A Regression Analysis for Predicting Student Academic Performance

The aim of the study is to identify the factors that accurately predict academic performance and the contribution that each factor makes to overall academic success. The collected dataset consists of 21 attributes for 97 students in one of the public universities in Malaysia. Several data pre-processing and feature selection tasks have been performed to ensure the quality of the data. A regression model is developed to predict the cumulative grade point average (CGPA) using the selected variables. The result discovered that variables such as gender, absence rate and GPA affect the CGPA of the students. Model evaluation also proves that it can be utilized to predict the CGPA of students. This study is expected to help educational institutions, particularly academic advisors, in identifying students who are at risk of failure. Thus, an early effective program can assist students in improving their academic performance.