Human Factors Assessment of Pedestrian Roadway Crossing Behavior

Discussion

An overarching goal of the present study is to determine which environmental factors influence where pedestrians cross the roadway. Pedestrian crossing behaviors were recorded and coded over a 2-week period at seven different locations. It was hoped that these data would help to identify factors that influence pedestrians to cross at unmarked non-intersection locations. As was discovered in phase 1 of this study, the number of non-intersection crossings is quite low. As a result, modeling and predicting pedestrian behavior is difficult using statistical methodologies. To more accurately describe pedestrian crossing behaviors, additional data are collected in phase 3 and described in more detail later. At present, the data and trends for the seven locations from phase 2 are discussed.

Crossing Location

As in phase 1, the majority of pedestrian crossed at marked intersection locations. Although none of the locations in phase 2 were considered outlying values in terms of crossing locations, three locations (11, 12, and 15) did have greater unmarked non-intersection crossing percentages than those in phase 1. These locations are discussed in more detail here.

Overall, Location 11 had very few crossings. However, this does not imply that there was no pedestrian traffic in this area. Rather, pedestrians did not have reason to cross the roadway at the specific location. While not specifically coded, it did appear that the majority of people crossing in the vicinity of Location 11 crossed at the far non-signalized marked crossing. This may have been the result of the increased perceived control of crossing when desired. This crossing is also quite close to a bus stop waiting area. Because of the small sample size (17 crossings), a small number of incidents can artificially influence the percentage of different types of crossings. As a result, Location 11 is excluded from further discussions.

Overall, Location 12 is quite representative of many city blocks in many cities. However, several factors, when combined, may contribute to a high number of unmarked non-intersection crossings. The first of these factors is that there are a large number of businesses on both sides of the street with parking located on both sides of the street. The result is that people frequently park on one side of the street and walk to a business on the opposite side. Further, the businesses along this block are mixed. This can result in people walking from their workplace to the other side of the street to obtain food. Although in many cases pedestrians may opt to walk to the marked crossing to cross the roadway, this portion of Georgia Avenue Northwest is relatively narrow, with a relatively low AADT. The low AADT provides gaps to cross the roadway that many pedestrians find acceptable. It is possible that pedestrians perceive that they will have more time to safely cross the roadway in these gaps than the forced 20-s crossing time (a crossing rate of 3.23 ft/s) at the marked intersection.

At Location 15, 36.55 percent of the pedestrians crossed at an unmarked non-intersection location. These crossings took place primarily in one specific area between the two marked crossings on Pennsylvania Avenue Southeast. On the south side of the relevant block, there are two primary pedestrian traffic originators/destinations: a Metro (subway) station and a bus stop that services multiple bus lines. On the north side of the relevant block, there is a large residential neighborhood. As such, many people travel between their homes and the public transportation stations/stops. Figure 45 shows there is a short section of roadway on the north side of Pennsylvania Avenue Southeast that connects westbound traffic to an intersecting street. This creates another vehicle intersection just west of 8th Street. While this is likely designed to alleviate vehicular traffic, it creates a “shortcut” for pedestrians. Pedestrians can see a clear pathway between the Metro station/bus stop areas to the residential area. In using this path to cross, pedestrians are able to make half of their crossing and wait on the grassy area that divides the north and south portions of Pennsylvania Avenue before completing the crossing, as needed. This is a tactic that would likely increase pedestrians’ perceived control of their crossing. As figure 45 shows, pedestrians do indeed use this pathway regularly, so much so that the grass in the median area has been worn away.

At Location 13, most of the crossings did occur at the marked intersection (89.76 percent). However, more than half (62.57 percent) of these took place at least partially during the don’t walk phase. This extreme difference warrants discussion. There are several factors in this area that would draw people to cross at the marked intersection rather than an unmarked non-intersection. As figure 39 shows, there is a large park just to the south of the relevant block of H Street. This park provides the opportunity to walk “diagonally” through several blocks without encountering vehicle traffic. There are, however, small barriers around the park that inhibit entrance to the park outside pre-specified areas. As a result of pedestrians’ perceived advantages of walking through the park, they may also perceive an advantage to cross at the marked intersection to gain easier access to the park. This however, does not explain why pedestrians may have been persuaded to cross outside the walk phase.

As was mentioned in phase 1, it appears that the rate at which pedestrians must cross during walk phase may influence when and where pedestrians decide to cross the roadway. At Location 13, pedestrians must cross at a rate of 5 ft/s, which greatly exceeds the MUTCD recommendation of 3.5 to 3.7 ft/s.⁽²⁸⁾ At a rate of 5 ft/s (10 s to cross the roadway) it is possible that many pedestrians are simply not able to complete the crossing during the allotted time and are forced to complete the crossing during the don’t walk phase. In addition, other environmental factors in this area may influence pedestrians to cross during the don’t walk phase. There are two general expectation violations involved in this crossing. Because of southbound traffic on Connecticut Avenue turning left, pedestrians crossing west of the marked crossing of interest are allotted more time to cross the roadway. In other words, pedestrians waiting to cross in the marked intersection where crossings were recorded, were likely able to see pedestrians making a similar crossing in the adjacent crosswalk. Furthermore, the pedestrians may have been able to see the adjacent walk sign and perhaps presumed that there was an error or other problem with their respective walk sign and entered the roadway. In this case, pedestrians’ expectations to be able to cross at the same time as the pedestrians in the adjacent crossing were violated. Pedestrians crossing at this marked intersection can also view the time countdown on the perpendicular crossing light (i.e., the amount of time left for pedestrians travelling perpendicularly to cross). They might expect their own crossing light to change to walk shortly thereafter and preemptively begin crossing the roadway. This behavior may be particularly risky for pedestrians crossing from north to south because they cannot see the vehicles making left turns (and may not make appropriate evasive maneuvers). As mentioned in phase 1, it is also possible that pedestrians may not be able to see that vehicles have a turning arrow and inappropriately begin their crossings. This certainly could have been the case for pedestrians crossing from south to north in this area.

Pedestrian Yielding

In total, 20 of the 3,189 crossings (.62 percent) involved pedestrian yielding. Overall the mean percentage of pedestrians who yielded to vehicles was .86 (i.e., the mean of the percentage of pedestrians who yielded to vehicles at each location). There were no significant differences between locations when looking at total percentage of pedestrian yielding behaviors.

Only at Location 10 were there instances of pedestrians crossing in the marked intersection. However, the value of .22 percent of the total crossings within the marked intersection is within the range of observed values in phase 1. As a result, these crossings are not discussed further here. There were also no significant differences in the percentage of pedestrians who yielded to vehicles in unmarked non-intersections. All values were similar to those found in phase 1.

Vehicle Yielding

In total, 226 of the 3,189 crossings (7.09 percent) involved a vehicle yielding to a pedestrian. Overall, the mean percentage of vehicles that yielded to pedestrians was 9.71 (i.e., the mean of the percentage of vehicles that yielded to pedestrians at each location). This value is quite similar to phase 1.

Overall, Location 10 (28.41 percent) had a significantly greater percentage of vehicles that yielded to pedestrians. This high percentage is likely the result of turning vehicle traffic. Further exploration of this location reveals that the majority of these occurred in the marked intersection during the walk light phase. Vehicles originating from Van Ness Street have a green light to turn north on to Connecticut Avenue while pedestrians have a walk light phase. This then, would commonly result in vehicles yielding to pedestrians in the marked crosswalk.

There were also no significant differences in the percentage of drivers who yielded to pedestrians based on the area where pedestrians crossed the roadway.

Yielding Comparison

In phase 2, there were no significant differences in the percentage of pedestrians who yielded to vehicles and the percentage of vehicles that yielded to pedestrians.

Evasive Pedestrian Actions

In total, 71 of the 3,189 crossings (2.23 percent) involved an evasive pedestrian action. This value was similar to that for phase 1 (2.83 percent). The mean percentage of pedestrians who took evasive actions was 2.32 (i.e., the mean of the percentage of evasive pedestrian actions at each location). The percentage of evasive actions was fairly consistent across locations and across crossing types. At the present time, the data do not suggest any clear environmental reason for these evasive pedestrian actions.

Evasive Vehicle Actions

In total, 2 of the 3,189 crossings (.06 percent) involved an evasive vehicle action. This value is similar to that for phase 1 (.02 percent). The only evasive vehicle actions took place in Location 12. This, combined with the small total number of vehicle evasive actions, does not provide the opportunity to make inferences at this time.

General Discussion

In the present set of data, non-intersection crossings are relatively rare. This makes it difficult to make predictions about where pedestrians will cross the roadway. In this second phase, there were no recorded close calls or near misses. In phase 3, researchers recorded crossings at five sites in vivo. The results of these observations are discussed in the next section.