Bridging the gender gap in mathematics: Examining STEM enrolment rate by gender in Higher Education, in Italian Provinces

Reducing gender disparities in STEM education is essential both for addressing broader gender inequalities and for advancing the Sustainable Development Goals. This paper investigates the relationship between gender differences in mathematics achievement at school—one of the strongest predictors of gender segregation in higher education—and female enrolment in STEM degree programmes across Italian provinces. Drawing on Bourdieu’s conceptualisation of field and habitus, relative mathematics achievement is interpreted as an indicator of how gendered dispositions and social structures shape educational choices. The analysis relies on an original dataset combining INVALSI standardised mathematics test scores with university enrolment data at the provincial level, covering the period from 2019 to 2023. Exploiting a panel dataset, we estimate a fixed-effects regression model to control for unobserved heterogeneity and address potential endogeneity through an instrumental variable approach. The results show that female STEM enrolment is driven primarily by mathematics achievement relative to male peers rather than by absolute performance: the smaller the gender gap, the higher the female enrolment rate. Moreover, provinces with lower average mathematics scores exhibit higher female STEM participation. This apparent paradox can be explained by the fact that in lower-performing provinces gender gaps in mathematics are minimal, thereby confirming that relative rather than absolute achievement is the key predictor of STEM enrolment. Our findings contribute to the literature by quantifying the central role of relative achievement and by providing novel evidence that reducing gender differences in mathematics at school can foster female participation in STEM.