A TOTAL OF 48 INMATES HOUSED AT THE MAXIMUM SECURITY INSTITUTION AT MARION, ILLINOIS, VOLUNTEERED TO COMPLETE THE ISB, 22 MEMBERS OF THE GROUP HAD RECEIVED ASSIGNMENT OF AT LEAST 9 MONTHS' DURATION TO A SPECIALIZED HOUSING UNIT DUE TO DANGEROUSNESS OR ACTS OF VIOLENCE. AN ASSUMPTION OF THIS STUDY WAS THAT AGGRESSION, WHATEVER THE MOTIVATION, IS ALWAYS A PRECEDING FACTOR IN A DANGEROUS ACT. AGGRESSION BASED ON HOSTILITY IS AN INTEGRAL PART OF THE PERSONALITY MAKEUP OF THE DANGEROUS PERSON. ACTS OF MURDER, RAPE, ASSAULT, ARSON, BATTERY, ROBBERY, EXTORTION, ESCAPE, AND ATTEMPTED ESCAPE ILLUSTRATE WHAT IS MEANT BY VIOLENCE OR DANGEROUS BEHAVIOR IN THE STUDY. THE REMAINING SUBJECTS (26 INMATES), WHILE CONSIDERED DANGEROUS, HAD NOT RECEIVED ASSIGNMENT OT THE SPECIALIZED UNIT, NOR HAD THEY EVER ACTED, WHILE CONFINED, IN SUCH A MANNER THAT WOULD BE CONSIDERED CAUSE FOR ASSIGNMENT TO THE UNIT. THE HYPOTHESIS TESTED WAS THAT A SIGNIFICANT DIFFERENCE COULD BE EXPECTED IN ADJUSTMENT TO INCARCERATION, AS MEASURED BY THE ISB, BETWEEN INMATES WHO HAD COMPLETED LONG ASSIGNMENTS IN HIGHLY CONTROLLED UNIT SITUATIONS AND INMATES WHO HAD NOT BEEN ASSIGNED TO THE UNIT. THE ISB WAS SELECTED BECAUSE IT WAS THE ONLY PROJECTIVE INSTRUMENT OF ITS KIND WITH WHICH THE AUTHOR HAD HAD WORKING EXPERIENCE, WITH AND BECAUSE IT HAD A DEVELOPED SCORING STRUCTURE AND MANUAL WHICH WERE EFFICIENT WITHIN THE CONTEXT OF THE STUDY. TREATMENT AND CONTROL GROUPS WERE COMPARABLE WITH RESPECT TO ALL MEASURED DEMOGRAHIC VARIABLES. A BIVARIATE ANALYSIS SHOWED THE EFFECT OF THE INMATE GROUP VARIABLE (TREATMENT V CONTROL GROUP) TO BE STATISTICALLY SIGNIFICANT. THE ISB SCORE WAS THE BEST PREDICTOR OF INMATE GROUP STATUS. A TOTAL OF 12 REFERENCES AND A TABLE ARE INCLUDED. (LWM)