Pennsylvania State Police was warned about possible racial bias in car searches. The agency’s answer? End the research.

Last fall, under scrutiny for its decision years earlier to stop tracking the race of drivers who get pulled over, the Pennsylvania State Police pointed to a series of independent studies that officials said proved there was no consistent bias in the way troopers did their jobs.

But a review of those studies reveals a far more nuanced picture of what was actually occurring, raising new questions about why the data collection was halted and how troopers are trained to prevent people of certain races from being unfairly targeted.

The findings, contained in 2,000 pages of research produced by the University of Cincinnati and obtained by Spotlight PA under the state public records law, indicated no consistent evidence over the study period that troopers stopped drivers, issued citations or made arrests based on race.

But they did reveal a persistent problem after stops had occurred: Year after year, according to the reports, troopers were roughly two to three times more likely to search black or Hispanic drivers than white drivers. At the same time, the researchers found, troopers were far less likely to find contraband on black and Hispanic drivers compared to white drivers.