Abstract

Meehl's statement "in most psychological research, Improved power of a statistical design leads to a prior probability approaching 1/2 of finding a significant difference in the theoretically predicted direction" (philosophy of science, Volume 34, Pages 103-115), Is without foundation. The computation of prior probabilities of accepting or rejecting a hypothesis presupposes knowledge of the prior probabilities that this hypothesis or any of its conceivable alternatives are true. As we do not have such knowledge, We cannot give any numerical values of prior probabilities of accepting or rejecting hypotheses in any statistical test procedure. Only topological statements are possible, As for example: when the region of acceptance of a hypothesis narrows, The prior probability that it will be accepted remains constant or increases