Nonresident fathers' involvement and adolescents' smoking.

Although prior research has shown that adolescents from divorced and separated households are more likely to smoke than their peers from intact families, few studies have addressed factors that may minimize this risk, such as the role of involvement by nonresident fathers. A sample from the National Longitudinal Study of Adolescent Health (Add Health) is used to examine the longitudinal effects of nonresident fathers 'involvement, changes in involvement, and fathers' modeling of smoking behavior on the probability that adolescents will begin smoking regularly. Results indicate that adolescents who are more involved with their fathers are less likely to begin smoking regularly, that changes in involvement over time predict changes in the probability that adolescents will begin to smoke regularly, and that fathers' smoking also affects this outcome. Implications for theory and public policy are discussed.