Stereotype threat and inflexible perseverance in problem solving

The present research examines whether women burdened by stereotype threat, a threat of confirming negative ingroup stereotypes (Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69(5), 797-811), are less able to abandon old strategies and employ newer, more efficient ones when conditions change. In two studies, stereotype threat was found to increase inflexible perseverance: women made to believe they were taking a diagnostic math/spatial ability test, compared to those not threatened by stereotypes, were more likely to use previously successful but presently inefficient or incorrect strategies. In Study 1, participants under stereotype threat also suppressed relevant stereotypes to the greatest degree, and their inflexible perseverance was predicted by the degree to which they suppressed these stereotypes. Implications for test performance and potential decision-making effects of stereotype threat are discussed.     

When women can’t do math: The interplay of self-construal, group identification, and stereotypic performance standards

The present research examined the interplay of self-construal (independent vs. interdependent), gender group identification, and performance standards (positive vs. negative) on women’s math performance. Female participants were given a subtle self-prime prior to completing a math test under conditions where either a positive or negative group-based performance standard was rendered accessible. We report an interactive effect of self-construal, gender identification, and performance standard such that a negative (compared to a positive) standard decreased performance under interdependent self-prime (“we”) conditions, whereas the reverse pattern emerged under independent self-prime (“I”) conditions. Importantly, we observed this interplay of performance standards and self-construal only in individuals who self-identify with their gender group whereas performance outcomes of low identifiers were not affected by the experimental manipulations.     

Category and stereotype activation revisited

In Study 1 (N= 230), we found that the participants' explicit prejudice was not related to their knowledge of cultural stereotypes of immigrants in Sweden, and that they associated the social category immigrants with the same national/ethnic categories. In Study 2 (N= 88), employing the category and stereotype words obtained in Study 1 as primes, we examined whether participants with varying degrees of explicit prejudice differed in their automatic stereotyping and implicit prejudice when primed with category or stereotypical words. In accord with our hypothesis, and contrary to previous findings, the results showed that people's explicit prejudice did not affect their automatic stereotyping and implicit prejudice, neither in the category nor stereotype priming condition. Study 3 (N= 62), employing category priming using facial photographs of Swedes and immigrants as primes, showed that participants' implicit prejudice was not moderated by their explicit prejudice. The outcome is discussed in relation to the distinction between category and stereotype priming and in terms of the associative strength between a social category and its related stereotypes.     

The effect of stereotype threat on student-athlete math performance

ObjectivesTo assess whether a subtle stereotype threat of student-athletes would cause a decrease in both academic effort and performance.DesignA 2 (Male/Female) x 2 (Athlete Prime/No Athlete Prime) design was used to assess effort and performance on a math test.MethodA subtle threat manipulation was used to prime half of 60 NCAA Division III student-athletes with their athletic identity prior to taking a difficult math test.ResultsSupporting the hypotheses, student-athletes who were primed with their athletic identity attempted significantly fewer problems and received lower mean math scores than those who were not primed. Contrary to hypotheses, gender did not impact effort or performance, and there was no evidence of buffering effects of priming non-athlete identities.ConclusionsThe results of this experiment provide evidence for stereotype threat effects across genders and into Division III athletes, which potentially impact student-athlete academic performance.     

The “Obama Effect”: How a salient role model reduces race-based performance differences

Barack Obama, the first Black-American president, has been widely heralded as a role model for Black-Americans because he inspires hope. The current study was conducted to assess whether, beyond simply inspiring hope, this “Obama Effect” has a concrete positive influence on Black-Americans’ academic performance. Over a three-month period we administered a verbal exam to four separate groups of Black- and White-American participants at four predetermined times. When Obama’s stereotype-defying accomplishments garnered national attention - just after his convention speech, and election to the presidency - they had a profound beneficial effect on Black-Americans’ exam performance, such that the negative effects of stereotype threat were dramatically reduced. This effect occurred even when concerns about racial stereotypes continued to exist. The fact that we found performance effects with a random sample of American participants, far removed from any direct contact with Obama, attests to the powerful impact of ingroup role models.     

Approaching math increases math = me and math = pleasant

In two studies, we examined the effect of extensive practice in approaching math on implicit identification with math, implicit math attitudes, and behavior during a math test. The results from Study 1 demonstrated that women trained to approach math showed more identification with and positive implicit attitudes toward math than women trained to avoid math. Notably, this latter pattern of findings was only evident for women low in initial identification with this field. The results from Study 2 replicated these findings by showing that women who were initially low in math identification and trained to approach math showed more implicit identification with math and attempted more items on a math test than women trained to respond to math in a neutral way. The implications of these findings for current theorizing on the gender gap in women’s representation in math related careers are discussed.     

Performance boosts in the classroom: Stereotype endorsement and prejudice moderate stereotype lift

The present study (N = 293) examined whether stereotype endorsement and prejudice moderate stereotype lift (i.e., a performance boost caused by the salience of a negative out-group stereotype in the testing situation). The stereotype in the focus of inquiry was the belief that immigrant students have lower intellectual ability than native students. French native high school students performed an intellectual test in a condition of low stereotype salience (the test was presented as assessing individual differences) or in a condition of high stereotype salience (the test was presented as assessing group differences between African immigrants and native students). As expected, results indicated that native students high in stereotype endorsement and those high in prejudice performed better in the high than in the low stereotype salience condition, whereas those low on these constructs did not. By identifying two moderators of stereotype lift, this study sheds new light on the achievement gap between immigrant and native students in educational institutions.     

Trait expectancies and stereotype expectancies have the same effect on person memory

Two studies examined whether trait expectancies and stereotype expectancies similarly affected memory for expectancy-relevant behaviors. The results of both studies showed that, when activated in identical ways, trait expectancies and stereotype expectancies had similar effects on recall. Better recall was obtained for expectancy-incongruent items when expectancies were activated in relatively overt ways, while better recall for expectancy-congruent items was obtained when expectancies were activated in less overt ways. Additional analyses examining the order of events that emerged in the recall protocols provided little evidence that participants attempted to reconcile the items, as would be predicted by some earlier models of person memory (e.g., Srull, 1981). The implications of these findings for how social expectancies guide social information processing are discussed.