Mass open on-line courseware (MOOCs): An analysis of attrition from principles of micro

In Massive Open Online Courses (or MOOCs), which are courses given over the web, for free, to anyone who is interested in signing up and has the basic technical capabilities to do so, only a very small fraction of the thousands of students enrolled actually finish the course and earn a certificate of completion. We use data from a 9 week Microeconomics to explore patterns of attrition using a Cox proportional hazard model. We find that students who engaged in the course during the first week, by either completing the first week’s quiz or by submitted a peer assessment exercise, were less likely to drop out of the course. Among students who completed the first week, those who also completed the peer assignment were less likely to drop out of completing future quizzes. This suggests that the pattern of dropping out of the course is not uniform among all enrollees, but rather there are distinct sub groups of participants who reveal their type early on. For the sub sample of enrollees for whom we have demographic data, we find that younger students (ages 18-25) and full time college students are more likely to drop out of the course. Female enrollees were also more likely to drop out. Older students, those over 35, are most likely to stay in the course. Interestingly, among students who took the first quiz, neither female nor younger students, are more likely to drop out from future quiz submissions then others. This suggests that submitting the first quiz is a strong signal of commitment, since the differences between categories found in video retention disappear when looking at quiz retention, at least for those who took the survey.