Doctoral Dissertation Research in Political Science: Information Diffusion in Complex Electoral Environments: Voting Behavior in Two Brazilian Cities

University Of Pittsburgh, Pittsburgh PA

Investigators

Abstract

This Doctoral Dissertation addresses the issue of how voters learn about politics during elections in a complex electoral environment (CEE). A CEE is includes the following: voters weakly identify with parties, there are several parties, more than one seat is in dispute by district, and many candidates run for the same office. An environment with at least a few such characteristics does not correspond to Sniderman, Brady, and Tetlock's (1991) description of the political system in the United States (an extremely well-studied case) where voters are able to pick up effective cues during electoral campaigns because the "world of politics is so organized" (p. 29). There are, however, systems where the complexity of the electoral process is much greater. In several Latin American countries, for example, political parties are weaker than in the US (Stokes 1999, Mainwaring and Scully 1996) and more than one representative is elected per district (Nicolau 1999). This research extends the studies of information diffusion during electoral campaigns to political systems with CEEs, such as those of Latin America. The impact of several distinct sources of information diffusion is analyzed using a multiplicity of research techniques -including survey data, media content analysis data, and in-depth interviews with political and social activists. Two Brazilian cities during the 2002 Chamber of Deputies election are the locus of the study. These cities provide enough variation in electoral complexity to test hypotheses about voters' learning process in milieus with distinct levels of electoral competition and strength of party cues.