This study tested the hypthesis that the reading gains of compensatory education pupils in grades one and four exposed to intensive test wiseness instruction would not differ significantly from the reading gains in a randomly assigned control group. Schools offering ESEA Title I classes (N=48) were randomly assigned to the treatment and control groups. Approximately four hundred students in each grade received twenty hours of test wiseness instruction as part of one hundred hours of supplemental reading instruction between the pretest and the posttest administration of the California Achievement Test. The implementation procedures of RMC Model A were rigorously observed and the NCE gains of the school units were analyzed separately for the two grades using a t test. The null hypothesis was supported in both cases. Extensive evidence was collected to show that the materials were used by the treatment teachers. Survey results indicated that teachers were quite positive about the commercially prepared mini-tests and felt that they would improve reading gains. (Author)