Modeling Gender Inequality in Argentine ICT Degrees through Multiple Regression

Abstract

This paper presents a quantitative analysis of the gender gap in Computer Science, Information Systems, and Informatics (CSI) degree programs at Argentine universities during the 2010–2015 period. Using a Data Science approach, supervised machine learning was applied—specifically, a multiple linear regression model—to identify the main factors influencing student graduation. The results show that being a female student significantly decreases the probability of graduation, while studying at private universities or in the province of Buenos Aires increases this probability. This study provides relevant empirical evidence for the design of public policies aimed at reducing gender inequality in STEM fields. As future work, the analysis will be extended by incorporating more advanced machine learning techniques and including other STEAM-related degrees.