Estimation of sex from cranial measurements in an Australian population

It has been empirically demonstrated that the application of anthropological standards, developed using populations geographically or temporally disparate from that of a decedent, negatively impacts on classification accuracy. Therefore, there is a real need for population specific methodologies to ensure that accurate estimations of biological attributes are provisioned the present study reports the first morphometric cranial sexing standards formulated specifically for application in, and based on the statistical analysis of, a broad representation of the contemporary Australian population. The primary aim of the present paper is to quantify cranial sexual dimorphism and to subsequently develop statistically robust cranial sex estimation standards for the complete cranium and/or its associated fragments. The sample analysed comprised computed tomographic (CT) cranial scans of 771 (385 female and 386 male) individuals collected from five Australian states/territories. Scans are visualized as three-dimensional volume rendered reconstructions using OsiriX®, from which 76 three-dimensional cranial landmarks are then acquired. A total of 36 linear inter-landmark measurements are calculated using MorphDB. All cranial measurements were statistically significantly sexually dimorphic in mean values; the three most dimorphic being bizygomatic breadth, bizygotemporal breadth and basion-nasion height. A total of five univariate and nine multivariate discriminant functions were formulated, achieving a maximum of 90.3% classification accuracy according to sex, with an associated sex bias of −4.4%. The findings of the present study indicate that within the Australian population sex can be accurately estimated using cranial measurements. Furthermore, this study is the first to incorporate a collective and inclusive sample of contemporary living Australians thus better informing the accuracy and reliability of casework and forensic practice across the country.