Research

Peer-reviewed publications

The Electoral Effects of State-Sponsored Anti-LGBTQ Measures

(co-authored with Violeta Haas, Tarik Abou-Chadi, Heike Klüver and Lukas Stoetzer)

Forthcoming in the Journal of Politics, accepted manuscript available on the UChicago Press website

Do strategies of state-sponsored homophobia translate into electoral gains? While a growing body of literature documents the increasing politicization of LGBTQ- and gender-related issues by illiberal elites, little is known about the electoral effects of these strategies. We address this important question by studying whether anti-LGBTQ mobilization pays off electorally for government parties. Empirically, we study the adoption of anti-LGBTQ resolutions in many Polish municipalities prior to the 2019 parliamentary election. Using a synthetic difference-in-differences design, we find that these resolutions significantly depressed turnout in affected municipalities, with opposition parties showing less mobilization capacity. By contrast, turnout for the incumbent Law and Justice Party increased substantially. Overall, this study's findings are relevant for understanding the electoral consequences of both elite-led mobilization against stigmatized and discriminated groups, and policies of subnational democratic backsliding.

Estimating Treatment Effects on Proportions with Synthetic Controls

(co-authored with Lukas Stoetzer)

Political Analysis, open access on the CUP website.

Synthetic control methods are widely used for causal inference in case studies and panel data settings, often applied to model counterfactuals for proportional outcomes. However, conventional synthetic control methods are designed for univariate outcomes, leading researchers to model counterfactuals for each proportion separately. We make the case for jointly estimating synthetic controls across multiple compositional outcomes. Using the same weights for each proportion establishes a constant control comparison, improving comparability while adhering to compositional constraints on treatment effects. We illustrate the benefits of the method through a simulation and two applications to recent empirical studies. This implementation integrates naturally with a wide range of synthetic control approaches, providing interpretable estimates for compositional panel data common in political science.

RED: Russian Electoral Data

(co-authored with Georgy Tarasenko and Nikita Savin)

Scientific Data, open access on the Nature website.

Electoral data at fine-grained administrative levels are essential for analyzing political competition, voter behavior, and electoral institutions. Yet detailed electoral records have been notably restricted, fragmented and inconsistent in many contexts, including contemporary Russia. We introduce a comprehensive, standardized and longitudinally comparable dataset of Russian federal election results at the precinct level, spanning the period 2000–2024. Integrating digital and archival resources, the data include official records on electorate size, turnout, and voting results for all major parties and presidential candidates. The dataset is harmonized across six presidential and five parliamentary election cycles to accommodate frequent precinct renumbering, administrative boundary changes and inconsistencies in reporting. The paper describes the dataset, data sources, and harmonization process, contextualizing them within Russia's institutional framework and research on authoritarian elections. By offering consistent and harmonized electoral returns, the dataset enables robust empirical analyses of political competition, electoral manipulation, and voter behavior in the period of Russia's transition to a consolidated autocracy.

Working papers

Immigration and Far-Right Government Support

How do citizens evaluate far-right governments in office? I conduct a survey experiment to study how immigration affects the support for a far-right government. The experiment, embedded in a representative survey of 2055 individuals in Italy, exposes different respondent groups to information regarding the government’s pledges and outcomes in immigration and economic growth. Drawing upon the literature, I find evidence that under the far right in office, citizens react negatively to non-attainment of pledged immigration outcomes. No heterogeneous treatment effects w.r.t. political views are evident, as the effect is neither more nor less pronounced among those who consider immigration important, people with left, right or far-right attitudes and among far-right party voters. By examining the dynamics between policy pledges, outcomes, and voter responses, this research sheds light on the evaluation of far-right governments by the electorate.

Presented at Bocconi University, University of Bern and NYU's Rebecca Morton Conference on Experimental Political Science.

Draft available and citeable via my 2025 PhD dissertation.

Fault Lines: Partisanship and Accountability after Turkey's 2023 Earthquake

Natural disasters test government capacity in ways that are difficult to conceal, yet their electoral consequences vary widely. I argue that prior partisan identity is the central moderator: when a disaster implicates government responsibility,

supporters of the incumbent face different cognitive incentives than opposition-aligned voters and engage in motivated reasoning to protect their political identity. I test this argument using the February 2023 Kahramanmaras earthquakes in Turkey, which killed over 50,000 people and were followed three months later by parliamentary and presidential elections. Combining a locality-level electoral panel of roughly 45,000 voting units with monthly survey data from 69,720 respondents, I find that earthquake exposure reduced AKP vote share on average, but this masks a sharp partisan gradient. In the most pro-AKP localities within the disaster zone, party support partially recovered.

Individual-level survey evidence provides the mechanism directly: AKP supporters in exposed districts reported higher political satisfaction after the earthquake while non-AKP voters in the same districts moved in the opposite direction. These results show how partisan identity constrains electoral accountability under electoral authoritarianism.

Presented at Bocconi, IE, LSE, University of Southampton, APSA, EPSA and SPSA conferences.

Detecting AI-Generated Responses in Outsourced Survey Data Collection

(co-authored with Georgy Tarasenko and Andrei Tkachenko)

Researchers who outsource survey fieldwork to third-party vendors cede oversight over the collection process. We present evidence that this practice is vulnerable to AI-generated responses from vendors. Analyzing an online survey of Russian residents in 2026, we detect a sharp discontinuity in data quality as the fieldwork deadline approached. Several independent indicators, spanning open-text characteristics, completion-time distributions, geographic-location clustering, and closed-ended response distributions, shift simultaneously during the affected period. The patterns are consistent with the vendor's batch AI-generated survey responses and are difficult to reconcile with scattered respondent-level use of AI. While the current literature on AI contamination in surveys focuses on respondent-side threats on established Western platforms, our case highlights a risk that data providers within an outsourced data collection pipeline use AI to fabricate responses.

R&R, draft available upon request.

Page updated

Google Sites

Report abuse