In this paper we investigate the effect of political sophistication on turnout and whether this effect differs in second-order national elections. Political sophistication is thought to influence turnout because the more sophisticated voters have access to more information about the electoral and the party system. In this paper, we start from the expectation that these effects should be even stronger in the context of secondorder national elections, where information about the stakes of the election is not readily available. We analyse citizens’ willingness to turn out to vote at different levels of government in Belgium and the Netherlands. The results show that a higher degree of political sophistication increases the probability to turn out at the national as well as the European level. Our expectation that this effect would be larger at the European level, however, is not supported by these results.

Feasibility considerations are important for political theories. An infeasible theory may contain valuable ideals, but is not suited to reality. Utilizing a precise and complete conceptualization of feasibility is essential to considering the feasibility of theories. An existing definition of feasibility is presented here. It is argued that this conception of feasibility is incomplete. The impact of parts of a theory as well as the feasibility of the parts of a theory themselves should be considered as well – something that has not been addressed in the literature thus far. A new feasibility framework is developed, in which the impact of parts of a theory, the feasibility of the parts of a theory as well as the feasibility of the preferred outcome of a theory have a place. Thus, the notion of feasibility is conceptualized in a more complete way.

This article studies the extent to which differences or inequality in policy congruence between higher and lower educated voters are moderated by policy domains. Instead of measuring inequality across all areas of policy, this study takes a policy domain-specific approach. The analyses are based on a dataset containing voters and party positions on 50 policy statements, gathered in the run-up to the 2009 regional election in Belgium largest region, Flanders. We find, overall, only small and unsubstantial, though significant, differences, in policy congruence between higher and lower educated voters, in favor of the former. However, we find a much larger representational bias towards higher educated when we look at transportation, culture and media, immigration, taxand budgetary policy, and economic policy. At the same time, differences in policy congruence are lower as regards spatial planning. Studying inequality in policy congruence across policy domains thus hides more complex patterns of representational bias.