Analyses of descriptive statistics were conducted in the computer program Mplus (20). Age and sex differences in diversity of substances used were examined to confirm that the measure being used behaved in a manner consistent with findings from previous research (e.g. 1–3). Main effects of age and sex on substance use diversity were examined while correcting for the non-independent nature of the twin pair observations using the robust maximum likelihood estimator in Mplus. All tests for age and sex differences were two-tailed. Twin data were analyzed to estimate genetic and environmental influences via the computer program Mx (21) using a multivariate Cholesky model. Phenotypic variance was decomposed into three components: variance due to additive genetic factors (a2); variance due to shared environmental, or family-level, factors (c2); and variance due to non-shared environmental, or individual-specific, factors (e2). Calculation of variance accounted for by each of these factors is performed by comparing within-pair monozygotic twin correlations to within-pair dizygotic twin correlations. Additive genetic effects are indicated when rMZ>rDZ, shared environmental effects when rDZ>0.5rMZ, and non-shared environmental effects when rMZ<1.0. The primary model
