Understanding Self-Directed Learning with Sequential Pattern Mining

We describe a study on the use of an online laboratory for self-directed learning through the construction and simulation of conceptual models of ecological systems. We analyzed the modeling behaviors of 315 learners and 822 instances of learner-generated models using a sequential pattern mining technique. We found three types of learner behaviors: observation, construction, and exploration. We found that while the observation behavior was most common, exploration led to models of higher quality.