DOES CORRECTIONAL TREATMENT WORK? A CLINICALLY RELEVANT AND PSYCHOLOGICALLY INFORMED META‐ANALYSIS *

TLDR

Literature on criminal conduct and prior reviews indicate that neither sanctioning without rehabilitation nor service without clinical principles succeeds in reducing recidivism, and a psychologically informed perspective may guide renewed service, research, and conceptual efforts. The authors contend that effective correctional service requires targeting higher‑risk cases, addressing criminogenic needs, and matching cognitive or behavioral treatment styles to client learning preferences. They applied these principles to 154 studies of juvenile and adult correctional treatment, producing phi coefficients that quantified treatment effects on recidivism. Appropriate correctional service yielded a mean phi of .30, significantly higher than unspecified service (.13) and markedly better than inappropriate service (−.06) or non‑service sanctioning (−.07), and was effective across juvenile and adult settings, pre‑ and post‑1980, randomized and non‑randomized designs, and diversionary, community, and residential programs, though attenuated in residential contexts.

Abstract

Careful reading of the literature on the psychology of criminal conduct and of prior reviews of studies of treatment effects suggests that neither criminal sanctioning without provision of rehabilitative service nor servicing without reference to clinical principles of rehabilitation will succeed in reducing recidivism. What works, in our view, is the delivery of appropriate correctional service, and appropriate service reflects three psychological principles: (1) delivery of service to higher risk cases, (2) targeting of criminogenic needs, and (3) use of styles and modes of treatment (e.g., cognitive and behavioral) that are matched with client need and learning styles. These principles were applied to studies of juvenile and adult correctional treatment, which yielded 154 phi coefficients that summarized the magnitude and direction of the impact of treatment on recidivism. The effect of appropriate correctional service (mean phi = .30) was significantly (p <.05) greater than that of unspecified correctional service (.13), and both were more effective than inappropriate service (−.06) and non‐service criminal sanctioning (−.07). Service was effective within juvenile and adult corrections, in studies published before and after 1980, in randomized and nonrandomized designs, and in diversionary, community, and residential programs (albeit, attenuated in residential settings). Clinical sensitivity and a psychologically informed perspective on crime may assist in the renewed service, research, and conceptual efforts that are strongly indicated by our review.