Evaluating the Impact of Demographic and Technical Background Factors on Academic Performance of First-Year Computer Engineering Students

Traditional approaches to predicting student academic performance have largely focused on academic indicators such as high school grades and entrance examination scores. However, these metrics often fail to account for socio-demographic and experiential factors that may significantly influence a student success in higher education. This study investigates the predictive value of non-academic attributes—including gender, place of origin, prior exposure to computing, participation in developer communities, and extracurricular technical engagement—on the academic performance of first-year Computer Engineering students. Using a dataset comprising 25 diverse features collected through structured surveys, we apply a combination of statistical methods and supervised machine learning models to evaluate the relative importance of these predictors. The results are compared against findings from prior work focused exclusively on academic variables such as subject-specific grades and cumulative average. Our analysis reveals that certain experiential and behavioural factors, particularly technical project involvement and developer community participation, demonstrate strong predictive power—comparable to or exceeding traditional academic measures. These findings support the integration of non-academic dimensions into predictive models of student success and highlight the value of a holistic assessment framework for educational policy, admissions, and academic support planning.