||The current research used Eccles' expectancy-value theory to assess the relationship between sex-typing and non-traditional career choice. I predicted that sex-typed individuals would be less likely than cross-sexed or non-sex-typed individuals to pursue college majors that are stereotypically at odds with their biological gender. I also predicted that work-related motivations and self-efficacy would mediate the relationship between sex-typing and major choice. Multiple regression analyses assessed (1) the main effect of sex-typing, and (2) whether self-efficacy and / or work-related motivations mediated the relationship between sex-typing and major choice; Sobel tests assessed the significance of indirect effects in the mediation models. Masculinity and femininity predicted college major. Self-efficacy, but not work-related motivations, partially mediated the relationship between sex-typing and college major choice, but only in the femininity model. Results suggest that targeted interventions could be useful to increase female involvement in STEM and to increase non-traditional career choice in sex-typed individuals who would otherwise be hesitant due to gender-role stereotypes.