THE AUTHOR APPROACHES THE TOPIC OF JUVENILE DELINQUENCY PREVENTION BY WAY OF A DISCUSSION OF THE OBSTACLES AND DIFFICULTIES ASSOCIATED WITH THE DEVELOPMENT AND EVALUATION OF DELINQUENCY PREVENTION PROGRAMS. THERE HAS BEEN AN INCREASED INTEREST IN THE EVALUATION OF SUCH PROGRAMS, AND THIS INTEREST INFLUENCES THE PROGRAMS THEMSELVES. THE OBSESSION WITH EVALUATION AND THE SOCIETAL NEED TO 'DO SOMETHING,' DESCRIBED AS THE 'GREAT STUMBLE FORWARD,' ARE DISCUSSED ALONG WITH THE POSSIBLE DANGERS OF THIS ACTIVITY. THE AUTHOR ARGUES THAT IT IS EXTREMELY DIFFICULT TO CONDUCT AN OBJECTIVE EVALUATION, THAT ATTEMPTS TO DO SO MAY ACTUALLY DEFEAT OTHER IMPORTANT PURPOSES, AND THAT THE PRESSURE FROM POLITICAL AND OTHER FORCES TO CONDUCT EVALUATIONS CAN MAKE THE SITUATION EVEN WORSE. SPECIFIC EVALUATIVE STUDIES WHICH ILLUSTRATE INHERENT DIFFICULTIES ARE EXAMINED, AND THEIR ASSUMPTIONS, DESIGN, AND CONCLUSIONS ARE ANALYZED. THE AUTHOR FINDS FAULT WITH THE TRADITIONAL EXPERIMENTAL DESIGN IN THE EVALUATION OF JUVENILE DELINQUENCY PROGRAMS AND PRESENTS ALTERNATIVES TO THIS TYPE OF DESIGN. POLITICAL FACTORS WHICH CAN INFLUENCE THE KINDS OF PROGRAMS ATTEMPTED AND THE ROLE OF EXPERTS IN DELINQUENCY PREVENTION ARE EXPLORED. THE WORK NEXT FOCUSES UPON ALTERNATIVE DIRECTIONS TO 'STUMBLE'; IT PAYS PARTICULAR ATTENTION TO INSTITUTIONALIZATION AND ARGUES THAT THIS APPROACH TO DELINQUENCY MAY BE MORE EXPENSIVE THAN CRIME ITSELF. SPECIFIC TYPES OF PROGRAMS ARE ANALYZED, INCLUDING COMMUNITY TREATMENT, GROUP THERAPY, INDIVIDUAL COUNSELING, BEHAVIOR MODIFICATION, CONVENTIONAL APPROACHES (SUCH AS TRAINING PROGRAMS AND PROBATION), AND LESS CONVENTIONAL APPROACHES (SUCH AS CORRECTIVE SURGERY AND FOSTER HOMES). FINALLY, RECOMMENDATIONS ARE MADE TO INCREASE THE LIKELIHOOD OF LAUNCHING SUCCESSFUL PROGRAMS; THE SUGGESTIONS ASSUME THAT FUNDAMENTAL CHANGES WOULD BE NECESSARY TO MAKE A MEANINGFUL IMPACT ON DELINQUENCY. A BIBLIOGRAPHY AND SUBJECT AND NAME INDEXES ARE PROVIDED. (VDA)