How Hate Speech On Social Media Can Predict Hate Speech In Real Life

Hate crimeAnyone that has spent any time on social networks such as Twitter is almost certain to have encountered hate speech of some kind.  Is such language particular to social media where people are emboldened by their anonymity, or is it more reflective of wider changes across society?

That was the question posed by new research from New York University, and the data suggests there does indeed appear to be correlation between online racism and reports of hate crimes related to race, ethnicity or national origin in a region.

The researchers analyzed around 532 million tweets posted between 2011 and 2016, capturing both the location of the tweet and its linguistic features.  The data was then used to train a machine learning model to identify tweets that share discriminatory views or describe or comment upon discriminatory remarks or acts.  This dataset was then compared to recorded incidences of hate crimes in the same locations during the same timeframe.

“We found that more targeted, discriminatory tweets posted in a city related to a higher number of hate crimes,” the researchers say. “This trend across different types of cities (for example, urban, rural, large, and small) confirms the need to more specifically study how different types of discriminatory speech online may contribute to consequences in the physical world.”

Black mirror

The team attempted to include a diverse range of cities to capture varying levels of population density, diversity and social media usage.  They focused on tweets and crimes related to racial, ethnic or national origin-based discrimination because data is captured by the FBI on such crimes, hence why crimes related to sexual orientation were not included as data is not so readily available across the United States.

In addition to chronicling the correlation between hate speech online and hate crimes in the real world, the team were also able to identify a number of terms and phrases that were commonly used on social media, which the team believe will help to better identify groups that are possible targets of racially motivated crimes or discrimination.

They plan to continue to explore the topic, and aim to focus specifically on the impact of online hate speech on the victims.  Whilst the implications of hate crimes and the psychological stress caused has been a topic of considerable study when occurring offline, there has been much less study of the implications of online hate speech.

There is also no real understanding as to the connection between online and offline instances of hate speech, so there remains much to be explored, but the study presents a useful contribution to the debate.

Facebooktwitterredditpinterestlinkedinmail