A comparison of different delinquency measures derived from self-report data

The purpose of this research was to explore different ways of using self-report data to derive measures of juvenile delinquent behavior. The subjects were 161 public school children in the Omaha area, and 1,030 public school children and 665 adjudicated delinquents in the metropolitan areas of Baltimore, Indianapolis, and Phoenix. The Omaha youths responded to twenty-eight self-report items two times with an intervening period of three weeks and then rated the items for seriousness. Four measures of delinquency were derived from the data: frequency of activity, diversity of activity, seriousness of activity, and progression into delinquent behavior. The four measures were shown to be highly reliable and strongly intercorrelated; none was shown to be consistently better than the other three. Rated seriousness and reported frequency of behavior were highly negatively correlated, giving support to the use of frequency measures in building delinquency scales. All four scales showed differences within the Omaha sample as a function of sex, age, and birth order; but the differences were not always in the same directions when the four measures were used.