A reassessment of the Acsádi and Nemeskéri complex method for adult age estimation using South African and Dutch skeletal sample

Abstract

The Acsádi and Nemeskéri Complex Method is a multifactorial method that enables the use of different combinations of skeletal age indicators to assess adult age-at-death. The Complex Method employs the use of four skeletal elements in assessing age, including the degree of endocranial suture closure, age-related changes to the pubic symphyseal surface, and the internal structural changes observable in the proximal humerus and femur. Acsádi and Nemeskéri (1970) reported a very optimistic error margin of ±2.5 years when all four skeletal elements were used. Despite these promising results, the Complex Method’s statistical validity and effectiveness have been questioned, suggesting that it results in a constant over- and underestimation of age in younger and older groups, respectively. In this study, the Complex Method was tested on a Dutch archaeological population (n = 167) and two modern South African skeletal samples (n = 381) with individuals of known age-at-death. In addition, the use of Bayesian statistics was employed, using the scoring criteria of Acsádi and Nemeskéri (1970), to see whether a statistically improved method might yield better results than the averaging technique used in the Complex Method. Results from this study to some extent justify the critique levelled against the Complex Method’s tendency to over- and underestimate age. Nevertheless, the Complex Method still performed surprisingly well for certain skeletal element combinations, substantiating its continued use as an age estimation method for European archaeological populations. Contrary to Acsádi and Nemeskéri’s (1970) findings, the use of all four skeletal elements did not provide more accurate age estimates. For the most part, combinations making use of three or two skeletal elements outperformed the use of all four skeletal indicators. Population- and sex-specific differences were noted in this study, and therefore, it is suggested that the Complex Method may be optimised by making use of sex- and population-specific skeletal element combinations and weighting systems in the future. The Bayesian method provided some improvement as it decreased levels of over- and underestimation in the younger and older age groups, especially for females. It seems that there is a benefit in using a weighting system, given that for the SA white males the Complex Method outperformed the Bayesian method in combinations that required weighting.