Identifying Learning Activity Sequences that Are Associated with High Intention-Fulfillment in MOOCs

Learners join MOOCs (Massive Open Online Courses) with a variety of intentions. The fulfillment of these initial intentions is an important success criterion in self-paced and open courses. Using post course self-reported data enabled us to divide the participants to those who fulfilled the initial intentions (high-IF) and those who did not fulfill their initial intentions (low-IF). We used methods adapted from natural language processing (NLP) to analyze the learning paths of 462 MOOC participants and to identify activities and activity sequences of participants in the two groups. Specifically, we used n-gram analysis to identify learning activity sequences and keyness analysis to identify prominent learning activities. These measures enable us to identify the differences between the two groups. Differences can be seen at the level of single activities, but major differences were found when longer n-grams were used. The high-IF group showed more consistency and less divergent learning behavior. High-IF was associated, among other things, with study patterns of sequentially watching video lectures. Theoretical and practical suggestions are introduced in order to help MOOC developers and participants to fulfill the participants’ learning intentions.