Paradoxes, Contradictions, and Illusions

Abstract

There is no contradiction between a powerful signif icance test based on a difference score and the neces sity for reliable measurement of the dependent mea sure in a controlled experiment. In fact, the former requires the latter. In this paper we review the conclu sions that were drawn by Humphreys and Drasgow (1989) and show that Overall's (1989) "contradic tion" is an illusion derived from imprecise language. Index terms: analysis of covariance, baseline correc tion, control of individual differences, difference scores, measurement of change, reliability of the mar ginal distribution, statistical power, within-group reli abilities.