The statistical evaluation of accident experience eliminates the inconsistencies which result from the individual analysis of accident frequencies because it distinguishes between the number of events that might have occurred as a matter of chance and th9se incidents exceeding this level, while revealing the true relationship of the study interval to the overall route. It is of even greater importance to the general route analysis as a tool to pinpoint locations which warrant closer inspection and analysis. It should facilitate the institution of a state-wide traffic analysis program, which in turn should effect lower accident rates, more efficient use of personnel and monied, and improved public relations. The methods used in this study were based upon the methods described in the Manual for Application of Statistical Control Techniques to Analysis of Highway Accident Data by Dunlap and Associates, Inc. 1956. The techniques described there were modified to meet the needs of the present application. In the text of this paper, certain levels of probability of false detection are discussed; because of the modifications in the methods, these are only nominal probabilities of false detection.
Abstract