A Geometric Morphometric Approach to Sex Determination of the Human Adult Os Coxa

The os coxae, the two bones that comprise the left and right halves of the pelvic girdle, are the most reliable bones of the adult human skeleton to determine sex and when used alone, can achieve approximately 90% accuracy. Sex determination using the os coxa can be accomplished through visual observation of anthroposcopic characteristics or metric analysis of linear measurements, and both methods have shown high accuracy percentages.

Some of the problems with using anthroposcopic traits of the pelvis are 1) the negative correlation that exists between age and accuracy of identification due to degeneration of bony features; 2) some features, such as the ventral arc, do not become distinct until the third decade of life; and 3) some of the features are commonly missing due to taphonomic or post-mortem disturbances, such as scavenging, erosion, or breakage.

Traditional metric analyses of the human skeleton have been claimed by some to be more repeatable and more case-inclusive methods of determining sex, which can be performed by less experienced practitioners and can sometimes expose significant areas of variation that may not be readily recognizable via visual observation.

The purpose of this study was to apply modern three-dimensional morphometric techniques to quantify the shape of the os coxae and possibly identify new areas on the os coxa that could be used for sex estimation. Because methods from the geometric morphometrics may make it possible to locate regions of shape variation using a reduced set of landmarks, locating new areas of the pelvis to use for sex estimation could have applicability for forensic fragmentary skeletal remains and in bioarchaeology.

Statistical results showed that the pubis, ilium, and ischium are the most sexually dimorphic regions of the os coxa, whereas differences were not evident in either the acetabulum or obturator foramen. T-test results showed that European American females and African American females differ in size as do the males, also suggesting variation in levels of sexual dimorphism across populations. These results identified the exact areas and direction of shape changes in the os coxa between males and females that is not possible with more traditional analyses. Knowledge of the areas of variability will allow us to develop sexing criteria using a reduced suite of three-dimensional landmarks, which could have applicability to both the forensic and bioarchaeology setting, particularly for fragmentary remains.

Member of The Texas State University System