A review of research conducted between 1975-1986 has been inconclusive. Many programs appear successful, but methodological problems often make results questionable. Particularly in nonbehaviorally based treatment programs (e.g., psychodynamic, cognitive), claims of effectiveness have little empirical validation. Experiments evaluating inhouse behavior modification and token economy programs are usually more methodologically sound. Studies of such programs suggest that behavioral techniques may be valuable in influencing residents' short-term, inhouse behaviors. However, there are no conclusive data about maintenance of behavior change following treatment. 22 references.