This study examined the effects of a 1-year high-visibility pedestrian right-of-way enforcement program on yielding to pedestrians at uncontrolled crosswalks, some of which received enforcement and some of which did not. The program included four 2-week enforcement waves supported by education and engineering components that increased the visibility of enforcement. The study produced five results: (a) enforcement led to a slow and steady increase in the percentage of drivers yielding the right-of-way to pedestrians over the year; (b) the program produced a large change in yielding over the course of the year; (c) the program produced higher levels of yielding to natural pedestrian crossing than to staged crossings, and the changes in both were highly correlated; (d) the effects of the program generalized to crosswalks that were not targeted for pedestrian right-of-way enforcement; and (e) the amount of generalization to unenforced sites was inversely proportional to the distance from sites that received enforcement.

This study examined the effects of LED rectangular rapid-flash yellow beacons (RRFBs) in uncontrolled marked crosswalks in three experiments. In Experiment 1, the RRFB system was evaluated with a two- and a four-beacon system at four multilane crossings. Results showed a marked increase in motorist yielding behavior over the baseline for the two-beacon system and a small but statistically significant further increase with the four-beacon system. The second experiment compared a traditional pedestrian overhead yellow flashing beacon and a traditional side-mounted yellow flashing beacon with the RRFB system. The results showed that the traditional overhead and side-mounted yellow flashing beacons produced a minimal increase in yielding, whereas the RRFB system produced a marked increase in yielding behavior. A third experiment examined the effectiveness of the RRFB system at 19 sites in St. Petersburg, Florida, as well as three additional sites, two in the suburbs of Chicago, Illinois, and one in the Washington, D.C., area. Results indicated that baseline daytime yielding behavior increased from an average of 2% to 86% at the 19 St. Petersburg sites and was 85% at the 2-year follow-up. A time-series intervention regression modeling was employed that estimated the parameters of the model based on a double bootstrap methodology. The results of this analysis confirmed a highly significant level change following the introduction of the RRFB that showed no sign of decay over time. Similar results were obtained at the District of Columbia and Chicago suburb sites. Probe data collected after dark revealed an even larger effect with yielding levels at the middle to high 90% level.

This study is a follow-up to a previous study that implemented high-visibility enforcement with social norming to produce a cultural change in driver yielding behavior. The objective was to determine the extent to which observed increases in driver yielding obtained in the previous study persisted over a follow-up period of nearly 4 years after the program of high-visibility enforcement intervention ended. The study involved limited enforcement and no new publicity. Observers collected data on staged and naturally occurring crossings at the same six sites at which enforcement took place in the previous study and at the same six spillover-effect sites (referred to as generalization sites in this report) where no enforcement had taken place. Observers employed the same observation procedures used in the original study. Results showed that yielding behavior continued on an upward trend with both the enforcement and generalization sites, exhibiting significantly higher rates of driver yielding during the follow-up period than at the end of the intervention period almost 4 years earlier. Yielding rates averaged 76.5% at the enforcement sites and 77.0% at the generalization sites. Thus, above and beyond the significant increase documented by the original study from before to immediately after the intervention, this study showed an additional significant increase in yielding from the end of the intervention to the follow-up period. The results suggest a fundamental change in driver behavior likely resulting from a tipping-point effect.

No abstract