Abstract
In a subject where individual tasks are intermingled with collaborative tasks, complex learning dynamics emerge. When these tasks are performed using a computer tool, the student's interactions leave some evidence traces in the tool. Process mining techniques allow us to analyze these traces and discover and understand these learning dynamics. In this work, we applied process mining to the traces of a computer tool used to implement a studentgenerated questions activity, in which students propose questions and improve them through collaboration with other students. The study has provided insight into the work habits of the students. It has also shown that students with greater involvement in the activity obtain better academic results. Moreover, some student deviations from the intended process have been detected and an unexpected tendency towards non-collaborative behaviors has also been identified, resulting in an unforeseen aspect of intergroup competition within the activity.

This publication has 11 references indexed in Scilit: