Fatal shootings by the police are just one piece of the deadly force “puzzle.” Until recently, we all have been questioning national patterns, practices and trends of these fatal shootings because no comprehensive data were available. Amazingly, and embarrassingly, this information has not been collected and analyzed so that ways to avoid police shootings could be explored at the national level.
We must acknowledge the media and The Washington Post specifically for pulling back the curtain so the public, policymakers and researchers can see partially what is inside. The Post recently completed its third year of tracking fatal shootings by U.S. police officers. Its data show that police have shot and killed almost 1,000 people per year each of the past three years.
{mosads}That fact alone is important, and provides an opportunity to examine trends and patterns in order to garner a better understanding of the circumstances under which police officers use deadly force. The loss of 1,000 lives a year is a critical reality, but we must pose at least three concerns as we examine The Post’s data for answers to our questions about officer-involved shooting deaths.
The Post’s data tell us nothing about nonfatal shootings, an officer’s action that is the same as one that takes a life. It remains unknown how often nonfatal shootings occur nationally (in other words, when a citizen is shot but survives, or when the officer’s bullets miss the citizen). Recently, VICE News reached out to the 50 largest U.S. police departments and obtained from 47 their data on all officer-involved shootings from 2010 to 2016. In some cities, fatal shootings were as little as 17 percent of all officer-involved shootings; in others, they were more than half of all firearm discharges. Academic studies likewise show that police shootings frequently do not result in death.
In other words, because officers are trained to stop the threat, there were likely thousands of incidents over the past three years in which officers fired their guns but the citizens survived because bullets missed vital organs, or the citizens received lifesaving medical treatment. These incidents do not appear in The Post’s data, but we need to know about them as well.
The Post’s data cannot tell us whether police officers are biased when deciding to use deadly force. To do so, we need an appropriate measure of each racial group’s “exposure” to police, so that we could compare the number of citizens fatally shot against the number of citizens who could have been fatally shot.
Thanks to The Post, the numerator (number of citizens fatally shot) in the equation no longer is unknown. But how do we come up with an appropriate denominator (number of citizens who could have been fatally shot)? To be certain, U.S. population estimates are not the answer because they assume everyone has an equal chance of being stopped or contacted by police officers. Studies consistently have demonstrated this is not the case: black citizens are more likely to be stopped, searched and arrested than white citizens.
Thus, the fact that black males account for 22 percent of fatal shootings, but just 6 percent of the U.S. population, tells us nothing about whether officers exhibit racial bias in their decision to use deadly force. After all, to be shot by a police officer, one must first be approached, contacted or stopped by a police officer (and bear in mind, forming suspicion and making the decision to stop or approach a citizen is entirely separate from the perceived need to shoot).
A better denominator might be arrest figures available from the FBI’s Uniform Crime Report, but these also must be viewed cautiously. Consider that, according to The Post, many fatal shootings stemmed from non-criminal incidents — or, at least incidents that would not necessarily require an arrest. For example, about 12 percent of fatal shootings occurred after a traffic stop, and another 16 percent from domestic disturbances. Another 14 percent began as a “suspicious activity” stop. If officers are more likely to perceive black citizens as suspicious, that would increase black citizens’ exposure to police, which might explain some of the racial variation we observe in fatal shootings.
Finally, the best denominator that could be used to make sense of fatal shootings is the number of citizens who have face-to-face contact with police each year. Unfortunately, the most recent data available are from a Bureau of Justice Statistics survey administered in 2011. Still, this survey gives us an estimate of how many police-citizen interactions occur in the United States annually. In 2011, BJS estimated 62.9 million citizens had at least one face-to-face interaction with a police officer. Three years prior the estimate was 40 million, and six years prior, the estimate was 43.5 million.
If we assumed, conservatively, that there were 40 million police-citizen interactions in 2017, then we would conclude using The Post’s data that police fatally shot a citizen in approximately 0.002 percent of those interactions. Furthermore, the majority of those fatal shootings involved a citizen who was armed with a deadly weapon and/or posing an imminent threat to the safety of officers or others. We raise these points not to dismiss concern about excessive use of force or possible racial disparities, but rather, to put The Post’s tally in the appropriate context.
The FBI has pledged to collect national use-of-force data. We sincerely hope the FBI follows through and can get agencies to participate by providing the rich data most agencies collect on these most serious police activities. Hopefully, the creation of a repository on all uses of force — and particularly discharges of firearms — and analyses of those data will help reduce the number of fatal shootings in coming years as the links among policies, training, supervision and accountability systems continue to improve and are informed by evidence.
Geoffrey Alpert is a professor in the Department of Criminology and Criminal Justice at the University of South Carolina. An expert on police use of force, he has been conducting research on high-risk police activities for more than 25 years.
Justin Nix is an assistant professor in the School of Criminology and Criminal Justice at the University of Nebraska Omaha. His research focuses on policing, procedural justice and use of force. Follow him on Twitter @jnixy.