Deficiency of education concerning the methodological issues in accepting null hypotheses

Abstract

Although there have been frequent warnings in the methodology literature about the serious problems which are entailed in accepting null hypotheses, empirical research often reveals a casual acceptance of null hypotheses simply because the obtained differences are not statistically significant. The present article examines this disparity between the methodology and the empirical research literature in order to suggest a specific cause for it in the training programs for psychological researchers. The analysis indicates that the training programs have concentrated on inculcating rules and narrow methods of hypothesis testing, without a proper emphasis on the comprehensive evaluation of all information, as demanded by modern views of the scientific method. The significance test is only one part of the multifarious scientific process for confirming hypotheses and theories.