“STOP THE AMOUNT!” President Trump tweeted November 5, the Thursday after the election. Twitter surrounded the tweet with warnings. Before the tweet was a label that said, ‘Some or all of the content shared in this tweet is disputed and may be misleading about an election or other civil process.’ It then offered a click-through link to learn more about U.S. election safety efforts. While the countdown continued, Twitter tagged four such tweets in a 24-hour period.
For months, Twitter has been identifying and contextualizing Trump’s misleading tweets, aimed at preventing misinformation from spreading on his platform. Such warnings may have been a victory for public relations for Twitter and the other social media companies that took similar steps in 2020, at least among some audiences. But have these attempts influenced public opinion on the matter itself?
We wanted to know if fact-checking of social media posts corrects the perceptions of the audience. Or does it flare up again – which could reinforce Americans’ beliefs in false or misleading allegations?
Our research suggests that corrections to presidential tweets may not work as intended – but rather further polarize Democrats and Republicans.
Scholars have investigated the fact-checking of politically incorrect information that citizens believe is effectively changing. Sometimes research finds that such refutations can catch fire again; individuals confronted with conflicting views double their beliefs. Other research suggests that correction corrections are relatively rare. But scholars have never investigated the consequences of labeling misinformation as president of the United States.
To study this, we conducted a survey experiment in May shortly after Twitter issued its first rebuttal of a presidential tweet. Under two Trump tweets, it was unfounded to claim that ballots were “materially fraudulent”, Twitter stamped: “Get the facts about ballots.” The link brought readers to a full fact check, with information refuting Trump’s claim.
We recruited a sample of 1,003 adult Americans through the online marketplace Lucid, which uses quota sampling to produce samples that reflect the demographics of the U.S. population as a whole.
We asked respondents if they felt that ballots were often fraudulent. Before we did that, we randomly assigned it in several groups. The first group conducted the survey without reading any additional information. The second group read the president’s tweet. The third read the tweet, as well as Twitter’s “get the facts” cuddle, which gently indicates that Trump’s allegations were inaccurate. The final group saw the tweet, the hug and the full fact check with rebuttal information.
In the group that read the tweet alone, Democrats and Republicans reacted differently. Compared to the group that did not read anything, Republicans were more likely to believe that e-mail fraud was common and that Democrats believed the claim slightly less. In addition, both the more modest pressure and the complete refutation further polarized biased responses: Democrats became less likely, and Republicans more likely to believe in voter fraud. In the real world, this polarizing effect can be even stronger if publicity about Twitter’s fact checking attracts more readers to the tweet than it would have seen without the correction.
Of course, our study only examined a single case – by definition, because it was the first time Twitter actually checked a presidential tweet. It will be interesting to investigate whether such contradictions always polarize opinions, or whether they differ under different circumstances.
During the waves of misinformation about the coronavirus and the presidential election in 2020, social media platforms were hampered by conflicting pressure. On the one hand, the Democrats put them under pressure to cage incorrect information on their platforms. And by contrast, Republicans have accused the social media platforms of intervening too aggressively and playing favorites.
These two camps turned up on Capitol Hill last month when Twitter executives Jack Dorsey and Facebook’s Mark Zuckerberg answered questions during a Senate Judicial Committee hearing entitled Breaking the News: Censorship, Oppression and the Election in 2020. Senator Marsha Blackburn (R-Tenn.) accused the platforms of “using this power to silence conservatives,” while Senator Dianne Feinstein (D-California) asked if Twitter alerts were even effective. Democrats believe these platforms have not moderated the content thoroughly enough. Republicans argue that platforms censor conservative thinking.
Our research suggests that the parties have inflated the impact of these corrections on public opinion. Instead of swaying public opinion, corrections to Trump’s posts could only further polarize the public anyway.
This column was originally published in The Monkey Cage blog at the Washington Post on November 25 Reprint with permission.
‘POV’ is an opinion page that provides timely comments from students, faculties and staff on various issues: on campus, locally, state, nationally or internationally. Anyone interested in submitting a piece, which should be approximately 700 words in length, should contact John O’Rourke at [email protected]. BU Today reserves the right to reject or amend submissions. The views expressed are those of the author only and are not intended to represent the views of Boston University.