Sunday, September 11, 2016

Add racial profiling at traffic searches to law enforcement 'early warning' triggers

The Austin Statesman's investigative team today published an excellent analysis of DPS racial profiling data regarding discretionary searches conducted by troopers, coupled with an analysis of video from traffic stops which generated racial profiling complaints. This effort culminated months of team reporting on the topic and responded to DPS Col. Steve McCraw's ostensible rebuttal to their earlier findings. For background:
In December, the Statesman found that African-American and Hispanic motorists were searched more often when stopped, and contraband found less often for Hispanic drivers than white drivers.
Earlier this year, DPS Director Steve McCraw testified before state lawmakers that troopers do not engage in racial profiling and encouraged the public to instead look at the individual cases in which officers are accused of racial profiling. There have been about 40 in the last five years, and the Statesman requested the dashcam videos from those incidents through a request under the Texas Public Information Act.

While DPS said they haven't found evidence to support any allegations of racial profiling, some of the videos raise more questions about how the department handles those investigations than they answer about racial bias in policing. Highlights of them are below, along with Sean Collins Walsh's report.

In addition to the videos, the Statesman looked again at stop data since 2009 to study how frequently individual troopers searched motorists, and how those search rates differ by race. Reporting and analysis of 14 million traffic stop records by Eric Dexheimer, Jeremy Schwartz and Christian McDonald found that 35 percent of officers studied searched minority motorists at more than twice the rate of white motorists, and most found contraband less often as a result of those searches.
In 2001, the Texas Legislature required law enforcement agencies performing traffic stops to collect data which would allow an assessment of racial disparity. Your correspondent was part of a team back in the day successfully pushing the bill at the Lege and working with local departments afterward on implementation, so I'm particularly gratified to see that these data are proving useful in contemporary public debates after all this time.*

Documenting racial disparity at traffic stops
Some disparity by race at traffic stops may be explained by factors like deployment patterns, higher poverty rates among minorities, and the rates at which people of different races are arrested for crimes. That's why the most useful possible analysis from these data focus on after-stop activities, especially the rate at which drivers of different races are searched and the contraband "hit rate" that results (although that's not the only useful analysis which may be performed). An academic statistician who studies these matters told the paper that “officers who search minorities at twice the rate of whites are statistically significant 'high disparity officers.' A search rate that is four or more times higher, he added, is 'pretty astronomical. … It’s a very egregious ratio.'” With that in mind, check out these top-line results:
  • 35 percent of the 1,138 troopers included in the analysis searched black and Hispanic motorists at least twice as often as white drivers.
  • 231 of the officers who searched black and Hispanic motorists at two times or more the rate at which they searched white drivers were less likely to find contraband while searching the minority drivers.
  • 65 DPS officers searched minority drivers at least three times more often than the white motorists they stopped yet found contraband less often.
  • 16 officers searched minority motorists more than four times as often as Anglos, with lower contraband hit rates.
The key datapoint here isn't just the disparity in searches but in outcomes: There could be numerous explanations for top-line disparities in the number of stops and searches, but if troopers use their discretion to search minorities' vehicles more often and find contraband less frequently, then they're not basing the extra searches on factors which empirically target crime.

At a minimum, then, that should be a red flag signalling a need for retraining or at least supervisory correction. But neither DPS nor any Texas department of which Grits is aware uses this valuable metric as part of their "early warning system" to identify officers with behavioral problems. (See a discussion of Austin PD's minimalist early warning system here.)

Grits should also mention that most local agencies don't even collect the contraband hit-rate data discussed here. It's probably time to go back into Texas racial profiling statute to adjust those data-collection elements now that we have a decade-and-a-half of experience regarding what information from that dataset is and isn't particularly useful.

Closed records on complaints
Col. McCraw used his discretion to release video from stops resulting in racial profiling complaints, but the paper's analysis of complaint videos failed to corroborate McCraw's evaluation. "Some show mundane police work or motorists making accusations at seemingly well-intended officers. But others show troopers asking to search vehicles or aggressively questioning motorists despite there being little apparent reason for suspicion."

Equally important, the article identified gaps in public information which prevent a more probative, external analysis: "The department withheld the motorists’ complaints as well as the investigative reports that resulted from them, which the Statesman also requested, for all but one of the incidents, citing a state law that makes confidential records from investigations that do not result in significant disciplinary action."

At one time, open records laws governing Texas law enforcement were among the strongest in the nation. But beginning in the late '80s, culminating in the codification of a bad Texas Supreme Court ruling in 1997, Texas' open records statutes on police discipline and law enforcement generally were gutted like a fish, letting agencies cover up wide swaths of misconduct which previously could be discerned by reporters or interested members of the public through public information requests. (Earlier this year, Grits expressed nostalgia for the remarkable level of since-lost law-enforcement openness in Texas in the halcyon days of my youth.)

The Statesman's big contribution to the racial profiling literature over the last year has been to dig deep at the Department of Public Safety, whereas all prior analyses have focused on local departments. That's both a good way to make it a statewide story without analyzing hundreds of departments and also useful because DPS issues more tickets than any other agency and collects the full panoply of racial-profiling data, whereas most agencies do not. Really good, important story; congrats to everyone involved.

* Note to Will Harrell, Ana Correa, Molly Totman, Dwight Steward, and Eva Owens: Ain't this a hoot?


Anonymous said...

Do we know the race of the officers making the stops? There is an assumption that white officers are making the stops, but is that the case?

Anonymous said...

4:55pm - great point. I had a black female officer once tell me that she could get away with a lot more on a traffic stop than a white male officer could.

Gritsforbreakfast said...

Is it a great point? Do motorists care about the race of the officer discriminating against them? I tend to think not.

Anonymous said...

I don't think motorists care but knowing the race of the officers can point to how to solve the problem.
If black officers are as likely to stop and hassle black motorists as white officers, then that suggests a different problem and different solution than if its almost entirely white officers stopping blacks.

Anonymous said...

Well, until the borders are protected what do you expect?

Gritsforbreakfast said...

@texastankee, sounds to me like the same problem. If officers violate civil rights on behalf of the government it's a problem no matter what race they are. And the means for stopping them are the same either way.

@9.54, none of this has anything to do with immigration. Your comments speak more to your own obsessions them to the issue at hand.

Baron of Greymatter said...

So why isn't Col. McCraw being charged with perjury, for testifying under oath that TxDPS does not profile?

Anonymous said...

Grits - do we know if the stats include traffic encounters where events like ether DWI arrests, motorists with arrest warrants or a traffic accident then cause a vehicle inventory? If a motorist is arrested or transported to a hospital in the case of a traffic accident, officers will usually perform an inventory if there is no one else to release the person's vehicle to besides a wrecker. An inventory is a search.

Gunny Thompson said...

It appears that the Good Old Boys, in Brown Shirts, feel that they are above the law and will not suffer any adverse actions from complaints filed by citizens. In very rare cases, when violating citizens protection, they may be terminated, but are not restricted from reemployment by a different agency. They take care of their own! Terrorist motorist has a right to ask why he/she is being stopped and object to his/her vehicle being searched. In order to pass muster, the pattern and practice of unlawful stops and searches must be authorized by a statewide policy emanating from a politically accountable governing body. See,,, Holt v. State, 887 S.W. 2d at 19 [emphasis added] In the Sandra Bland case, it is clear that the Racist Brown Shirt's actions must be prosecuted for felony assault while in possession of a handgun. If he was Black, he would still be in jail!

Anonymous said...

Gunny you make Grits seem to the right and that's hard to do.

Gunny Thompson said...

Tanks Anonymous @ 08.12.00. Shouldn't we all be for right? I'm in appreciation to Grits for the form and its views. And you for your response. There is still hope for change.

Anonymous said...

I'm curious where those 35% of troopers are stationed/assigned. If you're assigned to the border region, your search stats are going to be skewed because the majority of your contact is going to be with Latinos. If you're assigned to Abilene, the majority of your contact is going to be with Anglos. Do these stats really show that these troopers searched minorities at a higher rate overall or based on those minority populations. There is a difference. Stats don't lie, but they don't always tell the whole story.

It does appear, though, that TxDPS needs to revamp their searches if the hit rate is that low, especially given how much we always hear about how understaffed they are.

Anonymous said...

The late 80's is when republicans became the majority in Texas...

john said...

eeyikes, the trouble with that is I'm white, and though I don't have contraband, NOW I'm statistically more likely to have some??
But the trouble is the cops will absolutely try to frighten or anger whoever they stop into resisting or anything that inflates the charge, increasing their chance of revenue enhancement. Those of you that believe it's about "enforcing law" are living in a dream world, albeit highly publicized & politically correct.
The Cops War on America has long been promoted by giving them outlandish discretion and protection--even immunity. Of course, judges, or anyone in power, also seek that. We The Poor People no longer have real representation--court-appointed (OR ANY) attorneys who took oaths to the guild Bar and the Court certainly do not count--we're a distant third place, to them.

DLW said...
This comment has been removed by the author.
DLW said...

This problem gained momentum with the SCOTUS ruling in Whren v US. The case abolished the pretext stop doctrine. A pretext stop or arrest is an objectively valid stop for an improper reason. It occurs where the police employ a stop based on probable cause or reasonable suspicion as a device to search for evidence of an unrelated offense for which probable cause is lacking.

They want to see if you might have drugs in the car but they don't have any probable cause to believe you do. So they stop you for driving in the left lane or following to closely or not giving a turn signal for 100 feet before the turn or any other real or imagined Transportation Code offense. After the stop PC might develop but it is much more likely that an intimidated motorist will consent to a search without knowing that he could refuse.

Anonymous said...

I'll kindly ask again.

Grits - do we know if the stats include traffic encounters where events like ether DWI arrests, motorists with arrest warrants or a traffic accident then cause a vehicle inventory? If a motorist is arrested or transported to a hospital in the case of a traffic accident, officers will usually perform an inventory if there is no one else to release the person's vehicle to besides a wrecker. An inventory is a search.

DPS investigates thousands, if not tens of thousands of traffic accidents each year as well as arrests tens of thousands of fugitives and DWIs. Do the vehicle inventories of these encounters count as searches? Stolen vehicle recoveries with or without a driver could also cause an inventory/search.

So if a minority driver is stopped for speeding and the driver has an arrest warrant and is arrested, the trooper inventories the car. Would it be the trooper's fault if the next 4 or 5 arrests he/she made followed that similar pattern with a variety of warrant arrests, traffic crashes, and/or DWI arrests? An inventory is a search.