For many citizens, online discussions are important for their political decision-making (Stromer-Galley, 2017). However, little is known about whether following and participating in online discussions motivate individuals to engage in other, more demanding forms of political participation such as protest (e.g., signing petitions) or institutional forms of engagement (e.g., donations; Theocharis et al., 2021). Particularly, it is unclear how different discussion environments in terms of the represented political attitudes affect further forms of participation – especially because usually studies only measure intentions to participate, but not real participative actions. Resorting to the Theory of Planned Behavior/TPB (Ajzen, 1985), we hypothesize that joining online discussions affect individuals’ attitudes, subjective norms, and perceived behavior control, which in turn influences the intention to participate, which, finally, affects actual participative actions (e.g., Ziegele et al., 2018).
For an empirical test of this hypothesis, a three-wave panel survey with two embedded experiments was conducted in 2020 with a German quota sample (N = 1,287). The respondents were exposed to online discussions on the questions of whether plastic packaging (PP) for fresh foods should be banned in supermarkets (wave 2) and whether genetically modified organisms (GMO) should be banned in food production (wave 3). In each wave, participants were randomly assigned to one of three discussion environments in which they were exposed to either (a) arguments for a ban, (b) balanced arguments, or (c) arguments against a ban (the manipulation was successful). After exposure, participants could sign a petition calling for a ban of PP (wave 2) or donate (parts of) their incentive to an organization that advocates for or against GMO (wave 3). Before the respondents could act, the central variables of the TPB were surveyed (attitude, subjective norm, perceived behavior control, intention). Socio-demographics and the central variables of the Civic Voluntarism Model (Verba et al., 1995) were used as control variables. The hypotheses were tested with logistic (wave 2) and linear hierarchical regressions (wave 3).
The results of the logistic regressions show that respondents who were exposed to arguments for banning PP were more likely to sign the petition than respondents who were exposed to arguments against PP. The results of the linear regressions show that participants who were exposed to balanced arguments donated more money to a pro-GMO organization compared to those who were exposed to arguments against GMO. Detailed results will be presented at the conference.
Our study indicates that following and debating in online discussions can have positive spill-over effects on other, more demanding forms of political participation. These “upstream effects” (Holbein & Rangel, 2020) emphasize the importance of political online discussions for citizens’ political decision-making. Furthermore, the direction of this impact seems to depend on the specific discussion environment in terms of the arguments citizens are exposed to (pro vs. contra vs. balanced). These observed effects can be well explained by the TPB. In the discussion of the findings, limitations such as the external validity of the experimental design and the potential influence of further political preferences must be considered.
