Data Mining and Learning Analytics

CHAPTER 12 UNDERSTANDING COMMUNICATION PATTERNS IN MOOCs: COMBINING DATA MINING AND QUALITATIVE METHODS

Rebecca Eynon, Isis Hjorth, Taha Yasseri, and Nabeel Gillani

Oxford Internet Institute & Department of Education, University of Oxford, Oxford, UK

12.1 INTRODUCTION

Massive open online courses (MOOCs) offer unprecedented opportunities to learn at scale. Within a few years, the phenomenon of crowd‐based learning has gained enormous popularity with millions of learners across the globe participating in courses ranging from popular music to astrophysics. They have captured the imaginations of many, attracting significant media attention—with The New York Times naming 2012 “The Year of the MOOC.” For those engaged in learning analytics and educational data mining, MOOCs have provided an exciting opportunity to develop innovative methodologies that harness big data in education.

At these early stages of exploring how learning unfolds in large‐scale learning environments, it is becoming clear that significant methodological challenges remain. In particular, we argue that qualitative or quantitative approaches are not, on their own, sufficient to extract meaningful insights into how people learn in these settings. We suggest that particularly constructive ways of addressing these challenges include the adoption of pragmatic research paradigms (Tashakkori and Teddlie, 1998), embracing multilevel exploration of data (Wesler et al., 2008), informed by critical engagement with ...