Job Training Partnership Act (JTPA) National RCT
Mathematica Policy Research / US Department of Labor · United States · 1994
Summary
The JTPA evaluation is one of the most important experiments in US social policy — not because it confirmed what worked, but because it documented what didn't. The null result for adult men and the negative short-term effect for youth directly contradicted the program's rationale. The design contributed to a shift in federal workforce policy: from generic skills training toward employer-linked sectoral programs. The contrast between JTPA's null effects and Year Up's 30% earnings effect illustrates why mechanism matters: generic training produces credentials; sectoral training produces employment. The JTPA study established a baseline for what 'rigorous evaluation of job training' requires and influenced the design of every subsequent workforce program evaluation.
Research question
"Does federally funded job training increase earnings for adult men, adult women, and out-of-school youth?"
Methodology
Intervention
Eligible applicants to JTPA programs randomly assigned to receive services (classroom training, on-the-job training, job search assistance) or to control group barred from JTPA for 18 months; services varied by site and individual assessment
Assignment
Randomized controlled trial (individual, multi-site)
Sample size
20,601 participants across 16 program sites
Primary outcome
Earnings at 30 months; employment rate; hours worked
Effect estimate
Adult women: +$1,500 earnings at 30 months (+12%); adult men: no statistically significant effect; out-of-school youth: negative effect during training; female youth: small positive effect
Decision
JTPA replaced by Workforce Investment Act in 1998 with emphasis on employer linkages and sectoral approaches; null and negative results for men and youth drove significant design changes in federal workforce policy
Result
Mixed
Adult women: +$1,500 earnings at 30 months (+12%); adult men: no statistically significant effect; out-of-school youth: negative effect during training; female youth: small positive effect
Evidence strength
Strong
Randomized trial, replicated across multiple sites or studies.
Replication status
Replicated
Institution
Mathematica Policy Research / US Department of Labor
Location
United States
Year
1994
Policy area
Education
Mechanism
Human capital