Bio

Brandon M. Stewart is Professor of Sociology at Princeton University, where he is also affiliated with the Politics Department, the Office of Population Research, the Princeton Institute for Computational Science and Engineering, the Center for Information Technology Policy, and the Center for Digital Humanities. He develops quantitative statistical methods for applications across computational social science.

His work spans computer-assisted text analysis, causal inference, machine learning, and the connections among them. Across these projects, a recurring theme is how social scientists can use complex data sources while keeping measurement, research design, and substantive interpretation tightly linked.

With Justin Grimmer and Molly Roberts, he wrote Text as Data: A New Framework for Machine Learning and the Social Sciences with Princeton University Press. With Molly Roberts and Dustin Tingley, he developed the Structural Topic Model and accompanying software. With collaborators, he has also worked on design-based supervised learning, estimands in quantitative social science, text matching, contextual word embeddings, and measurement problems in computational social science.

He teaches undergraduate and graduate statistics as well as courses on text analysis. Course materials, preprints, and replication materials are linked from this site and from Google Scholar and Dataverse where available.