items assess one’s "personal
independence and self-reliance;" (b) the next ﬁve items measure the "importance
accorded to competitive success;" (0) the next three items assess the "value attached to
working alone;" ((1) the next four items measure the "espousal of norms about the
subordination of personal needs to group interests;" and (e) the ﬁnal three items assess the
"beliefs about the effects of personal pursuits on group productivity." In sum, the
complete individualism—collectivism measure consists of twenty items with a Likert-
response format ranging from 1 (completely disagree) to 5 (completely agree). See Table
1 for the individualism-collectivism scale intercorrelations and reliabilities. Refer to

Appendix C for the individualism-collectivism scale.

38

An individual ’5 awareness of and acceptance or belief in the reported difference in
cognitive ability test scores between African Americans and Euro-Americans was
assessed with six items. Responses to the items were indicated on a Likert-format
ranging from 1 (completely disagree) to 5 (completely agree). The six test items were: (1)
I am aware that African Americans score lower on intelligence tests than Caucasians; (2)
I am aware that research has found Caucasians to score 15 points higher than African
Americans on intelligence tests; (3) I have read reports that African Americans do not do
as well as Caucasians on intelligence tests; (4) Caucasians are smarter than African
Americans; (5) Caucasians do better on intelligence tests than African Americans; and (6)
African Americans possess less intellectual ability than Caucasians. See Appendix D for
a reproduction of the Awareness/Belief Measure.

The self-efﬁcacy measure consisted of 5 items assessing the subject’s belief about
his or her capability to perform well on the given test. Responses to four of the questions
were indicated on a Likert-format, with a range from 1 (completely disagree) to 5
(completely agree). These items were: (1) I am conﬁdent that I will do well on the verbal
analogies subtest; (2) I will earn one of the top scores on the verbal analogies subtest; (3)
I believe that I will have no problems on the verbal analogies subtest; and (4) Compared
to other subjects, I will perform well on the verbal analogies subtest. The ﬁnal self—
efﬁcacy item also has a Likert-format response scale ranging from 1 = 20% to 5 = 100%,
in which subjects responded to "What percentage of the 25 verbal analogy problems do
you think you will answer correctly?". See Appendices H and K for reproductions of the

self-efﬁcacy measures.

39

A demographics questionnaire requested information about the subject, such as
race, gender, self-reported GPA and self-reported past performance on similar tests such
as the SAT or ACT. See Appendix 2 for the reproduction of the Demographic
questionnaire. These items were used to describe the sample and the self-reported ACT
or SAT test scores were utilized to evaluate Hypothesis 4.

Procedure.

Before the administration of either cognitive ability subtest, all subjects were
given a demographics questionnaire and measures to assess their individual-collective
orientation. Subjects’awareness of and belief in the disparity between African Americans
and Euro-Americans on cognitive ability tests were assessed but counterbalanced. Half of
the subjects received the beliefs measure following completion of the individual—
collective orientation measure; the other half of the sample completed the beliefs scale
following completion of the math knowledge test. Prior to completing the cognitive
ability subtests, subjects were given general directions about the subtests and one
example (See Appendix E). Next, a brief description of the verbal analogies subtest
(Appendix F) along with ﬁve example items were provided (See Appendix G), then
subjects were asked to respond to self-efﬁcacy items corresponding to the verbal
analogies subtest. Once the verbal analogies subtest was completed, subjects were given
a brief description of the math knowledge subtest (See Appendix I) in addition to ﬁve
example items (See Appendix J). Then, subjects were asked to respond to the self-
efﬁcacy items about the math knowledge subtest. Once subjects responded to the self-
efﬁcacy items, the math knowledge subtest was administered. Subjects were then

debriefed.

 

40

Table 1a

Scale Reliabilities and Factor Intercorrelations of Individualism-Collectivism Measure

 

 

 

 

 

 

 

 

Factors 1 2 3 4 5
Collectivism 1 .72 "
Collectivism 2 .50" .79 b
Collectivism 3 .30M .09 .83 b
Collectivism 4 .03 .15M .05 .80 "
Collectivism 5 .19" .19** -.05 .29" .765

 

 

 

 

 

a Reprinted from Wagner, 1995.

b Statistic is a coefﬁcient alpha reliability estimate.

**p<.01

Chapter 5

DATA ANALYSES

W: Differences in self-efﬁcacy on cognitive ability tests will exist
between African Americans and Euro-Americans, with that of Euro-Americans being
higher. This hypothesis was tested with a t-test to determine if there was a significant
mean diﬁ’erence between African Americans and Euro-Americans in the level of self-
eﬂ‘icac'y for both math and verbal tests .

ﬂyppthesis 2: The relationship between race and self-efﬁcacy is moderated by
one’s individual-collective orientation. Self-eﬁicacy for both math and verbal tests was
regressed on race, individualism-collectivism and the product of race and individualism-
collectivism. A signiﬁcant product term in this regression contributed support for
Hypothesis 2. The individualism-collectivism construct was multidimensional and the
analyses were repeated for each dimension. A plot of this interaction would indicate
whether the form of the interaction supported the hypothesis (See Figure I).

W1 African Americans who are collectively oriented and are aware of
the difference in performance scores on cognitive ability tests between African Americans

and Euro-Americans will possess a lower sense of self-efﬁcacy than those

41

42

African Americans who are collectively oriented and are not aware of the differences in
performance scores. On the other hand, African Americans who are individually-oriented
and are aware of the difference in performance scores between African Americans and
Euro-Americans will not possess a signiﬁcantly different level of self-efﬁcacy than those
African Americans who are individually-oriented and are unaware of the difference in
performance scores between African Americans and Euro-Americans. Moderated
regression was used to test this hypothesis. Self-efﬁcacy for both math and verbal tests
was regressed on race, individualism-collectivism, beliefs about group, and the three
two-way interactions and the three-way interaction. The individualism-collectivism
construct was multidimensional and the analyses were repeated for each dimension. A
statistically signiﬁcant three-way interaction was taken as evidence of support for
Hypothesis 3a. The interaction was plotted to determine if the form of the interaction
supported the hypothesis (See Figure 2a).

Hypothpsis 39: Euro-Americans who are collectively-oriented and are aware of
the difference in performance scores on cognitive ability tests between African Americans
and Euro-Americans will possess a higher sense of self-efﬁcacy than those Euro-
Americans who are collectively-oriented and unaware of the difference in performance
scores. On the other hand, Euro-Americans who are individually-oriented and are
unaware of the difference in performance scores between African Americans and Euro-
Americans will not possess a signiﬁcantly different level of self-efﬁcacy than those Euro-
Americans who are individually-oriented and are aware of the difference in performance
scores. Moderated regression also was used to test hypothesis 3B. Self-efﬁcacy for both

math and verbal tests was regressed on race, individualism-collectivism, beliefs about

43

group, and the three two-way interactions and the three-way interactions. The
individualism-collectivism construct was multidimensional and the analyses were
repeated for each dimension. A statistically significant three-way interaction was taken
as evidence of support for Hypothesis 3b. Again, the interaction was plotted to
determine if the form of the interaction was as hypothesized (See Figure 2b).

Hypothesis 4: A positive relationship exists between performance on cognitive
ability tests and self-efﬁcacy when previous performance on cognitive ability tests is held
constant. This hypothesis was analyzed by computing the correlation between
performance on cognitive ability tests and self-efﬁcacy for both math and verbal tests
when prior peq'ormance on cognitive ability tests was held constant. Support for
Hypothesis 4 would be the statistical signiﬁcance of this partial correlation.

Hymthesis 5: The interaction between race and level of self—efﬁcacy will
inﬂuence performance on the cognitive ability test. The mean difference in performance
between African Americans and Euro-Americans will be smaller at higher levels of self-
efﬁcacy than the mean difference between the two subgroups observed at lower levels of
self-efficacy. Self-eﬂicacy for both math and verbal tests was regressed on race,
individualism-collectivism and the product of race and individualism-collectivism A
signiﬁth product term in this regression would constitute support of Hypothesis 5. The
individualism-collectivism construct was multidimensional and the analyses were
repeated for each dimension. A plot of this interaction indicated whether the form of the

interaction supported the hypothesis (See Figure 3 ).

Chapter 6

RESULTS

Descriptive statistics (i.e., means and standard deviations), reliabilities and
variable intercorrelations for the entire sample are shown in Table 2. Descriptive
statistics for each subgroup are found in Table 3 and variable intercorrelations are
presented in Table 4.

Prior to conducting tests of the study hypotheses, the individualism-collectivism
measure was analyzed to determine the factor structure. Five factors--(1) "personal
independence and self-reliance;" (2) "importance accorded to competitive success;" (3)
"value attached to working alone;" (4) "espousal of norms about the subordination of
personal needs to group interests;" and, (5) "beliefs about the effects of personal pursuits
on group productivity"-were identiﬁed through principal factors analysis with varimax
rotation of ﬁve factors. Five factors were rotated based on a consideration of factor
loadings and eigenvalues. These factors were consistent with previous analyses of the
same instrument (Wagner, 1995). Since ﬁve factors seem to be replicated in various
studies of this instrument including the data collected in this study, all subsequent

analyses were repeated for each of the ﬁve factors. The relatively low

45

3. v a . Sign 206503 e..- aomosﬁan 5 23:52 .523. 8......“ a e. noun—3:. 2a 223.. .23 o...
2:. _ 85.... 8338230 3.. 2......” Sou .m n 8.25... e..- . u 832 3 cocoa 83 .855 .m u §u_.oE<.8=m e..- . a 53.054. 58...: 3 .893 as... 83. .202

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.2... .2: .2. S. 2.- S. 2. 2.. S. S. .8. a... .2: .2. 8... 3.2 .5. £2
35 .z. .2. m...- S: 8.- 2..- S: .2. .2. .3. 2.- .2. man :4. e... 23>
.2... .2. S. 2. a... 2.- 8.- 2..- .2. 2. .2: .2.- ..2 2.... 5...... 5:.
.2... .2. 2.. 2. 2.- 2.- .2. .2. .2. .2: .2: S... 3.2 5.5 €>
.2... .3. .2. 2.. .2. S: S. m... 2.- .- and 3... 2.5
.2... .2. 8.- 2.- 8.- 8. m... 8. .- $2 3.2 2.5
.2... .2. 2. e... 8.- .- .3.- S: 43 G... new
3.... .2. 2. .2: .2.- .2. 8.. Ed 8.: 2.8
3.... S. .7 2: .- .- 2... 2.2 :26
5.. ... a... .2: 2.- 2... 2.2 ale-.2.
.3. S: .3. on... 2.: 2...... use
.3533
8.. .2. cm...- ﬁ..~ 2...... use
22.3.2.2
.- a... a... 32.5
cm. 2... 8....
I 2 N. : 2 a a .- e m 4 n N . $5.2m 53:.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

29:3 0.5.5 3.. mee_.a_o:eo.o.£ 23m .23 2.2.250 2.65% .2302 - N 0.2....

 

46

nay-Va.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

53-.

 

caumhoE<IO5m~

 

§_.0E<-=SE<

 

.tha- 3... on.”— and 2.: 35:235.— 5a:
.momm- S .m 3.: Sam :6— oosscoubm 30>
.mmoN no... 2.: 2mm no.2 xuaoEm..=om 5a:
53$ :6 onw— aoé mm: xuqum-hom 133>
New. om.~ $0.: and 0a.: m 52:60:00
Sm. m: .N 2.0- mud 3.0- v 82:60:00
mmm. mmd «To SN 86 m F5.236230
mmm. Inn 5.0— :3“ 8.: N 82:80:00
wmm. 2... No.2 voé 3mm— . 52380.30
«.9... om... Emm— mm.m 3.: mtg—om 353.0%
3.2.0- mod mhdm vmd Quo— o.oom 5:5... o>EcmoU 3:350
.33.- mod waNN mm.m Exam b=3< 03::on cognac—om
soc Em 2.32 .50 Em 2.82

mom-.302

 

“EutoE<-2:m 93 233.32. §E< .8 2.2.2.8— Eavgm E... 2.82 - m use...

 

47

mqvno

.529... 8:88 a 5 62.2%... o... 828.. 85¢ 05 E... . .38.. 58.68.30 3.. 35.8 :30 .m n 3.3.0..—
28 . n 6.22 mm 3.5.. 83 32.00 .5895. o... 26.... 2a m5..toE<-9.=m .32. E... .8885. o... 323 08 23:25. sat-F. .082

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- a. 8. 8. .7 8. 8. a... 8.. 8. .8...- ...n. 2.- t... 5.8
. 2.. - 8. .2. 8.- 8.. 8. 8.- :. ... .8. .8. 8..- t... ..:>
8. .8 .. - .8. 8. 8. 8. N..- 8. 8.- ...N. 8. .8: 8889.3 5.2
.2. .8. z. - :. ... :- n..- 8. 8.. 8.. 2. .8: 5383.5 e;
8.- 8. 8. 8. - .3. 8.. 8. 8. n..- 8. 8. 8.- 8.5
8. 8 . .- m .. 8. 8.. - 8.. 8.- 8.- 8.- 8. 8. 8.- 2.5
8. 8.- 8.- .8.- m.. .8. - 8. 8.- ... 8.- ...- 8.- 8.5
8.- m..- 8..- 8..- 8. 8.- .8. - .3. 8.- .- 8.- .8. 85
8.- . ..- .5 . .- .5 . .- .8. o..- .8. .8. - 8. 8. .- 8.- :.5
8.. 8.. 8.- 8.. 8.- 8.- 8. z. 8. - 2. a.- 8.- a8......
.3. .8. .8. .8. 2. 8. 8.- n..- 8.- 8.. .8. 8.- 8...... .89

- .5553

.8. .8. .8. .8... 8. 8.- 8. 8..- 8.- 8. 8. 8.- 38.... .89

- term-.3

.8..- 8.- .8; 8.- m... 8. 8.- 8. 8.- .8. 8.- 8. - 3.8.9

m. N. .. 8. 8 a b e n e n N .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

08a .3 238.8328..— oﬁom - v 03....-

 

48

intercorrelations of the ﬁve factors and the reliabilities of measures of the ﬁve factors are
presented in Table 2.

First, in order to assess order affects due to the stereotype beliefs measure, a
hierarchical regression was computed. The stereotype beliefs scale was counterbalanced
with some of the subjects completing the measure before the self-efﬁcacy measures and
some completing it afterwards. The dummy.coded variable, form, was entered into the
equation after race. There was no main effect for order (form; AR2= .001, p > .05). The
race it form interaction was entered into the equation and also did not account for a
signiﬁcant portion of the variance (AR2 = .007, p > .05). Thus, the order in which the

stereotype beliefs scale was presented did not affect levels of self-efﬁcacy for cognitive
ability tests.

Hypothesis 1 was not supported by a t-test analysis of subgroup mean differences
in level of self-efﬁcacy for cognitive ability tests. African Americans and White
Americans did differ in level of self-efﬁcacy; however, the difference was in the opposite
direction of that predicted. African Americans reported having a higher sense of self-
efﬁcacy for verbal analogies than their White American counterparts (t = 3.25, p < .05).
Similarly, African Americans reported a higher sense of self-efﬁcacy for math knowledge
items than White American subjects (t = 2.04, p < .05).

Hierarchical regression analyses were used to analyze the moderated relationship
between race and self-efﬁcacy as related to the individualism-collectivism factors (see
Hypothesis 2). Race accounted for a signiﬁcant portion of the variance in math

knowledge test self-efﬁcacy. This main effect reﬂects the earlier ﬁnding that African

49

Americans possess a higher level of self efﬁcacy for the math knowledge test than their

White American counterparts (R2 = .019, p < .05). Although the demographic variable
race did account for a signiﬁcant portion of the variance in math knowledge test self-

efficacy, there was no main effect for individualism-collectivism once entered into the

equation arterrace(AR12= .00057, p> .05; AR22= .,01189 p) .05; AR32= .00715, p)

.05; M42: .00919, p > .05 AR52= .,00350 p > .05 for the ﬁve factors respectively).
More speciﬁcally concerning the hypothesized relationship, the interaction between the

ﬁve individualism-collectivism factors and race did not account for a signiﬁcant portion

of the variance in math self—efﬁcacy (AR12 .,01479 p) 0,5 AR22= 000.33 p> .;05

A1132: .00722, p > .05; AR42= .,00190 p > .05; AR52= .,07277 p > 0.5 respectively).
The results of these analyses are presented in Table 5.

Similar to the results found for math knowledge test self-efﬁcacy, there was a
main effect for race (R2: 04.702 p < .05) in verbal analogies test self-efﬁcacy indicating
that African Americans possess a higher sense of self-efﬁcacy for the verbal analogies
test than White American students. As before, the ﬁve individualism-collectivism factors

did not account for a signiﬁcant portion of the variance in self-efﬁcacy for the verbal

analogies test (AR12 .00424, p> .;05 AR22= .,01318 p) .;05 AR32= .,00386 p> .;05

AR42= .,00614 p> .;05 AR52= 0.1231 p) .,05 respectively). Contrary to the predicted

relationship (Hypothesis 2), nonsigniﬁcant ﬁndings for the interactions were observed
when self-efﬁcacy for verbal analogies was regressed on race, the ﬁve factors of

individualism-collectivism, and the ﬁve interactions between race and individualism-

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8m. 8m..- 88. .8.- 8m. 8m... 88. m8.- 85.....8..oo

“anony—
8m. 88.. 8... mg

E. 83 8... 8e. 88. e8. 8:. 8.. 88.88.80
88. 88.. 88. Iqlm 2m

8... 8.. 888 .8. a8. 88.n- E. 88... 83.
.8. 88... 88. glam

.8. 8m. 8.. .8. 8... 8m. 8.... .8. 2.3.3.8

.303—
88. 88. 88. 4|. 2m

88. 8... 8m. 88.- s 8.. 8e..- 88. 8..- «5......8..oo
8.. e8.~ N8. 1...; o.

88. 8..- 82 3.8- .8. M88- .8. 8....- 83.
.8. 88..- 88. 3%

N8. 8.....- 8.. 2m.- 88. 88..- 8.. 2....- 3.3.62.8

x35.
N8. 88.... 28. Ialm 2m

8:. 8e. .8. e8. .8. .8.- ..8. .8.- .sa.>..oo..oo
8... 8.. .8. 8.8m-

..8. 88.. 82 8m... .8. 88...- .8. 88..- 83.
.8. 88... 88. lql. em

a .88 m e .8 ”.8 o . .88 m
.25... .8... 8cm. 2

 

 

 

 

 

 

hence-“mica «mo-H omen—3e5— 53.‘ 88.252. game-aux . m via--

 

51

{533038.— q..8w :e 838.... 8:080: .«o «.00-:0
0... 5.5: 20:3: 0... 2 m 82380.30 egg-.25 as...» o. 2.00: 8:080: we 5.85.0.3...- 05 Son: m:...o: .3 825.80: 0... 2
v 52:80:00 20:20 $383.2 00:08... 020?. 0... 2 m 82:80:00 :.80003 9 3.0.80: 005288.: 2 N 82:80:00
a.00:..=0.-:0m .05. 00:0::0..0::_ 8:082... 0.0:: 2 . 82:80:00 82:80:00 .80 8200.. 02.. 0... o. 8.0. 9. 82280.30 ”0.02

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.8. 88.- 8m. 88.- 8... 88.- 8.. 8...- “5.58:8
x3
8.. 8... .8... 3%
8... 8... 82. .2. 8m. 8... 82. .2. 2.8.8.8:
8.. 8.. 8... Mama
.8. .8..- .8. 82...- 88. m8..- .8. 82-... 8a:
.8. 88... 88. .Iﬁw
.. . m. 2e. 8.. 8.. 2m. .8. 8m. 8.. c 5:88.80
x003.
2m. 2... 8... mun-mm
8... e2.- ..8. m8.- 8.. 8.... .1. 8m. ace-.8880
8.. 8.8 8... moonw-
8: .8.- 8....- 83- .8. .88- .8. 82...- cos.
.8.- 88... 2... j. o.
e . .88.. m e --.< .8. e . .88 m
.252 .8... 8cm 2.

 

.28 m .58

 

52

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8.. 8..- 2.. .2. 8.. 8...- 2.. .2: 2.5.8.3:

xooad
8.. .... 2.... I..|. 2m

2... .2. h... a... .... .8. 8.. .8. 2.2.8.3:
.... 8... 2.... Iql. 2.

a... 2.... .2... .2. S... 2....- v... .8..- as.
.8. an... 2.... Iql. 0..

.8. .... a... .8. .8. .... 3... .8. 223.8..8

x3
.8. .m... 8... 3%

2.... E.- .... E.- .8. .8..- .8. ....- .em.>..8..oo
2.... a... 2... .Iqlsm

.... v8.- 8... 3...- 8... 2....- v... .2..- 83.
8... 3.... 2... 3%

E. o...- .2. E.- E. e...- .... E.- 3:33.30

.303.
E. a... 2.... udl. 0..

.... .3. 8.. 8.. ..... .8.- .8. 3....- .em.>..8..oo
c... .3. 2.... Iml. 0..

2.... .2. .2... 8.. 8... 2....- v... 2.....- 85.
.8. an... 2.... Iql. 0..

. 8.... m .. ".4 .2. .. . 8.... m
3.52 .2... 9.... .<

 

 

 

 

 

 

.80.-um...0m 80... .2389... 8...0> 60...? 5.8030... - e 03:...

 

53

:.00:....0.....0. 0:8 00:0..:0:0..:. 8:080:: 20:0 2 . 82280.30 82280.30 -.0 8088-. 02.. 0... 0. 80.0.. n. 82280.30 ”0.02

:.b.28....0.: :..0..m :0 8380: 8:080: :0 .80.-.0
0... 80.8 22.0.... 0... 2 m 82280.30 :.8.0..0.:. :..0.w 0. 2.00: 8:080: .3 3.88.800... 0... 2.0.8 .83: .3 823:3: 0...
2 .. 82280.30 2.0:08 w:.....03 0. 00:00.8 008?. 0... 2 m 82280.30 :.8000... 0. 00.8000: 008.888.: 2 m 82280.30

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.8. 2...- .... 3...- .8. 2...- .... 2....- 2.2.8.30

x005.
.... .... .oc. ..nmw

o... a... ..... o... 8... 8.... .2. o... .8..>..8._o0
2.... 8... .8. Nam-a

.... 2..- :0. 8..- 8... 2....- ..... .2..- 8...
8... 8.... 2.... Iql. 2.

o... .8...- .... 2...- .... .8...- ..... 2...- «5.8.8.30

x000...
.... .... .oc. ..mmw

...... ..... .... 8.. ..... 8... 8.. .2. «8.8.8.30
.v.. ... . .oo. ..mmw

o... ...... .8. .8. S... 2....- ..... .2..- 8...
8... ....... 2.... Iql. 2.

. 8.... m .. n8 .8. .. . 8.0. 0
.082 .88 8...... .<

 

 

 

.88 . 0......

 

54

collectivism. None of the ﬁve interactions involving race and individualism-collectivism

accounted for a signiﬁcant portion of variance in verbal analogies self-efﬁcacy (ARI2 =
.00816, p > .05; AR22 = .00022, p > .05; AR32 = .00366, p > .05; AR42 = .00098, p >

.05; ARS2 = .00077, p > .05, respectively). The results of these regression analyses are

displayed in Table 6.

The predicted relationship concerning beliefs about African Americans and their
cognitive ability and its differential impact on self-efﬁcacy depending on group
membership and individualistic-collective orientation was not observed (Hypothesis 3).
Self-efﬁcacy was regressed on race, individualism-collectivism, beliefs about group, the
three two-way interactions, and the three-way interaction. The three main effects were
entered in the ﬁrst block, followed by the three two-way interactions, and ﬁnally the
three-way interaction was entered in block 3. The ﬁrst block did not account for a

signiﬁcant (p < .05) portion of the variance in self-efﬁcacy for math knowledge tests;
however, some of the ﬁndings were marginally signiﬁcant (R12 = .02254, p = .1889; RZ2

= .03357, p = .0672; R32 = 02992, p = .0952; R42 = 03335, p = .0686; R52 = .02575, p

.1407). Betas for the variables in block 1 indicate that there was a main effect for race,
but not for beliefs about group or individual-collective orientation and this pattern was
consistent across factors (see Table 7). Block two contained the three two-way

interactions which also did not account for a signiﬁcant portion of variance in self-

efﬁcacy for math knowledge tests (AR12 = .01937, p > .05; AR22 = .00491, p > .05;

AR32 = .01126, p > .05; AR42 = .01057, p > .05; AR52 = 00430, p > .05). The focal

55

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w _ v. _ s. to, So. a... 3m. 58. 25. 595m 955 x N :8
£9. E. g. as. St m2. me. 59. £23 955 x 85:
N8. 58. 5%. g. 5%. am. 2:. 95. N 5225255 9 83.

9:. RM. 85. Immlem
5mm. $5.- $2 8m. _ - 5mm. 28.- 58. 58.- £25m 955
5am. 58.- me ._ So. _- E. 52. _- 58. 52.- .5. 5225255
5:. ”3.- 85.2 93.2- ms. 23- 3. an _- 83.

So. 89.“ 98. Al_ 2m
35.. 5%.- m3. m8- :5. 5%.- 95. 85.- £23 9 _ .55 x 83.

an am. so. Imlm 2m
wax 5%. 58. $5. :5. 5%. Go. :o. 523 955 x _ =8
8... 52. 85. NS; Sm. $5. 9:. 5. £23 955 x 85m
85. 58.- 58. 25.- mo. 35. _- 5:. :5- _ 5235255 9 85m

£5. £2 55. Ala 2m
an. E7 «9 ._ :5; $5. 2.- S 8.- 55:5 955
as. a: .- ”8; 52 .- 52. 2 .- 8. 8.- _ 5238.55
85. an- 25.5 MEX- So. 2 .N- m5 m2- 83.

9:. 85.. m8. Iml_ 2m

. Ev mm m a "2 E4 5 . 8V mm m
582 55m 5:5 Z

 

 

 

 

 

 

homo—twig amok. onto—3°5— 6“: “mommﬁﬁx cosmouwom - 5 via.

 

56

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5:. :5- 59: m3. _- an. 5.- as. 85.- 522 955
a. .39. :2 8m. _- E. 82 «1. 2m. 5 59:55:55
58. an- a 5.2 :87 95. 23. NE. E. _- 85.

58. 82 m8. 4|. 2m
95. N8. :5. m8. :5. NS. :5. 85. 523 x m :55 x 85¢

95. m8. coo JIM 2m
58. 98.- 52. N8; 2m. 25.. 58 m8; 522 955 x m =8
:5. :5. E. MS. 58. 5%. 2:. 2:. £23 955 9 85¢
«3. 35.- $3 9:..- of. 5;. _- gm. 8,..- m 5935255 9 8.;

N55. 58. :o. Imlm 2m
M55. as. ”MS 85. 5. £5- 95. 58.- 52.5 955
E m. 3.5. 3“: 22 EN. £2 oz. 5:. M 5235255
5;. NE. 32: Sf 58. :3. g. 52 82

35. 5: .N 58. |5|_ 2m
5. an- 55. ”8.- 5. 3n- ”3. ”8.- £25m x N :55 x 85“

an. 5:. So. Imlm 2m

5 . EV mm m a “2 ”NZ 5 5 EV mm m
582 5:: 55m 2

 

 

4:8 5 ".33.

 

57

_
o3...—

 

"g

.58 K. as:

98..

«5.-

523 x m =8 5. as.

x
m Sagan—BU x 83.

55:3 x 5 :8 x 85x

x
v Em_>_.ou=ou x 85—

 

58

relationship--the three-way interaction--did not account for a signiﬁcant portion of the

variance in self-efﬁcacy suggesting the lack of the relationship referred to in Hypothesis 3

(AR12= .,00141 p) 0.5; AR22= .,00162 p) .05; AR32= 0000.2 p> .05; M42:

.00023, p > .;05 AR52= .,00073 p > .05). The results of these regression analyses are

presented in Table 7.

As was done for self-efﬁcacy for math knowledge tests, self-efﬁcacy for the
verbal analogies test was regressed on the three main effects in stepl, the three two-way
interactions in step two, and the three-way interaction in step 3. Contrary to what was
found for the self-efﬁcacy for math knowledge tests, block l-—main effects of race,

individualism-collectivism, and beliefs about group performancendid account for a
signiﬁcant portion of the variance in self-efﬁcacy for verbal analogies (RI2 = .08275, p <

.;05 R22: .,08959 p < .05, R32: 07,792 p < .;05 R42: 0,8383 p < .05; R52: .,09081
p < .05). The amount of variance accounted for by the individual variables was similar
across the ﬁve factors. For the variables in block 1, there were main effects for race and
beliefs about group performance consistently as shown by the betas in Table 8.
Individual—collective orientation did marginally (p < .10) impact reported level of self-
efﬁcacy for the verbal analogies test for two of the factors: block 12 (working alone) and
blocle (impact personal pursuits have on group productivity) as shown by the betas in
Table 8. Next, the block containing the two-way interactions was entered. As with the

math knowledge test self-efﬁcacy, the two-way interactions did not account for a

signiﬁcant portion of the variance in self-efﬁcacy for verbal analogies test (ARI2 =

.01900, p> .;05 AR22== 00,572 p> .0;5 A1132: .,00651 p> .05; AR42= .,00218 p>

59

.05; ARS2 = .00197, p > .05). Similar nonsigniﬁcant ﬁndings for the three-way

interaction were found for self-efﬁcacy for verbal analogies as for self-efﬁcacy for math
knowledge items (AR12 = .00106, p > .05; AR22 = .001 16, p > .05; AR32 = .00003, p >

.05; AR42 = .00828, p > .05; ARS2 = .00476, p > .05, respectively). The results of these
regressions for Hypothesis 3 are reported in Table 8.

The predicted relationship between self-efﬁcacy and performance was evaluated
while controlling for prior performance on a cognitive ability test. Subjects’ self-reported
standardized test scores (ACT and/or SAT) and, for most of the subjects (n = 161), ACI‘
and/or SAT scores were obtained from the university the subjects were attending. Self-
report scores of prior performance on cognitive ability tests and obtained scores on the
tests were signiﬁcantly correlated with each other (r = .936, p < .05). The
intercorrelations of self-report cognitive ability test scores, obtained cognitive ability test
scores, self-efﬁcacy for each cognitive ability test, verbal analogies test performance, and
math knowledge test performance are shown in Table 2. A partial correlation between
self-efﬁcacy and test performance, controlling for self-reported prior cognitive ability test
performance, was computed in order to test Hypothesis 4. As predicted, a signiﬁcant
relationship between self-efﬁcacy for verbal analogies and performance on the verbal

analogies test was observed when controlling for self-report prior performance (ran.

emmym ”manna” prior m perform-nee = .1217, p < .05). However, the relationship

between self-efﬁcacy for verbal analogies and performance on the verbal analogies test,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

aw. mm _ .- ”8. N8. EN. 33 we. 5N5. 22.5 955 x N :55
Se. Nan- N5. NmN. 2e. 5%. N2. «8. 525m 955 x 82
De. 3w- e2. SN. 85. 95.. mi. 58.- N 5225255 9 82

NNN. 59.. 58. .N 9%
53. oz. 33 me. NS. ENN So 5:. 225m 955
35. may 5%. 2a.- :5. 22- NS. om? N 5235255
Em. NE. 32.. em: :5. ONE- Em. 3:- 85m

58. Re 58. 5.15%.
mNe. 98.. N8. 25.. mNe. 95.. N8. 25.- 22.5 x _ .55 x 85m

nNe. 5mN. :5. In: em.
59m. 95. N8. 53. o. _. 83 ea. mNo. 55:5 955 x _ :8
.mm. an 99.. SN. 95. N55. 52. «8 225m 955 x 85m
95. ea. 59.. 58. m2. 8?. N2. 3:. _ 5225255 2 82

SN. we: 55. .le 2m
N _ e. NNw. EN. 35.. 2o. 5me N8. 9:. 225m 955
8m. wee; 58. NE. N2. 22- eeo. $5.- _ 5225255
55. 3.- NmNN em; :5. P3. 5mm. 23. 85m

:5. mNN5 m8. Elam

8 EV mm m 5 "2 R4 5 8 EV mm m
.252 5:5 55m 2

 

 

 

 

 

 

hoaoEm..=om amok mo_wo-=< $355 ”moms—mi :oumucmom - N. 2an

 

61

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

555N. N55..- 555.. 555..- 5.5. 555.N N55. 55.. 22.55. 955
$5. 35.- $5.. 555..- 55.. .55.. 55.. 55.. 5 E22.8.55
55.. 555... 555.5. 555.5N- N55. 5NN.5- 555. 555..- 83.

555. 5.5.5 555. I5I. 25

:5. 555.. 555. .55.- :5. 555.- 555. .55.- 22.85. 5 5 .55 5 88.
:5. 555. 555. |5|5 25

555. 555. .55. N55. 555. 5.5. 5N5. 5N5. 22.55. 955 x 5 .55
:5. 55.. N55. 555. N55. 555. 55.. 5.5. 22.55. 955 5 88.
N55. 555.- 5N5. .55.- 555. 555.- 5.N. 5N..- 5 5225255 5 88.
.55. 55... 555. 4N 5|.5

5. 5. 5.5.- N55. 5NN. 5N5. N5N.N N55. 55.. 22.55. 955
55. 55..- :5.. 55..- .5... 555. 55.. 555. 5 52:52.55
5N5. NNN.- 5. .5 .55..- .55. 55N.5- 555. NE. .- 855.
.55. .555 555. I5l. 25

.5.5. :5. 555. 5.5. 5.5. :5. 555. 5.5. 552.55. 5 N .55 5 83.
5.5. .5N. .55. 15 55.15

5 . .5.. 55 5. 5 ”.4 55.5. 5 5 .5.. 5.5 5.
.25.). .225 5.55 5<

 

.255 a 535...

 

62

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N55. .55.. 555. 555. N55. 555.. 555. 555. 2255. x 5 .55 x 83.
N55. 555.. 555. 5 525
555. 555.- 555. 555.- 555. 555. 5N5. N55. 2255. 955 x 5 .55
5 . 5. 555..- .55. 555.- 555. 555. 55.. 555. 5.255. 955 x 83.
55N. 55...- .55. 5.5..- .N5. N5.. 5NN. 5....- 5 5225255 5 83.
.55. 5.... N55. JIN 25.
555. 555.. 5.5.. .2 5.5. 535 555. 55.. 2255. 955
5NN. 55N.. .55.. 555.. 555. 555.. N... :N. 5 5225255
555. 555. 555.5. 555.5. N55. 55.5- 555. 5.5..- 83.
555. N555 .55. I5I. 25
55 .. 555..- .55. 555.- 55.. 555..- .555. 555.- 2255. x 5 .55 5 83.
55.. 555.. 555. |5|5 25
:.N. 55... .5.. 5. .. 555. 555.- N55. N.5.- 2255. 955 x 5 .55
55.. 555.. _ 555.. 5N5.. 555. N5.. 55.. 5.5. 2255. 955 x 83.
55N. 55... 555. 55... 555. 555.- 5NN. 5N..- 5 5225255 5 83.
NN5. 55.. N55. JIN 25.

5 . .5.. 55 5. 5 55 552 5 5 a. 5.5 5.

.252 55... 5.55. .<

 

 

 

 

5:8 5 555.—-

 

63

when controlling for obtained ACT scores, was not observed (rself-efﬁcncy/test perfomccmoms,
“mm = .0918, p > .05). Furthermore, a signiﬁcant relationship was found between
self-efﬁcacy for math knowledge items and performance on the math knowledge items
when controlling for self-report prior cognitive test performance (rsclf,cmcacy,,w Momcc,,c;f.
“pm pm, 15: Mom“ = .2665, p < .05). Likewise, a similar relationship between self-
efﬁcacy for math knowledge tests and performance on the test when controlling for
obtained past performance scores was signiﬁcant (rsdfcmmynw ”mam-o, a, performance =

.25, p < .05).
Hierarchical regression analyses were used to assess Hypothesis 5. Test

performance was regressed on race, self-efﬁcacy, and the product of race and self-

efﬁcacy. For math knowledge test performance, there was a main effect for race (R2 =
.04751, p < .05). Similarly, the amount of variance accounted for by self-efﬁcacy for
math knowledge tests was signiﬁcant after race was entered into the equation (AR2 =
.15456, p < 05). However, the focal relationship concerning the amount of variance in
math knowledge test performance accounted for by the race x self-efﬁcacy for math
knowledge test interaction was not signiﬁcant (AR2 = .00078, p > .05). The results of
these regression analyses are also presented in Table 9.

There was a main effect for race on verbal analogies test performance (R2 =
.02333, p < .05), suggesting that Euro-Americans outperformed their African American

counterparts. Likewise, self-efﬁcacy for verbal analogies accounted for a signiﬁcant

portion of the variance in verbal analogies test performance after race was entered into the

equation (AR2 = .05983, p < .05). Although there were main effects for race and self-

64

efﬁcacy for the verbal analogies test, the prediction that differences in test performance
scores between subgroups will be smaller at higher levels of self-efﬁcacy for verbal
analogies than the mean difference between the two subgroups at lower levels of self-
efﬁcacy for verbal analogies (Hypothesis 5) was unsupported by the data. The
hierarchical regression analysis indicated that the race x self-efﬁcacy interaction was not

related to performance on the verbal analogies test after race and the test self-efﬁcacy

were entered into the regression (AR2 = .00634, p > .05). The results of these regression

analyses are presented in Table 10.

65

 

 

 

 

 

 

 

 

 

 

 

 

 

555. 55... 5N.. 555. 555. 55... 5N.. 555. 558555.555

x05
55. 55N. .55. 15l5 8.5

.55. 555.. 55.. 555. 555. 555.5 555. 555. 525555-555
555. N55.... 55.. 5.55

555. 5N5.. 555.. 5N5.. .55. 55N.5 .55. 55.5 83.
.55. 5N5.5. 555. I5I. 25

5 . 8.5.5 5. 5 ...< 53.5 5 5 .55 5.5 m
.252 55... .555. 2

 

 

 

 

 

 

555555-5555. 5555. once—35:2 552 ”5.5.5.554. 5.5.5583”— .. m 538.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5555. 555.. 5:. 555. 5555. 555.5 5:. m5. 558855.555m

x05
5555. 58.5 85. glam

£5. 55555.. 85. 85.- 85. 5555.5 58. £5. 558£m£5m
85. an: 553. .Nlﬂw

Sn 5%.. 555.5 85.5- 58. $5.5 E. 58.5 855
58. 553. 58. 3 55
5 85mm m 5 54 5% 5 5 85mm m .

.5552 .5555 b555,“. 5<

 

 

 

 

 

 

35555853 58,—. 83255555 ﬁn5o> "mam—£2 commmawom - 2 035.

 

Chapter 7

DISCUSSION

This study was conducted in an attempt to assess the possible contributions of
self-efﬁcacy--beliefs about one’s capability--in accounting for differences between
African Americans and Euro-Americans in cognitive ability test performance. For
reasons, some suggested, most of the expected ﬁndings were not observed as discussed
below. Potential explanations are offered for the observed ﬁndings.

The ﬁndings of the study suggest that African Americans possess a higher sense
of self-efﬁcacy for cognitive ability tests than Euro-Americans. Orientation toward the
self or others did not inﬂuence one’s belief in his or her ability to perform well on the
cognitive ability tests. Likewise, stereotypical beliefs about African Americans’ level of
intelligence did not impact the self-efficacy of group members. As expected, however,
one’s belief in his or her ability was related to performance even when prior performance
was taken into consideration. Finally, some of the difference in cognitive ability test
performance between groups was not accounted for by the interaction of racial group

membership and increasing level of self-efﬁcacy.

67

68
Contrary to the expected relationship between self-efﬁcacy for cognitive ability tests and

subgroup status, African Americans reported a higher sense of self—efﬁcacy for cognitive
ability tests than their Euro-American counterparts. Potential explanations for this
ﬁnding may include differences in general or overall perception of oneself--self-esteem--
and its relationship to beliefs of capability in speciﬁc skill areas, and different reference
groups African Americans and Euro-Americans use to assess their self-competence.
Research has shown that African Americans have a more positive self-perception
than Euro-Americans. More speciﬁcally, contrary to predictions/expectations, African
Americans possess a more positive assessment or evaluation of themselves in comparison
to Euro-Americans (Turner & Turner, 1982; Tashakkori & Thompson, 1991). While
self-esteem can be considered a general attitude or evaluation about the self based on
many speciﬁc beliefs, self-efﬁcacy has been deﬁned as general or task-speciﬁc. General
self-efﬁcacy refers to a global conﬁdence one has to successfully execute tasks; whereas,
task-speciﬁc self-efﬁcacy concerns beliefs about one’s capability in performing a speciﬁc
task (Bandura, 1977; Gist & Mitchell, 1992; and Stanley & Murphy, 1997). Research has
found a strong relationship between general self-efﬁcacy and self-esteem (W oodruff &
Cashman, 1993; Schwarzer, 1993; and Saracoglu, Minden, & Wilchesky, 1989: as cited
in Stanley & Murphy, 1997). Gist and Mitchell (1992) predicted that self-efﬁcacy would
be lower for individuals with low self-esteem compared to those with high self-esteem.
More recently, Stanley and Murphy (1997) attempted to determine: (1) the
relationship between self-esteem and general self-efﬁcacy, (2) whether self-esteem and
general self-efﬁcacy were distinct constructs, and (3) the relationship between general

self-efficacy and task-speciﬁc self-efﬁcacy. Findings suggest that self-esteem and general

69

self-efﬁcacy are highly correlated which suggests that the two constructs are not distinct.
Additionally, general self—efﬁcacy was found to be related to task-speciﬁc self-efﬁcacy.
Thus, if African Americans do possess a more positive self-esteem than Euro-Americans
and self—esteem is related to a general sense of competency which in turn is related to
one’s belief in his or her ability to perform well on a given task, then the ﬁnding that
African Americans have a higher sense of self-efﬁcacy for cognitive ability tests than
Euro-Americans is not surprising.

The ﬁnding in this study contributes to current research regarding issues
surrounding construals of self-esteem. Heiss and Owens (1972; see also Martinez &
Dukes, 1987) proposed two types of self-esteem--public vs private domain--and
investigated racial group differences in self-esteem depending on the domain. Heiss and
Owens classiﬁed sense of self-worth or importance and intimate interactions or
relationships with others in the private domain. They found that African Americans and
Euro-Americans did not differ in regards to self-evaluations on private domain concerns.
On the other hand, Heiss and Owens (1972) found that self-evaluations of African
Americans were more negative than those of Euro-Americans for traits referenced by
public (majority) standards such as those relevant to work-related competence and
success, in addition to their ability, achievement, or intelligence (public domain).
Consequently, research by Heiss and Owens (1972) would suggest that African
Americans and Euro-Americans would differ in regards to self-efﬁcacy for cognitive
ability tests and that African Americans would report a lower sense of self-efﬁcacy as
found by Pajares and Kranzler (1995). However, as indicated, this ﬁnding was not

observed in this study suggesting that perhaps the line between these two domains is not

70

as distinct as predicted for African Americans or simply the public domain for African
Americans does not refer to the majority or public standards.

As previously suggested, African Americans and Euro-Americans may refer to
different groups for comparison when attempting to construct one’s evaluation of the self.
White and Parham (1990) suggest that African Americans and Euro-Americans use
different reference groups or standards for some of their self-evaluations. In particular,
how an African American feels about himself/herself and what he/she is capable of doing
is affected by individuals in their immediate surroundings (i.e., family, friends and others
in African American community), not by the evaluations made by those of the majority or
Euro-Americans, in particular (Rosenberg & Simmons, 1972; Rosenberg, 1979; Krause,
1983: as cited in Stanley & Murphy, 1997). If this is true, you would also expect African
Americans in this study to have a high self-esteem because among their own group, they
may be more superior (i.e., not as many African Americans go to college) than they are
relative to the general population.

Thus, the observation that African Americans reflect a more positive belief in
their capability than Euro-Americans may result from a more positive self—esteem in
general and that African Americans are not affected by assumptions held by
"nonsigniﬁcant" others and/or do not compare themselves to standards held by the
majority.

This study also investigated differences in individual-collective orientation
between African Americans and Euro-Americans and how these differences may impact
the relationship between self-efﬁcacy for cognitive ability tests and subgroup status.

Once again, contrary to expectations, individual-collective orientation did not influence

71

the relationship between race and self-efﬁcacy. As already discussed, subgroup status did
influence level of self-efﬁcacy for cognitive ability tests, but individual-collective
orientation did not impact perceptions of capability. African Americans and Euro-
Americans did not differ signiﬁcantly in their orientation toward the self or others
supporting some past research (Oyserman, et a1, 1995).

According to Tajfel (1981: as cited in Arroyo & Zigler, 1995), a person’s
relationship with his or her social group is affected by how the group is evaluated.
Arroyo and Zigler (1995) found that African Americans who believed that African
Americans were viewed negatively lowered their identiﬁcation with the social group.
This maybe a factor of an attempt to protect self from group-based stereotypes
(Oyserman,, et al, 1995, stereotype vulnerability; Steele & Aronson, 1995, stereotype
threat). More speciﬁcally, in the construction of one's identity, African Americans are
engaged in both the process of building a positive sense of self and disavowing negative
stereotypes concerning the group of which they are members. This attempt to discredit
negative group perceptions may be manifested in low group identiﬁcation.

Next, the focal relationship regarding the impact of beliefs about African
Americans and their cognitive ability and its differential impact on self-efﬁcacy
depending on group membership and individual-collective orientation was addressed.
This relationship may not have been observed for a number of reasons. First, there may
be a difference between collectivism and group identiﬁcation or ethnic pride. In
particular, as research has suggested, one’s collective or individual-orientation changes
depending on the target individual. Thus, in order to ascertain the impact that beliefs

have on self-efﬁcacy for African Americans in relation to identiﬁcation with the group, a

72
measure of ethnic pride or afﬁliation (Whittler, Calatone, & Young, 1994) may have been

more explanatory. For example, those with low ethnic pride or afﬁliation with the group
may not see stereotypes relating to the self and may not engage in protective behaviors;
on the other hand, those with a high sense of ethnic identity may be more reactive to the
statements, thereby protecting one’s identity. Also, the greater one’s racial self-esteem or
ethnic pride, the more reactive African Americans may have been in their responses to
belief statements. This ﬁnding may have resulted because as Arroyo and Zigler (1995)
found, for African Americans, the greater the group identiﬁcation, the lower the belief in
group-based stereotypes held by the individual; thus fostering a reaction.

In regards to stereotype threat (Steele & Aronson, 1995) African Americans may
not have responded to being aware of the stereotypes or believing them because
agreement may have indicated conﬁrmation of the stereotype beliefs. More speciﬁcally,
African Americans who were aware of the cognitive ability assessment of the measure,
such as in this study, were “the most avoidant of conforming to stereotype images of
African Americans” (Steele & Aronson, 1995, p. 804). Also, with regard to stereotype
vulnerability (Oyserman, et al, 1995), by denying awareness and belief of the stereotypes
African Americans were protecting their sense of identity. African Americans could have
been protecting their positive views of self and responding accordingly to questions about
stereotypes (Oyserman, et al, 1995, stereotype vulnerability; Steele & Aronson, 1995,
stereotype threat).

Finally, Euro-Americans did exhibit a relationship between beliefs about the
"cognitive inferiority" of African Americans and verbal self-efﬁcacy and verbal

performance but this relationship was not evident for African Americans. For African

73

Americans, the stereotypes about "lack of intellectual ability" were not related to
perceptions of ability. Arroyo and Zigler (1995) found that for African Americans,
stereotypical beliefs were not related to perceptions of self-efﬁcacy. This ﬁnding may
have been observed because in order for the stereotypes to impact African Americans’
perceptions of capability, they ﬁrst had to be aware of the stereotypes, accept them and
then believe that they pertain to them (Rosenberg, 1979). For the observed relationship
for Euro-Americans, whether a person believes the stereotype to be true or not and
whether he or she is a member of the targeted group or not, the biased information is still
accessible and influential when making judgments or decisions about one’s actions
(Devine, 1989); thus the beliefs about the "cognitive inferiority" of African Americans
may be related to Euro-Americans level of self-efﬁcacy and subsequent performance on
the verbal analogies test.

The predicted relationship between self-efﬁcacy and performance was found
further supporting current research ﬁndings (Locke, et al, 1984; Wood & Bandura, 1989;
Zimmerman, et al (1992) and Pajares & Kranzler, 1995). Going a step further,
controlling for prior performance, the relationship was still present. The more one
believes in his or her ability to do something, the better he or she will perform even when
taking prior performance on a similar task into consideration. However, self-report prior
performance was related to perceptions of ability for Euro-Americans, but not for African
Americans in some instances. This once again may be the lack of African Americans
comparing themselves to Euro-Americans or the standards set by the public (Rosenberg

& Simmons, 1972; Rosenberg, 1979; Krause, 1983).

74
In addition, although self-efﬁcacy and performance were positively related,

African Americans possessed a higher sense of self-efﬁcacy; whereas, their Euro-
American counterparts outperformed them. This ﬁnding may be reﬂecting research
suggesting that overconﬁdence may be detrimental to performance or the overarching
positive self-attributions and perceptions of African Americans. In particular, Zakay &
Glicksohn (1992) found that overconﬁdent individuals (e.g., those possessing an
unrealistically high sense of self-efﬁcacy for ability level) were more likely to make
mistakes on multiple-choice tests, speciﬁcally. Conﬁdence was positively related to
performance, but overconﬁdence was found to be detrimental (Zakay & Glicksohn,
1992).

Finally, the lack of the predicted relationship between racial groups and increasing
self-efﬁcacy for cognitive ability tests may be explained by the overarching positive
perceptions of oneself by African Americans across situations.

Conclusions

The central question addressed by this study was: "Does self-efﬁcacy account for
differences between African Americans and Euro-Americans in cognitive ability test
performance?" Self-efﬁcacy did not explain cognitive ability test performance
differences between these two groups. African Americans reported a higher sense of self-
efﬁcacy which may be related to reports suggesting that African Americans possess a
more positive perception of the self than Euro-Americans. This perception of self may be
resilient to stereotypes held by the majority and perhaps the denial of stereotypes leads to
overcorrection/overcompensation fostering a higher sense of self-efﬁcacy, and subsequent

overconﬁdence in ability. Furthermore, this overconﬁdence may have actually

75
contributed to the poorer performance by African Americans on the cognitive ability test.

When examining the relationship between self-efﬁcacy and performance on a complex
decision task, Cervone and Wood (1995) found a negative relationship between one’s
belief in his or her ability to perform on the task and actual performance. Those who
believed that they were highly competent “tended to overestimate their capabilities which
resulted in marked discrepancies between expectations and attainment” (Wood, George-
Falvy, Debowski, unpublished manuscript, p. 2)

Although a majority of the research previously presented existed before
undertaking this study, this study does contribute to research. First, conﬂicting research
exists concerning the “public” self-esteem of African Americans (i.e., African Americans
have a higher sense of self-esteem than Euro-Americans and vice versa; see Peterson &
Ramirez, 1971; Rosenberg & Simmons, 1972; Harris & Stokes, 1978; Porter &
Washington, 1979; Osborne & LeGette, 1982; Wade, Thompson, Tashakkori, & Valente,
1989; Tashakkori, Thompson, Wade, & Valente, 1990; White & Parham, 1990). The line
of research that suggested African Americans would display a lower sense of conﬁdence
in one’s ability in regards to public standards (e.g., cognitive ability; Heiss & Owens,
1972; Martinez & Dukes, 1987) was used to generate the hypotheses tested in this study.
However, the ﬁndings of this study appear to be more consistent with the research
suggesting that African Americans may not employ the same references and standards
when determining one’s perception of oneself (see White & Parham, 1990). Thus,
African Americans would not necessarily report lower perceptions of ability.

Also, conﬂicting research exists concerning whether self—esteem and self—efﬁcacy

are two distinct constructs. Once again, this study was based the line of research

76
suggesting that the two constructs were different (i.e., Hunt and Hunt (1977) reported that

African Americans have a higher level of self-regard than Euro-Americans, but a lower
sense of self-efﬁcacy), however a relationship between the two does exist (Lorr &
Wunderlich, 1986). For example, research conducted by Hughes and Demo (1989)
suggests that self-efﬁcacy is lower for African Americans, not selfoesteem and that
neither self-efﬁcacy nor academic success are related to self-esteem for African
Americans, as it is for Euro-Americans, perhaps due to different reference groups. Also,
Lorr & Wunderlich (1986) suggest that self-efﬁcacy is a subcomponent of self-esteem. In
light of the ﬁndings of this study, for African Americans, one’s perception of oneself and
one’s perception of his or her ability to perform a given task may not be as distinct as
expected when conducting this investigation.
Limitations

In an attempt to recruit a sufﬁcient number of African American participants,
African Americans from other universities and minority programs were asked to
participate in the study. A signiﬁcant portion of the African American sample were
participating in programs that had recognized or identiﬁed the students based on their
academic performance and potential. Such recognition may have inﬂuenced their belief
in their ability to perform well on the cognitive ability tests. As a result, African
. American student responses to the self-efﬁcacy measure may have been affected.

While recruiting from other universities increased the sample size, this also
resulted in an inability to obtain objective scores of prior cognitive ability test
performance for these participants. Although the correlation between the obtained test

scores and self-report scores for the remainder of the sample was high, the participants

77

could have falsely reported prior performance scores due to lack of accurate recall or bias.
This is not likely, however, since the relationships between variables for obtained
cognitive ability test score were the same for the same relations reported for the total
sample and self-reported cognitive ability test score with only one exception.

Additionally, with the need for African American participants, the focused
recruitment may have primed African Americans in regards a race factor. This
heightened racial sensitivity may have inﬂuenced responses and performance.

In relation to the study design, the overt presentation of stereotype beliefs may
have elicited reactive responses or demand characteristics from both African Americans
in an attempt to protect self-identity or self-esteem and for Euro-Americans possibly in an
attempt to avoid appearing racist in an acknowledgment of awareness of and belief in
stereotypes. Further, for a race-based study, the race of the experimenter/researcher was
consistent across sessions (i.e., African American researcher present for all sessions),
potentially impacting student responses to stereotype statements whether by African
Americans or Euro-Americans.

Also, collectivism was not assessed in regards to the speciﬁc target group, i.e.,
racial group membership. Research has shown that individual-collective orientation
differs in relation to target person. Consequently, a more appropriate measure would
have assessed orientation toward the individual’s racial group (i.e., social identity; Tajfel,
1978) or, more speciﬁcally, the assessment of racial pride (Porter & Washington, 1979),
collective self-esteem (Ethier & Deaux, 1990) or ethnic afﬁliation (W hittler, et al, 1994)
may have been more appropriate for the hypothesized relationships for African

Americans.

78

Future Research

Self-efﬁcacy is inﬂuenced by past performance, modeling, and persuasion
(Bandura, 1977; Gist & Mitchell, 1992). If African Americans have a higher level of
self-efﬁcacy than Euro-Americans and this was not related to past performance, such as
found in this study for verbal self-efﬁcacy, how does past performance relate to self-
efﬁcacy for African Americans? Additionally, the standards set by the majority or
stereotypes held (verbal persuasion) may not impact level of self-efﬁcacy for African
Americans. Consequently, future research might explore the determinants of African
Americans’ task-speciﬁc self—efﬁcacy. Perhaps, more importantly, do African Americans
possess task-speciﬁc self-efﬁcacies or do they perceive their ability to perform tasks in
general as related to self-esteem? For example, there was a strong positive relationship
between verbal and math self-efﬁcacy for African Americans, but not for Euro-
Americans.

In spite of the prevailing notion that African Americans are more collective;
future research should investigate what is meant by this notion of collectivism and in
regards to what targeted population. In what situations are African Americans more
collective, and is this relationship moderated by the situation such that if negative
stereotypes surround the situation then African Americans will not identify with the
group? Also, what is the relationship between collectivism and ethnic pride, and is ethnic
pride what African Americans possess more of in comparison to their Euro—American
counterparts?

In addition, if African Americans do not make a distinction between the public

and private domains as suggested, then the theories concerning the construals of self-

79

esteem may not be applicable to African Americans when stereotype threat or
vulnerability becomes relevant. For example, although research hypothesizes that
African Americans will possess a lower sense of self-esteem in regards to public
domains, this does not take into account their attempt to disavow negative stereotypes
which may manifest themselves in an "overcorrection" or more positive sense of self in
the "public" domain for protective purposes.

Along those lines, does "overcorrection" in the development of self-esteem lead to
a sense of "overconﬁdence" in ability? Thus, the relationship between overconﬁdence
and self-esteem, self-efﬁcacy in particular, should be investigated in African American
groups.

In sum, belief in one’s ability to perform well on cognitive ability tests did not
explain the difference between African American and White American test performance.
Future research should attempt to determine if these are true differences in ability or a

function of test-taking motivating.

APPENDICES

APPENDD( A

Power Analysis

Effect sizes:

(Locke, et al (1984); Wood & Bandura (1989); Zimmerman, et a1 (1992);
& Pajares & Kranzler (1995))

self-efﬁcacy and performance-4 = .39-.64, p < .05

(Oyserman, et a1 (1995)
collectivism-F(l, 102) = 3.69, (MAA = 2.99 vs. MBA = 2.78). P < .05
individualism--F(1,102) = 1.46, (MAA = 3.79 vs. MBA = 3.66), p > .10

(Pajares & Kranzler, 1995)

self-efﬁcacy and race-r = .21 (AA score lower), p < .001

self-efﬁcacy and race--mean difference: MAA = 74.9 vs. MBA = 83.2, diff =
8.3, SD = 16.3, t = -4.09, p < .0001

(Arroyo & Zigler, 1995):

For African Americans:
stereotypical beliefs and self-efﬁcacynr = -.08
stereotypical beliefs and identityur = -.27, p < .05
self-efﬁcacy and identitynr = .24, p < .05

For Euro-Americans (Arroyo & Zigler, 1995):
stereotypical beliefs and identityur = -.12
stereotypical beliefs and self-efﬁcacyur = -.12

(Note: AA = African Americans and EA = Euro-Americans)

80

For hypothesis 1: Mean difference between African Americans and Euro-Americans will
be analyzed and the expected effect size is moderate therefore requiring 64 African
Americans and 64 Euro-Americans since the desired alpha level is .05 and the desired

power = .80.

For hypotheses 2 & 5: The number of independent variables in the moderated regression
are 3 and with an anticipated moderate effect size therefore requiring 76 African '
Americans and 76 Euro-Americans since the desired alpha level is .05 and the desired

power = .80.

For hypotheses 3a & 3b: The number of independent variables in the moderated
regression are 7 and with an anticipated moderate effect size therefore requiring 102
African Americans and 102 Euro-Americans since the desired alpha level is .05 and the

desired power = .80.

For hypothesis 4: The anticipated effect size is a medium to large correlation therefore
requiring between 28 and 85 subjects per group since the desired alpha level is .05 and
the desired power = .80.

Consequently, in order to perform all necessary analyses with power = .80 and an alpha

level = .05, the necessary sample size would be at least 204.

81

APPENDIX B

Demographic Questionnaire

Circle your responses or fill in your response in the

 

appropriate blank.

Race

1 = African American 2 = White 3 = Other
Gender

1 = Male 2 = Female
Age
Year in College

1 = First-year 3 = Third Year

2 = Second Year 4 = Fourth Year

SAT score (Highest score possible 1600)

ACT score (Highest score possible 36)

gpa

 

82

APPENDIX C

 

 

Individualism-Collectivism Measure
Mark the number on your answer sheet that most corresponds

to how much you agree with the statement.

1 2 3 4 5
Strongly Neutral Strongly
Disagree Agree

 

. Only a person who depends on him/herself gets ahead in life

2. To be superior a person must stand alone

3. If you want something done right, you have to do it yourself

4. What happens to me is my own doing

5. In the long run the only person you can count on is yourself

6. Winning is everything

7. I feel that winning is important in both work and games

8. Success is the most important thing in life

9. It bothers me when other people perform better than me

83

 

 

 

Mark the number on your answer sheet that most corresponds

to how much you agree with the statement.

1 2 3 4 5
Strongly Neutral Strongly
Disagree Agree

 

 

10. Doing your best isnt good enough; it is important to win

11. I prefer to work with the group rather than working alone

12. Given the choice, I would prefer a job where I could alone rather than in a group.

13. Working with a group is better than working alone

14. People should be made aware that if they are going to be part of a group then they are
sometimes going to have to do things they don't want to do

15. People who belong to a group should realize that they’re not always going to get what
they personally want

16. People in a group should realize that they sometimes are going to have to make
sacriﬁces for the sake of the group as a whole

17. People in a group should be willing to make sacriﬁces for the sake of the group’s
well-being

18. A group is more productive when its members do what they want to do rather than
doing what the group wants them to do

19. A group is most efﬁcient when its members do what they think is best rather than
doing what the group wants them to do

20. A group is more productive when its members follow their own interests and

CODCCI’DS

84

APPENDIX D

 

Awareness of and Belief in Difference Items

 

 

Mark the number on your answer sheet that most corresponds

to how much you agree with the statement.

1 2 3 4 5
Strongly Neutral Strongly
Disagree Agree

 

 

1. I am aware that African Americans score lower on intelligence tests than Caucasians.

2. I am aware that research has found Caucasians to score 15 points higher than African

Americans on intelligence tests.

3. I have read reports that African Americans do not do as well as Caucasians on

intelligence tests.

4. Caucasians are smarter than African Americans.

5. Caucasians do better on intelligence tests than African Americans.

6. African Americans possess less intellectual ability than Caucasians.

8S

APPENDIX B

General Directions

Use the pencil provided to indicate your answer to each question on the answer
sheet provided and use the scratch paper provided for any ﬁguring you need to do. You
should ﬁll in one answer and only one answer. If you mark more than one answer for any
question, you will receive no credit for that question. If you change your mind about the
answer to a question, completely erase your ﬁrst answer and ﬁll in the letter for the other

answer.

Example item:

A square has:

a) two sides

b) three sides

c) four sides

d) ﬁve sides

e) six sides

The correct answer is "c. " A square has four sides. Therefore, the answer "c"
should be ﬁlled in.

Work as rapidly and accurately as you can. Do not spend too much time on any
one question. If you ﬁnish a part of the test before time is called, you may check over
your work. .

You will not be penalized for guessing. If you are unsure of the answer to a

question, select the response that you think is best.

86

APPENDIX F

Verbal Analogies

Directions: This test measures your ability to reason and see relationships between
words. You are to choose the answer that best completes the analogy developed at the

beginning of each question.

This test consists of 25 questions. Fill in the appropriate letter on your answer sheet. If
you complete all the questions before the allotted time has elapsed, you may go back over

this test.

N T NT TI-IENEXTPA E NTILY ARET LDT D S

87

APPENDIX G

Air Force foicer Qualifﬂng Test (AFOQT) Form 0: verbal analogies subtest

The verbal analogies subtest measures your ability to reason and see relationships

between words. There are a total of 25 questions.

Example item I:
- A word is to letter as zip code is to
a) number
b) sign
c) symbol
d) address

The answer is "a. " A word is to letter as zip code is to number. Therefore, the

letter "a" should be ﬁlled in.

Example item 2:
Applicant is to employee as
a) customer is to salesperson
b) distributor is to competitor
c) pledge is to member
d) secretary is to engineer

The answer is "c. " An applicant is to employee as pledge is member. Therefore,

the letter "c" should be ﬁlled in.

88

Example item 3:
Condemn is to praise as
a) evaluate is to judgment
b) peripheral is to critical
c) success is to advantage
d) discern is to represent

The answer is "d. " Condemn is to praise as discern is to represent. Therefore,

the letter "d" should be ﬁlled in.

Example item 4:
Purpose is to direction as
a) stamina is to endurance
b) transition is to illustrate
c) eﬁ‘iciency is to quality
d) repetition is to achievement
The answer is "a. " Purpose is to direction as stamina is to endurance. Therefore,

the letter "a" should be ﬁlled in.

Example item 5:
Company is to merger as
a) auditor is to ﬁnance
b) monopoly is to takeover
c) person is to partnership
d) leader is to group
The answer is letter "c. " Company is to merger as person is to partnership.

Therefore, the letter "c" should be ﬁlled in.

89

 

APPENDIX H

Self-Efficacy Measure for AFOQT Verbal Analogies Subtest

 

Mark the number on your answer sheet that most corresponds

to how much you agree with the statement.

1 2 3 4 5
Strongly Neutral Strongly
Disagree wee

 

 

 

1. I am conﬁdent that I will do well on the verbal analogies subtest.
2. I will earn one of the top scores on the verbal analogies subtest.
3. I believe that I will have no problems on the verbal analogies subtest.

4. Compared to other subjects, I will perform well on the verbal analogies subtest.

 

l 2 3 4 5
20% 40% 60% 80% 100%

 

 

 

5. What percentage of the 25 verbal analogy problems do you think you will answer

correctly?

90

APPENDIX I

Math Knowledge

Directions: This test measures your ability to use learned mathematical relationships.
Each problem is followed by ﬁve possible answers. Decide which one of the ﬁve
answers is most nearly correct, and mark your answer on the separate answer sheet. Use

the scratch paper provided for any ﬁguring you need to do.
This test consists of 25 questions. Fill in the appropriate letter on your answersheet. If
you complete all the questions before the allotted time has elapsed, you may go back over

this test.

DO NOT GO ON TO THE NEXT PAGE UNTIL YOU ARE TOLD TO DO SO

91

APPENDIX J

Air Force Officer Qualifying Test (AFOQT) Form O: math knowledge subtest

The math knowledge subtest measures your ability to use learned mathematical

relationships. There are a total of 25 questions.

Example item 1:
Ifa=b+4andb=3a+6, then

a)a=-9
b)a=7

c)a=13
d)a=-5

The correct answer is "d. " If a = b + 4 and b = 30 + 6, then a = -5. Therefore,

letter "(1" should be ﬁlled in.

Example item 2:
If 3x-2 = 7, then 4x =
a)12
b) 5

C)2_0
5

d)9

The correct answer is "a. " If 3x-2 = 7, then 4x = 12. Therefore, the letter "a"

should be filled in.

92

Example Item 3:

(19-18-17-16) - (20-19-18-17) =
a) ~36
b) 2
c) -4
d)!

The correct answer is letter "b. " (19-18-17-16) - (20-19~]8-1 7) = 2. Therefore,
the letter "b" should be ﬁlled in.

Example Item 4:

Which one of the following is equal to 1 of 0.01 percent?
4

a) 0.000025
b) 0.00025
c) 0.0025
d) 0.025

The correct answer is "a. " One fourth of 0.01 percent = 0.000025. Therefore,
the letter "a" should be ﬁlled in.

Example Item 5:

Which of the following is a solution to x + x2 = 1 .7
a) -1
b) I
C) l
2
d) None of the above

The correct answer is "d. " The solution to x + x2 = I is "none of the above. "
Therefore the letter "d" should be ﬁlled in.

93

APPENDIX K

Self-Efficacy Measure for AF OQT Math Knowledge Subtest

 

 

 

 

Mark the number on your answer sheet that most corresponds

to how much you agree with the statement.

1 2 3 4 * 5
Strongly Neutral Strongly
Disagree Agree

 

 

1. I am conﬁdent that I will do well on the math knowledge subtest.
2. I will earn one of the top scores on the math knowledge subtest.
3. I believe that I will have no problems on the math knowledge subtest.

4. Compared to other subjects, I will perform well on the math knowledge subtest.

 

l 2 3 4 5
20% 40% 60% 80% 100%

 

 

5. What percentage of the 25 math knowledge problems do you think you will answer

correctly?

94

LIST OF REFERENCES

Arroyo, C. G. & Zigler, E. (1995). Racial identity, academic achievement, and the
psychological well-being of economically disadvantaged adolescents. Journal of

Personality and Social Psychology, 69, 903-914.

Bandura, A. (1977). Self-efﬁcacy: Toward a unifying theory of behavioral change.
Psychological Review. 84. 191-215.

Bandura, A. (1982). Self-efﬁcacy mechanism in human agency. American

Psychologist, 37(2), 122-147.

Bandura, A. (1993). Perceived self-efficacy in cognitive development and

functioning. Educational Psychologist, 28(2), 117-148.

Bandura, A., & Adams, N. E. (1977). Analysis of self-efﬁcacy theory of
behavioral change. Cognitive Therapy and Research. 1(4). 287-310.

Bemal, E. M. (1984). Bias in mental testing: Evidence for an alternative to the
heredity-environment controversy. In C. R. Reynolds, & R. T. Brown (Eds.), Perspectives
on Bias in Mental Testing. (pp. 171-188). New York, NY: Plenum Press.

Bersoff, D. N. (1981). Testing and the law. American Psychologist, 36(10), 1047-
1056.

Blaine, B., Crocker, J., & Major, B. (1995). The unintended negative

consequences of sympathy for the stigmatized. Journal of Applied Social Psychology, 25,
889-905.

Bobo, L. (1988). Group conﬂict, prejudice, and the paradox of contemporary
racial attitudes. In P. A. Katz, & D. A. Taylor (Eds). Eliminating Racism: Proﬁles in
Controversy. (pp. 85-114). New York, NY: Plenum Press.

Bouffard-Bouchard, T. (1989). Inﬂuence of self-efﬁcacy on performance in a
cognitive task. The Joumgl of Social Psychology, 130(3), 353-363.

Cervone, D. (1989). Effects of envisioning future activities on self-efﬁcacy
judgments and motivation: An availability heuristic interpretation. Cogpitive Therapy and

Research, 131 3 1, 247-261.

Cervone, D., & Wood, R. (1995) Goals, feedback, and the differential inﬂuence of
self-efﬁcacy processes on cognitively complex performance. Cognitive Therapy &
Research 19 519-545.

 

95

96

Chan, D., Schmitt, N., Deshon, R. P., Clause, C. S., & Delbridge, K. (1997).
Reactions to cognitive ability tests: The relationships between race, test performance, face
validity, and test-taking motivation. Journal of Applied Psychology. 82. 300-310.

Cooper, S. E., & Robinson, D. A. G. (1991). The relationship of mathematics self-
efﬁcacy beliefs to mathematics anxiety and performance. Measuremen_tand Evaluation in
Counseling and Development. 24. 4-11.

Crane, J. (1994). Exploding the myth of scientiﬁc support for the theory of Black
intellectual inferiority. Journal of Black PsychologyL20(2). 189-209.

Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled
components. J oum_2_tl of Personality and Social Psychology, 56, 5-18.

Earley, P. C. (1994). Self or group? Cultural effects of training on self-efﬁcacy
and performance. Administrative Science Quartery. 39. 89-117.

Eden, D., & Kinnar, J. (1991). Modeling galatea: Boosting self-efﬁcacy to
increase volunteering. Journal of Applied Psychology, 76, 770-780.

Erez, M., & Barley, P. C. (1993). Culture, Self-identity, and Work. New York:
Oxford University Press.

Eyring, J. D., Steele Johnson, D., Francis, D. J. (1993). A cross-level units-of-
analysis approach to individual differences in skill acquisition. Journal of Applied

Psychology, 78, 805-814.

Eysenck, H. J. (1984). The effect of race on human abilities and mental test
scores. In C. R. Reynolds, & R. T. Brown (Eds), Perspectives on Bias in Mental Testing.
(pp.249-292). New York, NY: Plenum Publishing Corporation.

Fordham, S. & Ogbu, J. V. (1986). Black students’ school success: "Coping with
the burden of ’acting white’." The Urbpan_ Review. 18. 176-206.

Gaines, S. O. (1994). Generic, stereotypic, and collectivistic models of
interpersonal resource exchange among African American couples. Journal of Black

Psychology, 20(3), 294-304.

Gecas, V. & Schwalbe, M. L. (1983). Beyond the looking-glass self: Social
structure and efﬁcacy-based self-esteem. Social Psychology Quarterly, 46, 77-88.

Gist, M. E., & Mitchell, T. R. (1992). Self-efﬁcacy: A theoretical analysis of its
determinants and malleability. Academy of Management Review, 17(2), 183-211.

97

Gottfredson, L. S. (1988). Reconsidering fairness: A matter of social and ethical
priorities. Journal of Vocational Behavior. 33. 293-319.

Graham, S. (1994). Motivation in African Americans. Review of Educational
Research. 64(1). 55-117.

Grubb, H. J. & Ollendick, T. H. (1986). Cultural-distance perspective: An
exploratory analysis of its effect on learning and intelligence. International J oumam
Intercultml Relations. 10. 399-414.

Harrington, G. M. (1984). An experimental model of bias in mental testing. In C.
R. Reynolds & R. T. Brown (Eds.), Perspectives on Bias in Mental Testing, (pp.101-138).
New York, NY: Plenum Publishing Corporation.

Harris, A. R., & Stokes, R. (1978). Race, self-evaluation, and the Protestant
ethnic. Social Problems 26 71-85.

 

Heiss, J., & Owens, S. (1972). Self-evaluations of Blacks and Whites. American
Journal of Sociology, 78, 71-85.

Helms, J. E. (1992). Why is there no study of cultural equivalence in standardized
cognitive ability testing? American Psychologist, 47(9), 1083-1101.

Hogg, M. A. & Turner, J. C. (1987). Intergroup behaviour, self-stereotyping and
the salience of social categories. British Journal of Social Psychology, 26, 325-340.

Hughes, M., & Demo, D. H. (1989). Self-perceptions of Black Americans: Self-
esteem and personal efﬁcacy. American Journal of Sociology, 95, 132-159.

Hui, C. H. (1988). Measurement of individualism-collectivism. Journal of
Research in Personality. 22. 17-36.

Hui, C. H., & Triandis, H. C. (1986). Individualism-collectivism: A study of
cross-cultural researchers. Journal of Cross-Cultural Psychology, 17(2), 225-248.

Humphreys, L. D. (1971). Theory of intelligence. In R. Cranco (Ed) Genetic and
Environmental Inﬂuences, New York: Grune & Stratton.

Hunt, J. G., & Hunt, L. L. (1977). Racial inequality and self-image: Identity
maintenance and identity diffusion. Sociology and Social Research. 6L 539-559.

Hunter, J. E., & Hunter, R. F. (1984). Validity and utility of alternative predictors
of job performance. Psychological Bulletin, 96(1), 72-98.

Jensen, A. R. (1980). Bias in Mental Testing. New York, NY: Free Press.

98

Jensen, A. R. (1984). Test bias: Concepts and criticisms. In C. R. Reynolds & R.
T. Brown (Eds.), Perspectives on Bias in Mental Testirg (pp.507-586). New York, NY:
Plenum Publishing Corporation.

Krause, N. (1983). The racial context of Black self-esteem. Social Psychology
Quarterly, 46, 98-107.

Locke, E. A. (1991). The motivation sequence, the motivation hub, and the
motivation core. Organizational Behavior and Human Decision Processes. SQ 288-299.

Locke, E. A., Frederick, E., Lee, C., Bobko, P. (1984). Effect of self-efﬁcacy,

goals, and task strategies on task performance. Journal of Applied Psychology, 69(2),
241-251.

Loehlin, J. C., Lindzey, G., & Spuhler, J. N. (1975). Race differences in
intelligence. R. C. Atkinson, J. Freedman, Lindzey, & R. F. Thompson (Eds.), San
Francisco, CA: W. H. Freeman and Company.

Lorr, M. & Wunderlich, R. A. (1986). Two objective measures of self-esteem.
Journal of Personality Assessment, 50, 18-23.

Lynn, R., & Owens, D. (1993). Spearrnan’s hypothesis and test score difference
between Whites, Indians, and Blacks in South Africa. The Journal of General Psychology,

121(1), 27-36.

Manning, W. H., & Jackson, R. (1984. College entrance examinations: Objective
selection or gatekeeping for the economically privileged. In C. R. Reynolds, & R. T.
Brown (Eds.), Perspectives on Big in Mental Testing. (pp.189-220). New York, NY:
Plenum Press.

Markus, H., & Kitayama, S. (1991). Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98, 224-253.

Martinez, R. & Dukes, R. L. (1987). Race, gender and self-esteem among youth.
Hispanic Journal of Behﬂoral SciencesL9. 427-443.

Matsui, T., Ikeda, H., & Ohnishi, R. (1989). Relations of sex-typed socializations
to career self-efﬁcacy expectations of college students. Journal of Vocational Behgion .
3;, 1-16.

Multon, K. D., Brown, S. D., & Lent, R. W. ( 1991). Relation of self-efﬁcacy
beliefs to academic outcomes: A meta-analytic investigation. Journal of Counseling

Psychology, 38(1), 30-38.

99

Osborne, W. L. & LeGette, H. R. (1982). Sex, race, grade level and social class
differences in self-concept. Measurement and Evaluation in Guidance. 144 195-201.

Oyserrnan, D., Gant, L., & Ager, J. (1995). A socially contextualized model of
African American identity: Possible selves and school persistence. Journal of Personality

and Social Psychology, 69(6), 1216-1232.

Pajares, F., & Kranzler, J. (1995). Self-efﬁcacy beliefs and general mental ability
in mathematical problem-solving. Contemporary Educational Psychology. 20. 426-443.

Peterson, B., & Ramirez, M. (1971). Real-ideal self-disparity in Negro and
Mexican American children. Psychology, 8, 22-26.

Pettigrew, T. F., & Martin, J. (1987). Shaping the organizational context for Black
American inclusion. Journal of Social Issues. 43(1). 41-78.

Phinney, J. S. (1996). When we talk about American Ethnic groups what do we
mean? American Psychologist, 51, 918-927.

Pintrich, P. R., & DeGroot, E. V. (1990). Motivational and self-regulated learning
components of classroom academic performance. Journal of Educﬁatiorpl Psychology, 82,
33-40.

Porter, J. R. and Washington, R. E. (1979). Black identity and self-esteem: A
review of studies on Black self-concept 1968-1978. Annual Review of Sociologyé, 53-
74.

Rosenberg, M. (1979). Conceiving the self. New York: Basic.

Rosenberg, M. (1981). The self-concept: Social product and social force. In M.

Rosenberg and R. H. Turner (Eds.), Social Psychology: Sociological Persppctives, p. 593-
624. New York; Basic.

Rosenberg, M. , & Simmons, R. G. (1972). Black and White Self-Esteem: The
Urbﬂ School Child. Rose Monograph series. Washington, D. C.: American Sociological
Association.

Ruvolo, A. P., & Markus, H. R. (1992). Possible selves and performance: The
power of self-relevant imagery. Social Cogpition, 10(1), 95-124.

Saragcoglu, B., Minden, H., & Wilchesky, M. (1989). The adjustment of students
with learning disabilities to university and its relationship to self-esteem and self-efﬁcacy.
Journal of Learning and Disabilities, 22, 590-592.

100

Schmidt, F. L. (1988). The problem of group differences in ability test scores in
employment selection. Journal of Vocational Behavior. 33. 272-292.

Schmidt, F. L., Greenthal, A. L., Hunter, J. B., Berner, J. G., & Seaton, F. W.
(1977). Job sample vs. paper-and-pencil trades and technical tests: Adverse impact and
examinee ~.; .1des. Personnel Psychology, 30, 187-197.

Schmitt, N ., Clause, C. S., & Pulakos, E. D. (1996). Subgroup differences
associated with different measures of some common job-relevant constructs. In C. L.
Cooper, & I. T. Robertson (Eds.), International Review of Indusﬂ and Organizational
Psychology (Vol. 11, pp. 115-139). John Wiley & Sons Ltd.

Schmitt, N ., Gooding, R. Z., Noe, R. A., & Kirsch, M. (1984). Metaanalyses of
validity studies published between 1964 and 1982 and the investigation of study
characteristics. Personnel Psychology, 37, 407-422.

Schunk, D. H. (1984). Self-efﬁcacy perspective on achievement behavior.
Educational Psychologist, 19, 48-58.

Schunk, D. H. (1991). Self-efﬁcacy and academic motivation. Educational

Psychologist, 26 (3&4), 207-231.

Seymour, R. T. (1988). Why plaintiffs’ counsel challenge tests, and how they can
successfully challenge the theory of "validity generalization." Journal of Vocaﬂonal
Behavior. 33. 331-364.

Skinner, J. & Ree, M. J. (1987). Air Force Ofﬁcer Qualifying Test (AFOQT):
Item and Factor Analysis of Form 0. Brooks Air Force Base, Texas: Manpower and
Personnel Division.

Stanley, K. D. & Murphy, M. R. (1997). A comparison of general self-efﬁcacy
with self-esteem. Genetic Social and General Psychology Monoggaphs, 123, 81-99.

 

Steele, C. & Aronson, J. (1995). Stereotype threat and the intellectual test
performance of African Americans. Joumal of Personality and Social Psychology, 69,
797-8 1 1 .

Tajfel, H. (1978). Differentiation between social groups: Studies in the social
psychology of intergrgoup relations. London: Academic Press.

Tajfel, H. (1981). Human ggoups and social categories: Studies in social
psychology. Cambridge, England: Cambridge University Press.

101

Tashakkori, A., & Thompson, V. D. (1991). Race differences in self-perception
and locus of control during adolescence and early adulthood: Methodological
implications. Genetic. Social-and General Psychology Mongraphs. 117(2). 135-152.

Tashakkori, A., Thompson, V. D., Wade, T. J ., & Valente, E. (1989). A
longitudinal analysis of sex by race differences in predictors of adolescent self-esteem.
Personality and Individual Differences. 10. 717-729.

Triandis, H. C. (1989). The self and social behavior in differing cultural contexts.
Psychological Review. 96. 506-520.

Triandis, H. C., Bontempo, R., Villareal, M. J., Asai, M., & Lucca, N. (1988).
Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships.
Journal of Personality and Social Lalchology. 54(2). 323-338.

Tuckman, B. W., & Sexton, T. L. (1990). The relation between self-beliefs and
self-regulated performance. Journal of Social Behavior and Personality, 5(5), 465-472.

Turner, M. E., Pratkanis, A. R., & Hardaway, T. J. (1991). Sex differences in
reactions to preferential selection: Towards a model of preferential selection as help.
Journal of Social Behavior and Personality, 6(4), 797-814.

Turner, C. B., & Turner, B. F. (1982). Gender, race, social social class, and self-
evaluations among college students. The Sociological Quarterly, 23, 491-507.

Wade, T. J ., Thompson, V. D., Tashakkori, A., & Valente, E. (1990). Structure
and stability of self-esteem in late teens. Personality and Individual Differences, 11, 885-
893.

Wagner, J. A. (1995). Studies of individualism-collectivism: Effects on
cooperation in groups. Academy of Management Journal, 38(1), 152-172.

Weinberg, R., Gould, D., & Jackson, A. (1979). Expectations and performance:

An empirical test of Bandura’s self-efﬁcacy theory. Journal of Sport Psychology, 1, 320-
331.

White, J. L. & Parham, T. A. (1990). The psychology of Blacks: An African-
American pprsmctive (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall.

Whittler, T. E., Calatone, R. J ., & Young, M. R. (1994). Strength of ethnic
afﬁliation: Examining Black identiﬁcation with Black culture. The Journal of Social

Psychology, 131, 461-467.

Wollack, S. (1994). Confronting adverse impact in cognitive examinations. Public
Personnel Management, 23(2), 217-224.

 

102

Wood, R., & Bandura, A. (1989). Social cognitive theory of organizational
management. Academy of Management RevieLM. 361-384.

Wood, R. E., George-Falvy, J ., & Debowski, S. Motivation and information
search on complex tasks, unpublished manuscript.

Woodruff, S. L. & Cashman, J. F. (1993). Task, domain, and general self-efﬁcacy;
A re-examination of the self-efﬁcacy scale. Psychological Reports, 72, 423-432.

Zakay, D. & Glicksohn, J. (1992). Overconﬁdence in a multiple-choice test and its
relationship to achievement. The Psychological Record. 42. 519-524.

Zimmerman, B. J ., Bandura, A., Martinez-Pons, M. (1992). Self-motivation for
academic attainment: The role of self-efﬁcacy beliefs and personal goal setting. American

Educational Research Journal, 29(3), 663-676.