This study examines the relative effectiveness of two means of analyzing the pre-test/post-test control group experimental design. Samples were randomly drawn from a standardized normal population and assigned to one of the four cells of the design. A set of experimental differences were induced in the post-test experimental cell. Each case was subjected to a t-test of pre-test and post-test difference scores, and an analysis of covariance (ANCOVA) using the pre-test scores as the covariate. Analysis of 200 cases in each of four sample sizes indicated that: (1) the ANCOVA is a more sensitive test having greater power to detect the induced differences but also being more susceptible to a Type I error; and (2) the findings were essentially the same when cases involving heterogeneity of regression were included and when they were excluded. (Author)