Christopher Claassen

My CV is available here.

Current Research

At the microlevel, comparative public opinion data are abundant. But at the macrolevel, the level where many prominent hypotheses in political behavior are believed to operate, data are scarce. In response, we develop a Bayesian dynamic latent trait modeling framework for measuring smooth country-year panels of public opinion even when data are fragmented across time, space, and survey item. We derive six models based on this framework, and test their accuracy using data on support for democracy. Our best model is reasonably accurate, with predicted responses that deviate from the true response proportions in a hold-out dataset by six percentage points. We show, furthermore, that our smoothed country-year estimates of support for democracy have both construct and convergent validity, with spatiotemporal patterns and associations with other covariates that are consistent with previous research.

Political tolerance has long been regarded as one of the most important democratic values because intolerant political cultures are believed to foster conformity and inhibit dissent. Although widely endorsed, this theory has rarely been investigated. Using multilevel regression with poststratification to measure levels of macro-tolerance in U.S. metropolitan areas, and event data to measure rates of protest, we test whether cultures of intolerance do indeed inhibit public expressions of dissent. We find that they do: levels of macro-tolerance are positively and strongly associated with higher rates of protest in American metropolitan areas. Our findings have implications for the study of political tolerance, for normative theories of free speech and other civil liberties, and for scholarship on protest and collective action.

An important paper (Halperin, Canetti-Nisim, & Hirsch-Hoefler, 2009) brings emotion squarely into the study of political intolerance, finding that (1) hatred of outgroups, but not anger or fear, leads directly to intolerance; (2) group-based anger and fear influence intolerance but through the mediation of hatred; and (3) hatred has a markedly stronger influence among the unsophisticated. These conclusions challenge much of the conventional wisdom in the study of intolerance. Yet we suspect that the findings may primarily be a function of the context of the study: Israel, a deeply divided society where intergroup hatred likely plays an outsized role in political life. Using a large representative sample of the American population, we re-examine the influence of emotion and sophistication on intolerance. Our findings differ dramatically and in nearly every respect from those of Halperin, Canetti-Nisim & Hirsch-Hoefler. Hatred is indeed associated with intolerance, but only quite modestly, and no more so than are anger or fear; emotions play a less significant role than traditional predictors of intolerance; finally, we find that the effects of emotion on intolerance are not consistently stronger among the unsophisticated. All of these findings speak to the role political contexts play in structuring mass political intolerance.

Religious group size, demographic composition, and the dynamics thereof are of interest in many areas of social science, including migration, social cohesion, parties and voting, and violent conflict. Existing estimates however are of varying and perhaps poor quality because many countries do not collect official data on religious identity. We propose a method for accurately measuring religious group demographics using existing survey data: Bayesian multilevel regression models with post-stratification, or MRP. We illustrate this method by estimating the demography of Muslims, Hindus, and Jews in Great Britain over a 20-year period, and validate it by comparing our estimates to UK census data on religious demography. Our estimates are very accurate, differing from true population proportions by as little as 0.29 (Muslim) to 0.04 (Jewish) percentage points. These findings have implications for the measurement of religious demography as well as small group attributes more generally.

We argue that there may be some utility to bringing groups back into the study of intolerance. Even with controls for group threat, groups may differ in ways that are highly significant for tolerance. In particular, people may react differently to groups perceived to be anti-democratic, and other group attributes may be influential as well. Consequently, we analyze political intolerance using a multilevel model that takes into account both micro-level determinants and group-level determinants of intolerance. We conclude that the “least-liked” approach can be usefully supplemented by including perceived group attributes. When it comes to tolerance, not all groups can be treated equally.

Immigrants are regularly blamed by natives for taking jobs, committing crime, being a burden on public resources, and rejecting the dominant culture. Researchers have shown, moreover, that blame facilitates aggression by focusing attention on a target and triggering anger. The purpose of this paper is to explicate and test a theory of why immigrants are blamed. There are two hypotheses. The 􏰀first hypothesis argues that blame is rooted in threats to native entitlement. In other words, natives blame immigrants when the intergroup distribution of resources is incongruent with the distribution that is believed to be normatively desirable. The second hypothesis proposes that these contextually-shaped threats moderate the effects of individual differences; perceptions of threat will have a particularly pronounced effect on blame for individuals who subscribe to authoritarian values. These hypotheses are tested using survey data gathered in an urban slum in South Africa, where widespread violence against immigrants occurred in 2008 and where tensions remain. Analysis of the survey data supports both hypotheses.

Published Research

After widespread violence in 2008 and 2015, South Africa is now clearly one of most hostile destinations in the world for African migrants. Existing research on the determinants of South African xenophobia has focused on developing and advancing theories, with little attention paid to testing which theories, if any, actually account for mass xenophobia. This is the goal of this paper. By combining individual-level Afrobarometer survey items with municipal-level census indicators, we produce a rich, quantitative data set of numerous factors that have been proposed as determinants of South African xenophobia. The results of multilevel regression analyses show support for the explanations of poverty, relative deprivation, frustration with government, and social mobilization, with mixed evidence for resource competition. Taken together, the results point toward a mechanism of scapegoating, where frustrations and hopelessness produce aggression that is targeted at African immigrants.

There is little research on the thousands of individuals who take part in intergroup violence. This article proposes that their participation is motivated by the emotion of intergroup anger, which, in turn, is triggered by a comparison between the intergroup distribution of resources and the distribution that is believed to be desirable. Thus, when another group is perceived to violate group entitlements – by taking jobs thought to belong to the ingroup, for example – anger is experienced and individuals become more willing to take part in violence against the outgroup. Support for this theory is found in a new survey dataset, collected in a slum in South Africa where anti-immigrant violence occurred in 2008.

An earlier version of this paper received an honourable mention in the 2012 Otto Klineberg Intercultural and International Relations Award from the Society for the Psychological Study of Social Issues.

I use a Bayesian hierarchical latent trait model, and data from eight different university ranking systems, to estimate university quality. By combining information from different systems, I obtain more accurate ratings than are currently available from any single source. And rather than dropping institutions that receive only a few ratings, the model simply uses whatever information is available. In addition, while most ratings focus on point estimates and their attendant ranks, I focus on the uncertainty in quality estimates, showing that the difference between universities ranked 50th and 100th, and 100th and 250th, is insignificant. Finally, I also estimate the error implicit in each rating, allowing me to measure the accuracy of the various ranking systems.

This paper extends recent research on the operational-symbolic paradox in American politics: the presence of “conflicted conservatives” who hold conservative symbolic identities but are liberal on the issues. Using issue-level measures of ideological incongruence, we find that substantial numbers—over 30 percent—of Americans experience conflicted conservatism, particularly on the issues of education and welfare spending. We also, however, find that 20 percent of Americans exhibit conflicted liberalism. Our analysis of the determinants of ideological conflict confirms existing findings that conflicted conservativism is a function of low sophistication and religiosity. We also find, however, that partisan, ideological and ethnic identities influence the extent of conflicted conservatism and conflicted liberalism.

Little is known about the thousands of people who take part in communal violence. Existing research is largely based on interviews, impressionistic accounts and government records of arrestees. In contrast, this paper examines data from a novel survey of a representative sample of residents of Alexandra, a township in South Africa where a 2008 nation-wide wave of anti-immigrant riots began. Data on participation in the attacks were collected using a method ensuring the privacy of responses, thus potentially reducing response bias. In contrast to the conclusions of existing research, which emphasize the participation of young males, the survey data reveal that a significant number of participants were female and participants were not particularly young, being 34 years old on average. Participants are more likely to support an opposition party, attend community policing meetings and have a high school education.

We test a model of US electoral politics where activist groups contribute resources to their favoured parties. We find that presidential candidates in the United States are pulled from their convergent equilibrium position by the influence of activists. In particular, using American National Election Study survey data, we find that in the 2008 presidential election social activists were influential in the Democrat party and economic activists in the Republican party.

We investigate interracial “reconciliation” in South Africa, comparing how group levels of reconciliation have shifted from 2001 to 2004. Noting that contact is one of the most important drivers of reconciliation, we then examine the changing levels of intergroup contact over this time period.

We test the congruence hypothesis—the claim that democracies requires popular support. Using World Values Survey data, we find that demand for democracy, or mass preference for democracy over its authoritarian alternatives, is indeed associated with the supply of democracy. However, it is the level of demand in relation to the exiting level of supply that results in regime changes: democracy tends to increase where popular demand outstrips supply and contract where demand lags behind supply.