Tech | Visa | Scholarship/School | Info Place

False information works, 80% of false information in 2020 was sent by a small number of social “super sharers”

A pair of studies published Thursday in the journal Science offer evidence not only that misinformation on social media can change people’s minds, but also that a small group of devoted “super-sharers” — mostly older Republican women — were responsible for the vast majority of fake news during the study period.

The research was conducted independently but complementary by researchers at MIT, Ben-Gurion University, the University of Cambridge, and Northeastern University.

In an MIT study led by Jennifer Allen, researchers noted that misinformation is often blamed for vaccine hesitancy in 2020 and beyond, but the phenomenon remains poorly documented. This is understandable: Not only is the data from the world of social media vast and complex, but the companies involved are reluctant to participate in research that might paint them as prime vectors of misinformation and other data warfare. Few doubt that they are, but that’s not the same as scientific validation.

The study first showed that exposure to vaccine misinformation (the researchers collected data in 2021 and 2022), especially any information claiming negative health effects, does reduce people’s willingness to get vaccinated. (Previous research has shown that willingness is associated with actual vaccination.)

Second, the study showed that articles flagged as misinformation by moderators at the time had a greater impact on vaccine hesitancy than unflagged content — so, well done on flagging. But the fact is that there is far more unflagged misinformation than flagged content. So while each article has a smaller impact, its aggregate impact is likely much larger overall.

This kind of misinformation, they clarified, is more like misleading information published by large news outlets that falsely portrays risks or studies. For example, who remembers the Chicago Tribune headline “A healthy doctor dies two weeks after getting COVID-19 vaccine; CDC investigating why”? As the magazine’s commentators pointed out, there was no evidence that the vaccine had anything to do with his death. Yet, despite being deeply misleading, it was not flagged as misinformation, and the headline was subsequently viewed about 55 million times—six times more than all flagged material combined.

The data shows that the amount of unlabeled misinformation far outweighs the labeled reports.
Image Source: Allen et al.

“This contradicts the common belief that fake news on Facebook is contributing to low vaccination rates in the U.S.,” Allen told TechCrunch. “It’s possible that Facebook user numbers are associated with low vaccination rates (as other studies have found), but it may be this ‘grey area’ content that’s causing the effect — not the outrageously false content.”

The findings suggest that while suppressing obvious false information is beneficial and justifiable, it is ultimately just a drop in the ocean of the chaotic environment that social media users are operating in at the time.

So who are the swimmers who spread the most misinformation? It’s a natural question, but one that went beyond the scope of Allen’s research.

In a second study released Thursday, a multi-university team came to a pretty shocking conclusion: 2,107 registered U.S. voters spread 80% of the “fake news” (their term) during the 2020 election.

That’s a big claim, but this study cuts through the data pretty convincingly. The researchers looked at the activity of 664,391 voters matched with active X (at the time, Twitter) users and found that a small group of them accounted for a large share of the spread of false and misleading information.

These 2,107 users (with the help of algorithms) exerted a huge network effect in promoting and sharing links to political fake news. Data shows that 1 in 20 US voters follows one of these super sharers, which puts them far ahead of the average user. On a given day, about 7% of political news links to plausible news sites, but 80% of those links come from these few people. People are also more likely to interact with their posts.

Yet these aren’t government-backed “plants” or “bot farms.” “A large number of super-sharers’ tweets do not appear to be automatically generated, but rather generated through manual and sustained retweets,” the researchers wrote. (Co-author Nir Grinberg clarified to me, “We cannot be 100% certain that super-sharers are not sock puppets, but from using state-of-the-art bot detection tools and analyzing temporal patterns and app usage, it does not appear that they are automatically generated.”)

They compared super-sharers to two other groups of users: a random sample of users and those who shared the most non-fake political news. They found that these fake news spreaders tended to belong to a specific demographic: older, female, white, and overwhelmingly Republican.

This graph shows the demographics of super sharers (purple) compared to others (grey, entire panel; yellow, non-fake news sharers; magenta, average fake news sharers)
Image Source: Baribi-Bartov et al.

Compared to the average share of the panel, only 60% of the super-sharers were women, and were significantly more white than the already majority white group, but not dramatically so. They were older (58 years old on average, compared to 41 for all), and about 65% were Republicans, compared to about 28% of Twitter users at the time.

The demographics are certainly illuminating, but remember that even the vast majority are not the majority. Millions of people retweeted the Chicago Tribune article, not 2,107. Even the super-sharers, the Science commentary notes, “are diverse and include political pundits, media celebrities, naysayers, and anti-vaxxers who spread discredited content for personal, financial, and political motivations.” That’s not true. only Older women in red states, though, do feature prominently. Very prominently.

As Baribi-Bartov et al. pessimistically conclude: “These findings highlight the vulnerability of social media to democracy, where a small minority distorts the political reality for the majority.”

This brings to mind Margaret Mead’s famous quote: “Never doubt that a small group of thoughtful, dedicated citizens can change the world. In fact, it is the only thing that can change the world.” Somehow I doubt if this is what she really meant.

#False #information #works #false #information #small #number #social #super #sharers

Leave a Reply

Your email address will not be published. Required fields are marked *