Understanding Student Performance in Foundation Year: Insights from Logistic Regression, Naïve Bayes, and Random Forest Models

Abstract

Foundation programs enhance students’ essential skills, equip them for degree programs, and impact academic performance, retention, and intrinsic motivation. Previous studies focused mostly on demographic factors and statistics. Limited literature has focused on students’ performance in the foundation year. This study uses machine learning techniques to investigate the factors influencing foundation year students’ performance. The study assesses 22 predictor factors, including demographics, secondary school achievement, language proficiency, and university experiences, using Logistic Regression (LR), Naïve Bayes (NB), and Random Forest (RF) algorithms. The study’s findings revealed that gender, school type, secondary school scores, desired college major, and English and math proficiency levels were the significant determinants of students’ performance in their foundation year. Random Forest (RF) showed higher accuracy than both Naïve Bayes (NB) and Logistic Regression (LR). The study indicated that identifying performance factors can improve support services by maximizing learning and results via data-driven methodologies. In conclusion, this study revealed the potential of machine learning in evaluating student performance determinants, supporting targeted interventions, and individualized training through advanced machine learning algorithms and longitudinal data. Moreover, the study helps predict students’ performance in the second semester. Consequently, it projects the enrollment figures for each college along with the anticipated dropout rates.