Is There Better Evidence on Mastery Learning? A Response to Slavin
James A. Kulik, Chen‐Lin C. Kulik, Robert L. Bangert‐Drowns · Review of Educational Research · 1990
Our purpose in writing Effectiveness of Mastery Learning Programs: A MetaAnalysis was to present as full a picture as possible of evaluation findings on mastery learning (Kulik, Kulik, & Bangert-Drowns, 1990). We wanted to get out all the facts on mastery learning so that we and others could draw conclusions about its effectiveness. We therefore described a variety of results from a variety of programs. For example, we covered cognitive and affective outcomes; effects of programs on course completion and student study time; findings from Grade 1 through college; and results from programs modeled after Bloom's system of Learning for Mastery (LFM) and Keller's Personalized System of Instruction (PSI). Further, we examined effects on average performance and variation in performance. And we looked at main effects and interactions. If we can judge from his published writing, Slavin's interest in mastery learning is narrower than ours (Slavin, 1987, 1990). He focuses exclusively on examination effects from precollege studies of Bloom's LFM. He thus ignores evidence on college applications of LFM, although more than half of all evaluations of LFM are carried out in colleges. He also ignores Keller's PSI, although two-thirds of the evaluations of mastery teaching involve PSI. Slavin does not look at affective outcomes of mastery teaching, nor does he look at its effects on variation in achievement. He gives scant attention to aptitude-achievement interactions and measures of instructional time. Thus our estimate is that Slavin's focus is on no more than 10 percent of the mastery picture. Although we believe that researchers need to see the whole picture to answer questions about mastery, we are restricting our comments here to the 10 percent of the story that interests Slavin: the effects of Bloom's LFM on precollege examination scores. We first try to determine whether Slavin agrees with us on the facts about studies of precollege LFM. Our conclusion is that he does not dispute the facts that we presented in our review. We then attempt to formulate some