CIRSS Seminar: Nigel Bosch

Abstract: Growth mindset interventions (which foster students’ beliefs that their abilities can grow through effort) are more effective in some contexts  than others; however, relatively few variables have been explored that could identify contexts in which growth mindset interventions are most effective. In this talk, I will describe a machine learning approach to predict growth mindset effectiveness in a nationwide experiment in the U.S. with over 10,000 students. I will also discuss how we applied Shapley feature importance analysis to extract interpretable findings and practical implications from "black box" machine learning models. Unique to this study, we found that low intervention compliance (attempting to navigate through the intervention software without completing all steps) resulted in as much as -0.2 difference in predicted intervention effect on GPA.

Bosch is an assistant professor at the iSchool and the Department of Educational Psychology and a faculty affiliate with Illinois Informatics and the National Center for Supercomputing Applications (NCSA). His research revolves around applications of machine learning to problems in education, especially focusing on ways to reduce algorithmic bias and improve transparency in analyses of student-student and student-computer interactions. He holds a PhD in computer science from the University of Notre Dame and was a postdoctoral researcher at NCSA for two years prior to joining the iSchool.