We examine the individual-level characteristics and political and economic conditions associated with talking about politics. To build our model of discursive engagement, we draw on existing research on political participation as well as our own theoretical reasoning. Our data cover two million individuals in 28 European countries over 45 years, and we employ a little-used approach to multilevel analysis that distinguishes variations in engagement attributable to cross-country differences from those stemming from within-country changes. Our primary findings reveal that elections, multiparty competition, economic recession, and unionization are positively associated with discursive engagement within countries but bear no observable relationship with discussion across countries. We also find that men, the higher educated, the middle aged, the employed, and students are less likely to eschew political discussion. Secondary analyses of narrower data sets indicate that, across countries, citizens are less likely to forgo political discussion where there is less income inequality and less corruption, while, within countries, political discussion is more likely in periods with higher ethnic diversity. Our approach is wide-ranging, but it is also deliberately correlational. Future observational and experimental studies might expand on and identify the causal underpinnings of the associations we establish here.

Existing research mainly analyzes mass attitudes towards the European Union (EU) from the national and individual-level perspective. This paper adds to this literature by focusing on the relationship between EU support and subnational economic conditions, using harmonized survey data covering 40 years and 1.1 million respondents in 197 European regions. Our analyses reveal important temporal dynamics in the relationship between EU support and subnational economic conditions. In particular, we find clear evidence of a spatial polarization of public opinion during and since the Great Recession, with the odds of EU support being significantly higher in economically resilient regions. During the years of the Great Recession, this spatial polarization emerged due to a sharp increase in euroscepticism in economically challenged regions. Our research indicates that this spatial disparity has widened since the Great Recession, not because of changes in public opinion in economically challenged regions but rather due to an increase in the odds of EU support among citizens living in economically resilient regions. Finally, our study reveals evidence of changes in public opinion that challenge received wisdom regarding the relationship between EU support and education. Specifically, our analyses suggest that, in the post-Great Recession period, the odds of EU support are higher among the lower-educated living in economically resilient regions than the higher-educated in economically challenged regions. We end by discussing the implications of these findings for future research on euroscepticism, the winners-and-losers-of-globalization thesis, and party competition.

In recent years, scholars of descriptive representation have paid growing attention to the issue of class. This paper contributes to this line of research by examining the educational (mis)match of elected officials and the citizens they serve. Using data from an original paired elite-mass survey experiment, the paper investigates whether judgements of democratic quality are affected by education-based descriptive representation. Our study reveals limited evidence in support of the idea that citizens’ and politicians’ democratic assessments are shaped by a norm of education-based descriptive representation. What we find instead is strong evidence of affinity effects where democratic judgments are influenced by whether descriptive representation, or the lack thereof, favors citizens and politicians based on their own educational background. An important exception though are citizens without higher education, whose assessments of democratic quality are unaffected by education-based descriptive representation. The paper ends with a discussion of the implications of these findings for existing and future research.