log in  |  register  |  feedback?  |  help  |  web accessibility
Using Natural Language Processing to Measure and Improve Teaching Practices at Scale
Wednesday, February 16, 2022, 11:00 am-12:00 pm Calendar
  • You are subscribed to this talk through .
  • You are watching this talk through .
  • You are subscribed to this talk. (unsubscribe, watch)
  • You are watching this talk. (unwatch, subscribe)
  • You are not subscribed to this talk. (watch, subscribe)

Valid and reliable measurement of teaching quality is key to effective education policy making. However, conventional use of classroom observations that rely on human raters is resource intensive and subject to a range of measurement issues. An emerging literature starts to leverage natural language processing techniques to measure teaching practices using classroom transcript data. In this talk, I will introduce a fully automated approach my collaborators and I have developed to measure teachers’ uptake of student ideas, a high-leverage teaching practice that supports dialogic instruction and makes students feel heard.  I will then present results from a randomized controlled trial we have designed to evaluate how the use of this automated measure on uptake can improve instructors’ teaching practices in a large-scale online computer science class.

Zoom: https://umd.zoom.us/j/98806584197?pwd=SXBWOHE1cU9adFFKUmN2UVlwUEJXdz09
(passcode if needed: clip)


Jing Liu is an Assistant Professor in Education Policy at the University of Maryland College Park. Named as a National Academy of Education Sciences/Spencer Dissertation Fellow, he earned his Ph.D. in Economics of Education from Stanford University in 2018. Before he joined UMD, he spent two years as a Postdoctoral Research Associate at Brown University’s Annenberg Institute. Dr. Liu's research uses rigorous quantitative evidence to evaluate and inform education policies at the national, state, and local levels, with the goal of improving learning opportunities for historically marginalized students in urban areas. His work broadly engages with critical policy issues including student absenteeism, exclusionary discipline, educator’s labor market, school reform, and higher education, with a special interest in the intersection of data science and education policy. In his most recent work, he is working with a group of scholars from computer science, linguistics, and curriculum and instruction, to create an automated system that can provide teachers immediate feedback on their teaching practices using classroom transcript data and natural language processing techniques. His work has appeared in peer-reviewed journals such as the Journal of Public Economics, Journal of Human Resources, Journal of Policy Analysis and Management,  and Educational Evaluation and Policy Analysis.

This talk is organized by Wei Ai