CONSTRUCT VALIDITY REFERS TO CLEARLY AND COMPLETELY DEFINED EXPERIMENTAL VARIABLES. STRENGTH OF TREATMENT IS AN ISSUE IN CONSTRUCT VALIDITY BECAUSE IT IS EASY TO MAKE FALSE ASSUMPTIONS WHEN EVALUATING TREATMENT RESULTS. STRENGTH OF TREATMENT CAN BE PLANNED FOR AND ASSESSED AT ANY TIME DURING TREATMENT. IT IS POSSIBLE ALSO TO ASSESS STRENGTH OF TREATMENT IN TERMS OF OUTCOME. INTEGRITY OF TREATMENT, THE FIDELITY WITH WHICH A TREATMENT PLAN IS IMPLEMENTED, CAN BE FULLY ASSESSED ONLY AFTER TREATMENT HAS BEEN COMPLETED. CAREFUL ANALYSIS IN THE PLANNING STAGE OF AN INTERVENTION MAY SUGGEST THE LIKELIHOOD OF THRESHOLD EFFECTS DUE TO TREATMENT. IN THIS REPORT ON STRENGTH AND INTEGRITY OF TREATMENT IN EVALUATION STUDIES, ATTENTION IS PAID TO RECIDIVISM, THEORETICAL PREMISES ON WHICH TREATMENT IS BASED, THE VIABILITY OF STRONG TREATMENT PROGRAMS FOR CORRECTIONS AND OFFENDER REHABILITATION, SERVICE DELIVERY AND MONITORING, PRIOR ASSESSMENTS, POSTERIORI ASSESSMENTS, THE NEED FOR BETTER DOCUMENTATION OF TREATMENT PROGRAMS, REASONS WHY A TREATMENT PROGRAM MAY BE WEAK, THE DOCUMENTATION OF NORMS AND COSTS RELATED TO TREATMENT, THE ESTABLISHMENT OF STANDARD PARAMETERS FOR COMPARISON IN TREATMENT EVALUATION, AND COMPONENTS OF A TREATMENT PLAN. STUDIES DEALING WITH WORK RELEASE, JOB TRAINING, COUNSELING, PROBATION, PAROLE, AND OTHER AREAS OF THE CRIMINAL JUSTICE SYSTEM, ARE REVIEWED. THE GENERALITY OF TREATMENT STRENGTH AND INTEGRITY ISSUES IS ADDRESSED. NOTES AND REFERENCES ARE INCLUDED. (DEP)