INTRACLASS CORRELATION AMONG COMMON MEASURES OF ADOLESCENT SMOKING - ESTIMATES, CORRELATES, AND APPLICATIONS IN SMOKING PREVENTION STUDIES

Most adolescent smoking prevention studies employ designs in which cla ssrooms, schools, school districts, or sometimes whole communities are assigned to treatment conditions while observations are made on indiv idual students. The critical design feature in such community trials i s the nesting of intact social groups within treatment conditions. Thi s combination requires that the treatment effect be assessed against t he between-group variance; unfortunately, that variance is usually lar ger than for randomly constituted groups and its precision is usually less than that for the within-group variance. These factors often comb ine to reduce power so that it is almost impossible to detect importan t treatment effects in an otherwise well designed and properly execute d study. To address these problems, investigators need good estimates of the intraclass correlation for the variables of interest, which tog ether with the number of observations per unit determine the magnitude of the extra variation in the nested design. The purpose of this pape r is to describe the methods and results from a study designed to gene rate estimates of intraclass correlation for common outcomes in adoles cent smoking prevention studies and to discuss the use of these estima tes in the planning of new studies.