After a Minneapolis police officer killed George Floyd, hundreds of thousands of protesters took to the streets, both across the United States and around the world. Americans’ reactions both to Floyd’s death under officer Derek Chauvin’s knee and to the ensuing protests have followed a familiar script. Some see it as further evidence that U.S. policing is deeply racist; others think protesters are overreacting to officers trying to do their jobs. Opinion largely divides along racial lines, with some variation by party affiliation.
Why do black and white Americans have such different perceptions of what happened? Here’s what our research reveals.
White and black Americans interpret police killings differently
The Floyd protests echo the 2014 Black Lives Matter protests that spread after a Ferguson, Mo., police officer shot and killed Michael Brown. Three weeks later, we surveyed a broad national sample of 3,729 Americans, using Qualtrics Panels, to better understand reactions both to Brown’s death and the protests.
We found a stark racial divide in American opinion, as you can see in the figure below. Black Americans and white Americans had very different beliefs about whether the officer should be charged with murder, and whether the protests were appropriate. Black Americans were more likely than whites to think that race played a significant role in the shooting and less likely to think that the police response to the protests was appropriate. Further, blacks and whites disagreed about what had actually happened between the two men — whether Brown had attacked Wilson and whether he had a weapon.
Different interpretations come from different beliefs and expectations
What could plausibly explain why black and white Americans interpret these events differently? First, it could be that they received different information about the events. Perhaps they followed different news sources that reported what happened in different ways. Second, it could be that they received similar information, but they themselves interpreted it in different ways.
If the key factor is interpretation, individuals may have interpreted the information differently for one of two reasons. First, individuals may want to view members of their own race in a positive light, and therefore focused on information that showed the member of their racial group more favorably. We describe this as “race-based motivated reasoning.” “Motivated reasoning” here means that individuals are interpreting information with the goal of making their racial group look good. This is similar to partisan motivated reasoning that Democrats and Republicans use to judge politics. Alternatively, black and white Americans might have different beliefs and expectations about police bias and black victims’ culpability, perhaps because of their own lives or the experiences of others they know.
To figure this out, we conducted another study in 2016. We asked a broad national sample of 370 black and 356 white Americans, recruited using Qualtrics Panels, what they thought of a fictitious scenario we had written in which a police officer killed a black man. Because we had made up this scenario, we knew that black and white respondents could not have seen different information about the event. Still, their reactions divided similarly by race. This meant that the racial opinion divide was not from where respondents got their news but from how they interpreted that information.
Racial differences in perceptions of fictitious shooting
To examine the possibility that the racial divide resulted from race-based motivated reasoning, we randomly assigned some respondents to answer questions about their racial group attachments before they answered the questions about the shooting. Here, we used standard survey questions that ask respondents about the strength of attachment to their racial group.
If bringing race to the forefront of respondents’ minds before hearing about the incident widened the racial divide, reactions were likely driven by race-based motivated reasoning. We would also suspect race-based motivated reasoning if people who had strong attachments to their race answered differently than people who had weak attachments to their race. Finally, to assess whether the racial divide comes about because blacks and whites encounter these incidents with different beliefs and expectations, we rely on questions that tap into respondents’ perceptions of the police and their perceptions of black people. These questions were asked a week before they read about the fictitious shooting.
Here is what we find. Answering questions about racial group did not widen the racial divide — but black and white Americans who reported being more strongly attached to their racial groups did divide more than those less strongly attached. We also find strong evidence that differing expectations and beliefs about the police and the likely culpability of black victims explain differences in interpretations of the scenario. That suggests that racialized experiences, including experiences with the police, likely shape interpretations of these events more than individuals’ desire to view their racial groups positively.
What this tells us about the current protests
Our work suggests that Americans’ existing beliefs influence how they interpret police violence. Americans disagreed about Ferguson because they had different beliefs about whether police generally behave fairly — and therefore whether someone who was killed must have been culpable in some way. These beliefs are shaped by the different experiences that black and white Americans have with the police and with the criminal justice system more broadly.
To be sure, in the years since Michael Brown’s death, many more white Americans have seen video evidence of excessive police violence. A recent Monmouth poll suggests that Americans are far more united in believing that the Minneapolis officer acted inappropriately than they were about the Ferguson shooting. But they split along racial lines in whether they do or do not support the protests.
Our data suggest that one plausible reason for this divide is that black Americans are more likely to have been mistreated by the police or known someone who has, while white Americans who have not observed this personally may think that biased policing is rare. Indeed, that same Monmouth poll finds that 87 percent of black Americans think police are more likely to use excessive force when a culprit is black, while only 49 percent of white Americans do. However, both numbers have increased since 2014, suggesting that protests and online video evidence may be persuading more Americans of all races that police bias is a problem. Widely shared videos of police attacking protesters may change opinions further still.
Hakeem Jefferson (@hakeemjefferson) is an assistant professor of political science at Stanford University.
Fabian G. Neuner (@FabianNeuner) is an assistant professor in the School of Politics and Global Studies at Arizona State University.
Josh Pasek (@jmping) is an associate professor of communication and media at the University of Michigan.