The aim of the paper sought to identify the factors that influence the dropout risk of university students. We worked with data from a sample of 476 students, including academic, institutional and socio-economic information. Logistic regression was applied to identify the variables with the greatest impact on the dropout risk and thus propose actions that could help mitigate this phenomenon. The model correctly predicted 84% of cases. Among the most important results of this study was that the youngest students taking classes during the day were those who tended to abandon their studies during the first four semesters. Additionally, the effect exerted by age of entry, number of semesters completed, accumulated average, total number of credits approved, financing and obtaining educational aid with the institution was greater when compared with other factors analysed in the present study. Copyright © 2022 Inderscience Enterprises Ltd.