The Danger of Fake News to Our Election

One of the biggest questions everyone asks about fake news is whether it actually affected the 2016 US Presidential election. Congress and the FBI have been investigating this, and concluded (so far) that fake news was intended for that purpose. No one has suggested that people didn’t vote their minds, so without a doubt, the election was legitimate. But did fake news change people’s minds?

This question is harder to answer. To try, we need to find out two things: First, how many people have been exposed to fake news and, second, how many of them changed their opinions in reaction to it?

Exposed?

We know from its CEO, Mark Zuckerberg, himself, that 126 million Americans were shown Russian-backed, politically-oriented fake news stories via Facebook during the 2016 US presidential election campaign. Not everyone who was presented with fake news headlines on Facebook necessarily read the stories, or consciously perceived them, or was affected by them. However, even if only a fraction of those who were shown the stories saw them and even if an even smaller fraction opened the stories, and if some of those people were influenced by them, then—because the election was so close in some of the swing states—it could have changed the voting choice of enough people to affect the election. All hypothetical, but a real potential.

It wasn’t just Russians paying for Facebook users to see fake news that could have played a role. Another figure that raises cause for concern is that leading up to the 2016 Presidential election, the 20 most popular fake news stories received more shares, reactions, and comments (8.7 million engagements) than the most popular 20 real news stories (7.3 million engagements) [1]. Again, these numbers have to be taken with caution. The top 20 stories are only a small fraction of all new stories. Moreover, most of the engagement with fake news stories came from a small fraction of Americans: 10% of Americans accounted for 60% of the traffic on fake news sites.

To sum up the answer to our first question: Many people were exposed to fake news just before the 2016 election but it is difficult to estimate how many people actually saw it and whether they were influenced by it.

Influenced?

In the aftermath of the 2016 elections as the extent of fake news became clear, three scientists from the Ohio State University—Gunther, Beck, and Nisbet—explored whether people who might have changed their votes from Democrat to Republican were affected by fake news [2]. Why these particular people? The scientists reasoned that, like most people, those who voted for Obama in 2012 would probably have voted for Clinton in 2016 unless something changed their minds.  So they found a group of voters who voted for Obama in 2012. Of these Obama voters, only 77% of them voted for Clinton in 2016; 10% voted for Trump, and the others didn’t vote or voted for someone else.

The researchers asked the voters how much they believed in three statements, each of which, according to independent analysis, had been promoted by fake news but were actually false: that Hillary Clinton was in poor health due to a serious illness, that Pope Francis endorsed Donald Trump, and that, during her time as Secretary of State, Hillary Clinton approved weapon sales to Islamic jihadists including ISIS.

Although most people didn’t think that these statements were true, there was a very strong link between the belief that they were true and the way people voted:

“Among those who believed none of the three fake news stories, 89 percent cast ballots for Hillary Clinton in 2016; among those who believed one fake news item, this level of electoral support fell to 61 percent; but among those who had voted for Obama in 2012 and believed two or all three of these false assertions, only 17 percent voted for Clinton.”

And if these percentages of voters—even 17% of former Obama supporters—had voted for Clinton in the states that went from Blue to Red in 2016, instead, it would have changed the outcome of the election.

Again, it’s important to be cautious interpreting these results. Just because these people said that they thought these statements were likely to be true doesn’t mean they actually saw the fake news stories before the election. So, it doesn’t prove that they crossed sides because they’d been exposed to fake news. But it may be as close as we can get to identifying if fake news affected voters.

But there’s also reason to believe that it wasn’t fake news, but good old fashioned clickbait, that helped sway voters. A WIRED article by Antonio Martinez explains how the Trump campaign got more ads on Facebook, cheaper, than the Clinton campaign did [3]. Facebook charges advertisers different rates depending on (a) the location of and competition for users, so that urban Facebook users (primary targets for the Clinton campaign) cost more to advertise to than do rural voters (who the Trump campaign sought), and (b) Facebook also prioritizes bids for ads that are seem more likely to generate a click, a like, or a comment. “Because Trump used provocative content to stoke social media buzz,” according to Martinez, his campaign was able to win the bidding contest to place more ads for more people online. This explanation seems to have merit, too.  Look at the section of our website about the way analytics help target ads to reach certain types of social media users, and how humans and bots pass along certain messages, you can see how powerful these strategies may have been.

Some Take-Aways

There are several important take-aways here. First, it’s hard to say how big the impact of fake news was on the 2016 election. The Ohio State study strongly indicates that when people believe fake news, it affects their votes. So when elections are close, fake news can impact who ultimately wins and who loses.  Finally, whether it was through fake news or provocative advertising, the use of social media seems to be able to affect elections differently than we’re used to seeing. The economics, computer algorithms, and social dynamics that make up social media platforms are doing things to politics that traditional media cannot. 

 

References

[1]       D. Kurtzleben, “Did Fake News On Facebook Help Elect Trump? Here’s What We Know,” NPR.org, 11-Apr-2017. [Online]. Available: https://www.npr.org/2018/04/11/601323233/6-facts-we-know-about-fake-news.... [Accessed: 03-Aug-2018].

[2]       R. Gunther, P. A. Beck, and E. C. Nisbet, “Fake News Piece for ‘The Conversation’ with Methodological Appendix,” 15-Feb-2018. [Online]. Available: https://cpb-us-west-2-juc1ugur1qwqqqo4.stackpathdns.com/u.osu.edu/dist/d.... [Accessed: 03-Aug-2018].

[3]       A. G. Martínez, “How Trump Conquered Facebook Without Russian Ads,” Wired, 23-Feb-2018. [Online]. Available: https://www.wired.com/story/how-trump-conquered-facebookwithout-russian-.... [Accessed: 02-Aug-2018].