About

Shiro Kuriwaki is an Assistant Professor of Political Science at Yale University and a Resident Fellow at the Institution of Social and Policy Studies. His primary research focuses on American Politics, particularly on how individual preferences aggregate into electoral and legislative outcomes. He also develops statistical methods to improve the measurement of electoral behavior and public opinion. His work includes studying the structure of voters' party choices across different levels of government using cast vote records, as well as examining representation in Congress. Kuriwaki coauthored the forthcoming book "Representation in America," published by the University of Chicago Press. His research addresses key questions about the nature of representation, electoral behavior, and the dynamics of partisanship in U.S. politics, contributing to a deeper understanding of how legislative decisions reflect public opinion and how voters make choices in both partisan and nonpartisan contests.

Research topics

• Political Science
• Computer Science
• Sociology
• Economics
• Law
• Data Mining
• Political economy
• Geography
• Demography
• Computer Security
• Mathematics
• Public economics
• Medicine
• Business
• Biology
• Epistemology
• Statistics
• Internet privacy
• Public administration
• Environmental health
• Econometrics
• Actuarial science
• Law and economics
• Engineering

Selected publications

• The Role of Confounders and Linearity in Ecological Inference: A Reassessment

  ArXiv.org · 2026-01-12

  articleOpen access1st authorCorresponding

  Estimating conditional means using only the marginal means available from aggregate data is commonly known as the ecological inference problem (EI). We provide a reassessment of EI, including a new formalization of identification conditions and a demonstration of how these conditions fail to hold in common cases. The identification conditions reveal that, similar to causal inference, credible ecological inference requires controlling for confounders. The aggregation process itself creates additional structure to assist in estimation by restricting the conditional expectation function to be linear in the predictor variable. A linear model perspective also clarifies the differences between the EI methods commonly used in the literature, and when they lead to ecological fallacies. We provide an overview of new methodology which builds on both the identification and linearity results to flexibly control for confounders and yield improved ecological inferences. Finally, using datasets for common EI problems in which the ground truth is fortuitously observed, we show that, while covariates can help, all methods are prone to overestimating both racial polarization and nationalized partisan voting.

  PublisherOA PDF

• Candidates in American General Elections

  Open MIND · 2026-02-15 · 1 citations

  dataset

  Comprehensive and standardized constituency-level election results of winning and losing candidates for the offices of President, U.S. House, U.S. Senate, and Governor. Currently from 2006-2024. See the end of the documentation for version history.

  DOI

• The Role of Confounders and Linearity in Ecological Inference: A Reassessment

  arXiv (Cornell University) · 2026-01-12

  preprintOpen access1st authorCorresponding

  Estimating conditional means using only the marginal means available from aggregate data is commonly known as the ecological inference problem (EI). We provide a reassessment of EI, including a new formalization of identification conditions and a demonstration of how these conditions fail to hold in common cases. The identification conditions reveal that, similar to causal inference, credible ecological inference requires controlling for confounders. The aggregation process itself creates additional structure to assist in estimation by restricting the conditional expectation function to be linear in the predictor variable. A linear model perspective also clarifies the differences between the EI methods commonly used in the literature, and when they lead to ecological fallacies. We provide an overview of new methodology which builds on both the identification and linearity results to flexibly control for confounders and yield improved ecological inferences. Finally, using datasets for common EI problems in which the ground truth is fortuitously observed, we show that, while covariates can help, all methods are prone to overestimating both racial polarization and nationalized partisan voting.

  PublisherDOI

• Collective Representation in Congress

  Perspectives on Politics · 2025-08-27 · 3 citations

  articleOpen accessSenior author

  The aspiration of representative democracy is that the legislature will make decisions that reflect what the majority of people want. The US Constitution, however, created a Congress with both majoritarian and counter-majoritarian forces. We study public opinion on 103 important issues on the congressional agenda from 2006 to 2022 using the Cooperative Congressional Election Study. Congress made decisions that aligned with what the majority of people wanted on 55% of these issues. Analysis of each issue further reveals the circumstances under which Congress represents the majority and the many ways that representation fails. The likelihood that the House passes a bill is usually a reflection of public support for that policy, but Senate passage depends on how divided the public is on the issue and whether party control of the two chambers of Congress is divided. Legislative institutions make it difficult to pass popular bills but even more difficult to pass unpopular ones. As a result, most representational failures occur because Congress failed to pass a popular bill, rather than because it passed a bill that the public did not want.

  PublisherOA PDFDOI

• Ticket Splitting in a Nationalized Era

  2025-11-03

  articleOpen access1st authorCorresponding

  Party loyalty in U.S. congressional elections has reached heights unprecedented in decades. Some theories of would predict that deviation from national partisanship is even more rare in low-information, down-ballot offices with a party label. Yet, here I show that ticket splitting in state and local offices is often higher than in Congress. I use cast vote records from voting machines that overcome measurement challenges, and develop a clustering algorithm to summarize such ballot data. For example, about one in three South Carolina Trump voters are part of a bloc whose probability of ticket splitting is 5 percent for Congress, but 23 percent for county council and 48 percent for sheriff. A model with candidate valence differentials can explain these patterns. These results show that even with nationalized politics, some voters cross party lines to vote for the more experienced and higher quality candidate in state and local elections.

  PublisherDOI

• Privacy violations in election results

  Science Advances · 2025-03-12

  articleOpen access1st authorCorresponding

  After an election, should election officials release a copy of each anonymous ballot? Some policy-makers have championed public disclosure to counter distrust, but others worry that it might undermine ballot secrecy. We introduce the term vote revelation to refer to the linkage of a vote on an anonymous ballot to the voter's name in the public voter file and detail how such revelation could theoretically occur. Using the 2020 election in Maricopa County, Arizona, as a case study, we show that the release of individual ballot records would lead to no revelation of any vote choice for 99.83% of voters as compared to 99.95% under Maricopa's current practice of reporting aggregate results by precinct and method of voting. Further, revelation is overwhelmingly concentrated among the few voters who cast provisional ballots or federal-only ballots. We discuss the potential benefits of transparency, compare remedies to reduce or eliminate privacy violations, and highlight the privacy-transparency trade-off inherent in all election reporting.

  PublisherDOI

• Cast vote records: A database of ballots from the 2020 U.S. Election

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access1st authorCorresponding
  PublisherDOI

• Cumulative CES Common Content

  Harvard Dataverse · 2025-10-08 · 51 citations

  datasetOpen access1st authorCorresponding

  The <a href="https://tischcollege.tufts.edu/research-faculty/research-centers/cooperative-election-study">Cooperative Election Study</a> (CES), previously called the Cooperative Congressional Election Study (CCES), is one of the largest political surveys in the United States. This cumulative dataset contains the respondents from the Common Content of the CCES (n = 701,955), combining all available Common Content datasets from 2006 - 2024. It includes select standardized variables including demographics, geography, vote choice, validated vote, representative approval, and views on the economy. See the attached guide for a full list of variables, methodology, and ways to load the data. V11 includes vote validation variables from 2024, and several other changes noted at the end of the Guide.

  PublisherOA PDFDOI

• Privacy Violations in Election Results

  SSRN Electronic Journal · 2025-01-01

  articleOpen access1st authorCorresponding
  PublisherDOI

• How Partisan Are U.S. Local Elections? Evidence from 2020 Cast Vote Records

  American Political Science Review · 2025-09-26

  articleOpen accessCorresponding

  Analyzing nominally partisan contests, previous literature has argued that state and local politics have nationalized. Here we use individual ballots from the 2020 general elections covering over 50 million voters to study the relationship between individual national partisanship and voting in over 5,700 contested down-ballot contests, including nonpartisan races and ballot measures. Voting in partisan contests can be explained by voter’s national partisanship, consistent with existing literature. However, we find that voting for local nonpartisan offices and ballot measures is much less partisan. National partisanship explains more than 80% of the within-contest variation in voting for partisan state and local offices but less than 10% for local nonpartisan contests and local ballot measures. The degree of partisanship in local spending measures varies by the type of service—for example, education, roads, public safety, housing. Finally, we find evidence of structure in the pattern of votes on local spending measures.

  PublisherDOI

Frequent coauthors

• Stephen Ansolabehere

  Harvard University

  21 shared

• Kosuke Imai

  Harvard University

  15 shared

• Andrew B. Hall

  Stanford University

  14 shared

• Cory McCartan

  14 shared

• Connor Huff

  Rice University

  14 shared

• Tyler Simko

  Princeton University

  14 shared

• Christopher T Kenny

  12 shared

• Soichiro Yamauchi

  8 shared

Labs

• American Politics & Public Policy WorkshopPI

Education

• Ph.D., Government

  Harvard University

  2021

• A.B., Woodrow Wilson School of Public Policy

  Princeton University

  2014