Randomized experiments involving education interventions are typically implemented as cluster randomized trials, with schools serving as clusters. To design such a study, it is critical to understand the degree to which learning outcomes vary between versus within clusters (schools), specifically the intraclass correlation coefficient. It is also…

Background: Many university departments use students' evaluations of teaching (SET) to compare and rank courses. However, absenteeism from class is often nonrandom and, therefore, SET for different courses might not be comparable. Objective: The present study aims to answer two questions. Are SET positively biased due to absenteeism? Do…

Evaluation designs for social programs are developed assuming minimal or no disruption from external shocks, such as natural disasters. This is because extremely rare shocks may not make it worthwhile to account for them in the design. Among extreme shocks is the 2010 Chile earthquake. Un Buen Comienzo (UBC), an ongoing early childhood program in…

Recent meta-analyses of drug-court studies recognized the poor methodological quality of the evaluations, with only a few being randomized. This article critiques the design of the randomized studies from a statistical perspective. Learning points are identified for future drug-court studies and are applicable to evaluations both of other…

The determinants of sentencing are of much interest in criminal justice and legal research. Understanding the determinants of sentencing decisions is important for ensuring transparent, consistent, and justifiable sentencing practice that adheres to the goals of sentencing, such as the punishment, rehabilitation, deterrence, and incapacitation of…

Describes the history of the random experiment in the Home Office in the United Kingdom and demonstrates that research and the conduct of research is not an altogether rational process and that fashion, personality, and politics play a role in research policy and methodology choice. (SLD)

Determining which program evaluations in developing countries are appropriate/feasible for randomized experiments to plan and evaluate social programs is discussed. Five threshold conditions are defined. Reviews of experiments from Barbados, China, Colombia, Kenya, Israel, Nicaragua, Pakistan, Taiwan, and the United States illustrate the issues…

A methodology is developed for producing a well-designed problem in policy analysis. This evaluative context methodology is defined as an iterative process based on a series of checklist questions. Use of the methodology is demonstrated in a policy development project in New Zealand devoted to encouraging use of the Maori language. (SLD)