research

journal articles

1. Who Should Politicians Represent? How Winning an Election Influences Citizens’ Views on Representation

   Tim Lars Allinger

   (Ahead of Print at The Journal of Politics)

   Abstract URL Replication

   Should elected politicians represent their own voters or all citizens, regardless of who they voted for? Though a fundamental question in representative democracies, we lack insights into citizens’ perspectives on this tension despite their essential role in democratic stability and legitimacy. I argue that while there is generally strong support for the idea that an elected politician should represent all citizens in their constituency (partisan surrogation), this support diminishes markedly due to self-interested considerations once citizens find themselves on the winning side of an election. Relying on a regression discontinuity design in close district races across three German national elections, I present quasi-experimental evidence in line with this argument. The findings indicate that many citizens expect partisan surrogation when losing without being willing to concede that to others upon an election victory. This highlights the need for a more dynamic perspective when studying citizens’ expectations of the representative system.

   

2. Social Desirability Bias In Online Surveys: An Experimental Assessment

   Daniel Bischof, Tim Lars Allinger, Kristian Frederiksen, and Morgan Le Corre Juratic

   (Just Accepted at The Journal of Politics)

   Abstract URL Replication

   Social scientists have long debated the extent to which social desirability bias affects online survey responses. We argue that the a priori most promising way to measure it is through a global manipulation at the survey’s outset. If effective, this approach allows researchers to (1) isolate social desirability from confounders, (2) verify manipulation success, and (3) assess effects across multiple outcomes. Using well-powered survey experiments in the United States and Denmark, we provide a comprehensive assessment of how social desirability can be successfully manipulated and to what extent it biases online survey responses. Our findings show that some treatments fail to induce perceptions of being observed, while others succeed but do not alter outcomes most likely to be affected by social desirability bias. Beyond offering guidelines for future research designs, these results imply that researchers should be less concerned of social desirability bias in online surveys than previously thought.

   

working papers

1. Do Exclusionary Policies Increase Hate Crimes? Evidence from the Nomad Camps in Italy

   Tim Lars Allinger

   (Available upon request)

   Abstract

   Hostility and hate crimes against immigrants and ethnic minorities are a persistent problem, which have surged again across Europe in recent years. Yet, the role of state policies in shaping such violence remains understudied. I argue that exclusionary policies targeting minority groups can legitimize hostility, increasing hate crimes not only against the targeted group but also against other minorities. To test this, I examine Italy’s nomad camps – institutionalized, mono-ethnic settlements for Sinti and Roma – using municipality-level hate crime data. Leveraging spatial spillover effects combined with an instrumental variable approach, I show that proximity to these camps increases hate crimes against individuals with a migration background. Notably, this effect extends beyond camp residents, suggesting a broader shift in intolerant actions rather than mere opportunity-driven violence. These findings highlight the consequences of exclusionary policies and underscore the state’s role in shaping intergroup relations, with implications for policymaking in societies grappling with rising anti-minority violence.

   

2. Is this my Last Resort? Understanding Public Support for Hate Crimes against Refugees

   Tim Lars Allinger

   (Available upon request)

   Abstract

   Violent attacks against refugees remain a pervasive problem – also because they enjoy notable public support, especially among far-right voters. Yet little is known about the political conditions under which anti-refugee grievances escalate into support for such violence. I argue that this support is not only driven by hostility toward refugees but also by perceptions of political unresponsiveness. Specifically, individuals are more likely to endorse violence when refugee policies are perceived as overly permissive and non-violent efforts as having failed previously, making violence appear as a last resort. To test this, I fielded a preregistered vignette experiment with 2,356 far-right voters in Germany, the group most susceptible to such violent appeals. The experiment independently varied local refugee policy and prior citizen protest, and then measured respondents’ endorsement of an attack on a refugee home. The findings confirm the argument: support for violence increases in response to permissive policies, but only when preceded by protest. The results have important implications for understanding how institutional behavior can shape public support for violence in polarized and migration-challenged democracies.

   

3. Democratic Support Is Conditional on Social Norms: How Citizens’ Perceptions of a Democratic Consensus Matter

   Daniel Bischof, Tim Lars Allinger, Morgan Le Corre Juratic, and Kristian Frederiksen

   Abstract Preprint

   Democratic stability is often said to rest on a social consensus over democratic values. Yet this consensus is usually inferred from aggregating individual attitudes rather than measured in its social nature – as citizens’ perceptions of what others support. We conceptualize democratic consensus as a perceived social environment (social norm) and argue that democratic support is conditional on it: citizens should be more willing to support and defend democracy when they believe democratic commitments are widely shared. We study this perceived consensus by mapping it across 14 democracies and assessing its consequences with embedded vignette experiments and quasi-experimental interview-mode evidence. Citizens generally perceive broad support for democracy, but this consensus is uneven and socially fragmented. Perceiving no consensus reduces democratic support, increases acceptance of attacks on political institutions, and weakens willingness to defend democratic commitments publicly. Democratic resilience may therefore be fragile even where democratic attitudes appear stable.

   

4. Democratic Transgressions Embedded in Reality

   Kristian Frederiksen, Tim Lars Allinger, Morgan Le Corre Juratic, and Daniel Bischof

   Abstract Preprint

   Research on citizens and democratic backsliding has skyrocketed over the past decade. Most of this research has focused on why citizens might tolerate hypothetical undemocratic behavior carried out in the abstract by hypothetical actors. We present a theoretical framework and use a two-wave panel survey from six challenged democracies (the United States, Hungary, Poland, Brazil, Mexico, and India) to assess the consequences of this approach. In the first wave, we develop an alternative survey instrument based on real-world democratic transgressions and compare support for such transgressions to typical hypothetical behaviors. In the second wave, we construct an intervention to compare how easily support for real-world transgressions is moved vis-à-vis support for hypothetical transgressions. Our results show that support for real-world undemocratic behaviors is substantially higher but also easier to intervene against. Prior research may simultaneously have underestimated both support for undemocratic behavior and the prospects of intervening against it.

   

5. Democratic Transgressions Embedded in Reality

   Mirko Wegemann, Elena Leuschner, Tim Lars Allinger, Kristian Frederiksen , and 2 more authors

   Abstract Preprint

   Do citizens agree on what democracy means? We analyze more than 25,000 open-ended survey responses from representative samples in 14 countries, classifying answers with a fine-tuned language model and sentence embeddings. We find that citizens most often define democracy one-dimensionally, in terms of individual and civil rights, while elections are mentioned by only 20% of respondents. Yet there is partisan divergence. Far-right voters and non-voters are less likely to mention civil rights, elections, and egalitarianism than other voters, and far-right voters more often invoke majoritarian principles. Among respondents mentioning the rights dimension, left-wing voters connect rights to equality and citizenship, whereas right-wing voters emphasize freedom of speech. Democratic contestation thus unfolds within democratic language, creating opportunities for selective elite reframing.

   

work in progress

1. Hate Crime Support Across 13 Countries

   Tim Lars Allinger, and Daniel Bischof
2. Political Reactions to Anti-Refugee Violence

   Tim Lars Allinger
3. Does Violence Work? Perceived Consequences and Support for Anti-Refugee Attacks

   Tim Lars Allinger, and Geraldine Bustos
4. MrP Measurements and Causality

   Tim Lars Allinger, and Lucas Leemann
5. Social norm interventions increase resistance to real-world anti-democratic elite behavior

   Kristian Frederiksen, Morgan Le Corre Juratic, Tim Lars Allinger, and Daniel Bischof
6. Ballots and Backyards: The Influence of Objective and Perceived Income Mobility on Voter Turnout

   Tim Lars Allinger, Jonas Ineichen, Thomas Kurer, Lucas Leemann , and 1 more author