Speeding, the driver behaviour of exceeding the posted speed limit or driving too fast for conditions, has consistently been estimated to be a contributing factor to a significant percentage of fatal and nonfatal crashes. The U.S. Department of Transportation has instituted the Speed Management Strategic Initiative to seek more effective ways to manage the crash-related effects of speeding. In support of this initiative, this study conducted a detailed examination of recent crash data through the development of a speeding-related (SR) crash typology to help define the crash, vehicle, and driver characteristics that appear to result in a higher probability of SR crashes. Thus, the goal is to determine variables associated with SR crashes–such as what, where, when, and who–in order to provide guidance to the future development of new treatments and to more effectively target new and existing treatments. Recent Fatality Analysis Reporting System (FARS) and National Automotive Sampling System General Estimates System (NASS GES) data were used to answer these questions. Because these national databases only allow the use of a definition of SR that combines both exceeding the speed limit and too fast for conditions, two State databases (North Carolina and Ohio) were used to determine if different findings resulted from using the combined definition versus the exceeding the speed limit definition. Two analysis methodologies were used: (1) single-variable table analysis and (2) classification and regression tree (CART). In the first, for a series of both crash-related and vehicle/driver-related variables (e.g., crash type and age of driver), individual codes within each variable were examined to determine which showed an overrepresentation of SR crashes or SR vehicles/drivers (e.g., rear-end crashes for 16—19-year-old drivers). The second method involved CART analyses which automatically define which factors/variables are the most critical with regard to SR crashes or drivers and which combinations of variables/codes are the most important. Similar single-variable and CART analyses were also conducted for five high-priority subsets of the data (e.g., pedestrian crashes and intersection crashes). As might be expected, the results differed between fatal and total crashes, national and State, and among States. Few differences were seen in the results based on the two definitions. The single-variable table results were consistent with two earlier studies in indicating higher SR percentages in single-vehicle crashes, rural crashes, crashes on curves, night-time crashes, motorcycle crashes, as well as crashes involving young drivers, male drivers, drivers not using restraints, and drivers under the influence of alcohol. No consistent pattern of speeding was seen in either pedestrian or bicycle crashes or in work zone crashes. The CART results from the different databases were less consistent and more difficult to interpret. The crash-based results consistently identified single-vehicle crashes during adverse weather as a high-priority subgroup. The vehicle-based findings indicated almost no consistency across databases, with young male showing up more than other descriptors. (Author/publisher)
Samenvatting