Predictive Analysis of Online Course Completion: Key Insights and Practical Implications

Abstract

The rapid expansion of online education has brought significant attention to understanding factors that influence student engagement and course completion. This study aims to predict online course engagement using a dataset from Kaggle, encompassing user demographics, course-specific data, and engagement metrics. Employing a Decision Tree model with 5-fold cross-validation, the research identifies key predictors of course completion, including time spent on the course, the number of videos watched, and quiz scores. The model demonstrates robust performance with accuracy, precision, recall, and F1-scores consistently above 92%, indicating its effectiveness in predicting student outcomes. This predictive capability allows educators and online course providers to identify at-risk students early and implement timely interventions to enhance engagement and completion rates. The study's contributions lie in pinpointing critical engagement metrics and validating the use of Decision Trees in educational data mining. The findings align with existing educational theories that emphasize the importance of active engagement for academic success. Practical implications suggest that online platforms should focus on strategies to increase interaction with course content and provide timely feedback. Future research should explore additional datasets and machine learning models to further refine predictive accuracy and broaden the understanding of factors influencing online learning success. This research provides a foundation for developing more effective online education strategies, ultimately aiming to improve student retention and outcomes

Keywords

• Online Education
• Course Completion
• Student Engagement
• Decision Tree
• Predictive Modelling