Next wave of police departments face racial disparity analysis

Central Connecticut State University researchers released their third annual statewide report Thursday that identified seven Connecticut police departments for further study because of racial or ethnic disparities in their traffic stop patterns.

The departments are Berlin, Monroe, Newtown, Norwich, Ridgefield, Darien and State Police Troop B in North Canaan.

In these jurisdictions, minority drivers were more likely to be stopped during daylight hours than at night. The assumption is that it’s generally easier to see a driver to determine their apparent race or ethnicity during the daytime. Applying this so-called “Veil of Darkness” analysis to Ridgefield, for example, researchers found Hispanic drivers were 2.5 times more likely to be stopped in daylight than at night.

Researchers used two other statistical methods to identify departments’ racial disparities — a “synthetic control” test, which compares departments to other departments considered similar by a number of measures, and a test of how often drivers or their vehicles were searched versus how often those searches actually turned up anything illegal. In Monroe, for instance, officers were far more likely to search Hispanic drivers without finding anything, with a “hit rate” of 8.3 percent compared to 42.9 percent for white drivers.

The report is the first of two that come out each year as part of the Racial Profiling Prohibition Project. The second part will focus just on the departments identified in Thursday’s report, and researchers will use more data specific to each jurisdiction to give the numbers more context and try to determine the cause of the disparities.

The goal of the Racial Profiling Prohibition Project is to prevent racial profiling by police, but these reports alone don’t do that, or claim to. The authors take pains to state that the disparities they identify don’t prove profiling is the cause. The follow-up reports shed light on factors other than profiling that can influence the disparities, such as high-traffic-volume roads that might lead to more enforcement in high minority areas.

But the Connecticut Police Chiefs Association doesn’t think those distinction will be clear to the public and fears these reports can be taken out of context or used to jump to premature conclusions. They take issue both with the presentation of the reports, and with the methodology.

“One of the problems is that they name many police departments before they do a secondary review with the chief, and although they don’t indicate the police departments are racial profiling, the fact that they’re being named indicates the public may believe they are,” said Watertown Police Chief John Gavallas, president of the Connecticut Police Chiefs Association.

“Certainly as chiefs we all want to know if racial profiling is going on so we can correct it, but so far there’s been no definitive evidence that has demonstrated there’s been any racial profiling going on.”

Police chiefs long critical of the state’s approach had new fodder Thursday — a peer review they requested by academic heavyweights in the field of racial profiling and criminal justice research that challenges some of the state’s analytical methods.

The peer review, dated Sept. 7, was made public by police chiefs on the Racial Profiling Prohibition Project Advisory Board Wednesday night only about 12 hours before Thursday’s  meeting.

The review was based on past years’ reports from IMRP and was composed of comments from three well respected researchers in the field of criminal justice, including one of the creators of the Veil of Darkness analysis, University of Chicago professor Jeffrey T. Grogger. The other researchers were Michael Smith, chairman of the criminal justice department at the University of Texas at San Antonio, and Edward Maguire, a criminology and criminal justice professor at Arizona State University.

The reviewers generally agreed on a few things. They were highly critical of the IMRP’s use of methods that rely on estimating a town’s “driving population” for each race or ethnicity using Census and other data, and comparing that to the mix of drivers actually pulled over to see if they were stopped at a pace approximately equal to their proportion of the population. The reviewers said those comparisons were “invalid” and should be dropped from the report.

The reviewers leaned toward the Veil of Darkness data, which doesn’t rely on estimating who’s on the road. It only compares actual stops during different visibility conditions.

Bill Dyson, chairman of the Racial Profiling Prohibition Advisory Board said he didn’t appreciate the peer review’s being made public just hours before the meeting, and he hadn’t reviewed it fully.

“It’s the timing, you know. They had it for how long, and why are you presenting it now? You have to come away with the conclusion that it’s intended to raise questions about the report. That’s not a good sign. It’s an indication of being defensive or somewhat apprehensive or just resistant to wanting to make an adjustment or some change.”

Dyson said he understands the chiefs’ concerns about their departments being prematurely judged.

“I can appreciate that, not wanting to be on the list,” Dyson said  “It reflects on the department itself, but I would imagine there are people that don’t want to be profiled … so it’s six of one, half a dozen of another. So we set out to try to […] collect the data, analyze the data. If it’s there, the data will show it. If it’s not there, the data will show it, you know, so, I’m OK with that. And it’s my hope that they would be OK with it as well.”

He said examining departments where there isn’t any profiling to be found should be a benefit to the departments’ community relations.

“It removes any uncertainty. It kind of lets the public know that you’re being open about what it is and you can deal with being scrutinized by others, so I’m OK with it, and hoping they’ll be just as OK with it as I am. And the ultimate aim is that the public will win. That’s what I hope.”