Design, Sites, and Data Sources
The study used a random assignment research design. Returning citizens who agreed to participate in the study were assigned, at random, either to a program group that had access to the DOL-funded programs or to a control group that did not have access to the DOL-funded programs but could seek out other services in their communities. A total of 24 grantees in 18 states participated in the evaluation.
The evaluation team followed both groups for at least two years, using administrative records and surveys to measure employment and recidivism outcomes. The random assignment process ensures that the two groups were comparable at the start of the study, so any significant differences that emerge over time can be attributed to the services provided by the DOL grantees.