The purpose of the present research was to establish relationships between traffic conflicts and accidents, and to identify expected and abnormal conflict rates given various circumstances. The data upon which the conclusions and recommendations are based were collected during the summer of 1982 at 46 signalized and unsignalized intersections in the Greater Kansas City area. The conclusions are limited to daytime (0700 to 1800) and weekday (Monday-Thursday) traffic, and to dry pavement conditions. Accident/conflict ratios have been statistically determined for several types of collisions for each of four types of intersections (signalized high volume; signalized medium volume; unsignalized medium volume; unsignalized low volume). These ratios can be applied to comparable intersections to obtain an expected accident rate of a specific type after the appropriate conflict data are collected. Also, statistical procedures were developed to determine conflict rate values that could be considered "abnormally" high.