Trends and Issues in MOOC Learning Analytics Empirical Research: A Systematic Literature Review (2011–2021)

Abstract

AbstractLearning analytics (LA) is a growing research trend and has recently been used in research and practices in massive open online courses (MOOCs). This systematic review of 166 articles from 2011–2021 synthesizes the trends and critical issues of LA in MOOCs. The eight-step process proposed by Okoli and Schabram was used to guide this systematic review in analyzing publication outlets, research purposes and methods, stakeholders, and researchers’ geographic locations and subjects. The review showed that MOOC LA research is primarily published in top-tier journals, such as the Journal of Learning Analytics, Journal of Computer Assisted Learning, and Computers & Education, as well as conference proceedings. The review also revealed that LA in MOOCs was used more frequently for the purpose of research than for practice (i.e., learning and teaching). Approximately 60% of the reviewed studies adopted learners’ log data and achievement data as the primary data sources. Statistics, machine learning, content analysis, social network analysis, text analysis, and data visualization were the top six specific data analysis techniques used in the MOOC LA studies. Regarding collaboration, more than half of the reviewed studies involved interdisciplinary collaborations, and approximately one-third involved international collaborations. We suggested future studies on MOOC LA interventions to improve learning and teaching practices, and the active interdisciplinary collaboration to increase the rigor of the studies and the dissemination of the knowledge. More detailed discussion and implications for research and practice are presented. This research provides insights on future research and practices regarding LA use in MOOCs.