Statistical Analysis of Students' Productivity

Student productivity is shaped by a complex interaction of psychological, academic, environmental, physical, and social factors. This study investigates the key determinants affecting productivity among 18–21-year-old undergraduates, using a dataset of 843 responses collected via Google Forms on a 5-point Likert scale. Multiple Linear Regression (R² = 0.2395, F = 88.08, p < 0.001) identified rapid heartbeat, anxiety, and academic conflict as primary stress drivers. Machine learning classifiers — Random Forest, XGBoost, Logistic Regression, Association Rule Mining, and K-Means Clustering — were applied to classify students by productivity level. XGBoost achieved the highest accuracy (90.5%, AUC = 0.976), with workload overwhelm, low academic confidence, and poor concentration emerging as the three strongest predictors. Approximately 29% of students were flagged as at-risk. Findings support the development of data-driven, early-warning academic support frameworks. The findings highlight the multidimensional nature of personal productivity and identify the importance of emotional well-being and academic environment management. This study describe the usefulness of statistical and machine learning approaches in understanding and predicting productivity patterns among students.