Hypothesis Testing in Relation to Statistical Methodology

Abstract

THE SHORTCOMINGS in the methodology of statistical hypothesis testing used in educational and psychological research have been emphasized repeatedly in recent behavioral science and statistical literature (Binder, 1963; Edwards, Lindman, and Savage, 1963; Grant, 1962; Lubin, 1962; McNemar, 1960; Mowrer, 1960; Nunnally, 1960; Rozeboom, 1960; Savage, 1957). This chapter reviews the salient points of the criticisms. Since this issue of the REVIEW marks the first time an entire chapter has been devoted to the statistical methodology of hypothesis testing, a brief account of several theories of statistical inference is included to provide a background for evaluating the rationales of significance tests compared with other methods for statistical inferences, as well as the modifications in the function of the null hypothesis in testing statistical hypotheses. The problems in statistical inference which have been associated with the widespread use of significance tests are reviewed. The limitations in and the effectiveness of significance tests as methods for informative inference are described. The applications of interval estimation contrasted with significance tests in the investigation of hypotheses and models and in the development of a body of empirical data are summarized.