Body mass estimation in skeletal samples using the hybrid approach: the effect of population-specific variations and sexual dimorphism
Abstract
Body mass is estimated from skeletal records with low accuracy, and it is expected that population-specific equations derived by a hybrid approach may help to reduce the error in body mass estimates. We used 204 individuals from five Central European Early Medieval sites to test the effect of population-specific femoral head breadth equations on the accuracy of body mass estimates.
The baseline for living body mass was computed using the biiliac breadth and stature. We also analyzed the agreement of five general femoral head techniques that are used in body mass estimation (Elliott et al. (Archaeol Anthropol Sci 1-20, 2015b; Grine et al. (Am J Phys Anthropol 97:151-185, 1995); McHenry (Am J Phys Anthropol 87:407-431, 1992); Ruff et al. (Am J Phys Anthropol 148:601-617, 2012); Ruff et al. (Am J Phys Anthropol 86:397, 1991)).
Our results support previous findings showing that body mass is predicted with lower accuracy than stature, even when population-specific equations are derived. However, the population-specific approach increases the agreement with the body mass estimated from the biiliac breadth and stature, particularly when sex-specific equations are used.
Thus, our results advocate for the employment of sex-specific equations when possible and show that the possibility of deriving equation for each sex separately is the main advantage of the population-specific approach. The best agreement among the body mass techniques in the Central European Early Medieval samples was observed using the femoral head equations reported by Ruff et al. (Am J Phys Anthropol 148:601-617, 2012) and McHenry (Am J Phys Anthropol 87:407-431, 1992), whereas other studied equations provided lower agreement.
The particularly low performance obtained using the technique reported by Elliott et al. (2015b) questioned the use of their equations to estimate body masses.