Skip to content
Breaking News Alert Secret Service Director: We Didn’t Put Snipers On The Roof Because It Was Kind Of Sloped

More Evidence Twitter Skews The Media’s Understanding Of What Americans Think


It’s a given that Twitter is a bubble. The size of that bubble is what’s worth reemphasizing, especially for the benefit of the media. According to a new Pew survey, only 22 percent of American adults say they ever use Twitter. That’s compared with 69 percent who say they use Facebook, and 37 percent who say they use Instagram.

It gets worse. Of that 22 percent, a full 58 percent say they use Twitter “less often” than once a day. Twenty-nine percent say “once a week,” 10 percent say “every few weeks,” and 18 percent say “less often.” Only 42 percent of the already small 22 percent are daily users. That number was 74 percent among Facebook users, 63 percent among Instagram users, and 61 percent among Snapchat users. (Similar to Twitter, only 24 percent of adults say they use Snapchat.) The Pew survey was conducted from Jan. 8-Feb. 7 among 1,502 adults.

Journalists love Twitter, and for some obvious reasons. But even if you’re conscious of the platform’s limited representative value, it’s difficult to avoid internalizing or perpetuating false narratives—both about actual news (see: Covington Catholic) and the national mood (see: Joe Biden).

After the press spent days mulling the fate of Biden’s potential candidacy—on Twitter, airwaves, and in print—polling found Democratic voters were less concerned. According to a Politico-Morning Consult survey, “Half of Democrats polled said the allegations from women that Biden made them uncomfortable by invading their personal space will make no difference in their vote. Twenty-nine percent said the allegations would make them less likely to vote for Biden.”

“The responses from male and female Democrats were nearly identical,” Politico reported. “And very few voters see the allegations as truly disqualifying. Only 14 percent say the stories make them ‘much less likely’ to vote for Biden.”

The disconnect between Democrats on and off social media was probed by Nate Cohn and Kevin Quealy in the New York Times this week:

The outspoken group of Democratic-leaning voters on social media is outnumbered, roughly 2 to 1, by the more moderate, more diverse and less educated group of Democrats who typically don’t post political content online, according to data from the Hidden Tribes Project. This latter group has the numbers to decide the Democratic presidential nomination in favor of a relatively moderate establishment favorite, as it has often done in the past… The relative moderation of Democrats who are not sharing their political thoughts on social media, and therefore of Democrats as a whole, makes it less surprising that Virginia Democrats tolerated Mr. Northam’s yearbook page. It makes it easier to imagine how Joe Biden might not merely survive questions about whether he touched women in ways that made them feel uncomfortable, but might even emerge essentially unscathed.

It also helps explain why recent polls show that a majority of Democrats would rather see the party become more moderate than move leftward, even as progressives clamor for a Green New Deal or Medicare for all.

Journalists who spend a good deal of time on Twitter hear disproportionately from partisans, and it naturally infects their outlook on the 78 percent of people who don’t use the platform. As I argued here, the result is an overrepresentation of divisive stories lacking in news value.

If Twitter is buzzing about a video of high school kids interacting with Native American activists on the National Mall, journalists breathlessly confer it with news value and mine it for clicks. It’s easy to see how a video like the first Covington Catholic clip, which functioned as partisan clickbait, would generate debate on Twitter. But that some portion of 42 percent of 22 percent of adults who use Twitter daily are interested in a video does not indicate it’s worthy of national news coverage. Pausing to consider that question is an important step.

Of course, Twitter has other disadvantages, (and some advantages too). But the more we’re all aware of the platform’s shortcomings, and the more we act accordingly, the healthier our coverage will be. It’s one thing for someone without a media megaphone to make the mistake of allowing Twitter to distort his perception of the country; it’s another thing when your perception of the country shapes the media’s depiction of it.