An Analysis Framework for Video-Based Teacher Education Curriculum Through the Lens of Topic Modeling

• In Event: Poster Session 1
  In Poster Session: Online Teaching and Learning SIG Poster Session 1

Abstract

In this paper we propose VICTOR, a theoretical framework for analyzing video-based teacher education curricula through Latent Dirichlet Allocation (LDA), a topic model method for identifying underlying relations between various learning components including the reading materials, videos, and discussions. A preliminary analysis reveals the discussion topics are concentrated in our video-driven discussion data, suggesting LDA is a feasible method to be used with the VICTOR framework for analyzing online discussions around videos.

Authors

• Zhou Zhou, Teachers College, Columbia University

• Xiang Liu, Teachers College, Columbia University

• Hui Soo Chae, Teachers College, Columbia University

• Gary J. Natriello, Teachers College, Columbia University