Skip to main content
. 2014 Mar 11;16(8):1063–1069. doi: 10.1093/ntr/ntu023

Figure 2.

Figure 2.

Modeled associations between rated craving intensity and changes in the probability of smoking (vs. randomly selected nonsmoking occasions). Data are presented as changes relative to the group average probability of smoking, because the absolute probability is influenced by the sampling scheme for smoking and nonsmoking occasions, which differs between groups (i.e., the prompting algorithm aimed to gather the same number of randomly sampled nonsmoking occasions for each group, even though the number of smoking occasions was very different for the two groups). The value of 0, marked by a line, represents each group’s average probability of an observation representing smoking (vs. a nonsmoking observation) within the dataset and is not comparable between nondaily intermittent smokers, ITS, (0.31) and daily smokers, DS, 0.71.