Competency 7.4

Competency 7.4: Describe how models might be used in Learning Analytics research, specifically for the problem of assessing some reasons for attrition along the way in MOOCs.

Experience reveals that there are couple of reasons for MOOC learners to attrit, typically:

• the course content is not what the learner expect;
• the learner finds the learning system tiresome;
• the learner is frustrated with the learning system;
• the learner finds the pace of the course too slow and is feeling boring;
• the learner fails to catch up with the progress of the course;

All these will lead to a disengaged behaviour of the learner concerned.

In a MOOC learning environment, teachers and course leaders alike are bound to face some tens or even hundreds of thousand of learners at a time.  Apparently it is extremely challenging for teachers and course leaders to pay close attention to each individual learner to assess if they are  following the course comfortably, or that whether any one of them has problem in keeping pace with progress, or that whether any one of them is feeling bored, frustrated or disengaged with the pace.

This situation can be improved with support from analytic models.  By applying predictive models upon learning data, learner’s behaviour can be illuminated and monitored.  Some of these behaviour are clues of attrition such as when the learner is “gaming” the system, or that the learner submit an assignment earlier than normal, or that the learner falls behind the progress too far, and the like.  These are all clues of a potential attrition.  Predictive models can help single out these clues so that teachers or course leaders are able to “find the needle from the hay” and provide appropriate intervention to the learner in need.

Survival Modeling:

Survival model is a regression model that captures the changes in probability of survival over time. It captures the probability at each time point and it is measured in terms of  hazard ratio which indicates how much more or less likely a student is to drop out. If Hazard ratio>1, the student is significantly more likely to drop out in the next time point.

Sentiment analysis in MOOC forums looked at Expressed sentiment and Exposure to sentiment. The four independent variables Individual Positivity, Individual Negativity, Thread Positivity and Thread Negativity were used to calculate the dependent variable Dropout. The effects were relatively weak and inconsistent across courses.

Some factors that may contribute to student attrition like student's prior motivation, skill set/ knowledge in the area, previous experience in learning MOOCs are difficult to capture. We can link different analysis methods like social network analysis, text mining, predictive modeling and survey data analysis to try to get the complete picture of an individual student for more consistent results.