Modeling the Self-Regulated Learning Behaviors of Graduate Students in Online Academic Reading and Writing Environments

Abstract

In this chapter, the authors demonstrate two cases of self-regulated learning behavior analysis in academic reading and writing. In the first case, graduate students are participated in a negotiated academic reading assessment, which provided additional opportunities to negotiate with the system for changing their estimated ability of academic reading comprehension after tests. The negotiation actions in the system mainly include checking results, explaining results, taking tests, evaluating ability, proposing questions, answering questions, and making decisions. In the second case, graduate students participated in an online academic writing system, which provided the scaffolding of an academic writing process. The process mainly includes the phases of literature searching, literature reading, paper planning, peer commenting, and paper writing. Behavioral data in both cases was collected for more than 2 months. The authors adopt hidden Markov models to analyze the data in order to capture the temporal and dynamic nature of self-regulated learning. In the models of self-regulated learners, hidden states are assumed as cognitive, metacognitive, or resource management strategies, consisted of actions with various probabilities. By interpreting and comparing the models in both cases, the authors may conclude several patterns of self-regulated academic reading and writing behaviors.