ABSTRACT
The classical definition of party identification is that it is a “sense of personal attachment which the individual feels towards the [party] of his (sic) choice” (Campbell, Gurin and Miller 1954: 88–89). That is, voters have long-running attachments to particular parties regardless of candidates or issues in specific elections. Voters may defect from “their” party every now and then – they may choose a candidate from another party – but over the long run, more often than not, voters will have a homing tendency and return to support “their” party for which they have a sense of attachment. Party identification is probably the central conceptual building block in behavioral research and is a standard, one might even say required, factor to be included in models of vote choice, being seen as a precursor to the vote and party preference. One crude indication of its importance is found in Google Scholar where a search for the terms “party identification” AND “political science” produces over 27,000 results. Clearly, such a large literature presents challenges for any review. This chapter on party identification is therefore necessarily limited and organizes a discussion of party identification around three main questions. The first question is: what does party identification do? The second is: how (and why) do people develop party identification? And the third question asks: what kinds of variations do we see in party identification?
