Revisiting the Influence of Crash Report Forms on Work Zone Crash Data

Researchers present an analysis of work zone fatal crashes nationwide to assess possible underreporting due to differences in how information about a work zone crash is captured on standard state crash reporting forms. The possible effects of differences in crash report forms on work zone crash statistics were first identified in the mid-1990s, by using data from the Highway Safety Information System. The influence of different crash report forms on work zone crash data contained in the Fatality Analysis Reporting System (FARS) is examined. An investigation of the data contained in FARS from 1998 to 2000 indicates a statistically significant dependence between the way in which work zones are denoted on a state's crash report form and the percentage of fatalities that are coded as occurring in a work zone. From this analysis, it appears that existing data may underreport the number of fatalities that occur in work zones nationally by as much as 10%.