Study raises issues with LMPD traffic stops

An analysis of 2014 vehicle stops by Louisville Metro police officers has found again that black drivers are searched twice as often as whites.

That finding was also a main takeaway of a similar department-commissioned report made public in October 2014 that was unable to conclude if racial bias played a factor in the stops.

“There’s no measure in the world that can tell you if police officers are engaging in biased policing,” said Deborah Keeling, professor in the Department of Criminal Justice at the University of Louisville.

Keeling conducted the study, analyzing 82,590 vehicle stops from 2014.

Many of the report’s findings remain unchanged from the prior study. Women make up about one-third of those stopped. Black drivers are more likely to be the subject of a probable cause search. Two-thirds of drivers stopped were white.

But new in 2014: more searches resulting in no findings.

Take for example probable cause searches, those in which an officer has reason to believe a crime has been committed. In the 2013-2014 study, 76 percent of such searches resulted in some type of “positive” finding – contraband, evidence or fruits of the crime. In 2014, the number fell to 43 percent – a 44 percent drop.

Positive findings in “incident to arrest” searches fell by 70 percent and consent searches saw a 47 percent drop.

“These differences are significant and may indicate a change in officer decision-making that may require additional training on the various justifications for vehicle searches and/or training on how to conduct a more complete search,” the study reads.

LMPD launched focus group meetings with officers to find out why searches were turning up with fewer incriminating results.

In those groups, officers said more training is needed for new recruits to conduct “more complete searches,” according to the study. Some officers said they were fearful of being stuck by a drug needle.

The department plans to beef up training to help officers make the call of whether or not it’s justified – and legal – to search a vehicle.

Also different in 2014: fewer drivers, once stopped, are being searched. That percentage fell from 9.5 percent in the 2013-2014 study to 6.5 percent in 2014.

Unchanged from the last study is the lack of a critical piece of data.

While officers do indicate on their reports if the stop started as a suspected criminal or traffic offense, Police Chief Steve Conrad said an issue with a state-provided electronic records system prevents aggregate analysis of that information.

The 2013-2014 study recommended such data be added, saying “the original reason for the stop determines much of what follows during the stop as well as the outcome of the stop.”

Keeling told reporters in 2014 she intended to incorporate that data in the 2014 report but in an interview said the technical problems prevented that.

“If you look at the other reports, when you controlled for the type of stop, a lot of the differences in race disappear,” she said, adding that she's found a way to include that info in the 2015 analysis.

The study noted any possible effects of post-2014 training, as well as body cameras that rolled out in 2015 and early this year, won’t be seen until the 2015 study at the earliest.

Conrad and Keeling also addressed the timing of the study’s May release – nearly 17 months after the last vehicle stop analyzed. The report was a part of a three-study contract with the university – including a 2014 report on citizen attitudes – that cost taxpayers a total of $134,000, Conrad said.

Keeling gave the police department her findings for review last fall, she said. Conrad acknowledged the age of the data and said it could have been released earlier. He explained the department took the winter and spring to add an extensive appendix to the study that lists ways LMPD has woven concepts such as procedural justice, implicit bias and de-escalation into its training and practices.

LMPD is again contracting with U of L for a 2015 study, one that Keeling said will be turned over to Conrad for review by August or September.

"I think this is just one piece of the puzzle," Conrad said of the study. "All of this together are our steps to try and do everything we can to demonstrate to the community that we want and need their trust..."

Other findings:

• Months with the highest percentage of stops: January, February, March and May
• Divisions with the highest percentage of stops: 8 (18.2 percent), 5 (17.9 percent) and 7 (13.8 percent)
• 37.4 percent of drivers stopped were women
• More than 60 percent of drivers stopped were between ages 20 and 40.
• Of all drivers stopped in 2014, 66 percent were white, 28.8 were black and 3.8 were Hispanic, nearly identical to the 2013-2014 study’s 67, 28 and 3.5 percentages.
• Not included in the 2013-2014 study was a racial breakdown of the search outcome. The new study shows that searches of vehicles with black drivers were twice as likely to have a “positive” finding (2.6 percent) compared with white drivers (1.3 percent).
• When searched, regardless of the type of search, both whites and blacks were found to be in possession of contraband, evidence or some fruit of the crime at nearly the same rate.
• Two-thirds of all drivers were given a warning and about a quarter were cited. Of all stops, blacks and Hispanics were most likely to be arrested, at 8.8 percent and 8.6 percent, respectively. There was not a large racial difference how often blacks, whites and Hispanics were given a warning or citation. 

Reporter Matthew Glowicki can be reached at 502-582-4989 or mglowicki@courier-journal.com.