ALSO INVESTIGATED WERE THE RATE OF RECIDIVISM FOR OFFENDERS WHOSE SELF-REPORT OFFENSE HISTORIES DIFFER BY TYPE (PERSONAL VS. PROPERTY) AND FREQUENCY OF OFFENSE; THE ADEQUACY OF AN OFFICIAL RECORD OF VIOLENT BEHAVIOR FOR EXAMINING THE RELATIONSHIP BETWEEN TYPE OF OFFENSE (VIOLENCE VS. NO VIOLENCE) AND RECIDIVISM; AND THE MANNER IN WHICH SOCIOECONOMIC STATUS MAY AFFECT SELF-REPORTED AND OFFICIAL INDICATORS. THE STUDY SAMPLE CONSISTED OF 1,681 ADJUDICATED DELINQUENTS (AGES 16 TO 20 YEARS OLD) HOUSED AT THE PRESTON SCHOOL OF INDUSTRY NEAR SACRAMENTO, CALIFORNIA. NINETY-EIGHT PERCENT HAD A PRIOR OFFENSE RECORD AND 45 PERCENT HAD PREVIOUSLY BEEN COMMITTED TO A JUVENILE INSTITUTION. WHILE MANY PREVIOUS STUDIES OF OFFICIAL RECORDS SHOW THAT PERSONAL OFFENDERS HAVE A LOWER RATE OF RECIDIVISM THAN DO PROPERTY OFFENDERS, THIS STUDY SHOWED THAT, WHEN TYPE OF OFFENSE IS MEASURED BY SELF-REPORTS, PERSONAL OFFENDERS HAVE A HIGHER RATE OF RECIDIVISM THAN DO PROPERTY OFFENDERS HAVING THE SAME FREQUENCY OF OFFENSE. ALSO, THE MOST FREQUENT PROPERTY OFFENDERS HAD A RECIDIVISM RATE SIMILAR TO A LESS FREQUENTLY INVOLVED CATEGORY OF PERSONAL OFFENDERS. (AUTHOR ABSTRACT MODIFIED)...ELW