ESTIMATION OF EFFECT SIZE FROM A SERIES OF EXPERIMENTS INVOLVING PAIRED COMPARISONS

This article develops the distribution theory for a Glass-type (1976) estimator of effect size from studies involving paired comparisons. We derive an unbiased estimator of effect size following Hedges's (1981) work on combining effect sizes from independent samples. It is shown that studies involving mixtures of paired and independent samples can also be combined in this way. Approximate confidence limits for effect size are derived. The method is illustrated using an example from psychiatric research.