Open Access

Julian Noseck

• Participating in an election is by far the most prevalent form of political participation in modern democracies. Turnout rates, however, not only vary considerably between countries but also over time: By trend, in many Western democracies turnout levels have declined over the last decades. Electoral systems depict a prominent factor that has always been discussed with respect to its impact on turnout. In this respect, a high number of empirical studies found aggregate turnout predominantly to be higher in countries using proportional representation compared to countries using a less proportional electoral system. Based on these findings, one should expect turnout to increase when the electoral system changes towards higher proportionality. However, empirical evidence of the actual lasting impact of changes in electoral institutions on voter turnout is all but conclusive. In this dissertation, I aim at answering the following question: What are the consequences of electoral system change for voter turnout? I argue that it is necessary to examine the relationship between electoral systems and turnout more detailed as most studies did to date by taking the level of electoral constituencies and the temporal dimension of electoral system change into account. To assess the impact of electoral systems and further proposed causal factors associated with electoral systems, party system size for instance, on turnout empirically, I make use of a comparative research design, analyzing longitudinal data with time-series cross-sectional regression models. These data, being the basis for my empirical analyses, represent a unique data set covering 9.639 electoral districts from 146 national legislative elections in eleven European countries. The dissertation generally finds an increasing district magnitude to boost turnout, while a decreasing magnitude has negative consequences for electoral participation. The positive effect of district magnitude on turnout seems to depend on the size of the population in the respective district, however. In addition, the analyses show that a higher number of parties in a district, respectively an increase in the number of parties in a district, has a negative impact on turnout.