Abstract
The authors have been concerned about the fact that many economics education researchers concentrate on aggregate scores when attempting to measure the impact of various experimental approaches. Noting that the effect of an innovation might be positive for some students and negative for others, they argue that the members of an experimental group should not be treated as if they were all equal in skills and abilities. Then they suggest disaggregation procedures and give the results of their own analyses of various subgroups of students who were exposed to computer-assisted instruction. Their findings lend support to those who favor disaggregated analyses.