![[PDF]](/pb-assets/Icons/adobe-pdf-icon-1498649896243.png){kind=icon}
Abstract
Glass's estimator of effect size, the sample mean difference divided by the sample standard deviation, is studied in the context of an explicit statistical model. The exact distribution of Glass's estimator is obtained and the estimator is shown to have a small sample bias. The minimum variance unbiased estimator is obtained and shown to have uniformly smaller variance than Glass's (biased) estimator. Measurement error is shown to attenuate estimates of effect size and a correction is given. The effects of measurement invalidity are discussed. Expressions for weights that yield the most precise weighted estimate of effect size are also derived.
REFERENCES
| Glass, GV Primary, secondary, and meta-analysis of researchEducational Researcher1976538 Google Scholar | SAGE Journals | |
| Glass, GV, Smith, ML Meta-analysis of research on the relationship of class-size and achievementEducational Evaluation and Policy Analysis19791216 Google Scholar | SAGE Journals | |
| Johnson, NL, Welch, BL Applications of the noncentral t-distributionBiometrika193931362389 Google Scholar | |
| Lehmann, EL, Scheffé, H Completeness, similar regions and unbiased estimationSankhyá195010305340 Google Scholar | |
| Lord, FL, Novick, MR Statistical theories of mental test scores1969Reading, MassAddison-Welsey Google Scholar | |
| Smith, ML, Glass, GV Meta-analysis of psychotherapy out-come studiesAmerican Psychologist197732752760 Google Scholar | |
| Walker, GA, Saw, JG The distribution of linear combinations of t-variablesJournal of the American Statistical Association197873876878 Google Scholar |
