Influence of item content and stereotype situation on gender differences in mathematical problem solving

Abstract

On standardized tests of mathematical problem solving, the typical finding has been that women score lower than men. Experiment 1 manipulated gender labeling (female character, male character, or gender neutral) within the problem question to see whether this accounted for gender differences in mathematical problem solving. Sixty-four seventh and eighth graders were tested on modified versions of the Canadian Test of Basic Skills (CTBS) with the results showing that although gender labeling affected students performance, it did not account for gender differences. Experiment 2 manipulated both gender labeling and gender stereotype threat for 174 university students writing modified versions of a model Standardized Achievement Test (SAT). Again, gender labeling within problem questions did not account for gender differences. However, women scored lower than men when they believed that the test had previously shown gender differences. There was no gender difference in the performance of the same women and men when they believed that the test was merely comparing Canadian students with American students. This suggests that gender stereotype threat could be a key factor in explaining gender differences in mathematical problem solving.