Cronbach, L. J., & Webb, N. (1975). Between-class and within-class effects in a reported aptitude * treatment interaction: Reanalysis of a study by G. L. Anderson. Journal of Educational Psychology, 67(6), 717-724.
http://dx.doi.org/10.1037/0022-0663.67.6.717
Reanalyzed a study by G. L. Anderson (1941) which found an interaction of drill and meaningful methods of arithmetic instruction with student ability and achievement. Drill was superior for "overachievers" and meaningful instruction for "underachievers" in 18 4th-grade classrooms. Pretest measures were the Minnesota School Ability Test and the Compass Survey Test. In a reanalysis to separate between- and within-class components of the outcome on aptitude regression, the Aptitude * Treatment interaction finding disappeared. An apparent interaction in the between-class analysis was dismissed as unreliable. No interaction was found within classes. The importance of separating class, individual, and individual-within-class interaction effects is discussed. (PsycINFO Database Record (c) 2016 APA, all rights reserved)