No abstract

Recent commentary points to clear increases in ideological polarization between the major American political parties. We review the theoretical and empirical literature on party polarization and partisan change. We begin by comparing the current period both to earlier political eras and to theories of partisan change. We argue that in the current period the parties have grown increasingly divided on all the major policy dimensions in American politics-a process that we term conflict extension. We discuss various perspectives on increases in polarization between the parties in government, the parties in the electorate, and the parties' activists, and we consider the causal links between polarization at each of these levels. We consider whether American society itself, and not just the parties and their identifiers, has become increasingly polarized. Finally, we discuss the consequences of growing party polarization for American political life.

Scholars have long debated the individual-level relationship between partisanship and policy preferences. We argue that partisanship and issue attitudes cause changes in each other, but the pattern of influence varies systematically. Issue-based change in party identification should occur among individuals who are aware of party differences on an issue and find that issue to be salient. Individuals who are aware of party differences, but do not attach importance to the issue, should evidence party-based issue change. Those lacking awareness of party differences on an issue should show neither effect. We test our account by examining individuals' party identifications and their attitudes on abortion, government spending and provision of services, and government help for African Americans using the 1992-94-96 National Election Study panel study, finding strong support for our argument. We discuss the implications of our findings both for the microlevel study of party identification and the macrolevel analysis of partisan change.

No abstract

No abstract