A Threatening Intellectual Environment: Why Females Are Susceptible to Experiencing Problem-Solving Deficits in the Presence of Males

Being in the minority matters for women when participating in gender-stereotyped tasks. Working in groups where women are in the numerical minority (e.g. two men and one woman) can elicit stereotype threat and result in a performance deficit, performing below their ability, on negatively stereotyped tasks, like math tests. 

• Women in the minority test-taking group (two men and one woman) had lower accuracy on math tests with a score of 55% than women participating in all-women groups (three women) who scored on average 70%, demonstrating the effects of stereotype threat.
• There are no significant differences in performance on verbal tests for women in the minority groups and women in same-sex groups.
• There are no significant differences in performance on verbal or math tests for men in the minority groups and men in same-sex groups, demonstrating that men are not experiencing stereotype threat.
• Interestingly, women’s lower performance in math is proportional to the relative number of men in a group: as the number of men in a group increases, the stereotype threat increases and their scores decrease:
  • In same-sex groups, women demonstrated 70% accuracy.
  • In a mixed-majority (two women, one man), women demonstrated 64% accuracy.
  • In a minority-group (one woman, two men), women demonstrated 58% accuracy.

Because women underperformed only on math tests, but not on verbal tests, the authors conclude that this comes from stereotype threat, rather than tokenism (being the only one in a group). Women did not need to be directly reminded of negative stereotypes to underperform, rather the presence of men in a group had the same effect by eliciting stereotype threat. The authors suggest that this finding has implications for same-gender education, as women participants show these deficits not just when they are outnumbered, but when men are present.

The researchers conducted two experiments with a sample of undergraduate students at Brown University. In the first experiment, 72 women undergraduates were asked to take a math or a verbal test in a group of three participants. Women students were randomly assigned to a same-gender condition, in which they took the test with two other women, or to a mixed-gender condition, in which they took the test with two men. All participants in the study expected to interact with one another and compare their experiences with the test itself, which therefore included an element of social comparison within the experiments. By comparing women’s performance in the same-gender and mixed-gender treatments, the researchers intended to measure whether being in a group with men decreased women’s performance on the exam. By varying whether the women took a math or verbal exam, the researchers could examine if the change in performance was specific to domains for which there exists gender stereotypes or was more general.

In the second experiment, 92 men and women undergraduates were asked to take a math test. In this experiment, researchers varied the proportion of men and women in a group to determine if women participants’ experience performance deficits on math tests even when they were not a minority member of the group.