Files
Abstract
This thesis investigates whether increasing high school math graduation requirements can reduce the gender gap in math self-efficacy among American students. Math self-efficacy, defined as students’ belief in their ability to succeed in math, is a key predictor of ad- vanced course-taking, STEM major selection, and long-term persistence. Using nationally representative data from the Educational Longitudinal Study of 2002, I apply a difference- in-differences design with propensity score matching to estimate the effects of moving from two to three and from three to four years of required math coursework. While the two-to- three-year change shows no effect, the shift to four years leads to statistically significant gains in self-efficacy, particularly after accounting for gender. Girls report consistently lower confidence than boys, even after controlling for academic ability, and further interaction anal- yses show that race and baseline math beliefs influence how students respond to increased requirements. These findings provide new causal evidence that structural policy can shape student confidence and suggest that graduation mandates may play a role in reducing gender disparities in math and STEM pathways.