Luke Miratrix

During his fellowship year, Luke Miratrix plans to write a book on randomized trials in the social sciences, with a focus on multisite and cluster randomized trials. After laying out an overall framework for understanding these from a statistical point of view, the book will cover tradeoffs between analytic choices, evaluating treatment effect heterogeneity, handling noncompliance, and using machine learning to augment traditional analyses. Miratrix’s goal is to connect a body of work that currently exists as scattered individual (and often overly technical) contributions into an accessible, unified framework. CASBS offers an ideal setting for this project: writing a book aimed at social science researchers is best done in conversation with them. He also hopes to get distracted by the interesting problems of his fellows and learn even more about how statistics can help us learn about the world.

Miratrix is a professor at the Harvard Graduate School of Education and affiliate faculty of the department of statistics. His research focuses on improving quantitative methods in the social sciences, with particular emphasis on causal inference, randomized trials, and the use of text as data. Miratrix develops statistical tools for large-scale program evaluations, with work spanning education, criminal justice, elections, and public policy.

For more information, please visit: https://cares.gse.harvard.edu/about-luke-miratrix