Using an alternative to classical statistics, this paper reanalyzes results from three published studies of interventions to increase employment and reduce welfare dependency. The analysis formally incorporates prior beliefs about the interventions, characterizing the results in terms of the distribution of possible effects, and generally confirms the earlier published findings.
Relying on 427 classroom observations conducted over a three-year period, this study traces changes in teachers’ instructional practices in the First Things First schools.
Planning for the Jobs-Plus Demonstration
Statistical Implications for the Evaluation of Education Programs