Research and evaluation, according to a classic report authored by Steve Bell of Urban Institute, needs to be “part of the justification, oversight, reappraisal, and ongoing improvement of every employment and training program” (p. 2).
This report considers optimal ways to measure program impacts on diverse populations of clients and the economy overall—and to judge their cost-effectiveness in terms of both client and broader social benefits. It offers alternative designs when random assignment is not possible.The report focuses on gold-standard ways to measure national, state, and local programs’ impacts. “Three main messages stand out:
- Information on policies and programs taken from research and evaluation must be rigorous and thorough;
- Different methods work best to gather information of this sort in different policy and program settings; and
- All the research techniques involved pose large technical challenges and utilize substantial amounts of data, necessitating a major, sustained commitment of resources by the government to be successful in improving the nation’s workforce investment system.
The discussion developing these themes covers several topics. Section 1 discusses the role of research in policymaking and the types of information gathering and evaluation activity to be considered. The best way to do research in different workforce investment program and policy areas comes next, in Section 2.... [Section 3] shows how random assignment experiments can be conducted to rigorously measure impacts in almost all cases if the determination is to get the best information to guide policy—even when barriers to the approach appear to arise from the very nature of the intervention involved…. Recommendations continue in Section 4 in two other important areas of strategic planning: making good evaluation choices when the best choice has not been achieved for some reason, and supporting improvements to existing tools for measuring the impacts, costs, and benefits of employment interventions in second-best situations in the future.” (pp. 2-3)