Asynchronous online discussions are commonly used in hybrid and online courses to facilitate social interaction. However, evaluating these discussions can be a daunting task for instructors. This paper proposes a solution using Latent Dirichlet Analysis (LDA) and the instructor’s keywords to automatically extract codes from a small dataset. These codes are then used to build an Epistemic Network Analysis (ENA) model, which is compared to a model created by human coders. The results reveal no statistical difference between the two models. The paper also provides an analysis of these models and discusses the potential use of ENA as a visualization tool to assist instructors in evaluating asynchronous online discussions.
Generating ENA Visualizations for Asynchronous Online Discussion Using a Combination of Automatic Coding and Instructor Input (arXiv:2308.13549v1 [cs.HC])
