Back to Conference page
The 13th World Conference on Tobacco OR Health
Building capacity for a tobacco-free world
July 12-15, 2006, Washington, DC, USA
Objective: To show that the importance of peer influence as predictor of smoking uptake may be overestimated, and to show the importance of selection and parenting processes.
Methods: The data of 16.000 students of four waves of the 6 country ESFA trials were used to compare cross-sectional analyses with longitudinal analysis (Study A) and to analyze the importance of peer influences, selection processes and parenting behaviors using structural equation techniques (Study B).
Results: Study A shows that cross-sectional analysis confirm that adolescent smoking was most strongly associated with friends' smoking. Longitudinal regression analysis, however, showed that friend influences were comparable to parental influences. Study B shows (using AMOS and LISREL) that not much support was found for peer smoking as an important predictor of smoking onset, which was only significant in Portugal. Much support was found for the selection paradigm in all countries, implying that adolescents choose friends with similar smoking behavior. Furthermore, influences of parents on adolescent behavior and the choice of friends were also found in some countries. Thus, the importance of peer smoking as the most important predictor of smoking uptake is overestimated. For many smokers a social inoculation approach will not be relevant, but they may profit from programs clearly outlining the pros of non-smoking. However, more in-depth analyses are needed to identify the reasons for selecting smoking friends in order to be able to fine-tune programs for this group.