Agentic and Emotional Engagement in Science Discourse: A Natural Language Processing Illustration

• In Event: The Role of Emotion and Gesture in Science Learning Environments

Abstract

This study explores the application of natural language processing (NLP) for providing new insights into science classroom discourse processes. This study presents an illustration of NLP to science classroom textual data to 1) examine patterns in the quantity and nature of spoken words by teacher and students, 2) co-occurrence of discourse spaces (academic, hybrid, everyday) and students’ agentic and emotional engagement, and 3) the sentiments associated with different discourse opportunities and questioning as a high leverage practice for emotional engagement in science discourse. Based on the NLP findings, we conducted a contextualized interpretation of discourse patterns that will inform educators’ effort to expand science discourse opportunities in urban classrooms for all students.

Authors

• Christine Lee Bae, Virginia Commonwealth University
  Presenting Author

• Morgan DeBusk-Lane, Virginia Commonwealth University
  Non-Presenting Author

• Kimberly Williamson, Cornell University
  Non-Presenting Author