Tweets, Electoral Fraud, and Political Violence: Evidence from Two Experiments

• In Event: Experimental Studies of U.S. Voters

Abstract

Using experimental data, this paper investigates how and to what extent public confidence in elections can be influenced by statements made by political elites. Respondents were recruited on Amazon Mechanical Turk and were randomly assigned to actual tweets by prominent politicians and then asked about their perceptions of electoral fraud in two different experiments. The first experiment was conducted in the context of the 2018 Florida Senate Race which was in dispute for 12 days while a statewide manual recount triggered by a margin of less than one quarter of one percentage point. In this case, both Senate candidates and President Trump made statements about the integrity of the election. The experiment was repeated following the 2020 presidential election when Trump again made statements about vote rigging without any evidence. The findings demonstrate that a single tweet can influence perceptions of electoral fraud and that these perceptions can influence attitudes about violence. These results suggest that confidence in elections can be easily undermined which can create an atmosphere where violence is condoned by far more than a small group of radical activists.

Author

• Jeffrey A. Karp, Brunel University London