In assessing the validity of a norm-referenced model used in evaluating large-scale federal educational programs for disadvantaged children, gain estimates were shown as approximately equal with randomized control group model estimates compared by retrospective analyses of two databases. (Author/CM)

Meta-analytic techniques were used to study the overall impact of the Title I program on student achievement and to examine the effects of mediating methodological and programmatic factors. Data from 17 Federal studies with 657 effect sizes indicate a modest impact of Title I that increased as the program matured. (SLD)