In fact, more intensive monitoring and more frequent minimal intervention may be more effective than administering severe sanctions. The 4,079 subjects for this longitudinal study were the members of 1942, 1949, and 1955 birth cohorts who had continuous residence in Racine, Wisc. The multiple regression analysis focused on their nontraffic contacts with police when the subjects were between ages 13 and 22. Most subjects whose police contacts did not result in referral to courts or other agencies had no future police contacts. Those whose referred contacts were less serious than a felony tended to drop out of delinquency. Furthermore, those with referred felonies but not institutionalized were less likely to have another referred felony in the next 2 years than were those who had been institutionalized at an early period in their lives. The frequency of intervention rather than the severity of sanctions seemed to have the most desirable effect. Most demographic, ecological, and prior offense variables generally did not account for the variance in the seriousness of the present or future offenses. However, younger age at the time of any given police contact was associated with greater probability of future police contacts. The difficulty of identifying future career offenders suggests the need for broadly based and positive programs for youth development. Footnotes and data tables. For the summary of this report, see NCJ 103973.