THE MMPI, USED TO ASSESS INMATES' PERSONALITIES, DIFFERENTIATES 10 TYPES OF OFFENDERS. OPERATIONAL DEFINITIONS AND GUIDELINES FOR PROFILE CLASSIFICATION PRESENTED FOR EACH TYPE ENABLE THOSE FAMILIAR WITH THE MMPI TO CLASSIFY OFFENDERS' PROFILES EASILY AND RAPIDLY. A COMPUTER PROGRAM CAN ALSO BE MADE AVAILABLE TO CLASSIFY MOST OF THE PROFILES ENCOUNTERED, WITH CLINICAL JUDGMENTS RESERVED ONLY FOR THE MORE DIFFICULT CASES. DURING THE 8 YEARS OF RESEARCH TO DEVELOP AND VALIDATE THE SYSTEM THUS FAR, THE 10 GROUPS HAVE BEEN FOUND TO DIFFER SIGNIFICANTLY ON A BROAD ARRAY OF BACKGROUND FACTORS, SOCIAL AND DEMOGRAPHIC VARIABLES, PERSONALITY CHARACTERISTICS, ATTITUDES, AND BEHAVIOR PATTERNS. PROGNOSTIC VALUE IS ALSO SHOWN IN THAT THE 10 TYPES OF OFFENDERS DIFFERED MARKEDLY IN THEIR SUBSEQUENT PATTERNS OF BEHAVIOR AND ADJUSTMENT IN PRISON, IN THEIR PROPENSITY FOR VIOLENCE, AND IN THEIR RESPONSE TO INSTITUTIONAL PROGRAMS. A FOLLOWUP STUDY CONDUCTED SEVERAL YEARS AFTER THEIR INCARCERATION REVEALED MARKED DIFFERENCES IN EVENTUAL RECIDIVISM. ALTHOUGH THE TYPOLOGY WAS DERIVED AND TESTED USING YOUTHFUL OFFENDERS IN A FEDERAL CORRECTIONAL INSTITUTION, STUDIES INDICATE THE TYPOLOGY CAN BE APPLIED TO OTHER GROUPS IN OTHER SETTINGS. DESCRIPTIONS OF THE 10 TYPES OF OFFENDERS ARE ACCOMPANIED BY RECOMMENDATIONS FOR THE MANAGEMENT AND TREATMENT OF EACH OF THE GROUPS, INCLUDING THE OPTIMAL PLACEMENT, THE MOST APPROPRIATE CHANGE AGENT, AND THE BEST STRATEGY FOR INDUCING POSITIVE CHANGE. THE RECOMMENDATIONS HAVE YET TO BE TESTED EMPIRICALLY. TABULAR AND GRAPHIC DATA ARE PROVIDED, ALONG WITH REFERENCES AND NAME AND SUBJECT INDEXES. (RCB)