A Stereotype Boost or Choking Under Pressure?
Positive Gender Stereotypes and Men Who Are Low
in Domain IdentificationA STEREOTYPE BOOSTSMITH AND JOHNS N
Jessi L. Smith and Camille S. Johnson
The Ohio State University
In 3 studies, we examined the effect of a positive gender stereotype (e.g., men are superior to
women in math and computer science) on performance and motivation as a function of domain
identification, with a special emphasis on men low in domain identification. Drawing from past
research on stereotype boost and choking under pressure, we predicted that for men low in do-
main identification, performance would ironically suffer in the positive stereotype condition
compared to a nullified stereotype condition. In all 3 studies, we found that men lower in do-
main identification performed better in nullified stereotype conditions and poorly in gen-
der-stereotypical conditions. Results suggest that the interaction between the expectancies con-
veyed by a stereotype and the value placed on a domain by an individual may influence
performance and motivation regardless of stereotype valence.
Stereotypes are frequently thought of as conveying negative
beliefs about members of a particular group. Often those
same stereotypes contain positive beliefs about members of
another group; they suggest that members of one group are
more proficient than members of another group within a par-
ticular domain (Shih, Ambady, Richeson, Fujita, & Gray,
2002). For instance, there is a well-known gender stereotype
that women are not as good as men at math. This gender math
stereotype conveys expectations for both groups: Women are
expected to be uninterested in and to perform poorly on
math-related tasks, whereas men are expected to take interest
in and excel at math-related tasks (e.g., Hyde, Fennema,
Ryan, Frost, & Hopp, 1990; Smith & White, 2002; Walton &
Cohen, 2003). The dual nature of most stereotypes in regards
to targeted versus nontargeted group members is a relatively
new area of study in research on stereotypes and perfor-
mance. Traditionally, research on stereotypes and perfor-
mance has examined the effects of negative stereotypes on
marginalized group members. Dominant group members are
often used only as a control comparison with the targeted
marginalized group member. As suggested by Leyens,
Desert, Croizet, and Darcis (2000), “clearly more research is
needed to understand how stereotype threat acts on dominant
group members” (p. 1191).
Dominant group members are often positively stereo-
typed, perhaps as a way to maintain the status quo (called
system justification; Jost & Kay, 2005). In this project, we
examined whether and how positive stereotypes affect per-
formance (and motivation) for dominant group members
with an emphasis on the role of domain identification.
POSITIVE STEREOTYPES AND
PERFORMANCE: A STEREOTYPE BOOST
OR CHOKING UNDER PRESSURE?
Some research has suggested that positive stereotypes may
boost performance for dominant group members. This per-
formance benefit, associated with the reminder of the relative
advantage of one’s own group over another, has been dubbed
“stereotype boost” or “stereotype lift” (Shih et al., 2002;
Shih, Pittinsky, & Ambady, 1999; Walton & Cohen, 2003).
Stereotype boost/lift refers to the enhanced performance re-
sulting from a stereotype that is directly relevant to one’s
group. For example, in their meta-analysis of stereotype
threat studies, Walton and Cohen (2003) found that the con-
ditions that produced stereotype threat effects (lowered per-
formance) among negatively stereotyped participants led to
moderate stereotype lift effects (improved performance)
among the nonstereotyped participants (see also Croizet et
al., 2004). For example, although the gender math stereotype
is framed as “women cannot do math,” the implication is that
BASIC AND APPLIED SOCIAL PSYCHOLOGY, 28(1), 51–63
Copyright © 2006, Lawrence Erlbaum Associates, Inc.
Correspondence should be addressed to Jessi L. Smith, Department of
Psychology, The Ohio State University, 1179 University Drive, Newark,
OH 43054. E-mail: Smith-Klaphake.1@osu.edu
D
ow
nl
oa
de
d
by
[A
mh
ers
t C
oll
eg
e L
ibr
ary
] a
t 1
0:2
2 2
0 D
ec
em
be
r 2
01
1