Multivariate analysis of craniofacial measurements in twin and family data

Thirty‐three cephalometric variables and height have been measured on each of 630 individuals (316 male and 314 female) from 157 families. After age and sex differences were adjusted for each measurement, a rotated factor analysis was undertaken to account for the variation by a limited number of linear combinations of the adjusted measurements. It was found that most of the variation could be explained by nine independent factors. Finally, correlation coefficients were computed on twins, siblings and parent‐offspring data for factor scores. The results suggest that each factor which is measured by a linear combination of a set of variables could result from the interaction of independent sets of genes with the environment.