This thesis analyses the extent students are stratified into subjects depending on their social background, and the consequences of this in the labour market. I draw on analysis from three longitudinal cohort studies; Next Steps, the 1970 British Cohort Study, and the US study National Longitudinal Survey of Youth (NLSY79). It makes four unique contributions to the literature on educational inequality and subject choice. Firstly, in a joint-authored paper, it offers an overview of the use of intersectionality as a method in quantitative educational research. We make the case that the method should be used more readily in research measuring inequalities in education. Secondly, I empirically test the relationships between students’ characteristics, including their social background, ethnicity and gender, and field of study in higher education. I find that parents’ level of education is more strongly associated with subject choices than either social class or financial resources, suggesting a preferred focus of future research into stratification by subject. I also find that gender and social background interact in determining choice of degree subject. Thirdly, I go on to explore the psychological mechanisms that may drive differences in subject choices. I find differing relationships between students’ personal attitudes and university choices depending on social background. Students from more advantaged backgrounds appeared most likely to choose subjects they enjoyed and thought they were good at. My final chapter compares the relationship between social background and subject choice in the UK and the US. I find that parental education was associated with subject choice for the US cohort, but not the UK cohort. I further test how far these differences explained disparities in earnings in adulthood, and do not find evidence that differences in field of study by background contribute to earnings inequalities in later life.

content: "@mcmasterStratificationFieldStudy" -file:@mcmasterStratificationFieldStudy