Statistical classification methods for estimating sex based on five skull traits: A nonmetric assessment using 3D CT...

Area of Science:

• Forensic Anthropology
• Biological Anthropology
• Human Osteology

Background:

• Cranial nonmetric traits are influenced by population of origin.
• The effectiveness of specific cranial traits for sex estimation varies across different populations.
• Developing population-specific methods for sex estimation is essential.

Purpose of the Study:

• To identify the most effective cranial nonmetric traits for sex estimation in a modern Korean population.
• To demonstrate the necessity of tailored sex estimation methods for diverse populations.
• To compare the accuracy of discriminant analysis and decision tree analysis for sex estimation using cranial traits.

Main Methods:

• Evaluation of 135 three-dimensional skull images from 21st-century Korean autopsy cadavers.
• Application of an ordinal scoring system for five cranial nonmetric traits (Buikstra & Ubelaker, 1994).
• Analysis of trait scores using linear discriminant and decision tree analyses.

Main Results:

• High accuracy rates for sex estimation: 88.1% (discriminant analysis) and 90.4% (decision tree).
• Glabella and mastoid process exhibited the highest accuracy in both analytical methods.
• Significant differences in trait score patterns were observed when compared to other global populations, including those in Asia.

Conclusions:

• Cranial nonmetric methods, particularly using discriminant and decision tree analyses, are highly accurate for sex estimation in modern Korean populations.
• The observed variations in trait frequencies underscore the need for population-specific sex estimation techniques.
• This study highlights the importance of adapting forensic anthropological methods to regional demographic characteristics.