Short communicationDifferential contribution of parents and friends to smoking trajectories during adolescence
Introduction
Both parental and friends' smoking have often been indicated as important contributors to smoking initiation. However, their relative role has been debated during the past two decades. On the one hand, many studies have shown that friends' smoking is a better predictor of adolescent's smoking than is parental smoking Chassin et al., 1996, Hu et al., 1995. On the other hand, there are also studies showing that both friends' and parents' smoking affect adolescent initiation of tobacco use (Greenlund, Johnson, Webber, & Berenson, 1997). Finally, in at least one case, parental, but not friends', smoking predicted smoking initiation (Engels, Knibble, de Vries, Drop, & van Breukelen, 1999).
Studies reporting a stronger contribution of friends used samples aged 13 years or older, whereas studies reporting a strong parental influence used preadolescent samples. If, as suggested by Hu et al. (1995), the influence of friends' smoking increases from preadolescence to adolescence, whereas the influence of parent smoking remains stable or even diminishes, then, it should be possible to resolve the conflicting results about the influence of friends' smoking versus parents' smoking by distinguishing the age of onset of adolescents' smoking trajectories. Specifically, parents' smoking might contribute to smoking initiation in those adolescents who start smoking early (i.e., by age 12 years or earlier). In contrast, friends' smoking may become the main predictor for smoking initiation during adolescence (i.e., by age 13 and later). The goal of the present study was to help clarify these issues by showing that the contribution of parents' and friends' smoking to smoking initiation during adolescence varied according to the participants' smoking trajectories based on the age of onset and level of use.
Section snippets
Design and participants
An accelerated longitudinal design that included 802 participants (405 boys and 407 girls) aged 9 through 11.5 years at T1 was used. Five data waves (i.e., from T1 to T5) were collected over a 4-year period (T1 and T2 corresponded to Fall and Spring of the same school year, whereas T3, T4, and T5 corresponded to Spring of the following 3 years). The participants' socioeconomic status, as assessed by Statistics Canada, was in the middle to upper range, and the vast majority (>90%) were
Results
In the first set of analyses, we examined whether groups with distinct longitudinal trajectory profiles of smoking could be empirically identified. For this purpose, we used a recently developed semiparametric clustering technique for longitudinal data (TRAJ, Nagin, 1999). The results from the TRAJ procedure revealed that a four-group model showed the best fit to the data (BIC=−1655.69) compared with the competing models specifying one, two, three, and five groups (BIC=−2492.37, −1871.34,
Discussion
Four smoking trajectory groups were found for children 10 through 15 years old,: From age 11–12 through age 13–14 years, there was a new trajectory group every year with new smokers. No group of new smokers was found after age 13–14 years. Overall, by age 15 years, one out of four adolescents in the present sample reported smoking between 52 and 96 cigarettes over the course of a week, depending on their trajectory group.
Parents' and friends' smoking seem to play different roles for different
References (5)
- et al.
The natural history of cigarette smoking from adolescence to adulthood: Demographic predictors of continuity and change
Health Psychology
(1996) - et al.
Influences of parental and best friends' smoking and drinking on adolescent use: A longitudinal study
Journal of Applied Social Psychology
(1999)