Multivariate Comparison of Cephalometrictraits in Iranian Azeris and Persians

Abstract

Turkic-speaking Azaris and Iranian-speaking Persians living in Iran were compared by cephalomentric traits. Two multivariate statistical approaches, principal component and discriminant analyses, were applied. The populations are described by almost the same average values (p > 0,05) of the first two principal component variables accounting in aggregate for 73.6% of total diversity of the initial traits. The average level of correct reclassification for the objects while using the analytical recognition method is about 52%, i.e. the probability of their attribution to any of the two groups is almost equal. In whole, the two samples do not differ significantly from each other according to multivariate characteristics analysed that witnesses to the same generalized cephalomentric peculiarities in the ethnic groups considered.

Key words: Iranian people, Iranian Azaris, Persians, craniofacial traits, multivariate analysis