The first dilemma is the nature of adolescent smoking. Smoking during adolescence is best characterized by change. Many of the family characteristics used to predict adolescent tobacco use are relatively stable characteristics. The dilemma is how one conceptualizes the processes that relate stability to change. The second dilemma is the nature of the family itself. Families come in many diverse forms. Family member characteristics are inherently interdependent, posing problems for traditional statistical techniques. The third dilemma is the nature of the family during adolescence. Adolescents spend more time in the company of peers and less time with family members. Stable family characteristics can have a direct effect on child characteristics that change the likelihood that children will be exposed to triggering events. Stable family characteristics can have a direct effect on child characteristics that increase the likelihood that triggering events will cause a change in behavior. Stable family characteristics can have a direct effect on the likelihood that children will be exposed to triggering events. Although the study of smoking during adolescence is fundamentally the study of change, smoking is often predicted from relatively stable family characteristics, such as demographics, parent smoking status, or relationship quality. The complexity of the family system makes it imperative that research be grounded theoretically. Approaches that recognize the nested nature of family data or take a community systems approach to modeling family influences offer advantages over more commonly employed statistical techniques. Because adolescents spend increasing time outside the family, the connections between the family and the larger social world may be a critical area of research. 96 references