Black Americans are more than twice as likely as white Americans to be unarmed when they’re shot and killed by police officers, according to a study published Wednesday in the journal Criminology & Public Policy. And the new study finds that racial bias is a likely contributing factor.
While this point is a well-known issue of contention between police departments and many of the communities they work in, it has been very difficult to actually study due to a lack of data. To determine whether racial bias could account for the discrepancy, Justin Nix, lead study author and assistant professor at the University of Louisville, analyzed 900 fatal police shootings that occurred in 2015, collected from The Washington Post’s two-year-old database of fatal police shootings.
There’s no government database on fatal police shootings, so relying on a journalism source was Nix’s alternative. State and local jurisdictions aren’t required to collect information about officer-involved shootings, and they’re also not required to submit that information to the federal government if they do track it. Non-fatal officer-involved shootings are particularly difficult to study.
Because there’s no good national data on police shootings, racial bias has frequently been measured using lab simulations to test officers in hypothetical scenarios ― something researchers want to move beyond to determine whether racial bias is a factor in police shootings in the real world.
In his analysis, Nix found that even after controlling for factors such as age, mental illness, region, crime rate in that jurisdiction and city size, black citizens shot and killed by police were still more likely to be unarmed than citizens of other racial groups.
“Long story short, we think some of these findings are suggestive of implicit bias,” Nix said.
Still, he stressed that his work has limitations and that implicit racial bias isn’t the only possible explanation for racial disparities in fatal police shooting. One factor Nix wasn’t able control for: the information officers were working with when they responded to a 911 or dispatch call (while dispatch decisions could also involve implicit racial bias, it’s not on the part of the police).
“If by chance [police officers] were more likely to have been provided with information that black citizens were armed, then that might explain why they went in and we observed the outcomes that we did,” he said.
Shooting deaths don’t account for all gunfire in the field
Nix lauded the Post for its commitment to documenting fatal force in policing, but acknowledged that the Post’s database is likely incomplete.
“It doesn’t include all those instances where officers shot and missed,” Nix explained. “Or if they shot and wounded but didn’t kill a suspect, those events are still deadly force incidents. The officer had the intent to kill but didn’t. We need to know about those incidents as well.”
Civilian death only occurs in 15 to 25 percent of police shootings, meaning that thousands of non-fatal shootings are potentially unaccounted for, according to a study published in the journal Criminology and Public Policy in 2015.
“We have relatively little rigorous scientific evidence on either side of the case,” Sharad Goel, a professor at the Stanford School of Engineering, previously told HuffPost. Goel, who was not involved in the study, is a member of Law, Order and Algorithms, a research project that aims to analyze 100 million traffic stops in the the U.S. for evidence of racial disparities and discrimination.
The current lack of reliable data is “not a great state of affairs for understanding police encounters with the public,” Goel said.
A year of controversy in policing research
The new study follows a year of often conflicting research on racial bias and use of force in policing. The most controversial among the 2016 reports was a “working study” published by Harvard economist Roland Fryer this summer that claimed there was no racial bias in police shootings.
The media extensively covered Fryer’s report, including a New York Times story that ran with the headline “Surprising New Evidence Shows Bias in Police Use of Force but Not in Shootings.”
“I agree that blacks are more likely to be stopped, more likely to be harassed and more likely to be arrested,” he clarified. “Ideally we would be able to set up an experiment to understand potential differences before an encounter. Unfortunately, that would require us to randomly assign civilian race in encounters of police, which isn’t possible!”
Despite Fryer’s clarification, his paper reached the highest levels of government. In October, FBI director James Comey cited the study during a speech he gave to a gathering of police chiefs, during which he criticized viral videos of police brutality as not being representative of the state of policing in America. He also called for better data collection on police encounters with civilians.
“There were 10.7 million arrests in this country last year, and many times that number of encounters between officers and civilians,” Comey said. “Out of those tens of millions of encounters, how many people were shot? What did they look like? What were the circumstances? Is deadly force use trending up or down? Where is it worst and where is it best? Nobody knows.”
(Last year the Justice Department also announced a 2017 pilot project to collect use-of-force statistics and create a national database of fatal and non-fatal police interactions, though it remains to be seen if President Donald Trump’s administration will prioritize that project.)
Nix, for his part, is hopeful about the prospect of such a project. “We implore the federal government to move forward with this database that they’ve been talking about and go a step further,” he said. “Not just mimic what the Post has done and collect fatal shootings, but all shootings.”
And beyond government’s responsibility to require data collection and reporting on police shootings, the media has a responsibility to accurately translate that information about police shootings to the public.
While researchers and the scientific community understand that one study is a single data point among a landscape of information, blockbuster headlines calling attention to a single study can skew the public’s opinion of nuanced public health issues and ultimately impact policy.
It’s also crucial that news organizations put the facts we know ― and even more importantly, the facts we don’t know ― about shootings into context.