Confirming the Stereotype: How Stereotype Threat, Performance Feedback, and Academic Identification affect Identity and Future Performance

This study investigates the post-performance effects of stereotype threat. Undergraduate students (N = 130) classified as either stronglyor weaklyidentified with academics were told a diagnostic anagram task either typically shows poorer performance for their gender (stereotype threat) or no gender differences (no stereotype threat), and received arbitrary positive or negative feedback on an initial task. They later performed a second anagram task. Results indicate a 2-way interaction between stereotype threat and academic identification among those who received negative feedback. Negative feedback under stereotype threat did not harm performance for participants strongly-identified with academics, but did harm performance for participants weakly-identified with academics. This same 2-way interaction within the negative feedback condition also predicted post-feedback levels of identification as a college student, though it did not seem to affect post-feedback levels of academic identification. Strongly-identified participants receiving negative feedback identified less as a college student if they were under stereotype threat while weakly-academically identified participants identified more. Levels of post-feedback identification as a college student negatively predicted performance. CONFIRMING THE STEREOTYPE 3 Confirming the Stereotype: How Stereotype Threat, Performance Feedback, and Academic Identification affect Identity and Future Performance Threats to social identity remain one of the most powerful but publically underestimated social processes in contemporary society. The worry that we will be evaluated based on our group memberships rather than out own personal characteristics can feel just like threats to our own selves—both psychologically and physiologically. The implications of social identity threat have been researched both in the long term and the short term, and it has been found to affect many domains including attributions to discrimination (Eccleston & Major, 2006; Kaiser, Dyrenforth, & Hagiwara, 2006; Verkuyten, 1998), academic performance (Osborne & Walker, 2006; Steele & Aronson, 1995) interpersonal interactions (Castelli, Pavan, Ferrari, & Kashima, 2009; Leary & Schreindorfer, 1998; Romero-Canyas, Anderson, Reddy, & Downey, 2009), and even sports performance (Beilock, & McConnell, 2004; Stone, Lynch, Sjomeling, & Darley, 1999; Stone & McWhinnie, 2008). While long-term exposure to social identity threat (e.g., having a stigmatized identity) has been hypothesized to cause several detrimental outcomes (including poor health, reduced involvement in academics, etc.; Allison, 1998; van Laar, 2000), shortterm-effects of social identity threat have more immediate and measurable deleterious effects. One of the most powerful short-term effects of social identity threat involves the extent to which negative stereotypes affect performance on important academic tasks and tests. Known as stereotype threat (Steele, 1997; Steele & Aronson, 1995), this form of social identity threat specifies that cuing negative stereotypes about one’s group in an evaluative setting will cause depressed performance because negatively-stereotyped group members are CONFIRMING THE STEREOTYPE 4 threatened by their devalued social identity. While the phenomenon of stereotype threat has been replicated among many different social identities and skills, little research has attempted to link the short-term effects of experiencing stereotype threat (i.e., depressed performance) to the more long-term effects of social identity threat (i.e., less academic striving and lower engagement with academics). This study will investigate what happens when we either confirm or disconfirm the negative stereotypes that exist about our groups in academic settings. By looking at how receiving positive or negative performance feedback on diagnostic tasks presented as either gender-fair or gender-biased affects subsequent performance on similar tasks, we can better understand how short-term experience with social identity threat may affect whether we are able to strive on similar tasks in the future, or whether we disengage.