THE TESTED HYPOTHESIS STATES THAT THERE WILL BE NO CHARACTERISTIC DIFFERENCES BETWEEN THOSE MENTALLY RETARDED JUVENILE OFFENDERS FOUND IN RETARDATION FACILITIES AND THOSE FOUND IN CORRECTIONAL FACILITIES. FOUR SUBHYPOTHESES ALSO TESTED DIFFERENCES IN SOCIOLOGICAL CHARACTERISTICS. ALL 61 SUBJECTS WERE RESIDENTS OF TEXAS INSTITUTIONS WHICH HOUSED JUVENILES WHO MET THE FOLLOWING CRITERIA: 10 TO 20 YEARS OF AGE, INTELLIGENCE QUOTIENT BELOW 75, AND COMMISSION OF A CRIMINAL ACT. STANDARDIZED SCALES WERE USED IN THE DESIGN OF THE QUESTIONNAIRE WHICH WAS COMPLETED BY OFFICIAL STAFF PERSONS. THE FIRST PART OF THE SURVEY INSTRUMENT CONTAINED DATA COLLECTED FROM OFFICIAL RECORDS; THE SECOND PART IDENTIFIED PHYSICAL, INTELLECTUAL, AND BEHAVIORAL CHARACTERISTICS. THE CHI-SQUARE WAS USED TO TEST FOR CHARACTERISTICS AND OBSERVERS' RESPONSE. THE T-TEST WAS USED TO COMPARE DIFFERENCES IN SOCIOLOGICAL FACTORS. THERE WERE NO SIGNIFICANT DIFFERENCES IN THE FOLLOWING CHARACTERISTICS: ORIENTATION TO ADULTS, VAGUENESS OF VOCATIONAL GOALS, ABSENCE OF HELPLESS FEELINGS, AND GOOD RELATIONSHIP WITH THE MOTHER. THERE WERE SIGNIFICANT STATISTICAL DIFFERENCES CONCERNING BEHAVIOR CONTROL, INTELLIGENCE QUOTIENT, AND RELATING TO ADULTS. RECOMMENDATIONS WERE MADE IN REFERENCE TO FUTURE RESEARCH, SCHOOLS, LEGISLATION, THE JUSTICE SYSTEM, AND STATE AGENCIES FOR THE RETARDED. TABULAR DATA AND REFERENCES ARE INCLUDED. THE QUESTIONNAIRE IS APPENDED. (JCP)