Equations Which Include Reliability of the Dependent Variable for Estimating the Power of a Group ANOVA Design

Abstract

The purpose of this study was to obtain regression equations for estimating the power of the completely randomized, fixed-effects, one-factor Analysis of Variance (ANOVA) for two groups from four independent variables: effect size, alpha level, sample size, and the reliability of the dependent measure. Of particular interest was the degree and nature of the contribution of the reliability variable (the reliability with which the dependent variable in the ANOVA design is measured) to estimating a design's power.