In experimental designs with nested structures, entire groups (such as schools) are often assigned to treatment conditions. Key aspects of the design in these cluster-randomized experiments involve knowledge of the intraclass correlation structure, the effect size, and the sample sizes necessary to achieve adequate power to detect the treatment…

In social policy evaluations, the multiple testing problem occurs due to the many hypothesis tests that are typically conducted across multiple outcomes and subgroups, which can lead to spurious impact findings. This article discusses a framework for addressing this problem that balances Types I and II errors. The framework involves specifying…

A method is presented for analyzing data collected in a multiple group time-series design. This consists of testing linear hypotheses about the experimental and control group-means. Both a multivariate and a univariate procedure are described. (Author/GK)

The desirability of combining tests of theory with evaluations of treatment modalities is argued in an investigation of the effectiveness of a juvenile diversion program. Using a true experimental design (with randomization), recidivism analyses dependent on court record data failed to demonstrate the relative superiority of any of three treatment…

Statistical methods for program evaluation with nonexperimental data are reviewed with emphasis on circumstances in which nonexperimental data are valid. Three solutions are proposed for problems of selection bias, and implications for evaluation design and data collection and analysis are discussed. (SLD)