Profiling MOOC Course Returners

Abstract

Massive Open Online Courses represent a fertile ground for examining student behavior. However, due to their openness MOOC attract a diverse body of students, for the most part, unknown to the course instructors. However, a certain number of students enroll in the same course multiple times, and there are records of their previous learning activities which might provide some useful information to course organizers before the start of the course. In this study, we examined how student behavior changes between subsequent course offerings. We identified profiles of returning students and also interesting changes in their behavior between two enrollments to the same course. Results and their implications are further discussed.

The Promise of MOOCs Revisited? Demographics of Learners Preparing for University

“…They determine four prototypical user types: completing, auditing, disengaging, and sampling, defined by steep drop-out points and deeply unequal levels of participation. The methods employed by Kizilcec et al (2013) have been leveraged across several different MOOC environments (e.g., Arora et al, 2017;Kovanović et al, 2016;. With few exceptions (e.g., , these papers consistently report four types of engagement patterns; this is further confirmed in the literature beyond research using strictly cluster analysis (Li & Baker, 2018;Anderson et al, 2014;Ramesh et al, 2014).…”

“…Hierarchical clustering divides or agglomerates data into groups as small as one to as large as the entire data set (Kaufman & Rousseeuw, 1990). In the learning analytics literature, partitioning methods are dominant (Khalil & Ebner, 2017;Arora et al, 2017;Kovanović et al, 2016;Kizilcec et al, 2013), though hierarchical methods have been used (Chen et al, 2015), as well as other techniques that form clusters (Anderson et al, 2014;Ramesh et al, 2014). Our analysis explores the more common approach of partitioning methods.…”

Participating by activity or by week in MOOCs

“…acquisition, participation and selfdirection), which were characterized by the pedagogical philosophies enacted within the MOOCs they studied. Explorations of these user engagement personae have even gone so far as to characterize how students will morph from one personae to another between different offerings of the same MOOC (Kovanovi c et al, 2016) or how they will cheat the system by creating multiple profiles to ensure certification (Ruiperez-Valiente et al, 2016). These types of analyses align with claims that there are a small number of ways in which students interact with MOOC resources (Kizilcec et al, 2013).…”

“…Overall, these explorations of student use of MOOC resources and features fail to provide a nuanced understanding of how students interact with MOOC resources. They instead identify stereotypical patterns in student learning activities that are formulated as personae describing a user type (Anderson et al, 2014;Corrin et al, 2017;Gasevic et al, 2014;Guo and Reinecke, 2014;Kovanovi c et al, 2016;Liu et al, 2016). For example, Anderson et al (2014) identified five types of users based on student interactions with video lectures and graded ILS 119,9/10 assignments.…”

Temporal Analysis in Massive Open Online Courses – Towards Identifying at-Risk Students Through Analyzing Demographical Changes

“…Other studies were focused on different approaches to identifying subgroups, but most of them did not consider behavioral changes over time from the clustering [18,22,26,28]. It is important to explore behavior patterns of subgroups of the students on a specific time scale, since the characteristics of each subgroup, and the proportion of its total interaction, vary along a MOOC progresses.…”