The current study used national survey data to identify which aspects of family relationships are predictive of experimental and daily cigarette smoking. A multiwave longitudinal survey periodically assessed adolescents aged 11 to 17 (N = 1,725) over a seven year follow-up period. Parent interviewers were obtained at the initial screening. Nineteen indicators of family relationships were used based on parent and child interviews. Several indicators of smoking were used including first time cigarette smoking, a nine-point scale of intensity of experimentation with cigarettes, an indicator of daily smoking, and an indicator of smoking at least five or more cigarettes per day. Logistic and multiple regression analyses that controlled for ethnicity, gender, and socioeconomic status found that family relationships were only predictive for subjects under the age of 18. Significant predictors were (a) indices of parental attachment, (b) indices of time spent with one's family, (c) having older siblings who smoke, (d) family structure, (e) family stress, (g) parental negative labeling of the child, and (h) parental leniency towards delinquent acts. We conclude that several aspects of family relationships are important predictors of adolescent cigarette smoking and predict daily smoking more strongly than initial smoking.

INTRODUCTION

Parent-child interventions designed to prevent smoking are based on social-- psychological theories of the development of adolescent cigarette smoking. These theories suggest that a lack of positive family relationships are an antecedent to adolescent cigarette use (Petraitis, Flay, & Miller, 1995). Although theory suggests that family relationships are important, the empirical literature has not consistently identified the critical aspects of family relationships that predict cigarette smoking (Conrad, Flay, & Hill, 1992). Identifying the critical aspects of family relationships that are associated with adolescents not smoking may provide valuable information for designing successful parent-child interventions. Family relationships have been operationalized in a variety of ways, ranging from family structural characteristics to the quality of the parent-child relationship (Conrad et al.). Some of these aspects of family relationships may be risk factors for adolescent cigarette smoking while others are not. The purpose of the current paper is to identify which aspects of family life are predictive of adolescent initial experimentation and daily cigarette use, utilizing a wide-ranging set of indicators of family relationships.

PREVIOUS THEORY ON THE ROLE OF FAMILY RELATIONSHIPS IN ADOLESCENT SUBSTANCE USE

There are many theories of the etiology of substance use, and most of these theories suggest that family bonding is a cause of substance use. For a review, see Petraitis et al. (1995). Most argue that a lack of family bonding produces more time spent with peers and a rejection of conventional norms, which in turn leads to bonding with deviant peers who engage in illicit behaviors such as underage cigarette smoking. The bonding with deviant peers leads to more positive attitudes regarding cigarette use and a further deterioration of norms against using cigarettes and eventual cigarette use (see Petraitis et al.). Although most theories mention family bonding, the definition of family appears to vary widely among theories. Previous theory suggests that several aspects of family relationships may be predictive of cigarette smoking, including family disorganization (Akers, 1977; Elliott, Huizinga, & Ageton, 1985; Jessor & Jessor, 1977), the relative influence of parents versus peers (Elliott et al.; Hawkins & Weis, 1985; Kaplan, 1975), unrealistic parental achievement expectations for their child (Akers; Hawkins & Weis), family closeness (i.e., warmth and cohesion) (Akers; Simons, Conger & Whitbeck, 1988), negative labeling of the child by their parents (Jessor & Jessor; Kaplan), time spent with their family (Elliott et al. …