Who spreads fake news?

During the 2016 US presidential elections, the phrase “fake news” emerged as a significant talking point in political conversations around the world. But how much it was actually disseminated during the campaign?

According to a new paper published in the journal Science Advances, the answer is: not a lot. Written by researchers led by Andrew Guess, a political scientist from Princeton University in the US, it says “the sharing of articles from fake news domains was a rare activity”.

It also says that Americans over the age of 65, and especially those who describe themselves as “very conservative”, were more likely than any other demographic to share fake news to their Facebook friends.

Both the supply and consumption of fake news during the campaign showed an “overwhelming” bias in favour of the Republican presidential candidate – and eventual winner – Donald Trump.

Guess described the age-related discovery as “our most robust finding”. It remained dominant even after potential confounding factors such as education, ideology, and partisanship were taken into account. 

“No other demographic characteristic seems to have a consistent effect on sharing fake news, making our age finding that much more notable,” he says.

The report notes that scholars and commentators have raised concerns about the implications of fake news for the quality of democratic discourse, along with the spread of misinformation more generally. 

As for assertions that such content had a persuasive impact that could have affected the election outcome, “the best evidence,” the researchers conclude, “suggests that these claims are farfetched”. 

Guess and colleagues admit the term “fake news” is amorphous and subject to interpretation. For the sake of this report, they settled on stories that contained “knowingly false or misleading content created largely for the purpose of generating ad revenue”.

“Given the difficulty of establishing a commonly accepted ground-truth standard for what constitutes fake news, our approach was to build on the work of both journalists and academics who worked to document the prevalence of this content over the course of the 2016 election campaign,” the authors write.

They also adopted the work of BuzzFeed journalist Craig Silverman, who took the lead in covering the “fake news” story as it developed, compiling lists of websites known to be purveyors of intentionally false election-related stories generating the most Facebook engagement. 

They were also careful to exclude websites that could be construed as partisan or hyperpartisan, rather than intentionally or systematically factually inaccurate.

Using this approach, the researchers combined multiple sources across the political spectrum to generate a list of fake news stories specifically debunked by fact-checking organisations.

Using a Facebook survey and web application as a primary source of data, they found that people visited the platform more often than they did outlets from which the fake news was originating, “suggesting a powerful role for the social network”.

With the permission of survey respondents and Facebook itself, Guess and colleagues accessed each person’s social media activity and share history.

They found that “the vast majority of Facebook users in our data did not share any articles from fake news domains in 2016 at all, and this is not because people generally do not share links”. 

“While 3.4% of respondents for whom we have Facebook profile data shared 10 or fewer links of any kind, 310 (26.1%) respondents shared 10 to 100 links during the period of data collection and 729 (61.3%) respondents shared 100 to 1000 links. Sharing of stories from false news domains is a much rarer event than sharing links overall.”