Abstract

The growing demand for data expertise has heightened the need to understand and address persistent gender disparities in statistics and data science. A primary area of concern regarding these gender disparities centers on differences in self-efficacy among graduate students. This study investigates the psychological mechanisms underlying these disparities by testing a Hayes’ conditional process model grounded in Social Cognitive Theory. A sample of 299 master’s students from statistics and data science programs completed measures assessing gendered perceptions, domain identification, grade point average (GPA), and self-efficacy in statistics. Results revealed a gender gap in self-efficacy, with women reporting overall lower self-efficacy than men. However, our findings showed a significant indirect path from gender to self-efficacy through gendered perceptions, suggesting that women who are less likely to perceive statistics as a masculine domain show increased self-efficacy. Furthermore, domain identification was a strong positive predictor of self-efficacy, while GPA was a surprising negative predictor. This research provides novel insights into the motivational dynamics at the graduate level and suggests that supporting views of statistics and data science as gender inclusive  and increasing domain identification can have a positive effect on self-efficacy.

This work is licensed under a Creative Commons Attribution 4.0 License.