Stature estimation using measurements of the cranium for populations in the United States

摘要

Highlights ? Weak to moderate correlations with stature found for some cranial measurements. ? Sex- and ancestry-specific stature equations perform better than pooled equations. ? Validation testing of regression equations indicates that stature can be estimated accurately. ? Confidence interval ranges are too large to be useful in most forensic casework. Abstract Stature estimation is an important component of the biological profile. Human crania are sometimes recovered in the absence of other skeletal material in forensic casework, and stature estimation equations using cranial measurements have not been developed for populations in the United States. Both simple and multiple regression equations for estimating stature were developed from standard cranial measurements taken from both a cranial CT dataset and the Forensic Data Bank, and the resulting equations were tested using a separate dataset. A weak to moderate correlation with stature was found for some of the measurements tested. Tests of the sex- and ancestry-specific equations and pooled sex equations produced accurate estimated stature ranges for most of the individuals in the test dataset, but large 95% confidence intervals (±14–16cm) for these equations can produce only an imprecise estimated stature range for an unidentified individual. Pooled ancestry equations also produced accurate estimated stature ranges for many of the individuals in the test dataset, but with even larger 95% confidence intervals (±18–20cm). The results of this study indicate that stature can be estimated using cranial measurements, but the resulting 95% confidence intervals produce stature ranges that are too broad to use in most forensic casework.