Massive online open courses (MOOCs) are revolutionizing online education with their accessibility. MOOCs provide a wealth of information about the way large numbers of learners interact with educational platforms and engage with the courses offered. This extensive data on student usage information is a gold mine for the specific application of data science in the field known as educational data science (EDS). EDS builds on older sub-disciplines of education to solve educational problems with data gathered from educational environments and settings.
In the review article “Educational data science in massive open online courses” recently published in WIREs Data Mining and Knowledge Discovery, Cristóbal Romero and Sebastián Ventura from the University of Cordoba provide a comprehensive review of the existing literature needed to understand the application of EDS in MOOCs. Some researchers identify the next foundational areas for big data and data science in education as technology-mediated psychometrics, self-regulated learning and metacognition, analyses of complex performance and holistic disciplinary practice and practices that are formative and situated for classroom assessment. Other researchers view EDS as four related communities that form this new research and practice paradigm. These communities are Learning Analytics (LA)/Educational Data Mining (EDM), Learner Analytics/Personalization(LA/P)Educational Recommender Systems (ERS), Academic/Institutional Analytics (AA), and Systemic/Instructional Improvement (S/II).
Romero and Ventura discuss the main challenges facing EDS which include analyzing student interactions; predicting students at risk of dropout; grading, assessing, and providing feedback to students; and adapting learning and making recommendations. This review discusses each of these issues alongside the specific EDS techniques used.
Text contributed by the authors.
