Test of the multifactorial aging method using skeletons with known ages-at-death from the Grant Collection.

The multifactorial aging method has been shown to be a highly reliable method of skeletal aging because it incorporates age information from as many age indicators as are available for each skeleton (Lovejoy et al.; Am. J. Phys. Anthropol. 68:1-14, 1985). The present study was a blind test to assess its accuracy on a skeletal sample composed of 55 individuals with verified death certificates (Grant Collection, University of Toronto). Three authors (C.O.L., M.E.B., and K.F.R.), with no access to the death certificate ages, independently seriated and aged the sample using three to four criteria: auricular surface, pubic symphysis, and radiographs of the proximal femur and clavicle. Summary ages were then calculated for each individual in the sample. The authors' independent summary age estimates showed strong correlations with one another (r = 0.84-0.89). Multifactorial age estimates correlated better with real age than did those from any single indicator used. The mean error (averaging 8.7 years) for summary age was at least 1 year less than that for any single indicator. Average bias ranged from -0.7 (underage) to 1.4 (overage) years. These results indicate that utilization of several age indicators, weighted according to their reliability, helps control for variation in the changes that occur with age in any single morphological indicator. This method may therefore be considered one of the most accurate available for the determination of skeletal age-at-death, particularly for paleodemographic analysis.