Scores of intersections throughout the metro area are marked with white lines meant to prevent crashes by keeping drivers from pulling too far into intersections or blowing through stop signs.

But in a first-of-its-kind study carried out over 14 years, the Minnesota Local Road Research Board found the lines, also known as stop bars, neither improved intersection safety nor are all that effective at getting drivers to stop. The study, which examined data collected in five metro suburbs from 2004 to 2018, also found that drivers commonly stopped on top of the bar or up to 10 feet past it.

On the positive side, stop bars were found to be somewhat effective at slowing drivers down as they approached an intersection even if they didn't come to a full stop.

"Stop lines do affect the driver's approach behavior, so if they are used, they need to be used appropriately," said John Hourdos of the University of Minnesota Traffic Observatory and the study's lead investigator. "Since they are not globally beneficial, they can be used on locations where vehicles are approaching at high speeds to slow down the unavoidable roll-and-stoppers."

Stop bars, which are 12 to 24 inches wide, are commonly used at intersections with bad sight distances to prevent drivers from creeping out beyond the curb to peer around a corner. They also are used when the proper location where a vehicle should stop is not obvious.

The lines have long been a part of the traffic landscape, but many drivers may not understand their function. Though motorists are required by law to stop at stop signs, there is no law governing stop bars. Their placement has been haphazard, used in some places but not others, Hourdos said.

Researchers wanted to find out if their presence led to a reduction in crashes and learn how stop bars influence driver behavior. They collected data from intersections in Richfield, Edina, Roseville, St. Louis Park and Golden Valley that have stop signs controlling traffic in two directions. They also collected video from 16 different intersections to assess driver behavior before and after a stop line was put down.

The results showed stop bars alone did not reduce crashes. The study also found that placement plus sight distances and speed limits did influence driver behavior — the more space between the stop bar and the curb, the more likely drivers were to ignore the stop line.

In Richfield, for example, drivers on westbound 73rd Street at Portland Avenue stopped at the curb line about 50% of the time before a stop bar was put down, but none encroached beyond the curb line into the bike lane. After a stop bar was added, about 25% of drivers stopped beyond the curb and encroached into the bike lane.

"Drivers will always try to find the spot that provides the best visual information regarding conflicting traffic," Hourdos said.

Except in certain situations where there are bad sight lines, stop bars may not be worth the cost. A single stop bar costs anywhere from $250 to more than $1,000 depending on material, equipment and the number of lanes they cover. There also are annual maintenance costs.

Edina stopped putting in stop bars in 2016, and as result of pavement repair projects there are only about 30 left. Having fewer of them might be better, said Nick Bauler, Edina's traffic safety coordinator.

"Stop bars can be beneficial if used properly based on intersections with sight-line concerns," he said. "If they are installed too frequently in unnecessary areas, drivers tend to disobey them in general, even when truly needed."

Tim Harlow • 612-673-7768