Multivariate Analysis of Determinants of Student Learning Achievement: Discriminant Analysis and Random Forest Approach

Abstrak

This study aims to identify factors that influence student learning achievement using the Random Forest method and linear discriminant analysis (LDA). The data used include variables such as gender, part-time work, days of absence, extracurricular activities, weekly hours of independent study, and grades from various subjects. The results of the analysis show that weekly hours of independent study are the most dominant factor influencing student academic achievement, followed by involvement in extracurricular activities. In addition, student attendance was also found to be an important factor, with a significant correlation between days of absence and part-time work and gender. These findings provide valuable insights for educators and policy makers to encourage independent learning practices, support student involvement in extracurricular activities, and reduce student absenteeism. Educational strategies that focus on these factors are expected to significantly improve student academic achievement.