Content uploaded by Marc Bendick
Author content
All content in this area was uploaded by Marc Bendick on Oct 18, 2017
Content may be subject to copyright.
Bias in Hiring 1
Running Head: BIAS IN HIRING
February 27, 2010
This paper is an invited contribution to the issue
“The Reality of Contemporary Discrimination,”
forthcoming in the Journal of Social Issues, 2011
DEVELOPING THE RESEARCH BASIS FOR CONTROLLING BIAS IN HIRING
MARC BENDICK, JR.*
BENDICK AND EGAN ECONOMIC CONSULTANTS, INC., WASHINGTON, DC
ANA P. NUNES
DEPARTMENT OF PSYCHOLOGY AND NEUROSCIENCE, UNIVERSITY OF COLORADO AT BOULDER
* Bendick and Nunes, who are listed in alphabetical order, contributed equally to this paper. Correspondence
should be addressed to Marc Bendick, Jr., Bendick and Egan Economic Consultants, Inc., 4411 Westover Place NW,
Washington, DC 20016 USA [www.bendickegan.com; bendickegan@mindspring.com].
An earlier version of this paper was presented at the Monash University Conference on Field Experiments on
Discrimination in Markets, Prato, Italy (2005).
Bias in Hiring 2
Abstract
Discrimination based on race, ethnicity, national origin, gender, age, disability, gender
orientation, and other personal characteristics continues to distort employers’ hiring decisions
and thereby limit employment opportunities for historically excluded groups. Research in
psychology, sociology, economics, and management provides many insights concerning the
mechanisms of bias and interventions to mitigate their effects, but important questions remain
unanswered. The innovative research technique of matched pair testing offers laboratory-like
controlled conditions in quasi-experiments set in real-world hiring situations. We propose seven
types of matched pair testing studies to advance conceptual understanding of hiring bias and
improve hiring practices.
Bias in Hiring 3
DEVELOPING THE RESEARCH BASIS FOR CONTROLLING BIAS IN HIRING
Employers in America’s private sector decide who to hire about 60 million times a year
— more than 250,000 times each work day (U.S. Bureau of Labor Statistics, 2009). For
successful job candidates, the hiring process provides employment and also determines initial job
titles, work assignments, and wages, which in turn often influence that employee’s career for
years thereafter. For both successful and unsuccessful job applicants, it also provides job-
seeking experience, career information, and encouragement or discouragement. Few human
resource management processes rival hiring in terms of impact on the distribution of
employment opportunities and rewards.
Controlling bias that potentially distorts these high-stakes decisions has long been a goal
of American society. This concern is reflected in federal, state, and local laws against hiring
discrimination, as well as legal provisions for affirmative action in recruitment and hiring. It has
also motivated many employers to adopt highly structured hiring procedures (Bielby, 2000), train
hiring decision-makers (Bendick, Egan & Lofhjelm, 2001), and actively “manage workforce
diversity” (Kochan et al., 2003).
As Section II of this paper reviews, research in psychology, sociology, economics and
management offers many insights into the mechanisms of bias and the likely effects of
interventions to mitigate their effects. However, as Sections III and IV document, despite
considerable progress, bias continues to pervade the American hiring system. Accordingly, the
paper proposes additional research to support further changes in employer hiring practices.
Section V proposes four types of studies to expand conceptual understanding of hiring bias,
while Section VI proposes three types of studies to enhance the impact of such research on the
behavior of hiring practitioners.
Bias in Hiring 4
A number of other publications synthesize research on employment bias and identify
future research needs (e.g., Brief, 2008; Dipboye & Colella, 2005; Stockdale & Crosby, 2004).
The present paper differs from those reviews by focusing on hiring and one innovative research
methodology — matched pair testing — with unique potential for studying hiring. For readers
not familiar with matched pair testing, it is briefly described in an appendix.
II. Research-Based Predictions about Hiring Bias
The Civil Rights Movement of the 1960s was a reaction to widespread, blatant and
sometimes legally-codified discrimination against African-Americans and other demographic
groups. Correspondingly, ensuing anti-discrimination legislation — prominently, the Civil
Rights Act of 1964 (mandating equal opportunity based on race, gender and selected other
personal characteristics), Executive Order 11246 (establishing affirmative action), the Age
Discrimination in Employment Act, the Americans with Disabilities Act, and counterpart state
and local laws — aimed foremost to address employers’ conscious behavior. Violations of these
statutes are most commonly proven through documentation of discriminatory acts (e.g., use of
racial epithets in the workplace) or measurement of employment outcomes (e.g., few minority
employees when many qualified minority job applicants are available).
Behavioral science research raises important concerns about the limitations of this
approach. In particular, it questions the effectiveness of seeking to change employer behavior
without explicitly addressing the often-unconscious attitudes and perceptions underlying that
behavior. Equally, it suggests the need to improve specific employment “micro processes”
(such as interviewing, performance evaluation, or succession planning), not simply the
employment outcomes they help to shape. At least three lines of research raise such concerns
either explicitly or implicitly.
Bias in Hiring 5
Stereotypes Infect Us All
The first such line of research explores stereotypes and their role in discriminatory
behavior. Implicit bias –– unconscious association of specific traits with members of a
demographic group –– has been extensively measured with the Implicit Association Test (IAT)
(Greenwald, McGhee, and Schwartz, 1998; www.implicit.harvard.edu). Studies employing the
IAT use experimental subjects’ reaction times to illustrate that, regardless of their expressed or
self-perceived attitudes toward racial equality, individuals tend to associate negative traits with
racial minorities (Nosek, Banaji, & Greenwald, 2002; Nosek, Greenwald, & Banaji, 2005).
Those implicit associations, in turn, correlate with biased behavior (Dovidio, Kawakami, &
Gaertner, 2002). When such associations are activated in the hiring process, the predominantly
negative content of stereotypes about traditionally-excluded groups (e.g. African Americans are
uneducated; women are not career-committed), tends to handicap members of these groups in
competing for jobs.
Decades of related research has further elucidated the processes by which stereotypes
unconsciously influence how individuals are perceived and evaluated, processes central to
employment decisions such as hiring. For example, studies have documented that in evaluating
members of a stereotyped group, individuals pay more attention to information consistent with a
stereotype than to inconsistent information (Koomen & Dijker, 1998), interpret ambiguous
information to confirm stereotypes (Hilton & Van Hippel, 1996), seek out information
confirming stereotypes at a greater rate than that contradicting them (Erber & Fiske, 1984), and
are unaffected by information that a stereotype is invalid (Nelson, Acker & Manis, 1996).
Individuals also make memory errors consistent with stereotypes (Eberhardt, Dasgupta, &
Banaszynski, 2003), even when recalling objective facts such as scores on skill tests (Darley &
Bias in Hiring 6
Gross, 1980). Together, such processes explain how hiring decision-makers may honestly
perceive themselves as making unbiased selections that reflect objective differences in
applicants’ qualifications when in fact they have not.
Individuals Cannot Readily Counter Stereotypes
A second line of research demonstrates the difficulties that individual members of
stigmatized groups face when they attempt to ameliorate the adverse effects of negative
stereotypes about the groups to which they belong. For example, when an individual performs
in a way inconsistent with a stereotype, that performance gets discounted as reflecting
exceptional circumstances such as luck (Swim & Sanna, 1996). Furthermore, social congruity
theory (Eagly & Karau, 2002; see also Fiske, Bersoff, Borgida, Deaux, & Heilman, 1991)
predicts that members of a stereotyped group who behave or occupy social roles inconsistent
with a stereotype experience stronger adverse reactions from others than if they had conformed
to the stereotype.
Compounding these difficulties, biased treatment itself or anticipation of it, can adversely
affect the actual performance of employees or job applicants. For example, Word, Zamma, and
Cooper (1974; see also Shelton, Richeson, & Salvatore, 2005; Richeson, Trawalter, & Shelton,
2005) demonstrated that White job interviewers sat further away from African American job
applicants than White applicants, made less eye contact, and made more speech errors during
their interviews, and that this interviewer behavior caused applicants to perform less well during
interviews.
Deficits in minority applicants’ performance during job interviews may also be triggered
by stereotype threat. Stereotype threat research (Steele & Aronson, 1995; Roberson & Kulik,
2007) suggests that minorities and women tend to perform less well when they are aware that
Bias in Hiring 7
their performance may confirm a negative stereotype (e.g. when gender is made salient prior to a
math exam). A plausible explanation for such effects is that stereotype threat enhances anxiety
(O’Brien & Crandall, 2003).
Shared Traits Confer Advantage
A third line of research focuses on ingroup bias, or the tendency for individuals to prefer
members of their own ingroup and derogate those of other groups, as a function of the need to
maintain a positive social identity (Brewer, 1979; Tajfel, 1982). Studies have documented that
this unconscious process may lead to employment decisions not based on applicants’ abilities to
perform on the job, but rather the decision-maker’s categorization of the applicant as an ingroup
member. This ingroup preference has been empirically demonstrated in the laboratory even with
minimal groups, a fabricated group to which an experimental subject is lead to identify as a
member (Tajfel, Billig, Bundy, & Flament, 1971; Hertel & Kerr, 2001). The same processes
are likely to operate even more powerfully when based on more salient traits such as race,
gender, age, or social class. For instance, Finkelstein, Burke, & Raju (1995) concluded that
young persons with authority to hire tended to rate young job applicants more favorably than
older ones.
Employment advantage conferred by shared traits may also be created by social patterns
that extend beyond the workplace. Many social relationships in American society today remain
highly segregated. Persons of different racial, ethnic, and class backgrounds tend to reside in
separate neighborhoods and attend different schools and churches. Persons of different genders
and ages tend to socialize in different groups and social networks (Alba, Logan, & Stutts, 2000).
But the social relationships created through such interactions often are resources in the labor
market (Granovetter, 1995; Adler & Kwon, 2002; Ibarra, 1995). For instance, many job
Bias in Hiring 8
vacancies are never publicly advertised (e.g., on internet job boards or in newspapers), so that the
opportunities become known primarily to the friends, relatives, neighbors, classmates, or
colleagues of current employees (Kuhn & Skuterud, 2000). Similarly, in pre-employment skill
testing and job interviewing, informal coaching and insider information may equip applicants
who have ties to current employees to perform better than their competitors (Hulett, Bendick,
Thomas, & Moccio, 2008). Limited access to information-rich social networks helps to keep
traditionally-excluded groups excluded.
Bias May Evolve Rather than Disappear
Over the four decades since the Civil Rights Movement, anti-discrimination legislation
has certainly reduced many blatant forms of employment discrimination, such as the traditional
division of job vacancy announcements between “Help Wanted-Male” and Help Wanted-
Female.” Over the same period, public opinion polls report consistent downward trends in
individuals’ self-reported prejudice. For example, in public opinion polling in 1965, 59% of
Americans agreed that they would vote for a qualified African American for president, but by
2005, that figure had risen to 93%; parallel questions concerning a qualified woman elicited 55%
agreement in 1965 but 87% in 2005 (Gallup, 2005; see also Smith, 2000).
However, research cautions that these developments may not translate into substantially
reduced bias in employment. Rather, blatant discriminatory behavior toward minority groups
may have simply evolved into what McConahay, Hardee, and Batts (1981) label modern racism
-- a shift in social norms such that explicit expression of prejudiced attitudes is no longer
tolerated and rarely seen, yet prejudiced attitudes remain in more covert forms.
Consistent with the modern racism model, researchers have observed such covert
prejudice reflected in the behavior of actual supervisors and managers. For example, Moss and
Bias in Hiring 9
Tilly (2002; see also Neckerman & Kirschenman, 1991, and Rooth, 2007) describe how, in
rejecting minority job applicants, hiring decision-makers now often cite those applicants’
deficiencies in “soft skills” (such as “inability to communicate” or “lack of work commitment”)
as a socially-acceptable proxy for a job applicant’s race. Other researchers have described some
employers’ initiatives to advance equal employment opportunity as primarily symbolic acts,
creating the appearance of compliance with the law and social norms without the intention to
change employment outcomes substantially (Edelman & Petterson, 1999; Bendick & Egan,
2009).
Hiring is Particularly Vulnerable to Bias
Of course, cognitive and organizational processes such as those just discussed can lead to
biased outcomes in post-hiring employment processes such as performance evaluations,
promotions, raises, transfers, and terminations. However, hiring decisions differ from post-
hiring decisions in several ways tending to make bias more influential in hiring than in those
other processes.
The first of these circumstances is the limited information on which hiring decisions are
based (Altonji & Pierret, 2001). Post-hiring decisions tend to be made by managers and
supervisors who have observed an employee’s performance over months or years. Hiring
decisions, in contrast, typically require assessing job applicants who are virtual strangers.
Applicants’ resumes, application forms, and work samples typically provide only a sketch of
their qualifications and require inference about the relevance of those qualifications to the
positions for which they are being hired. Job interviews tend to be brief — for entry-level
positions, often as short as 10 minutes (Bendick, Rodriguez, & Jayaraman, 2009). Interviews
also tend to be highly subject to “impression management,” or deliberate manipulation of self-
Bias in Hiring 10
presentation to create a favorable impression that does not necessarily translate to successful job
performance (Giacalone & Rosenfeld, 1989; Higgins & Judge, 2004). For these and other
reasons, traditional job interviews have limited power to predict post-hiring job performance (Le,
Oh, Shaffer, & Schmidt, 2007).
Another factor that negatively impacts members of stigmatized groups is the time
pressure under which hiring decisions are often made. Supervisors and managers often must
assess numerous job applications while facing other demands on their time. Stereotypes tend to
exercise particular influence in time-pressured situations (Macrae, Bodenhausen, & Milne,
1999). Concurrently, limited contact between job applicants and hiring decision-makers imbues
every interaction with intense pressure to perform making it particularly susceptible to stereotype
threat.
A different circumstance enhancing bias in hiring compared to post-hiring decisions is
differences in the opportunity to correct errors. Many post-hiring employment decisions are part
of ongoing or repeated decision processes in which an error can be corrected later. For instance,
an employee granted too generous a raise in one year may be brought back into line by a smaller
raise the next year, or an employee passed over for one promotion may be subsequently
considered for other opportunities. In fact, the continued visibility of a minority employee who
is not progressing may itself pressure managers to correct the situation (Zitek & Hebl, 2006;
Stangor, Sechrist, & Jost, 2001). In hiring, in contrast, applicants who are not hired tend
immediately to go elsewhere to find employment, eliminating opportunities for the employer to
whom they had applied to reconsider or adjust.
External pressure to correct bias, such as through enforcement of anti-discrimination
laws, is also less likely for hiring decisions than post-hiring decisions. An employee who feels
Bias in Hiring 11
aggrieved about a post-hiring decision often has both the information necessary to file a
complaint (e.g., Who was promoted instead of me?) and a sufficient stake in the outcome to
invest the resources need to seek redress. In contrast, a job applicant who suspects hiring
discrimination often lacks sufficient information to be sure that discrimination has occurred (e.g.,
When the company said the job vacancy was already filled, was that the truth or an excuse?), and
is more likely to pursue other job opportunities than to dispute the decision. For such reason,
hiring complaints comprise only 6% of formal discrimination complaints filed with the Equal
Employment Opportunity Commission (Bendick, Jackson, & Reinoso, 1994; see also Donohue
& Siegelman, 1991).
III. Empirical Evidence of Employment Bias
The research reviewed in Section II can be translated into a testable hypothesis: bias
continues to adversely affect hiring outcomes for historically excluded groups in the American
labor market today. Does empirical evidence support this hypothesis?
Certainly, research documents substantial progress toward employment equality over
recent decades. Women now constitute 51% of managerial and professional workers in the
American labor force, and race/ethnic minorities more than 22% (U.S. Bureau of Labor
Statistics, 2008), a very different situation from the early 1960s, when their numbers in most
such occupations were so small that essentially every individual was a highly visible exception.
Women’s earnings, which averaged about 60% of men’s until the 1970s, rose to nearly 80% by
the 1990s (Blau & Kahn, 2007), and over the same period, African Americans’ average earnings
rose from 57% of Whites’ to more than 73% (Smith & Welch, 1989).
Yet, such indications of change reveal as much about remaining problems as they do
about progress. Earnings ratios of 73% or 80% are still far below the 100% which would signal
Bias in Hiring 12
simple equality. Worse, in recent years, upward movement of these race and gender wage ratios
has slowed and perhaps plateaued (Rodgers, 2006). Concurrently, the representation of women
and minorities remains very limited in many occupations, especially more prestigious, well paid
ones; these patterns of occupational segregation are often referred to as “glass walls” and “glass
ceilings” (Altonji & Blank, 1999; Reskin & Bielby, 2005). For example, although Hispanics
now constitute 11.6% of police patrol officers, they are only 6.1% of police sergeants and
lieutenants. African Americans are 7.7% of construction laborers but only 2.2% of structural
steel workers. Women still account for only about 1% of skilled construction craft workers, such
as carpenters and electricians (U.S. Bureau of Labor Statistics, 2008). And although women now
account for 31% of medical doctors, they are only 9% of cardiologists and 4% of orthopedic
surgeons. Moreover, female physicians earn an average of 18% less than male physicians with
equal credentials, medical specialties, years in practice, and hours of work per week (HRSA,
2010).
Research in the form illustrated by these findings on physicians’ earnings -- showing
different employment outcomes among demographic groups that are not explained by
differences in education, experience, other measures of qualifications, or work effort -- provides
the most rigorous evidence that substantial bias remains in the American labor market. Such
studies have been conducted on many different occupations, industries, and localities for many
different demographic groups, producing results such as the following:
• After controlling for their education, age, work experience, residence, and criminal
records, dark-skinned black males had a 52% lower chance of being employed than
lighter-skinned black males (Johnson, Bienenstock, & Stoloff, 1995).
Bias in Hiring 13
• Although surgery did not alter their job qualifications, when transgender individuals who
were men transformed into women, their earnings fell 12%, while women who
transformed into men had their earnings rise 8%, (Schilt & Wiwsall, 2008).
• Among American professionals and managers working in international business, an
additional year of international work experience raised men’s salaries an average of
$2,500 but women’s only $1,300; working an extra 8 hours a week increased men’s
salaries an average of $7,300 but women’s only $4,300 (Egan, Bendick, & Miller, 2002).
• In the state of Georgia, 28% of large employers employed statistically-significantly fewer
minority workers than peer companies in the same industry hiring workers with the same
qualifications and local labor market (Blumrosen, Bendick, Miller, & Blumrosen, 1999;
see also Bendick, 2000).
• Leading symphony orchestras which auditioned musicians behind screens so that judges
could not see the job seekers’ sex hired more women players than those which conducted
auditions in the open (Golden & Rouse, 2000).
Such findings are consistent with less rigorous evidence based on perceptions of bias
reported by adversely affected individuals. In one typical public opinion survey with a
nationally-representative sample, 81% of African Americans, 60% of Hispanics, and 53% of
Asian respondents felt that they would have a lower chance of promotion to a managerial
position than an equally-qualified White (Smith, 2000). In another national survey, 31% of
Asians, 26% of African Americans, 18% of Hispanics, and 22% of women reported having been
discriminated against in their workplace during the previous year (Gallup, 2005). In still
another national survey, 53% of workers of all backgrounds recalled hearing jokes at work about
women, 44% for jokes about sexual orientation, 38% about older people, 37% about African
Bias in Hiring 14
Americans, 32% about Hispanics, 29% about Jews, 29% about Muslims, 26% about people with
disabilities, and 25% about Asians (Harris Interactive, 2002). In narrower examples, in a survey
of 194 self-identified transgender adults in San Francisco, 40% of respondents reported having
been discriminated against in applying for jobs, and 24% reported having been sexually harassed
at work (Guardian, 2006); and in a nationwide survey of women firefighters, 85% reported
having been treated adversely at work, in forms ranging from hostile comments to chronic sexual
advances in firehouses and being placed in unnecessary danger on fire scenes (Hulett, Bendick,
Thomas, & Moccio, 2008).
Finally, evidence of perceived discrimination is provided by formal legal complaints of
employment discrimination filed with the U.S. Equal Employment Opportunity Commission,
which in 2008 numbered 93,000 (U.S. Equal Employment Opportunity Commission, 2010). In
parallel, thousands of anti-discrimination lawsuits, both by the public anti-discrimination
enforcement agencies and by private litigators, continue to be filed each year and settled in favor
of plaintiffs, with some settlements including tens of millions of dollars in compensation for
discrimination-based earnings losses (Darity & Mason, 1998; Bendick & Egan, 2009).
IV. Testing Provides Direct Evidence of Hiring Bias
Although the previous section provides considerable evidence of continuing bias, this
evidence often does not separate hiring from post-hiring employment processes; may be based
on perceptions rather than objectively-verified facts; sometimes requires extrapolation to the
behavior of actual employment decision-makers from findings demonstrated in laboratory
settings; or is indirectly inferred from observing differences in employment outcomes that
remain after other job-related explanations have been eliminated. Since about 1990, an
additional research technique has been increasingly used that uniquely provides direct, objective
Bias in Hiring 15
observation of hiring bias in real world settings under controlled, quasi-experimental conditions.
This technique is matched pair testing (see Appendix: A Primer on Matched Pair Tests of
Hiring).
Over the past two decades, several dozen matched pair testing studies have examined
hiring in the Unites States in labor markets from Boston to Los Angeles and occupations from
entry-level retail sales to professional and managerial positions (Bendick, 2007; Pager, 2007;
Bendick, 1999). Several dozen additional studies have been conducted in other industrialized
nations, ranging from the Netherlands to Australia (Riach & Rich, 2002; European Commission,
2006; ILO, 2010). The demographic groups whose experiences were analyzed have included
women; older workers; persons with disabilities; transgender individuals; and race/ethnic
minority groups ranging from African Americans and Hispanics in the United States to
immigrant Turks, North Africans, West Indians, and South Asians in various European nations.
These studies have been essentially unanimous in documenting considerable bias in
hiring. Among studies conducted in a rigorous manner which analyzed the complete hiring
process, the estimated net rate of discrimination typically ranges between 20% and 40%. That
is, the studies found that job applicants from historically excluded groups had a 20% to 40%
chance of a substantially less favorable outcome because of that group membership each time
they applied for a job. Equivalently, these studies estimate that bias infects the hiring decisions
of 20% to 40% of employers.
Such estimates are striking consistent with the non-testing evidence reviewed in Sections
II and III. In falling far below 100%, the estimate rates of discrimination reflect the progress
toward employment equality over the past four decades; in falling far above 0%, they confirm
that substantial bias remains.
Bias in Hiring 16
One obvious direction for future matched pair testing studies is to continue to measure
the extent of discrimination in labor markets, systematically “mapping” the prevalence of
discrimination in different locations (Are workplaces in the suburbs more discriminatory than
central cities?), occupations and industry (Are higher paid jobs more subject to bias than entry-
level jobs?), and demographic groups (Do women or African Americans face worse problems?)
Such studies would be particularly useful if they were repeated annually using a consistent
methodology on a nationally-representative sample to generate a periodic “national report card”
on discrimination tracing trends over time (Fix & Turner, 1999).
A related form of “mapping” has been conducted in recent years through innovative
studies that have applied testing to individuals facing bias on bases other than the “usual” race,
gender, age, and disability characteristics. These studies have involved, for instance, Arab
Americans facing backlash following the terrorist attacks of September 11, 2001 (Discrimination
Research Center, 2004), persons with a criminal record (Pager, 2003), transgender individuals
(Make the Road New York, 2010), and overweight persons (Rooth, 2009).
These testing studies have typically revealed relatively modest variation in the prevalence
of discrimination among demographic groups, locations, and occupations, suggesting that
continued testing to explore how rates of discrimination vary across demographic groups or labor
market circumstances should perhaps no longer command high priority. Furthermore, although
these studies typically capture media attention when released, there is little evidence that they
have powerful or lasting impact on public opinion or public policy. For example, although
testing results were discussed in debates about California’s ballot proposition to abolish
affirmative action (Proposition 209) (Bendick, 1995), the proposition passed.
Bias in Hiring 17
Accordingly, future matched pair testing research is likely to address bias more
effectively if, rather than primarily measuring the prevalence of bias, studies are designed to
advance fundamental understanding of hiring bias and its remedies. Compared to past testing
studies, research in this spirit would be more grounded in behavioral science theory, coordinated
with non-testing research that addresses the same issues, and designed to provide data analyzable
with non-testing methodologies. In short, future research should “test deeper” rather than simply
“test broader.”
V. Testing to Understand Hiring Bias
This section presents four research proposals illustrating that “test deeper” approach.
Study Employer-Job Candidate Interactions
The first proposal concerns using testing to analyze employer-candidate interactions,
particularly job interviews.
Behavioral science research suggests that bias, especially bias that occurs at the implicit
level, is often embodied in small, subtle but crucial differences in words or actions sometimes
referred to as micro-inequities (Valian, 1998). These small, subtle differences often leave
perpetrators of bias unaware of these unconscious processes and their cumulative impact.
Equally, they pose measurement challenges to researchers that seek to document exactly how
bias operates.
To date, research on the details of employer-employee interactions has often examined
written materials such as letters of recommendations and performance evaluations (e.g., Trix &
Psenka, 2003; Bison-Rapp, 1999). These studies have documented striking systematic
differences in statements about equally-qualified individuals of different demographic
backgrounds. For example, one comparison of performance evaluations for men and women
Bias in Hiring 18
professional employees who all received high performance ratings in the same financial services
firm found that women were praised primarily for activities within their own work group and
men for activities involving external relationships; criticisms of men tended to be accompanied
by mitigating explanations but those for women were not; and men were commonly
recommended for advancement while women were described as valuable in their current position
(Townsend, 1997). Other studies have revealed Linguistic Intergroup Bias, language describing
ingroup and outgroup behavior that itself tends to perpetuate stereotypes — for instance, by
describing outgroup positive behavior and ingroup negative behavior in concrete terms that
suggest situational and therefore not enduring traits, while reporting outgroup negative behavior
and positive ingroup behavior in abstract terms that suggest persistent qualities (Maass, Salvi,
Arcuri, & Semin, 1989; see also Semin & Fiedler, 1992, and Wenneker, Wigboldus, & Spears,
2005).
Only a few studies in this tradition have examined face-to-face interactions rather than
written material (Binning, Goldstein, Garica, & Scatteregia, 1988, and Shelley & Shelley, 2009.
The data required for such research are samples of the exact words used to describe
demographically different individuals who are equivalent in actual qualifications or actual
performance. Filling a gap in existing research, matched pair testing could provide such data for
in-person interactions, such as job interviews. Through modern miniature recording technology,
voice and video recordings of conversations can now be collected while leaving employers
unaware of being tested or recorded. Transcripts of these encounters could then be analyzed
using the psycholinguistic techniques and concepts illustrated in the previous paragraph.
Some testing studies have examined employer and employee behavior in job interviews.
For example, Bendick, Jackson, & Reinoso (1994) examined interviews of African American
Bias in Hiring 19
and White applicants for entry-level positions that do not require college degrees. They
observed that, although the two groups received very different numbers of job offers, applicants
experienced no substantial differences in treatment while being interviewed. Here, applicant
treatment was measured using observable behavior such as interview length, proportion of the
interview devoted to job-relevant topics rather than social chit chat, and rank of the interviewer.
Another study examined interviews of Whites and Persons of Color applying for waitstaff
positions in upscale restaurants (Bendick, Rodriguez, & Jayaraman, 2009). It documented that
interviewers tended to accept past restaurant experience without probing when it was claimed by
White applicants but posed aggressive, skeptical questions to non-White applicants claiming
equivalent experience. Such studies are far from definitive, but they suggest the richness of
insights that more sophisticated analyses of interview transcripts could generate.
Coordinate Testing Studies of Multiple Outgroups
Our second research proposal is to study patterns of bias against multiple out-groups by
the same employer.
Many analyses of discrimination against historically excluded groups focus on issues
specific to each group, for example, the impact of slavery and segregation on African Americans,
the relationship between family responsibilities and women’s careers, or the effect of English
language requirements on recent immigrants (Bell, 2007). But other threads in behavioral
science research suggest a fundamentally different approach. More than 50 years ago,
psychologists began to consider prejudice as a consistent characteristic of an individual, so that
the same person harboring bias against women would also tend to be prejudiced against
minorities (Bierly, 1985; Peterson, Doty, & Winter, 1993). In parallel, some sociologists have
argued that the essence of workplace discrimination is not adverse attitudes toward specific
Bias in Hiring 20
outgroups but rather social closure to preserve the power, status, and privilege of a dominant
ingroup against all alternative claimants (Tomascovic-Devey, 1993; Murray, 1988; Moore, 1990;
Freshman, 1990). In this model, an employer that discriminates against women will tend to do
the same against race-ethnic minorities because both actions preserve White male dominance.
This debate is important not only to social theory but also to practical anti-discrimination
efforts. For instance, the first perspective implies that it is important to contradict the stereotypes
associated with individual outgroups (e.g., older workers are obsolete; women are not good
leaders). In contrast, the second perspective considers stereotypes primarily ex post rationales
for negative treatment of outgroups rather than independent causes and suggests controlling the
generic process of stereotyping rather than addressing the content of specific stereotypes
(Bendick, Egan, Lofhjelm, 2001; Egan & Bendick, 2008).
To date, matched pair testing has typically examined one outgroup per study. However,
it could be organized to, for example, field teams of males and females, African Americans and
Whites, and older and younger workers to apply for jobs with the same sample of employers.
Positive correlations in bias against multiple outgroups would support the social closure model,
while zero or negative correlations would support the alternative and imply that stereotype
content does matter (Fiske, Cuddy, Glick, & Xu, 2002).
Behavioral science research also suggests other ways in which an employer’s behavior
toward an outgroup may differ depending on whether that group is considered in isolation or in a
broader context that includes other outgroups. For instance, moral credentialing (Monin &
Miller, 2001) suggests that an employer who has hired an applicant from a stigmatized group
(e.g., women) and therefore feels that he or she has adequately demonstrated egalitarian values,
may be less likely to evaluate favorably a subsequent female or minority candidate. Again,
Bias in Hiring 21
testing studies could be designed to test this hypothesis, in this case, by appropriately sequencing
multiple tests of the same employer and then analyzing the relationship between each test
outcome and that of preceding tests.
Evaluate Strategies for Minimizing Bias
Practical advice routinely offered to outgroup job seekers includes a variety of
suggestions concerning how to minimize bias they may encounter. For example, with respect to
stereotypes, some advisors counsel explicitly refuting stereotypes about groups to which the job
seeker belongs (Kawakami, Dovidio, Moll, Hermsen, & Russin, 2000); others suggest refuting
using individuating information to counter the applicability of the stereotype to the job seeker
(Beckett & Park, 2005; Glick, Zion, & Nelson, 1988); and still others advise emphasizing
alternative, positive stereotypes associated with those groups (Gawronski, Deutsch, Mbirkou,
Seibt, & Strack, 2008). Some advisors suggest implementing these strategies proactively — in
anticipation that stereotypes will inevitably bias hiring decision-makers — while others suggest
doing so only if some indication triggers a concern.
Matched pair testing could be designed to evaluate such alternative strategies in actual
hiring situations. Such research would provide advice to job seekers based on empirical
evidence in place of the current basis, which is largely intuitive.
To date, only rare testing studies have examined alternative strategies for combating bias.
One study of age discrimination using mailed resumes randomly assigned different cover letters
to accompany the older workers’ resume. In one letter, the older applicant described
herself/himself as career-committed, energetic, and technologically up-to-date — positive
attributes stereotypically associated with younger workers; an alternative letter described the
older applicant as experienced, mature and stable — positive attributes stereotypically associated
Bias in Hiring 22
with older workers; and a third letter contained neither statement. The first cover letter generated
a substantially higher rate of favorable employer responses to the older applicant than the other
two (Bendick, Jackson, and Romero, 1996).
Employers also have implemented a variety of initiatives designed to minimize bias, and
again the empirical evidence supporting these practices is often intuitive rather than empirical.
In particular, most major employers in the United States today invest in “workforce diversity
training” for their employees, despite research which questions the effectiveness of many of
these efforts (Kalev, Dobbin, & Kelly, 2006; Bendick, Egan & Lofhjelm, 2001).
Matched pair testing studies could be structured to provide more credible assessments of
these efforts. For example, matched pair tests could be conducted in a workplace prior to a
diversity training program and then again after the training. Or a firm such as a large retail chain
could implement training in one subset of its establishments and not in a matched subset and then
conduct hiring tests in both the experimental and control locations. Here, testing could measure
not only the effect of the training on overall rates of discrimination but also on specific hiring
practices and behaviors that employees were trained to adopt or avoid.
Testing could also improve guidance for employers on other complex issues in bias
control. For example, in sincere efforts to avoid violating social norms, some hiring interviewers
become hyper-vigilant and consciously avoid any reference to race. However, Apfelbaum,
Sommers, & Norton (2008) demonstrate that such strategic colorblindness may backfire,
leading to nonverbal unfriendliness interpreted by Black observers as negative and prejudiced.
Similarly, Chartrand & Bargh (1999) discuss a chameleon effect in which interviewees
unconsciously mirror the behavior of interviewers, generating awkward behavior not
representative of their potential post-hiring work performance. Matched pair testing could be
Bias in Hiring 23
used to study the prevalence in actual job interviews of these hypothesized effects and whether
they vary among different demographic combinations of interviewers and interviewees.
Findings could then be translated into practical guidance to interviewers in interviewing and
assessing job candidates.
Study the Effects of Hiring Rejection
Behavioral science research has long documented the negative psychological and social
consequences of unemployment, including increased physical and mental illness, loss of self
esteem, family stress, and discouragement in seeking work or investing in training (e.g., Linn,
Sandifer, & Stein, 1985; Clark, Georgellis, & Sanfey, 1999). A few studies have focused on
these effects when bias was a factor (Goldsmith, Sedo, Darity, & Hamilton, 2004). These studies
employ a range of research techniques, including epidemiological studies of morbidity and
mortality, worker surveys utilizing scales of self-efficacy and psychological well being,
laboratory studies tracking physiological responses to disappointment, or diaries tracking
attitudes and perceptions over time (Bolger, Davis, & Rafieli, 2003).
Matched pair testing offers a number of advantages as an additional methodology for
measuring these effects. It provides unusually detailed information about the experimental
stimulus whose effects are being studied -- exactly what jobs were sought, what testers
experienced, and the bias encountered. It allows examining hiring separately from other aspects
of workers’ unemployment experiences, such as job separation. And by forming appropriate
tester teams, it can isolate differences in the responses of different types of workers -- for
instance, youth just entering the workforce compared to workers with more experience.
Using such measures, researchers might study for example, the cumulative effects of
multiple job disappointments; whether the effects of failing to be hired are different when the
Bias in Hiring 24
decision was biased or unbiased; and whether the same disappointing hiring experience creates
different effects on different demographic groups; for example, it might be hypothesized that,
because greater expectations of eventual employment create a greater sense of self-efficacy
(Bandura, 1997), white males might experience smaller adverse effects on their self-esteem from
failure to be hired than comparable women or minorities. Testing could also be used to examine
the efficacy of strategies to minimize the adverse effects of job disappointment. For example,
self affirmation theory (Steele, 1988) suggests that individuals frustrated in one endeavor (e.g.,
job seeking) might preserve their self-image by achieving success in a different domain (e.g.,
volunteer work).
In most testing studies, testers are research assistants who did not actually want the jobs
to which they apply, and therefore their reactions to their testing experiences may differ
substantially from those of actual job seekers. Accordingly, for research such as is suggested
here, real job seekers should be hired as testers and dispatched to apply for jobs they would be
allowed to take if offered.
However, testing using conventional research assistants can be used to study the effect on
individuals’ attitudes and empathy of being personally exposed to the reality of discrimination
which they otherwise might not encounter. Research by Batson, et al. (2003; see also Gaertner
& Dovisio, 2005, and www.janeelliott.com) concludes that putting an individual “in the shoes of
others” expands the range of persons about whose well-being they feel concerned; Fogelman
(1994) has documented the role of striking personal incidents in persuading bystanders to act on
behalf of victims of persecution. In the same spirit, it can be hypothesized that ingroup testers
(e.g. Whites or males) who observe their paired testing partners encounter bias may be made
Bias in Hiring 25
more empathetic by the experience. This hypothesis could be studied by measuring the attitudes
and behavior of ingroup testers before and after their testing experience.
VI. Using Testing to Change Employer Behavior
A perennial challenge in the behavioral sciences is to mobilize research findings to
influence real world practices. How can employers be convinced and assisted to incorporate the
lessons from research discussed throughout this paper into their firm’s employment practices and
the behavior of their employees? Matched pair testing offers new approaches for doing so, of
which this section sketches three proposals.
Testing on Behalf of Employers
Employers commonly monitor the behavior of their own staff using unobtrusive data
gathering techniques. For example, retailers often employ “mystery shoppers” to pose as
customers and record detailed data on their interactions with sales staff (www.mysteryshop.org).
They then use these data to improve the quality of service provided to their customers, such as
through staff retraining.
Most employment testing studies have been conducted without employers being aware of
being tested or presented with findings about their individual firms. Our first proposal suggests
testing by or for employers as a means of self-audit.
One example of which we are personally aware took place in 2009 at the teaching
hospital of a major medical school. This employer desired to offer jobs to residents of its
surrounding community as well as employ staff who could relate to its demographically diverse
patients, and therefore wanted to ensure its hiring practices were not biased. To support this
goal, the hospital required all its supervisors to be trained to conduct job interviews in a job-
related, structured manner (Le, Oh, Shaffer, & Schmidt, 2007). However, the hospital had never
Bias in Hiring 26
measured the extent to which these supervisors subsequently followed those procedures when
interviewing job candidates or whether their hiring decisions resulted in a more diverse
workforce. Accordingly, the hospital’s vice president of human resources secretly arranged with
a local non-profit organization to send matched pairs of White and minority applicants to apply
for positions at the hospital and two comparable hospitals in the same city. Although this project
was not completed due to a hiring freeze, it exemplifies a potentially fruitful, cooperative
relationship between testing researchers and an employer.
Businesses currently invest huge numbers of staff hours as well as many millions of
dollars each year in initiatives to address bias and increase staff diversity. However, 62% of
them collect no information on the impact of these efforts (Esen, 2005). This lack of data limits
their ability to improve these initiatives, as well as to hold managers accountable for following
prescribed practices (Bendick, 2008). As the hospital example illustrates, testing conducted by
or with employers can provide direct data on the extent to which initiatives effectively modify
employment practices “on the shop floor” and the extent to which those modifications make a
difference in hiring outcomes.
Testing for Training
When testing results are presented to audiences of non-researchers — for example, in
legislative hearings or articles in the mass media — their attention-grabbing power is
immediately evident. This power appears to derive from testing’s combination of rigorously
controlled statistical evidence and vivid anecdotes that put a human face on those statistics
(Cialdini, 2000).
This persuasive power has yet to be extensively taken advantage of in practical employee
training. Among larger employers, more than 90% provide some form of anti-discrimination or
Bias in Hiring 27
pro-diversity training for employees, often enrolling everyone from senior executives and mid-
level supervisors to non -managerial employees (Esen, 2005; Kalev, Dobbin, & Kelly, 2006).
The most effective forms of this training use vivid examples from real workplaces to persuade
trainees of the continued presence of discrimination and to communicate desired changes in
employee behavior (Adamson, 2000; Bendick, Egan, & Lofhjelm, 2001).
Testing results are ideally structured to provide specific, vivid examples for use in
training, especially if tests are documented through audio or video recording as was suggested in
Section V. As Section V also suggested, post-training testing of hiring outcomes could be used
to assess whether training that includes testing-based illustrations is more effective than other
approaches.
Testing for Litigation
Employers in the United States who violate the equal employment provisions of federal
or state laws can be sued either by government agencies, such as the federal Equal Employment
Opportunity Commission, or in private litigation brought by the victims of discrimination. If the
plaintiffs prevail, the employer may be liable for substantial damage payments as well as
mandatory, court-supervised changes in its employment practices. Since the late 1960s, such
litigation -- or employers’ desire to avoid it -- has been a major motivator for reductions in
employment bias (Blumrosen, 1999).
American law gives matched pair testing potential roles in enforcement. If an employer
is sued based on evidence other than testing, documentation of an employer’s discriminatory
behavior through testing can be used to corroborate that evidence. In addition, individual testers
and non-profit organizations employing testers have “standing” to become plaintiffs in litigation
based on testing evidence alone (Yelnosky, 2010; Boggs, Sellers, & Bendick, 1993).
Bias in Hiring 28
Despite this potential, only a handful of testing-based enforcement actions have been
brought. One pioneering lawsuit, Fair Employment Council et al. v. BMC Marketing, was filed
in the District of Columbia in 1990. In it, the plaintiffs were two African American university
students and the non-profit organization that had hired them as testers. The defendant was a
local office of Snelling & Snelling, one of the nation’s largest job placement agencies. The
agency had interviewed, coached, and found entry-level office jobs for two testers while it failed
to do so for the African Americans who were their testing partners. This litigation was
subsequently settled with payment of damages to the non-profit organization and commitments
by the defendant to retrain its staff and implement other actions to eliminate discriminatory
practices (Boggs, Sellers, & Bendick, 1993).
That lawsuit happened to have involved plaintiffs from a stigmatized group -- African
Americans -- for whom testing is only one of several ways to generate evidence of
discrimination. For other groups, however, testing-based evidence may be the only feasible
basis for litigation. For example, in anti-discrimination, statistics from sources such as the
Census are often used to demonstrate that an employer has fewer female or race/ethnic minority
employees than would be expected based on their availability in the local labor market.
However, such Census data is not collected for many historically excluded groups-- for instance,
persons with psychiatric disabilities (Tal, Moran, Rooth, & Bendick, 2009) or gay, lesbian,
transgender and bisexual individuals (Make the Road, 2010). In that circumstance, direct
evidence of employer discrimination obtained by matched pair testing may be the only feasible
way to document hiring discrimination in a manner that will hold up in court.
Litigation is only one adversarial approach in which testing could be used to pressure
employers. To date, publicly-released results from testing studies have typically not named
Bias in Hiring 29
employers against whom incriminating findings were obtained. However, the results of situation
tests on specific employers could be released to the general public, news media, insurers,
investors, unions, customers and others, providing information by which these stakeholders can
pressure discriminating employers to change their behavior (Egan, Mauleon, Wolff, & Bendick,
2007).
VII. Summary
The seven types of matched pair testing studies discussed here form an ambitious
research agenda. However, these examples by no means exhaust the potential of matched pair
testing to generate theoretical and practical advances against hiring bias. It is time for
researchers in the social, behavioral, and managerial sciences to mobilize the power of this
technique creatively to advance this important societal goal.
Bias in Hiring 30
References
Adamson, J. (2000). The Denny’s story: How a company in crisis resurrected its good name and
reputation. New York: John Wiley.
Adler, P., & Kwon, S. (2002). Social capital: Prospects for a new concept. Academy of
Management Review, 27, 17-40.
Alba, R., Logan, J., & Stutts, B. (2000). How segregated are middle-class African Americans?
Social Problems, 47, 543-558.
Altonji, J., & Blank, R. (1999). Race and gender in the labor market. In O. Aschenfelter &
Card, D. (Eds.) Handbook of Labor Economics, Volume 1 (pp. 3243-3259). Amsterdam:
Elsevier.
Altonji, J., & Pierret, C. (2001). Employer learning and statistical discrimination. Quarterly
Journal of Economics, 116, 313-350.
Apfelbaum, E., Sommers, S., & Norton, M. (2008). Seeing race and seeming racist? Evaluating
colorblindness in social interactions. Journal of Personality and Social Psychology, 95,
918-932.
Arrow, K. (1998). What has economics to say about racial discrimination? Journal of Economic
Perspectives, 12, 91-100.
Bandura, A. (1997). Self-efficacy, The exercise of control. New York: Freeman.
Batson, C., Lishner, D., Carpenter, A., Dulin, L., Harjusola-Webb, S., Stocks, E., Gale, S.,
Hassan, O., & Sampat, B. (2003). “As you would have them do unto you”: Does
Bias in Hiring 31
imagining yourself in the other’s place stimulate moral action? Personality and Social
Psychology Bulletin, 29, 1190-1201.
Beckett, N., & Park, B. (1995). Use of category versus individuating information: Making base
rates salient. Personality and Social Psychology Bulletin, 21, 21-31.
Bell, M. (2007). Diversity in Organizations. Mason, OH: Thomson South-Western.
Bendick, Jr., M. (1999). Adding testing to the nation’s portfolio of information on employment
discrimination. In M. Fix and M. Turner (Eds.) A national report card on discrimination:
The role of testing (pp. 47-68). Washington: The Urban Institute.
Bendick, Jr., M. (2008) Measure Inclusion, Not Diversity. Presentation, Society for Human
Resource Management National Diversity Conference.
Bendick, Jr., M. (1995). Research evidence on racial/ethnic discrimination and affirmative action
in employment. In Discrimination and affirmative action: Are there any facts out there?
(pp. A58-A78). Sacramento: Assembly Judiciary Committee, California State
Legislature.
Bendick, Jr., M. (2007). Situation testing for employment discrimination in the United States of
America. Horizon Strategiques, 5, 17-39.
Bendick, Jr., M. (2000). Using EEO-1 data to analyze allegations of employment discrimination.
Presentation, American Bar Association National Conference.
Bias in Hiring 32
Bendick, Jr., M., and Egan, M. (2009). Research Perspectives on Race and Employment in
Advertising. Washington, DC: Bendick and Egan Economic Consultants, Inc. for the
Madison Avenue Project.
Bendick, Jr., M., Egan, M., & Lofhjelm, S. (2001). Diversity training: From anti-discrimination
compliance to organization development. Human Resource Planning, 24, 10-25.
Bendick, Marc, Jr., Brown, L., & Wall, K. (1999). No foot in the door: An experimental study of
employment discrimination against older workers. Journal of Aging and Social Policy,
10, 5-23.
Bendick, Jr., M., Jackson, C., & Reinoso, V. (1994). Measuring employment discrimination
through controlled experiments. Review of Black Political Economy, 23, 25-48.
Bendick, Jr., Marc, Jr., Jackson, C., Reinoso, V., & Hodges, L. (1991). Discrimination against
Latino job applicants: A controlled experiment. Human Resource Management, 30, 469-
484.
Bendick, Jr., M., Jackson, C., & Romero, J. (1996). Employment discrimination against older
workers: An experimental study of hiring practices. Journal of Aging and Social Policy,
8, 25-46.
Bendick, Jr., M., Rodriguez, R., & Jayaraman, S. (2009). Race-ethnic employment discrimination
in upscale restaurants: Evidence from matched pair testing. Washington, DC: Bendick
and Egan Economic Consultants, Inc, for the Restaurant Opportunity Center of New
York.
Bias in Hiring 33
Bertrand, M., & Mullainathan, S. (2004). Are Emily and Brendan more employable than Lakisha
and Jamal? A field experiment on labor market discrimination. American Economic
Review, 113, 991-1001.
Bielby, W. (2000). Minimizing workplace gender and racial bias. Contemporary Sociology 29,
120-129.
Bierly, M. (1985). Prejudice toward contemporary outgroups as a generalized attitude. Journal of
Applied Social Psychology, 15, 189-199.
Binning, J., Goldstein, M., Garcia, M., & Scattereregia, J. (1988). The effects of preinterview
impressions on interviewer questioning strategies in same- and opposite-sex employment
interviews. Journal of Applied Psychology, 73, 30-37.
Bison-Rapp, S. (1999). Bulletproofing the workplace: Symbol and substance in employment
discrimination law practice. Florida State University Law Review, 26, 959-1047.
Blau, F., & Kahn, L. (2007). The gender pay gap: Have women gone as far as they can?
Academy of Management Perspectives, 21, 7 – 23.
Blumrosen, A. (1993). Modern Law: The law transmission system and equal employment
opportunity. Madison: University of Wisconsin Press.
Blumrosen, A., Bendick, Jr., M., Miller, J., & Blumrosen, R. (1999). Employment discrimination
against women and minorities in Georgia. Rutgers, NJ: Rutgers University Law School.
Bias in Hiring 34
Boggs, R., Sellers, J., and Bendick, Jr., M. (1993). Use of testing in civil rights enforcement. In
Fix, M. and Struyk, R. (Eds.), Clear and convincing evidence: Measurement of
discrimination in America (pp. 345-376). Washington: Urban Institute Press.
Bolger, N., Davis, A., & Rafieli, E. (2003). Diary methods, Capturing life as it is lived. Annual
Review of Psychology, 54, 579-616.
Bovenkerk, F. (1992). Testing discrimination in natural experiments: A manual for international
comparative research on discrimination on the grounds of “race” and ethnic origin.
Geneva: International Labour Office.
Brewer, M. (1979). In-group bias in the minimal intergroup situation: A cognitive motivational
analysis. Psychological Bulletin, 86, 307-324.
Brief, A. (Ed.) (2008). Diversity at work. New York: Cambridge University Press.
Chartrand, T., & Bargh, J. (1999). The chameleon effect: The perception-behavior link and
social interaction. Journal of Personality and Social Psychology, 76, 893-910.
Cialdini, R. (2000). Influence: Science and Practice. Boston: Allyn & Bacon.
Clark, A., Georgellis, Y., & Sanfey, P. (1999). Scarring: The psychological effect of past
unemployment. Economica, 68, 221-241.
Darity, W., & Mason, P. (1998). Evidence on discrimination in employment: Codes of color,
codes of gender. Journal of Economic Perspectives, 12, 63-90.
Darley, J., & Gross, P. (1983). A hypothesis-conforming bias in labelling effects. Journal of
Personality and Social Psychology, 44, 20-33.
Bias in Hiring 35
Dipboye, R., & Colella, A. (2005). Discrimination at work: The Psychological and
Organizational Bases. Mahwah, NJ: Lawrence Erlbaum.
Discrimination Research Center (2004). Names make a difference: The screening of resumes by
temporary employment agencies in California. Berkeley, CA: Discrimination Research
Center of The Impact Fund.
Donohue, J., & Siegelman, P. (1991). The changing nature of employment discrimination
litigation, Stanford Law Review, 43, 983-1033.
Dovidio, J., Kawakami, K., & Gaertner, S. (2002). Implicit and explicit prejudice and interracial
interaction. Journal of Personality and Social Psychology, 82, 62-68.
Eagley, A., & Karau, S. (2002). Role congruity theory of prejudice toward female leaders.
Psychological Review, 109, 573-598.
Eberhardt, J., Dasgupta, N., & Banaszynski, T. (2003). Believing is seeing: The effects of racial
labels and implicit beliefs on face perception. Personality and Social Psychology
Bulletin, 29, 360-370.
Edelman, L., & Petterson, S. (1999). Symbols and substance in organizational response to civil
rights law. Research in Social Stratification and Mobility, 17, 107-136.
Egan, M., & Bendick, Jr., M. (2008). Combining multicultural management and diversity into
one course on cultural competence. Academy of Management Learning and Education,
7, 387-393.
Bias in Hiring 36
Egan, M., Bendick, Jr., M., & Miller, J. (2002). US firms’ evaluation of employee credentials in
international business. International Journal of Human Resource Management, 13, 78-
88.
Egan, M., Mauleon, F., Wolff, D., & Bendick, Jr. M. (2009). France’s Mandatory “Triple
Bottom Line” Reporting: Promoting Sustainable Development through Informational
Regulation. International Journal of Environmental, Cultural, Economic, and Social
Sustainability, 7, 27-47.
Erber, R., & Fiske, S. (1984). Outcome dependency and attention to inconsistent information.
Journal of Personality and Social Psychology, 47, 709-726.
Esen, E. (2005). 2005 workplace diversity practices: Survey report. Alexandria, VA: Society for
Human Resource Management.
European Commision (2006). Equality and non-discrimination annual report. Brussels:
Commission of the European Union.
Finkelstein, L, Burke, M., & Raju, N. (1995). Age discrimination in simulated employment
contexts: An integrative analysis. Journal of Applied Psychology, 80, 625-663.
Fiske, S., Bersoff, D., Borgida, E., Deaux, K., & Heilman, M. (1991). Social science research on
trial: The use of sex stereotyping research in Price Waterhouse v. Hopkins. American
Psychologist, 46, 1049-1060.
Fiske, S., Cuddy, A., Glick, P., Xu, J. (2002). A model of (often mixed) stereotype content:
Competence and warmth respectively follow from perceived status and competition.
Journal of Personality and Social Psychology, 82, 878-902.
Bias in Hiring 37
Fix, M. and Turner, M. (Eds.) (1999). A national report card on discrimination: The role of
testing. Washington: Urban Institute Press.
Fogelman, E. (1994). Conscience and courage, Rescuers of Jews during the Holocaust. New
York: Anchor Doubleday.
Freshman, C. (1990). Beyond atomized discrimination: Use of acts of discrimination against
“other” minorities to prove discriminatory motivation under federal employment law.
Stanford Law Review, 43, 241- 273.
Gaertner, S., & Dovidio, J. (2005). Understanding and addressing contemporary racism: From
aversive racism to the common ingroup identity model. Journal of Social Issues, 61, 615-
639.
Gallup (2005). Employee discrimination in the workplace. Washington, DC: The Gallup
Organization.
Gawronski, B., Deutsch, R., Mbirkou, S., Siebt, B., & Strack F. (2008). When 'Just say no' is not
enough: Affirmation vs. negation training and the reduction of automatic stereotype
activation. Journal of Experimental Social Psychology, 44, 370-377.
Giacalone, R., & Rosenfeld, P. Impression management in the organization. Hillsdale, NJ:
Lawrence Erlbaum.
Glick, P., Zion, C., & Nelson, C. (1988). What mediates sex discrimination in hiring decisions?
Journal of Personality and Social Psychology, 55, 178-186.
Bias in Hiring 38
Golden, C., & Rouse, C. (2000). Orchestrating impartiality: The impact of “blind” auditions on
female musicians. American Economic Review, 90, 715-741.
Goldsmith, A., Sedo, S., Darity, W., & Hamilton, D. (2004). The labor supply consequences of
perceptions of employer discrimination during search and on-the-job: Integrating
neoclassical theory and cognitive dissonance. Journal of Economic Psychology, 25, 15-
39.
Granovetter, M. (1995). Getting a job, A study of contacts and career. Chicago: University of
Chicago Press.
Greenwald, A., McGeer, D., and Schwartz, J. (1998). Measuring individual differences in
implicit cognition, The Implicit Association Test. Journal of Personality and Social
Psychology, 74, 1464 –1480.
Guardian (2006). Good jobs NOW! A snapshot of the economic health of San Francisco’s
transgender communities. San Francisco: The San Francisco Bay Guardian.
Harris Interactive (2002). Workplace discrimination against, and jokes about, African
Americans, gays, Jews, Muslims, and other. The Harris Poll 61, at
http://www.harrisinteractive.com, downloaded February 2, 2010.
Heckman, J. (1998). Detecting discrimination. Journal of Economic Perspectives, 12, 101-116.
Hertel, G. & Kerr, N.L. (2001). Priming in-group favoritism: The impact of normative scripts in
the minimal group paradigm. Journal of Experimental Social Psychology, 37, 316-324.
Bias in Hiring 39
Higgins, C., & Judge, T. (2004). The effect of applicant influence tactics on recruiter perceptions
of fit and hiring recommendations: A field study. Journal of Applied Psychology, 89,
622-632.
Hilton, J., & Von Hippel, W. (1996). Stereotypes. Annual Review of Psychology, 47, p. 237-271.
HRSA (2010). The physician workforce: Projections and research into current issues affecting
supply and demand. Washington, DC: U.S. Department of Health and Human Services,
Health Resources and Services Administration.
Hulett, D., Bendick, Jr., M., Thomas, S., & Moccio F. (2008). Enhancing women’s inclusion in
firefighting in the USA. International Journal of Diversity in Organizations,
Communities, and Nations, 8, 1-24.
Ibarra, H. (1995). Race, opportunity, and diversity of social circles in managerial network.
Academy of Management Journal, 38, 673-703.
ILO (2010). Discrimination: The evidence, at
www.ilo.org/public/english/protection/migrant/equality/discrimination/evidence.htm,
downloaded February 17, 2010.
Johnson, Jr., J., Bienenstock, E., & Stoloff, J. (1995). An empirical test of the cultural capital
hypothesis. Review of Black Political Economy, 23, 7-27.
Kalev, A., Dobbin, F., and Kelly, E. (2006). Best practices or best guesses? Diversity
management and the remediation of inequality. American Sociological Review, 71, 589-
617.
Bias in Hiring 40
Kawakami, K., Dovidio, J.F., Moll, J., Hermsen, S., & Russin, A. (2000). Just say no (to
stereotyping): Effects of training in the negation of stereotypic associations on stereotype
activation. Journal of Personality and Social Psychology. 78, 871-888.
Kenney, G., and Wissoker, D. (1994). An analysis of the correlates of discrimination facing
young Hispanic job-seekers. American Economic Review, 84, 674-83.
Kochan, T., Bezrukova, K., Ely, R., Jackson, S., Joshi, A., Jehn, K., Leonard, J., Levine, D., &
Thomas, D. (2003). The effects of diversity on business performance: Report of the
Diversity Research Network. Human Resource Management, 42, 3-21.
Koomen, W. & Dijker, A.J. (1998). Ingroup and outgroup stereotypes and selective processing.
European Journal of Social Psychology, 27, 589-601.
Kuhn, P., & Skuterud, M. (2000). Job search methods: Internet versus traditional. Monthly Labor
Review, 123, 3-11.
Le, H., Oh, I., Shaffer, J., & Schmidt, F. (2007). Implications of methodological advances for
practices of personnel selection: How practitioners benefit from meta-analysis. Academy
of Management Perspectives, 21, 6-15.
Linn, M., Sandifer, R., Stein, S. (1985). Effects of unemployment on mental and physical health.
American Journal of Public Health, 75, 507-512.
Lodder, L. (1994). Employment testing for civil rights enforcement, An operations manual.
Chicago: Legal Assistance Foundation of Chicago.
Bias in Hiring 41
Maas, A., Salvi, D., Acuri, L., Semin, G. (1989). Language use in intergroup contexts: The
linguistic intergroup bias. Journal of Personality and Social Psychology, 57, 981-993.
Macrae, C., Bodenshausen, G., & Milne, A. (1999) Saying no to unwanted thoughts, Self-focus
and the regulation of mental life. Journal of Personality and Social Psychology, 74, 578-
590.
Macrae, C., Bodenshausen, G., Milne, A., & Jetten, J. (1994) Out of mind but back in sight,
Stereotypes on the rebound. Journal of Personality and Social Psychology, 67, 808-817.
McConahay, J., Hardee, B, & Batts, V. (1981). Has racism declined in America? It depends on
who is asking and what is asked. Journal of Conflict Resolution, 25, 563-579.
Make the Road (2010). Transgender need not apply: Gender identity job discrimination in New
York City’s retail sector. New York: Make the Road New York.
Monin, B., & Miller, D. (2001). Moral credentials and the expression of prejudice. Journal of
Personality and Social Psychology, 81, 33-43.
Moore, R. (1990). The formation of a persecuting society, Power and deviance in Western
Europe, 950-1250. Oxford: Blackwell.
Moss, P., & Tilly, C. (2002). Stories employers tell, Race, skill, and hiring in America. New
York: Russell Sage Foundation.
Murray, R. (1988). Social closure, The theory of monopolization and exclusion. Oxford: Oxford
University Press.
Bias in Hiring 42
Neckerman, J., & Kirschenman, K. (1991). “We’d love to hire them, but…” the meaning of race
for employers. In C. Jencks and P. Person (Eds.) The Urban Underclass (pp. 203-34).
Washington, DC: Brookings Institution.
Nelson, T., Acker, M., & Manis, M. (1996). Irrepressible stereotypes. Journal of Experimental
Social Psychology, 32, 13-38.
Nosek, B., Banaji, M., Greenwald, A.(2002). Harvesting implicit group attitudes and beliefs
from a demonstration website. Group Dynamics, 6, 101-115.
Nosek, B. Greenwald, A Banaji, M.(2005). Understanding and using the implicit association test:
II. Method variables and construct validity. Personality & Social Psychology Bulletin, 31,
166-180.
Nunes, A., and Seligman, B. (2000). A study of the treatment of female and male applicants by
San Francisco Bay Area auto service shops. Berkeley, CA: Discrimination Research
Center of The Impact Fund.
Nunes, A., & Seligman, B. (1999). Treatment of Caucasian and African-American applicants by
San Francisco Bay Area employment agencies: Results of a study utilizing “testers.”
Berkeley, CA: Discrimination Research Center of the Impact Fund.
O’Brien, L.,& Crandall, C. (2003). Stereotype threat and arousal: Effects on women’s math
performance. Personality and Social Psychology Bulletin, 29, 782-789.
Bias in Hiring 43
Pager, D. (2007). The use of field experiments for studies of employment discrimination:
Contributions, critiques, and directions for the future. Annals of the American Academy of
Political and Social Science, 609, 104-133.
Pager, D. (2003). The mark of a criminal record. American Journal of Sociology, 108, 937-975.
Peterson, B., Doty, R., & Winter, D. (1993). Authoritarianism and attitudes toward
contemporary social issues. Personality and Social Psychology Bulletin, 19, 174-184.
Reskin, B., & Bielby, D. (2005). A sociological perspective on gender and career outcomes.
Journal of Economic Perspectives, 19, 71-86.
Riach, P., and Rich, J. (2002). Field experiments in discrimination in the market place, The
Economic Journal, 112, F480-F518.
Richeson, J., Trawalter, S., & Shelton, J. (2005). African Americans’ implicit racial attitudes and
the depletion of executive function after interracial interactions. Social Cognition, 23,
336-352.
Roberson, L., & Kulik, C. (2007). Stereotype threat at work. Academy of Management
Perspectives, 21, 24-40.
Rodgers III, W. (Ed.) Handbook on the economics of discrimination. Northampton, MA:
Edward Elgar.
Rooth, D. (2007). Implicit discrimination in hiring: Real world evidence. Discussion Paper 2764.
Berlin: IZA.
Bias in Hiring 44
Rooth, D. (2009) Obesity, attractiveness, and differential treatment in hiring: A field experiment,
Journal of Human Resources, 44, 710-735.
Schilt, K., & Wiswall, M. (2008). Before and after: Gender transitions, human capital, and
workplace experiences. The Bell Economic Journal of Economic Analysis and Policy, 8,
1-26.
Semin, G., & Fiedler, F. (Eds.) (1992). Language, interaction, and social cognition. Newberry
Park, CA: Sage Publications.
Shelley, R., & Shelly, A. (2009). Speech content and the emergence of inequality in task groups.
Journal of Social Issues, 65, 307-333.
Shelton, J., Richeson, J., & Salvatore, J. (2005). Expecting to be the target of prejudice:
Implications for interethnic interactions. Personality and Social Psychology Bulletin, 31,
1189-1202.
Smith, J., & Welch, F. (1989). Black economic progress after Myrdal. Journal of Economic
Literature, 27, 519-564.
Smith, T. (2000). Taking America’s Pulse II, NCCJ’s 200 survey of intergroup relations in the
United States. New York: National Conference for Community and Justice.
Stangor, C., Sechrist, G., & Jost, J. T. (2001). Changing racial beliefs by providing consensus
information. Personality and Social Psychology Bulletin, 27, 486–496.
Steele, C. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. In
Berkowitz, L. (Ed.) Advances in Experimental Social Psychology, Vol, 21 (pp. 261-302).
Bias in Hiring 45
Orlando, FL: Academic Press.
Steele, C., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of
African Americans. Journal of Personality and Social Psychology, 9, 797-811.
Stockdale, M., & Crosby, F. (2004). The psychology and management of workplace diversity.
Malden, MA: Blackwell.
Swim, J., & Samma, L. (1996). He’s skilled, she’s lucky, A meta-analysis of observers’
attributions for women’s and men’s successes and failures. Personality and Social
Psychology Bulletin, 22, 507-519.
Tajfel, H., Billig, M., Bundy, R., & Flament, C. (1971). Social categorisation and intergroup
behaviour. European Journal of Social Psychology, 1, 149–178.
Tal, A., Moran, G., Rooth, D., & Bendick, Jr., M. (2009). Using situation testing to document
employment discrimination against persons with psychiatric disabilities. Employee
Relations Law Journal, 35, 82-102.
Tomaskovic-Devey, D. (1993). The gender and race composition of jobs and the male/female,
white/black pay gaps. Social Forces, 72, 45-76.
Townsend, B. (1997). Confidential Study of Performance Evaluations. New York: Catalyst.
Trix, F., & Psenka, C. (2003). Exploring the color of glass, Letters of recommendation for
female and male medical faculty. Discourse and Society, 14, 191-220.
Bias in Hiring 46
U.S. Bureau of Labor Statistics (2008). Employed Persons by Detailed Occupations, Sex, Race,
and Hispanic or Latino Ethnicity, at www.bls.gov/cps/cpsaat11.pdf, downloaded
February 10, 2010.
U.S. Bureau of Labor Statistics (2009). Job Openings and Labor Turnover-August 2009, at
www.bls.gov/news.release/pdf/jolts.pdf, downloaded October 26, 2009.
U.S. Equal Employment Opportunity Commission (2010). Charge statistics FY 2007 through
FY 2009, at www.eeoc.gov/eeoc/statistics/enforcement/charges.cfm, downloaded
February 22, 2010.
Valian, V. (1998). Why so slow? The advancement of women. Cambridge, MA: MIT Press.
Wenneker, C., Wigboldus, D., & Spears, R. (2005). Biased language use in stereotype
maintenance: The role of encoding and goals. Journal of Personality and Social
Psychology, 84, 470-484.
Word, C., Zamma, M., & Cooper, J. (1974). The nonverbal mediation of self-fulfilling
prophecies in interracial interactions. Journal of Experimental Social Psychology, 10,
109-120.
Yelnosky, M. (2010). Testers Revisited. Legal Studies Paper 74. Bristol, RI: Roger Williams
School of Law.
Zitek, E.M. & Hebl, M.R. (2007). The role of social norm clarity in the influenced expression of
prejudice over time. Journal of Experimental Social Psychology, 43, 867-876.
Bias in Hiring 47
Appendix
A Primer on Matched Pair Tests of Hiring
Matched pair testing (synonyms: situation testing, paired-comparison testing,
employment auditing, field experiments, and employment testing) is a systematic research
procedure for creating controlled experiments analyzing employers’ candid responses to
employees’ personal characteristics (Bendick, 2007). Economists define employment
discrimination as valuation in the labor market of workers’ characteristics not related to
productivity (Arrow, 1998). In matched pair testing, pairs of research assistants apply for the
same actual job vacancy. Within each pair, employee characteristics related to a worker’s
productivity on the job — such as education, work experience, professional certifications, and
technical skills — are made equal by selecting, training, and credentialing testers to appear
equally qualified for the positions they seek. Simultaneously, personal characteristics unrelated
to job performance are experimentally manipulated by pairing testers who differ in only one
characteristic — a White paired with an African American, a male with a female, or a person age
32 with one age 57. If testers within these pairs experience substantially different responses to
their job-seeking efforts, few assumptions are required to attribute that difference to the
employers’ reaction to that characteristic.
Of course, this interpretation is appropriate only if employers are presented with pairs of
job candidates who truly appear equally qualified. This condition is relatively easy to achieve in
situation testing studies that involve only resumes which are mailed, faxed, or e-mailed to
employers (Bertrand & Mullainathan, 2002; Bendick, Jackson & Romero, 1996). In these
studies, testers’ resumes describe equivalent education, work experience, and job skills while
varying details and wording to avoid appearing obviously similar. The resumes communicate the
Bias in Hiring 48
demographic characteristics of the applicants through signals such as gender-specific names,
ethnically-related activities, or age-revealing dates of academic degrees, and similar clues. In
nations where resumes customarily contain applicant photographs, these images can also
communicate demographic characteristics.
Resume-based tests probe only the initial stages of the hiring process, ending with
employers’ decision to invite job candidates for in-person interviews. To study the complete
hiring process, it is necessary to provide “live” testers to file applications, return messages,
undergo pre-employment skills tests, and be interviewed. Ensuring that the testers in each pair
appear equally throughout those processes requires substantial care and effort (Bovenkerk, 1992;
Lodder, 1994).
The first step is to recruit research assistants who meet a daunting set of requirements:
ability to play the job-seeker role convincingly while simultaneously making accurate
observations about the hiring process; willingness to approach the study objectively; similarities
between testing partners in general appearance and demeanor; and the differing demographic
characteristics required by the study design. Recruiting individuals meeting all the requirements
is often a time-consuming, painstaking process; in one typical study, 93 potential testers were
interviewed before 4 were selected to form 2 testing teams (Nunes & Seligman, 1999). College
students, professional actors, actual job seekers, and community volunteers have all served as
testers.
The second step is training to make pairs of testers equally credible job applicants.
During training, testers develop and memorize their false resumes, receive coaching on effective
job interviewing techniques, and rehearse similar answers to common interview questions.
Concurrently, testers are trained to be “human tape recorders” by drilling them to notice and
Bias in Hiring 49
remember important details of their testing experiences and report them objectively. In typical
well-run testing studies, training requires about 3 days.
A third step in maintaining tester equivalence involves closely supervising testers’
actions. The testers within each pair usually present themselves to employers in random order,
with the second tester applying within minutes of the first. Each tester documents his or her
experiences as soon as practical after the event and prior to being told the experiences of her/his
testing partner. Testers typically record their data using pre-structured questionnaires, and they
are constantly reminded to focus on observable facts rather than interpretations. Such careful
management requires continuous, hands-on monitoring of each test by a trained “Test
Coordinator,” who can usually supervise no more than 3 testing teams concurrently.
Testing studies typically repeat their job application quasi-experiment for dozens or
hundreds of job vacancies, to “average out” random circumstances which may affect the
outcome in any single test. In analyzing test outcomes, one key summary statistic is the net rate
of discrimination, the proportion of job applications in which testers with the characteristic
hypothesized to be disfavored (e.g., African American) is successful minus the proportion of
applications in which testers with the characteristic hypothesized to be favored (e.g., Whites) is
successful. Another important summary measure is the ratio of the proportion of tests in which
testers with the disfavored characteristic are successful divided by the proportion of tests in
which the other tester is successful. In both computations, “successful” is typically defined as
reaching an identifiable milestone in the hiring process, such as being offered a job interview or
being offered a job.
Given the 20% to 40% net rates of discrimination observed in typical testing studies,
statistically significant estimates of the main effect studied in the experiment -- e.g., whether the
Bias in Hiring 50
net rate of discrimination is above zero – can be obtained with as few as 40 to 50 completed
tests. Samples of about 100 tests have sometimes proved sufficient to observe statistically
significant effects of mediating factors on net rates of discrimination – e.g., whether the net rate
of discrimination is higher in one location than another. Analyses that involve multivariate
techniques, such as multiple regression analysis (e.g., Kenney & Wissoker, 1994), often require
larger samples.
The following are examples of hiring outcomes that testing studies have concluded
indicate bias:
• A large-circulation newspaper carried an advertisement for a restaurant supervisor in an
affluent neighborhood. An African American tester who presented himself at the
restaurant was told that he would be called if the restaurant wished to pursue his
application. Minutes later, a similar-appearing White tester whose resume showed the
same level of education and restaurant experience followed the same procedure. He was
called later that day to schedule an interview, interviewed the day after that, and
subsequently offered the position. The African American tester made 4 follow-up calls to
reiterate his interest, including one shortly after the White tester declined the job offer,
with no response (Bendick, Jackson & Reinoso, 1994).
• A vacancy for a receptionist in an optometrist’s office was advertised in a local
newspaper in an affluent neighborhood. When a tester with a Latina name and slight
accent telephoned the following day, she was put on hold, called Carmen when she had
given her name as Juanita, and told that the office was not taking any further applications.
When her testing partner with an Anglo name and no accent called 13 minutes later, she
Bias in Hiring 51
was given an appointment for an interview the following morning (Bendick, Jackson,
Reinoso & Hodges, 1991).
• An employment agency advertised for an “account representative” to do executive
recruiting. Two white males, whose resumes and appearance portrayed them as age 32
and 57 respectively, responded by telephone and were granted interviews. The older
tester’s interview lasted 48 minutes, during which the tester was cautioned against
making a precipitous career change and instructed to call back if he was still interested
after reading books on sales techniques. The younger tester’s interview lasted 85
minutes, during which the interviewer discussed a variety of work and non-work topics in
a friendly manner and commented enthusiastically on the tester’s questions and
responses. This tester was invited back for a second interview, after which he was
offered a job (Bendick, Brown & Wall, 1999).
• An automobile service shop advertised in a newspaper for a technician to lubricate and
repair automobiles. When a female applicant whose resume showed experience in
physically demanding jobs applied for the position, the manager who interviewed her told
her that “the auto lube job is hard for a woman,” said that he liked her smile, and offered
an alternative, lower-paying customer service position in the on-premise coffee kiosk.
When her male testing partner applied for the position several hours later, he was
interviewed for the advertised technician position (Nunes and Seligman, 2000).
The most frequent criticism of testing studies is voiced by “free market” economists who
reason that, because employers are forced by a competitive labor market to consider only
productivity-related characteristics of job applicants, differences in hiring outcomes between
paired testers must signal failure by the testing researchers to match the testers on some subtle
Bias in Hiring 52
but important productivity-related characteristics (e.g., Heckman, 1998; see also Pager, 2007).
Such criticism is tautological, as well as contradicted by the research reviewed throughout this
paper. However, it provides an important caution that only researchers committed to conducting
their research with sustained thought, care, and rigor should undertake matched pair testing.
Bias in Hiring 53
About the Authors
MARC BENDICK, JR., is a principal in Bendick and Egan Economic Consultants, Inc., in
Washington, DC. An economist (Ph.D., University of Wisconsin), his 125 scholarly publications
concern poverty, employment, and public policies to enhance inclusion of individuals,
businesses, and communities in the economic mainstream. He is a consultant on workforce
diversity management to major employers, a frequent expert witness in employment
discrimination litigation, and has led or participated in several dozen matched pair testing
studies.
ANA P. NUNES is a Ph.D. candidate in social psychology at the University of Colorado at
Boulder. Her research interests focus on person perception processes and the effect of diversity
mandates on judgments and behavior, especially in high-stake decision contexts such as hiring.
She formerly ran matched pair testing projects investigating housing discrimination for Project
Sentinel in Palo Alto, California, and employment discrimination for the Discrimination
Research Center in Berkeley, CA.
