Supporting MOOC Instruction with Social Network Analysis (1401.5175v1)

Abstract: With an expansive and ubiquitously available gold mine of educational data, Massive Open Online courses (MOOCs) have become the an important foci of learning analytics research. In this paper, we investigate potential reasons as to why are these digitalized learning repositories being plagued with huge attrition rates. We analyze an ongoing online course offered in Coursera using a social network perspective, with an objective to identify students who are actively participating in course discussions and those who are potentially at a risk of dropping off. We additionally perform extensive forum analysis to visualize student's posting patterns longitudinally. Our results provide insights that can assist educational designers in establishing a pedagogical basis for decision-making while designing MOOCs. We infer prominent characteristics about the participation patterns of distinct groups of students in the networked learning community, and effectively discover important discussion threads. These methods can, despite the otherwise prohibitive number of students involved, allow an instructor to leverage forum behavior to identify opportunities for support.