Estimating stature in South African populations using various measures of the sacrum.

The accurate estimation of stature can meaningfully contribute to the positive identification of skeletal remains of unknown individuals. Regression equations are most commonly used to estimate stature and have been derived from numerous bones. Little is however known about stature estimation from the sacrum and therefore the aim of the current study was to evaluate whether it is possible to estimate stature from various sacral measurements in South African populations. A total of 210 sacra representing 50 black males, 51 white males, 58 black females and 51 white females were measured. These measurements included anterior sacral height, anterior sacral width, maximum transverse diameter of the base, anterior sacral height of individual sacral segments (S1, S2, S3, S4, S5) as well as combinations of anterior sacral heights of segments, (i.e. S1+S2). The correlation of these measures to stature was calculated using Pearson's correlation coefficient, followed by the calculation of linear least square and multiple regression equations for the estimation of stature. Statistical analyses indicated differences between sexes and population groups with multiple regressions being slightly more accurate than regression equations derived from single variables. Standard error of estimate values for the present study fall within the ranges stated by previous researchers, with smaller stature estimation intervals reported for black South Africans and females. This study suggests that regression equations derived from the sacrum of South Africans can be used to calculate reasonable estimates of stature, in the absence of bones that yield more accurate results.