TEAM CLASSIFICATION HAS THREE MAIN FEATURES: CASELOADS STRUCTURED AROUND HOUSING UNITS, CLASSIFICATION TEAMS CONSISTING OF A FEW SPECIALISTS, AND TEAM ACCEPTANCE OF THE FULL RANGE OF CASE MANAGEMENT RESPONSIBILITY. THE PURPOSE OF TEAM CLASSIFICATION IS TO DECENTRALIZE DECISIONMAKING REGARDING INDIVIDUAL INMATES, WITH A VIEW TO IMPROVING INTERPERSONAL RELATIONSHIPS AMONG STAFF AND BETWEEN INMATES AND STAFF. TEAM PARTICIPANTS INCLUDE THE INMATE HIMSELF, HIS CASEWORKER, PAROLE OFFICER, AND IMMEDIATE WORK SUPERVISOR, AND THE CORRECTIONAL SUPERVISOR ASSIGNED TO HIS HOUSING UNIT. THE TEAM'S CENTRAL TASK IS TO DEVELOP A PERSONALIZED PLAN FOR EACH INMATE AND TO CONTINUALLY ASSESS INMATE BEHAVIOR AND ATTITUDES. TO EVALUATE THE IMPACT OF TEAM CLASSIFICATION AND THE EXTENT TO WHICH THE PROGRAM HAS BEEN IMPLEMENTED AS PLANNED, DATA WERE GATHERED FROM INMATES AND STAFF ON THEIR PERCEPTIONS OF TEAM OPERATIONS, THE EFFECTIVENESS OF TEAM CLASSIFICATION, AND GENERAL CONDITIONS IN THE PRISONS. THE EVALUATION, WHICH WAS CONDUCTED AFTER THE PROGRAM HAD BEEN IN OPERATION FOR 2 YEARS, REFLECTS THE RESPONSES OF 27.6 PERCENT OF THE STATE'S INMATE POPULATION AND 67.6 PERCENT OF TREATMENT AND CUSTODY STAFF. THE FINDINGS SHOW THAT PARTICIPATION IN TEAM CLASSIFICATION IS IN ITSELF NOT ENOUGH TO PRODUCE THE PROGRAM'S INTENDED RESULTS. THERE MUST BE A PERCEPTION ON THE PART OF THE PARTICIPANT THAT TEAM CLASSIFICATION IS BEING CARRIED OUT EFFECTIVELY. WHEN WELL IMPLEMENTED, TEAM CLASSIFICATION IMPROVES INMATE ATTITUDES TOWARD STAFF, REDUCES INMATE ALIENATION, IMPROVES STAFF INTERRELATIONSHIPS, REDUCES ROLE CONFLICT AMONG STAFF, AND IMPROVES STAFF MEMBERS' OUTLOOKS ON THEIR JOBS AND ON INMATES. THE NEED FOR SYSTEMATIC, LONGITUDINAL EVALUATION OF THE PROGRAM IS NOTED. SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED.