Although electoral volatility is a central concept in research on voting behaviour, we still know relatively little about which voters are most likely to switch parties. Given the scarcity of comparative surveys that have information on more than one election, the contextual variables influencing vote-switching have hardly been studied at all.
A contextual variable that most of the theoretical literature identifies as essential for explaining electoral volatility is political polarisation. The more polarised a political system is, the larger the ideological distances between political parties. The central assumption on the individual level is that voters will be more reluctant to change parties if these parties are ideologically very dissimilar. Because it is a voters’ perception of the distance between parties that drives his/her decision to switch, the link between political polarisation and electoral volatility is one that has to be investigated at the individual level. Especially with regard to the link between polarisation and electoral volatility, distances between parties and how distant voters perceive different parties to be has an important impact on voters likelihood of switching. Therefore the focus of this paper is on a continuous measure of volatility.
In this paper, we will investigate the link between polarisation and electoral volatility in a comparative way. Therefore, we will make use of data of the second and third module of the Comparative Study of Electoral Systems (CSES) project. Doing so provides us with data that allow to include and investigate the effects of both country- and individual-level variables. The individual-level voter survey data provided by the CSES project will be complemented with party level data from the Comparative Manifesto Project (CMP). This combination of datasets will allow us to enhance our knowledge about the link between party polarisation and electoral volatility in a comparative way.
