What If There Were No More Bickering About Statistical Significance Tests
Joel R. Levin · 1998
Questions and concerns are directed to those who advocate replacing statistical hypothesis testing with alternative data-analysis strategies. It is further suggested that: (1) commonly recommended hypothesis-testing alternatives are anything but perfect, especially when allowed to stand alone without an accompanying inferential filtering device; (2) various hypothesis-testing modifications can be implemented to make the hypothesis-testing process and its associated conclusions more credible; and (3) hypothesis testing, when implemented intelligently, adds importantly to the story-telling function of a published empirical research investigation. From the local pubs to our professional "pubs, " everyone in social-science academic circles seems to be talking about it these days. Not that there's anything wrong with talking about it, mind you, even to a more practically oriented crowd such as the readership of this journal. But as with the "gates " of Washington politics on the one coast and the Gates of Washington state on the other, when do we stand up and say "Enough already!"? When do we decide that ample arguments