Analyzed data from two data sets in the National Longitudinal Study of Adolescent Health Wave I to identify discrepancies between in-school questionnaire and in-home interview results. Identified a group of "jokester" adolescents likely to have identified themselves as adoptees, when they were not. Found evidence suggesting that this group provided extreme responses to questions about delinquency. Concluded that previously reported differences between adopted and non-adopted adolescents were overstated. (Author/KB)