Longitudinal tracking and retention in a school-based study on adolescent smoking: costs, variables, and smoking status.

Methods used to track a cohort of Grade 6 students through Grades 8 and 11, and costs involved for survey completion in school and by mail for ever and never smokers from the original group are detailed. At baseline, 1,598 students in Scarborough, Canada, completed a questionnaire on smoking, drinking, and health, and again in Grade 8 (N = 1,543/1,598) and Grade 11 (N = 1,454/1,598). In Grades 8 and 11, tracking and administering the questionnaire was more costly per participant when the survey was administered by mail than in school. Average completion costs were highest for Grade 11 students who used tobacco at baseline ($52.44). Students categorized as ever smokers in Grade 6 were harder to locate at each phase of testing, which suggests that this group should be identified at baseline so that closer tracking procedures may be employed between data collection points.