Research talk "MPowering Teachers: Using Natural Language Processing to Measure and Improve Teaching Effectiveness"

Classroom observations are central to education evaluation and improvement efforts but face challenges of cost, scalability, and measurement error. Advances in natural language processing (NLP) techniques provide a potentially transformative approach for instructional measurement and feedback. This talk will introduce M-Powering Teachers (MPT), an innovative system that combines rich educational theory, cutting-edge NLP techniques, and state-of-the-art machine learning. MPT measures high-leverage teaching practices by analyzing classroom transcripts and then provides teachers with automated feedback to improve their instructional quality and student outcomes. Through several randomized controlled trials, MPT has demonstrated its promise in virtual and in-person learning contexts. However, major technical and practical challenges remain before achieving MPT's full potential. Drawing from the development and research process of MPT, this talk will also discuss the promises and challenges of using artificial intelligence to improve teaching and learning more generally.