Misinformation is a growing concern among the public and policymakers. Yet we still lack a clear understanding of its political effects. As misinformation often contains politically motivated content that sends unambiguous signals in favor of or against a group, it may increase the salience of political identities and make individuals more oriented toward partisan goals. Consequently, misinformation may be effective at mobilizing voters. We assess this argument using novel descriptive and experimental data from Brazil. We find that, while misinformation is connected to online and offline engagement, this relationship does not follow the patterns anticipated by conventional wisdom. First, we analyze 4.1 million posts of Brazilian politicians across different social media platforms and observe that posts with misinformation indeed tend to receive more engagement from their social media audiences. However, two online survey experiments show that, when misinformation reaches the general public, it primarily lowers willingness to act in favor of the group targeted by the posts. All in all, misinformation may be more likely associated with silencing than with raising voices in public debates and elections.
The winner-loser gap in attitudes towards democracy and election fairness has been widely documented by scholarship across different contexts. At the same time, studies on partisan misperceptions and social desirability biases suggest that those attitudinal dynamics could be overstated by the use of direct survey questions. To address this puzzle, we focus on the case of the 2022 Brazilian presidential election, when a highly polarized environment provided favorable conditions for survey respondents to falsify their reported beliefs about the electoral process. We conducted a panel-survey with waves before and after the second round of the election asking direct survey questions on beliefs about electoral integrity, as well as a list experiment and an honesty nudge gauging insincerity in those attitudes. Our results confirm the winner-loser gap and suggest that direct survey responses are genuine even among politically motivated voters.