| This thesis evaluates the effectiveness of high-visibility crosswalks (HVC) to improve pedestrian safety at uncontrolled locations using the SHRP2 naturalistic driving study (NDS) data. This is accomplished by analyzing the driving behavior of SHRP2 participants at three uncontrolled locations in the Erie County, New York test site. At two intersections, traversal data were available both before and after HVC installation allowing for a before/after analysis. At the third location, only post HVC installation data were available. Because no pedestrian -- motor vehicle crashes were observed, crash surrogates (i.e., speed, acceleration, and gas pedal position) were used to evaluate driving behavior. Random effects linear regression models with fixed and random parameters were estimated for the change in the surrogate measures between predetermined benchmark points and the crosswalks, while controlling for a variety of other factors. The results show that presence of both the HVC and pedestrian crossing sign decreased the change in speed and acceleration between the benchmark and crosswalk points. In addition, there was a greater deceleration between the benchmark and crosswalk points after HVC installation. This exploratory work shows that HVCs have the potential to improve pedestrian safety and modify driving behavior, and that NDS data are useful for analyzing their effectiveness. |
