Online Newsletter of the UC Berkeley Traffic Safety Center: Volume 3, No. 2, Summer 2006
When Crosswalks Work—and When They Don't
Charles Zegeer discusses his study on marked vs. unmarked crossings.
As part of the Traffic Safety Center's ongoing effort to foster information exchange across disciplines and institutions, Director David Ragland invited noted pedestrian safety expert Charles Zegeer to teach a class in his traffic safety course and address the Traffic Safety Seminar. (see also interview with Zegeer: link in sidebar).
Zegeer is associate director of the University of North Carolina’s Highway Safety Research Center and director of the Pedestrian and Bicycle Information Center.
Zegeer's lecture was closely modeled on a September 2005 report he prepared for the Federal Highway Administration (FHWA), "Safety Effects of Marked Versus Unmarked Crosswalks at Uncontrolled Locations" (see sidebar). This is a well-documented contribution to much-debated topic of the appropriate use of crosswalks, and how much safety they truly provide.
Pedestrians are legitimate users of the transportation system. The design issue is not "if" pedestrians are part of the equation, but "how" they can best be included.
The crosswalk investigation is part of a larger discussion about designing the transportation system so that all users, not just those in motorized vehicles, experience the greatest degree of safety, Zegeer emphasized.
"Pedestrians are legitimate users of the transportation system," Zegeer notes in the opening of the FHWA report. The design issue is not "if" pedestrians are part of the equation, but "how" they can best be included.
The study analyzed five years worth of data about pedestrian crashes at 1,000 marked crosswalks and 1,000 matched unmarked sites in comparable locations. All of them were "uncontrolled," meaning there was no signal or stop sign regulating traffic.
Although crosswalks are typically the first, minimal step toward enhancing pedestrian safety, there are no firm guidelines for when it is best to use them and with what combination of measures they work best.
A crosswalk is considered to be an extension of the sidewalk, whether it is marked or not.
The study notes that under the Uniform Vehicle Code, a crosswalk is considered to be an extension of the sidewalk, whether it is marked or not. Under that definition, mid-block crosswalks only exist if they are marked.
Another key document, the Manual on Uniform Traffic Control Devices, warns that crosswalks should not be installed "indiscriminately," but offers no specific advice on choosing where they should go.
The confusion has been compounded over the years by studies that have suggested that marked crosswalks may lead to increased pedestrian crashes under certain circumstances. (A commonly cited study is one carried out in 1972 in San Diego, CA, although many of its findings may not be as broadly applicable as once thought.) The most common explanation researchers have proposed is that pedestrians might be less cautious because they trust the crosswalk to provide more protection than is reasonable.
Zegeer's study sought to obtain a more precise comparison, by matching marked and unmarked locations, taking into account numerous factors, including traffic and pedestrian volume, geometry and the like.
Key results from the study were as follows:
Zegeer emphasized that the important finding was not that marked crosswalks should not be used, but that they should be used appropriately.
He cautioned that the study did not address possible causes for the findings and urged additional work on behavior, an expensive and time-consuming proposition. He noted that the TSC was undertaking work in this area. (See sidebar link, TSC crosswalk survey.)
Although it didn't address behavior explicitly, Zegeer's study provides some information about different types of pedestrians. The youngest and oldest pedestrians were most likely to use a crosswalk. The oldest pedestrians, those over 65, were consistently more likely to be involved in crashes, out of proportion to their exposure. The youngest pedestrians were not overly involved. The report's authors speculate that the older pedestrians may tend to cross higher-speed streets. Other possible explanations for their vulnerability are older people's reduced faculties and slower walking speeds.
A Unique Crosswalk "Threat"
In his talk, Zegeer emphasized the dangers of a phenomenon unique to marked crosswalks at uncontrolled intersections on multi-lane roads: the "multiple-threat crash."
This type of crash happens when one car stops for a pedestrian in a marked crosswalk, but a car in the neighboring lane continues to proceed through. The pedestrian often will not be able to see the approaching car in time to avoid being hit.
As Zegeer notes, "the marked crosswalk sets up the events" for this type of crash.
One possible solution would be to incorporate and "advanced stop lines," (pictured at let) forcing cars to halt substantially in front of the crosswalk markings in order to improve visibility. Other measures, more costly and harder to implement, could include sensors or automated signals to show that the crosswalk was being used (though caution is needed to avoid creating confusion.)
Ultimately, Zegeer said, "this doesn't mean you should take out all the crosswalks. But you need more substantial steps than just paint alone, especially on high-volume, multi-lane roadways."