Digging deeper into racial disparities in Connecticut traffic stops June 14, 2016

Andrew Ba Tran
Data Editor

The annual report on traffic stops in Connecticut is one of the nation’s most ambitious attempts to examine potential racial profiling by police. This week Trend CT will dig deeply into the report's data and methodology to illuminate how it identified police departments that its authors said warrant further scrutiny.

The study, by Central Connecticut State University's Institute for Municipal and Regional Policy, identified dozens of departments for showing varying levels of racial and ethnic disparities in their traffic-stop patterns.

No single measure or statistic provides an accurate picture of any potential disparity, so researchers designed a series of measures to develop a more complete portrait. Even so, researchers were careful to say that evidence of racial and ethnic disparities do not, by themselves, provide conclusive evidence of racial profiling.

They also said it was likely that specific departments were driving the statewide trends and that some departments' disparities were affected by a small number of individual officers.

Researchers wanted to measure traffic stops not only against the number of  minority residents in a town but also account for commuter traffic and for drivers who work but do not live in a town. Points were assigned to a department if they exceeded certain thresholds, as illustrated in the chart below.

Among the measures researchers examined were:

  • How much a town's minority population and minority traffic-stop rates departed from state averages.
  • How the number of minorities pulled over in a town compared to the town's estimated minority driving population, including commuters and those driving through.
  • How the number of minority residents pulled over in a town compared to their representation in the town population.

It should be noted that some parts of this analysis depend on demographic and population data which isn't available for some non-municipal police departments, like campus police and state trooper jurisdictions.

Statewide average comparison

To understand how the comparisons to statewide averages work when examining a particular town, we can look at Wethersfield.

The statewide average for minority residents (16 and older) was 25 percent. Of the 31 towns that exceeded the statewide average for the proportion of minority drivers stopped, 20 also had populations that exceeded the statewide average.

The police department in Wethersfield stopped minority drivers 47 percent of the time, though the state average was less, at 31 percent. Also, Wethersfield's minority population is just 12 percent compared to the state average of 25 percent. The combination of those gaps is 31 percentage points. Towns with combinations of more than 10 were given a point.

Using this standard, 12 towns were found to have a gap of more than 10 points between their percentage of  minority drivers stopped and the percentage of their minority driving age population.

All but three of the towns identified for pulling over minorities disproportionately were towns that border municipalities with minority populations that exceed the state average.

Five towns were found to have high disparities in the rate of black drivers pulled over. In four of the five towns— Woodbridge, Trumbull, Orange and Wethersfield— more than 90 percent of the black drivers stopped were not residents of the town. The statewide average for stopped black drivers who were not residents of the town in which they were stopped was less than 60 percent. Stratford was the fifth town with a high disparity for black drivers.

For Hispanic drivers, Wethersfield, Newington, Darien, and Berlin had the highest disparities in rates.

Six additional towns, Fairfield, Wilton, Orange, Trumbull, Meriden and New Britain, fell just below the 10-point threshold.

Estimated Driving Population Comparison

Researchers sought to account for nonresidents likely to be driving in a town because they work there by estimating the makeup of the driving population during typical commuting hours.

For example, Bloomfield has a proportion of minority residents – about 62 percent – but the commuters coming into the town to work are estimated to be about 73 percent white.

This analysis looks at the estimated driving population compared to the rates of traffic stops conducted during peak commuting hours in the morning and afternoon.

Wethersfield, once again, ranks high when measuring the gap between the percentage of traffic stops involving minorities and estimated minority percentage of the driving population. The department had gaps of 26 percentage points for minorities, 17 for Hispanics, and 11 for blacks.

In a couple of dozen towns, the disparity was negative, meaning that more whites were stopped than expected out of the estimated driving population

Resident only comparison

Researchers then did a comparison of whether minority residents of the town a department covered were pulled over in numbers disproportionate to their representation in the town population. They cautioned that disparities in this comparison might be explained by a greater police presence in high-crime areas.

Meriden, East Hartford and New Britain showed the greatest disparity between the proportion of minority residents stopped and the proportion of minorities living in town. New Haven and Bloomfield pulled over more resident black drivers compared to the resident population.