Gender equality in education remains a global policy priority, as gendered patterns of participation persist despite overall improvements in educational attainment (UNESCO, 2023).

In Italy, females are more likely to enrol in university (Tocchioni, 2025) and have a lower propensity to drop out (Atzeni et al., 2022) than males.

However, marked horizontal segregation across fields of study persists: females remain underrepresented in most STEM fields and overrepresented in the humanities, nursing, and social sciences (Priulla et al., 2021).

The choice of the field of study represents a crucial life-course moment with long consequences. Indeed, gender inequalities in career prospects are largely due to the underrepresentation of women in engineering and computing and to their overrepresentation in the humanities and the social sciences (Barone & Assirelli, 2020).

Within this framework, numerous studies have investigated gender differences in enrolment, focusing often on STEM degrees and examining the role of prior educational pathways, concentrating on a specific upper-secondary school track (Priulla et al., 2024; Giannantoni & Falzetti,2025), as well as other individual and contextual factors shaping students’ entry choices. Some contributions rely on administrative data for the Italian university context (Contini et al., 2018; De Angelis et al., 2016), while others rely on surveys or large-scale experiments such as information campaigns that can improve both university enrolment and the quality of students’ educational choices, as demonstrated by Ballarino et al. (2022).

The aim of this contribution is to analyse how gender interacts with other potentially relevant factors—such as socio-economic background, migration background, and local labour market characteristics—in shaping students’ choices of university fields of study and how these factors operate within different tracks of upper secondary education (academic, technical, and vocational).

By integrating administrative data on university enrolment with INVALSI assessment data, the analysis also examines the potential interaction between the choice of university field and students’ competencies in Italian language and mathematics. This allows us to assess, for instance, whether within each gender the propensity to enrol in scientific or humanities-related university programmes is associated with differential skill profiles, or whether gender stereotypes persist in directing students towards specific fields of study regardless of their acquired competencies during secondary education.

A key aspect of our approach lies in extending the analysis of gender-related issues beyond the well-documented under-representation of women in STEM disciplines. We symmetrically address patterns of both over- and under-representation of male and female students across a broad range of university fields. This is made possible by a more fine-grained classification of fields of study, which distinguishes, among other dimensions, between technological STEM disciplines and more abstract or bio-mathematical STEM fields, as well as between different subdomains within the humanities (e.g. literary versus socio-economic areas).

This study highlights how gender-based selection mechanisms already operate at the point of university entry, shaping differentiated academic trajectories within the Italian higher education system.

The analysis is conducted on a cohort of students graduated in the 2018-2019 school year and enrolled in university in the academic year 2019-2020. The focus is on students enrolled in Italian Bachelor’s degrees, as recorded in the University Register with the exclusion of students enrolled abroad or in online universities. The dataset is constructed through the integration of multiple data sources: Italian Ministry of Education, National University Register and the National Institute for the Evaluation of the Education System (INVALSI). The data include detailed information on individual characteristics and prior educational pathways, such as parental education, school regularity and the Socio-Economic and Cultural Status of Families (ESCS) indicator (Campodifiori et al., 2010), upper-secondary school track as well as INVALSI scores in grade 13. At the university level, degree programmes are grouped into the following disciplinary areas: Arts and Literature, Economics, Law and Social Sciences, Health Studies, and STEM. For this analysis STEM area is further disaggregated into two distinct categories, core (e.g., bio-mathematical sciences) and technological STEM, based on the specific degree programmes included in each field. This distinction is motivated by the markedly different composition of these subfields. The analysis relies on a sequence of multinomial logistic regression models of increasing complexity. The models estimate the probability of enrolling in each disciplinary area as a function of gender and pre-university characteristics. The set of control variables includes geographical macro-area, origin, upper-secondary school track and student regularity, INVALSI scores in Italian and Mathematics and additional contextual controls. This sequential modelling strategy allows us to control the conceptual and interpretative complexity arising from the simultaneous consideration of a wide range of individual and contextual determinants of students’ educational choices and outcomes. The large-scale nature of the dataset—covering a cohort of several tens of thousands of students enrolled in university—makes it possible to estimate not only main effects but also interactions between key covariates, including higher-order interactions. In particular, this approach enables us to assess whether the relationship between gender and university field choice varies systematically according to the type of secondary school attended. Model selection is supported by AIC and BIC comparisons, which indicate a better fit for the more complex specification with interaction terms. Additionally, predicted probabilities and conditional marginal effects are computed.

The distribution of university field choices within the cohort summarises enrolment patterns across disciplines. Socio-economic fields are the most frequently chosen (30.5%), followed by technological STEM (24.3%), humanities (19.0%), core STEM (14.2%), and health-related fields (11.9%). Consistent with existing literature, from model results men are more likely to enrol in STEM, while women are more concentrated in humanities-related fields. However, this pattern masks substantial heterogeneity within STEM: gender differences are largely confined to technological disciplines and do not extend to core scientific or bio-medical fields. These patterns underscore the importance of distinguishing between core and technological STEM. Predicted probabilities indicate that upper secondary school type plays a key role in structuring gendered enrolment patterns. Among scientific high school graduates, women are more likely to enrol in core STEM, while men are predominantly oriented towards technological STEM; in technical tracks, the gender gap in technological STEM is particularly pronounced in favour of men. By contrast, graduates from classical and other academic tracks mainly enrol in humanities and socio-economic fields, with more balanced gender distributions. Focusing on student competencies, marginal effects based on INVALSI mathematics scores show a positive gradient in enrolment in technological STEM fields, particularly among male students. By contrast, mathematics performance is only weakly associated with enrolment in core STEM fields and corresponds to lower probabilities of enrolling in literary programmes among graduates from classical upper secondary schools. Beyond additional analyses of interactions between gender, school type, socio-economic background, and student competencies—which are omitted here for reasons of space—a promising avenue for future research consists in extending the analysis to students who do not enrol in university. This would allow for a more comprehensive assessment of the mechanisms shaping post-secondary educational choices, while addressing potential selection issues inherent in analyses restricted to university entrants.

ALMALAUREA (2024). Rapporto 2024 sul profilo e sulla condizione occupazionale dei laureati, https://www.almalaurea.it ANVUR. (2023). Rapporto biennale sullo stato del sistema universitario e della ricerca. https://www.anvur.it/it/dati-e-pubblicazioni/rapporto-biennale. Atzeni, G., et al. (2022). Drop-Out Decisions in a Cohort of Italian Universities. In: Teaching, Research and Academic Careers: An Analysis of the Interrelations and Impacts. Cham: Springer International Publishing, pp. 71-103. Ballarino, G., Filippin, A., Abbiati, G., Argentin, G., Barone, C., & Schizzerotto, A. (2022). The effects of an information campaign beyond university enrolment: A large-scale field experiment on the choices of high school students. Economics of Education Review, 91, 102308. Barone, C., & Assirelli, G. (2020). Gender segregation in higher education: an empirical test of seven explanations. Higher Education, 79(1), 55–78. Campodifiori, E., Figura, E., Papini, M., & Ricci, R. (2010). Un indicatore di status socioeconomico-culturale degli allievi della quinta primaria in Italia (Working Paper No. 2). INVALSI. http://www.invalsi.it/download/wp/wp02_Ricci.pdf. Contini, D., Cugnata, F., & Scagni, A. (2018). Social selection in higher education. Enrolment, dropout and timely degree attainment in Italy. Higher Education, 75(5), 785-808. Correll, S.J. (2001). Gender and the career choice process: The role of biased self-assessments. American Journal of Sociology, 106(6), 1691–1730. De Angelis, I., Mariani, V., Modena, F., & Montanaro, P. (2016). Immatricolazioni, percorsi accademici e mobilità degli studenti italiani. Questioni di economia e finanza (Occasional Papers) n. 354 della Banca d’Italia. Giannantoni, P., & Falzetti, P. (2025). EQUAL OPPORTUNITIES TO “COUNT”? A LONGITUDINAL POPULATION STUDY ABOUT CHOICE OF STEM AT UNIVERSITY BY GENDER. Italian Journal of Applied Statistics. Priulla,A., Albano, A., D'Angelo, N., & Attanasio, M. (2024). A machine learning approach to predict university enrolment choices through students' high school background in Italy. ArXiv abs/2403.13819. Priulla, A., D’Angelo, N., & Attanasio, M. (2021). An analysis of Italian university students’ performance through segmented regression models: gender differences in STEM courses. Genus, 77(1), 11. Tocchioni, V., Milone, S., & Lombardi, G. (2025). Gender disparities in school-to-university transition in Italy: the role played by the socio-economic condition and the type of high school. RIVISTA ITALIANA DI ECONOMIA, DEMOGRAFIA E STATISTICA, 67-78. UNESCO. (2023). UNESCO in action for gender equality: 2022–2023. https://unesdoc.unesco.org/ark:/48223/pf0000387300.