Charles Cowger's recent recommendation that statistical tests of significance be deemphasized or omitted from social work research is shown to be unwarranted on the basis of the arguments he presents. Four errors in Cowger's position are identified and the value of statistical tests of significance for social work research is established in terms of the unique evidence they provide the researcher. It is suggested that the misuse and misinterpretation of statistical tests of significance be alleviated by improved research education rather than by the de-emphasis or omission of such tests in published research.
No takes yet. Share an insight, caveat, or question.
Charles Glisson (1985) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: