To follow trends of truck accident involvement rates requires reliable information on truck accidents and travel. Procedures for estimating truck accident involvement rates and their confidence limitson the basis of variabilities inherent in the sample design of the trb-proposed national monitoring system (nms) are presented. Formulas for computing confidence limits of national and state truck accident involvement rates per mile of travel are given for any level of disaggregation. The quality of truck accident and travel data that may be expected from implementing the nms, together with consistent estimation of confidence limits of accident involvement rates, would represent significant improvement over truck safety statistics available from existing data programs. This paper appears in transportation research record no. 1322, Large vehicle safety: transit and trucks 1991
Abstract