Our goals in this study are to investigate the following questions: Can examining variants in the cholinergic nicotinic receptor subunit genes jointly using the restricted partition method (RPM) account for more of the variance in nicotine dependence than univariate analyses?Are joint effects detected by the RPM also detected by logistic regression, a traditional statistical approach?Do the joint effects replicate in an independent sample?How much does the explained variance increase when the joint effects are included?
