The emphasis of mastery learning instruction is unique in that concern is not only with raising the level of achievement but also with reducing the variability of achievement among the students instructed. The Glass effect size estimator, presently used an an index to measure the effectiveness of mastery learning experiments, is examined relative to these two conditions. This paper advances the thesis that the Glass effect size estimator represents only one dimension in the assessment of mastery learning experimental outcomes. The second dimension of effect, relevant to mastery learning outcomes, is the variability of the effect. The index to be proposed is directed at assessing this condition, in addition to the information provided by the effect size estimator. (Author)