Delinquent behavior and drug abuse are common and persistent problems among American youth. Deviance is """"""""familial"""""""": when one family member exhibits deviance, the risk to other family members is much greater than the risk to people in general. Most previous research on deviance has identified parental role models and child rearing practices as the probable cause of familial resemblance. Neglected in this analysis is the possibility that the major environmental effect within families is sibling mutual interaction. The proposed project is a sibling study of drug use and deviant behavior (minor to serious delinquent acts). The sample population consists of siblings aged 11-14 years at Round 1, who will be surveyed by anonymous questionnaire on three occasions at one year intervals. The questionnaire contains scales to measure four proposed processes of sibling mutual influence: coercion, identification, modeling, and direct influence. The effect of each process on the behavioral similarity of siblings will be determined; if the influence process affects behavioral similarity, we can infer that it is operative (which is different from accepting the verbal testimony of the siblings). Two complementary forms of statistical analysis will be used: a multiple regression model of cross-sectional data structural equation model of longitudinal data. In sum, the proposed project will 1) demonstrate the existence of sibling interaction effects, 2) evaluate theoretical models of those effects, and (3) place boundaries on the heritabilities of deviance and drug abuse.
