Overview
As a comparativist studying the effects of public policy on ethnic identity and political behavior, I focus on a variety of countries and regions including South Asia, the states of the former Soviet Union, China, the United States, and Brazil. Combined with my undergraduate background in biochemistry and graduate training in research methodology, my research portfolio is an eclectic mix of judicial politics, ethnic politics, survey methods, and Bayesian data analysis.
Boundaries of Belonging
Book project · Under development
Boundaries of Belonging: The Structure and Stability of Ethnic Identity
The book asks how ethnic boundaries and support for ethnically-targeted policies shift in response to variation in the way potential beneficiaries are defined. It draws on two original experiments fielded in May 2024 under a Doctoral Dissertation Grant from the Rapoport Family Foundation: a survey experiment among 1,800 White, Black, and Latino self-identifying adults in the United States, and a conjoint experiment among 900 Latino self-identifying adults. I find that individuals are more likely to perceive affirmative action as promoting equality of opportunity when eligibility is framed in more ascriptive terms, or when respondents themselves employ more ascriptive definitions of ethnicity. The conjoint complicates the picture – the two groups do not behave identically, and the difference is not one a single summary measure of ethnic attachment would capture. The next stage of the project expands the project to India, where reservation policies make state classification both materially and politically consequential. The pieces below are selections from the larger project.
Peer-Reviewed Publication
The Dimensions of Ethnicity: Understanding the Cognitive Architecture of Ethnic Group Membership
What makes someone belong to an ethnic group? This article examines how individuals organize and prioritize ethnic attributes frequently used to denote membership and asks when simplified measurement strategies provide a reasonable approximation of ethnic identity in practice. Using original survey data from two online experiments fielded in 2024, I analyze respondents’ evaluations of eight common identity attributes among White, Black, and Latino Americans. Drawing on principal components analysis, multidimensional preference scaling, and circular-linear regression, I show that while some identities are well approximated by a single dominant dimension, others exhibit weaker alignment and greater heterogeneity in how attributes are prioritized. These differences have direct implications for measurement: reliance on any single attribute can meaningfully capture some identities while systematically misrepresenting others. More broadly, the findings underscore the need for group-sensitive measurement strategies when studying ethnic identity and its political and social consequences.
Peer-Reviewed Publication
The Myth of Ethnicity: Understanding the Relationship Between Genotype, Phenotype, and Ethnotype
The explosion of at-home DNA ancestry testing has unsettled how people understand their ethnic heritage and how they choose to identify. Behind these tests lies a critical intuition: if all ethnic identities have some combination of descent-based attributes in common, then they must have some combination of genetic material in common as well. But does genetic ancestry testing actually allow us to determine ethnicity? This article places the literature on ethnic identity into conversation with advances in genetic science and commercial genealogy to explain why the latter is incapable of predicting the former. The subjectivity inherent in selecting a system of categorization and the properties of gene expression and replication illustrate the difficulties involved in linking one concept to the other. Ultimately, this paper suggests that future technological advancements are unlikely to resolve these tensions, and urges the field of political science to be vigilant in educating communities about the differences between genetic ancestry and ethnotype.
Working Paper
Fixed or Fluid? Testing the Effects of Institutional Classification on Ethnic Boundaries
Preparing for submission to Nationalities Papers
How malleable are ethnic group boundaries? Although existing research demonstrates that ethnic identification can respond to political, economic, and institutional incentives, less is known about whether such interventions alter individuals’ underlying beliefs about who counts as a member of an ethnic group. This paper examines whether top-down institutional definitions of ethnicity can shift these boundaries. Drawing on two survey experiments with White, Black, and Latino adults in the United States, I test whether respondents’ understandings of ethnic membership change when affirmative action policies define eligibility using either more easily acquired cultural attributes, such as cuisine and neighborhood, or more ascriptive traits, such as ancestry and appearance. Across measures of both the importance of individual identity attributes and the relative ranking of those attributes, I find substantial variation in how individuals understand ethnic boundaries but relatively little evidence that these understandings change in response to the experimental treatments. A supplementary conjoint experiment among Latino respondents similarly finds little evidence that the treatments alter classification behavior. These results suggest that ethnic boundaries are more resistant to short-term, top-down intervention than accounts emphasizing the malleability of ethnic identity might imply. More broadly, the findings highlight the importance of distinguishing changes in self-identification from changes in the social rules individuals use to determine who belongs to an ethnic group.
Working Paper
Defining the Beneficiary: Group Boundaries and the Politics of Affirmative Action Support
Preparing for submission to Political Behavior
Why do people support or oppose affirmative action? Existing explanations often emphasize racial attitudes, beliefs about merit, or group conflict, but relatively little attention has been paid to how the definition of the policy’s beneficiary group shapes evaluations of the policy itself. This paper examines whether support for affirmative action depends on how ethnic group membership is defined. Drawing on survey experiments with White, Black, and Latino adults in the United States, I compare responses to affirmative action policies that define eligibility using more ascriptive, inherited traits and those that rely on more acquirable cultural and behavioral attributes. I find that the framing of ethnic membership significantly affects White respondents’ evaluations of affirmative action: policies that define eligibility using more ascriptive traits receive substantially more favorable evaluations than those that rely on more easily acquired characteristics. In contrast, Black and Latino respondents’ views remain comparatively stable across treatments. Additional analyses show that stronger racial or ethnic identification is associated with more favorable views of affirmative action across groups, while the relationship between socioeconomic status and policy attitudes varies by racial and ethnic identity. Together, these findings suggest that opposition to affirmative action is shaped not only by beliefs about redistribution or racial inequality, but also by perceptions of who can legitimately claim membership in the groups such policies are intended to benefit. More broadly, the results demonstrate how institutional definitions of ethnic identity can shape political attitudes even when they do not substantially alter individuals’ underlying understandings of ethnic boundaries.
High Courts and Court-Curbing
Book project · In progress
With Brandon Bartels and Eric Kramon, I am building a global account of when public support protects high courts from the leaders who would curb them. The project rests on a Bayesian dynamic latent trait model, borrowing from Christopher Claassen’s work on democratic mood, that pools survey data across time and space in order to produce country-year estimates of public support for high courts in 145 countries from 1981 to 2024. Merged with V-Dem measures of court-curbing, the estimates show that elite attacks raise the likelihood of court-curbing, that public support significantly attenuates that link, and that the attenuation is strongest precisely where theories of judicial independence have long assumed it would be, which is to say in more open societies and during moments of salient threats. The pieces below are selections from the larger project.
R&R · Comparative Political Studies
Can Public Support Insulate High Courts from Court-Curbing? Testing a Comparative Theory with a New Global Dataset
With Brandon Bartels and Eric Kramon
The idea that public support protects judicial independence is central to numerous theories in judicial politics. Yet this claim has been subjected to limited, direct empirical scrutiny in global context. We argue that public accountability and civil society mobilization are mechanisms through which the public can protect the judiciary, and that these mechanisms should matter most when courts face salient court-curbing threats. When elite attacks on courts increase the likelihood of court-curbing, public support significantly attenuates that connection. Our results have implications for literatures on judicial independence, judicial decision-making, and democratic erosion.
R&R · Political Science Research and Methods
Public Support for the Judiciary around the World: A New Cross-National Measure, 1981–2024
With Brandon Bartels and Eric Kramon
Public support features prominently in theory related to judicial independence and judicial power. Yet a lack of globally comparable data, across countries and over time, has limited researchers’ ability to test these ideas. This paper introduces the first global measure of public support for the judiciary that addresses this gap. We produce country-year estimates by gathering courts-related measures of trust, satisfaction, and approval from international surveys and estimating a Bayesian group-level dynamic item-response-theory model, and we provide evidence of construct, convergent, discriminant, and predictive validity using independent V-Dem data. These data create new opportunities to study how the public shapes judicial independence, power, and the rule of law.
Presented · APSA 2025
The Thermostatic Effect of Judicial Independence on Public Support for Courts
Presented at the Annual Meeting of the American Political Science Association, Vancouver, 13 September 2025 with Brandon Bartels and Eric Kramon
Public support is critical to maintaining judicial power and independence in a political system. In comparative perspective, we know little about the drivers of aggregate, country-level public support around the world. We seek to fill this gap with (1) the development of a country-level measure of public support; and (2) an empirical analysis of the drivers of such support. We augment cross-national studies with a comparative case study of Kenya and Poland. Performance-based indicators like judicial corruption and judicial accountability strongly predict public support in expected directions. While de facto judicial independence is positively associated with public support, case studies suggest that independence matters more in combination with higher levels of performance. Partisan political factors are also associated with public support, though less so than performance-based factors. Finally, we show that many of the predictors of specific support also predict indicators of diffuse support, which suggests that both forms of support have similar foundations.
Planned Work
Prospective
I plan to adapt my measurement approach to other relevant contexts, including members of the Han and Zhuang minzu in China and self-identified Tamils and Vanniyars in India. In the case of India, these identities are highly salient and Tamil Nadu maintains reservations for both, which makes the state an unusually direct test of whether the American findings travel. These studies have not yet been fielded.
Prospective
A group-deliberation conjoint
A replication of the conjoint experiment that replaces individual respondents with small discussion groups, in order to test whether classification norms emerge from aggregated individual views or from interaction itself.
Identifiers
ORCID: 0000-0001-9772-6374