On the emergence of typical behaviours in LMS

The emergence of Massive Open Online Courses (MOOCs) has enabled new research to analyze typical behaviours of learners. Inspired by this research, we characterize individual learning behaviours, taking into account specificities of the LMS we use. We then apply clustering techniques to uncover typical behaviours in university courses. In this contribution, we consider a classical face-to-face course on Advanced Web Technologies (AWT) delivered in a Master degree. This course has been offered in the winter semesters 2016/17 and 2017/18. Three typical behaviours appear in each course, reminiscent of those found by other researchers in MOOCs. Aggregating the data week by week, we investigate when these typical behaviours emerge. It turns out that they emerge only shortly before the exam for the two instances of the AWT course. We discuss implications of these findings.