Data from monozygotic (MZ) and dizygotic (DZ) twins can be utilized to parse out the relative contributions of additive genetic (A), shared environmental (C) and non-shared environmental (E) influences on population variation in behaviors of interest (in this case, onset of alcohol, cigarette, and cannabis use). Non-additive genetic influences (including dominance, and thus denoted as D) can be estimated in place of shared environment when the correlation between members of DZ twin pairs is less than half the correlation between their MZ counterparts for a given behavior, but C and D cannot be jointly estimated when data from twins alone are used. In our sample, the DZ correlations for age at initiation were greater than half the MZ correlations for all three substances, indicating that a model incorporating C rather D (i.e., an ACE model) would be most appropriate for these data.
