Key points are not available for this paper at this time.
Recently there has been great interest in possible mediators and moderators of family history risk for alcoholism. However, previous studies have failed to employ appropriate designs and data analytic strategies to identify moderators and mediators. This article uses a large data set to illustrate such analyses. In the current data, both presumed personality risk and dispositional self-awareness were found to play moderator (rather than mediator) roles. The conceptual, methodological and data analytic implications of the mediator-moderator distinction are discussed.
Rogosch et al. (Sun,) studied this question.