Using the first four waves of the National Youth Survey, the study focused on the predictive ability of variables central to social control, labeling, differential association, and urbanism perspectives. Data were collected over a 4-year period beginning in 1976. At the start of the study, the data came from a national probability sample of youths from ages 11 to 17 in the United States. The dependent variable of interest was whether juveniles self-reported involvement in any of the specified delinquent acts during the fourth wave of data collection. Independent variables used in the analysis were measured at the third wave of data collection. The variables included conventional aspirations, family involvement, moral beliefs, negative labeling, and exposure to delinquent peers. Demographic variables of race, gender, and age were also included. The study found that although many variables predict initial and continued offending in a similar fashion, other variables predict only one of these decisions. Moral beliefs and residence in a rural place affect only decisions to begin offending, and negative labeling contributes exclusively to the continuity of delinquency. Hence, the factors related to committing one's first offense are not identical to those that explain continued offending. 39 references