Pedestrian Safety Guide and Countermeasure Selection System

 

Case Study No. 69

Advance Yield Markings

Halifax, Nova Scotia

Prepared by Ron Van Houten, Ph.D., Center for Education and Research in Safety, Dartmouth, Nova Scotia.

Problem

A “Yield Here To Pedestrians” sign.

Crosswalks on streets with multilane, uncontrolled approaches are often associated with a type of high-energy pedestrian crash termed a multiple threat crash (Snyder, 1972; Zegeer, et. al., 2001). Multiple threat crashes involve a vehicle in one lane stopping to allow a pedestrian to cross the street while the driver of an oncoming vehicle travelling in the same direction, in an adjacent lane, strikes the pedestrian. A major factor contributing to this kind of crash is the fact that the yielding vehicle stops (or slows) too close to the crosswalk, screening the pedestrian from the view of another motorist fast approaching in the lane that the pedestrian is crossing next.

Background

Problems with screening and multiple threat situations have always been a safety issue on urban streets and highways, and some rural roads. For example, buses and trucks have always been capable of totally screening the pedestrian; however the popularity of ever larger sport utility vehicles and minivans has increased the percentage of vehicles on the road that can completely screen the view of pedestrians crossing the street. Moreover, children and persons of short stature can be completely screened by even small- or medium-sized passenger cars.

Traditionally, crosswalks have been painted to increase pedestrian safety and level of service, where previously legal crossing areas were unmarked. Zegeer, Stewart and Huang (2002) compared 1,000 marked and 1,000 unmarked crosswalks in 30 U.S. cities. They observed no significant difference in crashes between marked and unmarked crosswalks with one exception: crosswalks on multilane roads which are not controlled by a traffic signal or stop sign were associated with significantly more crashes than unmarked crosswalks if the road had an average daily traffic volume (ADT) above 12,000. It has been suggested that marking crosswalks can lead to a false sense of security (Herms, 1972). However, behavioral data collected before and after crosswalks were installed at a number of sites contradict this hypothesis. These data show that marked crosswalks were associated with somewhat higher levels of pedestrian observing behavior by motorists and somewhat lower driving speeds (Knoblauch, Nitzburg, & Seifert, 1999).

Zegeer et. al. (2002) found that the greatest difference in pedestrian crash types between marked and unmarked crosswalks involved multiple threat crashes. This makes sense because multilane roads with a high ADT are more likely to have cars approaching in adjacent lanes than roads with a low ADT, and therefore, provide greater exposure for multiple threat crashes. Zegeer et. al. recommended that an unmarked crosswalk is insufficient to confer safety benefits on multilane roads with a high ADT, and should only be installed in conjunction with other traffic engineering improvements.

Solution

One treatment that addresses the issue of multiple-threat crashes is the use of yield markings placed 30-50 ft in advance of the crosswalks along with a "Yield Here To Pedestrians" sign placed adjacent to the markings. Data show that this treatment can produce a marked reduction in multiple threat conflicts.

Prior research (Van Houten, 1988; Van Houten & Malenfant, 1992, Van Houten, McCusker and Malenfant, 2001) has demonstrated that the use of advance stop lines or yield markings in conjunction with signs directing motorists to yield 50 ft in advance of the crosswalk will reduce motor vehicle/pedestrian conflicts and increase motorists yielding to pedestrians at multilane crosswalks with an uncontrolled approach.

When motorists yield in advance of the crosswalk, they enhance pedestrian safety in three ways. First, the yielding vehicle does not screen the view of motorists in the pedestrian's next lane of travel. Second, they reduce the likelihood that a vehicle travelling behind the yielding vehicle will cross the centerline to pass it striking the pedestrian. Third, they reduce the chance that an inattentive driver who strikes the yielding vehicle from behind will push it into the pedestrian.

In a recently completed study conducted in Halifax, Nova Scotia, Canada, 24 crosswalks were randomly assigned to a treatment or control condition. Following a baseline measurement period, twelve of the streets had advance yield markings and the "Yield Here To Pedestrians" sign installed, 23-65 ft in advance of the crosswalk. The remaining half of the crosswalks remained in the baseline condition and served as control sites. Each of the streets used in the study included multiple travel lanes in both directions or multiple lanes on a one way street. The posted speed limit was 30 mi/h, yet actual speeds were higher on some streets, up to 40 mi/h. Street settings included urban and suburban contexts.

The study cost was $25,000 and was funded by the Halifax Regional Municipality and Province of Nova Scotia. To ensure unbiased road user behavior, no public outreach or education was conducted at the same time that the signs were being tested.

Results

The sign and markings increased the percentage of motorists yielding to pedestrians and decreased the percentage of motor vehicle/pedestrian conflicts at all 12 sites. For the control crosswalks, driver-yielding behavior remained almost unchanged between the before- and after-treatment measurements. However the percentage of drivers who yielded to pedestrians at crosswalks with the added sign and markings increased from around 70-75 percent to around 80-85 percent. Further, vehicle-pedestrian conflicts remained nearly constant for the control sites but declined from about 10 to 15 conflicts per 100 crossings to under 5 conflicts per 100 crossings at the treatment sites.

Follow-up data collected six months after the markings and signs were introduced show no reduction in treatment effectiveness. These data are in accord with previous findings, which show that effects are maintained over time.

The success of the "Yield Here To Pedestrians" sign and advanced stop bar is underscored by the decision of the local government to retain the treatments installed for the study. While a formal user opinion survey was not conducted, data collectors and study principals received favorable reactions from roadways users and more people were aware of multiple threat crashes and conditions. This study only looked at changes in yielding rates and not in collision rates overall.

Contacts

Dr. Ron Van Houten
Western Michigan University
Department of Psychology

Dave McCusker
Director of Transportation
Halifax Regional Municipality
Halifax, Nova Scotia, Canada
Phone: (902) 490-6696

References

Knoblauch, R.L., Nitzburg, M., & Seifert, R.L. (1999). Pedestrian Crosswalk Case Studies. Center for Applied Research, for Federal Highway Administration.

Snyder, M.B. "Traffic engineering for pedestrian safety: Some new data and solutions." Highway Research Record, 406, 21-27, 1972.

Van Houten, R. (1988). The effects of advance stop lines and sign prompts on pedestrian safety in crosswalk on a multilane highway. Journal of Applied Behavior Analysis. 21, 245-251.

Van Houten, R. & Malenfant, L. (1992). The influence of signs prompting motorists to yield 50 ft (15.5 m) before marked crosswalks on motor vehicle-pedestrian conflicts at crosswalks with pedestrian activated flashing lights. Accident Analysis and Prevention, 24, 217-225.

Van Houten, R., McCusker, D., & Malenfant, J.E.L. (2001). Advance Yield Markings: Reducing Motor Vehicle--Pedestrian Conflicts at Multilane Crosswalks with Uncontrolled Approach. Transportation Research Record: Journal of the Transportation Research Board, 1773(1), 69-74.

Zegeer, C.V., Stewart, J.R., & Huang, H. (2001). Safety Effects of Marked vs. Unmarked Crosswalks at Uncontrolled Locations: Analysis of Pedestrian Crashes in 30 Cities. Transportation Research Record: Journal of the Transportation Research Board, 1773(1), 56-68.