Space & Innovation

Racist Tweets Can Be Tempered by Popular, Same Race Peers

A study found that stopping harassment on Twitter was most effective if the intervenor was of a similar race, and had a large or comparable number of followers to the harasser.

<p> Chesnot/Getty Images<span></span></p>

Today, Twitter unveiled new anti-harassment tools that would allow users to mute or block certain negative or derogatory words from their notifications. But how can people stop such an outpouring of hate to begin with?

In a study that will be published in an upcoming issue of the journal Political Behavior, NYU PhD student Kevin Munger found that stepping in and stopping harassment on Twitter was most effective if the intervenor was of a similar race, and had a large or comparable number of followers to the harasser.

Munger created four different profiles on Twitter to respond to racial harassment (identified by racial slurs in tweets directed towards other Twitter users). All four profiles presented as male. Two presented as white, two presented as black, and two had a higher number of followers and two had lower numbers of followers.

Munger then monitored Twitter for harassers, specifically white men who were not juveniles and used racial slurs in over 3 percent of their tweets. When he noticed one, he had one of his four personas respond with this tweet:

"@[subject] Hey man, just remember that there are real people who are
hurt when you harass them with that kind of language"

The corrective tweet was often met with outright hostility, but when the tweet came from a white man with a high number of followers, the harasser toned down the racial slurs in subsequent tweets for up to a month. Munger speculates that this is because the harassers had something in common (their race) with Munger's profile, and psychologically were more willing to listen to the correction.

WATCH VIDEO: How Powerful Is Twitter?

"Efforts to sanction people and encourage them to behave better are going to be more effective if the people have something in common" Munger told Seeker. "I think that it is really important to emphasize these commonalities, to make the person that you're trying to change feel like you're on the same side rather than attacking them."

That's in contrast to other anti-harassment methods which rely on moderators or companies to step in and regulate harassment from the top down.

"I think a lot of the top-down efforts don't work that well, and they do take a lot of effort, so if people could find a better way to talk to each other and encourage good behavior among each other that would both free up resources and be more effective." Munger said.

Munger's research is especially important in light of recent events. Donald Trump, the President-elect, has credited his extensive use of Twitter and Facebook for his win, solidifying social media's standing, in the political sphere.

But Trump's election has also been linked to the rise of a radical white nationalist movement. Could his often insulting tweets as a powerful white man help to normalize harassing behavior, just as censure of harassment helps to diffuse harassment? The answer isn't clear.

RELATED: Does Twitter Need A Robin Hood?

"It depends on how these norms [governing use of hate speech online] come about. Online groups tend to be quite diffuse and these things evolve. I don't think that even one influential user could change things too much by themselves, but I could certainly see that contributing to changes in norms." Munger said.

Reports of harassment and hate crimes have gone up since the election, making it more important than ever to find ways of reducing harassment online. And the conclusion of this study is that to call out harassing behavior, it helps to find common ground, even if you have to reach across a deeply divided aisle to do it. It also means that allies of minorities and oppressed groups have both the power and the responsibility to step in and prevent harassment as it happens.

"People have lots of different overlapping social identities and in some contexts different identities are more salient. It's actually not the case that we have nothing in common, it's just that the most salient thing in a political context is partisanship. If you don't have that in common, it feels like you have nothing in common, whereas if you find what you do have in common it becomes a lot easier to empathize with and understand other people." Munger says.

Munger hopes to do more work on the theory, and is currently looking to expand the work from racial-focused harassment to looking at politically-focused harassment directed at Hillary Clinton and Donald Trump. The results of that study are still being written up.