Thanks for stopping by! My name is Madison Coots and I am a second year Ph.D. student in Public Policy at the Harvard Kennedy School of Government and a Stone Ph.D. Scholar in Inequality and Wealth Concentration. I am also an affiliated researcher at the Stanford Computational Policy Lab. My research interests lie in computational approaches to public policy and decision-making, with applications in criminal justice reform and healthcare.
Outside of academics, I am a senior data scientist at STR. Previously, I've worked as a data scientist in academia and in the public sector. I also have experience working in the energy and utilities sector as a part-time data science and risk consultant with Aerospace Technical Services. Before beginning my Ph.D., I studied computer science (M.S.), management science and engineering (B.S.), and English literature (minor) at Stanford.
Public Policy, Ph.D.
Stanford University, Class of 2021
Computer Science, M.S.
Stanford University, Class of 2019
Management Science and Engineering, B.S.
Systems & Technology Research
Senior Data Scientist, 2023-Present
Stanford Computational Policy Lab
Data Scientist, 2020-2022
Aerospace Technical Services
Data and Risk Consultant, 2020-Present
Technical Founder and Head of Data Science, 2017-Present
United States Government
Data Science Graduate Fellow, Summer 2020
Data Science Graduate Fellow, Summer 2019
Data Science Intern, Summer 2018
Data Science Intern, Summer 2017
Systems Engineering Intern, Summer 2017
Stanford MS&E Department
Course Assistant, MS&E 252: Decision Analysis I, Fall 2019
Course Assistant, MS&E 125: Applied Statistics, Winter 2020
Stanford Code In Place
Volunteer Section Leader, Spring 2020
Stanford Athletic Department
Manager, Men's Volleyball, 2015-2020
Stanford Center for Spatial and Textual Analysis
Research Assistant, Mapping Old Regimes of Movement, 2016-2018
Research Assistant, Early Modern Mobility: Breton Corvee, 2018
Some things I've worked on
Reevaluating the Role of Race and Ethnicity in Diabetes Screening
M. Coots, S. Saghafian, D. Kent, and S.GoelWe show that incorporating race and ethnicity into diabetes risk predictions substantially improves statistical accuracy, confirming past analyses, but argue that these statistical benefits lead to surprisingly modest gains in clinical utility.
Automated Court Date Reminders Reduce Warrants for Arrest: Evidence from a Text Messaging Experiment
A. Chohlas-Wood, M. Coots, J. Nudell, J. Nyarko, E. Brunskill, T. Rogers, and S.GoelResults from a large field experiment in partnership with the Santa Clara County Public Defender Office that show that automated text message reminders reduced the number of warrants for arrest issued for missing court by over 20%.
Designing Equitable Algorithms
A. Chohlas-Wood, M. Coots, J. Nyarko, and S. GoelA discussion about the apparent conflicts between formal algorithmic fairness constraints and equitable outcomes, and approaches for practitioners to use in designing more equitable algorithms.
Learning to be Fair: A Consequentialist Approach to Equitable Decision Making
A. Chohlas-Wood, M. Coots, H. Zhu, E. Brunskill, and S. GoelA new theoretical framework for how to efficiently allocate limited resources to people in an equitable way. Working paper.