This research project is focused on helping teachers become better at helping their students learn. By exploring computer-assisted classroom observation coding, researchers will look for ways to enable teachers to better perceive when things are going well with their students, and when things are not going so well, improving student interaction.
Grant Title: “Towards Computer-Assisted Coding of Classroom Observations: A Computer Vision Approach to Measuring Positive Climate”
Principal Investigator: Jacob Whitehill, Assistant Professor of Computer Science
Funding Amount: $49,850 for One Year
Award Date: April 11, 2018
Sponsor: Spencer Foundation
Classroom observation is one of the chief methods of evaluating the quality of teaching and teacher-student interactions. The Classroom Assessment Scoring System (CLASS), which is an observational tool, has been used in thousands of classrooms nationwide. Manual CLASS coding has some important limitations, however, such as reliability, efficiency, and specificity. This project is focused on using machine learning and computer vision to automatically identify segments of the videos that provide the most evidence of positive climate in the classroom. The envisioned automated system will harness computers' ability to perceive students' and teachers' facial expressions, body language, and physical proximity to others. This project is a collaboration between WPI and the Center for the Advanced Study of Teaching and Learning at the University of Virginia. A PhD student in the computer science department will be working on the research project.