Key Determinants of Academic Performance: A Data-Driven Study Using Linear Regression

Abstract

Academic performance is a multifaceted outcome influenced by a variety of personal, social, and environmental factors. This study investigates the key factors influencing academic performance by applying a data-driven approach using linear regression analysis on a dataset collected from a diverse student population. Several variables—including attendance, study habits, socio-economic status, parental education levels, psychological well-being, and extracurricular involvement—were examined to understand their individual and combined effects on students’ academic outcomes. The findings reveal that consistent attendance, dedicated study time, and higher parental education significantly improve academic performance, while excessive extracurricular activities and poor time management are associated with lower grades. The proposed linear regression model demonstrated strong predictive capability, achieving an R-squared value of 0.72, indicating that 72% of the variation in academic performance can be explained by the selected determinants. This robust model provides not only insights into the relative importance of each factor but also serves as a practical tool for early identification of students at risk of poor performance. Although the study primarily focuses on linear relationships and does not incorporate nonlinear or interaction effects, the results emphasize the need for balanced student engagement and supportive environments. These insights can guide educators and policymakers in designing evidence-based interventions and strategies that target the most influential factors to enhance academic success. Future research is encouraged to extend this work by exploring more complex modeling techniques and longitudinal data to capture evolving patterns in student performance over time.