The journal of irreproducible results

Abstract

"Statistical significance," writes William L. Hays, "is a statement about conditional probability, nothing else. It does not guarantee that something important, or even meaningful, has been found." Hays, of course, is only one of many who have bemoaned this glaring shortcoming of conventional statistical tests. Given the increasing number of research reports vying for the attention of any serious scholar, surely the time has come for statistics which will aid us in ascertaining the importance and meaningfulness of our findings. The purpose of this paper is to provide such statistics.