and those with missing data were eliminated;
27% (n = 98) reported smoking
in the past 30 days. Thus, the ratio of
smokers to nonsmokers was within
the suggested limits (Kenny, 2007). In
determining the minimum sample
size, the sample size was calculated
using Cohen and Cohn’s classical procedure
for a priori power analysis
(Cohen, 1988). Additionally, both
Baron and Kenny’s (1986) and Kenny’s
(2007) recommendations for power
analysis with mediational models
were followed. The minimum required sample size was exceeded.