A recent study is among the first to use “big data” to dig into how our sense of “ingroup and outgroup” – basically who we see as part of our group and who we don’t – plays a role in making content go viral.
The researchers put together a huge collection of Facebook and Twitter posts from both liberal sources like The New York Times and MSNBC, and conservative ones like Fox News and Breitbart. They also gathered over half a million tweets from Republican members of Congress, plus the same amount from Democrats.
They then used lists of politicians’ names, identity words, and dictionaries filled with positive, negative, and morally charged language to count how often these appeared in each post. They matched all this against how many times those posts were shared, retweeted, commented on, or reacted to.
For example, some viral tweets using outgroup language included conservative media posting things like “Every American needs to see Joe Biden’s latest brain freeze,” while Democratic politicians might tweet “Donald Trump has lied more than 3,000 times since taking office but Republicans refuse to say Trump is a liar.”
Looking at all the posts from politicians and media outlets on Facebook and Twitter combined: every negative word bumped up the chance of a post being shared by 14%, while every positive word actually lowered that chance by about 5%. Words with moral-emotional weight gave sharing a 10% boost per word.
Interestingly enough, words tied to your own political group didn’t really affect shares much. But if you threw in an outgroup term? That shot up the odds of sharing by 67% per word.
The numbers got even clearer when zooming in on US Congress members’ social media. Negative words increased shares by as much as 45% each time they showed up; meanwhile positive words dropped sharing rates between 2-5% each.
The use of ingroup terms still didn’t do much for getting more shares. But every single outgroup word-usually aimed at insulting or mocking someone-led to a jump in shares ranging from 65% up to a whopping 180%, no matter if it was naming a politician directly or just using general identity phrases.
“Viral content can help campaigns or social movements to succeed,” said study co-author Prof Jay Van Bavel from New York University. “But when hostile and hyper-partisan language is most likely to go viral, generating superficial engagement may ultimately harm politics and society.”
Van der Linden added: “Unless social media companies start penalising polarising content and rewarding more constructive posts, these platforms will continue to be swamped by political animosity that risks spilling into real-world turmoil. It may mean a radical rethink of their models for revenue generation.”
Rathje, Van Bavel, and van der Linden have also recently launched a research project allowing people to gauge the political slant of news shared by Twitter accounts – whether their own or other public feeds – as well as how reliable it is. The site includes “fake news scores” for all US Members of Congress.
