2012-2013 Winter Colloquia

ABSTRACT: Research found that prejudiced people avoid contact with outgroup members whereas those with pre-existing positive attitudes toward the outgroup willingly engage with members of this group. It thus seems paradoxically that intergroup contact interventions successfully promote intergroup friendships even among those high in prejudice. This paper challenges the assertion responsible for this apparent paradox; that outgroup attitudes directly cause or prevent intergroup friendships. Instead, network processes such as transitivity mediate and intensify the effect of outgroup attitudes. Majority group member with more positive attitudes towards a minority group are more likely to have friends who have minority friends and because of a tendency to form transitive relationships, these majority group members then become friends of their friends’ outgroup friends. This assertion is supported by longitudinal network analysis of data from a three-wave study of 12-13 year-olds (N = 1,113). No evidence was found for reciprocity as an alternative explanation; students with more positive outgroup attitudes did not simply receive more friendship invitations from outgroup members. Neglecting the role of network processes as determinants of outgroup contact has most likely led to an overestimation in past research of the apparent direct impact of outgroup attitudes on the number of intergroup friendships. Transitivity did not mediate the effect of outgroup attitudes on friendship selection among minority group students. Given the larger size of the majority group, positive attitudes can directly guide friendship choices of minority members whereas majority members have to be introduced to the fewer minority members by their friends.

ABSTRACT: Climate change mitigation requires international cooperation and for this cooperation to be sustainable over the long term, formal global agreements to reduce CO2 emissions need broad public support. Using data from an experimental conjoint survey, we provide estimates of the political demand for different types of climate agreements in France, Germany, the United Kingdom, and the United States. Specically, we explore how three key dimensions of climate cooperation | costs and distribution, participation, and enforcement | aect demand for global climate agreements. We contend that citizens’ sentiment toward climate agreements most strongly depends on costs. Our estimates imply that an increase of household costs equivalent to 0.5% of gross domestic product decreases the probability that an individual supports an agreement by 20% percent. Our results, however, also suggest that citizens are sensitive to the principles that govern the international distribution of costs, prefer more encompassing forms of climate cooperation, and support agreements that include a low sanction for failing emission reduction targets. Overall, our findings suggest that an important mechanism through which interests, norms, and institutions can support international cooperation is their influence on public opinion.

ABSTRACT: While water is the most important commodity for life, our economy, and our environment, it is something that few people understand, appreciate, or even think about. It is out of sight, out of mind. There has been very little research into citizen understanding of water – where it comes from, where our waste goes, how it is used and how vulnerable it makes us. It is hard to get anyone to support changes and reforms to something they don’t understand and don’t recognize as important or in need of fixing. How do we ask consumers to pay more for something that they have become accustomed to paying so little? I would like to engage the group in a conversation about the need to acquire better and more frequent data about public attitudes and understanding of freshwater in the Western US.

ABSTRACT: In previous research, and consistent with stereotypes linking blackness and criminality, we demonstrated that racial perceptions are affected by contact with the criminal justice system; people are more likely to be classified as black and less likely to be classified as white after being incarcerated. We extend this research using data from the National Longitudinal Study of Adolescent Health, which includes information on whether respondents have been arrested, convicted or incarcerated, and the details of their most recent arrest. This allows us to ask whether any contact with the criminal justice system results in racialization or only certain types of contact. Additional racial categories also allow us to explore the potential racialization of crime beyond the black-white divide. We conclude by discussing the implications of the criminal justice system as an institution that "makes race" in the context of mass incarceration.

ABSTRACT: Optimization is a key ingredient in both model fitting and policy decisions. While social science research has embraced the use of optimization techniques for specific problems in model fitting, such as regression and logistic regression analyses, the use of mathematical optimization based on these models for the purposes of policy decisions has not proceeded apace. In this talk, we present an application of mathematical optimization to vote share optimization, using data from the American National Election Studies. The solution uses a novel mathematical optimization technique called sigmoidal programming, which maximizes a sum of sigmoidal functions over a convex constraint set. In this talk, we discuss the idea behind this algorithm, present some numerical results, and consider other lessons from mathematical optimization that may be of interest to the social science community.

ABSTRACT: Relative deprivation (RD) is the judgment that one is worse off compared to some standard accompanied by feelings of anger and resentment. Social scientists use RD to predict a wide range of significant outcome variables: willingness to join protests, individual achievement and deviance, intergroup attitudes, and physical and mental health. But the results are often weak and inconsistent. To determine whether these results reflect measurement or theoretical deficiencies, the authors conducted a meta-analytic review of 210 studies (representing 293 independent samples and 186,073 respondents). RD measures that (1) include justice-related affect, (2) match the outcome level of analysis and (3) used higher quality measures yielded significantly stronger relationships. Future research should focus on appropriate RD measurement, affect and the inclusion of theoretically relevant appraisals of the situation. Such methodological improvements would revitalize RD as a useful social psychological predictor of a wide range of important individual and social processes.