Both official records of police and court contact and self-reports of involvement in 69 chargeable offenses were used as well as other demographic and social information. Males, blacks, persons of lower socioeconomic status, and official delinquents were oversampled to facilitate analysis on those variables of etiological and methodological interest that are often underrepresented in general random or probability samples. Data were collected within a quasi-experimental design to assess the effects of different measurement methods on estimates of the prevalence, incidence, and correlates of juvenile crime. Except for a small number of social-class group differences on a few self-report items, the findings support the general conclusion of a systematically weak relationship between social class, measured by a number of different indicators, and a variety of self-reported delinquency incidence scales, a number of official crime indexes, and a large number of self-reported crime items, all at the individual level of measurement. 6 tables.