Series # :
  RERS 2001
Title :
  Testing Different Methods of Estimating the Impacts of Worker Profiling and Reemployment Services Systems
Release Date :
Abstract :
  Prior to the implementation of the Worker Profiling and Reemployment Services (WPRS) program, USDOL sponsored a demonstration to test different program models that are consistent with the regulations governing WPRS. In 1995, the Job Search Assistance (JSA) Demonstration was implemented in the District of Columbia and in selected counties in Florida. Because the demonstration was based on the random assignment of eligible claimants to treatment and control groups, impacts were estimated by comparing treatment group members to control group members. Random assignment ensured that the pre-existing differences between the two groups were negligible.

In this evaluation, data from the JSA Demonstration in Florida are used to mimic the treatment and comparison samples from the WPRS evaluation, and to test different methods of estimating impacts from these samples. These methods include the regression method used in the WPRS evaluation, but also include variants of the matching methods used in other evaluations. Matching is designed to select a subgroup of comparison group members who are similar to treatment group members. Impacts are then estimated by comparing treatment group members to the subgroup of similar comparison group members.

Publication Author(s)
Mathematica Policy Research, Inc.
  • Robert B. Olsen
Full Text Document(s)
Additional Information

Hard copy available: YES