Statistical comparisons of the crash characteristics on highways between construction time and non-construction time.

Many studies in the past reported that crash frequencies during construction time were higher than during non-construction time at the same highway sections. A question was raised whether this trend applies to work zones in Utah. Descriptive statistics (mean, standard deviation and histogram), a paired t-test, a two-way ANOVA, and a Tukey test were used to test whether the crash rates are statistically higher during construction time than during non-construction time at the same highway sections using 202 work zone sites located on Utah's highways. These highway sections were grouped into four highway classes and the crashes were grouped into six crash severity levels. It was found that the difference in mean crash rates between construction time and non-construction time was not statistically significant at the 95% confidence level, indicating that the trend of higher crash rates during construction time reported by previous work zone safety-related studies was statistically not supported by Utah's work zone crash records. The observance by contractors of Utah Department of Transportation's (UDOT's) guidelines for work zone traffic control may have helped them achieve a similar level of traffic safety during construction and non-construction times.