Topics in Evaluation

This article, which appeared in the GAO Review, Vol. 19, Issue 4, Fall 1984, discussed nonequivalent comparison group design, an evaluation method which is used to estimate the effect of a program by comparing outcomes after a program is implemented with outcomes which would have been expected in the program's absence. In such comparisons, the nonequivalent comparison design avoids the difficulty of making random assignments, but units are not equivalent in a number of ways. However, it was suggested that members of the groups can be chosen to minimize selection differences by matching or blocking, and statistical adjustments can be made to minimize the bias due to confounding variables.