The concept of party identiﬁcation has been a matter of dispute ever since it was ﬁrst introduced by a team of US scholars based at the University of Michigan in the 1950s (Belknap and Campbell 1952; Campbell et al. 1954, 1960).1 These debates are wide-ranging but essentially boil down to four major issues. The ﬁrst relates to the nature of party identiﬁcation: what is this concept exactly? The second concerns the sources of party identiﬁcation and its stability: how does it develop? The third is strongly related to both these issues: how should party identiﬁcation be measured? The fourth and ﬁnal question relates to applicability of the concept outside the United States: is it useful in parliamentary systems, such as those of Western Europe, or only relevant in the country in which it was developed?