Skip Navigation

Produced by the Office of Marketing and Communications

Subscribe Now

Study: Anti-Black Racism Fell—Temporarily—Following Floyd’s Killing

UMD Researchers Studied Millions of Tweets for Evidence of Bias

By Kelly Blake

Illustration of Twitter bird with profanity speech bubbles

A team led by a School of Public Health researcher examined millions of race-related tweets from November 2019 to September 2020, and compared trends from the periods before and after Ahmaud Arbery, Breonna Taylor and George Floyd were killed.

Illustration by Valerie Morgan

George Floyd’s murder at the hands of police last May sparked horror, outrage and demands for change worldwide. But a new study led by University of Maryland researchers used Twitter data to take the nation’s pulse on race-related sentiments and suggests that changes in attitudes—particularly about anti-Black sentiment—were short-lived following the high-profile killings of Floyd and other unarmed Black people in early 2020.

Nationwide, the proportion of tweets referencing Black people in a negative way after the Floyd killing declined by 32%, but these negative tweets returned nearly to baseline levels in July, according to a new study published this month in the journal SSM-Population Health.

“Things that are said online are a proxy that can reflect things about the greater society and social and cultural norms,” said Thu Nguyen, associate professor of epidemiology at the UMD School of Public Health, who led the study. She said, however, that a decline in expressions of racist sentiment does not necessarily equate to an overall decline in racism, which is an organized, interconnected social system that perpetuates inequality through both personal and institutional dynamics.

Her research team examined millions of race-related tweets from November 2019 to September 2020, and compared trends from the periods before and after Ahmaud Arbery, Breonna Taylor and Floyd were killed. Using machine learning models, they analyzed the sentiment of tweets that contained one or more keywords related to Black people. Not all tweets characterized as “negative” were expressing racist or prejudiced ideas; in fact, some were calls to action for change.

To complement the sentiment analysis, the team also looked at a subset of 3,000 tweets, which revealed themes of desire for long-lasting social change and specific calls for actions. Yet, Nguyen also noted a backlash of negative sentiment against Black Lives Matter protests, as well as victim-blaming and an overall denial of the impacts of racism.

Nguyen’s research group, which also includes her sister, Quynh Nguyen, associate professor of epidemiology, uses social media data to gauge attitudes about race and expressions of racism because she says that areas with greater negativity toward minoritized communities are likely to be areas of increased stress for racial and ethnic minorities.

“We know that race-related stress is harmful and contributes to many health conditions,” Thu Nguyen said. “We look at racial sentiment by geographic area because there may be important implications for support for programs and policies that promote equity and health.”

While this study demonstrates the durability of negative attitudes toward Black people, there’s an optimistic takeaway, Nguyen said: Transitory but powerful shifts in racial sentiment due to important and tragic events can provide windows of opportunity to create lasting social change.

“Activists can seize these moments to motivate grassroots organizing, mutual aid projects and other efforts to promote equity,” Nguyen wrote in the paper. “In tandem, policymakers and political leaders can leverage the public’s outcry and calls for justice to write and pass legislation to create sustainable and systemic change planting further seeds to promote egalitarian societal norms and attitudes.”



Schools & Departments:

School of Public Health

Maryland Today is produced by the Office of Marketing and Communications for the University of Maryland community on weekdays during the academic year, except for university holidays.