Abstract
This study utilizes a large sample of 8th, 10th, and 12th-grade youth in Arizona (n = 34,373). Five cycles of this biannual survey were used to examine the longitudinal trends. Multivariate logistic regressions were used to identify the key risk factors for past 30-day and lifetime cigarette use. Across all survey cycles, peers played a consistently significant role in smoking behaviors. Those youth with drug-free friends and friends who disapproved of substance use were less likely to engage in smoking behaviors. Further, various community, school, and familial factors were predictive of smoking behavior, though less consistently than peer-level determinants. This study demonstrates that peers consistently play a critical role in cigarette smoking behavior. As such, interventions and future research need to examine further how prosocial peers can be leveraged to reduce smoking behaviors and the associated health outcomes.