The present study examines the construct validity and reliability of the Prison Group Climate Instrument (PGCI) in a sample of 77 adolescents placed in a Dutch youth prison and 49 adult prisoners living in a Dutch psychiatric prison with a therapeutic living group structure. Confirmatory factor analysis of a four-factor modelwith "repression," "support," "growth," and "group atmosphere" as first-order factorsand "overall group climate" as a second-order factor shows an adequate fit to the data, indicating construct validity of the PGCI. Cronbach's alpha reliability coefficients are good for all factors. The PGCI is a parsimonious instrument, enabling future research on group climate in youth prisons and secure forensic psychiatric institutions. The instrument can be used as an assessment tool for judicial interventions that use group climate to improve outcomes in delinquent youth and adult delinquents receiving treatment for psychiatric problems. (Published Abstract)