Of welfare-to-work programs evaluated by random assignment, two stand out as having exceptionally large estimated effects: one in Riverside, California, and the other in Portland, Oregon. The authors use data from 24 evaluations and the tools of meta-analysis to examine why. The findings...

In social experiments, individuals, households, or organizations are randomly assigned to two or more policy interventions. Elsewhere, we have summarized 143 experiments completed by autumn 1996. Here, we use the information we have gathered on these experiments and findings from informal...

"When evaluating a pilot or demonstration program, there are risks from drawing inferences from a single test. This paper reviews the experiences of replication efforts from demonstrations using randomized controlled trials in the initial evaluation and the replications. Although replications of...