Study: Trump’s tweets can lead Republicans to lose faith in elections

Rep. Jim Jordan (R-OH) stands with dozens of people at a “Stop the Steal” rally on November 5 at the Pennsylvania State Capitol. | Spencer Platt/Getty Images

Trump supporters indicated a “decrease[d] willingness to accept election results peacefully.”

President Donald Trump’s tweets attacking the legitimacy of the 2020 presidential election might not sway the outcome — but they might sway Americans’ faith in democracy.

A study by political science researchers from Stanford and five other universities found that exposure to those tweets “erodes trust and confidence in elections and increases the belief that elections are rigged among his supporters.” However, among those who oppose the president, the study found that their trust in elections actually increased after seeing his tweets, albeit by a slightly smaller magnitude.

The study’s survey, which was conducted before the election (from October 7-24), presented roughly 2,000 participants from both political parties, as well as some independents, with a variety of Trump tweets about election integrity and other topics. The methodology gets a little complicated from there, but in essence, people then answered questions about their faith in democratic processes and their emotional reactions to the president’s statements.

Because the participants were shown the tweets outside the usual context of news commentary or a larger Twitter feed, it’s hard to know how other signals — from Twitter labeling the president’s tweets as being “potentially misleading” to other behavior by elected officials — could affect their responses. But it’s clear the president’s rhetoric impacted at least their self-reported beliefs.

“We’ve never had a president who is attacking the legitimacy of an election at the level that Trump is,” said Stanford University researcher Katherine Clayton, the study’s lead author.

Over the last eight years, Trump has attacked mail-in voting or made allegations of election fraud more than 130 times on Twitter, according to a September Wall Street Journal review. Around 66 percent of those tweets were posted this year.

Since the election was called for Democrat Joe Biden, Trump has intensified his attacks on the legitimacy of the process, and many congressional Republicans have fallen in line and bolstered his claims about widespread fraud and the appropriateness of legal action. For instance, Sen. Roy Blunt (R-MO) told reporters that Trump “may not have been defeated,” even as most major media outlets projected Biden as the winner. More concerning is Senate Majority Leader Mitch McConnell (R-KY), who legitimized Trump’s legal challenges when he said on the Senate floor that “President Trump is 100 percent within his rights to look into allegations of irregularities and weigh his legal options.”

Comments like these could be cementing Republican voters’ loss of faith in elections, Clayton said. But even if other GOP lawmakers push back against the president’s claims (as some have), “It will come down to [whether they’re] a Trump supporter or a Republican more strongly,” she added.

To be sure, people’s self-reported views may have less to do with their actual opinions and more to do with staying in line with their party. Previous research has shown that survey respondents often follow partisan cues: Politico, for instance, found that Republicans’ and Democrats’ views on whether the economy was improving flipped after Trump’s 2016 win, but that those shifts in reported attitudes only sometimes affected people’s actual behavior.

With regard to the Stanford study, this makes it hard to know whether the participants’ increased distrust in US elections could turn into action. Overall, researchers determined that support for political violence was unaffected by Trump’s tweets. Responses to survey questions asking when it’s okay to “send threatening and intimidating messages” to members of the opposing political party; whether it’s justified for members of their own party to “use violence in advancing their political goals,” and when it’s okay to “harass an ordinary [member of an opposing party] on the internet in a way that makes [them] feel unsafe” did not change after exposure to Trump’s Twitter statements.

However, Trump supporters indicated a “decrease[d] willingness to accept election results peacefully” when asked how strongly they agree or disagree with the idea that “sometimes regular people need to be a little violent to make sure votes are counted correctly.”

“I would almost think that a survey measure like that could understate the amount of violence people would be willing to engage in,” Clayton said. People aren’t predisposed to confess their readiness to engage in violent acts, so when research finds that people are willing to admit to it, it’s worth paying attention.