Sex Roles [sers] pp671-sers-454960 October 30, 2002 18:21 Style ﬁle version June 3rd, 2002
Sex Roles, Vol. 47, Nos. 3/4, August 2002 (
An Examination of Implicitly Activated, Explicitly
Activated, and Nulliﬁed Stereotypes on Mathematical
Performance: It’s Not Just a Woman’s Issue
Jessi L. Smith
and Paul H. White
This study was designed to examine the different ways that stereotypes might become activated
in testing situations and the effects this activation has on task performance. In Experiment 1,
women undergraduates exposed to an explicitly activated stereotype (i.e., told men outperform
women in mathematics) performed worse than women exposed to a nulliﬁed stereotype (i.e.,
told men and women perform at the same level in mathematics). The stereotype threat also
was activated implicitly under “normal” conditions (i.e., just given the test with nothing else
stated) such that performance in this condition was at the same (low) level as the explicitly
activated threat. In Experiment 2, the results were replicated with White male undergraduates
using the stereotype that “Asians are better than Whites” in mathematics. In addition, in a
small ﬁeld survey we found that this belief about ethnicity did occur spontaneously for White
men in college calculus courses. Taken together, the results of these studies suggest that even
under normal circumstances, math test situations may lead to nonoptimal performance for
both stigmatized (women) and traditionally nonstigmatized (White men) group members.
Implications for threat nulliﬁcation techniques are discussed.
KEY WORDS: stereotype threat; mathematics; group processes; motivation.
“Math class is tough.”
Teen Talk Barbie, 1992
When Mattel introduced its second talking
Barbie doll, the company inadvertently created an up-
roar by including the phrase “Math class is tough” in
Teen Talk Barbie’s repertoire. Members of the pub-
lic, the media, and particularly the American Asso-
ciation of University Women (AAUW) vehemently
objected to the phrase (e.g., Brandon, 1992), which
Mattel eventually dropped from Barbie’s lexicon. It
University of Utah, Salt Lake City, Utah.
Present address: Department of Psychology, Ohio State Univer-
sity, Newark, Ohio.
To whom correspondence should be addressed at Department
of Psychology, Ohio State University, 1179 University Drive,
Newark, Ohio 43055; e-mail: firstname.lastname@example.org. Department
of Psychology, University of Utah, 380 South 1530 East, Rm.
502, Salt Lake City, Utah 84112; e-mail: paul.white@psych.
appears that Mattel was working under the assump-
tion that mathematics is not for girls, thus illustrating a
widely held belief in our society that girls and women
do not, or can not, do math.
As with most stereotypes, there may be a “ker-
nel of truth” (Allport, 1954/1979) behind this be-
lief. Educational researchers have documented that
girls do not perform as well as boys in standardized
math tests such as the SAT (Scholastic Aptitude Test)
quantitative section (Benbow & Stanley, 1980, 1983;
Halpern, 1992, as cited in Neisser et al., 1996; Kim-
ball, 1998), the GRE (Graduate Record Exam) quan-
titative section (Grandy, 1994), and advanced math
placement exams, in which boys score on average 37–
38 points higher than girls (AAUW, 1992). Further-
more, girls are less likely than boys to enroll in math
classes (e.g., Eccles, l984), select math as a major in
college (LeFevre, Kulak, & Heymans, 1992; Stangor
& Sechrist, 1998), or pursue math as a career (e.g.,
Betz & Hackett, 1981). Remarkably, even girls who
2002 Plenum Publishing Corporation