FAILURE RATE ANALYSIS IS A STATISTICAL TECHNIQUE COMMONLY USED IN RELIABILITY ENGINEERING AND MORTALITY MODELING (BIOSTATISTICS) IN STUDIES OF FAILURE OR NONFAILURE WITHIN A GIVEN PERIOD OF EXPOSURE TO A HAZARDOUS ENVIRONMENT. THE TECHNIQUE IS RELATIVELY INSENSITIVE TO VARYING GRADATIONS OF OUTCOME DEFINITIONS BUT IS ADAPTABLE TO THE MOST UNIVERSAL SET OF PROBLEMS. THE FAILURE RATE IS CALCULATED AS THE NUMBER OF FAILURES OBSERVED DURING A PERIOD OF TIME DIVIDED BY THE TOTAL TIME IN WHICH FAILURE EVENTS COULD HAVE OCCURRED. IN LAW ENFORCEMENT AND CRIMINAL JUSTICE, FAILURE EVENTS CAN BE ARRESTS, CONVICTIONS, ESCAPES, INCARCERATIONS, ETC., DEPENDING ON THE CONTEXT OF THE PARTICULAR ANALYSIS. BY ADDRESSING MAJOR SHORTCOMINGS ASSOCIATED WITH TRADITIONAL EVALUATION OF RECIDIVISM DATA (E.G., DISREGARD OF THE TIMES AT WHICH FAILURES OCCUR), FAILURE RATE ANALYSIS PROVIDES CRIMINAL JUSTICE AGENCIES WITH A VALUABLE STATISTICAL TOOL FOR MEASURING AND COMPARING THE PERFORMANCE OF REHABILITATION TECHNIQUES AND PROGRAMS. METHODOLOGICAL DETAILS OF FAILURE RATE ANALYSIS ARE PRESENTED, AND ITS APPLICATION IS ILLUSTRATED IN STUDIES OF HALFWAY HOUSE PERFORMANCE IN THE DISTRICT OF COLUMBIA AND PAROLE OUTCOMES IN CALIFORNIA. GRAPHS AND SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED. (LKM)