How does wartime exposure to ethnic violence affect the political preferences of ordinary citizens? Are high-violence communities more or less likely to reject the politicization of ethnicity post-war? We argue that community-level experience with wartime violence solidifies ethnic identities, fosters intra-ethnic cohesion and increases distrust toward non-co-ethnics, thereby making ethnic parties the most attractive channels of representation and contributing to the politicization of ethnicity. Employing data on wartime casualties at the community level and pre- as well as post-war election results in Bosnia, we find strong support for this argument. The findings hold across a number of robustness checks. Using post-war survey data, we also provide evidence that offers suggestive support for the proposed causal mechanism.

In regions plagued by reoccurring periods of war, violence and displacement, how does past exposure to violence affect altruism toward members of different ethnic or religious groups? Drawing on theories of empathy-driven altruism in psychology, this article proposes that violence can increase individuals’ capacity to empathize with others, and that empathy born of violence can in turn motivate helping behavior across group boundaries. This hypothesis is tested using data on the hosting behavior of roughly 1,500 Liberians during the 2010–11 Ivorian refugee crisis in eastern Liberia, a region with a long history of cross-border, inter-ethnic violence. Consistent with its theoretical predictions, the study finds that those who experienced violence during the Liberian civil war host greater numbers of refugees, exhibit stronger preferences for distressed refugees and less bias against outgroup refugees, and host a higher proportion of non-coethnic, non-coreligious and distressed refugees. These findings suggest that violence does not necessarily lead to greater antagonism toward outgroups, as is often assumed, and that in some circumstances it can actually promote inter-group co-operation.

What explains citizens’ willingness to fight for their country in times of war? Using six waves of the World Values Survey, this study finds that individual willingness to fight is negatively related with country-level income inequality. When income inequality is high, the rich are less willing to fight than the poor. When inequality is low, the poor and rich differ little in their willingness to fight. This change in the willingness to fight between low and high inequality countries is greater among the rich than among the poor. This article explores several explanations for these findings. The data are consistent with the argument that high inequality makes it more attractive for the rich to buy themselves out of military service.

The process by which people transfer their allegiance from ethnic to national identities is highly topical yet somewhat opaque. This article argues that one of the key determinants of national identification is membership in a ‘core’ ethnic group, or Staatsvolk, and whether or not that group is in power. It uses the example of Uganda as well as Afrobarometer data to show that, when the core ethnic group is in power (as measured by the ethnic identity of the president), members of this group identify more with the nation, but when this group is out of power members identify more with their ethnic group. This finding has important implications for the study of nationalism, ethnicity and African politics.

Welfare states are exposed to a host of cost-inducing ‘reform pressures’. An experiment implemented in Germany, Norway and Sweden tests how various reform pressure frames affect perceptions about the future financial sustainability of the welfare state. Such perceptions have been shown to moderate electoral punishment for welfare reform, but little is known about their origins. Hypotheses are formulated in dialogue with newer research on welfare state change, as well as with older theory expecting more stability in policy and attitudes (the ‘new politics’ framework). Research drawing on ‘deservingness theory’ is also consulted. The results suggest large variations in impact across treatments. The most influential path to effective pressure framing is to ‘zoom in’ on specific economic pressures linked to undeserving groups (above all immigration, but also to some extent low employment). Conversely, a message emphasizing pressure linked to a very deserving group (population aging) had little effect. A second conceivable path to pressure framing entails ‘zooming out’ – making messages span a diverse and more broadly threatening set of challenges. This possibility, however, received weaker support.

What are the consequences of committing violent attacks for terrorist organizations? Terrorist attacks might broaden the base of supporters by increasing the perceived group efficacy. However, terrorist attacks might also lead its supporters to believe that the organization is excessively violent or involvement may become too dangerous. This article employs a unique dataset with 300,842 observations of 13,321 Twitter accounts linked to the Islamic State (IS), collected during a 127-day period, to empirically investigate the impact of terrorist attacks on the number of the organization’s supporters. By exploiting the exogenous timing of terrorist attacks as a natural experiment, we find that the number of followers of IS-related Twitter accounts significantly reduces in the aftermath of the attacks. Additionally, we provide some suggestive evidence to disentangle two mechanisms: disengagement – a change in supporters’ beliefs – and deterrence – demobilization due to fear. Because we do not find support for the latter, we conclude that the disengagement effect might explain our main result.

Media stories often reach citizens via a two-step process, transmitted to them indirectly via their social networks. Why are some media stories strongly transmitted and impact opinions powerfully in this two-step flow while others quickly perish? Integrating classical research on the two-step flow of political communication and novel theories from cognitive psychology, this article outlines a model for understanding the strength of political frames in the two-step flow. It argues that frames that resonate with cognitive biases (that is, deep-seated psychological decision rules) will be transmitted more and have a stronger influence on opinion when citizens recollect media frames in their social networks. Focusing on the case of episodic and thematic frames, the study tests this model. It introduces a novel research design: implementing the children’s game ‘Telephone’ in consecutive experimental online surveys fielded to nationally representative samples. This design helps gauge the reliability of transmission and the degree of persuasiveness in actual chains of transmission.