Since the 1950s, party identification has occupied a central place in the study of political behaviour. Although scholars may disagree about whether party attachments are on the decline or are stronger in some places than others, no one disputes the fact that party identification strongly predicts public opinion and vote choice.1 The main scholarly disagreements instead focus on issues related to the conceptualization and measurement of party identification. Is party identification primarily a social identity akin to ethnic identification insofar as it tends to be learnt relatively early in life and remains stable thereafter? Or is party identification primarily a reflection of voters’ stances on leading political issues, evaluations of party performance, or preferences regarding leading candidates for office? Why do people acquire and retain partisan identities – is it simply a reflection of their self-categorization, or do such identities emerge because they provide an instrumental benefit, such as simplifying complex choices or augmenting feelings of self-esteem?