Nonmetric subadult skeletal sexing traits: I. A blind test of the accuracy of eight previously proposed methods using prehistoric known-sex mummies from northern Chile

While a number of nonmetric sex-related traits have been proposed, the accurate assignment of sex to subadult skeletal materials is generally considered problematic. Eight previously proposed nonmetric traits of the ilia and mandible have been demonstrated by other researchers to be related to the known-sex of subadults, yet there has been relatively little research attempting to evaluate the utility of these traits using independent collections of known-sex subadult skeletal remains. These eight traits include: elevation of the auricular surface, angle of the greater sciatic notch, depth of the greater sciatic notch, the "arch criteria," curvature of the iliac crest, gonial eversion, mandibular protrusion, and mandibular arcade shape. The present study blindly tested these eight nonmetric traits using 85 autopsied prehistoric known-sex mummified subadult remains from northern Chile that range in age between newborn and 15 skeletal years of age. The two primary questions examined for each trait are: (1) are the different character states for each of the eight traits related to known-sex, and (2) which of the eight traits are accurate enough for use in forensic cases? These questions are examined for each trait by both sex and age class so as to uncover potential sex- and age-related strengths and weaknesses for each of the characteristics. The results indicate that all eight of the previously proposed subadult sex-related traits are indeed related to known-sex, but that there is a great deal of variation by both sex and age in terms of the strength of congruent sex-related associations. With the exception of gonial eversion, all of the traits produced statistically significant chi2 values for their associations with known-sex when all subadult remains were considered. However, when all subadults are considered, only four traits demonstrated acceptable levels of accuracy for forensic applications. These traits include the arch criteria (82.3%), angle of the sciatic notch (80.7%), depth of the sciatic notch (79.0%), and mandibular arcade shape (77.6%). For subadults ranging in age from newborn to five skeletal years of age, only depth of the sciatic notch (81.5%) and the arch criteria (81.5%) approach acceptable levels of accuracy for use in forensic cases. The implications of these results are discussed and recommendations for bioarchaeological and forensic applications are made.