The costs of accepting gender differences: The role of stereotype endorsement in women's experience in the math domain
Two experiments assessed the impact of individuals' stereotype endorsement on their performance under stereotype threat. In Experiment 1, women undergraduates who were majoring in a math-related field completed a survey in which they were asked about their endorsement of stereotypes (e.g., "In general, men may be better than women at math"), the legitimacy of gender-based status differences (e.g., "Differences in status between men and women are fair"), their perceptions of their own ability, confidence, and self-esteem in math, and their intentions of continuing in the field of mathematics. Results showed that stereotype endorsement was negatively correlated with confidence, performance self-esteem, and desire to attend graduate school in the major. Analyses controlling for GPA left these results unchanged, ruling out the possibility that actual performance in the major accounted for these effects. In Experiment 2, female undergraduates from a variety of majors indicated their degree of stereotype endorsement using the same questionnaire used in Experiment 1. Later in the semester, they were asked to complete a math test that focused either on "each individuals performance on the test...comparing their individual scores to those of other students" (control) or on "how women score on the test relative to men...comparing womens scores to mens scores...using each individuals score as an indicator of womens math ability" (stereotype threat). Women in the stereotype threat condition also indicated their sex before taking the test. Performance was marginally worse under stereotype threat than in the control condition, but this effect was moderated by stereotype endorsement. Under stereotype threat, stereotype endorsement was negatively related to womens math performance. In the control condition, however, endorsement of the stereotype was positively associated with test performance. Thus, women who tended to believe in the accuracy of gender stereotypes about womens math ability tended to perform worse on a stereotype-relevant test when their gender identity was highlighted.