Study 1
Cave and colleagues (1993) evaluated JOBSTART using an experimental design with random assignment to treatment and control groups across 13 sites from 1985 to 1988. The 13 sites ranged from community-based organizations to schools to Job Corp Centers. The six community based sites were Buffalo, N.Y.; New York, N.Y.; Hartford, Conn.; San Jose, Calif.; Chicago, Ill.; and Corpus Christi, Texas. The three vocational school sites were Pittsburgh, Pa.; Monterey Park, Calif.; and Denver, Colo. The three Job Corps Center sites were located in Atlanta, Ga.; Los Angeles, Calif.; and Phoenix, Ariz. The last site was in Dallas, Texas.
Youths who applied for the program were randomly assigned to receive JOBSTART services or a control condition. The evaluation sample across all sites was 2,312 youths who received follow-up surveys at 12, 24, 36, and 48 months after random assignment. The final analyses were conducted on 1,941 youth (84 percent of the initial sample) for whom there was complete follow-up data at all four assessments over the 4-year study. The treatment and control groups consisted of 988 and 953 youths, respectively. The treatment group participated in the JOBSTART program, while the control condition left youth free to access other services in the community without receiving anything from the JOBSTART program. There were no significant differences between the treatment and control groups in any background or demographic variables such as gender, ethnicity, or parental status.
The outcomes of interest included education, employment, earnings, receipt of public assistance, and drug use. These outcomes were measured by a self-reported survey at the 4-year follow up. The researchers also conducted a cost–benefit analysis to determine whether the investment in JOBSTART yielded subsequent returns to taxpayers. The study authors conducted subgroup analyses to determine the potential effects of gender on program outcomes.