Thad Dunning
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Thad Dunning has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
Marcelo Nazareno
Universidad Nacional de Córdoba, Argentina
3 shared publications- Susan C. Stokes
Department of Political Science, University of Chicago
3 shared publications Valeria Brusco
Universidad Nacional de Córdoba, Argentina
3 shared publications- Clara Bicalho
Stanford University
2 shared publications - Craig McIntosh
University of California, San Diego
2 shared publications - Gareth Nellis
University of California, San Diego
2 shared publications
Latest Publications
- Open Access
- 31 July 2026
- Published by Cambridge University Press (CUP) in Political Analysis
- p. 1-26
The Power of Prognosis: Improving Covariate Balance Tests with Outcome Information
Scholars frequently use covariate balance tests to test the validity of natural experiments and related designs. Unfortunately, when measured covariates are unrelated to potential outcomes, balance is uninformative about key identification conditions. We show that balance tests can then lead to erroneous conclusions. To build stronger tests, researchers should identify covariates that are jointly predictive of potential outcomes; formally measure and report covariate prognosis; and prioritize the most individually informative variables in tests. Building on prior research on “prognostic scores,” we develop bootstrap balance tests that upweight covariates associated with the outcome. We adapt this approach for regression-discontinuity designs and use simulations to compare weighting methods based on linear regression and more flexible methods, including machine learning. The results show how prognosis weighting can avoid both false negatives and false positives. To illustrate key points, we study empirical examples from a sample of published studies, including an important debate over close elections.
- 31 July 2026
- Published by Cambridge University Press (CUP) in Political Analysis
- p. 1-26
Open AccessThe Power of Prognosis: Improving Covariate Balance Tests with Outcome Information
Scholars frequently use covariate balance tests to test the validity of natural experiments and related designs. Unfortunately, when measured covariates are unrelated to potential outcomes, balance is uninformative about key identification conditions. We show that balance tests can then lead to erroneous conclusions. To build stronger tests, researchers should identify covariates that are jointly predictive of potential outcomes; formally measure and report covariate prognosis; and prioritize the most individually informative variables in tests. Building on prior research on “prognostic scores,” we develop bootstrap balance tests that upweight covariates associated with the outcome. We adapt this approach for regression-discontinuity designs and use simulations to compare weighting methods based on linear regression and more flexible methods, including machine learning. The results show how prognosis weighting can avoid both false negatives and false positives. To illustrate key points, we study empirical examples from a sample of published studies, including an important debate over close elections.
- 5 July 2021
- Published by Project MUSE in World Politics
- Vol. 73 (3) , 590-591
Race, Resources, and Representation: Evidence from Brazilian Politicians—CORRIGENDUM
- 5 July 2021
- Published by Project MUSE in World Politics
- Vol. 73 (3) , 590-591
Race, Resources, and Representation: Evidence from Brazilian Politicians—CORRIGENDUM
- 11 July 2019
- Published by Cambridge University Press (CUP)
- p. 287-312
Dilemmas and Challenges of Citizen Information Campaigns: Lessons from a Failed Experiment in India
Information, Accountability, and Cumulative Learning - edited by Thad Dunning July 2019
- 11 July 2019
- Published by Cambridge University Press (CUP)
- p. 287-312
Dilemmas and Challenges of Citizen Information Campaigns: Lessons from a Failed Experiment in India
Information, Accountability, and Cumulative Learning - edited by Thad Dunning July 2019
- Open Access
- 5 July 2019
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 5 (7) , eaaw2612
Voter information campaigns and political accountability: Cumulative findings from a preregistered meta-analysis of coordinated trials
A preregistered meta-analysis of six field experiments finds no evidence overall that information campaigns shape voter behavior.
- 5 July 2019
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 5 (7) , eaaw2612
Open AccessVoter information campaigns and political accountability: Cumulative findings from a preregistered meta-analysis of coordinated trials
A preregistered meta-analysis of six field experiments finds no evidence overall that information campaigns shape voter behavior.
- Open Access
- 1 December 2016
- Published by SciELO Agencia Nacional de Investigacion y Desarrollo (ANID) in Revista de ciencia política (Santiago)
- Vol. 36 (3) , 773-783
Transparency and Reproducibility in Multi-Method Research
- 1 December 2016
- Published by SciELO Agencia Nacional de Investigacion y Desarrollo (ANID) in Revista de ciencia política (Santiago)
- Vol. 36 (3) , 773-783
Open AccessTransparency and Reproducibility in Multi-Method Research
- 11 May 2016
- Published by Annual Reviews in Annual Review of Political Science
- Vol. 19 (1) , S1-S23
Transparency, Replication, and Cumulative Learning: What Experiments Alone Cannot Achieve
Replication of simple and transparent experiments should promote the cumulation of knowledge. Yet, randomization alone does not guarantee simple analysis, transparent reporting, or third-party replication. This article surveys several challenges to cumulative learning from experiments and discusses emerging research practices—including several kinds of prespecification, two forms of replication, and a new model for coordinated experimental research—that may partially overcome the obstacles. I reflect on both the strengths and limitations of these new approaches to doing social science research.
- 11 May 2016
- Published by Annual Reviews in Annual Review of Political Science
- Vol. 19 (1) , S1-S23
Transparency, Replication, and Cumulative Learning: What Experiments Alone Cannot Achieve
Replication of simple and transparent experiments should promote the cumulation of knowledge. Yet, randomization alone does not guarantee simple analysis, transparent reporting, or third-party replication. This article surveys several challenges to cumulative learning from experiments and discusses emerging research practices—including several kinds of prespecification, two forms of replication, and a new model for coordinated experimental research—that may partially overcome the obstacles. I reflect on both the strengths and limitations of these new approaches to doing social science research.
- 16 September 2013
- Published by Cambridge University Press (CUP)
- p. 261-278
Appendix A: Argentina Brokers' Survey
Brokers, Voters, and Clientelism - September 2013
- 16 September 2013
- Published by Cambridge University Press (CUP)
- p. 261-278
Appendix A: Argentina Brokers' Survey
Brokers, Voters, and Clientelism - September 2013
- 16 September 2013
- Published by Cambridge University Press (CUP)
- p. 152-172
Clientelism and Poverty
Brokers, Voters, and Clientelism - September 2013
- 16 September 2013
- Published by Cambridge University Press (CUP)
- p. 152-172
Clientelism and Poverty
Brokers, Voters, and Clientelism - September 2013
- 16 September 2013
- Published by Cambridge University Press (CUP)
- p. 96-129
Testing the Theory of Broker-Mediated Distribution
Brokers, Voters, and Clientelism - September 2013
- 16 September 2013
- Published by Cambridge University Press (CUP)
- p. 96-129
Testing the Theory of Broker-Mediated Distribution
Brokers, Voters, and Clientelism - September 2013