• Li Chen, Cynthia Chen, Reid Ewing, Claire E McKnight, Raghavan Srinivasan, and Matthew Roe.
• Department of Civil Engineering, Steinman Hall, City College of New York, New York, NY 10031, USA. lichen.cuny@gmail.com
• Accid Anal Prev. 2013 Jan 1; 50: 312-22.

AbstractTraffic fatalities and injuries constitute a major global public health problem and the United States has fallen behind other developed countries in traffic safety. Yet, New York City stands out as a traffic safety model in the nation with its low fatality rate and its significant reductions in various types of crashes. This study develops a safety framework that considers three principal axes that affect crashes: why, who, and where. While "why" concerns exposure, conflict, and speed, "who" and "where" consider the unique characteristics of the road users and the surrounding built environment. Grounded in this safety framework, the effectiveness of 13 safety countermeasures and street designs installed in New York City between 1990 and 2008 are evaluated using a two group pretest-posttest design. The potential regression-to-the-mean problem is addressed by applying the ANCOVA regression approach. The results show that signal related countermeasures that are designed to reduce conflicts: split phase timing, signal installations, all pedestrian phase, and increasing pedestrian crossing time, reduce crashes. Traffic calming measures, including road diets, are also found to have significant safety benefits. Countermeasures that are designed to alert drivers' cognitive attention, such as high visibility crosswalks and posted speed limit reduction signs, appear to have a lesser effect. The various safety countermeasures implemented in New York City considered all three important dimensions in the safety framework: why, who, and where. The study suggests these strategies are likely to contribute to the large reductions in crashes in New York City. We also demonstrate that a rigorous quasi-experimental design can be readily deployed in transportation safety evaluation studies.Copyright © 2012 Elsevier Ltd. All rights reserved.

15 keywords...

hide…