I develop statistical and data science methods for measuring social systems, evaluating AI and predictive models, and improving population health decision-making. Recent projects include estimating features of social networks (e.g. the degree of clustering or how central an individual is) using data from standard surveys, inferring a likely cause of death (when deaths happen outside of
hospitals) using reports from surviving caretakers, and quantifying & communicating uncertainty
in predictive models for global health policymakers. More information about OpenVA, our suite of open tools to manage and analyze verbal autopsy surveys, is available here.
I’m the former Editor of the Journal of Computational and Graphical Statistics and a 2019 recipient of the NIH Director’s New Innovator Award. I was elected as a Fellow of the American Statistical Association in 2023. I’m a member of the Committee on Applied and Theoretical Statistics (CATS) of the National Academies of Sciences, Engineering, and Medicine. I was a Visiting Faculty Researcher at Google PAIR from 2019 to 2020.
I will be visiting CEMFI for the 2026-2027 academic year on a María de Maeztu Visiting Professorship.
Adam Visokay and I organized the 2026 Summer Institute in Computational Social Science at the University of Washington, with an emphasis on AI-centered computational social science.
A description of some of our work recently appeared in the Wall Street Journal and it was used for a visual story in the Washington Post.
A link to my slides from the ASA Webinar on the role of statistics in AI is here.
A link to our ENAR short course is here.
Recent/featured work:
Robustly estimating heterogeneity in factorial data using Rashomon Partitions
This paper studies how to estimate heterogeneity in factorial experiments using Rashomon Partitions. Forthcoming in Journal of the Royal Statistical Society: Series B (with Discussion).
The Uneven Rise of AI-Generated Content on the Web
AI-generated content is becoming more prevalent, but spreading unevenly across the web.
When Surveys Become Conversations: Adaptive Matrix Validation for AI-Assisted Interviews
This paper presents a survey design that turns conversational responses into structured variables and uses direct questions to calibrate estimates.
The Epidemiology of Artificial Intelligence
We propose a framework for measuring AI exposure and studying its health effects at the population level. Forthcoming in Nature Health.
A Multimodal Benchmark for Evaluating Cause-of-Death Inference Using Child Health and Mortality Data
This paper presents a benchmark for evaluating cause-of-death inference using child health and mortality data. Accepted to the NeurIPS 2026 Evaluations & Datasets Track.
From Census to Sensors: The Bureaucratic Regime and the Data Revolution
With Marion Fourcade, Adam Visokay, and Sarah Quinn, I examine how data collection is shifting from government surveys to digital traces and private data brokers, and what this means for state governance.
See all technical reports and working papers.
Affiliations
- University of Washington
- Professor, Department of Statistics and Department of Sociology.
- Core faculty member, Center for Statistics and the Social Sciences.
- Senior Data Science Fellow, eScience Institute.
- Research affiliate, Center for Studies in Demography and Ecology.
- Faculty partner, Responsible AI Systems & Experiences (RAISE).