Citation

Hallmark SL, Qiu Y, Pawlovitch M, McDonald TJ. J. Transp. Saf. Secur. 2013; 5(2): 131-147.

Copyright

(Copyright © 2013, Southeastern Transportation Center, and Beijing Jiaotong University, Publisher Informa - Taylor and Francis Group)

DOI

10.1080/19439962.2012.711438

PMID

unavailable

Abstract

Single-vehicle run-off-road crashes in Iowa account for 36% of all rural crashes and 62% of rural fatal crashes. Paved shoulders are a potential countermeasure for run-off-road crashes, and several studies have indicated that paved shoulders are effective in reducing crashes. However, the number of studies that quantify the benefits is limited. This article summarizes a before-and-after study to evaluate the impact of paved shoulders on crashes in Iowa to provide additional information for the Iowa Department of Transportation who had adopted a paved shoulder policy for higher volume roads. Bayesian inference for negative binomial-Lindley generalized linear models were developed to assess this relationship. Results suggested a relationship between total crashes and total right shoulder width, presence of a median, speed limit, presence of paved shoulders, season, and year-after intervention. The model for run-off-road crashes indicated a relationship with total right shoulder width, presence of a median, speed limit, presence of paved shoulders, season, and year after intervention. The model for single-vehicle run-off-road crashes indicated a relationship with total width of right shoulder, presence of a median, speed limit, presence of paved shoulders, season, and year-after intervention.