Citation

Chen W, Cooper PJ, Pinili M. J. Saf. Res. 1995; 26(1): 9-18.

Copyright

(Copyright © 1995, U.S. National Safety Council, Publisher Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Driver adjudication systems that assign penalty points to various traffic law infractions, and establish a level of point accumulation at which action to suspend drivers' licenses is taken, are a feature of the activities of virtually all motor vehicle authorities. However, in many cases the number of penalty points assigned to each infraction type is simply a qualitative assessment without any empirical foundation. The ordering of infractions by "severity" is often a very subjective process that tends to result in a linear scale of relatively narrow range. The purpose of the research reported in this paper was to assess the relative impact on future crash-involvement risk of a number of different infractions and also of accident history.A total of 1,998,347 B.C. drivers' records was examined. Logistic regression analysis was employed to identify drivers who were most likely to have one or more at-fault accident involvements in a postperiod on the basis of their preperiod records of at-fault accident involvements and convictions. The results showed a consistent increase in postperiod accidents per driver with increasing preperiod numbers of both crashes and convictions. It was also found that prior at-fault accident involvements were a better predictor of future at-fault accident involvements than were prior convictions, and that up to 23% more high-risk drivers could be identified by making use of prior atfault accident involvements as opposed to convictions alone. Right-of-way infractions such as "failure-to-yield" and disobeying a traffic signal were found, after accidents, to be the type of preperiod incidents most strongly associated with postperiod crashes.