Metric characterization of the human coxal bone on a recent Italian sample and multivariate discriminant analysis to determine sex.

The ability of human pelvic bones to sexually differentiate has been of great interest in forensic anthropology for quite some time as it allows for the determination of skeletal sex by combining metric and morphological data. However, the criteria for determining the sex of a skeleton must be calibrated according to the variability of the population to which it belongs. The aim of this work is the metric characterization of the human coxal bone on a recent sample (of known sex) from the region of Apulia, in southern Italy, in order to establish its efficacy in sex determination by way of multivariate discriminant analysis. Seventeen standard anthropological measurements used in sex determination were taken from 168 coxal bones (78 males and 90 females) all belonging to 86 adult skeletons (40 males and 46 females). The bones used were taken from subjects who had died in the 1960s and 1970s in Apulia. The results obtained define the variability in size and proportion of the sample analyzed with respect to the variations of other skeletal populations. Nine discriminant functions, utilizing between 4 and 11 variables, have been shown to be useful in determining the sex of coxal bones, whether they be complete, partial, or fragmented. All of the functions selected resulted in an attribution error equal to zero, and differ only in the number of variables utilized and by the degree of separation between the groups. The results of this study confirm the validity and utility of diagnostic techniques based on discriminant functions as reported in the literature for other population groups. The combination of metric characteristics from various regions of the coxal bone is, therefore, a valid aid in the correct attribution of skeletal sex even when the combination of variables is numerically limited, but sufficient in sex determination from partial coxal bones.