Samenvatting
Earlier mathematical work on accident proneness is reviewed and attention drawn to the danger of assuming that a Poisson distribution shows the absence of accident proneness while a negative binomial shows accident proneness. The results of the average time interval model are outlined.