Field Experiments on Citizen Participation in Elections and Referenda

Research question/goal: 

Participating in elections is a central element of democracies. Over the last decades, however, turnout rates have continuously declined in most democracies. At the same time, there are clear social inequalities in turnout: Citizens with low income and little education participate less often in elections.

Using a large-scale field experiment, our project studied whether a non-partisan, personal mobilisation campaign can increase turnout in Germany and decrease the social inequalities in turnout. In the run-up to the state election in Baden-Württemberg, we randomised voting districts in Mannheim and Heidelberg, and student assistants visited citizens at their doors in the experimental condition. Those districts were then compared to a control group with no canvassing. In addition to analysing this aggregate-level data, we conducted surveys of citizens both in the experimental group and in the control group, which allows us to examine the individual background of mobilisation. In a second wave of the survey, we analysed the long-term effects of the mobilisation for the German federal election in 2017.

Our results confirm that a personal, non-partisan mobilisation campaign in Germany can increase voter participation. Citizens in the experimental group turned out more frequently than citizens in the control group. We found significant differences both at the aggregate level of voting districts and at the individual level. Additionally, our survey analysis revealed that there were differing mobilization effects for various social groups. For instance, the effect on citizens with low education was twice as large as on citizens with higher levels of education. The results of the second wave of the survey point to long-term effects of a one-time mobilisation, as citizens who have been mobilised for one election are more likely to participate also in the next election.

The central finding of our project is that citizens can be motivated to vote when they are addressed personally. Non-partisan mobilisation campaigns therefore have the potential to increase turnout in democracies and, at the same time, decrease the social inequality in participation.

Fact sheet

Baden-Württemberg Stiftung
2015 to 2020
Data Sources: 
field experimental data, official voting statistics, survey data
Geographic Space: