New research published in the Proceedings of the National Academy of Sciences provides evidence that voters in Georgia who embraced Donald Trump’s claims of widespread election fraud were less likely to cast their ballot in a pivotal runoff election.
In the aftermath of the 2020 election, Trump made a series of baseless allegations that the election had been stolen from him. These claims were quickly debunked by election experts, but Trump continued to push these conspiracy theories in an attempt to undermine the legitimacy of Joe Biden’s victory. The study authors were interested in exploring how this rhetoric might have impacted the 2021 runoff elections in Georgia, where Democrats were able to flip two U.S. Senate seats.
“Trends in U.S. politics are such that Democratic and Republican politicians are both vocally suspicious of their opponents’ respect for the ‘rules of the game’ in our democracy,” said study author Jon Green, a senior research scientist in the Network Science Institute at Northeastern University, and a fellow at the Shorenstein Center on Media, Politics and Public Policy at the Harvard Kennedy School.
“To be clear, these suspicions come with very different levels of empirical support. These conspiracy theories about widespread voter fraud on the part of Democrats really are baseless. But that doesn’t change the fact that both parties’ core voters are hearing from their leaders that the other side is trying to take away their ability to vote their preferred candidates into office – either by making it harder to vote, loosening the relationship between votes and outcomes (through gerrymandering, e.g.), or through fraud that ignores legally cast votes altogether.”
“And there are debates within the parties about whether this kind of rhetoric is counterproductive,” Green continued.” People who work in Democratic politics, for instance, are sometimes worried that Democrats’ allegations that Republicans are engaging in voter suppression could discourage potential Democratic voters from voting, on the margins, because it could create the impression that voting will be more difficult or inconvenient.”
“During the Georgia runoffs, we saw local Republican activists raising similar concerns about these fraud allegations – that they were discouraging Republican voters from turning out because they were introducing uncertainty into whether voting was going to have any relationship with the runoffs’ outcome. So this question of whether there are any behavioral implications of this kind of elite rhetoric was what sparked my initial interest in the specific research question.”
The researchers analyzed Twitter data and voter records to explore the link between the endorsement of Trump’s election fraud claims and voter turnout in Georgia’s 2021 runoff elections. The sample included 45,431 Twitter users who had a voter record in the state of Georgia.
The study authors evaluated whether the Twitter users had liked or retweeted content supporting or detracting from election-fraud conspiracy theories. They also evaluated whether the users had liked or retweeted posts by the official Twitter account of then-President Donald Trump, and whether they had engaged with other prominent accounts that posted content supporting or opposing election-fraud conspiracies.
Green and his colleagues found that online engagement with content detracting from election-fraud conspiracy theories was associated with higher turnout than expected in the runoff election. On the other hand, those who liked or shared tweets supporting fraud-related conspiracy theories were slightly less likely to vote.
“We can’t say for sure whether the ‘Big Lie’ cost Republicans control of the Senate, but our study provides some evidence that it didn’t help their candidates in the Georgia runoffs,” Green told PsyPost. “All of our findings surprised us in the sense that there were good reasons to expect different results and we really weren’t sure what we’d find until we analyzed the data.”
The researchers proposed several mechanisms by which widespread claims of election fraud could affect turnout. They had theorized, for instance, that engagement with 2020 election conspiracy theories on Twitter could possibly help to boost Republican voter turnout due to their potential to elicit anger and strengthen partisan ties.
“It was reasonable to expect that endorsing election conspiracy theories would have been associated with higher turnout in the runoffs because the conspiracy theory made people angry, and prior work has linked anger to things like party loyalty and intentions to vote. So if we had observed that positive relationship, it would have made sense,” Green explained.
“However, translating anger into action typically requires it to be paired with efficacy – a belief that you can do something to address the threat you perceive. Here, a central premise of the conspiracy theory – that election results in Georgia are not directly based on the votes its citizens cast – likely reduces efficacy, so the results we found also make sense.”
Another proposed mechanism was negative evaluations of co-partisan candidates. In other words, voters who bought into election-fraud conspiracies might have been more likely to see Republican Senate candidates as being insufficiently supportive of Trump.
“It’s also possible that the way the Republican Senate candidates campaigned turned these voters off,” Green explained. This election determined control of the Senate when Democrats had already won control of the House and presidency. If you’re a Republican running for Senate in that situation, you typically want to appeal to the (small but non-trivial) slice of the electorate that sincerely prefers divided government. You want to pitch yourself as the last check against a Democratic trifecta.”
“But if you say this as a Republican candidate in the Georgia runoffs, you are implicitly acknowledging that Joe Biden is going to be the president. That is, you’re behaving as if the election was not stolen. If you’re really invested in the idea that the election was stolen and we need to do everything in our power to correct this obvious injustice, then why bother turning out to vote for a candidate like that? Again, we don’t know if that’s what drove our finding, we just can’t rule it out.”
The findings are in line with press reports, which claimed that some Trump supporters in Georgia were intent on “boycotting” the runoff election because of alleged fraud. But the observational nature of the study prevents the researchers from making strong causal claims.
“The big caveat is that because this is all observational, we can’t nail down why we see the relationships we see, and we can’t completely rule out that the turnout behavior is being driven by something other than the conspiracy theory,” Green said. “We don’t have causal leverage, and we spend some time in the paper talking about why causal inference is especially hard in a case like this. We also talk about a few different causal mechanisms that we see as plausible, and I’ve alluded to some above, but we can’t say for sure which mechanism(s) are producing the relationships we see.”
Trump continues to claim that the 2020 election was stolen from him, but whether this rhetoric will help Democrats in other elections remains to be seen. “In terms of questions that still need to be addressed, the big one is that we don’t know how generalizable our findings are,” Green said. “Georgia’s 2021 runoff elections were unique for a couple of reasons, so it’s unclear whether we’d expect the same results given overlapping-but-not-identical conditions in another context in the future.”
“I think it’s really important to keep in mind that conspiracy theories undermining faith in elections’ outcomes are bad regardless of any relationships they might have with voting behavior in the short term,” the researcher added.
The study, “Online engagement with 2020 election misinformation and turnout in the 2021 Georgia runoff election“, was authored by Jon Green, William Hobbs, Stefan McCabe, and David Lazera.