In order to describe the progression of nicotine craving and withdrawal over the course of 24 hours of abstinence, a series of polynomial mixed effects random coefficient models were analyzed with observations (level 1) nested within subjects (level 2). These models included a polynomial function of hours since last cigarette (Hours, a level 1 predictor), smoking group (Group: a level 2 predictor coded moderate smoker = 0, heavy smoker = 1) as well as their interactions. Prior to model estimation, intraclass correlations (ICC) were calculated for each dependent variable. ICCs for craving, withdrawal/NA and PA were 0.44, 0.62, and 0.62 respectively, suggesting substantial between-subjects variation in addition to a large proportion of within subject variability.
