Authors: Mackenzie Alston, William A. Darity Jr., Catherine C. Eckel, Lawrence McNeil, Rhonda Sharpe
Abstract: We conducted lab experiments at a historically black university (HBCU), replicating the design and procedure, but not the results, of previous stereotype threat studies. The experimental design has two factors: stereotype salience (priming) and the identity of the experimenter (a less-threatening black woman vs. a more-threatening white man). Unlike previous studies, we found no effect of stereotype threat on student performance. We find little evidence that black students at the HBCU are affected by stereotype threat, regardless of the identity of the experimenter. We found no significant difference in the number of questions answered correctly by subjects in the control and treatment conditions in either the white male or the black female experimenter sessions. Finally, we found little evidence to support our prediction that subjects would respond differently to the identity of the experimenter. Having a black female experimenter, as opposed to a white male experimenter, had no effect on the number of questions answered correctly.
Key Findings
- Experimental studies have found that when Black students are reminded of their racial identity before taking a test – and thus influenced by this negative stereotype – Black students have significantly higher blood pressure rates, are more cognizant of their racial identity, and perform worse on the test when it is framed as a measure of intelligence.
- However, these studies largely took place at predominantly white institutions (PWIs); there is reason to believe that the unique qualities of the HBCU experience – particularly diverse faculty and greater exposure to success of individuals in one’s identity group – might help diminish the deleterious effects of stereotype threat.
- To test this, the authors asked students at an HBCU in Texas to answer a set of 18 verbal Graduate Record Examinations (GRE) questions.
- The authors found no significant difference in the number of questions answered correctly by the subjects in the control and treatment groups – that is, the students who weren’t reminded of their race and those who were reminded performed equally well on the test.
- Moreover, the identity of the test administrator did not appear to have an effect: there was no significant difference in the number of correct answers given in the high-threat treatment (with a white male researcher) versus the low-threat treatment (with a Black female researcher).