By crunching 130 million mouse-clicks, two CSAIL researchers have developed a machine-learning model that can predict with surprising accuracy whether or not a MOOC student will drop out of a given course.
Kalyan Veermachaneni and Una-May O’Reilly used machine-learning techniques to analyze which factors most directly correlate to drop-out rates, including their average number of weekly submissions; their level of procrastination (the time between when a student starts to work on the problems and the actual deadline); and their weekly lab grade.
They divided students into four different cohorts:
1. learners who participated in forums (discussion generators)
2. learners who edited wikis (content generators)
3. learners who did both (fully collaborative), and
4. learners who did neither (passive collaborators)
To learn more, read their recent post on the Office of Digital Learning website: http://odl.mit.edu/who-is-likely-to-drop-out-and-why/