Issue Preferences, Socialization, and the Evolution of Party Identification

Abstract

The traditional view that party identification develops at an early age, is remarkably stable throughout life, and is relatively unaffected by other political attitudes has been recently challenged by evidence that adult partisanship is in fact quite responsive to other political attitudes such as policy preferences, retrospective evaluations, and past votes. The author tests predictions based on the traditional and revisionist views by estimating a model of party identification which considers the effects of parental socialization and policy preferences on the party identification of young adults. In addition, the author estimates the reciprocal effects of partisanship on policy preference. The traditional view is that party identification develops at an early age, largely through the influence of parents, is remarkably stable throughout life, is relatively unaffected by short-term forces, and acts as a central organizing force for other political perceptions and preferences. While the theoretical underpinning of this view, and much of the empirical evidence, was developed by the early studies of electoral behavior and childhood socialization, contemporary work continues to apply this view of an exogenous partisanship to theories of political behavior. 1 Recent research questions the adequacy of this formulation. There is growing evidence that party identification responds systematically to contemporaneous political perceptions. Thus party preferences are not exogenous with respect to other political perceptions, but are part of an endogenous system of preferences. Further, there is evidence that even if party identifications are learned early and from parents this learning is not as enduring as early theories seemed to imply. Revised models which incorporate these findings lead to interpretations of the dynamics of public preferences which differ substantially from previous discussions. These new structures present a dynamic model of partisanship in which preferences depend not only on past learning, but also on the current positions and performance of the parties. This paper extends the revisionist argument by considering the role of socialization in a revised model of partisanship.