Testing Different Methods of Estimating the Impacts of Worker Profiling andReemployment Services Systems

摘要

This evaluation is motivated by two goals: (1) to assess the reliability of theimpact estimates provided in the evaluation of the Worker Profiling and Reemployment Services (WPRS) programs, and (2) to compute revised estimates of the impacts of WPRS programs if a more accurate estimation method can be identified. The evaluation also provides general information on the accuracy of different methods for estimating impacts without random assignment. Under WPRS, states were required to establish systems for identifying Unemployment Insurance (UI) claimants likely to exhaust their UI benefits and referring them to reemployment services, such as resume preparation and training in job search methods. In an evaluation sponsored by the U.S. Department of Labor (USDOL), the impacts of WPRS were estimated by comparing UI claimants who were assigned to WPRS services (the treatment group) to claimants who were not assigned to WPRS services (the comparison group). Linear regression techniques were used to control for pre-existing differences between the two groups.