In Connecticut, African Americans are twice as likely to have their cars searched for contraband than white drivers during a police stop, even though the majority of contraband found in traffic stops were found in cars belonging to white drivers, a study conducted by Central Connecticut State University found.
The study looked at more than 360,000 traffic stops over an eight-month period in Connecticut, an found that while African Americans make up less than 8 percent of the state’s population, they make up 14 percent of traffic stops. Across the state, African-American drivers have their cars searched for contraband a little over twice as often as white drivers, but 77 percent of contraband found during traffic stops was found in cars driven by white drivers.
Since the study began, nine police departments have been investigated over complaints of racial profiling. For five of the investigated departments, the percentage of stopped drivers who were African American was twice their driving populations in those communities. Only one of the investigated departments, the Farmington Police Department, searched a higher percentage of cars driven by white drivers for contraband than those driven by African Americans. In the other eight, African American drivers had their cars searched for contraband between 1.5 and three times more than white drivers. In all nine of the towns, the majority of contraband found in searched cars belonged to white drivers. Two police departments, the East Windsor and Farmington departments, found contraband in none of the stopped cars driven by African Americans.
While these nine police departments were the only ones to be investigated during the course of the study, other communities within the state have shown even higher levels of disparity. In Plymouth, African Americans make up about 0.92 percent of the driving population but 4.95 percent of traffic stops, a rate over five times their percentage of the driving population. In the town of Union, African American drivers are stopped at a rate of over 30 times their share of the driving population.
Connecticut is obviously not alone when it comes to these levels of disparity in traffic stops. A study from early this year found that almost 10 percent of traffic stops in its neighbor, Rhode Island, involve African-Americans, while they only make up about 7.5 percent of the population. In 2013, a North Carolina-based study found that African American drivers made up 30 percent of traffic stops, despite making up only 21 percent of the population and that African American drivers were 77 percent more likely to be searched than white drivers. Studies in both Minnesota and Illinois have found that African Americans are more likely to be pulled over than white drivers but that contraband was more likely to be found in cars driven by white people. The Minnesota stud found that if the rates drivers were pulled over matched the population, 22,500 more white drivers and 18,800 fewer African American drivers in the state would be pulled over that year.
The study in Connecticut, which was conducted by Central Connecticut State University’s Institute for Municipal and Regional Policy, looked at data from 168 Connecticut communities and 102 different police agencies. The institute is still gathering a year’s worth of data and will release its complete findings early in 2015. So far, the study has looked at 366,060 separate stops.
Of all the communities that the study looked at, only 13 did not stop African Americans at a rate higher than their percentage in the state’s driving population. Only three of these communities used municipal police departments; the other ten used either residential state troopers or the state police.
The study isn’t perfect; there is still four months of data left to be collected and analyzed, and there are concerns that certain departments are not reporting all their traffic stops. For example, Redding’s population is less than 7,000 and has made about 1,800 traffic stops, while Stamford, with a population of about 86,000, has reported less than 650, showing that there are some large reporting discrepancies. However, despite these issues, the margin of error across the study was a relatively low 3.73 percent.
Amelia Rosch is an intern for ThinkProgress.