Lily An

I am an Assistant Professor of Quantitative Methodology at Georgia State University in the Educational Policy Studies department in the College of Education & Human Development. I graduated in 2025 with my Ph.D. from Harvard University in the Education Policy and Program Evaluation concentration. Previously, I was a Research Analyst in the Department of Education at Brown University, working on a Research-Practice Partnership with the Massachusetts Department of Elementary and Secondary Education. 

My research interests include educational measurement, program evaluation, and accountability policies. I have experience working with state leaders on their accountability systems, evaluating educational policies' effects on student outcomes using quasi-experimental methods, conducting meta-analyses, and using psychometric tools to evaluate standardized assessments.

My job market paper, Using Gaussian Process Regression in Two-Dimensional Regression Discontinuity Designs, introduces this nonparametric response surface estimation method to educational contexts. Gaussian process regression mitigates concerns of bias and information loss from existing analytic methods, making it a particularly promising tool for evaluations of education policies that often require meeting multiple conditions (such as both English and math test score cutoffs) to receive treatment. Gaussian process regression also allows one to target an estimand of an average treatment effect along the treatment boundary as well as understand treatment effect heterogeneity along the boundary. 

I hold an Ed.M. in Education Policy and Management from the Harvard Graduate School of Education and a B.A. in Statistics from Williams College. I have been a Fellow with the Center for Education Policy Research's Partnering in Education Research fellowship, an Equity and Inclusion Fellow at the Harvard Graduate School of Education, a Summer Institute for Computational Social Science participant, and a Just Education Policy Institute for Developing Scholars participant.