President-elect Donald Trump and Gen. John Kelly emerge from the clubhouse following their meeting at Trump International Golf Club on Nov. 20 in Bedminster Township, N.J. (Drew Angerer/Getty Images)

The presidential election has left emotions raw, relationships strained  and political conversation fraught. Just ask the millions of Americans who sat down to Thanksgiving dinner last week and spent the entire meal trying to avoid talking politics with relatives.

Things have gotten so bad in the wake of the election, according to several news reports, that users on social media have been cutting ties with people whose political views they disagree with. One news story called it an “unfriending frenzy.”

But how often does “political unfriending” occur? And who are the people most likely to drop someone from their social network because of their political views?

My recent research sheds new light on these questions.

I look into this with data from the Pew Research Center. In 2012, Pew conducted a survey in which people were asked about their use of social networking sites, and in particular whether they had ever “unfriended” someone for political reasons. Overall, only 18 percent of respondents said that they had unfriended someone because of politics.

Interestingly, the most common reason for unfriending someone over politics was not the content of their posts — it was because they were posting too much. About 10 percent of respondents said they had dropped someone because they were posting too frequently about politics. That poster who can’t help but offer a political rant every five minutes is a prime candidate for unfriending. There’s probably someone from your own news feed that comes to mind.

About 4 percent of respondents reported unfriending someone because the person had disagreed with them. Another 9 percent said they had dropped someone who they had disagreed with. Other reasons were that the person had argued with them (8 percent) or worried that they would offend their friends (5 percent).

Few people are purging their news feeds of unfriendly opinions.  

All told, political disagreement does encourage unfriending, but it is relatively rare, occurring only among a minority of users. Despite concerns about social media “filter bubbles,” where users only see confirming information, this is one reason that most people report seeing a mix of political perspectives on social media. Few people are purging their news feeds of unfriendly opinions.

Of course, a more recent survey by Pew notes that when you include all ways of opting out of information on social media — not just unfriending, but also muting, blocking, hiding, or otherwise changing settings — this number of people having done so grows from 18 percent to 39 percent. But again, a majority of users report never having used these tools to avoid political posts.

But who is doing the unfriending that does occur? If it’s people who are uninterested in politics, that would suggest they may be using it as a way to opt out of politics altogether. People more interested in politics, on the other hand, are still likely to see political information in contexts outside of social media, like talking with friends and using traditional media, even if they cut ties on social media.

I use the frequency with which people discuss politics with others as a rough proxy of general political interest, and then compare the frequency of political unfriending among more and less interested people.

The most politically engaged are the most likely to “unfriend” for political reasons

In fact, it is the most politically engaged who are most likely to unfriend someone for political reasons. Even after controlling for how much political content shows up in a user’s feed, I find that it is not the politically apathetic who are cutting ties because of political posts.

For example, just 2 percent of people who never talk about politics report having unfriended someone for political reasons. The numbers are significantly higher among people who say they “sometimes” (8 percent) or “very often” (10 percent) talk about politics.

We might further expect that those with the strongest political attachments might be most likely to alter their networks for political reasons. To consider this possibility, I look at the frequency of political unfriending by political ideology.


As can be seen in the figure, those who describe themselves as liberal, very liberal or very conservative are the most likely to unfriend for political reasons. This is consistent with research that finds that people with strong political views are the ones most likely to engage in selective exposure, or the tendency to avoid information that challenges their beliefs. Individuals toward the middle of the political spectrum, however, are less inclined to structure their social feeds to reflect their political preferences.

“Unfriending” may be more common when politics are especially heated

Although political unfriending is relatively uncommon, there is some evidence that it is more likely to occur in contentious environments. During the controversial 2012 gubernatorial recall election, 34 percent of people in Wisconsin said that they had stopped talking with someone because of conflict over the recall. And in amid the heated climate of the Israel-Gaza conflict in 2014, 16 percent of surveyed Israelis said they had unfriended another user.

That suggests Donald Trump could lead to something similar. Indeed, a Monmouth poll during the 2016 campaign suggested 7 percent of Americans had lost or ended a friendship over the presidential race. Whether Trump’s presidency leads social media users to purge their feeds of disagreeable political content is something we will find out over the next four (or eight) years.

Leticia Bode is an assistant professor in the Communication, Culture and Technology program at Georgetown University.