I have three projects currently in progress.
This paper proposes a theoretical model called perceptual impact theory to explain how social geographic context can affect intergroup attitudes, a subject of interest since Key’s (1949) seminal work on racial threat in the American South. Perceptual impact theory is derived from recent research in political science, as well as social and evolutionary psychology. The foundation of this theory is that individuals possess an evolutionarily adaptive tendency toward self-preservation that is temporally prior to exposure to an external social environment, as well as an adaptive tendency to heuristically link perceived vulnerability to harm with the presence of an outgroup (Faulkner et al. 2004). The former tendency primes individuals who perceive themselves to be particularly vulnerable to physical harm to view related information in the external social environment, such as rates—and changes in rate (Kahneman & Tversky 1979)—of crime and infectious disease, with a relatively higher degree of salience. In turn, the strength with which such information is tied to evaluations of social identity outgroups is conditional of the “spatial impact” (Enos 2011a, b) of such outgroups. As Enos (2011a, b) argues, the spatial impact of an outgroup is conditional on its size, concentration (relative segregation or intergration), and proximity to an ingroup individual. Finally, the strength of such a linkage between environmental safety information and outgroup spatial impact is conditional on political communications that can further prime or attenuate the salience of such a linkage (Hopkins 2010). To test this theory, I propose survey and laboratory experiments in order to surmount unit specification issues that plague geographical analyses of intergroup attitudes (Baybeck 2006; Cho & Baer 2011; Openshaw & Taylor 1979; cf. Enos 2011a for empirical strategies to account for these issues within a geographic analysis).
2) Newburg, James. 2012. “Ethnic Identity and Trust in Parliament in a Consensus Democracy: A Case Study of Belgium.”
Consociationalism is suggested as an institutional arrangement for mitigating ethnic conflict in deeply divided states. The most important feature of consociationalism is the amount of veto power granted to minority groups, which puts a brake on drastic policy changes favoring an ethnic majority. Does this institutional reality, though, affect the opinion ethnic groups have toward their government? I examine the case of such an ethnically divided democracy, Belgium, using pooled data from four rounds of the European Social Survey to examine attitudes toward trust in parliament. My theory is that, typically, individuals of a given ethnic group use identity as a heuristic for evaluating trust in parliament. However, in a consensus government, the heuristic salience of ethnicity is attenuated. Instead, constraints placed on majorities lead to members of the Dutch-speaking Flemish majority to trust government less than the French-speaking Walloon minority, presumably because Flemish interests are less well represented than they otherwise would be in a majoritarian system. In two statistical models, I find that French-speaking Walloons trust parliament to a greater degree than Dutch-speaking Flemish.
This paper studies the election results of a local growth-control measure, Proposition L in the 2000 San Francisco general election, to examine two questions. First, how does residential location relative to the proposed project influence vote choice on this type of initiative? Second, do variables previously shown to explain voting behavior on similar initiatives have the same effect across racial groups? Using a Geographically Weighted Regression version of King’s Ecological Inference model (Calvo and Escolar 2003; King 1997), I generate estimates of individual-level voting behavior by racial group, which are dependent variables in ordinary least squares and spatial lag regression models to test three hypotheses: 1) increased distance from the Prop. L project area reduces voter support; 2) the spatial structure of vote choice indicates behavioral contagion; 3) the model of voting behavior specified behaves similarly across racial groups. My findings appear to affirm the first two hypotheses, but provide limited support for the third.