Learning Analytics Metrics into Online Course’s Critical Success Factors

Abstract

E-learning has gained tremendous popularity. In recent decades, major universities all over the world offer online courses aiming to support student learning performance, yet often exhibit low completion rates. Faced with the challenge of decreasing learners’ dropout rates, e-learning communities are increasingly in need to improve trainings. Machine learning and data analysis have emerged as powerful methods to analyze educational data. They can therefore empower the Technology Enhanced Learning Environments (TELEs). Many research projects have interest in understanding dropout factors related to learners, but few works are committed to refine pedagogical content quality. To address this problem, this paper proposes descriptive statistics analysis to evaluate e-course factors contributing to its success. We take up, at first, the task of analyzing a sample of successful online courses. Then, we manage to identify features that are able to attract a large number of learners, meet their needs and improve their satisfaction. We report findings of an exploratory study that investigates the relationship between course success and the strategy of pedagogical content creation.