| Introduction. Student characteristics and characteristics of school environments can influence student smoking behaviour but a limited number of studies have examined those characteristics simultaneously.;Objective. The purpose of the present cross-sectional exploratory study was to examine how individual student characteristics and school community characteristics were related to student smoking behaviour.;Method. Using linked data from the School Smoking Profile (22,091 students at 29 secondary schools in Ontario) and the School Environment Pilot Project (environmental scan of the same 29 secondary schools), multi-level logistic regression analyses were used to examine how individual student and school community characteristics were related to being: (1) a susceptible never smoker compared to a non-susceptible never smoker, (2) a tried once smoker compared to a never smoker, (3) an experimental smoker compared to a tried once smoker, (4) a regular smoker compared to an experimental smoker, and (5) an ex-smoker compared to a regular smoker.;Results. Significant between-school variability existed for the odds of being a: susceptible never smoker [sigma2 mu0 = 0.05(0.02), p < .05], tried once smoker [sigma2 mu0 = 0.06(0.01), p < .001], experimental smoker [sigma 2mu0 = 0.12(0.04), p < .001], and regular smoker [sigma 2mu0 = 0.08(0.03), p < .01]. Three school characteristics were directly related to student smoking behaviour: a student at a school with student smoking on the school periphery was less likely to be a susceptible never smoker ( b&d4; = -0.22, p < .05), a student at a school where the prevalence of regular smokers was greater than the average across all schools was more likely to be a tried once smoker b&d4; = 0.17, p < .05), and the odds of being an experimental smoker increased as the senior student smoking rate at a school increased ( b&d4; = 0.05, p < .001). Individual student characteristics were related to all of the smoking behaviours examined.;Conclusion. This study identified that school community characteristics are related to student smoking behaviour, even when controlling for the influence of individual student characteristics. |
