Finally, we estimated the extent of genetic (rA), broad familial environmental (rC+T) and individual-specific environmental (rE) correlations for both racial/ethnic groups (Table 4). We were unable to resolve the extent to which rA or rC+T contributed to the covariance between cigarette and cannabis use in AA twins. However both genetic and family environmental sources of covariance could not be simultaneously constrained to zero, indicating overlapping sources of familial influence with insufficient power to determine the source of the familial overlap. For the EA twins, both rA and rC+T were significant and substantial; confidence limits indicated the possibility of complete overlap across substances in both sources of variance. Despite racial/ethnic differences in the magnitude of E for each substance, rE was moderate for both AAs and EAs, and could be equated in magnitude across the racial/ethnic groups.
