Skip to main content

Just How Biased Are Police Shootings?

Unarmed African Americans are as likely to be shot as armed whites, one study found.

By Nathan Collins


(Photo: Tim Sloan/AFP/Getty Images)

The anecdotes are, of course, hard to ignore. Last week, police in Minnesota fatally shot Philando Castile during a traffic stop, just one day after Baton Rouge police killed Alton Sterling outside a convenience store. Castile and Sterling’s deaths are just the latest two in a series of questionable police shootings that stretch back years and years. The lesson from all of these killings is clear: It’s not safe to be a black man (or boy) interacting with police in America.

Exactly how unsafe is it? Between 2011 and 2014, the probability of being black, unarmed, and shot by police was 3.49 times higher than the probability of being white, unarmed, and shot by police, according to a paper published last November in the journal PLoS One.

In some counties in America, that ratio soars to 20 to one, and unarmed African Americans are as likely to be shot by police as armed whites. And that’s got nothing to do with crime rates, writes Cody Ross, an anthropology graduate student at the University of California–Davis.

At first glance, the PLoS One study appears to contradict a more recent paper by Harvard University economist Roland Fryer, who found evidence of police bias in certain actions — pushing a suspect to the ground, for example — but not shootings. There are at least two likely reasons for the different conclusions: Fryer relies heavily on data reported by police officers, and, perhaps more importantly, he focuses on the number of people shot by police as a fraction of those already detained — a valid choice for Fryer’s purposes, but one that obscures the higher rate at which police detain blacks versus whites.

Armed African Americans were nearly three times as likely to get shot by police as armed whites.

Ross drew his data from the United States Police Shooting Database, a project conceived by journalist Kyle Wagner. Drawing on press reports, the Police Shooting Database is intended to be more accurate than statistics from the Federal Bureau of Investigation, which are collected in a decidedly non-uniform fashion and are based entirely on the reports of individual officers and departments, who of course may have an interest in obfuscating the truth.

Apart from estimating the relative risk of being black, unarmed, and shot by the cops, Ross also found that armed African Americans were nearly three times as likely to get shot by police as armed whites.

Ross identified Miami-Dade County, Florida; Los Angeles County, California; and Orleans Parish, Louisiana, as the worst places to be black, unarmed, and approached by police.

What explains the differences? “In effect, larger county population size, a higher proportion of black residents in the population, lower median income, and greater disparities in income all appear to be reliably associated with an elevated ratio of police shooting rate against unarmed black individuals relative to unarmed — and even armed — whites,” Ross writes. Meanwhile, crime rates—even race-specific crime rates—do not appear to be related to racial bias in police shootings.

Ross notes that these results are preliminary, in part because the database is as yet incomplete—the search was structured around specific days of the year, and volunteers have searched for shootings on only about half the days between 2011 and 2014.

“Attenuating the racial bias in police shootings in a given location requires a better understanding of the drivers of racial bias operating in that location,” Ross writes. “Perhaps police departments with disproportionate rates of racially-biased police homicide can provide justification for these patterns based on local context, or perhaps they are headed by individuals like Police Commander Jon Burge — the public needs to know which.”