No rehabilitation programs deal exclusively with youths convicted of serious crimes. However, this report describes behavior-changing programs that include serious juvenile offenders, programs based on: (1) psychology/psychiatry including behavior modification; (2) sociology/social work, (3) schooling, and (4) vocational education. (Juvenile offenders are nearly all school failures and lack occupational skills.) The four study teams reached similar conclusions. Successful programs were flexible and heuristic; maximized juveniles' freedom of choice and participation in their own rehabilitation; utilized clear tasks structured for early and frequent success, with significant incentives; and were as much as possible like the real-world environment where the new behavior was to be lived. Staff members set good examples. The report recommends to the National Institute for Juvenile Justice a hypothesis-testing strategy that studies and documents both planned and naturally occurring variations in treatment. Standard program evaluations are irrelevant, because programs are seldom implemented as planned. "Status offenses" should be decriminalized, and juvenile defendants allowed due process.