Review of Educational Research
Meta-analytic techniques for summarizing information about the relationship between school resources and student performance are capable only of addressing the narrow and uninteresting hypothesis that resources are never used effectively by schools. The data show clearly that resources are sometimes used effectively, although this happens infrequently and there is no description of the circumstances under which resources are used effectively. This article relates the basic evidence on school effectiveness to the specific application of meta-analytic methods employed by Greenwald, Hedges, and Laine (1996). Their analysis, suffering from the narrowness of the inquiry inherent in their statistical methods, is also based on a very highly selected sample of results that biases their analysis precisely toward their conclusions. As a result, their summary of existing work provides a distorted and misleading view of the potential implications of school resource policies. Both detailed econometric evidence and aggregate performance of U.S. schools point toward serious problems with inefficient use of resources. This evidence in turn suggests that lack of resources is not the largest problem facing schools and that more fundamental reforms are needed in schools.