Belief Network Analysis of American Public Opinion: Findings From A Novel Structural Methodology

• In Event: 472 - Section on Methodology Paper Session. Innovative Methodological Applications to Data Analysis

Abstract

Theories of the structure of political belief systems typically conceive of them as networks of interrelated opinions, in which some beliefs are central and others are derived from these more fundamental positions. Following in the tradition of social network analysis, we formally show how such structural features can be used to construct relational measures of belief centrality that are based on direct comparisons of structural positions of beliefs in the network of correlations. To demonstrate the usefulness of this methodology and contrast it with existing techniques, we examine the belief networks we construct from the 2000 American National Election Study dataset. While regression analyses of these data have been used to argue that political beliefs are organized around cultural schemas of parenting, our structural approach contradicts this interpretation. Instead, our results are broadly consistent with the conception of liberal-conservative political identity as a heuristic device for acquiring attitudes. Moreover, in our analyses of belief networks across 39 demographic subpopulations, we find strong and consistent evidence that belief systems generally vary in the extent to which they are organized, but not in the logic around which they are organized. While our analyses focus on political beliefs, techniques we introduce here can be applied to a wide variety of other cultural domains.

Authors

• Andrei G. Boutyline, University of California-Berkeley

• Stephen Vaisey, Duke University