Newsroom

Believe it or not, when COVID-19 first hit the United States in 2020, opinions across the country about the virus were not particularly polarized. But as election season approached, something dramatically shifted. Suddenly opinions about COVID-19 were dividing politically, and social media was flooded with misinformation. What happened?

Trying to understand this phenomenon is what prompted Cornelia (Connie) Pechmann, professor of marketing at the UCI Merage School of Business, and Xiajing Zhu, Ph.D. marketing candidate, to dive into the data to seek answers. After years of research, their study went beyond merely looking at the polarization around COVID to explore how polarization relates to the spread of misinformation.

The resulting research, “Political Polarization Triggers Conservatives’ Misinformation Spread to Attain Ingroup Dominance,” was published online in the Journal of Marketing on June 16, 2024.

The Spread of Misinformation Differs Among Partisan Groups

Their initial idea looked at how misinformation evolved around the topic of COVID prior to the election and after election season started. “We didn’t find much difference in the beginning between Republicans or Democrats spreading misinformation,” says Zhu, “but as election season came, we began to see much more misinformation coming from the Republican side about COVID. I was curious why so many people were posting misinformation on social media. Why do we see so much more misinformation in 2020 compared to before the election?” Zhu wanted to understand the context of the “what, why, and who.”

Pechmann says Zhu is an expert at collecting big data. “We started by looking at Facebook pages to get some idea about what was happening,” says Pechmann. “Then we looked at research in psychology and found it showed that Republicans, i.e., conservatives, were primarily the ones spreading misinformation online. Researcher John Jost, professor of psychology at NYU, confirmed that for us as well. There were theories about this, but no one was sure why it happened.” It seemed to be a pattern, and more studies confirmed this was true. Zhu suggested misinformation and polarization were related. “The more we explored, the more we kept finding that pattern,” Pechmann says.

Interpreting Misinformation vs. Disinformation From an Outsider’s Perspective

Pechmann was fascinated with Zhu’s interpretation of the research results. “She’s from China,” says Pechmann. “They don’t have a two-party system in China. They don’t have a free press, so Zhu learned about our culture and our politics at the same time.”

Zhu adds, “The very first time I heard about misinformation, I had no idea what is a Republican or what is a Democrat. That helped me keep neutral. I don’t have a bias either way. I’m just curious why one side is more likely to spread misinformation than the other.”

Their research showed that, if not provoked, both conservatives and liberals are similar when it comes to sharing misinformation, and both sides agree as to what qualifies as misinformation as opposed to disinformation. “Disinformation means you purposely spread something that’s a lie,” says Pechmann. “If you spread something you know is a lie, that’s intentional. That’s disinformation. Misinformation is something you think is likely not true, but you’re not 100% certain about it. By not taking the time to verify it—because it confirms your bias, and you maybe want it to be true—and sharing it anyway, that’s misinformation. They think it is probably untrue, but they don’t know it is definitely false. That gray area gives them license to spread it.”

Republicans More Likely Than Democrats To Spread Misinformation When Provoked

The researchers say people most likely to spread misinformation are Republicans, i.e., conservatives, but mainly only under certain conditions. “They only do it when provoked,” says Pechmann. “That’s where they differ. Both liberals and conservatives have a baseline tendency to do this, but we see these big spikes only among the Republicans when they’re in a situation that feels like there is political conflict. When they’re in a competitive situation, Republicans want to win—almost at any cost.”

When Republicans feel like they’re in competition, they are more likely to spread misinformation that they acknowledge might be false, says Pechmann. “It can be anything that triggers them, like news headlines, for example, or events in the news, or a political campaign, an election. Election periods are highly polarized because one party will win, and one will lose. That’s when we see these big spikes in which Republicans are much more likely to spread misinformation than Democrats.”

Democrats don’t prioritize winning when it comes to spreading misinformation. “What Democrats value more than winning is equity, fairness and spreading the wealth,” says Pechmann. “They aren’t willing to do whatever it takes to win because that’s not their priority.”

The priority for conservatives is their party winning, she says. “Psychologists call it in-group dominance because they want their group to dominate. For conservatives, their number one priority is protecting their party, circling the wagons, party loyalty and hierarchy.” This is why Republicans spread misinformation more often than Democrats, Pechmann says.

Collecting Data From PolitiFact and Pew Research

Pechmann and Zhu eventually dropped the Facebook data and moved on to a new dataset called LIAR from PolitiFact, a nonprofit group that has been collecting data on political statements in the news. “A lot of academics have been using PolitiFact data, mostly in computer science to help train AI to identify fake news,” says Pechmann. “They identify political figures who make statements, and they record, if they can, their political party. Then they rate the statements from completely accurate to ‘pants on fire,’ as they call it.” The researchers pulled from PolitiFact’s huge datasets to see what was polarized and/or misinformative and what wasn’t. Pechmann says this was the hard part where Zhu’s skills working with data came to the fore.

Zhu and Pechmann classified their data to determine two things: “We coded the statements by topic; then we used Pew Research to find out which political topics were the most polarized,” says Pechmann. “For example, veterans benefits is not a polarized topic, whereas the environment is very polarized. Then we classified these same posts based on whether they had been rated as ‘pants on fire’—or lies versus truth.”

Polarization Index Confirms Findings

Pechmann and Zhu found that for neutral topics, there was no difference between Democrats and Republicans on providing misinformation, but for polarized issues, they found a large jump for Republicans on providing misinformation. To confirm their findings were correct, they “also looked at Federal Reserve Bank of Philadelphia data on polarization in the news because it turns out polarization affects financial markets and the stock market,” says Pechmann. “For example, the Federal Reserve Bank puts out their monthly polarization index to help economists predict exchange rates.”

“Again, we found the same pattern: When there was polarization in the news, as reported by this polarization index from the Federal Reserve Bank, we saw a spike in Republican officials sharing misinformation, whereas Democrats stayed stable. Democrats had a level of misinformation, but it wasn’t striking.”

Online Headlines More Polarizing Than in Neutral Print

Pechmann and Zhu also found that news headlines in print often varied wildly from headlines online for the same events. News headlines in print were less polarizing than the ones online, Pechmann says. “For example, when George Santos got kicked out of Congress for lying, the headlines online called it a ‘polarized vote,’ but it wasn’t polarized at all. In the print version, the headline was simply that he had been voted out. We also found that The Guardian had more neutral headlines compared to The Wall Street Journal for the exact same event. Basically, the news media is creating this polarization, especially online, and especially outlets like Fox News. That’s what’s riling up the Republicans to post misinformation.”

Experiment Shows Republicans More Apt To Share Polarized Headlines

In their experiments, Pechmann and Zhu shared both neutral and polarized political headlines with participants who were either Republican or Democrat and asked them if they would post political misinformation. “We found that if we gave them the neutral headline, there was no difference between Republicans and Democrats,” says Pechmann. “But if we gave them the polarized headline, Republicans were much more likely to say, ‘Yes, I will go ahead and post that misinformation,’ even though they were quite certain it wasn’t true but weren’t positive it was false.”

‘Us vs. Them’ Conservative Language Dates Back Nearly 100 Years

The researchers also pulled data from presidential speeches from 1929 through 2023 to look for the same pattern. “We wanted to make sure what we were finding wasn’t a Trump effect,” says Pechmann. “Zhu grabbed all the data and looked at speeches before elections versus after elections over decades. Presidential speeches are labeled. Our dataset went all the way up to President Biden, and what we found was the same pattern of competitive ‘us versus them’ language from conservatives. It’s been driving Republicans for nearly 100 years. This language is significantly less evident among Democrats.”

Can We ‘Dampen’ the Polarization Problem?

“What we find is that polarization is the cause of the problem,” says Zhu. “Ideally, we hope we can somehow dampen the polarization among politicians and in the media and social media. But we acknowledge there are huge marketplace incentives for them to remain polarized, so that’s why we recommend more resources for media to do fact-checking and urge them to strategically allocate their resources to highly polarized time periods.”

Polarization isn’t likely to go away anytime soon, mostly because it drives engagement on social media, but there may still be a way to reduce it. “What if we made it impossible to share posts that were flagged as misinformation?” asks Pechmann. “Because if winning is your number one priority, which is what many Republicans believe, then they will likely share that misinformation if it’s available and if it’s easy to share it.”

Beyond the Headlines: Critical Thinking Courses Are Key

Another idea to curb the sharing of misinformation is educating young people. “Education is key,” says Pechmann. “If we can teach them from a young age to ask questions and check their sources, that could really help. For example, my daughter’s required English class at UMass was changed to a course on misinformation. They were taught how to check their sources and how to know what a good source was versus a bad source, and how to read beyond the headlines. This was a first-year college course that everyone is now required to take. Maybe those courses should be taught in public high schools as well.”