There are currently two competing views related to party identity: the expressive and the instrumental perspective. Rather than seeing these views as mutually exclusive, I combine them to develop a new typology with different types of partisanship. Partisanship can be seen as a matrix with expressive partisanship (emotional attachment) as one dimension and instrumental partisanship (cognitive analysis) as the second dimension. People may solely base their partisanship on one of the dimensions meaning their partisanship is either purely expressive or purely instrumental. However, it is also possible that people’s partisanship is determined by emotional and cognitive factors at the same time whereby it is likely that either the expressive or the instrumental partisanship is more salient. Hence partisans can be located at any position across the matrix depending on their individual mix of emotional and cognitive influences shaping their individual partisanship. Four generic types of partisanship can be derived in this model.
The model allows to extend party identification to multi-party systems which, so far, has been mainly studied in two-party systems. Each type of partisan, which can be found in a two-party system, can also be found in a multi-party system. In addition, the model accounts for the possibility of dual partisanships as combinations of the above mentioned generic partisanships are possible, e.g. people may have one partisanship which is mostly affective and another one which is mostly cognitive. In a comparative analysis of partisanship further potential applications of the model include the mapping of party families and countries with certain cleavages across the two dimensions of partisanship.
I plan to collect data in at least four European countries measuring the different types of partisans to find evidence for my typology and possible applications. Expressive partisanship is best measured using the Implicit Association Test (IAT) which measures response times to analyze the affective association of the respondent with a party. Instrumental partisanship can be elicited with standard explicit questionnaires. The data will then be used to analyze the validity of the two-dimensional model of partisanship and the mapping of voters and party families within the model.
- Paper on “A Typology of Party Identification in Multi-Party Systems”, at the ECPR General Conference, Prague, September 7-10, 2016.