Published online by Cambridge University Press: 04 July 2016
In this paper we examine the hypothesis that aeroplane accidents are grouped, ie that individual accidents are not randomly spaced in time. While it has not been possible to find any explicit statements of the non-randomness hypothesis, it seems to be a widely-held belief. In its least precise form, the belief is that as soon as there is one aeroplane accident, there will be “several more within a relatively short period of time”. Clearly one does not expect equal or even near-equal spacing between aircraft accidents; the interesting question is what amount of bunching leads to the rejection of randomness as an adequate description of the underlying distribution. And if non-randomness is encountered in practice, how is it explained or interpreted?