The Department of Justice has been required to track excessive force complaints since 1994 but has little to show for it.
The federal government recently released what some hoped would be the most accurate data on how police departments resolve civilian complaints of police brutality.
Collected from nearly 4,000 law enforcement agencies across the country — spanning jurisdictions as big as New York City and as small as Yakutat, Alaska — the data shows fewer than one in 15 excessive-force allegations in 2015 met the threshold for disciplinary action against the accused officer. The vast majority of allegations were ruled unfounded, unsustained or otherwise dismissed.
But a GateHouse Media analysis found several flaws calling into question the accuracy of the one-year snapshot, which was appended to a 2016 survey by the U.S. Bureau of Justice Statistics about adoption of body-worn cameras. The BJS released the report in June.
Among the findings:
- Several agencies, including the Fayetteville Police Department in North Carolina, Miami-Dade Police Department in Florida and Harris County Sheriff’s Office in Texas, disputed the data when questioned about it by GateHouse Media, offering dramatically different internal numbers instead.
- The survey claims that about one in seven agencies did not provide data because complaint records were not “easily retrievable,” or they did not “formally document and store” complaints, raising questions about the agencies’ own ability to analyze their track records.
- Two-thirds of agencies that provided data show no excessive-force allegations. Yet some of those same agencies posted their own statistics on their websites showing they did receive such complaints in 2015.
- Roughly 1,000 agencies reported at least one civilian allegation of excessive force, but the disparities in their outcomes were wide, suggesting a need for standardization in how civilian complaints are processed and tracked.
“BJS is still examining the 2016 data and has no further comments,” a spokesman said in an emailed statement, in which the agency acknowledged “inconsistencies” with two previous data collection attempts.
The problems underscore a more salient issue: Despite decades of credible allegations of police brutality, and a 25-year-old federal law requiring the government to collect national data on excessive force, the United States still has little or no reliable information to show for it.
“It’s a huge mess, there’s no question about it,” said Matthew Hickman, chair of the criminal justice department at Seattle University and a former BJS statistician. “There’s no regulation at the national level. We’ve got 18,000 police departments, and they can each do whatever the hell they want.”
Data collection falls flat
The U.S. Attorney General’s Office has been required by law to “acquire data about the use of excessive force by law enforcement officers” and publish an annual summary of the data since 1994, when President Bill Clinton signed the controversial Violent Crime Control and Law Enforcement Act.
But the mandate came without funding, which is one reason why the federal agency has never done it.
In a response to questions, the U.S. Department of Justice cited a decades-long effort to capture various aspects of law enforcement use of force, which BJS said has partially fulfilled the requirements.
Among the attempts: In its triennial survey of U.S. residents, it added questions about whether they thought police had used excessive force against them; it conducted a national census from 2003 to 2012 about people who died during an arrest; and it conducted a 2002 survey that asked inmates in local jails whether law enforcement used force against them.
In 1996, BJS and the National Institute of Justice also sponsored the International Association of Police Chiefs’ effort to collect national data from police on the use of force. It issued a report in 2001, but the effort has since been abandoned.
In the early 2000s, Hickman, the associate professor at Seattle University, was one of two staffers in the law enforcement division of BJS. He decided it was time for the agency to start collecting the data on its own.
BJS already had a national survey called the Law Enforcement Management and Administrative Statistics (LEMAS), which was respected among policing agencies and had a high response rate. So as part of the 2003 survey, it included questions about civilian complaints against sworn officers, a measure police chiefs have long used to gauge the quality of officers’ interactions with civilians.
The agency collected the civilian complaints data in its 2003 and 2007 surveys. But when Hickman, who left BJS in 2007, and his research assistant, Jane Poore, analyzed the data years later, they discovered it suffered from serious flaws.
For example, some agencies incorrectly reported the total number of citizen complaints of any nature, instead of those specifically alleging excessive force. Other agencies incorrectly included the total number of internal investigations and officer-reported uses of force, as opposed to only to citizen complaints.
All these factors contributed to erratic results.
“The findings of this study indicate that the LEMAS data on citizen complaints about police use of force suffer from serious, if not fatal, measurement flaws,” Hickman and Poore wrote in 2016 in Criminal Justice Policy Review, a peer-reviewed journal. “The LEMAS data on citizen complaints do not provide a valid and reliable basis for comparative statistical reporting and research purposes, and should not be used for these purposes.”
Riddled with inaccuracies
Following the 2007 LEMAS survey, BJS stopped collecting citizen complaint data on excessive force. In 2016, the agency undertook the effort again.
This time, BJS statisticians included the questions in a supplement to its LEMAS survey on the adoption of body-worn cameras. BJS statistician Shelley Hyland consulted with Hickman to phrase the questions to eliminate as much confusion as possible.
But GateHouse Media’s analysis of the results indicates the data is still unreliable.
For instance, the survey shows the Fayetteville Police Department in North Carolina substantiated 42 out of 46 excessive force allegations, meaning it agreed with 91% of the claims. That’s compared with a national rate of just over 7% sustained allegations.
Gina Hawkins, who became police chief in 2017, said that someone must have incorrectly entered the information.
“I think whoever entered the data on the survey halfway read the question and entered the number that was the opposite,” Hawkins said, adding that her department is looking into resubmitting the 2015 numbers. “We don’t know who entered it in the system.”
The Police Department’s own records show only one allegation of excessive force, which was not sustained, said Capt. Todd Joyce, who oversees internal investigations.
The survey also shows the Harris County Sheriff’s Office in the Houston area sustained 195 of 256 allegations, or 76%. But spokesman Jason Spencer said the agency sustained seven out of 267, or 2.6%.
Spencer said he was unable to determine the source of the numbers that appear in the federal data, but a different administration was in charge at the time.
“The bottom line,” Spencer said, “is that we believe the data in the report is extremely inaccurate.”
The Miami-Dade Police Department also disputed the numbers, which show it received 78 excessive force allegations, all of which were pending an outcome at the time of survey. But the department provided GateHouse Media a document showing it received 282 allegations, none of which were pending. Only seven, or 2.5%, were sustained.
Of the 4,000 agencies included in the federal report, one-fourth of them did not provide data at all because they did not “formally document and store complaints,” their complaint records were not “easily retrievable” or those questions were left blank, the survey said.
Among this non-reporting group were the Los Angeles and Chicago police departments, both of which said their complaint records were not easily retrievable, even though both submit their own use-of-force reports each year.
Neither the LAPD nor CPD responded to requests for comment.
Of the other 3,000 agencies, two-thirds reported no excessive force complaints in 2015. They include the San Diego Police Department and San Jose Police Department in California.
But both agencies pointed to local reports that show dozens of force allegations that year. SDPD said it wasn’t able to find a copy of the survey it submitted, but SJPD was able to obtain its copy from the DOJ, which indeed showed zeroes in the questions about excessive-force allegations.
Brian Matchett, a commander in the SDPD internal affairs unit, said the staffer who filled out the form had no recollection of doing so.
“Clearly those are not accurate numbers,” Matchett said. “I don’t know why that person wouldn’t have reached out to internal affairs for the actual numbers versus putting in zero.”
What the data tells us
The data might provide telling insights for the nearly 1,000 agencies whose reported numbers appear, at least on the surface, to be reliable.
For one, only about 7% of formal excessive force allegations are sustained. While some will suspect a low rate is a sign of police bias against accusers, Hickman cautions that the majority of allegations likely don’t meet legal thresholds to be considered “excessive.”
“The courts and the police, how they define excessive force, are very different from the public’s understanding of what may or may not be ‘excessive,’” Hickman said. “Low sustained rates for physical force complaints by themselves do not just indicate that the police are not doing a good job of policing themselves.”
Still, Hickman said he’s skeptical of agencies with high rates of unfounded allegations, in which they determined the incident didn’t occur or the allegation wasn’t based on facts.
The data shows, for example, that the Las Vegas Metro Police Department ruled 92% of allegations unfounded and the San Bernardino County Sheriff’s Office in California ruled 87% unfounded. Nationwide, around 26% of complaints are unfounded; among the rest, 34% end with exoneration and 22% are not sustained.
“Certainly there are cases where citizens are not telling the truth, lying or exaggerating what happened, but whether it would happen to that extent, I’m having trouble believing,” Hickman said. “It implies that either citizens are lying, or the police are just maybe not doing a thorough investigation.”
The Las Vegas department disputed the data without providing alternative numbers. The San Bernardino County Sheriff’s Office confirmed its numbers are accurate.
“In my experience, the majority of complaints are unfounded,” sheriff’s Sgt. Jeff Allison said. “It’s almost as if the complainant will make as many allegations as possible, but they’re not based on what actually occurred.”
The Sheriff’s Office classifies allegations as unfounded when an internal investigation proves it false, Allison said, which is different from claims dismissed due to lack of evidence.
San Bernardino deputies did not use body-worn cameras at the time of the survey — though deputies did use audio recorders, Allison said.
GateHouse Media’s analysis found agencies that reported having acquired body-worn cameras in any form, including testing, at the time of the survey were roughly twice as likely to sustain excessive force allegations, and less likely to deem allegations unfounded.
A yearlong pilot program initiated by the San Bernardino County Sheriff’s Office in 2017 equipped some deputies with body cameras, which Allison said were “highly successful” at proving false allegations against officers. Allison cited two examples of complainants alleging they were assaulted by deputies, but body-camera footage showed otherwise.
“I think that a lot of our unfounded cases is the fact that our sheriff’s deputies do a good job day in and day out,” Allison said. “At the end of the day, law enforcement has a job to do and some people are not going to be happy about it.”
But the Sheriff’s Office is no longer using body-worn cameras, Allison said, due to the high cost of data storage.
About half the agencies surveyed reported having acquired body cameras in any form at the time. Of those that did not, roughly three-quarters cited the costs of video storage and disposal, hardware and ongoing maintenance as primary reasons. About 4 in 10 agencies without cameras said they had not acquired them due to privacy concerns.
The Sarasota Police Department in Florida bought body cameras for officers four years ago, but city leaders still aren’t sure about whether they should use them. In a July meeting, Chief Bernadette DiPino told the City Commission she is a “huge advocate” of cameras but hasn’t been pushing for them due to privacy issues and cost.
“If you say you’ll fund the cameras, I’ll be very happy about it because, great, this will be a way for us to show what our police officers are doing; it helps us with evidence,” DiPino told the commission. “But I think it’s awfully expensive, and when I look at that number, that’s seven more police officers I could have in the community for about that same exact price over a 10-year period. I would rather have them.”
No national standards
The wide disparities in how complaints are resolved can be explained in part by the lack of national standards for how agencies should track and process complaints.
Some agencies accept complaints online while others require sworn complaints in person, which can be more intimidating. Hickman said the in-person method would probably lead to fewer total complaints and a higher sustained rate, because weaker complaints would be weeded out on the front end.
In addition, Hickman said, uniform definitions as to what constitutes force and excessive force still do not exist.
“Why would use of force in Seattle be defined differently than use of force in Baton Rouge?” Hickman said. “There’s no good reason for that. It should be the same.”
Hawkins, the Fayetteville police chief, recently sat on an FBI task force to address that issue. The National Use of Force Data Collection Task Force, which started in January 2016, set out to develop uniform use-of-force data on a national scale. The FBI officially launched its new National Use-of-Force Data Collection program in January.
“There’s no mandatory way of reporting it,” Hawkins said. “Now we’re making sure people are formulating a consistent way — how to calculate it and how to have it for the future where measures are comparing apples to apples.”
The new data collection program will include information about the incident, officer and subject of the force, but it will cover only incidents in which the officer discharges a firearm in a person’s direction, causes a death or inflicts serious bodily injury. That could exclude most use-of-force incidents.
As recently as the CNN Democratic primary debate on July 31, Democratic presidential candidate Julián Castro called for federal regulation in tracking excessive-force allegations.
“We have a police system that is broken, and we need to fix it,” Castro said. “Whether it’s the case of someone like Tamir Rice or Michael Brown or Eric Garner, where the Trump Justice Department decided not to pursue charges, we need to ensure we have a national use-of-force standard, and that we end qualified immunity for police officers so that we can hold them accountable for using excessive force.”
But Hickman says it may be some time before we see national standards how local agencies should structure and process civilian complaints of excessive force.
“We just have such a strong history and tradition of local control over law enforcement and this kind of overlaying respect for states’ rights,” he said, “that at least during more conservative administrations, you see a lot less emphasis on this kind of thing.”
Fayetteville (N.C.) Observer staff writer Michael Futch contributed reporting to this story.