In experimental evaluations of policy interventions, the so-called Bloom adjustment is commonly used to estimate the impact of the treatment on the treated. It does so by rescaling the estimated impact of the intention to treat--that is, the overall treatment-control group difference in outcomes for the entire experimental sample--by the…

Researchers and policy makers are increasingly dissatisfied with the "average treatment effect." Not only are they interested in learning about the overall causal effects of policy interventions, but they want to know what specifically it is about the intervention that is responsible for any observed effects. In the U.S., using…

To answer "what works?" questions about policy interventions based on an experimental design, Peck (2003) proposes to use baseline characteristics to symmetrically divide treatment and control group members into subgroups defined by endogenously determined postrandom assignment events. Symmetric prediction of these subgroups in both…

Using exogenous characteristics to identify endogenous subgroups, the approach discussed in this method note creates symmetric subsets within treatment and control groups, allowing the analysis to take advantage of an experimental design. In order to maintain treatment--control symmetry, however, prior work has posited that it is necessary to use…

In this article, we test an alternative approach to creating a counterfactual basis for estimating individual and average treatment effects. Instead of using control/comparison groups or before-measures, the so-called Counterfactual as Self-Estimated by Program Participants (CSEPP) relies on program participants' self-estimations of their own…

Use of multivariate analysis (e.g., multivariate analysis of variance) is common when normally distributed outcomes are collected in intervention research. However, when mixed responses--a set of normal and binary outcomes--are collected, standard multivariate analyses are no longer suitable. While mixed responses are often obtained in…

We tested for inflationary bias introduced through retrospective pretests by analyzing traditional pretest, retrospective pretest, and posttest evaluation data collected on a first-line supervisory leadership training program, involving 196 supervisors and their subordinates, across 17 organizational settings. Retrospective pretest ratings by both…

The paper discusses two common scenarios in which evaluators must conduct impact evaluations when working under budget, time, or data constraints. Under the first scenario the evaluator is not called in until the project is already well advanced, and there is a tight deadline for completing the evaluation, frequently combined with a limited budget…

This article uses propensity scores to identify subgroups of individuals most likely to experience a reduction in cash benefits because of sanctions in some of the programs that make up the National Evaluation of Welfare-to-Work Strategies. It extends program evaluation methodology by using propensity scoring to identify the subgroups of…

Outlines some nonrandomized constructed comparison strategies that can be used to evaluate interventions for children and illustrates their use for outcome variables that would be expected to change over age if no treatment were given. The proposed strategy consists of determining an expected age function for the dependent variable using pretest…