Black and Hispanic drivers in Vermont are significantly more likely than whites to be searched by police during traffic stops, but less likely to be found with illegal contraband, according to a study released Wednesday by researchers at the University of Vermont.
UVM Professor of Economics Stephanie Seguino, one of the authors of the study, says she believes the statistical disparities are symptomatic of racial bias in the ranks of Vermont police agencies.
“When you find consistently that drivers of color are more likely to be searched but less likely to be found with contraband, often an inference is made that that is the result of some kind of bias — implicit bias or explicit bias,” Seguino said. “I think we have to make that inference.”
The study, called “A Deeper Dive into Racial Disparities in Policing in Vermont,” follows up on a paper issued last year by the same authors.
That 2017 report, called “Driving While Black and Brown in Vermont,” found that drivers of color are twice as likely to be pulled over as white drivers, and also subjected to roadside searches at a greater frequency.
The 2017 paper drew scrutiny from members of the police community who, according to Seguino, “had questions about the data … about the methodology.”
"I've had calls from African Americans who have been stopped and searched by the police. And this is a traumatic event for them. It's humiliating, and it erodes trust in the police." — UVM Professor Stephanie Seguino
The report released this week, Seguino said, “is meant to address those questions, and also to conduct much more sophisticated statistical analysis.”
Even accounting for confounding factors, like the age of the driver, or the reason for the stop, Seguino says the disparities in search rates persist.
According to the report, which is based on 2015 traffic-stop data from 29 police agencies, black and Hispanic drivers are as much as four times more likely to be searched than white drivers, despite the fact that those searches are half as likely to yield contraband.
Seguino says the data also refute the common stereotype that drivers of color are more likely to be trafficking heroin and cocaine into Vermont, from metropolitan areas in New England and New York.
Read the UVM-Cornell Report: A Deeper Dive into Racial Disparities in Policing in Vermont
She says the Vermont State Police did a special analysis of the 440 roadside searches that turned up some kind of contraband.
“Of all of the heroin, cocaine and opioids that turned up in searches in 2016, only white drivers were found with these drugs. There we no black or Hispanic drivers,” Seguino said. “So that’s inconsistent with the stereotype around drug trafficking, and I think it should give us pause to revisit those stereotypes.”
Seguino says the new look at 2015 traffic-stop data also reinforces the finding that drivers of color are more likely to be pulled over than white people. She says that disparity was evident in the data submitted by every agency the paper analyzed.
“So the fact that almost every agency showed the same pattern tells us there’s something about these racial categories that matters,” Seguino said.
"Law enforcement has been under the microscope on this topic and has been forced to not just learn about this topic but to be introspective about how these biases may manifest themselves in behaviors." — Colchester Police Chief Jennifer Morrison
Jennifer Morrison, chief of the Colchester Police Department and president of the Vermont Association of Chiefs of Police, says Seguino’s latest paper “has addressed some but not all of those concerns” about data quality in the previous study.
Morrison says the racial disparities in search rates, in particular, should prompt some introspection by law enforcement agencies as to the discrepancies.
“She makes a compelling case for disparities in search rates across races, and also for hit rates, and that is a very concerning piece of evidence that we need to continue to try to peel the layers of the onion on and get to the bottom of it,” Morrison said.
Morrison, however, says the Vermont Association of Chiefs of Police continues to harbor concerns about the quality of the data Seguino is using, and the methods of her analysis.
Morrison says the data used in the analysis is from 2015, the first year law enforcement agencies were required to collect it.
“So this was like getting our training wheels up under us, and we have said repeatedly that the data provided to Dr. Seguino … it’s not very good data,” Morrison said. “And that’s not Dr. Seguino’s fault, that is the fault of law enforcement. We are not trained as statisticians or data collectors.”
Morrison says she expects police will be able to put more stock in analyses conducted with data from 2016 and 2017, which she says should be available soon.
But Morrison also takes issue in some cases with the way in which Seguino uses the data that is already available. Seguino’s initial study from 2017, for instance, found that drivers of color are more likely to be arrested or ticketed than white drivers.
Morrison, however, says Seguino did not take into account the fact that drivers of color are more likely to have a suspended license, or no license at all, which may help account for that variance.
“I think those are confounding variables that have explanatory power, and that’s not explored in her research,” Morrison said.
Morrison says Vermont police agencies are already working to address the issue of implicit bias, through officer training.
“I think we have to start with an acknowledgment that every single human has implicit biases,” Morrison said. “Law enforcement has been under the microscope on this topic and has been forced to not just learn about this topic but to be introspective about how these biases may manifest themselves in behaviors. And really that’s what we’re concerned with, is behaviors.”
Seguino says Vermont law enforcement agencies deserve credit for the work they’ve done on implicit bias, and that those agencies are in many cases ahead of policymakers in other arenas.
But Seguino says the state has not moved with enough urgency when it comes to addressing the problem of racial bias, and that the consequences of unchecked racial bias, in law enforcement and other public institutions, has been corrosive.
“Those of us who are white have no sense of the impact of perceptions of maltreatment, whether it is kids in school that are disproportionately disciplined because they are black or because it is people who are riding on the roads and fell that they are being targeted by the police,” Seguino said. “I’ve had calls from African Americans who have been stopped and searched by the police. And this is a traumatic event for them. It’s humiliating, and it erodes trust in the police.”
Both Morrison and Seguino agree on one thing: The state needs better protocols in place for the collection and analysis of traffic-stop data.
Morrison says her organization is already working to improve those protocols. By the end of summer, she says, Morrison says Vermont will have a “best practices manual for race data collection, analysis and interpretation.”
She says that manual will be compiled as the result of a collaboration with the International Association of Police Chiefs,
Morrison says the state should also be providing resources to fund those data collection efforts, especially for smaller, more meagerly resourced police agencies.
