Technical Issues In Measuring Scholastic Improvement Due To Compensatory-Education Programs

Evaluation studies which require measurement of the effect of an intervention have been plagued by problems due to the effect of regression to the mean. In this paper, we review some of the efforts made to overcome problems due to this phenomenon in the evaluation of compensatory programs for reading and mathematics. We report on a study of Chapter 1 projects in the School District of Philadelphia where a new index of performance was created to overcome the effects of regression to the mean.