Data on Facebook's fake news problem

 In 2016 Presidential Elections, Behavioral Analysis, Competitive Analysis, Consumer Behavior, Featured, Popular, Site Analysis, Social Marketing, Tech News, Traffic Sources, Trends

“Only Facebook has the data that can exactly reveal how fake news, hoaxes and misinformation spread, how much there is of it, who creates and who reads it, and how much influence it may have,” wrote Zeynep Tufekci at The New York Times earlier this month after Mark Zuckerberg said that less than 1 percent of Facebook content is fake news.

A lot has been said about Facebook’s connection to fake and hyperpartisan news, which got us thinking about whether our data can help clarify the murky waters of how these stories are shared.

We analyzed the unique visitors to more than 20 known fake and hyperpartisan news sites, one known satire news site (The Onion), and three reputable sites (The New York Times, CNN and The Huffington Post) between September 1st and November 15th, 2016 to better understand Facebook’s role in sharing news articles.

What we found

Overall, we found that Facebook referrals accounted for 50 percent of total traffic to fake news sites and 20 percent of total traffic to reputable news sites.

The domain with the highest portion of Facebook referrals is Occupy Democrats, which has more than 4.6 million likes on Facebook and 79 percent of its total traffic coming through Facebook. American News closely follows in second with 78 percent of its total traffic coming through Facebook. The fake news site has about 5.5 million likes on Facebook, and recently made headlines when a story about Megyn Kelly made its way to Facebook’s Trending section several weeks after it was exposed as false. In comparison, 29 percent of The Huffington Post’s traffic comes from Facebook, while 20 percent of The New York Times’ traffic and 11 percent of CNN’s traffic comes from Facebook as well.

Below is a chart illustrating how we compared the top ten fake and hyperpartisan news sites with the most referral traffic from Facebook, three major reputable news sources (The New York Times, CNN and The Huffington Post), and one satire website (The Onion) to better understand Facebook’s role in sharing various news articles.

Jumpshot Facebook data

So who is reading fake news?

Our data clarified various theories about whether Facebook referrals are important to hyperpartisan and fake news sites, but we were also curious to see who clicks on and reads these news stories.

When looking across the U.S., the data showed that fake news readers are bipartisan as clicks to fake news articles through Facebook are equally popular in red and blue states, based on 2016 election data.

In terms of age demographics, millennials are 16 percent less likely to click on fake news from Facebook compared to the rest of the population. This begs the question, which age demographic is the most likely to click on fake news on Facebook? We found that the oldest age group analyzed, 65 and over, was the most likely to click on fake and hyperpartisan news.

Bottom line: Politics and opinions aside, the data indicates that Mark Zuckerberg’s statement that Facebook doesn’t have a “fake news problem” is unsubstantiated. Our data directly opposes his statement, as it revealed that fake news sites garner 2.5x as much of their total traffic from Facebook as do real news sites. The two domains with the largest amount of referral traffic from Facebook were, in fact, fake or hyperpartisan news sites. The media’s financial struggles have been widely recognized for some time, but the 2016 election shined a light on the quality (or lack of) reporting. Rest assured we’ll be doing some major fact checking of our own in the coming weeks.

Recent Posts
Jumpshot travel dataJumpshot 2nd-party data