Gender differences in Italian STEM degree courses: a discrete-time competing-risks model

In this paper, we highlight possible gender differences in performance of students attending Italian university STEM degree courses, controlling for sociodemographic and territorial variables.We used the micro-data from the Italian Ministry of Education, Universities and Research database of the 2011-12 freshmen cohort enrolled at an Italian bachelor university degree course. The analysis was dealt with by fitting a discrete-time competing-risks model. Bachelor degree graduation, dropout, and change to non-STEM courses were considered as competing events. The aim is to highlight gender gaps in the student performance, and the STEM courses where this gap is more critical.