A survival model for course-course interactions in a Massive Open Online Course platform

TL;DR

This paper models student dropout in MOOCs using a survival analysis approach, revealing how course interactions and engagement influence success rates across a large dataset.

Contribution

It introduces a reliability-based survival model for course completion, capturing complex patterns missed by traditional regression methods.

Findings

Course interactions are characterized by a single engagement parameter.

User engagement correlates with certificate completion across all courses.

The survival model outperforms conventional regression in predicting success.

Abstract

Massive Open Online Course (MOOC) platforms incorporate large course catalogs from which individual students may register multiple courses. We performed a network-based analysis of student achievement, considering how course-course interactions may positively or negatively affect student success. Our dataset included 378,000 users and 1,000,000 unique registration events in France Universite Numerique (FUN), a national MOOC platform. We adapt reliability theory to model certificate completion rates with a Weibull survival function, following the intuition that students "survive" in a course for a certain time before stochastically dropping out. Course-course interactions are found to be well described by a single parameter for user engagement that can be estimated from a user's registration profile. User engagement, in turn, correlates with certificate rates in all courses regardless of specific content. The reliability approach is shown to capture several certificate rate patterns that are overlooked by conventional regression models. User engagement emerges as a natural metric for tracking student progress across demographics and over time.