Examining multiple identities under stereotype threat : a regulatory fit perspective

This study sought to examine how underlying characteristics of multiple social identities could explain why some people are not affected by stereotype threat. Specifically, it proposed that different identities are not only associated with positive or negative stereotypes, but also different regulatory foci. It additionally sought to address a common methodological issue in the literature by including non-targets of stereotype threat as a comparative group. Using a quantitative experimental design, math-identified male (N=104) and female (N=172) university students were randomly assigned to take a difficult math test under circumstances which varied both reward structure and salient identities. For math- identified females, their gender identity was believed to invoke a negative stereotype about female math ability and thus stereotype threat. However, college identity was proposed to be positively stereotyped about ability. When both were made salient, females would suppress their gender identity in order to maintain a good self-concept and would thus be protected from stereotype threat effects. Furthermore, it was predicted females under threat would enter into a prevention regulatory focus and thus perform better under a reward structure which focused on minimizing losses. A major criticism of stereotype threat research is that it fails to differentiate itself enough from stereotype priming. While the two are similarly activated, stereotype threat only affects those for whom the stereotyped identity is relevant. Thus it is important to include non-targets of threat to ensure that the experimental manipulations do not affect them. Males were included in this study because the negative stereotype about female math ability is not relevant. Results indicated that when gender identity was made salient, math-identified females performed worse than a control group. However, when both gender and college identity were made salient, math-identified females performed better than those only reminded of their gender, and equivalent to those in a control group. Reward structure showed no main effect on performance. While the interaction between identity salience and reward structure was marginally significant, more research is needed to determine if there is a true relationship. Males showed no differences across conditions however, which indicates this was a more valid manipulation of threat.