Persistent Misconceptions concerning Null Hypothesis Significance Testing

ABTRACT The Null Hypothesis Significance Test is a popular testing procedure used mainly in the social sciences and occasionally in the natural sciences. The procedure has been criticized on several grounds. Firstly, opponents have attacked the procedure itself, claiming that it either does not tell the researcher what he or she wants to know; and/or that the procedure is not logically valid. Secondly, opponents argue that the concept of statistical significance is flawed on the ground that is it either too arbitrary or, that statistical significance cannot confer substantive significance. This paper argues that critics of Null Hypothesis Significance Test either lack a proper understanding of how and why the test works, or they apply the test in an incomplete manner.