Who Completes, Who Gets a Certificate? Digital Traces and Learning Outcomes in MOOCs

ORMAN R., ÇAKIR H., ÇAĞILTAY N.

International Journal of Multidisciplinary Studies and Innovative Technologies, vol.9, no.2, pp.201-209, 2025 (Peer-Reviewed Journal)

Abstract

This study aims to investigate the factors affecting course completion and certification among participants in the context of Massive Open Online Courses (MOOCs). Logistic regression analyses were conducted based on participants' demographic characteristics and learning interactions in the online course "Elements of Structures" offered by the Massachusetts Institute of Technology (MIT). The analyses revealed that variables such as graduate education level, age groups (especially ages 45 and above), number of days of interaction, number of video plays, number of chapters completed, and writing activity had significant effects on the likelihood of both course completion and certification. The contribution of written interaction and content completion to success was found to be particularly significant. Model performance was evaluated using ROC curves and AUC (Area Under the Curve). The AUC values of 0.80 and 0.92, respectively, demonstrated high classification accuracy. The findings reveal that logistic regression analysis is an effective tool in developing predictive models for predicting user success on MOOC platforms and offer strategic recommendations for personalized, interaction-oriented design of learning environments.