AN INITIAL EVALUATION OF UDIS REPORTED A PROFOUND DECREASE IN THE DELINQUENT ACTIVITY OF YOUTH PLACED IN EITHER UDIS OR DOC AND NO SUBSTANTIAL DIFFERENCE BETWEEN THE PROGRAMS IN THE POSTRELEASE DELINQUENT BEHAVIOR OF YOUTH. A SUPPRESSION EFFECT OCCURRED AFTER YOUTH WERE TREATED BY BOTH PROGRAMS, WITH ONLY SLIGHT DIFFERENCES BETWEEN UDIS AND DOC. THE REANALYSIS, SENSITIVE TO POTENTIAL ARTIFACTS IN UDIS DATA DERIVED IN THE INITIAL EVALUATION, DEMONSTRATES NO SUPPORT FOR A SUPPRESSION EFFECT ATTRIBUTABLE TO CORRECTIONAL INTERVENTIONS. FOR DELINQUENTS WITH AT LEAST ONE POSTINTERVENTION POLICE CONTACT, THERE IS A STRONG RELATIONSHIPS BETWEEN PREINTERVENTION ACTIVITY AND PERCENT REDUCTION IN DELINQUENCY. FOR DELINQUENTS WITH NO POLICE CONTACTS DURING THE FIRST POSTINTERVENTION YEAR, SUPPRESSION EFFECTS ARE IMPOSSIBLE TO CALCULATE. THE TIME SERVED BY A DELINQUENT IN PROGRAMS OR INSTITUTIONS HAS A POSITIVE OR ZERO IMPACT ON THE DELINQUENT'S PERCENT REDUCTION SCORE. SUPPRESSION IS A MEANINGFUL MEASURE OF RECIDIVISM IN THE CASE WHERE EVERY DELINQUENT IS EXPECTED TO BE REARRESTED. IT IS LESS MEANINGFUL IN THE CASE WHERE ONLY SOME DELINQUENTS ARE EXPECTED TO BE REARRESTED. THE EXPERIMENTAL DESIGN OF THE INITIAL EVALUATION IS FLAWED BY UNCONTROLLED THREATS TO INTERNAL VALIDITY (MORTALITY, MATURATION, AND REGRESSION). PROCEDURAL ASPECTS OF ANALYSIS ARE DESCRIBED, AND SUPPORTING DATA AND FIGURES ARE INCLUDED. NOTES AND REFERENCES ARE PROVIDED. (DEP)