Research

My research sits at the intersection of political psychology and institutional analysis: how emotions move voters, how identities form against outgroups, and how federal structures shape who gets blamed. Project entries below; the data and code behind them are open wherever the data sources allow.

Identity · Cohesion 2026–2028

Not Like Them

Why do so many political conflicts now revolve less around what people are than around what they refuse to be? Why does "not Bavarian," "not woke," "not a boomer" mobilise faster than any positive identity? This project develops the concept of negational group identities (NGIs): identities anchored in the rejection of an outgroup rather than in attachment to an ingroup. NGIs work differently from classical affirmative identities. Affirmative identities build cohesion through shared traits; NGIs achieve unity through shared rejection. That makes them fast to mobilise, but structurally corrosive for pluralist democracy. The project combines digital laboratory experiments with two national surveys in Germany. It asks who is drawn to rejection-based identities, how emotions activate them, and how different NGI types (partisan, issue-based, spatial, generational) shape trust between citizens and in democratic institutions. All data will be openly released via GESIS and the Harvard Dataverse.

With Sabrina Mayer (University of Bamberg)
Funding: Fritz Thyssen Foundation
Data
Infrastructure Published 2026

StateParl

Parliaments are central arenas of political debate where political actors publicly justify their actions, and parliamentary speeches offer in-depth insights into their interests and strategies. Yet few resources exist for the analysis of subnational parliaments. StateParl is a text corpus containing machine-readable plenary protocols from all 16 German Länder parliaments, 2000 to 2023: more than 14 million speech paragraphs with rich metadata, linked to the StatePol database of Länder politicians. Its applications extend far beyond comparative subnational politics and parliamentary research to party research, German politics, and text-as-data analyses more generally. The data are freely available through GESIS; the build pipeline is open source.

With Eric Beltermann, Sabine Kropp & Antonios Souris
Papers
  • Decoding German politics with StateParl: A text corpus of plenary protocols in the 16 German Länder parliaments Politische Vierteljahresschrift, 2026 DOI Code
Data
Measurement Working papers

FEnDi: Measuring Federalism in Parliamentary Debate

How often, and how positively, do state parliamentarians talk about the federal system? FEnDi is a dictionary of federal-entity references, validated against hand-coded debate data, that turns the StateParl corpus into a panel of salience and sentiment measures. Two papers build on it: a methods note on reusing qualitative content analysis as dictionary-based measurement, with FEnDi as the empirical illustration, and a substantive party-politics paper on salience-sentiment divergence and the cross-level coordination tightrope parties walk in Germany's multilevel system.

With Antonios Souris, Jan Schwalbach & Sabine Kropp
Data
Classification In progress

Blame in Parliamentary Debate

A multi-method text classification system for detecting blame attribution in German parliamentary debates. The project analyzes 14 million paragraphs from the 16 German state parliaments, with classifiers trained on thousands of hand-coded debate segments, to identify where speakers attribute blame, pass responsibility, or claim credit across federal levels: how blame-shifting unfolds in crises, and how government and opposition status shapes it.

With Antonios Souris
Data
Polarization Working paper

Regional Affective Polarization

The existing literature on affective polarization is, in practice, affective party polarization. This project introduces the concept of Regional Affective Polarization (RAP): the affective distance between territorially defined ingroups and outgroups. Using original survey data from Germany, we show that RAP is empirically separable from partisan affective polarization, that perceptions of regional inequality are its strongest predictor (citizens who feel their state is "left behind" exhibit substantially higher RAP), and that more polarized respondents express weaker support for federalism and systematically discount federal policy arrangements. The survey data also fed an interactive Tagesspiegel feature on which Bundesländer like each other.

With Lena Masch, Johanna Schnabel & Antonios Souris
Papers
  • Regional affective polarization and voting behavior Working paper, 2024
Federalism Working paper

Multilevel Responsibility

Motivated reasoning in political attribution has boundary conditions set by institutional plausibility. Partisans can shift blame for a bad economy upward to the national level but not downward to the states, because national economic primacy is the default frame. Federal systems do not just confuse voters; they provide asymmetric opportunities for partisan blame deflection. Analyzing 22 German state election surveys, the paper shows consistent upward blame-shifting and no detectable downward shifting. The demand-side complement to Party Strategies in Cooperative Federalism.

Parliaments In progress

Maiden Speeches

How long does it take new MPs to give their first substantive speech, and how does this differ by gender? Drawing on the GermaParl corpus of Bundestag debates, this project treats time to first speech as strategic adaptation at the participation margin: the sharp test is whether electoral safety changes the gender gap in who takes the floor, and when.

With Sabrina Mayer & Alexander Herzog (University of Bamberg)

Completed Projects

Federalism 2020–2024

Party Strategies in Cooperative Federalism

Germany's cooperative federalism forces parties to balance regional interests against federal loyalty. Using hand-coded parliamentary debates, above all from the COVID-19 pandemic, this project with Sabine Kropp and Antonios Souris traced how parties attribute blame and choose between bundesfreundlich and opportunistic argumentation. Published in West European Politics, Publius, and der moderne staat; the coded debate data are archived at GESIS. The StateParl corpus and the FEnDi dictionary grew out of this work.

With Sabine Kropp & Antonios Souris
Papers
  • Attributing blame: How political parties in Germany leverage cooperative federalism West European Politics, 47(7), 2024 DOI
  • Navigating conflicting incentives: Discursive strategies of political parties in Germany's cooperative federalism Publius: The Journal of Federalism, 54(4), 2024 DOI
  • Zusammenhalt im Bundesstaat? Bundesfreundliche und opportunistische Argumentationsstrategien in der Pandemie der moderne staat, 15(1), 2022 DOI
  • Prägt Opportunismus das föderale System? Herausforderungen des "bundesfreundlichen Verhaltens" in der Corona-Krise Staat, Rechtsstaat und Demokratie (W. Muno et al., Eds.), Springer VS, 2022 DOI
Discrimination 2020–2022

Anti-Asian Racism in Germany

Funded by the Berlin University Alliance and run with the DeZIM Institute, this project documented the rise of anti-Asian racism in Germany during the COVID-19 pandemic, combining survey data with field experimental evidence. Findings appeared in The British Journal of Sociology, Aus Politik und Zeitgeschichte, and DeZIM Insights, and were covered by Spiegel Online, Zeit Online, and others.

With Sabrina Mayer (PI), Susanne Veit, Jörg Dollmann & Kimiko Suda
Funding: Berlin University Alliance
Papers
  • The hidden majority/minority consensus: Minorities show similar preference patterns of immigrant support as the majority population The British Journal of Sociology, 74(4), 2023 DOI
  • Antiasiatischer Rassismus in Deutschland Aus Politik und Zeitgeschichte, 70(42/44), 2020
  • Mein rechter, rechter Platz ist leer: Wie Schutzmasken, Geschlecht und zugeschriebene Ethnizität die Sitzplatzwahl in der U-Bahn beeinflussen DeZIM Insights, 13, 2023
Data
Inequality Completed

Economic Insecurity and Political Efficacy

A long-running collaboration with Paul Marx on political inequality across income groups. Poverty depresses internal political efficacy by undermining cognitive and emotional resources, and dissent in the party system reduces the efficacy gap to higher incomes: conflict between anti-elite and mainstream parties simplifies political decisions and stimulates political attention among poor voters. Published in the European Journal of Political Research, the European Sociological Review, and the Journal of European Public Policy.

With Paul Marx (University of Bonn)
Papers
  • Anti-elite parties and political inequality: How challenges to the political mainstream reduce income gaps in internal efficacy European Journal of Political Research, 57(4), 2018 DOI
  • Are the unemployed less politically involved? A comparative study of internal political efficacy European Sociological Review, 32(5), 2016 DOI
  • Political participation in European welfare states: Does social investment matter? Journal of European Public Policy, 25(6), 2018 DOI
Welfare states Completed

Welfare States, Trust, and Punitiveness

Work growing out of my dissertation and my postdoc at the Centre for Welfare State Research in Odense. Labour market insecurity has a strong and persistent negative effect on generalized social trust, and social policy shapes how much damage it does (European Sociological Review); with Peter Starke, a second strand analyzes the relationship between welfare state attitudes and penal punitiveness (Punishment & Society).

With Peter Starke (University of Southern Denmark)
Papers
  • Labor market insecurity and generalized social trust in welfare state context European Sociological Review, 33(2), 2017 DOI
  • To support or punish? Analyzing the relationship between welfare state attitudes and penal punitiveness Punishment & Society, 2025 DOI