Analysis of Productive Learning Behaviors in a Structured Inquiry Cycle Using Hidden Markov Models

This paper demonstrates the generality of the hidden Markov model approach for exploratory sequence analysis by applying the methodology to study students’ learning behaviors in a new domain, i.e., an asynchronous, online environment that promotes an explicit inquiry cycle while permitting a great deal of learner control. Our analysis demonstrates that the high-performing students have more linear learning behaviors, and that their behaviors remain consistent across different study modules. We also compare our approach to a process mining approach, and suggest how they may complement one another.