A meta-analysis of interventions to modify stereotypes about African Americans

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content

A META-ANALYSIS OF INTERVENTIONS TO MODIFY

STEREOTYPES ABOUT AFRICAN AMERICANS

by

Sidalia Garrett Reel

___________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION

May 2007

Copyright 2007 Sidalia Garrett Reel

ii
DEDICATION

I dedicate this dissertation to my late husband who encouraged me to
embark on this journey and passed away the weekend the doctoral program began,
and to my mother who spent her last days with me during the final stages of
writing. I hope that I have instilled in my three sons the value of education, the
importance of a supportive network of family, friends and colleagues, and the
perseverance it takes to achieve your goals.

iii
ACKNOWLEDGEMENTS

I gratefully acknowledge Dr. Richard Clark for his unwavering support,
encouragement and expert guidance during this process. His faith in me kept me
going. Thanks go to Dr. Jeanne Farrington and Dr. Keith Howard for being on my
committee and for being so supportive and helpful. I also want to thank my family,
friends and my colleagues in the USC Northern Cohort for believing in me, and
giving me the will and energy to complete this program.

iv
TABLE OF CONTENTS

DEDICATION ii
ACKNOWLEDGEMENTS iii
LIST OF TABLES vi
ABSTRACT vii
CHAPTER 1: INTRODUCTION 1
Statement of the Problem 4
Purpose 6
Goals of This Study 7
Research Questions 8
Importance of the Study 8
Methodological Overview 10
Delimitations 11
Limitations 11

CHAPTER 2: LITERATURE REVIEW 13
Section One: Definitions 13
Section Two: Review of Automated Stereotyping 17
Activating Stereotypes 19
Stereotype Application 25
Section Three: Racial Stereotyping 31
Interrupting Racial Stereotypes 32
Prejudice and Discrimination 34
Racial Stigma 36
Stereotype Threat and Vulnerability 38
Coping Strategies for Dealing with Stigma 44
Section Four: Explicit and Implicit Measures of Automated
Stereotyping 47
Explicit Measures of Automated Stereotyping 47
Implicit Measures of Automated Stereotyping 50
MODE Model 51
Implicit Association Test (IAT) 53
Response to Social Structure 61
Preference for Implicit Measures 63
Section Five: Interventions to Moderate the Effects of
Implicit Stereotyping 63
Interethnic Friendship 63
Social Influence 64
First Impressions 65
Shifting Goal Intentions 65
Positive Priming 66
Mental Imagery Prime 67

v
Consequences of Mood on Stereotyping 67
Diversity Education 68
Wise Strategies 69
Jigsaw Classroom 70
Micro-Inequities Training 71
Meta-Analyses and Reviews on Mediating Factors 72
Meta-Analysis Findings on Factors to Mediate
Stereotype Activation 73
Implicit Self-Related Processes 76
Meta-Analysis of the Correlation between the IAT
and Explicit Measures 78
Summary of Moderating Factors 81

CHAPTER 3: METHODS 83
Procedure 83
Criteria for Inclusion/Exclusion 84
Data Sources 85
Study Design Factors 86
Statistical Analysis 87

CHAPTER 4: RESULTS 88
Overview 88
Additional Descriptive Statistics 91
Summary of Study Mean Effect Size 91
Effects of Moderating Intervention Category 92
Frequencies for Reducing the IAT Effect with Intervention 93

CHAPTER 5: DISCUSSION 95
Overview 95
Research Question Findings 98
Summary & Conclusions 104
Summary 104
Conclusions 106
Considerations for Future Research 107

REFERENCES 109

vi
LIST OF TABLES

Table 1: Data Sources 85
Table 2: Coding Sheet Description 86
Table 3: Meta-Analysis Studies Overview 88
Table 4: Descriptive Statistics for the IAT Effect 91
Table 5: Summary of Study Mean Effect Size 92
Table 6: Intervention Category Mean Effect Size 93
Table 7: Frequencies for % Reduction in IAT Effect with Intervention 93
Table 8: Descriptive Statistics - % Reduction in IAT Effect with
Intervention 94
Table 9: Studies Reporting Reliability of Implicit and Explicit Measures 99
Table 10: Stereotype Prevention Interventions for Corporate Settings 102
Table 11: Stereotype Prevention Interventions That Will Not Work in a
Corporate Setting 103

vii
ABSTRACT

Since 1995, the Implicit Association Test (IAT) has been employed to
reveal unconscious stereotyping. This meta-analysis examined studies that used the
IAT in combination with treatments to eliminate automated stereotyping of African
Americans and to determine which treatments were effective in corporate work
settings. Of 134 studies identified only 43 specifically addressed Black-White
stereotyping and only 6 studies met the minimum criteria for inclusion in the meta-
analysis (i.e., the use of the Black-White IAT and an intervention to reduce racial
stereotyping.) Results reveal that interventions produced a moderate overall effect
size of .48. The small number of studies prevents clear conclusions, yet it appears
that existing African American stereotype-inhibiting interventions can reduce the
effects of stereotypes by approximately 15 percent. None of the studies were
conducted in work settings, but the most effective treatments, i.e., positive
interethnic social contact and exposure to affirmative African American images,
could be employed in work settings.

1
CHAPTER 1. INTRODUCTION
This meta-analysis seeks to examine studies conducted in the past decade
(1995-2005) where treatments attempted to reduce or eliminate the activation and
application of African American stereotypes. It will examine the effectiveness of
three categories of treatments that reflect a number of different strategies that have
been employed to reduce or eliminate the negative effects of African American
stereotypes, including:
Category 1: Social exposure. This category includes interethnic friendship,
social influence, manipulation of social context, and first impressions.
Category 2: Covert psychology, including shifting goal intentions, positive
expectancy, mental imagery, and mood alteration.
Category 3: Education and training, which includes diversity education,
positive role modeling, and specific training methodologies such as the WISE
strategy, the jigsaw classroom, and micro-inequities training.
Despite progress toward an egalitarian and multicultural society in the
United States, prejudice, i.e., judging people based on the color other skin or other
human characteristics, continues to exist in American culture. More often than not,
prejudice is a negative response to a stereotype about an individual or a group, and
stereotyping is a categorization process that draws upon a mental model of a group
according to race, gender, age, physical size, etc. that is held in common by
members of another group (Fiske, 1998).

2
Ever since Allport’s 1954 seminal book, The Nature of Prejudice, helped to
define the current thinking about stereotypes, prejudice and discrimination,
prominent social psychologists, as well as cognitive psychologists and other
researchers have worked to find ways to reduce the negative impacts of these
concepts. Several theorists believe that prejudice is the inevitable consequence of
stereotyping, and that stereotyping is simply an automated categorization process
(Allport, 1954; Erlich, 1973; Hamilton, 1981; Taijfel, 1981). Some research has
even shown that racial stereotypes are embedded in children’s memories when they
are as young as three years old, even before they have developed any cognitive
ability to understand stereotypes (Aboud, 1988; Hirschfeld, 1996; Katz, 1976;
Proshansky, 1966).
The prevailing theories of the 1950s-1980s held that the level of prejudice
attitudes of individuals could be determined through self-report instruments that
measured prejudice reactions to groups of people. The Modern Racism Scale
(McConahay and Hough, 1976), the Motivation to Control Prejudiced Reactions
Scale (Fazio and Dunton, 1997), and the Linguistic Inter-group Bias procedure
(Maas, Milesi, Zabbini and Stahlberg, 1995) are some of the more commonly used
explicit instruments to measure stereotypes. The drawback of such measures is that
the subjects exposed to these instruments could predict the socially acceptable
answer and respond accordingly, rather than reveal their true feelings.
The ability to tap into the unconscious stereotypes and automatic prejudice
reactions became the subject of an extensive body of research that is centered on

3
implicit measures of prejudice. Research over the last 25 years reveals that
stereotypes are learned knowledge that is stored in long-term memory and then,
when triggered, become an unconscious mental process that is automatically
activated and then applied (Banaji, 1995; Devine, 1989a; Dovidio, 1986; Fazio,
1986; Greenwald, McGhee & Schwartz, 1998). Research that combines explicit
self-report instruments with the Implicit Association Test (IAT), or with other
implicit measures, do not consistently reveal a correlation between implicit and
explicit measures. However, more prejudiced responses are revealed on the
implicit methods than on the explicit self-report instruments (Aberson, Shoemaker
& Tomolillo, 2004; Cunningham, Preacher & Banaji, 2001; Devine 1989b;
Monteith, Voils and Ashburn-Nardo, 2001; Ottaway, Hayden and Oakes, 2001).
Even when the IAT is used for other subjects such as shyness (Asendorf, Bause and
Mucke, 2002) , self-esteem (Greenwald and Farnham, 2000) or rudeness (Bargh,
Chen & Burows, 1996), the IAT predicts spontaneous behaviors but does not
appear to be subject to conscious effort (Egloff and Schmulke, 2002).
Thus, the more difficult question to be answered is can the automaticity of
stereotyping be controlled or interrupted? Even in the face of socially unacceptable
responses that reveal prejudice, what prevents people from being more accepting of
differences in others?
Since 1995 when implicit measures of automated stereotyping came to
prominence (Fazio, Jackson, Dunton & Williams, 1995); Greenwald & Banaj,
1995), a majority of the studies focused on stereotypes about African Americans.

4
Eighteen IAT-based studies specifically examined reactions to Black and White
subjects, while there were twenty-two other IAT-based studies on topics ranging
from other racial pairings, gender, sexual orientation, immigrants, age, and physical
appearance. The focus of this study is on the modification of prejudiced responses
to negative stereotypes about African Americans.
Statement of the Problem
Research in the area of stereotyping is moving away from explicit,
controlled stereotype measurements such as self-report questionnaires, to stimulus-
response reaction observation instruments that measure automatically activated
implicit and hidden biases. The Implicit Association Test (IAT) is a computer-
based instrument that measures the response time when visual stimuli, such as
Black and White faces, are paired with positive or negative words. Responses are
slower when the words and the faces are incongruent (Greenwald, McGhee and
Schwartz, 1998). The PsycINFO database cites 522 references to IAT studies
based on stereotypes. The IAT (Fazio, Jackson Dunton & Williams 1995;
Greenwald, McGhee & Schwartz, 1998) and other implicit measures of prejudice
(Devine, 1989a; Dovidio & Gaertner, 1993; Graham, 1985) have shown results in
the laboratory environment that imply automated responses to racial or other
differences can be altered in two ways.
The first laboratory modification is to prime the experiment participants
with information about the appropriate response prior to taking the test. Eleven
studies used priming with the IAT. Notably, the introduction of an African

5
American experimenter to help administer the test impacted the results of both the
IAT and the explicit survey instrument(s) used (Lowery, Hardin and Sinclair, 2001;
McConnell and Liebold, 2000).
The second laboratory modification is to provide information about what
would have been the more appropriate or socially acceptable response after the IAT
is administered, and then re-administer the test (Dasgupta and Greenwald, 2001;
Leonardelli and Brewer, 2001; Lieberman, Ochsney, Gilbert and Schacter, 2001;
Steffens, 2004). While a reduction in the evidence of automated stereotype
responses was realized in both testing situations, for the most part, the stereotype
reductions relate to modifying the experimental methodology rather than modifying
the perceiver of the stereotype.
The problem this research will examine is whether automatic negative
stereotyping about African Americans can be modified by using interventions
outside of the laboratory. Racial stereotyping, specifically biased attitudes towards
African Americans is the focus of the study. Devine (1989a) explored the question
of whether or not the automatic activation of a stereotype is controllable. In her
research, Devine acknowledges the ability to control an individual’s beliefs about a
stereotype through disassociation, but that the stereotype cannot be eliminated from
memory. The inevitability of stereotypes as the cognitive component of prejudice
attitudes has been reinforced by other researchers (Bargh, 1999; Harding, Kutner,
Proshansky, & Chein, 1954). If a stereotype is learned and stored in memory, it
will be unconsciously activated whenever a member of the stereotyped group

6
appears. The possibility of interrupting this process is limited, as there is research
to suggest that even amnesiacs can possess an involuntary explicit memory that
allows the retrieval of information that was previously stored (Schacter and
Badgaiyan, 2001). Therefore, this study will focus on ways to modify the
application of a stereotype rather than stereotype activation.
Purpose
Presumably the unconsciously applied stereotypes improve the efficiency of
processing knowledge about individuals and groups. Stereotyping is viewed as a
social categorization process that saves time and mental energy (Macrae, Milne &
Bodenhausen, 1994). However, negative consequences occur when such
stereotypical representations about individuals or groups are not accurate.
Particularly as they relate to race, gender, sexual orientation and nationality/culture,
the targets of the stereotype experience discrimination and prejudice based on the
stereotype, i.e., the negative preconceived notions about individuals or groups lead
to, or become the rationale for, disparate treatment.
In corporate environments, the question of whether stereotyping can be
controlled is important because of the consequences of human resources decisions
such as hiring, promotions, compensation, rewards, and recognition, on those
individuals who are the targets of those decisions.
The consequences of working in corporate environments--where the
negative effects of stereotyping impact how employees contribute to the
organization and how they are rewarded and compensated for their contributions--

7
extort both a psychic and physical price on those who are the target of the
stereotypes. Steele (1997) introduced the concept of stereotype threat to describe
the consequences of stereotypes on intellectual identification and performance.
Studies on stereotype threat have shown how these stereotypes have been
internalized and impact stereotype targets including, but not limited to: women
(Schmader and Johns, 2003; Smith and White, 2002), African Americans (Cullen,
Hardison and Sackett, 2004; McKay, Doverspike, Bowen-Hilton, and McKay,
2003; Morgan, 2004; Steele, 1997; Steel and Aronson, 1995), the elderly
(Kawakami, Young and Dovidio, 2002) and obese individuals (Teachman,
Brownel, Gapinski & Rawling, 2003).
Goals of This Study
The goal of this study is to synthesize the research on the automatic
negative stereotyping of African Americans to determine if there is empirical
evidence of interventions that prevent people from acting on these stereotypes.
Studies that rely on the use of implicit measures, specifically the Implicit
Association Test (IAT), to measure the existence of the negative stereotypes, will
be the focus of the study, rather than explicit, self-report questionnaires or other
overt measures.
A secondary goal of this study is to propose a process for applying the
stereotype reduction interventions in the corporate work environment setting.
Human Resources practitioners will benefit from having methodologies to

8
recommend to the managers and organizations they support that can be applied to
solving issues related to prejudice and discriminatory acts in the workplace.
Research Questions
The key research question is what interventions can prevent people from
acting on negative stereotypes about African Americans? Can the automated racial
stereotyping process be interrupted? Related questions that are investigated
include:
1. Are unconscious, implicit measures of automated stereotyping more
reliable than conscious, explicit measures?
2. Can the application of stereotypes be interrupted once a stereotype is
automatically activated?
3. Can you automate the act of counteracting stereotyping as a performance
goal?
4. Which interventions are most effective in corporate or workplace
settings?
5. Is there evidence that interventions aimed at modifying stereotypes have
long-term effectiveness?
Importance of the Study
This meta-analysis will focus on ways to reduce automated racial
stereotyping, specifically the socially pervasive stereotypes about African
Americans in the United States. This study is important because it brings attention
to the need to identify effective methods for use in a corporate setting.

9
From the standpoint of practical value, knowledge about automated
stereotyping, and the proposed interventions to interrupt it, will contribute to
reducing the work environment consequences of negative stereotyping about
African Americans. The consequences for African Americans in the work
environment include exclusion or unfavorable outcomes in human resources
decisions such as hiring, work assignments, job performance evaluations,
promotions, compensation, reward and recognition, personal reputation, and
violations of a harassment-free work environment and nondiscrimination policies.
Each year, the turnover costs to corporations related to losing employees who do
not feel valued or experience prejudice and discrimination averages a half-year of
salary for hourly workers, 18 months of salary per manager or professional, and
incalculable losses in compensation for executives (Hay Group, 2005). Moreover,
the litigation costs associated with workplace harassment and discrimination
complaints are costly from a financial standpoint, and they are damaging to the
reputation of the company (Institute of Management and Administration, 2004).
This study contributes insights about the advances in the field of social
psychology by identifying ways to counteract stereotyping in work environments.
These interventions may also be applicable to other groups that are the target of
prejudice and discrimination, including other non-majority racial groups,
immigrants, women, sexual orientation, physical appearance or size, disability, and
age.

10
Methodological Overview
This study is a meta-analysis of the effectiveness of stereotype reduction
interventions. The IAT is used as an indicator of racial prejudice against African
Americans. The meta-analysis research approach is a systematic integration and
coding of research articles to provide a consistent, valid and replicable model
(Bland, Meurer and Maldonado, 1995). The validity of each research article will
be measured against the criteria identified for the meta-analysis. The most essential
outcome of meta-analysis is that the individual studies will combine to generate
effect sizes that are statistically significant as compared to the findings realized in
individual studies.
To set the context of the study, Chapter Two is an extensive review of
literature that provides key definitions that are important to the identification of
articles that will be included in the meta-analysis. A review of implicit measures of
stereotype application, with a focus on the Implicit Association Test is provided to
help to explain the baseline measure of implicit stereotyping. The framework of
stereotypes, prejudice and discrimination, in addition to the concepts of racism
stereotype threat and stereotype vulnerability will complete the literature review so
that both the perpetrator and the target of stereotypes is described in relation to the
studies included in the meta-analysis.
Chapter Three describes the research design and procedure for conducting
the research, and introduces the coding form to rate the research in terms of the

11
category of the intervention, IAT results, content validity, and the effectiveness of
the intervention.
Chapter Four summarizes the results of the meta-analysis, including a table
describing each of the studies that met the meta-analysis criteria, and the statistical
analysis of the studies with respect to the study goals and study design criteria.
The final chapter discusses the results and offers suggestions for future
research on interventions that reduce or prevent automated stereotyping, with the
intent that stereotype modification focused on negative stereotypes about African
Americans can be generalized to stereotypes targeting other groups.
Delimitations
The study will focus on modification of negative stereotypes about African
Americans. The modification interventions can be organized into three major
categories: social exposure, covert psychology, and diversity or stereotype
reduction education. Although the meta-analysis will examine the interventions to
modify stereotypes in terms of their utility in corporate settings, the findings should
generalize and be transferable to other non-corporate environments.
Limitations
Research that focuses on modifying stereotypes by interrupting their
application, and that is conducted outside of the experimental setting, is extremely
limited. Thirty-five empirical studies aimed at modifying racial stereotypes have
been identified. An additional 17 studies focus on overcoming stereotype threat,
which is defined as the negative impact a stereotype has on the stereotyped group

12
members (Steele, 1995). Studies that incorporate the IAT in the study design and
specifically address African Americans in at least one experiment reduce the
number of studies to six. Three additional studies include stereotype modification
in the design, and use other non-African American IAT versions. There are 21
other studies that include stereotype modification in the design, and use other
implicit techniques instead of the IAT. The proposed meta-analysis will use the six
studies that use some form of the IAT and also include a stereotype modification
intervention. The other modification interventions will be discussed in the review
of literature and, if applicable, in the discussion about future research.
Given the limited number of empirical studies, the benefit of the meta-
analysis with respect to effect size may be minimal. However, the assessment in
terms of method reliability, content validity, and outcome effectiveness will make
an important contribution to advancing research in this area.

13
CHAPTER 2. LITERATURE REVIEW
This chapter is a review of literature related to the automaticity of
stereotyping and attempts to moderate it. The chapter is developed in five sections.
The first section provides definitions of the key terms that are used in this study.
The second section is a review of automated stereotyping. Section three discusses
the specific nature of racial stereotyping and the impact on the target of the
stereotype by providing a review of racial stigma and stereotype threat. The fourth
section is a review of explicit and implicit measures to identify and interrupt
stereotyping, and elaborates on the studies and results using the Implicit
Association Test (IAT). The final section reviews the interventions to moderate the
effects of racial stereotyping, including a description of a meta-analysis on how
stereotype activation and a series of mediating factors impact behavior.
Section One: Definitions
Stereotyping
A cognitive categorization process that draws upon a mental model that
contains standardized mental picture of a group according to race, gender, age,
physical size, etc. held in common by members of one group about a group that is
different from them (Fiske, 1998). The affective, attitudinal manifestation of
stereotyping is prejudice and the behavioral result of stereotyping is discrimination.

14
Prejudice
A stereotype, once internalized leads to a feeling, favorable or unfavorable,
toward a person or thing, prior to, or not, based on actual experience. (Allport,
1954).
Discrimination
Discrimination is the negative behavioral responses of individuals in the in-
group to stereotypical thoughts and prejudiced attitudes about individuals in the
out-group (Fiske, 1998).
Racism
Racism is the use of racial stereotyping and prejudices that lead to hatred
and outward aggression in the most blatant forms (Allport, 1954; Allport &
Kramer, 1946; Harding, Kutner, Proshansky, & Chien, 1954). In its covert forms
racial stereotyping may cause discomfort, ambivalence and avoidance of members
of the group targeted by the stereotype (Allport, 1954).
Stereotype Activation
Stereotype activation is the “increased accessibility of the constellation of
attributes that are believed to characterize members of a given social category.”
(Wheeler & Petty, 2001). The stereotypes are learned and ingrained in long-term
memory through repeated activation, making spontaneous and unintentional
activation possible. (Devine, 1989b).
Stereotype activation is most often the result of visual stimuli, including,
but not limited to, race, gender, and age. Stereotypes may also be triggered by

15
situational circumstances such as the perceiver is in a hurry, distracted, cognitively
overloaded, or in an ambiguous situation (Fiske, 1998).
Stereotype Application
Once a stereotype is automatically activated, the behavioral response or
action taken is the application of the stereotype. The application may be a negative
or positive reaction to the individual or individuals in the stereotyped category.
Preventing stereotype application may occur when the perceiver is unable to
activate the stereotype due to the number of distractions competing for cognitive
resources (Blair & Banaji, 1996).
The behavioral consequences of stereotype activation are felt by individuals
who are members of the group being stereotyped. Research regarding stereotypes
has shown that stereotypes can be acted upon when the stereotypes are about others
(i.e., people who are different from you in terms of race, gender, age, academic
ability, athletic ability, etc.) Just as compelling is the body of research
demonstrating that self stereotypes (i.e., individuals who are members of a group
being stereotyped believe the stereotypes hold true for themselves) are activated
just as easily as stereotypes about others (Wheeler & Petty, 2001). Once
stereotypes are automated, whether self stereotypes, or other stereotypes, they can
be unconsciously activated by environmental stimulus or clues (Shiffrin & Dumais,
1981).

16
Stereotype Threat
A stereotype threat is the concern that others will judge one negatively due
to a stereotype about one’s group. The social-psychological threat of being
negatively stereotyped, or fearing that one exhibits the stereotype occupies working
memory and negatively impacts performance (Steele, 1997). Stereotype threat is
one of the factors attributed to African Americans lower test scores on academic
standardized tests (Aronson, Fried & Good, 2001).
Explicit memory
Consciously controlled processing is sometimes referred to as explicit
memory, wherein the individual recalls past experiences and brings them to
consciousness. Explicit memory requires awareness of the stimulus, and invokes
the motivation to respond, and the ability to make a choice about how to respond.
(Schacter & Badgaiyan, 2001).
Implicit memory
Automaticity, or automated memory processing, is unintentional and results
in responding unconsciously and spontaneously without awareness of the stimulus.
This unconscious, implicit form of memory is outside conscious awareness, and is
often referred to as priming (Blair & Banaji, 1996; Devine, 1989b; Fazio, Jackson,
Dunton & Williams; 1995; Fiske, 1998; Greenwald & Banaji, 1995; Tulving &
Schacter, 1990).

17
Implicit Association Test (IAT)
The IAT is a computer-based categorization tool that seeks to reveal
implicit bias by assessing the strength of associations between groups and traits.
The IAT requires participants to categorize two objects or social categories (e.g.,
White and African American, or insect and flower) that are paired with two
opposite and evaluative descriptive words (e.g., pleasant and unpleasant), measures
the response time, and concludes implicit preference for faster response times.
(Greenwald, McGhee & Schwartz, 1998). The expected, or stereotypical pairings
are responded to more quickly (e.g., flower-pleasant) than unexpected pairings
(e.g., youth-senile.)
Section Two: Review of Automated Stereotyping
The troubling aspect of stereotypes about people is that they are learned
early and programmed in to long-term memory. In its simplest terms, automated
processing refers to unconsciously performing a task that taps into long-term
memory without requiring the consciousness of short-term memory. Gagne (1993)
stated that nearly 70 percent of our knowledge is automated, thereby allowing our
conscious attention to focus on new and novel events. Working memory, which is
also referred to as short-term memory, holds newly acquired information for
conscious use, and functions as a mental workspace.
Research on the phenomenon of the automatic reaction to people perceived
to be different from one’s own group dates back to the 1940’s with the study of the
authoritarian personality in Europe as a response to the anti-Semitism that spawned

18
the Holocaust (Adorno, Frankel-Brunswik, Levinson & Sanford, 1950). The study
of reactions to people as a group as opposed to the individual differences in people
revealed that people process stereotypic information faster than they process
information about, or the anticipation of, individual differences in people (Allport,
1954).
In the 1960s, attribution theory gave rise to controlled “effortful” processing
to look at the cause and effect of situations in the environment. (Jones & Davis,
1965; Kelly, 1967; and Weiner, 1974). The 1970s brought challenges to the
effortful processing theory when the notion of the cognitive miser was introduced
to explain the need to rely on heuristics and stereotypes to reduce the amount of
mental resources needed to perform an increasing number of behaviors and
decision-making tasks (Langer, 1978; Langer & Abelson, 1972; Taylor & Fiske
(1978); and Dawes, 1976). By the late 1970s automated processing was
introduced (Taylor & Fiske, 1978) and subsequently applied to most social
psychological phenomena, including stereotyping (Brewer, 1988; Deax & Lewis,
1984; Dovidio, Evans & Tyler, 1986).
The automaticity of stereotyping posed a crisis in social psychology
because if stereotyping was deemed as uncontrollable, the consequences,
particularly the legal ones, could create problems in litigation when defendants
engaging in prejudiced or discriminatory behavior used the argument that their
behavior was automatic and unintentional. This dilemma, often referred to as the

19
Cognitive Monster (Bargh, 1999), led to a need to determine if stereotypes, once
activated, could be controlled.
Research on automated stereotyping has received considerable attention in
the last ten years. The fact that stereotyping people is an automated process raises
concerns because if stereotypes, once learned, cannot be eliminated from long-term
memory, can the process of internalizing stereotypes be altered? There has to be a
shared social reality about stereotypes in order for them to exist outside of the mind
of an individual. In both social and cognitive psychology, research on automated
processes has identified stereotyping as an automated mental process, and therefore
stereotypes are permanently stored in long-term memory.
Activating Stereotypes
At the core of understanding the impact of stereotyping is how stereotypes
are activated. There are two complementary perspectives on how stereotypes are
conceived. Stangor and Schaller (2000) proposed that either stereotypes are
represented within the mind of the individual, or they are part of the social fabric of
a society (p. 64). These perspectives were shaped by early work in the 1920s
wherein the individual (Allport, 1924) versus the collective (McDougall, 1920)
psychology approaches basically result in the same outcome because to have
meaning, stereotypes rely on a culturally-shared understanding of them.
Stereotypes can be viewed as a specific type of expectancy whereby
information processing occurs more rapidly when familiar stereotypical
information is presented (Hamilton, Sherman & Ruvolo, 1990). Stereotypes serve

20
as a shortcut to facilitate perception, encoding, retrieval and interpretation of
stereotypic information. The limiting factor of stereotypes, as described by
Greenwald and Banaji (1995) in the IAT development work to define what
constitutes implicit representations, is that they are unconscious traces of past
experience that are introspectively identifiable as being accurate or inaccurate.
However, studies involving the introduction of cognitive constraints have
demonstrated that stereotype activation can be slowed or interrupted (Hamilton &
Sherman, 1994) by introducing counter-stereotypic information.
In addition to the discussion of the origin of a stereotype as emanating from
the mind of the individual or the collective society, there is another distinction
about how stereotypes lead to prejudice. Devine (1989b) distinguishes between
knowledge of a cultural stereotype that is automatically and unconsciously
activated, and the conscious attention required to accept or endorse the stereotype
and act on it. The distinction between stereotype activation and stereotype
application is significant because it allows the possibility of suppressing a response
to a stereotype once it is activated.
Stereotype Targets
Racial stereotyping is the focus of this study. However, in the 134 articles
reviewed for this study, only 43 specifically address White-Black racial
stereotyping. Therefore, it is instructive to identify the similarities in the
considerable body of research conducted that addresses stereotypes about other
groups as well.

21
What follows is a description of research in stereotype activation that covers
stereotypes for five human trait categories: gender, physical appearance, age, and
racial stereotypes in athletic performance and racial stereotypes in academic
performance.
Gender Stereotypes
Socially-applied stereotypes that differentiate men from women are based
on long-held beliefs about prescribed gender roles, relationships and power. As the
traditional male and female roles have shifted with the significantly high number of
women in the workplace, the social significance of gender roles has expanded. In a
study of the use of stereotyping in teamwork, the researchers found an increased
amount of social loafing when larger numbers of women were on the team, i.e.,
team members were less motivated to optimize their own level of effort to
accomplish the task when there are women on the team to do the work. (Plaks &
Higgins, 2000). For example, two experiments using word primes of gender
stereotypes found that the gender stereotypes persist even when a gender-irrelevant
task is being performed (Banaji & Hardin, 1996).
Physical Appearance
Social psychologists have pondered the impact of stereotypes about
physical appearance from a number of perspectives. Attractiveness, as well as
physical size with respect to ideal height and/or weight, has been compared to the
ideal Western-European standards of beauty. Attractiveness and appearance have
been studied in terms of work-related consequences including, but not limited to,

22
career advancement in the form of promotions, pay raises, high profile
assignments, etc. For individuals in the out-group, such as obese individuals,
negative attitudes persist, despite attempts to counteract the stereotypes. For
example, in a study of obesity, three experiments to determine if reducing blame or
increasing empathy for overweight individuals reduced anti-fat attitudes, found that
these measures did not reduce the implicit negative responses to obese individuals.
(Teachman, Brownell, Gapinski & Rawlins, 2003).
Age
In two studies to determine if automatically activated negative stereotypes
about the elderly could be interrupted, the experimenters introduced primes about
elderly people and found the primes to slow decision-making on an unrelated
lexical task (Kawakami, Young & Dovidio, 2002). Other studies involving age
found the fastest reaction times when comparing older people to negative or non-
word responses (Dasgupta & Greenwald, 2001; Rothermund & Wentura, 2004).
Moreover, the IAT was used to determine if there are age differences in
results. Two studies employed the IAT and explicit self-report instruments to
measure age attitudes, identity, and self esteem. The results showed that the IAT
revealed differences between young and old participants, while the explicit
measures did not reveal age-related differences (Hummer, Garstka, Greenwald,
Mellott, & O’Brien, 2002).

23
Athletic Ability
Two studies showed that athletic performance can be impacted by negative
racial stereotypes about athletes. The researchers compared the performance of
African American participants who were told that performance of a golf task was a
diagnostic of sports intelligence, while White participants were told the golf task
was a diagnostic of natural athletic ability. For both groups of participants, the
negative racial stereotypes resulted in poorer performance (Stone, & Sjomeling,
Lynch, Darley, 1999).
Academic Ability
There is extensive documentation about girls and women not performing as
well as men and boys on standardized math tests (AAUW, 1992). Continuing into
college and beyond, the result is that women continue to avoid math and not
pursuing math-related careers ( LeFevre, Kulak & Heymans, 1992).
Recent studies to examine stereotype threat and the impact of negative self-
talk, demonstrate that stereotypes can have an impact on anyone who is the target
of a stereotype. In a study of the ways stereotypes are activated in math testing
situations and the how the activation impacts task performance, Smith & White
(2002) evaluated the performance of two groups: women compared to men, and
White men compared to Asian men. The tests were administered under three
different conditions: (1) implicit activation where nothing was stated about how
different groups might perform, but the societal pervasiveness of the stereotype
increased the likelihood that the stereotype threat would be present; (2) explicit

24
activation, where the women were told that men usually outperform women, and
the White men were told that Asians are better than Whites in math; and (3) a
nullified stereotype threat condition in which women were told there were no
differences in math performance between men and women.
The women performed equally with the men in the nullified condition but
less well when both the implicit and explicit activation of the stereotype about
lower math performance for women was present. The White men performed
poorly in the implicit and explicit stereotype threat conditions, and performed
better in the nullified stereotype condition. These findings support earlier research
in which White men performed poorly on a math test when they were told that they
would be compared to Asian men (Aronson Good, Keough, Steele & Brown,
1999).
In a series of experiments conducted by Spencer, Steele and Quinn (1999),
the research revealed that test difficulty moderates stereotype threat effects across
gender lines. Easy and difficult math tests were administered to both men and
women who were considered to be highly skilled in math. Women performed most
poorly on the tests labeled difficult. In addition, the relevance of a task to the
activated stereotype diminishes task performance. Inzlicht and Ben-Zeev (2000)
found that the only woman in a group being administered tests in several topical
areas performed poorly on the stereotype-relevant math test, but performed well on
the other subject tests.

25
With respect to African American students, an academic gap between
African American and White students on standardized test scores begins around the
sixth grade and fuels the negative stereotype about the academic underperformance
of African American students through post-graduate education (Gerard, 1983).
Research shows that the explicit activation of a stereotype negatively impacts
performance. In a study conducted by Steele and Aronson (1995), African
American students performed worse than White students when they were told the
test was a diagnostic of intellectual ability, but performed equally to White students
when they were told the task was not related to intelligence.
Stereotypes are activated in testing situations involving other groups as
well. The expectations of others, including teachers, have been shown to influence
test performance, including IQ tests (Rosenthal & Jacobson, 1968).
Summary of Stereotype Activation
What the implicit stereotype activation studies presented show is the
immutable nature of automated stereotypes. Once a stereotype is internalized,
stereotype activation is unavoidable, whether it is in the mind of the perceiver of a
stereotype about others, or it is in the mind of an individual who is a member of the
stereotyped group.
Stereotype Application
The evidence that an individual has internalized a stereotype is that upon
activating it, the stereotype is acted upon as a prejudiced attitude or discriminatory
behavior. Research about stereotypes has shown how stereotypes may influence

26
responses without the subject’s awareness of that influence (Banaji & Greenwald,
1995, Devine 1989b; Macrae, Milne & Bodenhausen, 1994). However, is it
automatic that stereotypes once activated, are automatically applied? What factors
may prevent acting on a stereotype once it is unconsciously activated? The role of
context and environment in automated stereotyping has some significant
implications that change the way in which stereotypes are applied and experienced.
Factors such as distractions, deliberate attention, ambiguity, task characteristics and
other situational influences on automated stereotyping are considered next.
Distraction
The extent to which an individual’s conscious attention is divided among
two or more cognitive processes creates a distraction for attending to any one
process. This cognitive overload has been shown to interrupt stereotype activation
(Gilbert & Hixon, 1991). When two or more cognitive tasks are introduced
simultaneously and must compete for attention, the task that requires more
cognitive resources to process, such as a cognitive reasoning exercise, require more
conscious attention than activating an automatic stereotype. The result is that the
application of the stereotype is suppressed (Blair & Banaji, 1996).
Mulligan (1998) also found that dividing attention between tasks when the
tasks involve encoding new information into memory, in this case, a word
association task and a word-fragment completion task, the attention-demanding
processing reduces the performance, i.e., recall of the words, on both tasks. The
presence of distractions when stereotype information is being summoned from

27
memory may interrupt the process to move from stereotype activation to
application.
There are two exceptions to the result that distractions or cognitive overload
inhibits stereotype activation. The first is the role that a threat to self-image may
play when the individual experiencing cognitive overload is a member of the
stereotype target group, i.e., stereotype threat. In studies where participants
suffered a blow to their self-esteem that is related to the stereotype target group, the
stereotype will be activated and applied even in the presence of other cognitive
processes (Spencer, Fein, Wolfe, Fong & Dunn, 1998). Second, studies of
amnesiacs and elderly patients who are faced with performing a task while
experiencing cognitive overload were able to produce an attitude change
unconsciously without the use of explicit memory (Liebman, Ochsner, Gilbert &
Schacter, 2001).
Deliberate Attention
As the explanation about cognitive working memory suggests, deliberate
efforts to attend to some inputs of information while suppressing others indicates
that some pre-existing attitudes and systematic thought processes help to inform the
decision-making about what to give attention. Some theorists postulate that
humans can control or fail to take action on a stereotype after it has been activated.
For example, Ajzen and Fishbein’s (1980) theory of reasoned action assumes that
people are usually rational, making informed decisions based on the systematic use

28
of the information available to them, including evaluating their environment and
the current situation before taking action.
Ajzen’s (1991) subsequent theory of planned behavior takes into account
the notion that behavior stems from the intention to behave in a particular way, and
that the intention is a consequence of data-driven, systematic reasoning. With
respect to automated stereotyping, and the alternative situations just described, the
role of spontaneous versus deliberate processing appears to be an essential
component in distinguishing stereotype activation from stereotype application.
Stereotypes, by their very nature, do not stem from rational and systematic
information, rendering them less suited for deliberate processing than other data-
driven tasks.
Ambivalence
To some extent, many individuals in U.S. culture are ambivalent about their
feelings and reactions toward individuals in stereotyped target groups. Two
cultural values are at the root of this ambivalence: individualism and
egalitarianism.
Individualism is represented by the protestant work ethic, self-reliance,
devotion to work, and achievement. Negative reactions to people from stereotype
target groups are the result of the perception that individualism is not valued by the
stereotype target groups (Katz, 1981;Katz Wackenhut & Hass, 1986).
The response to egalitarian values, including such ideals as equal
opportunity, social justice and the worth of all human beings, is to sympathize with

29
the stereotype target groups, and to sometimes view target group members in
highly positive stereotypes such as heroic, brave and deserving (Katz, Wackenhut
& Hass, 1986).
It is often the case that in the U.S., people who are not a part of the target
stereotype groups hold both of these conflicting values simultaneously. Therefore,
ambivalence is the result of not bringing forward either the negative feelings
toward those who do not possess individualist ideals or the positive ideals of
egalitarianism in how people from target stereotype groups are perceived.
Distance from Stereotype Prime
There are some situations in which individuals act in ways that are opposite
to the activated stereotype (Wheeler & Petty, 2001). This is true of situations
where the individual who is the target is atypical of the stereotype (Manis, Nelson
& Shedler, 1998), or where issues exist regarding the nature of the prime, such as
individuals who are extreme exemplars of the stereotype. Wheeler and Petty
(2001) identified four possible mechanisms that impact stereotype priming:
automatic goal activation, behavioral tags, biased perception, and possible selves.
Each mechanism is briefly described below.
Automatic Goal Activation Mechanism - Automatic goal activation
indicates that environmental features can influence behavior without the awareness
of the subject, and the behavioral effects of the goal activation may increase over
time until the goal is completed. For example, participants primed with
achievement oriented words in a goal activation task will attempt to continue

30
working past the task time limit in an effort to achieve the goal (Bargh, Gollwitzer,
Lee-Chai, Barndollar & Trotschel, 2001).
Behavioral Tags Mechanism - Behavioral tags provide accessibility to
certain traits that activate certain behaviors. For example, in the experimental
prime condition of one study (Bodenhausen & Macrae, 1998), three sets of
participants each received a word list. One list title contained the name of a famous
person, the second list title contained an unknown name, and the third list contained
no name. The participants with the first list read the list significantly faster than the
participants with the other two lists.
Biased Perception Mechanism - Biased perception refers to the change in
behavior of individuals who have activated a stereotype but change their perception
based on changes in the behavioral setting. For example, participants primed with
moderately hostile and extremely non-hostile exemplars believed that their
ambiguous partner would behave more aggressively (Herr, 1986).
Possible Selves Mechanism - The possible selves mechanism describes
situations in which the perception of other individuals and the situation remain the
same, but the subject’s implicit or explicit self-perception changes as a result of the
environment. For example, the activation of the African American stereotype
could activate an underachieving or lazy possible self concept (Ruvulo & Markus,
1992).

31
Summary of Stereotype Application
In light of the processes and mechanisms just described, interrupting the
application of a stereotype is possible. Introducing additional cognitive tasks to be
performed simultaneously with stereotype activation creates distractions that
compete with stereotype activation. Because many cognitive tasks require
conscious attention, simultaneously introducing them can reduce the ability to act
upon a stereotype, even when it is unconsciously activated. Particularly in
situations where a person has no prior experience to support a stereotype, the
theories of reasoned action make it more likely that deliberate attention can be
bestowed on data-driven, systematic, cognitive tasks rather than acting upon an
automatically activated stereotype. Finally, a disconnect may occur between
stereotype activation and application when an individual of a stereotype target
group does not reflect the stereotype. Coping strategies, such as ambivalence or
setting up environmental cues to trigger reactions to counteract the stereotypes can
prevent automatically activated stereotypes from becoming automatically
responded to or acted upon.
Section Three: Racial Stereotyping
Much is written about the nature of racial stereotyping and the social
context in which stereotypes thrive (Allport, 1954; Aronson, 2001; Crocker, Major
& Steele, 1998; Devine, 1989a; Dovidio, Evans & Tyler, 1986; Fiske, 1998). This
section summarizes five racial stereotyping research topics: Interrupting racial

32
stereotyping; prejudice and discrimination; racial stigma; stereotype threat and
vulnerability; and coping strategies for dealing with stigmatization.
Interrupting Racial Stereotyping
Four contemporary factors related to racial stereotyping are presented here
in order to focus attention on aspects of racism that are related to how acting on
racial stereotypes may be interrupted. These factors are familiarity, curiosity,
negative priming and social group identity. Each is briefly described below.
Familiarity
Experiments involving the role of exemplars examined how exposure to
pictures of admired and disliked exemplars impacts automatic racial prejudice and
racial preferences (Dasgupta & Greenwald, 2001). Using the IAT and two explicit
attitude measures, the results showed that pictures of admired African Americans
and disliked White exemplars significantly weakened implicit automatic pro-White
attitudes within the first 24 hours of the treatment. However, the IAT results did
not correlate with the results of the explicit measures that reflected strong pro-
White attitudes.
Curiosity
A German study (Florack Scarabis & Bless, 2001) of the correlation
between automatic stereotyping about ethnic minorities and social judgments about
ethnic minority group members reveals that a moderating factor is the Need For
Cognition (NFC), i.e., the perceiver is curious about or wants to understand, the
target group. This study focused on the automatic associations and self-reported

33
attitudes of Germans towards Turkish immigrants. The results revealed that
automatic stereotypical associations are at higher levels when the need for
cognition is low, and the automatic associations contribute to self-reported negative
attitude assessment of the target. Curiosity or a desire to understand the target
group of a stereotype appears to reduce automated stereotyping about the group.
Negative Primes
The introduction of negative stereotypes, even indirect ones that may apply
to very specific segments of a stereotype target population, may evoke stereotype-
driven attitudes and responses. Violent rap music as a stereotype prime was
revealed to activate stereotypes about violent behavior in African American males.
The attribution of violent African American males were held by both African
American and White perceivers, and extended to generalizations about other
stereotypical traits of African American men, e.g., intelligence (Johnson, Trawalter
& Dovidio, 2000).
In another study of negative priming related to the attribution of violent
personalities, a survey of 823 British psychiatrists sought to determine if
psychiatrists still believed that Black men are more violent than White men, as had
been the case 10 years earlier (Minnis, Gillies, McMillan, & Shubulade, 2001).
Random distribution of a prospective patient profile included a questionnaire
accompanied by a photograph of either an African American or White man. The
survey revealed that while Black men are no longer considered more violent than

34
White men, other negative stereotypes surfaced that Black men are assumed to be
lower class and undereducated.
Social Group Identity
A final relevant area of inquiry regarding racial stereotyping involves social
identity theory (Brewer, 1979; Tajfel & Turner, 1986) wherein people are
categorized into social groups that are then evaluated and assigned value (i.e., in-
group or out-group.) The need to maintain a higher or more positive standing for
the in-group leads to a positive bias toward one’s own group and negative
behaviors and attitudes toward the other groups. The most prominent behaviors
exhibited are stereotyping, prejudice, hostility and conflict (Crocker, Thompson,
McGraw & Ingerman, 1987).
Prejudice and Discrimination
Underlying the prevalence of racial stereotyping is the progression from
stereotypes to prejudice attitudes and discriminatory behaviors. The inevitability of
prejudice as the attitudinal response to stereotypes is so ingrained that ethnic
attitudes and stereotypes are assumed to be part of the social heritage of society.
As such, no one can escape learning the prevailing attitudes and stereotypes
(Ehrlich, 1973). In general, the attitudinal responses to stereotypes are negative,
and they are based on a social comparison of the stereotype target group as
negative, while the individual’s own group is viewed as positive (Jones,1972).
As previously described, stereotyping is the mental process that activates
responses of prejudice attitudes and discriminatory behaviors. Dating back to the

35
mid 20
th
century, several authors have conceptualized the experience of being the
target of prejudice and discrimination (Allport, 1954, Fanon, 1952, 1967; and
Sartre, 1946 and 1965). The target internalizes the stigmatizing images and
stereotypes about his/her own group. Social group stigma links to Weiner’s
attribution theory (1971, 1986 and 1992) in that the stereotypes associated with a
particular group are external but become internalized, and they are largely
uncontrollable by the stigmatized individual. For this reason, stereotypes are also
considered a type of expectancy that shapes behavior (Hamilton, Sherman &
Ruvolo, 1990) and leads to the self-stereotype reactions that are encompassed by
the concept of stereotype threat presented by Steele (1995).
That prejudice may be the result of ordinary conflicts of interest between
groups suggests the possibility that stereotyping is not automatically applied.
Realistic conflict theory (Sherif, Harvey, White, Hood & Sherif, 1969) examined
conflict and hostility in a boys’ summer camp and found that conflicts of interest
arising from competitive activities intensified hostility and prejudice, while
activities requiring cooperation reduced inter-group hostility.
Discrimination is the behavioral manifestation of a stereotype in which an
individual responds to a stereotype category or an individual based on the group
stereotype. Discrimination occurs in very damaging ways. Discrimination in
employment and academic settings is well documented, and where unlawful, has
been litigated extensively. In her textbook description of stereotyping, prejudice
and discrimination, Fiske (1998) describes two types of discrimination: hot

36
discrimination, that is, prejudice toward out-group individuals that stems from
negative feelings, including resentment, disgust, hostility, and anger toward an
individual or group; and cold discrimination that is based on stereotypes about the
out-groups’ interests, knowledge or motivations. It is the act of discrimination and
the marginalizing effects discrimination has on those who are discriminated against
that emphasize the need to moderate stereotype application.
Racial Stigma
A significant influence on how racial stereotypes take hold is how the target
of the stereotype receives and processes the application of stereotypes in the forms
of prejudice attitudes and discriminatory behaviors. Social stigma, and specifically
for the purposes of this study, the stigma of race, plays a role in how racism persists
in society. Based on the definition of stigma as “a social identity that is devalued in
a particular social context” (Crocker, Major & Steele, 1998) a person who is a
member of a stigmatized group may suffer from internalized oppression and low
self esteem while a person who is not a part of a stigmatized group may experience
higher status and privileges not available to those in stigmatized groups.
Stigma is attached to particular racial groups because of negative
characteristics that are attributed to them. More than two decades of research has
revealed that, by age three children in the U.S. devalue people with dark skin,
people with disabilities and people who are overweight (Crocker, Major Steele,
1998). For example, this early programming was demonstrated in a study of
young African American and White children ages 3-6 years old. Both the African

37
American and White children reveal preferences for White children as potential
friends, and demonstrated a better memory of stereotypic situations in drawings
than for stereotype inconsistent drawings (Levy, 2000).
Stigmatization results in four experiences for the stigmatized: (1) the
potential for becoming the victim of prejudice and discrimination exists regardless
of the individual’s personal status, accomplishments or achievements (Cose, 1993);
(2) by the time an individual in a stigmatized group reaches puberty (Jarvie, Lahey
Graziano & Framer,1983; Rosenberg, 1979) low self-concept and self-esteem have
already taken hold; (3) stereotype threat, i.e., the feeling that the individual will be
judged or treated in terms of a stereotype, whether it is personally relevant or not,
because the individual is part of a stigmatized group (Steele & Aronson, 1995); and
(4) Attribution ambiguity arises due to the fact that the individual in a stigmatized
group may not know whether positive or negative outcomes are the result of the
stigma attributes or the individual’s own performance or personal attributes or
characteristics (Crocker, Cromwell & Major, 1993; Carver, Glass &Katz, 1977;
Schneider, Major, Luhtanen & Crocker, 1966).
In contrast to responding to stereotypes about one’s own stigmatized group,
even studies focused on the impact of priming “other” stereotypes show that
participants respond in accordance with the stereotype indicated in the prime. In a
subtle priming experiment that looked at the stereotype of African American
academic underachievement (Wheeler, Jarvis,& Petty, 2001) non-African
American participants were asked to write an essay about the day in the life of an

38
African-American student named Tyrone or a non-African American student
named Eric. Then all of the participants completed part of the math section of the
GRE. The participants who wrote about Tyrone performed significantly worse than
those who wrote about Eric.
The self-fulfilling nature of stereotypes on the stereotype target groups
causes those individuals to self-identify with the negative stereotype, and feel
stigmatized. Drawing on the attributional theories of Weiner (1974, 1992) to
identify how the stigma is personalized, the cause of an attribution, i.e., the
individual fits a negative racial stereotype, can be viewed as either internally
generated by the individual, or externally generated by a culturally-shared view of
all people in the stereotyped group.
Stereotype Threat and Vulnerability
The activation of negative stereotypes has a profound effect on members of
the target stereotype group, not only in terms of how they are treated by those
outside of the group, but also in terms of how an individual who is a member of the
target group personally receives the negative stereotype information. There are
consequences with respect to the target individuals’ affective responses to the
negative stereotype. Moreover, receiving information about a negative stereotype
disrupts working memory capacity, thereby dividing the individuals’ attention and
causing them to perform more poorly (Schmader & Johns, 2003).
In effect, the stereotype threat is that the individual has to deal with the
added pressure that poor performance could be viewed as substantiating that the

39
individual exemplifies a negative social stereotype about the out-group to which the
individual belongs (Steele, 1997). Although the stereotype threat research focused
on standardized test performance (Aronson & Steele, 1995), stereotype threat is a
social-psychological effect that exists across all socioeconomic levels. For
individuals in the group that is targeted for the negative stereotype, they may
internalize the stereotype and experience feelings of inferiority and inadequacy.
As Steele (1997) describes a stereotype threat as a “situational threat—a threat in
the air—that, in general form can affect the members of any group about whom a
negative stereotype exists … members of these groups can fear being reduced to
that stereotype.” (p. 614)
The other salient characteristic of stereotype threat is that it most often
affects the confident individuals who have not internalized the negative stereotype
as related to self. In fact, members of the target group do not need to believe the
stereotype or worry that it relates to them. It is the mere presence of the stereotype
that can distract attention away from performing a task (Steele, 1997).
Susceptibility to internalizing negative stereotypes about one’s own racial
group has been researched and analyzed under the research umbrellas of stereotype
threat and self-efficacy. Stereotype vulnerability manifests itself in many ways.
Five reactions to stereotype threat are: ambiguous stereotyping, the “Black Sheep”
effect, the “Cocktail Party” effect, tokenism and overconfidence. Each of these
reactions is briefly described next.

40
Ambiguous Stereotyping
With respect to recent studies on social stigma and stereotype threat,
attributional ambiguity arises for individuals in stereotyped groups because they
experience a lack of clarity as to whether outcomes related to personal interactions
result from negative reactions to the individual as a member of the group, or from
negative reactions to the individuals’ personal qualities. Major & Crocker (1993)
studied this attributional ambiguity phenomenon and found that as expressions of
prejudice shift from blatant and unambiguous prejudice, to more subtle and indirect
forms, individuals in the target stereotype group experience even more threats to
their self-esteem because there is no clear way of knowing whether the nature of
the prejudice is directed at the individual or the target group.
A more recent study of the attributions people make for their own in-group
and out-group attitudes showed that under high threat situations, people are most
likely to attribute negative out-group attitudes to external causes rather than their
own internal self-esteem (Kenworthy & Miller, 2002).
“Black Sheep” Effect
The extent to which members of a stigmatized group gain acceptance by
members both within and outside of the group may be hampered by another
condition referred to as the “Black Sheep” effect (Marques & Yzerbyt, 1988).
Members of stigmatized groups get labeled as black sheep when they are perceived
as being on the extreme edge of the positive social identity for the group because

41
they are marginal performers or they possess distasteful personality traits, such as
being aggressive or aloof.
More recently, a form of the “Black Sheep” effect has been identified with
an entire stigmatized group. In a series of studies on terror management theory and
in-group/out-group affiliation, members of out-groups identified as having ties to
terrorism or other negative attributions will distance themselves from the group
(Arndt, Greenberg, Schimel, Pyszczynski & Solomon, 2002). In a study of
distancing behaviors, individuals with high self-esteem distance themselves from
negative or unfavorable past experiences more readily than low self-esteem
participants (Ross & Wilson 2002). However, individuals in such stigmatized
groups often effectively use a variety of defense mechanisms to block the
psychological consequences of the negative attributions (Crocker, Voelkl, Testa &
Major, 1991; Tajfel &Turner, 1986; Brewer, Dull & Liu, 1981).
“Cocktail Party” Effect
Another recent body of research is the disruption that stereotype threats
cause in terms of the amount of cognitive working memory available to process
new information. In most instances, the effect that activating a stereotype has on
working memory is that some of the cognitive resources get channeled into
suppressing unwanted negative thoughts and feelings such as anxiety, physiological
arousal and apprehension (Schmader & Johns, 2003). The expanded definition of
working memory includes not only the traditional cognitive working memory
component of the central executive processor for the attention capacity of memory,

42
but it also includes auditory/phonological and a visual/spatial processors. Thus, the
sensory overload is compared to the physical and emotional stimulation of being at
a cocktail party with many people to attend to simultaneously.
In this more expansive definition of a three-pronged working memory
capacity, it can be argued that people are capable of simultaneously attending to
more than one piece of information. When higher memory capacity is invoked,
people can more readily suppress task-irrelevant information. Recent studies by
Conway, Cowan and Bunting (2001) reveal lower than expected susceptibility to
the “cocktail party” effect, i.e., divided attention across several new inputs of
information that may or may not be related to the task at hand, created added
disruptions and distractions but did not prevent task completion.
Tokenism
A specific phenomenon that arises from the notion of vulnerability to
stereotypes is the feeling of tokenism in creating a sense of self-doubt and
diminished self worth. Nieman’s (1999) case study of tokenism in academia
describes the stigma of incompetence that is ascribed to new faculty of color, the
increased workload expectations of the entire academic community to have the sole
Latina on the faculty to represent and address all of the needs of the Latino student
body and community, and the psychological and physiological personal
consequences these situations create. The road to tenure and professional
credibility is rife with roadblocks that are seemingly created by the individual.

43
In the corporate environment, the role of race in the career advancement of
minority professionals to the executive ranks revealed that prejudice is one of the
barriers to success. In a study conducted by Thomas and Gabarro (1999) the career
trajectory of minority executives rising to the top is longer than that of their White
counterparts, and due to the pressures of tokenism, stereotype threat, and low self-
efficacy, many minority executives take themselves out of consideration for senior
level jobs.
Overconfidence and False Hope
Overconfidence as a response to inflated expectations has been the subject
of several studies, particularly in self-change literature (Blanton, Pelham & Dehart,
2001; Polivy & Herman, 2000; Wegner, Erber & Zanakos, 1993). Overconfidence
has also been employed as a defense mechanism to counteract stereotype threat.
However, the false hope syndrome (Polivy and Herman, 2000) cites four
characteristics of overconfidence that lead to false hope, and provide valuable
insights for overcoming stereotype threats: (1) inflated success claims, particularly
with various self-change programs such as dieting and smoke cessation programs;
(2) overconfidence based on unrealistic expectations; (3) a focus on previous
successes and not enough on lessons to be learned from previous failures; and (4)
expending too much energy trying to avoid negative experiences, leaving little or
no energy for positive experiences.
The reactions to stereotype threat and the vulnerability of being negatively
labeled, means that members of the stereotyped groups are constantly on the

44
defensive, and frequently misunderstood. For those outside of the stereotyped
group, the effects are often lower expectations of performance, and ambivalence
toward protests about unequal treatment.
Coping Strategies for Dealing with Stigma
The stereotype threat reactions just described indicate the impact on the out-
group or stigmatized group. For those individuals who are the target of prejudice
and discrimination, the potential for experiencing threats to their individual and
collective self esteem looms large. The stigmatization that negative stereotypes
create impacts both the personal and professional lives of the stigmatized. Four
coping strategies for individuals in stigmatized groups have surfaced in a number
of experiments: dis-identification, external attribution, devaluing importance of
outcomes, and social comparison. These four psychological mechanisms can help
to prevent long-term negative outcomes, and they are briefly described below.
Dis-identification to Protect Self Esteem
In order to activate a stereotype threat, the target of the stereotype has to
identify, or care about, the stereotype threatening domain (Steele, 1997). It has
been noted that members of stigmatized groups find esteem-protecting strategies
that prevent stereotype threats from lowering their self esteem. In an evaluation of
dis-identification as an explanation for the African American-White achievement
gap, Crano and Chen (1998) found that African Americans are less likely to believe
that standardized tests are measures of academic competence. In addition, as
Crocker and Major (1989) reported, esteem protecting explanations for lower

45
performance may include blaming their failure on the prejudice of non-target
groups, comparing themselves only to others within the target group, or
disassociating or devaluing the stereotype threatening domain.
External Attribution
Attributing the negative stigmas to external sources rather than to
internalizing them, is another coping strategy that helps to preserve self-esteem.
Specifically, the stigmatized individual views societal prejudice and discrimination
as the cause of the stigma rather than personally identifying with the stereotypes
(Crocker, Cornwell & Major, 1993; Crocker & Major, 1989; Crocker, Voelkl,
Testa & Major, 1991; Dion & Earn, 1975; Major & Crocker, 1993).
Devalue Importance
A third coping strategy is to devalue the psychological importance of a
particular outcome resulting from being stigmatized by not identifying with the
stigma. Common examples of students who perform poorly on standardized tests
but do not experience negative impacts to their self esteem are able to do so
because they do not place importance on test and school performance (Rosenberg,
1965; Pelham & Swann, 1989; Tesser, Millar & Moore, 1988).
Social Comparison
Because people in stigmatized groups may experience feelings of
vulnerability in social comparisons with other individuals or groups, a fourth
coping strategy is to restrict social comparisons to those individuals that share the
same stigmatized status (Brickman & Bulman, 1977; Jones, Farina, Hastorf,

46
Markus, Miller & Scott, 1984; and Wills, 1981). However, when the legitimacy of
group differences is examined, the social group comparison may shift. For
example, in a study of pay expectations of women, when the pay criteria reflected
objective performance criteria women expected to be paid what other women were
paid. Nevertheless, when the pay criteria appeared to be biased in favor of men,
they compared themselves to the men, and expected to be paid the same as men
(Bylsma, Major & Cozzarelli, 1995).
Summary of Stigmatization
The targets of stereotypes experience prejudice and discrimination from
others and manifest internalized oppression and low self esteem in their own
reactions and behaviors. Stereotyping, and the accompanying stigma that comes
with being a member of a negatively stereotyped group, can have a lasting impact,
even when an individual does not exhibit the stereotype characteristics. The effects
of and responses to stigmatization reveal the damaging psychic costs of
stereotypes. The stereotype threat that is the vulnerability and susceptibility
individuals of a stigmatized group feel when a stereotype is evoked, has the power
to negatively impact the performance and the success of the individuals, and by
association, those who are professionally and personally connected to them.
Whether consciously or unconsciously activated and applied, stereotypes are
damaging, and measures to identify and modify them is an important research
endeavor. In the next section, explicit and implicit measures of automated
stereotyping are reviewed.

47
Section Four: Explicit and Implicit Measures of Automated
Stereotyping
Explicit Measures of Automated Stereotyping
The research of the last decade regarding the nature of automatic and
controlled components of racial prejudice has spawned an impressive array of self-
report instruments that provide explicit evaluations of racial prejudice. A brief
description follows of four self-report instruments aimed at uncovering racism.
The instruments that are described below include: the Modern Racism Scale, the
Motivation to Control Prejudice Scale, the Linguistic Inter-group Test, and the
Group-Based Dominance and Opposition to Equality Indicator.
Modern Racism Scale (MRS)
The MRS (McConahay & Hough, 1976) measures individual differences in
anti-African American feelings. The validity of the MRS was challenged in
Devine’s (1989b) studies using the scale when the results showed no difference in
the responses of high and low prejudiced individuals. Other studies reveal that
people with low prejudice MRS exhibited more negativity toward African
Americans (Fazio, Jackson & Dunton, 1995).
Motivation to Control Prejudiced Reactions Scale
Dunton & Fazio (1997) developed this scale to measure the extent to which
individuals are motivated to minimize the prejudiced reactions they experience.
The 17-item scale measures individual differences in a willingness to suppress

48
prejudiced attitudes. There is a correlation between a low motivation to control
prejudice and greater likelihood to express negative attitudes.
Linguistic Inter-Group Bias (LIB) Test
The LIB Test (Maass, Milesi, Zabbini & Stahlberg, 1995; von Hippel,
Sekaquaptewa & Vargas, 1997) is an explicit procedure that uses drawings
depicting members of different groups engaging in stereotypical and non-
stereotypical behaviors, and asks subjects to select one of four descriptive
statements that best describes the action in the drawing. The statements illicit the
degree to which respondents have internalized stereotypes by pairing a range from
concrete descriptions for stereotype incongruent actions to abstract descriptions for
stereotype congruent actions, e.g., a non-stereotypical picture of an Asian-
American women striking someone is described as “Serena Chu hit someone”
while a stereotypical picture of an African-American man striking someone is
described as “Jamal Washington is violent.”
Group-Based Dominance (GBD) and Opposition to Equality Indicator
The GBD and Opposition to Equality Indicator (Pratto, Sidanus, Stallworth
& Malle 1994; Jost & Thompson, 2000) are two attitudinal measures of social
dominance that seek to determine the impact of social dominance on African
American and European American subjects with respect to self-esteem,
ethnocentrism, political conservatism and racism. These scales look at self esteem,
ethnocentrism, economic system justification, political conservatism and racial
attitudes, as well as conservative social policies such as Affirmative Action.

49
With explicit measures there is a correlation between the responses on
explicit measures and the individual’s behaviors. As it relates to exhibiting
knowledge about the cultural stereotypes of African Americans, both high and low
prejudice individuals as indicated by explicit measures, possess the same
familiarity with the stereotypes, and differ in the amount of controlled behavior
they exhibit in response to activated stereotypes (Devine, 1989a). The major
drawback in the use of explicit self report instruments is that they are not reliable
indicators of prejudice attitudes. In responding to the questionnaires the subjects
are in control of how they will respond. When subjects are aware that stereotypes
are being activated, they may provide socially acceptable responses rather than
their true feelings, i.e., there is a reluctance to reveal the negativity of stereotyped
attitudes that the instruments seek to measure.
In general, compared to explicit measures, implicit measures are less
predictable and more indirect and unobtrusive, and therefore more likely to bring
out the automated stereotypes and spontaneous reactions to stereotypes. A number
of studies use an experimental design that includes both explicit and implicit
measures to uncover automated stereotyping, but there is little or no correlation
between the explicit and implicit measures. Moreover, the most widely used
implicit measure, the IAT, when used alone does not predict conscious behavior
(Karpinski & Hilton, 2001).

50
Implicit Measures of Automated Stereotyping
Limitations of the explicit, self-report measures to detect the presence of
automated stereotyping has lead to considerable interest in designing and
experimenting with implicit measures of automated stereotyping. The ability to
identify unconsciously an object or word as a result of a prior encounter with it
denotes automated processing, and the absence of conscious choice represents
implicit memory. (Shiffrin & Dumais, 1981). These two factors, evaluating prior
experience and measuring the unconscious rate of choosing, have become the
underlying principles of implicit measures of automated stereotyping. The body of
research that uses implicit measures of stereotype activation has as its origin the
priming procedure that matches pictures or words with an evaluative positive or
negative adjective (Fazio, Sanbonmatsu, Powell, and Kardes, 1986).
Moreover, research has shown that stereotypes influence responses without
the perceiver’s awareness of the influence (Banaji, & Greenwald, 1995; Bargh,
1992; & Jacoby & Kelley, 1992). Implicit bias exists even when explicit bias does
not exist (Dovidio, Kawakami, Johnson, Johnson & Howard, 1997).
As with other automated processes, the reaction times are faster on
automated stereotyping than with non-stereotypical stimuli. Blair & Banaji (1996)
studied the effects of manipulating the subjects’ intentions to stereotype and the
constraints on cognitive resources, and found that reaction times are faster when
automated concepts are involved.

51
A study of the role of attention on test performance showed that perceptual
implicit tests do not rely on attention-demanding cognitive processes as much as
other types of memory tests. Furthermore, distractions that can divide attention do
not reduce test performance (Mulligan, 2003). In another study of implicit
conceptual word association memory tests, the researchers expected words with
pictures to result in more priming than words alone, but found that the
distinctiveness of certain pictures was a greater factor than simply the presence of
pictures (Weldon & Coyote, 2003). Other studies of implicit test revealed that
manipulating attention during a procedure, or manipulating the amount of time
allocated affected explicit recall, but had no impact on automatic influences
(Jacoby, 1998).
In essence, implicit measures to uncover the existence of automated
stereotypes have taken many forms. The work of Fazio and his colleagues who
developed some of the original work in activating automated stereotypes is
presented next.
MODE Model
In the 1995 study conducted by Fazio, Jackson, Dunton & Williams, the
researchers sought determine if cultural stereotypes or personal evaluations are
activated when the subject is presented with faces of African American individuals.
Fazio’s research focused on characteristics of the perceiver and how visual
information about race activates racial stereotypes in memory (Fazio & Dunton,
1997; Fazio, Jackson Dunton & Williams, 1995). The Motivation and Opportunity

52
as Determinants (MODE) model of attitude-behavior processes (Fazio, 1986, 1990)
proposes that the automatically activated stereotypes are personal evaluations of
people that can be suppressed rather than socially shared, cultural stereotypes.
Additionally, in order to control automatic activation of a stereotype, the individual
must be motivated to appear less prejudiced. Where attitude plays a role in guiding
and motivating behavior, individuals are highly motivated to control prejudice, and
they may overcompensate in order to appear to be even less prejudiced than they
actually are (Fazio, Jackson, Dunton & Williams, 1995).
Four complex experiments were conducted that used five components: 1)
the Modern Racism Scale served as an introductory survey, 2) computer
experiments that required subjects to evaluate words as good or bad; 3)
presentation and recall of African American and White faces; 4) using timed
response prime of African American and White faces paired with stereotypic
positive and negative words; and 5) interaction with a African American or White
experimenter. The general findings were that there was no correspondence
between the MRS scores and how subjects performed on the unobtrusive tests. All
of the other test components that utilized African American and White differences
revealed a bias against African American subjects.
In a 1997 study on categorization by race, Fazio & Dunton again used
implicit measures to determine if the race of an individual will automatically attract
attention and cause a categorization by race. The speed of responding to a face and
word pairing indicated automated stereotyping, i.e., there appeared to be an

53
automatic activation of racial attitudes when African American photos were
presented.
Through a number of experiments involving the manipulation of racial
attitudes, three types of White participants were studied: individuals labeled as non-
prejudiced who do not experience automatically activated negativity towards
African Americans; prejudiced individuals who do experience automatically
activated negativity and are not motivated to control these attitudes; and individuals
who experience automatically activated negativity and are motivated to control
these attitudes. The response times of the participants who were motivated to
control prejudice had slower response times, indicating that they spent time to
counter the attempt to focus their attention on race (Fazio & Towles-Schwen, 1999;
Fazio, Jackson, Dunton & Williams, 1995).
To summarize the MODE findings, when the opportunity to express
prejudiced attitudes or exhibit prejudiced behaviors goes up, the motivation to
control those responses goes up, and the degree of automated response goes down.
Implicit Association Test (IAT)
The widespread acceptance of the (IAT) as an effective tool for measuring
implicit bias has generated new empirical questions about the true nature of
prejudice and the impact of contextual manipulations in moderating the
manifestation of prejudice in experimental environments. The IAT predicts
spontaneous behaviors, i.e., millisecond responses to words and pictures that are

54
not subject to the conscious control that exists in the use of explicit self-reporting
instruments (Egloff & Schmuckle, 2002).
Research in the 1980s demonstrated that repeated exposure to stereotypical
information automates it, and that people process stereotypical information faster
than nonstereotypic information (Brewer, Dull & Lui, 1981;Lalonde & Gardner,
1989). By 1998, Greenwald, McGhee & Schwartz turned the earlier work of Fazio
(et. al, 1986) and Banaji (et. al, 1996) on implicit measures of bias into the IAT, a
computerized response latency tool that could indirectly assess implicit attitudes.
Use of the IAT showed evidence of implicit racism on the part of Whites toward
African Americans, regardless of the subject’s scores on explicit Modern Racism
Scale. The faster the response time in matching stereotype congruent positive or
negative words to African American or White faces, the greater the level of
automated stereotype activation. To be specific, participants respond fastest when
White faces are displayed with positive words, and when Black faces are displayed
with negative words.
That prejudice exists, even when subjects responding to the IAT have no
prior experience with the stereotype target presented in the IAT, raises questions
about how stereotypes are formed (Ashburn-Nardo, Voils & Monteith, 2001).
These findings about the prevailing impact of pre-existing prejudice are contrary to
studies of implicit motivations that indicate subjects recall more information about
a group of people when they are a part of that group than when the information is
about another group (Woike, Lavezzary & Barsky, 2001).

55
Implicit bias researchers, primarily in Europe, have been proposing
alternative formats for the IAT in an effort to increase the correlation between IAT
results and the ability to predict prejudice attitudes and discriminatory behavior
(Brendl, Markham & Messner, 2001; Rothermund & Wentura, 2004; Mierke &
Klaner, 2003; and De Houwer, 2001). However, in the U.S., the basic IAT test
format remains unchanged, and instead of changing the IAT, additional priming
activities, explicit measures and other changes have been added to the experimental
design of IAT studies.
In terms of implicit measures, the IAT has consistently yielded results with
high reliability. Questions about variability that potentially could be attributed to
low internal consistency, have instead been attributed to independence (Blair, 2001;
Dovidio, Kawakami & Beach, 2001; and Cunningham, Preacher & Banaji, 2001).
The following descriptions identify four IAT formats: word-only
experiments, word and picture experiments, social situations and words
experiments, and stimulus familiarity experiments.
Word-Only IAT Experiments
In the initial IAT experiments, the IAT measured the differential association
of 2-choice tasks of objects (e.g., flower and insect names) paired with a negative
or positive attribute word evaluations (pleasant and unpleasant). In these word-only
IAT studies, the researchers found that the response time is slower when two less
associated words are together than when the expected evaluative associations (e.g.
flower and pleasant) appear together (Greenwald, McGhee & Schwartz, 1998).

56
Subsequent versions of the IAT used faces for the objects and covered a range of
subjects including race, gender, age, self-esteem and shyness.
People tend to respond quicker when generally liked items are paired with
positive words than when generally disliked items are paired with positive words.
Ashburn-Nardo, Voils & Monteith (2001) found that participants responded more
quickly when in-group names are paired with pleasant words and out-group names
with unpleasant words. The IAT also measured the correlation between self-esteem
with positive or negative words (Greenwald & Farnham, 2000).
Other word-only IAT experiments revealed the same results of faster
response times when the object word and attribute word pairings were the expected
or stereotypical pairs. In a series of German in-group and out-group experiments to
determine the extent to which inter-group bias occurs, the introduction of
unfamiliar objects and non-word attributes tested the pervasiveness of out-group
bias. American in-group and Surinam out-group objects were paired with pleasant
and unpleasant words, old and young were paired with good-bad and non-word
attributes, and East and West Germans were paired with good-bad words. In all
three experiments the in-group with positive attribute words were favored and
resulted in a faster response time (Rothermund, Wentura, 2004).
Similarly, in four studies pairing Black-White, homosexual-straight,
Americans-foreigners and high-low self esteem, to attributes of moral-immoral,
happy-unhappy, and pleasant-unpleasant, the stereotypical negative attribute words

57
and group pairings resulted in the fastest response times (McFarland & Crouch,
2002).
The word pair IATs have been used successfully to indicate the presence of
implicit bias. In fact, questions have been raised about the priming effects that
using pictures of people rather than words makes the IAT subject to faking, as does
repeated exposure to the IAT (Steffens, 2004).
Word-Picture IAT Experiments
As Greenwald made the IAT readily accessible as a test instrument, the
nature of the test expanded beyond the use of names and other words to describe
objects to tests that feature pictures of the objects. Most notably, IATs featuring
Black and White or male and female faces became standard by 2000 (Greenwald &
Farnham, 2000; Monteith, Voils & Ashburn-Nardo, 2001; Dasgupta & Greenwald,
2001).
Dasgupta and Greenwald (2001) found that exposure to pictures of admired
Black and disliked White Americans reduced automatic response times on the IAT,
although it did not reduce explicit racial attitudes. In three studies to determine the
moderating role of motivation to respond without prejudice, the IAT negative
response times to Black and negative pairings was slowed by the introduction of
cognitive tasks and cues to activate internal desires to respond without prejudice
(Devine, Amodio, Ashby-Plant, Harmon-Jones & Vance, 2002).

58
Social Situations and Words IAT Experiments
Manipulating the situational conditions presented in the IAT testing
environment moderates the results of the IAT. For example, different results were
achieved when the experimenter administering the test is White versus African
American (Lowery, Hardin & Sinclair 2001).
The IAT has also been used to assess dysfunctional beliefs of high anxiety
patients in psychopathology (Jong, Pasman, Kindt, & van den Hout, 2001). Using
social situations and neutral words with positive or negative outcomes, the
researchers found the predicted outcomes of the relation between social situations
and negative outcomes.
Additionally, a study (Egloff & Schmukle, 2002) adapted the IAT to
measure anxiety by matching associations of self compared to others’ using anxiety
vs. calmness-related words. The results revealed that the IAT reliably measures
anxiety and social desirability, and that the predictive validity of the IAT surpasses
explicit questionnaires for assessing anxiety.
Another social phenomenon that impacts IAT results is the implicit racial
evaluations of African American’s preference for Whites. Ashburn-Nardo,
Knowles & Monteith (2003) studied the relationship between African American
subjects IAT results and their preference for a White or African American work
partner. The study revealed that when African American subjects’ IAT results
were high for automated African American stereotyping, they preferred a White
partner, and conversely, if African American subjects’ IAT results were lower for

59
African American stereotyping, they preferred a African American work partner.
In addition to the result indicating that the IAT scores were related to work partner
preference, where African American subjects indicated a preference for White
work partners there was a greater acceptance of social inequalities and lower
confidence in the ability of a African American partner to successfully complete
intellectually challenging work tasks. Moreover, a significant correlation was found
between IAT scores of African Americans who implicitly preferred their own race
and the preference for a African American work partner.
Attempts to use the IAT to predict inter-group discrimination and to relate it
to explicit measures of prejudice have expanded the use of the implicit measures.
In a study to determine if there is a correlation between the IAT and explicit
measures, McConnell and Leibold (2000) used an explicit semantic differential
scale for African Americans and different scales for Whites; a feeling thermometer
for African Americans and one for Whites; and a word-based IAT comprised of
African American associated names (e.g., Jamal) and White associated names (e.g.,
Mary Anne) paired with desirable (e.g., wonderful) and undesirable (e.g.,
offensive) words; and interactions with a White experimenter, then a African
American female experimenter which were observed by two trained judges who
coded the subjects interactions with the experimenters using 16 behaviors (e.g.,
friendliness, openness of arms, number of speech hesitations, etc.) The findings of
the study revealed predictive validity of the IAT in that the IAT was related to
biases in inter-group social interactions, and that implicit measures of attitudes are

60
predictive of unintentional negative reactions, or behavioral “leakage” (Dovidio,
Kawakami, Johnson, Johnson & Howard, 1997).
Stimulus Familiarity IAT Experiments
A study of the effects of word familiarity and frequency on IAT results in
racial stereotyping (Ottaway, Hayden & Oakes, 2001) revealed that individual
differences outweigh the automatic attitudes shared by a culture, and that stimulus
familiarity can make the IAT task easier, confounding the response time results.
Karpinski & Hilton (2001) similarly conducted three studies to examine the
relationship between the IAT and explicit attitudes, and concluded that IAT scores
reflect environmental exposure to bias rather than the subjects’ personal
endorsement of the evaluative associations the IAT addresses.
The IAT was also used to implicitly measure Semitism attitudes using
Christian and Jewish names (Rudman, Greenwald, Mellott 7 Schwartz, 1999);
inter-group attitudes between Korean and Japanese subjects (Greenwald, McGhee
& Schwartz, 1998); and positive and negative judgments of familiar and unfamiliar
names of American and Russian leaders. The IAT produced the same results of
implicit prejudice when religion and nationality was the stereotype target, thus
demonstrating the IAT’s construct validity.
In another study to determine if IAT results are compromised by the
participants’ familiarity with the representations of Whites found no difference in
the results, i.e., the response to positive and White stimuli and negative and African
American stimuli remained unchanged when the stimulus is known to the

61
participants (Dasgupta, McGhee & Greenwald, 2000). An additional study of the
IAT found that the IAT reflects attitudes toward the target group or concept rather
than the individual exemplars of the target (De Houwer, 2001).
Response to Social Structure
A preponderance of the automated stereotype research has been conducted
in laboratory or academic settings, and with study populations comprised of college
students who are mostly White, and middle to upper class. However, there are a
few studies that take the issues of automated stereotyping out of the classroom
environment and that also consider other mitigating factors that are not relevant to
academic settings. A brief description of these studies conducted in non-academic
settings is presented below.
Lower and Working Class Study Population
Niemann (1998) examined African American and White stereotypes
amongst a population of African American and White flea market customers. Flea
market customers were invited to participate in social psychological research at a
flea market booth, and were asked to indicate on a list of stereotype terms those
terms that were true for each racial and gender group listed at the top of the page,
e.g., African American Male). The results indicated that non-college African
Americans have more complex views of themselves and others, and African
Americans identified both negative and positive self views that differentiate gender
differences. In contrast, Whites have more simplistic, one-dimensional negative
views of African Americans that focus on African American male stereotypes as

62
representative of all African Americans. One possible explanation of this
phenomenon is provided in studies of social distance (for example, Mindiola,
Rodriguez & Niemann, 1996). Regional demographics factor into the
homogeneity of some communities, but to a large extent, Whites can limit contact
with African Americans at work and at home, while African Americans generally
have more frequent contact with Whites outside the home.
Work Group Member Bias
In studies involving work groups, when African Americans perceived
negative bias toward African Americans they were inclined to prefer White
coworkers over African American coworkers (Livingston, 2002; Leonardelli,
Brewer & Geoffrey, 2001). Two theories support this finding: optimal
distinctiveness theory (Brewer, 1991; Jetter, Spears & Manstead, 1998) posits that
even for African Americans, the identification with the White in-group in order for
African American out-group discrimination to occur; and the system justification
theory of Jost and Banaji (1994) proposes that existing social hierarchy is observed
and legitimized at the expense of personal and group interest. Moreover,
attribution figures prominently in justifying decisions about in-group and out-group
preferences. Pettigrew (1979) described the “ultimate attribution error” as the
tendency to view everything related to the in-group as good and everything related
to the out-group as bad.

63
Preference for Implicit Measures
The IAT remains the implicit measure of choice for many researchers
seeking to expand knowledge about stereotype activation. A study designed to
assess the inter-item consistency, stability and convergent validity of implicit
attitude measures (Cunningham, Preacher & Banaji, 2001) revealed that implicit
measures were substantially correlated after correcting for inter-item consistency.
However, the most compelling research design for assessing racial stereotyping
uses the Fazio and Greenwald models consisting of a combination of explicit
survey measures, the IAT, and a treatment environment that includes a racial mix
of experimenters.
Section Five: Interventions to Moderate the Effects of Implicit
Stereotyping
As Allport stated, social categorization and prejudging people is normal, but
he also encouraged what he called “constructive interethnic contact” (Allport,
1954) among people whose prejudice is not too deeply rooted. These explanations
about stereotypes and prejudice stopped short of identifying whether these were
automatic or controlled processes. In this section, interventions are described that
show promise as treatments to interrupt unconscious stereotypes and result in
changes in thought, feelings and behavior.
Interethnic Friendship
Familiarity with people who are members of the stereotype target group can
moderate the stereotype activation. As originally proposed by Allport (1954),

64
contact with people from other/out groups improves the attitudes of people in the in
group when the out group members are perceived to have equal status with in
group members, in terms of possessing common goals in an interdependent setting
with support from authorities (Aberson, Shoemaker & Tomolillo, 2004). Further
research reframes and simplifies the concept of inter-group friendships by
identifying that friendships optimize inter-group contact (Pettigrew, 1998). Using
four explicit bias measurement scales and the IAT for measuring implicit bias,
Aberson’s and colleagues (2004) study revealed that White study participants who
have close friendships with African Americans exhibited less bias than those
without friendships. Similar results were found in a second study measuring the
impact of White participants who have close friendships with Latinos.
Social Influence
Shared social reality theory focuses on the common social influence of a
culture in developing shared stereotypes about a group of people. In a study of the
effects of social influence on racial prejudice, groups of White and Asian
Americans exhibited more automatic prejudice in the presence of an African
American experimenter compared to a White experimenter. Moreover, both groups
exhibited less automatic prejudice when primed to avoid prejudice reactions
(Lowery, Hardin, Sinclair, 2001).
In addition, common respect and admiration for famous people impacts
stereotypes. In an implicit bias study involving exposure to pictures of admired
African Americans and disliked Whites, the results revealed a reduction in pro-

65
White implicit bias, even though explicit racial attitudes were not effected
(Dasgupta & Greenwald, 2001). This study indicates that manipulation of the
social context may impact racial bias.
First Impressions
Implicit bias manifests itself even among professionals whose objectivity
about people is assumed to be part of the nature of their work. In studies of White
therapists/counselors who were asked to analyze the factors impacting their comfort
level in cross cultural counseling sessions, the IAT revealed a high rate of implicit
bias against African American clients. The impact of this bias is its influence on
the first impressions of African American clients and the validity of the therapists’
subsequent diagnosis of these clients (Abreu, 1999). Measures to identify the
existence of the implicit stereotypes and counteract them with tools to minimize
first impression assumptions about cross-cultural clients can improve the therapists’
diagnosis accuracy.
Shifting Goal Intentions
Setting goals and engaging in goal-directed behavior is central to positive
performance and behavioral outcomes. With respect to inhibiting automated
stereotyping and overcoming the negative impact that stigmatization has on
individuals in stereotype target groups, another challenge is to overcome the
attitudes and behaviors that cause people to retreat from efforts to attain goals.
Goal pursuit is fraught with opportunities to experience failure, and when the
failures are attributed to personal effort or other factors under the control of the

66
individual, the inability to achieve goals becomes one of the negative stereotypes
attributed to an individual or stigmatized group.
Gollwitzer (1999) describes the role that implementation intention plays in
helping people to stay focused on goal-directed behaviors. Recent studies
(Gollwitzer, Schaal, Moskowitz, Hammelbeck & Wasel, 1999) have shown that
implementation intentions reduce or inhibit the automatic activation of stereotypes.
In these studies, purposefully engaging in self talk to refrain from stereotyping
about elderly people when one encounters an older person, to ignore gender when
interacting with a woman, and to ignore the fact that a person is homeless when
encountering a homeless person, reduced the stereotypical judgment of that person.
Positive Priming
Another strategy for inhibiting automaticity, that has the field of
psychotherapy as its origin, is to focus on positive expectancies. Positive priming
to counter expectancies that a stereotype is true, works in two ways. Starting with
small examples that counteract stereotypes and moving to more complex examples,
coupled with increasing exposure to people in non-stereotypical roles and
situations, help to change attitudes and application of the stereotype (Kirsch &
Lynn, 1999).
Kawakami, Dovidio, Moll, Hermsen & Russin (2000) found that automated
stereotyping can be reduced through practice in denouncing the stereotypes. In a
series of experiments, stereotype words were presented to the participants who
were then asked to predict if the photograph to follow would be a White or African

67
American man. Over time, the participants paired non-stereotypic words with the
pictures of African American men. Similar word pairing experiments using
photographs of elderly people and skinheads produced the same results over time
of pairing the photographs with positive non-stereotypic words.
Mental Imagery Prime
Some research reveals that the automated activation of implicit stereotypes
may be reduced by mental imagery. In a series of five experiments, the use of
mental imagery that depicts a variety of people engaged in counter-stereotypic
behaviors resulted in participants using weaker implicit stereotypes as compared to
participants who viewed images of people engaged in neutral or stereotypic
behaviors (Blair, Lenton & Ma, 2001). These mental images results reinforce the
research previously described in the review of implicit measures in social cognition
research, wherein Fazio & Olson (2003) revealed the role that priming measures
play in interrupting automated stereotyping.
Consequences of Mood on Stereotyping
Mood and the factors that impact mood, such as optimism or pessimism,
may influence the degree to which stereotypes are applied during out-group/in-
group social situations. The basic research describes the impact of stress on mood,
and examines how stressors that disrupt goal-directed behavior negatively impact
mood (Zohar, 1997). The psychological and biological effects of coping with
stressors, including effects such as performance decrements, reduced effort,
reduced interpersonal tolerance and negative mood, are a part of a model that

68
explains the impact of mood on behavior that is called the adaptive-cost hypothesis
(Selye, 1956). It is assumed that stress results in these disruptive behaviors, and
that when a bad mood surfaces, the tolerance or ability to control stereotypes is
reduced. Conversely, studies on stress and mood indicate that optimism is
positively associated with better mood as well as stronger immune systems
(Segerstrom, Fahey, Kemeny & Taylor, 1998).
Anxiety is another mood-altering experience that is experienced by both the
targets of stereotypes as well as the individuals applying the stereotypes. Anxiety
in highly prejudiced individuals who hold negative stereotypes often experience
anxiety over the prospect of being viewed as prejudiced, and this anxiety may take
the form of feeling embarrassed, rejected, ridiculed or even exploited for their
views (Stephan & Stephan, 1985).
Diversity Education
In addition to the role of repeated exposure to non-stereotypic stimuli and
practice in unlearning the stereotypes, research also suggests that diversity
education reduces automatic stereotyping and prejudice. Students in a prejudice
and conflict reduction seminar demonstrated reduced implicit and explicit anti-
African American biases compared to students who did not take the seminar
(Rudman, Ashmore & Gary, 2001).
Steele’s (1997) research on stereotype threat, proposes remedies through
training directed at the stigmatized, or stereotype-threatened groups, as well as for
the individuals imposing the stereotypes on others. With respect to the stereotype-

69
threatened groups, for those individuals who do not identify with the stereotype
threatening domain (e.g., African Americans with high self-esteem in the face of
the low academic ability stigma or women with respect to the poor math
performance stigma) other strategies to increase their identification with the
performance domain may be needed.
Wise Strategies
For those individuals whose performance is impacted by the social stigma
of their group, and who may not be motivated by, or identify with, success in the
academic domain, the use of Wise schooling practices may work. “Wise” is the
label used by the gays and lesbians in the 1950s to identify heterosexual people
who understand the gay/lesbian issues beneath the social stigma associated with
their sexual orientation (Goffman, 1963).
Wise strategies employed on college campuses fall into three categories: 1)
for students who both identify with success in the academic domain, and for those
who do not, strategies include mentoring, relationships with faculty who are
optimistic about stigmatized student’s potential, substituting challenging projects
for remediation, and having the college students expand their identification with the
academic domain by mentoring elementary school children (Cohen, Steele and
Ross, 1997).
For academic domain identified students, strategies include: exposing them
to role models from the stigmatized group who are successful academically,
explicitly stated values supporting multiple approaches to academic success, and

70
affirmation of belonging despite the negative stereotype threat, e.g., belonging to
the academic domain despite the low ability stereotype attributed to African
Americans (Steele, 1997).
Finally, for the students who do not identify with academic success, two
efficacy-building strategies are employed: the use of a Socratic teaching strategy,
referred to as nonjudgmental responsiveness (Lepper, Woolverton, Mumme and
Gurtner, 1993), that uses limited feedback on right and wrong, and limited direct
praise that minimizes the cost of failure and gradually builds efficacy for these
students who are often poor academic performers, and who do not identify with
school; and focusing on building self efficacy and competence of the students
(Bandura, 1977, 1986).
Jigsaw Classroom
Aronson’s jigsaw classroom (Aronson & Gonzalez, 1988; Aronson,
Stephan, Sikes, Blaney & Snapp, 1978) leverages the use of superordinate goals as
a way to gain cooperation among students to successfully achieve the goals. The
jigsaw method of classroom instruction is a self-esteem building method that draws
attention away from the teacher and the competition for teacher attention because it
divides students into multiethnic groups and gives each group member a different
section of the lesson to learn and then teach to the rest of the group. Students are
allowed to meet with students from other groups assigned to the same section of the
lesson, and then students within each group teach the others their portion of the

71
lesson. In addition to building self esteem, the jigsaw classroom research results
revealed improved test performance and more favorable views about school.
Micro-Inequities Training
In a growing number of Fortune 500 companies, executives and employees
alike are experiencing how events of small, and even trivial, acts of bias can have a
large impact on the confidence, comfort and productivity of those who are the
targets of bias. Similar to Sartre’s “small acts of Anti-Semitism” (1946), Dr. Mary
Rowe at MIT (1990) coined the term “micro-inequities” to describe how people
communicate bias in subtle, often unconscious and unintentional ways, to another
person, often someone who is different. The messages are both verbal and
nonverbal in the form of looks, gestures and tone, and deliver the message that the
person is not valued. The cumulative effect of being ignored, treated differently,
etc., becomes a barrier to success that is hard to articulate to others. During the
training session, through vignettes and role plays, everyone experiences micro-
inequities. For example, the participants act out the morning ritual of a boss who
pleasantly chats at length with one subordinate, while simply saying good morning
to another. The interactions enable everyone to experience micro-inequities, i.e.,
how it feels to be discounted and devalued. One of the outcomes of the training is
that it gives everyone a common language to describe the subtle acts of bias when
they occur, thereby promoting a more open, inclusive and welcoming work
environment.

72
Meta-Analyses and Reviews on Mediating Factors
Stereotype threats are consciously experienced and are manifested through
the presence of anxiety, emotional distress, pressure and apprehension about being
evaluated. As mentioned earlier in the chapter, most often, stereotype threats
impact individuals in the stigmatized group who excel in the domain, who may be
among the highest achievers, possess high confidence in their own abilities, and
who have worked hard to disprove the stereotype (Steele, 1997). A number of
mediating factors have been tested to determine their impact on the stereotype
threat and the impact of the threats on performance.
In seeking the factors that appear to reduce the impact of stereotype threat
on performance, one mechanism enhances performance: activating a positive self-
stereotype is said to create a stereotype halo effect that boosts confidence and task
motivation, while reducing the stereotype threat, in this case, stereotypes about
women as negotiators (Kray, Thompson, & Galinsky, 2001).
The search to identify previous meta-analyses to identify treatments that
mediate stereotype activation revealed three meta-analyses that were conducted in
the last five years:
1. Wheeler and Petty (2001) conducted a meta-analysis on factors that
mediate stereotype activation.
2. Devos and Banaji (2003) conducted a review of implicit measures of self-
concept and identity to determine how self measures are related to social group
membership.

73
3. Hoffmann, Gawronski, Gschwendner, Le and Schmitt (2005) reviewed
studies to determine the correlation between the IAT and explicit self-report
measures.
A summary of these meta-analyses and their relevance to this meta-analysis is
presented next.
Meta-Analysis Findings on Factors to Mediate Stereotype Activation
Wheeler and Petty (2001) conducted a meta-analysis of the mechanisms
that cause stereotype activation to impact behavior. Their research sought to
determine if activating a stereotype leads to acting in a stereotypical way, and if the
behavioral response to stereotype activation is not the expected stereotype
response, what factors moderate the behavior.
The researchers evaluated 69 experimental studies that looked at two
primary features of stereotypes; the activated stereotype is evaluated as positive or
negative, and evaluated as a self-stereotype or a stereotype about other groups.
Once the stereotype was coded as positive/negative, and self/other, then the
behavioral effects were coded as assimilation effect, i.e. the behavior is consistent
with the stereotype, or coded as a contrast effect, wherein the behavior is
inconsistent with the stereotype. The research also distinguished between a hot
motivation “self” stereotype comprised of conscious feelings and motivation states
that are associated with stereotype threat, and a cold “other” stereotype that stems
from the ideomotor theory (Carpenter, 1874) of unconscious, automatically
activated cognitive processes that are mostly applied to stereotypes about others.

74
Wheeler and Petty expected that people would behave in ways that were
consistent with the stereotype but their findings did not consistently support that
hypothesis. 20 of the 25 self-stereotype experiments revealed that peoples’
behaviors were consistent with the stereotype and that in 20 of the 24 other
stereotype experiments the behavior was consistent with the stereotype. In total,
71% of the experiments found activated stereotypes lead to stereotype application.
This finding, coupled with the fact that the remaining 20 of the stereotype
activation experiments revealed no apparent effect on behavior, led the researchers
to conclude that there are mediating factors that were tested but did not produce
conclusive results in indicating the impact of primed stereotypes on performance.
These factors included: the presence of distracting thought; perceptions of test
bias; thoughts concerning academic performance and self-worth; state anxiety;
persistence; effort; perceived difficulty; perceived pressure or evaluation
apprehension; confidence and self-efficacy; performance expectancies or self-
perceptions of skills.
The meta-analysis revealed mixed results of other variables including
anxiety, distraction, time allocation, performance expectancies, guessing, and self-
handicapping (Wheeler & Petty, 2001, p. 806). Wheeler and Petty suggested
further study on the moderating conditions that prevent stereotype activation from
leading to stereotype application.
There are three major differences between the Wheeler and Petty meta-
analysis and this study. Wheeler and Petty’s meta-analysis provides a framework

75
for considering stereotype-based behavioral changes, and whether these changes
differ when comparing self or other stereotypes, positive or negative stereotypes,
and conscious or unconscious priming of stereotypes. This analysis led to the
identification of a large number of moderating factors. Rather than focusing on
what Wheeler and Petty’s meta-analysis focused on, i.e., whether stereotype
activation leads to stereotype application, this meta-analysis will look at the
mediating factors that may moderate the application of stereotypes, and will
specifically look at implicit, unconscious activation of stereotypes about African
Americans.
Second, the Wheeler and Petty meta-analysis included 6 self stereotype
studies involving consciously primed African American stereotypes, and 6 other
stereotype studies about African Americans, with 5 of the 6 studies using
consciously primed African American stereotypes. This study will focus on
unconscious priming of African American stereotypes, and the use of the IAT to
measure the extent to which stereotype activation occurs.
Third, the majority of the behavioral consequences of stereotyping
examined in the Wheeler and Petty study were related to academic performance
results. This meta-analysis will examine moderating factors that are not limited to
academic performance. In an effort to advance research in corporate environments,
this meta-analysis will focus on behaviors and performance that are applicable to,
or occur in corporate environments.

76
Implicit Self-Related Processes
One of the factors that is a fundamental underpinning of stereotyping is the
unconscious attitudes and beliefs that form our concept of self identity and how we
see ourselves. The degree to which our sense of self is static or dynamic is
important in determining whether interventions or treatments to prevent the
application of stereotypes are effective. Therefore, a brief description of some
research conducted to examine self-related processes is relevant to this meta-
analysis.
Devos & Banaji (2003) conducted a comprehensive critical review of
research on unconscious, self-related processes and how these processes relate to
the context of social group membership. The researchers explained that implicit
self and identity refer to processes that occur outside of conscious awareness
without conscious control, and that there are many ways that an individual may be
unaware of the sources that influence thoughts, feeling and behavior (p. 179).
Research examples presented include: evaluating the strength of the associations
between self and in-group or out-group (Smith, Coats & Walling, 1999); the study
of implicit national identity among U.S. citizens and foreigners (Devos & Banaji,
2003); and associating roles and attributes with gender categories (Greenwald,
Farnham, 2000; Lemm & Banaji, 1998).
The concept of implicit ethnocentrism, that is, an implicit preference
for the in-group in U.S. culture has been revealed to be a preference for White, rich,
American, straight, and Christian individuals (Cunningham, Nezlek & Banaji,

77
2001). Studies involving the activation of stereotypes unconsciously influence
social behavior, resulting in behaviors such as decrements in the memory of elderly
patients when subliminally activated negative stereotypes about the elderly were
introduced (Levy, 1996), priming participants with words related to rudeness or
politeness, resulting in more rude or more polite behavior (Bargh, Chen &
Burrows, 1996), or priming people with a subliminal presentation of Black faces
resulted in more hostile behavior on the part of both non-Black and Black
individuals (Chen & Bargh, 1977). Subsequent work has demonstrated that
subliminal messaging can result in behavior matching, ie., unconscious mimicry of
unconsciously induced behaviors of other people’s postures, facial expressions, etc.
(Chartrand & Bargh, 1999). This mimicry phenomenon is also called the
“Chameleon Effect” by Chartrand & Bargh (1999).
Devos & Banaji’s (2003) research reviewed several other studies about the
impact of implicit activation of stereotypes that have already been discussed in this
meta-analysis. One other important research area the researchers discussed is that
in some studies self descriptions shifted from a controlled to an automatic mode
when people are emotionally aroused (Paulhus & Levitt, 1987). In another study
that involved a selection task, the prime of encouraging people to rely on their
feelings lead to a preference to picking their own name letters and birth date
numbers out of an array of numbers and letters (Koole, Dijksterhuis & Van
Knippenberg, 2001).

78
Meta-Analysis of the Correlation Between the IAT and Explicit Measures
Hoffmann, Gawronski, Gschwendner, Le & Schmitt (2005) conducted an
extensive meta-analysis to examine the correlation between the IAT and explicit
self-report measures. Studies in this area reveal that the correlation between
implicit measures and explicit measures vary significantly, ranging from studies
that show a low correlation between the IAT and explicit measures (Blair, 2001;
Dovidio, Kawakami & Beach, 2001), those that show some correlation (Banse,
Seise & Zerbes, 2001), to those showing no correlation (Karpinski & Hilton, 2001).
The researchers identified five explanations for the variation in correlations
between implicit and explicit measures: 1. explicit measures are influenced by
social desirability motivations; 2. a lack of introspective access for explicit
measures may result in less awareness of subconscious biases that implicit
measures illicit; 3. the process of retrieving memories may impact explicit-implicit
correlations (for example, the correlation may be higher when people judge
spontaneously than when they are introspective); 4. method-related factors that are
brought out by characteristics of the implicit or explicit measure, such as
randomizing the order of trials, may reduce the correlation; and 5. the implicit and
explicit measures could be independent with no common constructs being assessed.
The meta-analysis coded 126 studies that met the following three criteria:
1. the study includes at least one IAT and one explicit self-report measure; 2. the
study reports IAT-explicit measure correlations precisely; and 3. there is no

79
duplication of data already reported in previously published work included in the
meta-analysis.
The meta-analysis sought to answer three questions: 1. Does the correlation
vary as a function of the topic under investigation? 2. Does the relationship
between the IAT and explicit measures vary as a function of characteristics of the
explicit measures? 3. Does the relationship between the IAT and explicit measures
vary as a function of characteristics of the IAT.
Twelve coders rated the studies and the meta-analysis process included
three major tasks: 1. estimate the mean and variance of the population correlation;
2. estimate the variance of population correlations, and 3. examine the effect of
potential moderators and the method-related factors that could influence the
direction (high or low) of the correlation. Seven topic categories were judged for
the influence of three moderators: social desirability, introspection, and
spontaneity.
For a total sample size of 12,289 participants, analysis of self-reported
representations, intentions, behaviors and demographics, and non-corresponding
self-reports revealed population correlations ranging from .011 to .471. The
finding revealed a reliable increase in correlations where an increase in spontaneity
in explicit judgments, but there was not a higher correlation with a higher level of
introspection, or an increase in the degree of social desirability.
The findings related to the study questions are briefly described next. The
response to the first study question was that the correlation does vary as a function

80
of the topic. The results indicate that for implicit and explicit measures, the
characteristics of the topic influenced the results. Above average correlations were
reported for the socially oriented topics of consumer research and group attitudes,
while lower correlations were reported for the more personal topics of stereotypes
and self esteem.
The results related to the second question reveal that the relationship
between the IAT and explicit measures does vary as a function of characteristics of
explicit tests, with affective self reports correlating with the IAT more than
cognitive self reports. The third study questions finding is that the relationship
between the IAT and explicit measures does vary as a function of the IAT. Studies
where the presentation of compatible and incompatible blocks of content was
varied, the results showed a significantly higher correlation when compared to the
studies that maintained a fixed order presentation.
Hoffman and colleagues (2005) provide a comprehensive analysis of the
IAT and explicit self-report measures across a broad spectrum of categories, and
their work provides insights with respect to the correlation between the IAT and
explicit measures. The correlation between the two measures is very broad, and
confounded by moderator variables. The role of stimuli or primes in the studies is
not addressed. In contrast, this meta-analysis focuses on one context, stereotyping,
and specifically looks at African American stereotypes. The Hoffman and
colleagues findings revealed that the IAT and explicit measure correlations for
categorical moderator variable of stereotyping were below average at .167. This

81
meta-analysis will further Hoffman and colleagues’ research by addressing the
issue of priming, the impact of what triggers the activation of the stereotype, and
what impact, if any, the correlation between the IAT and explicit measures
indicates about moderating the activation of stereotypes.
Summary of Moderating Factors
Armed with the mediating factors identified as producing inconclusive
results in impacting behaviors triggered by stereotype activation, it is apparent that
it is difficult to interrupt stereotype application once a stereotype has been
activated, and also that there are a number of ways to move from stereotype
activation to stereotype application. However, some promising results have been
realized in the dozen or so interventions that have been used with some success in
interrupting stereotype application. This study will synthesize a set of studies
aimed at addressing stereotypes about African Americans that use the IAT and
other implicit measures to interrupt behaviors that are in response to activated
stereotypes.
Eight moderating factors were reviewed in this chapter. For the purposes of
this meta-analysis, these factors have been grouped into three broad categories.
These categories will also be used to codify any other moderating factors that are
identified during the examination of studies for the meta-analysis.
Category 1: Social exposure. This category includes: interethnic
friendship, social influence, manipulation of social context and first impressions.

82
Category 2: Covert psychology, including shifting goal intentions, positive
expectancy, mental imagery, and mood alteration.
Category 3: Education and training, which includes: diversity education,
positive role modeling, and specific training methodologies such as the
aforementioned WISE strategy, the jigsaw classroom, and micro-inequities
training.
To summarize, this meta-analysis seeks to examine studies over the past
decade (1995-2005) that use treatments aimed at reducing or eliminating the
activation and application of African American stereotypes. Specifically, the
research exams the ways to moderate the negative effect of African American
stereotypes by evaluating the three categories of treatments described above: social
experience, covert psychology, and education and training.
This meta-analysis seeks to answer the following questions: Which
moderating factors apply to self stereotyping/stereotype threat, and which ones
apply to other stereotypes? Which moderating factors are applicable to corporate
environments? Are experimental designs that are comprised of implicit and
explicit measures more effective than designs with implicit-only measures?

83
CHAPTER 3. METHODS
This study is a meta-analysis of the treatments that modify or inhibit the
automatically activated stereotypes about African Americans that are revealed by
the IAT. The meta-analysis research approach is a systematic integration and
coding of research articles to provide a consistent, valid and replicable model
(Bland, Meurer and Maldonado, 1995). There is a very limited number of
individual studies that specifically employee the IAT to uncover automated
stereotyping about African Americans as a part of the research design, even when
explicit measures are also part of the design. Therefore the meta-analysis approach
to this study will increase the effect size of the interventions and garner statistical
significance for the results in a way that is not possible with individual studies.
Implicit measures of automated stereotyping is a new technology that has
received a significant amount of attention since the 1996 inception of the most
essential implicit measure of automated stereotyping, the IAT. The previous meta-
analysis that comes closest to addressing the topic of this study (Wheeler & Petty,
2001) did not provide conclusive evidence of the moderator and mediator variables
that interrupt stereotype application. A meta-analysis of the correlation between
the IAT and explicit self-report measures found that there is a low correlation for
stereotype and affective topics.
Procedure
This study will be conducted in four steps: a. locate studies conducted in
1996-2005 through computer searches of published studies in PsycInfo,

84
Educational Resource for Information Center (ERIC), ProQuest, Institute for
Science Information (Social Science), and Google Scholar. The identified studies
are cross-referenced with the previous meta-analyses conducted in 2001 and 2005.
Studies that meet the inclusion criteria are included in the meta-analysis; b. code
the studies based on the moderating variables identified for the study; c. conduct an
analysis and synthesis of study outcomes; and d. use statistical methods to measure
the effect size and the outcomes of the moderators as compared to the study
features.
Criteria for Inclusion/Exclusion
The studies included in this meta-analysis were limited to those that met the
following criteria:
• An empirical study that generates inferential statistics and contains data that
allows the calculation of effect size
• Conducted between 1996 and 2005
• Used the Black-White IAT, the current implicit measurement with validity
and reliability for identifying implicit stereotyping about African Americans
• Used an adult population (18 years or older)
• Includes a treatment or intervention to interrupt stereotype application
• Pre/post testing
Although the study focuses on stereotypes about African Americans, this
research is not limited to studies conducted in the U.S. Several studies regarding
African American stereotypes were conducted in Europe, and have been included

85
in the data collection (i.e., Egloff & Schmukle, 2002, Westerlundh, B. et al, 2004;
and Wickerts, 2005).
Studies were excluded from the meta-analysis if they did not include three
elements: the Black-White IAT, a treatment to reduce or eliminate stereotyping,
and a measure of the effectiveness of the treatment.
Data Sources
A variety of databases were used to conduct the search for studies of
automated stereotyping about African Americans that used the IAT. The
keywords and combinations of keywords used in the searches included: automated
stereotyping, IAT, African American stereotype modification, racial stereotype
modification, stereotype threat and implicit bias. The results of these searches
produced 134 articles about African American stereotype modification, and 522
articles that employ the IAT to measure the activation of a variety of stereotypes.
The number of IAT articles was reduced 330 to focus on Black-White IAT studies
as listed in Table 1.
Table 1. Data Sources
Database Search Results (# Articles)
APA Psycho Info 46
Emerald Management 33
ERIC 26
Google Scholar 105
Institute for Science Info 99
Psych Articles 21
Total 330

86
When the inclusion criteria were applied to the 330 articles and publications, six
studies were chosen for the meta-analysis that met all of the criteria (i.e., empirical
study, Black-White IAT, conducted between 1996 and 2005, adult study
population, and includes a treatment used to interrupt stereotyping.)
Study Design Factors
Coding sheets will capture the study categories and design considerations identified
in Table 2.
Table 2. Coding Sheet Description
Study Category Factor
Effectiveness • Data quality
• Pre-post test scores or treatment results
• Method reliability
Method used to identify
stereotypes
• Black-White IAT only
• Black-White and other IAT
• Combination IAT and other methods
• Explicit self-report questionnaire
• Experimenter as prime
• Other primes
• Interview
Sample Characteristics • Race
• Country
Stereotype Target • Self
• Other
Setting • Laboratory/University
• Corporate/industry
• Community/other
Performance Results • Academic performance
• Job skill
• Other
Moderating Interventions
(grouped into 3 broad
categories)
• Social exposure
• Covert psychology
• Education and training

87
Statistical Analysis
In keeping with commonly used meta-analysis research statistical analysis
(Glass, 1981; Rosenthal, 2001) the studies will be coded to measure the effect size
using standard deviations and percentages of standard mean difference. IAT
experiments report faster response times when pairings of pictures and words or
word-only (nouns and adjectives) reflect stereotypes. A preliminary review of the
IAT experiments used in this study reveals that many of the studies compute IAT
effect scores by comparing mean response latency, and conducting a regression
analysis. As the studies are reviewed and coded, additional statistical analysis
methods may be used.

88
CHAPTER 4. RESULTS
Overview
Six studies met the minimum criteria for inclusion in this meta-analysis.
The minimum criteria included the use of: (1) the Black-White IAT; (2) an
intervention to reduce or eliminate stereotyping; and (3) an effectiveness measure,
in this case, the effect size, i.e., a measure that will standardize the difference
between the control and experimental group. Table 3 provides an overview
description of the studies included in this meta-analysis.
Table 3. Meta-Analysis Studies Overview
Study # Citation and Study
Description
Intervention
Category
Stereotype
Identification
Method
Effectiveness
Measure
#1 Aberson, C.L.,
Shoemaker, C. &
Tomolillo, C. (2004).
Implicit bias and
contact: The role of
interethnic
friendships
Social
Exposure
Identify close
friendships with
African
Americans; 4
explicit bias
measures
IAT effect –
SD and mean;
1-African
American
friends or 2-No
African
American
Friends
#5a Olson, M.A. & Fazio,
R.H. (2004).
Reducing the
influence of extra-
personal associations
on the implicit
association test:
Personalizing the IAT
Covert
Psychology
Personal labels
(“I” statements)
in IAT to
separate personal
from cultural
attitudes
1-Traditional
vs.
2-Personalized
IAT
#5b Same as #5a;
replicate #5a plus
change task-order and
randomly assign
personalized or
traditional IAT
Covert
Psychology
Same as #5a Same as #5a
#8a Kim, D. (2003).
Voluntary
Education and
Training
Preliminary IAT
with suggestion
1-Pre vs. 2-Post

89
Study # Citation and Study
Description
Intervention
Category
Stereotype
Identification
Method
Effectiveness
Measure
controllability of the
implicit association
test (IAT);
to fake unbiased
results
#8b Same as #8a;
replicate faking with
strategy
Education and
Training
Preliminary IAT
with suggestion
to fake results
and given faking
strategy
1-Pre vs. 2-Post
#12a
(combined
with 12b)
Dasgupta, N. &
Greenwald, A.G.
(2001).
On the malleability of
automatic attitudes:
Combating automatic
prejudice with images
of admired and
disliked individuals
Social
Exposure
Pro-Black images
vs. Pro-White or
nonracial images
1-Nonracial vs.
2-Pro-Black
#12b
(combined
with 12a)
Same sample group
used as #12a, except
tested effects 24
hours later. The
average effect size of
12a & 12b is used in
the meta-analysis.
Social exposure Pro-Black images
vs. Pro-White or
nonracial images
1-Nonracial vs.
2-Pro-Black

The meta-analysis addresses the effectiveness of “interventions” (or
moderating variables) in reducing the automatic activation of African American
stereotypes and prejudice. While there is a large body of research on the IAT
effect, i.e., using the IAT to detect the automaticity of stereotyping, there are fewer
studies that examine the interaction of an intervention on the IAT effect. This is
the same approach used in the meta-analysis on the correlation between the IAT
and explicit self-report measures (Hofmann, et. al, 2005). When the studies are
limited to those that examine the effect of interventions on the Black-White IAT for

90
control vs. experimental groups, the number of studies is drastically reduced even
further.
Only four articles met the qualifications for inclusion in the meta-analysis.
Three of these articles (study numbers 5, 8 and 12) contained two studies within the
article for a total of seven studies that were included in the meta-analysis. For
studies 12a and 12b, the same sample is used. Scores for 12a are taken
immediately after the intervention and scores for 12b are taken 24 hours after the
intervention. The average of the two scores is included in the meta-analysis as one
study. Therefore, a total of six studies were used for comparison in the meta-
analysis.
Study sample sizes ranged from 24-109. Because of the wide range of the
sample sizes, an unbiased Cohen’s d (Cohen, 1988) was computed to get the
average sample size. The overall unbiased estimate of Cohen’s d is .43. The 95%
confidence interval ranges from .274 to .686, i.e., 95% of the time, the Cohen’s d
for the intervention will range from .274 to .686, based on an arithmetic average of
d=.48. Including all six studies, the effect size is homogeneous across studies
( χ2(5)=2.41, p=.780), meaning that the studies selected for this meta-analysis are
likely to be testing the same hypothesis. This study focused on the effect of
interventions on the IAT effect by looking at the difference in the IAT scores
between the control and experimental groups for which an intervention is applied.

91
Additional Descriptive Statistics
The average standard deviation for the mean IAT effects was determined. The
average SD ranged from 109.985 to 227.265. Table 4 presents the mean IAT effect
and SD for the comparison groups 1 (Control) and 2 (Experimental) of each study.
Table 4. Descriptive Statistics for the IAT Effect (*e.g., IAT_Mean1 is the mean
IAT effect for group 1)

ID Author(s) Groups N *IAT_
Mean1
IAT_
SD1
IAT_
Mean2
IAT_
SD2
Average
SD
1 Aberson,
Shoemaker
&
Tomolillo
1=No African
American
Friends
2= African
American
Friends

70
231.6 190.8 147.4 164.6 177.7
5a Olson &
Fazio
1= Traditional
2= Personalized

73

163.17 121.5 131.6 105.8 113.65
5b Olson &
Fazio
1= Traditional
2= Personalized

109

209.39 103.24 166.93 116.73 109.985
8a Kim 1= Pre, 2= Post 24 185.15 113.89 151.53 172.3 143.095
8b Kim 1= Pre, 2= Post 24 181.49 171.47 -32.21 283.06 227.265
12 Dasgupta
&
Greenwald
(Average
12a &
12b)
1= Nonracial
Exemplars
2= Pro-Black
Exemplars

48
150
Not
given
64.5
Not
given
--------

Summary of Study Mean Effect Size
Table 5 summarizes the study mean effect sizes and lists the w-reciprocal of
variance of d calculation result. Effect sizes range from .23 to .94. Overall, in this
study, the mean effects range from small (.23 - .39) to moderate (.47 - .58). The
exception is the effect size for study 8b which was found to be large (> .80) at .94.
This appears to be an outlier score, and the effect size for this study is substantially

92
larger because this study involved instructing the participants to respond slowly
during the IAT. As a result, this intervention led to a reversed effect – a slightly
positive bias towards African Americans. The IAT is intended to measure
automatic activation of stereotypes and attitudes. Therefore, it may be questionable
whether this reflects automatic activation if participants are consciously thinking of
their responses. This may be interpreted as creating a controlled process.
Table 5. Summary of Study Mean Effect Size
ID Authors Effect Size (d) w
1
1
Aberson, Shoemaker, &
Tomolillo
0.47 17.02
5a Olson & Fazio 0.28 18.07
5b Olson & Fazio 0.39 26.75
8a Kim 0.23 5.96
8b Kim 0.94 5.40
12 Dasgupta & Greenwald 0.58 11.51

Effects of Moderating Intervention Category
The mean effect size was sorted by moderating intervention category to
determine if there is a significant effect size difference. As Table 6 shows, the
mean effect size for the social exposure category and the education and training
category were in the moderate range (.53 and .58 respectively), while the effect size
for the covert psychology category was in the small range at .34.

1
Reciprocal of variance of d

93
Table 6. Intervention Category Mean Effect Size
Intervention Category Mean Effect Size for
Intervention Category
Social Exposure (1 & 12) .53
Covert Psychology (5a & 5b) .34
Education & Training (8a & 8b) .58

Frequencies for Reducing the IAT Effect with Intervention
The frequency for the IAT effect to be reduced by an intervention is
presented in Table 7. Using the Cohen’s d to look at the percent gain in reducing
the IAT effect, the percent gain revealed ranges from 18.15% to 117.75%. The
outlier result of 117.75% is found in study 8b, the study that increased the IAT
effect by using a faking strategy.
Table 7. Frequencies for % Reduction in IAT Effect with Intervention
Study ID Authors
% Reduction
in IAT Effect
1
Aberson,
Shoemaker, &
Tomolillo
36.36
5a Olson & Fazio 19.35
5b Olson & Fazio 20.28
8a Kim 18.16
8b Kim 117.75
12
Dasgupta &
Greenwald
57.00

94
Table 8 summarizes the IAT effect reduction frequency results as moderate
with a mean of 44.8146 and standard deviation of 38.73089.
Table 8. Descriptive Statistics - % Reduction in IAT Effect with Intervention
Statistic

N

Minimum

Maximum

Mean

Std.
Deviation
% Reduction in
IAT effect with
intervention
6 18.16 117.75 44.8146 38.73089

95
CHAPTER 5. DISCUSSION
Overview
The use of meta-analysis methodology seeks to evaluate the empirical
findings within a body of literature in order to provide conclusive evidence in
support of the research questions posed. This small and well-defined meta-analysis
examined research that used both the Black-White IAT and a treatment to reduce or
eliminate the effects of stereotypes about African Americans. Given the results of
this study, the findings indicate that treatments to reduce stereotypes have a low to
moderate (.23-.58 mean effects) impact on reducing unfavorable IAT responses
toward African Americans. The percent gain is moderate at 44.81%.
As reported in the literature review chapter, three related meta-analyses
were conducted that served to reduce the scope of this meta-analysis. The Wheeler
and Petty (2001) meta-analysis analyzed 69 studies to examine 17 factors that
could mediate stereotype activation, and found that stereotype activation impacted
behavior in 71% of the published studies. Devos & Banaji (2003) reviewed
research on four unconscious self-centered and other-centered processes across 155
articles, and concluded that some implicit attitudes, beliefs and motives about
oneself can be changed unconsciously through both psychological and social
processes. Hoffmann, Gawronski, Gschwendner, Le & Schmitt (2005) analyzed
126 studies to examine the correlation between the IAT and explicit self-report
measures and found a low to moderate (.011-.471) correlation between them.

96
At the outset of this study, the identification and review of a large number
of studies appeared to provide results that support the hypothesis that unconscious
stereotypes about African Americans can be interrupted before they are acted upon.
The popularity of implicit stereotype research over the last ten years is quite
evident in the vast array of IAT experiments and empirical research to find
conclusive evidence that strategies to reduce prejudice are effective. However,
once the minimum criteria of this study were applied to the studies identified in the
literature review, the total useable studies were reduced to six.
The sample sizes of the studies used in the meta-analysis were relatively
small, ranging from 24-109. The number of participants across the six studies
included in the meta-analysis totaled 348. Despite the small number of studies
compared in the meta-analysis, some empirical results were obtained that will
further the research to find ways to reduce the negative impact of automated
stereotyping.
There was sufficient data in the meta-analysis to examine the results by
treatment category (social exposure, cognitive psychology, and education and
training.) In the category of social exposure, exposure to interethnic friendships
(Aberson, Shoemaker & Tomolillo, 2004) and admired and disliked Black and
White exemplars ( Dasgupta and Greenwald, 2001) had moderate effect sizes (.47
and .58 respectively). The effects of exposure to admired and disliked exemplars
went further to include scores both immediately after the intervention and 24 hours

97
later. IAT images of pro-Black exemplars and disliked White exemplars resulted in
reduced pro-White preference.
In the category of covert psychology, two experiments conducted by Olson
& Fazio (2004) examined the potential contamination of IAT results by comparing
the results of the traditional IAT with a “personalized” version they developed that
used more personal language, such as “I do” and “I don’t” statements. Low effect
sizes (.28 and .39) were reported. An argument could be made that this
personalization of the IAT, i.e., to put more emphasis on the personal feelings and
reactions of the individual as opposed to focusing on the culturally acceptable
responses, may not be considered an intervention. However, because it is the
experimental version that is compared to the traditional IAT, this study was
included.
For the third treatment category, education and training, Kim (2003)
examined whether you can fake IAT results, both with a general directive that says
a pro-Black attitude can be faked, and with a specific strategy to slow down
responses in order to fake results. The effect sizes varied considerably between the
study with no strategy for faking (d=.23) and the study that provides a faking
strategy (d=.94) which instructs the participant to slow down their reaction.
Interestingly, the study participants were able to slow down IAT responses in what
is considered the easy condition (i.e., White and pleasant), but they were not able to
speed up IAT responses in the difficult condition (i.e., White and unpleasant.)

98
Based on the studies examined in this meta-analysis, the three treatment
categories can be summarized as follows, in order of effectiveness. The social
exposure category, with the interventions of interethnic friendships and exposure to
admired and disliked Black and White exemplars, are the most effective
interventions strategies, as represented by the moderate effect sizes. The education
and training category involving the suggestion of faking IT results to appear pro-
Black resulted in very high results (d=.94) when a specific faking strategy was
provided, but low effect size (d=.23) without the strategy. The extreme effect size
difference is noteworthy because the high effect size suggests that priming
interventions, such as the additional faking strategy instructions that interrupt
stereotyping, contaminates the IAT results. In this experiment, faking pro-Black
attitudes by slowing the response time acts as a conscious strategy for altering the
unconscious, automated stereotyping the IAT measures, i.e., it turns the
unconscious response into a conscious response.
The last intervention category, covert psychology, produced low effect
sizes. As mentioned earlier, the use of personalizing language is questionable as an
intervention to the IAT effect because it is an alteration of the IAT itself, and not an
intervention outside of the IAT.
Research Question Findings
A discussion regarding the five research questions posed for this meta-
analysis follows.

99
Question 1: Are unconscious, implicit measures of automated stereotyping more
reliable than conscious, explicit measures?
In this meta-analysis, the review of several studies containing both implicit
and explicit measures of anti-Black bias did not consistently address reliability.
Therefore, the question of whether implicit measures are more reliable than explicit
measures could not be answered. As corroborated by the meta-analysis to
determine the correlation between implicit and explicit measures (Hofmann, et al,
2005), measures of reliability are not applied to implicit measures. Table 9 shows
the number of studies reviewed in this meta-analysis that include reports of
reliability for implicit and explicit measures.
Table 9. Studies Reporting Reliability of Implicit and Explicit Measures
Implicit Measure Explicit Measure
Reliability reported 3 4
No reliability reported 11 9

Three findings from previous research with implicit measures to identify
unconscious stereotyping still hold true: (1) The IAT reliably uncovers implicit
bias, but does not predict behavior (Karpinski & Hilton, 2001); (2) In studies that
contain both implicit and explicit measures, implicit bias exists even when explicit
bias does not exist (Dovidio, Kawakami, Johnson, Johnson & Howard, 1997;
Devos & Banaji, 2003) and (3) the motivation to respond without prejudice is
exhibited in studies where the IAT results show high levels of race bias while the

100
explicit self-report surveys show low levels of race bias (Devine, Ashby-Plant,
Amodio, Harmon-Jones and Vance, 2002).
Question 2: Can the application of stereotypes be interrupted once a stereotype is
automatically activated?
There is some evidence that stereotype activation can be slowed or
interrupted before behavioral responses occur. The moderating factors that were
described in the meta-analysis results included social exposure experiences that
influenced whether a stereotype would be acted upon, including interethnic
friendships with stereotype target individuals, and exposure to admired individuals
of the stereotype target group. Education and training aimed at unlearning the
stereotypes, and faking non-stereotypical responses (Kim, 2003) also prohibited
stereotype application.
Several other methods of interrupting stereotype application were described,
including: establishing a shared social reality with the target group (Lower, Hardin,
Sinclair, 2001), first impression management (Abreu, 1999), shifting goal
intentions (Gollwitzer, Schaal, Moskowitz, Hammelbeck & Wasel, 1999), positive
priming (Kirsch & Lynn, 1999 and Kawakami, Dovidio, Moll, Hermsen & Russin,
2000), mental imagery (Blair, Ma, & Lenton, 2001), introducing counter-
stereotypic information (Hamilton & Sherman, 1994), and various forms of
education and training interventions.
Unfortunately, the reported number of successful attempts to de-activate
stereotypes is not compelling. In reviewing the 35 studies that specifically focused

101
on attempts to modify stereotypes about African Americans, 9 of the studies did not
use a clear intervention strategy or did not provide complete data on the moderating
effect.
In addition, the meta-analysis requirement that the Black-White IAT is one
of the minimum criteria used to determine which studies are analyzed in this meta-
analysis limited the studies to those that are conducted in an academic, laboratory
environment.
Question 3: Can you automate the act of counteracting stereotyping as a
performance goal?
With few exceptions, the studies reviewed and analyzed in this meta-
analysis did not address the degree to which counteracting automated stereotyping
can be established as a performance goal. In the Kim study (Kim, 2003), the
argument could be made that a performance goal was established to use a faking
strategy to improve IAT results. The result of this experiment was that IAT scores
increased significantly (effect size = .23 without the faking strategy and .94 with
the faking strategy.) However, what the faking strategy experiment measured was
improving the IAT effect results, not at reducing the act of stereotyping.
Question 4: Which interventions are most effective in corporate or work settings?
The overall results of this meta-analysis did not reveal specific information
that leads to recommendations for which interventions are most effective in
corporate or work settings. There were only two studies that were conducted in a
non-academic environment: the study of first impression management for British

102
therapists regarding the treatment of Black patients (Abreu, 1999); and the
stereotypes about self and others revealed by Black and White working-class flea
market patrons (Niemann, O’Connor & McClorie, 1998). None of the other studies
included recommendations for generalizing the moderating interventions to a
corporate or work environment.
Despite the lack of non-academic settings included in this meta-analysis,
the prospect of being able to transfer the experimental designs and modifying
treatments from a classroom or laboratory setting to a corporate work environment
is a desirable goal. Therefore, Table 10 provides a speculative list of the
moderating interventions for reducing or eliminating racism that could be used in a
corporate setting, and the conditions under which the intervention would be
successfully implemented:
Table 10. Stereotype Prevention Interventions for Corporate Settings
Intervention Conditions for Success
Meta-analysis interventions:
1. Interethnic friendships Pre-existing friendships, diversity awareness
2. Admired and disliked
exemplars
Internal communications promoting diversity
champions and positive exemplars
Other Interventions:
3. Shared social reality Corporate values include diversity, cross-
cultural focus
4. First impression management Tools for behavioral modeling
5. Shifting goal intentions Behavioral goals to demonstrate inclusive
behaviors
6. Positive priming Publicize non-stereotypic role models
7. Diversity education Diversity as standalone training and
integrated into other training
8. Micro-inequities training Baseline of known micro-inequities in the
corporate culture

Table 11 presents the interventions identified in this meta-analysis that are
aimed at reducing stereotyping that would not work in a corporate setting and the
reasons why they would not work:
Table 11. Stereotype Prevention Interventions That Will Not Work in a Corporate
Setting

Intervention Reason for Exclusion from Corporate Setting
Meta-analysis interventions:
1. Personalization Personalization isolates personal attitudes from
cultural attitudes, but does not account for
corporate culture as a third influence on attitudes
2. Faking non-stereotypic
thoughts
Improves IAT effect result but promotes
inappropriate behavior (faking)
Other Interventions:
3. Mental imagery Temporary attitude change; no IAT effect results
reported
4. Shifting mood Temporary behavior change; no IAT effect
results reported
5. WISE education method Raises stereotype awareness; no IAT effect
results reported
6. Jigsaw classroom Designed for classroom, non-adult environment
At a lower level of impact, there is a set of moderating “factors” that could
mediate stereotype application. As reported in the Wheeler & Petty (2001) meta-
analysis, these factors include: distracting thoughts, perception of test bias,
academic performance and self-worth concerns, state anxiety, persistence, effort,
perceived difficulty, perceived pressure or evaluation apprehension, confidence and
self-efficacy, performance expectancies, self perception skills, time allocation, and
guessing. It seems plausible to take these factors into account when implementing
interventions to counteract stereotyping in a corporate environment. These factors
103

104
were described in the review of literature. However, the studies used in the meta-
analysis did not specifically address any of these moderating factors.
Question 5: Is there evidence that interventions aimed at modifying stereotypes
have long term effectiveness?
The findings of this meta-analysis did not include any references to the long
term effectiveness of the moderating interventions. One study, the Dasgupta and
Greenwald (2001) study examined voluntary controllability of IAT results by first
administering a priming exercise with pictures of famous and infamous Black and
White Americans to determine if presenting positive Black exemplars and negative
White exemplars reduced the typical pro-White results of the IAT. The results
showed that introducing admired Black and disliked White Americans reduced the
automatic preferences for White individuals on the IAT immediately after the
exemplars exercise, and 24 hours later the IAT effect size results remained low,
increasing only slightly from .60 and .68.
Summary and Conclusions
Summary
The goal of this meta-analysis was to synthesize research on implicit
stereotyping of African Americans to determine if there are interventions that
inhibit stereotypes. In addition, a secondary goal was that this meta-analysis would
reveal stereotype reduction interventions which could be generalized to the
corporate/work environment settings.

105
With the proliferation of studies over the last 10 years that used the IAT and
focused on implicit methods to uncover unconscious prejudice and bias, it was
hoped that these studies would provide quantitative evidence that stereotype
application can be inhibited. However, when the criteria for this meta-analysis
were applied to the literature base, a small number of studies met the criteria. This
is contrary to the aim of a meta-analysis, which is to combine study results across a
large number of studies to increase the statistical power of the results. Despite the
limitation of the small number of studies, the meta-analysis results find that there is
a low to moderate effect size for interventions to interrupt stereotypes as reported
on the Black-White IAT.
With respect to the five research questions of this meta-analysis, three
questions could not be answered due to a lack of information on these topics.
Question 1 sought to compare the reliability of implicit vs. explicit measures.
Reliability is not uniformly addressed in the studies examined in this meta-analysis.
Question 3 addressed whether the act of counteracting stereotyping can be
automated as a performance goal. The studies reviewed did not provide evidence
to support using treatments to inhibit stereotypes as a performance goal. Question
5 sought to find evidence of the long-term effectiveness of stereotype-inhibiting
interventions, but no longitudinal evidence was provided. Only the study
comparing the impact of pro-Black vs. pro-White or nonracial images tested the
effects immediately after the experiment and 24 hours later (Dasgupta &
Greenwald, 2001).

106
The remaining two meta-analysis research questions can be answered to
some degree. Question 2 asked if a stereotype can be interrupted once it is
automatically activated. In this meta-analysis, the intent was to determine if there
are interventions that can prevent stereotypes, once activated, from being acted
upon, or to reduce their effects. This study shows that interventions to interrupt
stereotype application have low to moderate effectiveness. Question 4 sought to
generalize the interventions analyzed in this study to a corporate setting. The
setting for all but two studies was an academic/laboratory environment, and all six
of the studies in this meta-analysis were conducted in an academic/laboratory
environment. The lack of research in corporate settings alludes to the difficulty in
gaining access and permission to use corporate environments for social science
research. Recommendations were made regarding the potential to use some of the
identified interventions in a corporate setting.
Based on these findings, and the study limitations described, this meta-
analysis provides some evidence that stereotypes about African Americans can be
reduced.
Conclusions
Implicit measures, the IAT effect, and the impact of stereotypes on behavior
have all been the topics of other studies. This meta-analysis attempted to tackle all
three topics, and to measure the interaction between the Black-White IAT and other
treatments. Low to moderate effect sizes were reported for the six meta-analysis
studies. The narrowly-defined scope of this meta-analysis resulted in reducing the

107
number of studies that could contribute to the findings from 330 to 6. The study
design combination of a sole focus on African Americans as the stereotype target,
the requirement that the Black-White IAT be used, and the use of an intervention to
moderate African American stereotyping.
Moreover, in an effort to make the number of intervention strategies more
manageable, the interventions were grouped into three categories. Ultimately, the
categorization of interventions into three categories was not as informative as it
might have been because such a small number of studies met the study criteria (six
studies were divided into three categories.)
Considerations for Future Research
Here are some opportunities for future research:
1. There is a need for research that tests the use of stereotype interventions
in non-academic settings. In particular, the role that social interaction and
education/training interventions play in work settings could make valuable
contributions to improving human interactions of individuals and teams, and also
inform human resource policy development and enforcement.
2. Harness the widespread use of the IAT on the internet by supplementing
how the IAT is administered. Develop web-based research designs that use the
moderating interventions described in this meta-analysis in conjunction with the
IAT.
3. Continue implicit stereotype research that focuses on race. The early
priming methods and the IAT initially began with non-human subjects such as

108
flowers and food choices. IAT subjects now include race, gender, sexual
orientation, age and disability. There is much to be learned and unlearned about
how stereotypes lead to prejudice and discrimination, in ways that limit people’s
ability to make a valuable contribution to their organizations, their families and
mankind.
4. The research on stereotype threat is an important area of implicit
stereotypes that warrants further study. The roles of self-efficacy, self-worth,
ethnocentrism and other inwardly-focused psychological constructs play a role in
how the stimulus for responding to bias, prejudice and discrimination is formed,
and how resistance and resiliency become mechanisms to avert these responses.
In closing, here are some final thoughts about subject of this meta-analysis.
Throughout the review of research, it was striking to note the seemingly endless
array of moderating variables, both conscious and unconscious, that may cause a
person to refrain from responding to others based on implicit stereotypes. It is
encouraging to see the many possibilities for ways to reduce stereotyping.
However, consistent, quantitative data was lacking that would enable comparisons
and correlations across the studies. The same holds true for some of the IAT
studies where the millisecond response times were recorded, but not reliability and
validity data, or the means, standard deviations and effect size data that would
enhance the statistical power of the results. Hopefully, stereotype reduction
research will continue, and future research designs will give equal weight to the
results of both the IAT and other modifying treatments.

109
REFERENCES

References marked with an asterisk indicate studies included in this meta-analysis

*Aberson, C.L., Shoemaker, C. & Tomolillo, C. (2004). Implicit bias and contact:
The role of interethnic friendships. Journal of Social Psychology, 144 ( 3),
335-347.

Abreu, J. (1999). Conscious and nonconscious african american stereotypes:
Impact on first impression and diagnostic ratings by therapists. Journal of
Counseling, 67, 387-393.

Aboud, F.E. (1988). Children and Prejudice. New York: Blackwell.

Adorno, T.W., Frenkel-Brunswik, E., Levinson, D.J. & Sanford, R.N. (1950). The
Authoritarian Personality. New York: Harper.

Ajzen, I., & Fishbein, M. (1980). Understanding Attitudes and Predicting Social
Behavior. Englewood Cliffs, NJ: Prentice-Hall.

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and
Human Decision Processes, 50, 179-211.

Allport, F.H. (1924). Social Psychology. New York: Houghton-Mifflin.

Allport, G.W., & Kramer, B.M. (1946). Some roots of prejudice. Journal of
Psychology, 22, 9-39.

Allport, M.D. (1954). The Nature of Prejudice. New York: Doubleday.

Ambady, N., Shih, M., Kim, A. & Pittinsky, T. L, (2001). Stereotype
susceptibility in children: Effects of identify activation on quantitative
performance, Psychological Science, 12 (5), 385-390.

American Association of University Women (1992). How schools shortchange
girls: The AAUW report: A study of major findings on girls education.
Washington, DC: American Association of University Women Educational
Foundation.

110
Arndt, J., Greenberg, J., Schimel, J. Pyszczynski, T & Solomon, S. (2002). To
belong or not to belong, that is the question: Terror management and
identification with gender and ethnicity. Journal of Personality and Social
Psychology, 83(1), 26-43.

Aronson, E., Blaney, N., Stephan, C. Sikes, J. & Snapp, M. (1978). The Jigsaw
Classroom. Beverly Hills, CA: Sage Publishing Company.

Aronson, E., & Gonzalez, A. (1988). Desegregation jigsaw and the Mexican-
American experience: A clear and concise description of the jigsaw
classroom technique for reducing overt prejudice and discrimination. In
Katz, P. & Taylor, D., Eliminating Racism: Profiles in Controversy, 301-
314. New York: Plenum Press.

Aronson, J., Fried, C.B., & Good, C. (2001). Reducing the effects of stereotype
threat on african american college students by shaping theories of
intelligence. Journal of Experimental Social Psychology, 38, 113-125.

Aronson, J., Lustina, M.J., Good, C., Keough, K., Steele, C.M. & Brown, J. (1999).
When white men can’t do math: Necessary and sufficient factors in
stereotype threat. Journal of Experimental Social Psychology, 35, 29-46.

Asendorpf, J.B., Banse, R. & Mucke, D. (2002). Double dissociation between
implicit and explicit personality self-concept: The case of shy behavior.
Journal of Personality and Social Psychology, 83(2), 380-393.

Ashburn-Nardo, L., Knowles, M.L. & Monteith, M.J. (2003). Black american’s
implicit racial associations and their implications for intergroup judgment.
Social Cognition, 21 (1), 61-87.

Ashburn-Nardo, L., Voils, C.I., & Monteith, M.J. (2001). Implicit associations as
the seeds of intergroup bias: How easily do they take root? Journal of
Personality and Social Psychology, 81 (5), 789-799.

Banaji, M.R., & Greenwald, A.G. (1995). Implicit gender stereotyping in
judgments of fame. Journal of Personality and Social Psychology, 68, 181-
198.

Banaji, M.R., & Hardin, C.D. (1996). Automatic stereotyping. Psychological
Science, 7 (3), 136-141.

Bandura, A. (1995). Self-Efficacy in Changing Societies. Cambridge: Cambridge
University Press.

111

Bandura, Albert. (1997). Self-Efficacy: The Exercise of Control. New York:
W.H. Freeman & Company.

Banse, R., Seise, J. & Zerbes, N. (2001). Implicit attitudes towards homosexuality;
Reliability, validity and controllability of the IAT. Zeitschrift fur
Experimentelle Psychologie, 48, 145-160.

Bargh, J.A. (1992). Does subliminality matter to social psychology: Awareness of
the stimulus versus awareness of its influence. Perception with Awareness,
Bornstein, R.F. & Pittman, T.S. (Eds), 236-255, New York: Guilford Press.

Bargh, J.A. (1999). The cognitive monster: The case against the controllability of
automatic stereotype effects. Dual-Process Theories in Social Psychology,
Chaiken, S., & Trope, Y. (Eds). New York, NY: Guilford Press, 361-382.

Bargh, J.A. & Chartrand, T.L. (1999). The unbearable automaticity of being.
American Psychologist, 54 (7), 462-479.

Bargh, J.A., Chen, M, & Burrows, L. (1996). Automaticity of social behavior:
direct effects of trait construct and stereotype activation on action. Journal
of Personality and Social Psychology, 71, 230-244.

Bargh, J.A., Gollwitzer, P.M., Lee-Chai, A., Barndollar, K., & Trotschel, R.
(2001). The automated will: Nonconscious activation and pursuit of
behavioral goals. Journal of Personality and Social Psychology, 81, 1014-
1027.

Blair, I.V. (2001). Implicit stereotypes and prejudice. In. G. B. Moskowitz (Ed.),
Cognitive Social Psychology. Mahwah, NJ: Lawrence Erlbaum.

Blair, I.V. & Banaji, M.R. (1996). Automatic and controlled processes in
stereotype priming. Journal of Personality and Social Psychology, 70 (6),
1142-1163.

Blair, I. V., Ma, J. E. & Lenton, A. P. (2001). Imagining stereotypes away: The
moderation of implicit stereotypes through mental imagery. Journal of Personality
and Social Psychology, 81 (5), 828-841.

Bland, C.J., Meurer, L.N., Maldonado, G. (1995). A systematic approach to
conducting non-statistical meta-analysis of research literature. Journal of
Academic Medicine, 70, 7, 642-653.

112
Blanton, H., Pelham, B.W. & Dehart, T. (2001). Overconfidence as dissonance
reduction. Journal of Experimental Social Psychology, 37 (5), 373-385.

Bodenhausen, G.V. & Macreae, C.N. (1998). Stereotype activation and inhibition.
Advances in Social Cognition, Wyer, R.S. (Ed.), 11, 1-52. Mahway, NJ:
Erlbaum.

Brendl, C.M., Markman, A.B. & Messner, C. (2001). How do indirect measures
of evaluation work? Evaluating the inference of prejudice in the implicit
association test. Journal of Personality and Social Psychology, 81 (5), 760-
772.

Brewer, M.B., (1979). In-group bias in the minimal intergroup situation: A
cognitive-motivational analysis. Psychological Bulletin, 86 (2), 307

Brewer, M.B. (1988). A dual process model of impression formation. In R. Wyler
& T. Srull (Eds.), Advances in Social Cognition, vol. 1, pp. 1-36, Hillsdale,
NJ: Erlbaum.

Brewer, M.B. (1991). The social self: On being the same and different at the same
time. Personality and Social Psychology Bulletin, 17, 475-478.

Brewer, M.B., Dull, V., & Lui, L. (1981). Perceptions of the elderly: Stereotypes
as prototypes. Journal of Personality & Social Psychology, 41, 656-670.

Brickman, P., Bulman, R.J. (1977). Pleasure and pain in social cognition. In J.M.
Suls & R.L. Miller (Eds.), Social comparison processes; Theoretical and
empirical perspectives (pp. 149-186). Washington, D.C.: Hemisphere.

Byars, A.M. & Hackett, G. (1998). Applications of social cognitive theory to the
career development of women of color. Applied & Preventive Psychology,
7, 255-267.

Bylsma, W.H., Major, B., & Cozzarelli, C. (1995). The influence of legitimacy
appraisals on the determinants of entitlement beliefs. Basic and Applied
Social Psychology, 17, 223-237.

Carpenter, W.B., (1874). Principles of mental psychology with their applications to
the training and discipline of the mind and the study of its morbid
conditions. New York: Appleton.

113
Carver, C.S., Glass, D.C., & Katz, I. (1977). Favorable evaluations of blacks and
the disabled: Positive prejudice, unconscious denial, or social desirability?
Journal of Applied social Psychology, 8. 97-106.

Chartrand, T.L., & Bargh, J.A. (1996). Automatic activation of impression
formation and memorization goals: Nonconscious goal priming reproduces
effects of explicit task instructions, Journal of Personality and Social
Psychology, 71, 468-478.

Chartrand, T.L. & Bargh, J.A. (1999). The chameleon effect: The perception-
behavior link and social interaction. Journal of Personality and Social
Psychology, 76, 893-910.

Chen, M. & Bargh, J.A. (1997). Nonconscious behavioral confirmation processes:
The self-fulfilling nature of automatically activated stereotypes. Journal of
Experimental Social Psychology, 33, 541-560.

Coats, S., Claypool, H. M., & Smith, E. R. (2000) . Overlapping mental
representations of self and in-group: Reaction time evidence and its
relationship with explicit measures of group identification. Journal of
Experimental Social Psychology, 36 (3), 304-315.

Cohen, G., Steele, C.M., & Ross, L. (1997). Giving feedback across the racial
divide: Overcoming the effects of stereotypes. Unpublished manuscript,
Stanford University.

Cose, E. (1993). Rage of a privileged class. New York: Harper Collins.

Crano, W. D., & Chen, X. (1998) . The leniency contract and persistence of
majority and minority influence. Journal of Personality and Social
Psychology, 74 (6), 1437-1450.

Crocker, J., Cornwell, B., & Major, B. (1993). The stigma of overweight:
affective consequences of attributional ambiguity. Journal of Personality &
Social Psychology, 64, 60-70.

Crocker, J., & Major, B., (1989). Social stigma and self-esteem: The self-
protective properties of stigma. Psychological Review, 96, 608-630.

Crocker, J., Major, B. and Steele, C. (1998). Social stigma. In Gilbert, D. T.,
Fiske, S. T. and Gardner, L, (Eds.), The Handbook of Social Psychology,
(fourth edition), Volume II, 28, 504-553. Boston, MA: McGraw-Hill.

114
Crocker, J., Thompson, L., McGraw, K., & Ingernan, C. (1987). Downward
comparison, prejudice and evaluation of others: Effects of self-esteem and
threat. Journal of Personality and Social Psychology, 52, 907-916.

Crocker, J., Voelkl, K., Testa, M., & Major, B. (1991). Social stigma: The
affective consequences of attributional ambiguity. Journal of Personality
and Social Psychology, 60, 218-228.

Crosby, F., Bromley, S. and Saxe, L. (1980). Recent unobtrusive studies of black
and white discrimination and prejudice: a literature review. Psychological
Bulletin, 87, 546-563.

Cullen, M.J., Hardison, C.M., & Sackett, P.R., (2004). Using SAT-grade and
ability-Job performance relationships to test predictions derived from
stereotype threat theory. Journal of Applied Psychology, 89 (2), 220-230.

Cunningham, W. A., Preacher, K. J., & Banaji, M.R. (2001). Implicit attitude
measures: Consistency, stability, and convergent validity. Psychological
Science, 12 (2), 163-170.

Cunningham, W.A., Nezlek, J.B. & Banaji, M.R. (2001). Conscious and
unconscious ethnocentrism: Revisiting the ideologies of prejudice.
Manuscript submitted for publication.

Dasgupta, N., McGhee, D. E., Greenwald, A. G., & Banaji, M.R. (2000) .
Automatic preference for white americans: Eliminating the familiarity
explanation. Journal of Experimental Social Psychology, 36 (3), 316-328.

*Dasgupta, N., & Greenwald, A. G. (2001) . On the malleability of automatic
attitudes: Combating automatic prejudice with images of admired and
disliked individuals. Journal of Personality and Social Psychology, 81 (5),
800-814.

Dawes, R.M. (1976). Shallow psychology. In Carroll, J.S. & Payne, J.W. (Eds.),
Cognition and Social Behavior. Hillsdale, NJ: Erlbaum.

De Houwer, J. (2001) . A structural and process analysis of the implicit association
test. Journal of Experimental Social Psychology, 37 (3), 443-451.

Deaux, K., & Lewis, L.L. (1984). Structure of gender stereotypes:
Interrelationships among components and gender label. Journal of
Personality and Social Psychology, 46, 5, 991-1004.

115
Devine, Patricia G. (1989a). Stereotypes and prejudice: Their automatic and
controlled components. Journal of Personality and Social Psychology, 56
(1), 5-18.

Devine, Patricia G. (1989b). Automatic and controlled processes in prejudice: the
role of stereotypes and personal beliefs, in A.R. Pratkanis, S.J. Breckler and
A.G. Greenwald (Eds.), Attitude Structure and Function, 181-212,
Hillsdale, NJ: Erlbaum.

Devine, P. G. (2001) . Implicit prejudice and stereotyping: How automatic are
they? Introduction to the special section. Journal of Personality and Social
Psychology, 81, 757-759.

Devine, P. G., Ashby-Plant, E., Amodio, D.M., Harmon-Jones, E., & Vance, S. L.
(2002) . The regulation of explicit and implicit race bias: The role of
motivation to respond without prejudice. Journal of Personality and Social
Psychology, 82, 835-848.

Devine, P.G., Monteith, M.J., Zuwerink, J.R., & Elliot, A.J. (1991). Prejudice with
and without compunction. Journal of Personality and Social Psychology,
60, 817-830.

Devos, T. & Banaji, M.R. (2003). Implicit self and identity. Annals New York
Academy of Sciences. 1001: 177-211, New York: Guilford Press.

Dion, K.L., & Earn, B.M. (1975). The phenomenology of being a target of
prejudice. Journal of Personality and Social Psychology, 32, 944-950.

Dovidio, J., Evans, N, & Tyler, R. (1986). Racial stereotypes: the contents of their
cognitive representation, Journal of Experimental Social Psychology, 22,22-
37.

Dovidio, J.F. and Gaertner, S.L. (1993). Stereotypes and evaluative intergroup
bias. In D.M. Mackie and D. L. Hamilton (Eds.) Affect, Cognition, and
Stereotyping, 167-193, San Diego, CA: Academic Press.
Dovidio, J.F., Kawakami, K., & Beach, K.R. (2001). Implicit and explicit
attitudes: Examination of the relationship between measures of intergroup
bias. In Brown, & Gaertner, S.L. (Eds.), Blackwell Handbook of Social
Psychology: Intergroup Processes (pp. 175-197). Malden, MA: Blackwell.

Dovidio, J. F., Kawakami, K., Johnson, C., Johnson, B. & Howard, A. (1997). On
the nature of prejudice: automatic and controlled processes. Journal of
Experimental Social Psychology, 33 (5), 510-540.

116

Dunton, B.C., & Fazio, R.H. (1997). An individual difference measure of
motivation to control prejudiced reactions. Personality and Social
Psychology Bulletin, 23, 316-326.

Egloff, B., & Schmuckle, S. C. (2002) . Predictive validity of an implicit
association test for assessing anxiety. Journal of Personality and Social
Psychology, 83 (6), 1441-1455.

Ehrlich, H.J. (1973). The social psychology of prejudice. New York: Wiley.

Fanon, F. (1952, 1967). Black Skins, White Masks. New York: Grove

Fazio, R. H. (2003) . Implicit measures in social cognition research. Annual
Review of Psychology, 54, 297-327.

Fazio, R.H. (1986). How do attitudes guide behavior? In R.M. Sorrentino & E.T.
Higgins (EDS.), Handbook of motivation and cognition, vol. 1 pp. 204-243.
New York: Guilford Press.

Fazio, R.H. (1990). Multiple processes by which attitudes guide behavior: The
MODE model as an integrative framework. In M.P. Zanna (Ed.), Advances
in Experimental Social Psychology, 23, 75-109, San Diego: Academic
Press.

Fazio, R.H., & Dunton, B.C. (1997). Categorization by race: the impact of
automatic and controlled components of racial prejudice. Journal of
Experimental Social Psychology, 33, 451-470.

Fazio, R.H., Jackson, J.R., Dunton, B.C., & Williams, C. J. (1995). Variability in
automatic activation as an unobtrusive measure of racial attitudes: A
bonafide pipeline? Journal of Personality and Social Psychology, 69, 1013-
1027.

Fazio, R.H., & Olson, M.A. (2003). Implicit measures in social cognition research:
Their meaning and use. Annual Review of Psychology, 54, 297-327.

Fazio, R.H., Sanbonmatsu, D.M., Powell, M.C., & Kardes, F.R. (1986). On the
automatic activation of attitudes. Journal of Personality and Social
Psychology, 50, 229-238.

117
Fazio, R.H. and Towles-Schwen, T. (1999). The MODE model of attitude
behavior processes, in Chaiken, S. & Trope, Y (Eds), et al. Dual-Process
Theories in Social Psychology. New York, NY: Guilford Press, 97-116.

Fiske, S.T. (1989). Examining the role of intent, toward understanding its role in
stereotyping and prejudice, in J.S. Uleman & J.A. Bargh (Eds.), Unintended
Thought, 253-283. New York, NY: Guilford Press.

Fiske, S. T. (1998). Stereotyping, prejudice, and discrimination, in The Handbook
of Social Psychology, (fourth edition), Volume II (Gilbert, D.T., Fiske, S.T.
& Gardner, L., 25, 357-410. New York, NY: McGraw-Hill.

Florack, A., Scarabis, M., & Bless, H. (2001). When do associations matter? The
use of implicit association towards ethnic groups in person judgments.
Journal of Experimental Social Psychology, 37, 518-524.

Gagne, E.D. (1993). The Cognitive Psychology of School Learning (2
nd
Ed), New
York: Harper Collins.

Gawronski, B. (2002). What does the implicit association test measure? A test of
the convergent and discriminant validity of prejudice-related IATs.
Experimental Psychology, 49 (3), 171-180.

Gerard, H. (1983). School desegregation: The social science role. American
Psychologist, 38, 869-878.

Gilbert, D.T. and Hixon, J.G. (1991). The trouble of thinking: Activation and
application to stereotypic beliefs, Journal of Personality and Social
Psychology, 60, 509-517.

Goffman, E. (1963). Stigma: Notes on the management of spoiled identity.
Englewood Cliffs, NJ: Prentice-Hall.

Gollwitzer, Peter M. (1999). Implementation intentions: Strong effects of simple
plans, American Psychologist, 54 (7), 493-503.

Gollwitzer, P.M., & Moskowitz, G. (1996). Goal effects on thought and behavior.
In E. T. Higgins & A. Kruglanski (Eds.), Social Psychology: Handbook of
Basic Principles (pp.361-399). New York: Guilford Press.

Gollwitzer, P.M., Schaal, B., Moskovitz, G.B., Hammelbeck, H.J.P. & Wasel, W.
(1999). Implementation intention effects on stereotype and prejudice
activation. Manuscript submitted for publication.

118

Greenwald, A.G., & Banaji, M.R. (1995). Implicit social cognition: Attitudes,
self-esteem and stereotypes, Psychological Review, 102, 4-27.

Greenwald, A. G., & Farnham, S. D. (2000). Using the implicit association test to
measure self-esteem and self-concept. Journal of Personality and Social
Psychology, 79 (6), 1022-1038.

Greenwald, A.G., McGhee, D.E. & Schwartz, J.L.K. (1998). Measuring individual
differences in implicit cognition: The Implicit Association Test, Journal of
Personality and Social Psychology, 74 (6), 1464-1480.

Hamilton, D.L. (1981). Stereotyping and intergroup behavior: Some thoughts on
the cognitive approach. In Hamilton, D.L. (Ed.), Cognitive Processes in
Stereotyping and Intergroup Behavior, 333-353, Hillsdale, NJ: Erlbaum.

Hamilton, D. & Sherman, J (1994). Stereotypes. In Wyer, R.S. & Srull, T.K.
(Eds.), Handbook of Social Cognition (2
nd
ed), 1-68. Hillsdale, NJ:
Eribaum

Hamilton, D., Sherman, S., & Ruvolo, C. (1990). Stereotype-based expectancies:
Effects on information processing and social behavior. Journal of Social
Issues.
46, 35-60.

Harding, J.B., Kutner, B., Proshansky, H., & Chein, I. (1954). Prejudice and ethnic
relations. In G. Lindzey (Ed.), Handbook of Social Psychology (2
nd
ed,,
Vol. 5, pp. 1-76). Reading, MA: Addison.

Hay Group (2005). Hay Group/MLA 2005 Salary Survey.

Herr, P.M. (1986). Consequences of priming: Judgment and behavior. Journal of
Personality & Social Psychology, 51, 1106-1115.

Hirschfeld, L.A. (1996). Race in the making. Cambridge, MA: MIT Press.

Hoffman, W., Gawronski, B., Gschwendner, T., Huy, L., & Schmitt, M. (2005). A
meta-analysis on the correlation between the implicit association test and
explicit self-report measures. Journal of Personality and Social Psychology,
31, 10, 1369-1385.

119
Hummer, M. L., Garstka, T. A., Greenwald, A. G., Mellott, D. S., & O’Brien, L. T.
(2002) . Using the implicit association test to measure age differences in
implicit social cognitions. Psychology and Aging, 17, 482-491.

Institute of Management & Administration (2004). Complete Guide to HR Cost
Control Best Practices.

Inzlicht, M., & Ben-Zeev, T. (2000). A threatening intellectual environment: Why
females are susceptible to experiencing problem-solving deficits in the
presence of males. Psychological Science, 11, 365-371.

Jacoby, L. L. (1998). Invariance in automatic influence of memory: Toward a
user’s guide for the process-association procedure. Journal of Experimental
Psychology, 24, 3-26.

Jacoby, L., & Kelly, C. (1992). Unconscious influences of memory: Dissociations
and automaticity. In Milner, A.D. & Rugg, M.D. (Eds.), The
Neuropsychology of Consciousness (pp. 201-233). New York: Academic
Press.

Jarvie, G.J., Lahey, B., Graziano, W., & Framer, E. (1983). Childhood obesity and
social stigma: What we know and what you don’t know. Development
Review, 3, 237-273.

Jetter, J., Spears, R. & Manstead, A.S.R. (1998). Defining Dimensions of
Distinctiveness. Journal of Personality and Social Psychology, 74 (6), 1481-
1494.

Johnson, J.D., Trawalter, S. & Dovidio, J.F. (2000). Converging interracial
consequences of exposure to violent rap music on stereotypical attributions
of blacks. Journal of Experimental Social Psychology, 36 (3), 233-251.

Jones, E.E., & Davis, K.E. (1965). From acts to dispositions: The attribution
process in person perception. In Berkowitze, L. (Ed.), Advances in
Experimental Social Psychology ,2, 220-266. New York: Academic Press.

Jones, E.E., Farina, A., Hastorf, A.H., Markus, H., Miller, D.T., & Scott, R.A.
(1984). Social Stigma: The Psychology of Marked Relationships. New
York: Freeman.

Jones, J.M., (1972). Prejudice and Racism. Reading, MA: Addison-Wesley.

120
Jong, P.J. de, Pasman, W., Kindt, M., Van den Hout, M. (2001). A reaction time
paradigm to assess (implicit) complaint specific dysfunctional beliefs:
Behavior Research and Therapy, 39, 1, 101-113.

Jost, J.T. & Banaji, M.R. (1994). The role of stereotyping in system justification
and the production of false consciousness. British Journal of Psychology,
33, 1-27.

Jost, J. T., & Thompson, E. P. (2000). Group-based dominance and opposition to
equality as independent predictors of self-esteem, ethnocentrism, and social
policy attitudes among African Americans and European Americans.
Journal of Experimental and Social Psychology, 36, 209-232.

Karpinski, A., & Hilton, J. L. (2001). Attitudes and the implicit association test.
Journal of Personality and Social Psychology, 81, 774-787.

Katz, P.A. (1976). The acquisition of racial attitudes in children. In P.A. Katz
(Ed.). Towards the Elimination of Racism (pp. 125-154). New York:
Pergamon Press.

Katz, I., Wackenhut, J., & Hass, R.G. (1986). Racial ambivalence, value duality
and behavior. In J.F. Dovidio & S.L. Gaertner (Eds.), Prejudice,
Discrimination and Racism (pp. 35-59). Orlando, FL: Academic Press.

Kawakami, K., Dovidio, J.F., Moll, J., Hermsen, S., & Russin, A., (2000). Just say
no (to stereotyping). Journal of Personality and Social Psychology, 78 (5),
871-888.

Kawakami, K., Young, H. & Dovidio, J.F., (2002). Automatic stereotyping:
category, trait, and behavioral activations. Personality and Social
Psychology Bulletin, 28, 3-15.

Kelly, H.H. (1967). Attribution theory in social psychology. In Levine, D. (Ed.),
Nebraska Symposium on Motivation, 15, 192-2380. Lincoln: University of
Nebraska Press.

Kenworthy, J. B., & Miller, N. (2002). Attributional biases about the origins of
attitudes. Journal of Personality and Social Psychology, 82 (5), 693-707.

*Kim, D. (2003). Voluntary Controllability of the Implicit Association Test (IAT).
Social Psychology Quarterly, 66 (1), 83-96.

121
Kirsch I. & Lynn, S.J., (1999). Automaticity in clinical psychology. American
Psychologist, 54 (7), 504-515.

Koole, S.L., Dijksterhuis, A. & Van Kippenberg, A. (2001). What’s in a name:
Implicit self-esteem and the automatic self. Journal of Personality and
Social Psychology, 80, 669-685.

Kray, L.J., Thompson, L. & Galinsky, A. (2001). Battle of the sexes: Gender
stereotype confirmation and reactance in negotiations. Journal of
Personality and Social Psychology, 80, 942-958.

Lalonde, R.N., & Gardner, R.C. (1989). An intergroup perspective on stereotype
organization and processing. British Journal of Social Psychology, 28, 289-
303.

Langer, E.J. (1978). Rethinking the roles of thought in social interaction. In
Harvey, J.H., Ickes, W.J., & Kid, R.F. (Eds.), New Directions in
Attribution Research, (vol. 2, pp. 35-58). Hillsdale, NJ: Erlbaum.

Langer, E.J. & Abelson, R.P. (1972). The semantics of asking a favor: How to
succeed in getting help without really dying. Journal of Personality and
Social Psychology, 24, 26-32.

Le Fevre, J.A., Kulak, A.G., & Heymans, S.L. (1992). Factors influencing the
selection of university majors varying in mathematical content. Canadian
Journal of Behavioral Science, 24, 276-289.

Lemm, K.M. & Banaji, M.R. (1998) Implicit and explicit gender identity and
attitudes toward gender. Paper presented at the annual meeting of the
Midwestern Psychological Association, Chicago, IL.

Leonardelli, G., Brewer, M. B., & Geoffrey, J. (2001) . Minority and majority
discrimination: when and why. Journal of Experimental Social Psychology,
37, 468-485.

Lepper, M.R., Woolverton, M., Mumme, D.L. & Gurtner, J.L. (1993), Motivational
techniques of expert human tutors; Lessons for the design of computer-
based tutors. In Lajore, S.P. & Derry, S.J. (Eds.), Computers as Cognitive
Tools, 75-104. Hillsdale, NJ: Erlbaum.

Levy, B. (1996). Improving memory in old age through implicit self-stereotyping.
Journal of Personality and Social Psychology, 71, 1092-1107.

122
Levy, G. D. (2000) . Individual differences in race schematicity as predictors of
african american and white children’s race-relevant memories and peer
preferences. The Journal of Genetic Psychology, 161 (4), 400-419.

Leyens, J.P., Desert, M. Croizet, J.C. & Durcis, C. (2000). Stereotype threat. Are
lower status and history of stigmatization preconditions of stereotype
threat? Personality and Social Psychology Bulletin, 26, 1189-1199.

Lieberman, M.D., Ochsner, K.N., Gilbert, D.T. & Schacter, D. L. (2001). Do
amnesiacs exhibit cognitive dissonance reduction? The role of explicit
memory and attention in attitude change, Psychological Science, 135-140.

Livingston, R.W. (2002). The role of perceived negativity in the moderation of
African Americans implicit and explicit racial attitudes. Journal of
Experimental Social Psychology, 38, 405-413.

Lowery, B. S., Hardin, C. D., & Sinclair, S. (2001) . Social influence effects on
automatic racial prejudice. Journal of Personality and Social Psychology, 81
(5), 842-855.

Maass, A., Milesi, A., Zabbini, S. & Stahlberg, D. (1995). Linguistic intergroup
bias; Differential expectancies or in-group protection? Journal of
Personality and Social Psychology, 68, 116-126.

Macrae, C.N., Bodenhausen, G.V. Milne, A.B., Thorn, T.M.J. & Casstelli, L.
(1997).
On the activation of social stereotypes: The moderating role of processing
objectives. Journal of Experimental Social Psychology, 33, 471-489.

Macrae, C.N., Milne, A.B., & Bodenhausen, G.V. (1994). Stereotypes as energy-
saving devices; A peek inside the cognitive toolbox. Journal of
Personality and Social Psychology, 66, 37-47.

Macrae, C.N., Stangor, C. & Milne, A.B. (1994). Activating stereotypes: A
functional analysis: Journal of Experimental Social Psychology, 30, 370-389.

Major, B., & Crocker, J. (1993). Social Stigma: The affective consequence of
attributional ambiguity. In Mackie, D.M. & Hamilton, D.L. (Eds.), Affect,
Cognition, and Stereotyping: Interactive Processes in Intergroup Perception
345-370. New York: Academic Press.

123
Manis, M., Nelson, T.E. & Shedler, J. (1998). Stereotypes and social judgment:
Extremity, assimilation and contrast. Journal of Personality and Social
Psychology, 55, 28-36.

Marques, J.M. & Yzerbyt, V.Y. (1988). The black sheep effect: Judgmental
extremity towards ingroup members in inter- and intra-group situations.
European Journal of Social Psychology , 18, 287-292.

McConnell, A. R. and Leibold, J. M. (2001). Relations among the implicit
association test, discriminatory behavior, and explicit measures of racial
attitudes, Journal of Experimental Social Psychology, 37, 435-442.

McConahay, J. B. (1986). Modern racism, ambivalence, and the modern racism
scale, in J.F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination and
racism, 91-126, Orlando, FL: Academic Press.

McConahay, J.B., & Hough, J.C., Jr. (1976). Symbolic racism. Journal of Social
Issues, 32, 23-45.

McDougall, W. (1920). The Group Mind: A Sketch of the Principles of Collective
Psychology with Some Attempt to Apply Them to the Interpretation of
National Live and Character. New York: G. Putnam & Sons.

McFarland, S.G. & Crouch, Z., (2002). A cognitive skill confound on the implicit
association test. Social Cognition, 20 (6), 483-510.

McKay, P.F., Doverspike, D., Bowen-Hilton, D. & McKay, Q.D. (2003). The
effects of demographic variables and stereotype threat on black/white
differences in cognitive ability test performance. Journal of Business and
Psychology, 18 (1), 1-14.

Mierke, J., & Klauer, K.D. (2003). Method-specific variance in the Implicit
Association Test. Journal of Personality and Social Psychology, 85, 1180-
1192.

Mindiola, T., Rodriguez, N. & Niemann, Y.F. (1996). Intergroup Relations
between Hispanics and African Americans in Harris County. Houston, TX:
Center for Mexican American Studies.

Minnis, H., Gillies, M., McMillan, A., & Shubulade, S. (2001). Racial stereotyping:
Survey of psychiatrists in the United Kingdom. British Medical Journal,
323, 905-906.

124
Monteith, M.J., Voils, C.I & Ashburn-Nardo, L. (2001). Taking a look
underground: Detecting, interpreting and reacting to implicit racial biases.
Social Cognition, 19 (4), 395-417.

Morgan, S. L. & Mehta, J.D. (2004) . Beyond the laboratory: Evaluating the survey
evidence for the disidentification explanation of black-white differences in
achievement. Sociology of Education, 77 (1), 82-101.

Mulligan, N. W. (1998) . The role of attention during encoding in implicit and
explicit memory. Journal of Experimental Psychology, 24 (1), 27-47.

Nieman, Y.F., O’Connor, E. & McClorie, R. (1998). Intergroup stereotypes of
working class blacks and whites: Implications for stereotype threat. Western
Journal of Black Studies, 22, 103-108.

Nieman, Y. F. (1999) . The making of a token: A case study of stereotype threat,
stigma, racism, and tokenism in academe. Frontiers, 20 (1), 111-134.

*Olson, M.A. & Fazio, R.H. (2004). Reducing the influence of extrapersonal
associations on the implicit association test: Personalizing the IAT. Journal
of Personality and Social Psychology, 86 (5), 653-667.

Ottoway, S. A., Hayden D. C., & Oakes, M. A. (2001). Implicit attitudes and
racism: Effects of word familiarity and frequency on the implicit
association test. Social Cognition, 19, 97-144.

Paulhus, D.L. & Levitt, K. (1987). Desirable responding triggered by affect:
Automatic egotism? Journal of Personality and Social Psychology, 51, 245-
259.

Pelham, B.W. & Swann, W.B. (1989). From self-conceptions to self-worth: On the
sources and structure of global self-esteem. Journal of Personality & Social
Psychology, 57, 672-680.

Perugini, M. (2005). Predictive models of implicit and explicit attitudes. The
British Journal of Social Psychology, 44(1), 29-45.

Pettigrew, T.F., (1998). Intergroup contact theory. Annual Review of Psychology,
49, 65-85.

Pettigrew, T.F. (1979). The ultimate attribution error: Extending Allport’s
cognitive analysis of prejudice. Personality & Social Psychology Bulletin,
5, 461-476.

125

Phan, K.L, Wager, T., Taylor, S.F. & Liberzon, I. (2002). Functional neuroanatomy
of emotion: A meta-analysis of em0tion activation studies in PET and
FMRI. Neuro Image, 16, 133-348.

Plaks, J. E., & Higgins, T. (2000) . Pragmatic use of stereotyping in teamwork.
Journal of Personality and Social Psychology, 79, 962-974.

Polivy, J. & Herman, P.C. (2000). The false-hope syndrome: unfulfilled
expectations of self-change. Current Directions in Psychological Science, 9,
4, 128-131.

Pratto, F. Sidanius, J., Stallworth, L.M. & Malle, B.F. (1994). Social dominance
orientation: A personality variable predicting social and political attitudes.
Journal of Personality and Social Psychology, 67, 741-763.

Proshansky, H.M. (1966). The development of intergroup attitudes. In L.W.
Hoffman & M. L. Hoffman (Eds.), Review of Child Development Research,
2, 311-371. New York: Russell Sage Foundation.

Richeson, J.A. & Shelton, J.N., (2005). Brief report: Thin slices of racial bias.
Journal of Nonverbal Behavior, 29 (1), 75-86.

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

Rosenberg, M. (1965). Society and the Adolescent Self-image. Princeton, NJ:
Princeton University Press.

Rosenthal, R. & Jacobson, L. (1968). Pygmalion in the Classroom. New York:
Rinehart & Winston.

Ross, M., & Wilson, A.E. (2002). It feels like yesterday: Self esteem, valence of
personal past experiences, and judgments of subjective distance. Journal of
Personality and Social Psychology, 82 (5), 792-803.

Rothermund, K., & Wentura, D. (2004). Underlying processes in the implicit
association test: Dissociating salience from associations. Journal of
Experimental Psychology, 133 (2), 139-165.

Rowe, M.P. (1990). Barriers to equality: The power of subtle discrimination to
maintain equal opportunity. Employee Responsibilities and Rights Journal,
3 (2), 153-163.

126
Rudman, L. A., Ashmore, R. D., & Gary, M. L. (2001) . “Unlearning” automatic
biases: The malleability of implicit prejudice and stereotypes. Journal of
Personality and Social Psychology, 81, 856-868.

Rudman, L. A., Greenwald, A. G., Mellott, D. S., & Schwartz, J. L. K. (1999) .
Measuring the automatic components of prejudice: Flexibility and
generality of the implicit association test. Social Cognition, 17, 437-465.

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

Sackett, P. R., Hardison, C. M & Cullen, M. J., (2005). On interpreting stereotype
threat as accounting for african-american-white differences on cognitive
tests. American Psychologist, 59, 271-272.

Sartre, J.P. (1946, 1965). Anti-Semite and Jew. New York: Schocken Books.

Schacter, D.L. & Badgaiyan, R.D., (2001). Neuroimaging of priming: New
perspectives on implicit and explicit memory. Current Directions in
Psychological Science, 10 (1), 1-5.

Schmader, T., & Johns, M. (2003) . Converging evidence that stereotype threat
reduces working memory capacity. Journal of Personality and Social
Psychology, 85 (3), 440-452.

Schneider, M.E., Major, B. Luhtanen, R., & Crocker, J. (1996). When help hurts:
Social stigma and the costs of assumptive help. Personality and Social
Psychology Bulletin, 22, 201-209.

Sears, D.O. (1988). Symbolic racism, in P.A. Ikatz & D.A Taylor (Eds.),
Eliminating Racism: Profiles in Controversy, 53-84, New York, NY:
Plenum.

Segerstrom, S. C., Fahey, J. L., Kemeny, M. E., & Taylor, S. E. (1998). Optimism
is associated with mood, coping, and immune change in response to stress.
Journal of Personality and Social Psychology, 74 (6), 1646-1655.

Selye, H. (1956), The Stress of Life. New York: McGraw-Hill.

Sherif, M., Harvey, O.J., White, J. Hood, W. & Sherif, C.W. (1961). Intergroup
Conflict and Cooperation: The Robbers Cave Experiment. Norman, OK:
University Book Exchange.

127
Shiffrin, R.M., & Dumais, S.T. (1981). The development of automatism. In J.R.
Anderson (Ed.), Cognitive Skills and Their Acquisition, 111-140. Hillsdale,
NJ: Eribaum.

Smith, E.R., Coats, S. & Walling, D. (1999). Overlapping mental representations
of self, ingroup, and partner: Further response time evidence and a
connectionist model. Personality and Social Psychology Bulletin, 25, 873-
882.

Smith, J. L., & White, P. H. (2002) . An examination of implicitly activated,
explicitly activated, and nullified stereotypes on mathematical performance:
It’s not just a woman’s issue. Sex Roles, 47 (3-4), 179-191.

Spencer, S.J., Fein, S., Wolfe, C.T., Fong, C., & Dunn, M.A. (1998). Automatic
activation of stereotypes: The role of self-image threat. Personality and
Social Psychology Bulletin, 24, 1139-1152.

Spencer, S.J., Steele, C.M. & Quinn, D.M., (1999). Stereotype threat and women’s
math performance. Journal of Experimental Social Psychology, 35, 4-28.

Stangor, C. & Schaller, M. (1996). Stereotypes as individual and collective
representations. In Macrae, C.N., Hewstone, M. & Stangor, C. (eds.).
Foundations of Stereotypes and Stereotyping (pp. 3-37). New York:
Guilford Press.

Steele, C.M. (1997). A threat in the air: How stereotypes shape intellectual identity
and performance. American Psychologist. 52, 613-629.

Steele, C.M. (1995). Suspected by reputation: The stereotype vulnerability
disidentification and intellectual performance of women and African
Americans. Manuscript in preparation.

Steele, C.M & Aronson, J. (1995). Stereotype vulnerability and the intellectual test
performance of African Americans, Journal of Personality & Social
Psychology, 69, 797-811.

Steffens, M.C., (2004). Is the implicit association test immune to faking?
Experimental Psychology, 51 (3), 165-179.

Stephan, W.G., & Stephan, C.W. (1985). Intergroup anxiety. Journal of Social
Issues, 41, 157-175.

128
Stone, J., Darley, J. M., Lynch, C. L., & Sjomeling, M. (1999) . Stereotype threat
effects on black and white athletic performance. Journal of Personality and
Social Psychology, 77 (6), 1213-1227.

Tajfel, H. (1981). Human groups and social categories: Studies in Social
Psychology. Cambridge, England: Cambridge University Press.

Tajfel, H. & Turner, J.C. (1986). The social identity theory of intergroup behavior.
In S. Worchel & W.G. Austin (Eds.), Psychology of Intergroup Relations,
7-24, Chicago: Nelson-Hall.

Taylor, S.E. & Fiske, S.T. (1978). Salience, attention and attribution: Top of the
head phenomena. In L. Berkowitz (Ed.), Advances in Experimental Social
Psychology , 11, 249-288. New York: Academic Press.

Teachman, B. A., Brownell, K. D., & Gapinski, K. D., Rawlins, M. (2003) .
Demonstrations of implicit anti-fat bias: The impact of providing causal
information and evoking empathy. Health Psychology, 22 (1), 68-78.

Tesser, A. Millar, M., & Moore, J. (1988). Some affective consequences of social
comparison and reflection processes: The pain and pleasure of being close.
Journal of personality and Social Psychology, 54, 49-61.

Thomas, David A. & Gabarro, John J. (1999). Breaking Through: The Making of
Minority Executives in Corporate America. Boston: Harvard Business
School Press.

Tulving, E. & Schacter, D.L. (1990). Priming and human memory systems.
Science, 247, 301-306.

Uleman, J.S., & Moskowitz, G.B. (1994). Unintended effects of goals on
unintended inferences. Journal of Personality and Social Psychology, 66,
490-501.

Von Hippel, W., Sekaquaptewa, D. & Vargas, P. (1997). The linguistic intergroup
bias as an implicit indicator of prejudice. Journal of Experimental Social
Psychology, 33, 490-509.

Wegner, Daniel M., (1994). Ironic processes of mental control, Psychological
Review, 101 (1), 34-52.

129
Wegner, D.M, Erber, R., and Zanakos, S. (1993). Ironic processes in the mental
control of mood and mood-related thought, Journal of Personality and
Social Psychology, 65, 1093-1104.

Weiner, B. (1974). Achievement Motivation and Attribution Theory. Morristown,
NJ: General Learning Press.

Weiner, B. (1986). An Attributional Theory of Motivation and Emotion. New
York: Spring-Verlay.

Weiner, B. (1992). Human Motivation: Metaphors, Theories and Research.
Newbury Park, CA: Sage Publications.

Weldon, M. S., & Coyote, K. C. (1996). Failure to find the picture superiority
effect in implicit conceptual memory tests. Journal of Experimental
Psychology, 22, 670-686.

Westerlundh, B., Hansson, S.B., Bjorklund, F. & Backstrom, M. (2004). On the
nature and expression of ethnic prejudice as seen in judgments of pictorial
stimuli. Lund, Sweden: Lund University, Department of Psychology.

Wheeler, M.E. & Fiske, S.T. (2005). Controlling Racial Prejudice: Social-
Cognitive Goals Affect Amygdala and Stereotype Activation. Princeton,
NJ: Princeton University, 1-21.

Wheeler, S.C., Jarvis, W.B.G. & Petty, R.E. (2001). Think unto others: The self-
destructive impact of negative stereotypes. Journal of Experimental Social
Psychology, 37, 173-180.

Wheeler, S. C., & Petty, R. E. (2001) . The effects of stereotype activation on
behavior: A review of possible mechanisms. Psychological Bulletin, 127,
797-826.

Wicherts, J. M. (2005) . Stereotype threat research and the assumptions underlying
analysis of covariance. American Psychologist, 60 (3), 267-269.

Wills, T.A, (1981). Downward comparison in principles in social psychology.
Psychological Bulletin, 90, 245-271.

Wittenbrink, B., Judd, C. M., & Park, B. (2001) . Evaluative versus conceptual
judgments in automatic stereotyping and prejudice. Journal of
Experimental Social Psychology, 37, 244-315.

130
Wolfe, C. T., & Spencer, S. J. (1996) . Stereotypes and prejudice: Their overt and
subtle influence in the classroom. The American Behavioral Scientists, 40
(2), 176-185.

Woike, B., Lavezzary, E. & Barsky, J., (2001) . The influence of implicit motives
on memory processes. Journal of Personality and Social Psychology, 81 (5),
935-945.

Zellars, Kelly L. & Kacmar, K. Michelle. (1999). The influence of individual
differences on reactions to co-workers’ ingratiatory behaviors. Journal of
Managerial Issues, XL, 2, 234-238.

Zohar, D. (1999). When things go wrong: The effect of daily hassles on effort,
exertion and negative mood. Journal of Occupational and Organizational
Psychology, 72, 265-283.

Asset Metadata

Creator Reel, Sidalia Garrett (author)

Core Title A meta-analysis of interventions to modify stereotypes about African Americans

School Rossier School of Education

Degree Doctor of Education

Degree Program Education (Psychology)

Publication Date 04/11/2007

Defense Date 10/10/2006

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag African American stereotyping,automated stereotyping,black stereotyping,IAT,implicit bias,implicit stereotypes,OAI-PMH Harvest,stereotype interventions,stereotype modification

Language English

Advisor Clark, Richard E. (committee chair), Farrington, Jeanne (committee member), Howard, Keith (committee member)

Creator Email sid.reel@hp.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m364

Unique identifier UC1268175

Identifier etd-Reel-20070411 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-325107 (legacy record id),usctheses-m364 (legacy record id)

Legacy Identifier etd-Reel-20070411.pdf

Dmrecord 325107

Document Type Dissertation

Rights Reel, Sidalia Garrett

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email uscdl@usc.edu

Abstract (if available)

Abstract Since 1995, the Implicit Association Test (IAT) has been employed to reveal unconscious stereotyping. This meta-analysis examined studies that used the IAT in combination with treatments to eliminate automated stereotyping of African Americans and to determine which treatments were effective in corporate work settings. Of 134 studies identified, only 43 specifically addressed Black-White stereotyping and only 6 studies met the minimum criteria for inclusion in the meta-analysis (i.e., the use of the Black-White IAt and an intervention to reduce racial stereotyping.) Results revealed that interventions produced a moderate overall effect size of .48. The small number of studies prevents clear conclusions, yet it appears that existing African American stereotype-inhibiting interventions can reduce the effects of stereotypes by approximately 15 percent. None of the studies were conducted in work settings, but the most effective treatments, i.e., positive interethnic social contact and exposure to affirmative African American images, could be employed in work settings.