Over the past 15 years, we have seen an increase in the use of cluster randomized trials (CRTs) to test the efficacy of educational interventions. These studies are often designed with the goal of determining whether a program works, or answering the what works question. Recently, the goals of these studies expanded to include for whom and under…

Much of the currently available evidence on the causal effects of public prekindergarten programs on school readiness outcomes comes from studies that use a regression-discontinuity design (RDD) with the age cutoff to enter a program in a given year as the basis for assignment to treatment and control conditions. Because the RDD has high internal…

In an effort to maintain internal validity while maximizing the generalizability of evaluation results, the theory-testing methodology that links program activities to outcomes (program theory) is proposed. To evaluate a program by explicating and testing the program theory is to perform evaluation as theory testing. (PN)

The design and conduct of a national evaluation study is discussed, demonstrating that a control group may not provide the no-cause baseline information expected. Resolution of this problem requires reexamination of paradigms, research practices, and policies, as well as the underlying real world constraints and views that generate them. (PN)

LISREL V computer program is applied to a weak quasi-experimental design involving the Head Start program, as a multiple analysis attempt to assure that differences between nonequivalent control groups do not confound interpretation of a posteriori differences. (PN)

Glass and Smith's (EJ 207 324) conclusion supporting small class size may be true, but it is not supported by their evidence. Their analysis does not recognize intervening factors and varying class sizes. (CP)

Problems that face evaluators, including field based evaluations, are discussed. The author states that open communication, access to data, and an ability to be accepted are paramount; knowledge of randomness, robustness and homoscedasticity may be secondary. (Author/PN)

A model is described which illustrates two general uses of multiple analyses to evaluate quasi-experiments: obtaining estimates of the treatment effect, and checking the validity of the experiment. (MH)

Conclusion about site-specific determinants of educational outcomes of the controversial longitudinal evaluation of the Follow Through programs (1976) are challenged again by reexamining the Direct Instruction model data. Reanalysis confirms earlier critiques and supports more recent research funding on the relationship between types of…

Among the guidelines offered for educational policy-relevant research on the effects of school integration on minority student academic achievement are: age at which to desegregate students; school racial composition; ethnic or racial groups to be bused; instruction for non-Black minorities; and the role of social class. (MH)