Men and women occupy very different sectors of the labor market and women continue to be disadvantaged by this segregation in prestige, pay and power. Much research has examined how segregation is reproduced by discrimination and coercion, but as such actions become increasingly culturally illegitimate, we must begin to ask why gender inequality persists in the face of its illegitimacy. This project examines how two culturally-informed, individually -held sets of beliefs, gender schemas (beliefs about the appropriate roles and "essential natures" of men and women) and self-conceptions (gendered beliefs about the self) inspire decision-making that reproduces occupational gender segregation. Using 5-year longitudinal panel data of students from four institutions (MIT, Olin, Smith and UMass), I study how gender schemas and self-conceptions influence students' major selection and their career decisions 18 months after graduation. This project theorizes a phenomenon I call "the self-expressive edge of sex segregation," the reproduction of occupational sex segregation through the individualistic, self-reflective, yet culturally-informed decisions of social actors. This dissertation shows, first of all, that gender schemas and self-conceptions are indeed co-constructed: our most individualistic beliefs--who we think we are as people-- are informed by our adherence to cultural beliefs about gender. These culturally-informed self-conceptions not only help direct the distribution of men and women into already-segregated college majors and career fields, they help to reproduce the cultural sex-typing of such fields as "masculine" and "feminine" domains. As I show with a case study of engineering students, these individual-level beliefs can also interact with professional cultures in such a way that segregation is reproduced in the very process of learning to be a professional. In these mechanisms of sex segregation, cultural gender beliefs act through self-conceptions to influence career decisions. This cloaking of gendered career decision-making in individualistic self-expression may be particularly resistant to social change, as undermining individuals' self-expressive freedoms would be a culturally unacceptable method to reduce gender inequality

Some of the most male-dominated science, technology, engineering and mathematics occupations and degree programmes are found in the world's most affluent societies. This article assesses whether gender gaps in attitudes follow similarly surprising patterns. Multivariate analysis of eighth-graders' affinity for mathematics and aspirations for mathematically related jobs in 53 countries shows that the attitudinal gender gap is indeed larger in affluent 'postmaterialist' societies. Moreover, both girls and boys view mathematics more negatively in these societies. The authors suggest that cultural ideals of individual self-expression, highly prevalent under conditions of broad-based existential security, operate to reduce girls' and boys' interest in pursuits thought to be economically practical but personally non-expressive. Girls may be particularly susceptible to this negative effect, because taken-for-granted cultural beliefs about core female personality traits (and girls' gendered understandings of their own authentic inner selves) are often at odds with dominant representations of mathematical and technical work. © 2014 Taylor & Francis.

Social psychological research on gendered persistence in science, technology, engineering,and mathematics (STEM) professions is dominated by two explanations: women leavebecause they perceive their family plans to be at odds with demands of STEM careers, andwomen leave due to low self-assessment of their skills in STEM’s intellectual tasks, net oftheir performance. This study uses original panel data to examine behavioral and intentionalpersistence among students who enter an engineering major in college. Surprisingly, familyplans do not contribute to women’s attrition during college but are negatively associated withmen’s intentions to pursue an engineering career. Additionally, math self-assessment doesnot predict behavioral or intentional persistence once students enroll in a STEM major. Thisstudy introduces professional role confidence—individuals’ confidence in their ability tosuccessfully fulfill the roles, competencies, and identity features of a profession—and arguesthat women’s lack of this confidence, compared to men, reduces their likelihood of remainingin engineering majors and careers. We find that professional role confidence predictsbehavioral and intentional persistence, and that women’s relative lack of this confidencecontributes to their attrition.