Key points are not available for this paper at this time.
Abstract The custom of only publishing research when it reaches a certain degree of significance is likely to lead to errors, not through repetition of the same experiments, but over many different experiments.
Gordon Tullock (Tue,) studied this question.