Re-thinking turnout : explaining within-individual variation in electoral participation

Defence date: 7 July 2017Examining Board: Professor Hanspeter Kriesi, EUI; Professor Harold Clarke, University of Texas at Dallas; Professor Geoffrey Evans, Nuffield College, University of Oxford; Professor Diego Gambetta, EUIThis thesis presents four essays that aim to explain within-individual variation in voter turnout. The motivation behind this thesis is not only the importance of voter turnout to democracy, both in theoretical and substantive terms, but also the methodological and theoretical weaknesses in the existing literature caused by the lack of attention given to why individuals vote at some points in their lives and not at others. This deficit stands in contrast to the vast literature explaining aggregate level turnout – both within and between countries– and individual level turnout solely between individuals. Each of the four essays seeks to rethink one of the explanatory models of individual level voter turnout – mobilisation, resources, psychology and socialisation by applying many of their determinants to within-individual variation, as well as, in some cases, adding new ones. The methodological approach to explaining within-individual variation is to use fixed effects panel data models, as well as intermittently random effects models, cross-sectional models and structural equations. The data for these models comes from the British Household Panel Survey, the Swiss Household Panel Survey and the British Election Study. This thesis makes a number of theoretical, methodological and substantive contributions. I show that within-individual variation in voting seems to be fairly unaffected by such issues as material resources, ‘antipolitical’ sentiments, household politicisation or even feelings of personal ability to vote effectively – all of which have been mainstays of the between individual literature. Rather, I conclude that individuals vote when they are interested in the politics of the time, feel affinity towards a party or when a party has bothered to contact them, on the one hand, and, more fundamentally, by the lifestyle of the individual at the time of the election – with lifestyles built on rootedness, social integration and roles demanding responsibility increasing the individual’s desire to turnout to vote