Education of disadvantaged children

There are at least two basic strategies for finding answers to the questions posed: what compensatory programs are “successful”, for which groups of children, and at what cost? The first strategy involves the design and execution of a series of controlled experiments in education; the second involves statistical analyses of existing compensatory education programs. In line with the first strategy, next year, the Follow Through program will fund a series of educational demonstration programs based on a “menu” of different curricula for children in the early elementary grades. Participating school districts will opt for one or another of the approaches offered, and will, in the course of the year, receive the advice and counsel of consultants in implementing the approach selected. A second—and complementary—basic strategy for finding out what works in compensatory education capitalizes on the wealth of information waiting to be collected on the operation of existing programs.

Since the spring of 1967, staff members of the Office of the Secretary of DHEW, the Office of Education, and a contractor—TEMPO, a division of General Electric Company—have been engaged in a pilot cost-effectiveness analysis of compensatory education programs. For the 11 districts studied, two distinct developments appear to have occurred in pupil performance: first, there appears to have been a slight decline in average pupil achievement in the sample schools. At the same time, there appears to have been a slight improvement in the achievement of pupils who are at the lowest achievement levels in their respective grades. The achievement data from the districts in the study indicate considerable variation in achievement results. The conduct and content of compensatory programs vary greatly. Preliminary analysis suggests that the amount of achievement increase is positively related to the level of expenditures for compensatory education programs.

To make real progress in assessing the effectiveness of alternative education programs and to overcome the difficulties discussed throughout this paper, we see the need to pursue two major courses of action. The first concerns the funding of controlled experiments introducing major variations into school programs, using comparable measures of program costs, inputs and effectiveness over time. The second is a frontal assault on evaluation in the form of a cooperative effort with a small number of interested States and local school districts to carry out a longitudinal study of compensatory programs.