The Effect of the Partisan Supply on Electoral and Protest Participation: A Cross-national Vignette Experiment

In this study, we make use of a factorial survey experiment to examine the effects of the political context on choices individuals make with regards to participating in different forms of politics. We start from the assumption that the choice to take part in both electoral and protest politics depends on the partisan supply and individuals’ predispositions. More specifically, we primarily designed the vignette to test the hypothesis that right-wing individuals generally are more likely to opt for protest when there is no political party reflecting their point of view, whereas left-wing people are more likely to be active in both arenas when their demands meet a partisan supply. Thus, the effect of the presence of a partisan ally should be mediated by the ideological orientation of individual respondents. This argument is based on the differing value orientations and strategic considerations of the left and right when it comes to combining electoral and protest politics. The vignette allows us to study the effects on two forms of (hypothetical) participation in electoral politics (voting and contacting a politician) and two forms of protest politics (street demonstration and signing a petition). We opt for an experimental survey design (a vignette experiment within a large scale survey) because it permits testing of causal assumptions. In addition, the factorial survey forces the respondent to make a judgment on the vignette based on a trade-off of several dimensions. Thus, the description of a situation in the vignette resembles real life more closely than single-item measures in a survey. Also, we can test whether the effects of the manipulations in the vignette depend on personal characteristics of the respondent. In our vignette, we evoke a grievance by first measuring the respondent’s preferences on certain policy areas (refugees, taxes, environmental protection). Next, we start the vignette by asking the respondent to imagine a situation in which the government is planning a new policy that goes against her/his preference. The vignette manipulates whether there is a political party that also opposes the government plans (just as the respondent) to test our main interaction hypothesis. Moreover, we manipulate whether it is election time, whether a group of citizens protests against the new policy by organizing a street demonstration or a petition, and the expected number of participants in these activities. The survey data include not only left-right self-placement but also scales for libertarian-authoritarian and populist attitudes. The original data comes from an online survey among 3,000 respondents in nine new and old democracies (the UK, the Netherlands, Germany, Switzerland, Hungary, Romania, Greece, Argentina, and Brazil, N = 333 per country). The country sample allows us to test our hypotheses beyond the context of Western Europe. Data will be analyzed using regression analysis.