“The Big Lie” and the Effects of Misinformation

The spread and severity of misinformation have grown dramatically in the United States, bolstered by Trump’s labeling of traditional media outlets as “fake news” (Oehmichen et al. Reference Oehmichen, Hua, Lopez, Molina-Solana, Gómez-Romero and Guo2019). Baseless allegations of voter fraud in the 2016 presidential election were levied by both right-wing pundits and Trump himself, who claimed that he won the popular vote “if you deduct the millions of people who voted illegally” (Wootson Reference Wootson2016). After assuming office, Trump created the Presidential Advisory Commission on Election Integrity, tasked with investigating voter fraud (Nelson Reference Nelson2017). Trump’s claims about voter fraud in the 2016 election established the framework for skepticism about the validity of the 2020 presidential election (Oehmichen et al. Reference Oehmichen, Hua, Lopez, Molina-Solana, Gómez-Romero and Guo2019).

Misinformation in the 2020 presidential campaign was in full force, with Trump telling his followers that the only way he could lose would be through voter fraud (Badger Reference Badger2020a). These lies espoused by Trump and right-wing pundits culminated in “The Big Lie,” which Trump voters bought into, and ultimately is credited with inspiring the Capitol insurrection on January 6, 2021 (Vachudova Reference Vachudova2021).

Misinformation spread online tactically caters to the political preferences of its viewers, driving polarization over conflict about what is the truth (Enders and Smallpage Reference Enders and Smallpage2019). This reflects a tenet of misinformation: it is commonly accepted by those whose preexisting beliefs, assumptions, and worldviews conform to the message (Swire et al. Reference Swire, Berzinksy, Lewandowsky and Ullrich2017). The supply of misinformation, therefore, is received by voters who interpret new information through the lens of motivated reasoning. In doing so, individuals accept new information in ways that they find desirable and reject evidence that contradicts their worldview. This renders partisans especially susceptible to believing misinformation that supports their political preferences, thereby substantially impacting their factual beliefs.

Presented with the same information, partisans interpret that information differently to support their own beliefs (Enders and Smallpage Reference Enders and Smallpage2019). When interpreting information such as the winner of an election, identity-protective cognition is used. This represents a way to avoid “dissonance and estrangement from valued groups” (e.g., political parties) that leads “individuals [to] subconsciously resist factual information that threatens their defining values” (Kahan, Jenkins-Smith, and Braman Reference Kahan, Jenkins-Smith and Braman2011, 149; Kahan et al. Reference Kahan, Peters, Wittlin, Slovic, Ouellette, Berman and Mandel2012, 733).

Identity-protective cognition made Trump voters more susceptible to Trump’s unfounded claims of voter fraud leading up to—and well after—the 2020 presidential election. Due to the prominence of misinformation catered to the political Right, surveys conducted before Election Day found that Republicans were more likely than Democrats to have concern abouts the integrity of the 2020 presidential election (Persily and Stewart III Reference Persily and Stewart2021). Furthermore, in a study on the effects of misinformation spread on social media, Berlinski et al. (Reference Berlinski, Doyle, Guess, Levy, Lyons, Montgomery, Nyhan and Reifler2021) concluded that exposure to any misinformation significantly reduced Republicans’ and Trump voters’ confidence in US electoral integrity—no matter how much they were exposed to it.

Therefore, we believed that most respondents who would support Trump resisting the results would justify their position on the grounds of perceived fraud in the election. Specifically, we predicted that most respondents would indicate that Trump would lose due to fraud involving the unprecedented use of mail-in ballots necessitated by the COVID-19 pandemic. Before Election Day, Trump claimed that this would be the only way he could lose to Biden, thereby warranting his resistance in his supporters’ view. These respondents thus would be supportive of Trump resisting the results due to misinformation about the integrity of the election spread from elites to the public.

Partisan Motivations for Undermining Election Results

In contrast to those who supported resistance due to concerns about the legitimacy of the election, some respondents may have supported Trump’s efforts to resist free and fair election results due to partisan “cheerleading” by purposely basing their responses on information they knew to be inaccurate (e.g., voter fraud and malpractice) merely from a desire to support their party (Bullock et al. Reference Bullock, Gerber, Hill and Huber2015). This form of expressive responding persists even when the information presented provides a clear truth. For example, when respondents were presented with photographs of Obama’s and Trump’s respective inauguration crowds, “the most politically engaged Trump supporters” falsely claimed that Trump’s inauguration photograph depicted a larger crowd. In doing so, they provided “expressive responses to a straightforward question related to controversy” in which the factual answer was clear (Schaffner and Luks Reference Schaffner and Luks2018, 142). These findings suggest that some Trump voters in our survey stated that Trump should resist disappointing results not because they necessarily thought the results were fraudulent but rather to signal their political support for him.

Partisan cheerleading, however, presents a novel issue in our survey. Fundamentally, Trump supporters’ engagement in partisan cheerleading in our study is antithetical to the principles of liberal democracy because those respondents expressed a willingness to undermine the democratic electoral process for partisan ends. Indeed, partisan support for undermining free and fair election results is a hallmark of competitive authoritarianism (see, for example, Levitsky and Way Reference Levitsky and Way2010; Levitsky and Ziblatt Reference Levitsky and Ziblatt2018). As such, it must be examined why partisans would support efforts to undermine the legitimacy of elections by refusing to accept credible results.

We suspected that some Trump supporters may have supported his efforts to resist defeat simply because he was their preferred candidate or because they are opposed to Democratic rule. The latter is an extreme case of negative partisanship and has proven to lead to public support for illiberal actions—particularly in comparative studies of executive aggrandizement (Cleary and Öztürk Reference Cleary and Öztürk2020). In their analysis of American voters serving as a check on undemocratic candidates, Graham and Svolik (Reference Graham and Svolik2020) found that when two candidates are ideologically opposed, voters are less likely to punish the candidate who aligns with their policy interests when they exhibit undemocratic behavior. Their findings clearly can be applied to the 2020 presidential election, given that both the candidates and the voters were more polarized in 2020 than in any election in recent history (Jacobson Reference Jacobson2021). Thus, in applying Graham and Svolik’s (Reference Graham and Svolik2020) methods to our survey experiment, we expected that some voters—rather than simply tolerate illiberal candidates—would actively endorse their preferred candidate’s undemocratic actions rather than tolerate opposition rule.

Specifically, we assumed that those who identified their partisan affiliation as “strong Republicans” would be significantly more likely to support Trump’s efforts at resisting the outcome than Independents or moderate Republicans who voted for Trump, given that their policy interests likely were more aligned with Trump than with those in the latter group. Therefore, the opportunity cost of not having Trump in office was higher for those voters than it was for more centrist voters. In their survey experiment, Graham and Svolik (Reference Graham and Svolik2020) found that moderate and centrist voters serve as a pro-democratic check on illiberal politicians, thereby confirming our expectation.

Those who supported Trump resisting the election results for partisan reasons likely would provide partisan explanations to our open-ended question. We expected that they would cite either that Trump was their preferred candidate and thus they wanted him to be in power no matter what or that they were highly opposed to Democratic rule and were willing to undermine the electoral process to ensure that such an outcome did not occur.

Should Trump Resist?

Figure 1 shows Biden’s randomized hypothetical popular-vote margin of victory against the percentage of Trump voters who responded that Trump should resist the election results if Biden won the Electoral College. About 45% opted for Trump to resist when Biden received a margin of victory between 1 and 12 points. Although support for resistance declined when respondents were given a scenario in which Biden won the election by 13 to 15 points, in total, 40% supported Trump resisting the election results. Even when respondents were given a hypothetical scenario in which Biden’s large margin of victory would make voter-fraud concerns especially irrelevant, there still was widespread support for Trump to resist the outcome.

Figure 1 Support for Trump Resisting the Results of a Biden Electoral College Victory Across Popular-Vote Margins

These findings were consistent with other polls conducted before the election that probed how Trump voters would respond to a Biden Electoral College victory. Of the scenarios we presented, the most reflective of reality was when Biden won the popular vote by 4%, given that he actually won by 3.9%. Among those presented with this then-hypothetical scenario, 44% responded that they would support Trump resisting the election results.

Despite a slight downward trend in figure 1, we failed to find a statistically significant relationship between the popular-vote margin and support for Trump resisting the election results (p=0.26). We also did not find a significant relationship between education, household income, and gender and support for resistance as shown in Table 1. Respondents older than 65, however, were significantly less likely to support resistance than younger respondents. Self-reported partisan identification also proved to have insignificant effects on support for resistance. To an extent, this contradicts Graham and Svolnik’s (Reference Graham and Svolik2020) theory because the application of their model would expect more centrist voters to not support undemocratic actions to the same extent as hyperpartisans. However, this difference may be the result of our model using partisanship as a measure of alignment with Trump’s policies, whereas their model used a linear continuum to measure policy alignments with a candidate. Furthermore, measurement of partisanship may not be a sufficient proxy for analyzing alignment with Trump’s policies, particularly because Independents and “leaners” typically are actual partisans in their policy preferences (Petrocik Reference Petrocik2009). No significant relationship was discovered between respondents’ interest in news and current events or their acknowledgment of racism and supporting Trump’s resistance.

Table 1 Trump Voters’ Support for Resistance of Election Defeat (Binary Logit Model)

Notes: Table entries are binary logit coefficients with standard errors in parentheses. The dependent variable is coded 0 for “Trump should concede defeat and commit to a peaceful transfer of power” and 1 for “Trump should resist the results of the election through measures such as discrediting the results as invalid, declaring a state of emergency, and/or taking any means possible to remain in office.” Self-reported news interest is coded as a continuous variable, with higher values indicating more engagement with the news and media. Acknowledgment of racism is a continuous variable for the extent to which respondents agree that “White people in the US have certain advantages because of the color of their skin,” with higher values indicating more agreement. ^***=Results are significant at the 0.01 level. ^**=Results are significant at the 0.05 level. *=Results are significant at the 0.10 level.

Why Should Trump Resist?

Table 2 shows the categories of responses to the open-ended question that asked respondents’ their rationale for support of Trump resisting the election results. The most common response was that Trump should resist because he was the respondent’s preferred candidate, which we labeled “Support Trump.” Examples of these responses include “[I] am saying this because I am for Trump 2020” and “because he has done more for this country in 4 years than Biden has in 47 years.” These respondents justified their support for Trump working to overturn the results simply because he was their preferred candidate.

Table 2 Trump Voters’ Reasons for Him to Resist Results of Potential Electoral College Loss

Note: Respondents could be coded under more than one category.

Similarly, opposition to the Democratic Party was common, with respondents perceiving Democrats as ideological radicals. “Democrats Are Radicals” was coded for those who suggested that a Biden presidency would undermine their perception of American traditions and values. These responses included general attacks on Democrats’ policies and governance, such as “Because the Democrats will ruin America” and “Democraps are destroying our nation”; allegations of Democrats supporting socialism, such as “I don’t want to live under Socialism and all that the Democrats stand for!”; and perceptions of Democratic governance serving an ethnocultural threat, such as “’cause we don’t want or need a woman of color as the VP!!!”

These two categories reflect hyperpartisanship for Trump and negative partisanship against the Democrats. For this group, the election appears irrelevant to the legitimacy of the presidency, with these respondents instead simply supporting resistance for partisan reasons. In total, this theme of supporting resistance because of hyperpartisanship or negative partisanship was present in 40% of responses.

Responses coded as “Election Irregularities” questioned the legitimacy of the election results without explicitly mentioning mail-in ballots or voter fraud, as well as those that alluded to the election being rigged, such as “there is no way Biden is gonna win the election if the election [is] fair” and “the Democrats are rigging the election.” The category “Voter Fraud/Mail-In Ballots” included respondents who mentioned mail-in ballots and cited personal stories speculating that they would be abused, such as “my experience, I’ve received 2 mail-in ballots and individuals have never lived here,” as well as statements arguing that these ballots would be used to commit fraud favoring Democrats. Despite being touted before and after the election—and being the basis for Trump’s challenge to the state election results—election irregularities, voter fraud, and issues with mail-in ballots were cited by only 37% of respondents who would support Trump in resisting the election results.

Responses coded as “Democrats Are Corrupt” included “the Democrat party is totally corrupt” and “I personally think there is a hidden agenda, I never thought that way until I saw the so-called [first] impeachment process and saw how absolutely corrupt the Democratic party is.” Those who levied ad hominem attacks against Biden and his capacity to govern but did not reference his policy positions were coded as “Biden Is Incompetent.” These included statements such as “because in reality Biden isn’t fit” and “because Biden is an idiot and there is no way he can lead this country as president.”

The variance in responses reflected conflicting rationales for supporting Trump’s resistance. Those who stated that Trump should resist because he was their preferred candidate or because they disliked the Democratic Party appear to be motivated by hyperpartisanship or negative partisanship. For those respondents, their support of Trump resisting the election results was not inspired by concerns of electoral malpractice; instead, they simply did not care for the election itself and wanted Trump in power no matter what. In contrast, those who cited reasons concerning the legitimacy of the election were inspired largely by misinformation—notably, The Big Lie—which likely led them to believe that the election would be rigged and fraudulent. Despite expecting issues concerning the legitimacy of the election to be the most common reason cited by respondents, we failed to find that it was cited more often than reasons guided by partisanship or negative partisanship (p=0.74).

As a result, we cannot conclude that support for resisting the election results was driven primarily by misinformation about election fraud. Instead, a significant percentage of respondents supported resistance for partisan reasons, indicating that the declining legitimacy of the electoral process was not caused by misinformation alone but also by a lack of tolerance for the opposition and hyperpartisanship.