IN 1974, ROBERT MARTINSON COMPLETED A REVIEW OF ALL RESEARCH STUDIES IN CORRECTIONS PUBLISHED FROM 1945 TO 1967 AND CONCLUDED THAT EVERYTHING THUS FAR TRIED IN CORRECTIONS HAS EITHER NOT WORKED OR HAS FAILED TO DEMONSTRATE CLEAR EVIDENCE OF SUCCESS. THIS STUDY AND OTHERS WITH SIMILAR CONCLUSIONS HAVE BEEN USED BY MANY TO ARGUE FOR THE INCREASED USE OF INCARCERATION AND A DECREASE IN COMMUNITY TREATMENT. MARTINSON'S FINDINGS ARE QUESTIONABLE FOR THE FOLLOWING REASONS: (1) METHODOLOGICAL COMPLICATIONS IN THE STUDIES REVIEWED MAKE IT DIFFICULT TO INTERPRET THE FINDINGS; (2) MARTINSON USED PRIMARILY RECIDIVIST RATES IN DETERMINING THE EFFECTIVENESS OF THE PROGRAMS REVIEWED, AND THIS PRESENTS PROBLEMS BECAUSE ONLY CONVICTED OFFENDERS ARE INCLUDED, THE RATE OF APPREHENSION AND CONVICTION FOR EXPERIENCED CRIMINALS IS NOT KNOWN, AND MANY OTHER FACTORS ASIDE FROM RECIDIVIST RATES ARE IMPORTANT IN MEASURING POSITIVE CHANGES IN OFFENDERS; AND (3) THERE ARE NO ESTABLISHED CRITERIA FOR DETERMINING RECIDIVIST RATES. OTHER SIMILAR STUDIES EXAMINING THE EFFECTIVENESS OF COMMUNITY TREATMENT AND ALSO IMPRISONMENT (ROBISON, GARABEDIAN, AND GREENBERG) HAVE FLAWS THAT UNDERMINE THEIR CONCLUSIONS. THERE IS THUS A LACK OF GOOD RESEARCH ON ALTERNATIVES TO PRISON AND THE EFFECTS OF IMPRISONMENT, AND MUCH OF THE COMPLETED RESEARCH EVIDENCES METHODOLOGICAL PROBLEMS. THE MEASUREMENT OF POSITIVE CHANGE IN AN OFFENDER DEPENDS ON MANY COMPLEX FACTORS, REQUIRING MORE RESEARCH THAN HAS THUS FAR BEEN UNDERTAKEN. THERE IS EVIDENCE FOR THE EFFICACY OF THE ALTERNATIVES TO IMPRISONMENT, BUT MANY QUESTIONS STILL NEED TO BE ANSWERED. IT WOULD BE A MISTAKE TO HALT THE DEVELOPMENT OF COMMUNITY TREATMENT PROGRAMS ON THE BASIS OF THE INADEQUATE EVIDENCE THUS FAR PRODUCED. ADDITIONAL AND MORE SOPHISTICATED RESEARCH IS NEEDED IN ORDER TO IDENTIFY WHICH PROGRAM DYNAMICS WORK WITH WHICH TYPES OF OFFENDERS TO PRODUCE PARTICULAR OUTCOMES. NOTES ARE PROVIDED. (RCB)