The Logic of Group Voting: Identity Mobilization in Political Institutions

• In Event: Race and Identity in Political Behavior

Abstract

How the size and characteristics of social group identities affects the voting decisions of its members remains one of the most central but difficult to address inquiries of political science. Ever since Lispet and Rokkan (1967) stressed the important role of social and economic cleavages in determining political support across various identity types, the topic of identity voting has found relevancy in every world region. There remain, however, a large number of methodological and theoretical challenges, and the topic of socio-political group behavior has been advanced separately by conflict scholars, economists, and social psychologists in addition to political scientists. This paper seeks to advance the debate forward by overviewing and collecting the key hypotheses from this diverse set of literatures, and testing these hypotheses while addressing the common critiques and pitfalls of measuring social group salience. The outcome of interest is the level to which a particular social identity groups supports a political party compared to the national popularity of the party. Increasing coordination of voting intentions signals the salience and capacity of a particular social group to advance their interests and access political power. The analysis is constructed as a multi-level model testing over 1,300 distinct social identity groups represented by public opinion data from 123,751 individuals across 80 countries and representing every world region. Findings show that individuals are systematically influenced by the characteristics of their social group identities including the number of competing groups, prominence within a potential majoritarian coalition, the actual magnitude of group identifiers, and the relative power and differences between rival groups. Critically, however, these phenomena are in most cases mediated through either the type of electoral system (PR or Plurality), executive system (Parliamentary or Presidential) or both. The coalescing of social group votes also increases as electoral competitiveness heightens and political parties emphasize more distinct platforms, but falls as education and access to information increases among group identifiers. These findings suggest that the logic of group voting is likely more generalizable than areas experts presume, given the geographic scope of the data and the fact that the type of social identity and survey the data was drawn from do not systematically increase or decrease voting mobilization in this analysis.

Author

• Paul Friesen, University of Notre Dame