Stereotype threat is "an uncomfortable feeling that arises when
people are at risk of confirming a negative stereotype in the eyes of
others" (Steele, James, & Barnett, 2002). Stereotype threat can have an
effect on an individual's performance if he or she knows "that some people
hold a negative stereotype about their group" (O'Brien & Crandall, 2003).
In a stereotype threat experiment, the participants that are concerned with
confirming a negative stereotype about their group will perform more poorly
than those who are not threatened by a stereotype (O'Brien & Crandall,
2003). If stereotype threat is removed in these experiments there is no
difference in the participant's performance. This may suggest that "many
of the racial or gender group differences that appear to characterize
groups are in fact the results of stereotype threat- an environmental
challenge rather than an essential difference" (O'Brien & Crandall, 2003).
Some examples of stereotypes that can have an effect on one's performance
are: Caucasian men are smarter than African American men, African Americans
are better at sports than Caucasians, or Asian men are better at academics,
especially math, than Caucasian men. In the realm of mathematics, women
are seen as inferior to men. If women are aware of this stereotype their
mathematical performance will suffer as a result.
The stereotype that men are better at math than women contradicts
gender differences in math achievement found among school age children.
Quinn and Spencer (2001) reported that throughout elementary school and
middle school girls are outperforming boys on computation tests and are
performing equally on word problem tests. Starting in junior high, this
pattern begins to reverse. The gap in performance on math word problems
grows through high school, college, and adulthood. Furthermore, women
appear to be leaving male-dominated academic areas faster than men. In
college, more men enter the areas of math, science, and engineering.
Although some women enroll in these traditionally male academic areas, very
few of them stay in these fields throughout their college years (Steele,
James, & Barnett, 2002).
Steele, James, and Barnett (2002) have hypothesized that the most
crucial factor responsible for these trends is stereotype threat. They
found that women in male-dominated areas (defined as math, science and
engineering) were "most likely to report feeling threatened by negative
gender stereotypes that allege they are not as capable as men" (Steele,
James, & Barnett, 2002). Also, women in male-dominated areas were more
susceptible to gender discrimination than any other students. These women
also expected the most discrimination if they pursued a career in their
field. Contrary to the experimenters' hypothesis, "women in math, science
or engineering were no less identified with their specific field of study"
than students in other fields (Steele, James, & Barnett, 2002). However,
as expected, these students were most likely to think about changing their
major. This finding is unsettling because women who identify with their
field are pressured into leaving it strictly based on their gender.
Previous experiments have addressed the effect stereotype threat has
on women's mathematical performance. Inzlicht and Ben-Zeev (2000) found
that when women were placed in a "threatening intellectual environment
(when they were outnumbered by males)" they scored lower on a math test
than women who took the test in a same-sex group. Women's performance only
suffered in stereotyped domains. In the presence of males, women scored
lower on mathematical tasks; however, their verbal performance was not
affected by being in the minority among men. Inzlicht and Ben-Zeev (2000)
also concluded that women did not have to be the minority among men for
their math performance to suffer. Women's math scores were negatively
correlated with the number of males present.
Other research has examined the effect stereotype threat has on
different kinds of tasks. O'Brien and Crandall (2003) found that
stereotype threat affected performance on easy and difficult tasks in
different ways. Heightened levels of arousal may account for this
difference in effects of stereotype threat. If the participant is
concerned with confirming to a negative stereotype or being an example of a
negative stereotype their arousal level will be higher (O'Brien and
Crandall, 2003). High levels of arousal should have an impact on
performance. "When the task is complex or difficult, arousal will lead to
an increased number of inappropriate responses;" however, "when the task is
simple, or well learned, arousal will lead to an increased number of
appropriate responses thereby increasing overall performance" (O'Brien and
Crandall, 2003). Theses findings were in agreement with the arousal
hypothesis. When compared to women who took a math test in a same-sex
group, women under stereotype threat (males present) scored higher on an
easy math test and scored lower on difficult math problems (O'Brien and
Crandall, 2003).
Quinn and Spencer (2001) examined the difference in women's
achievement on different kinds of tests, specifically word problem tests.
Women took a word problem test or a numerical test "composed of problems
that were the mathematical equivalents of the word problems" (Quinn and
Spencer, 2001). The only difference between the two types of problems was
that when participants were completing word problems they needed to use an
extra step to "convert the word problems into their numerical equivalents"
(Quinn and Spencer, 2001). On the numerical test, men and women's
performances did not differ. However, on the word problem test, women's
score were significantly worse than the men's scores (Quinn and Spencer,
2001). Quinn and Spencer (2001) also found that manipulating the level of
stereotype threat had an effect on women's performance. When stereotype
threat was high, women's scores suffered and they were less likely than men
to come up with a strategy to solve the math problem they were working on.
When the same test was presented as "gender fair," women's performance did
not differ from men and they were just as likely to formulate a strategy
for solving problems as men (Quinn and Spencer, 2001).
Current research shows that the mere thought of a negative stereotype
about one's group can impair performance. In regards to male-dominated
areas, such as math, science, and engineering, women steer clear of these
fields or they enter them with the risk that their performance will be
hindered. The question remains, what can be done about a stereotype that
is so deeply rooted in our culture? Perhaps some simple changes may help.
"Adding a line in the instructions communicating that a test is gender fair
or nondiagnostic" may close the gender gap in performance (Quinn and
Spencer, 2001). Furthermore, parents and teachers must be aware of how
their behaviors and attitudes can influence how girls and women view
mathematics. This may lead to a change in the attitudes women have toward
math, science, and engineering. Only when women are not constantly at the
mercy of stereotype threat will their level of achievement in mathematics
improve. If women experience stereotype threat less frequently it will
have a positive effect on their math performance.