Resemblances among first degree relatives such as dizygotic twins or sibling pairs can be summarized by correlations (r). For phenotypes assessed on a continuous scale r can be a product-moment or intra-class correlation. For ordinal and dichotomous traits, r can be a polychoric or tetrachoric correlation, which summarizes the familial resemblance on the liability scale [3]. The expectations for sibling or DZ twins correlations assuming an autosomal inheritance pattern can be expressed as: where h2 represents the narrow-sense heritability i.e. the additive genetic variance divided by the total phenotypic variance of the phenotype, and c2 gives the standardized common environmental variance shared by family members. In same-sex sibling pairs, under the assumption of random mating, h2 is weighted by 0.5 (e.g. Jacquard, 1974). In opposite-sex pairs, the genetic correlation between relatives is symbolized by γ. Common environmental factors, defined as all environmental factors that increase resemblance of relatives for non-genetic reasons, are correlated unity in same-sex relatives and φ is the correlation among environmental factors in male-female pairs. When γ<0.5 there is evidence for qualitative sex differences, i.e. for the
