Gerhard Riener (University of Southampton)
Abstract: This study examines the effectiveness of different invitation methodologies in recruiting citizens for a citizen assembly in Austria, focusing on engaging those who feel politically disenfranchised and have a lower external locus of control. The assembly was designed to develop strategies for future crisis management, initiated by the federal government in response to the Corona pandemic. 14,999 citizens were selected randomly, blocked over federal states, and invited to participate in a one-day moderated discussion. 360 citizens who responded positively were selected to participate.
The experimental phase was divided into the invitation stage and the citizen assembly stage. During the invitation stage, three treatments—Voice, Experience, and Ideas—were randomly assigned across different demographic groups. These treatments varied in cognitive demand, the ideas treatment (which is standard and the baseline of our study) asked citizens to bring their ideas for future crisis policies, while the experience treatment just asked for previous experience and the Voice treatment only emphasizing the importance of each individual's perspective, highlighting the inclusiveness of the event. The Voice treatment increased participation by 8% compared to the baseline, moreover those subjects had a lower external locus of control.
Data was collected before the treatments, during the assembly, and through a reference survey. The primary outcome variables were the acceptance of the invitation to the assembly and a measure of locus of control. Participants responded to a detailed questionnaire at the beginning, end, and two weeks after the assembly, providing insights into trust in institutions, polarization on Corona measures, and concern about polarization. Specific variables were collected at different stages to provide a comprehensive understanding of the participants' perspectives.
An additional online and telephone survey was conducted among a different representative sample of the Austrian population (2000 participants), mirroring the demographics and locus of control questions of the assembly participants. Here we find that the survey participants also had a smaller locus of control than assembly participants in all treatments but that the “Voice” treatment could lower this gap.
In summary, this study provides valuable insights into citizen engagement in political processes, highlighting the impact of communication strategies on participation. The findings can inform future citizen assemblies and participatory democratic exercises on the issues of representativeness on relevant unobservable variables.