Ageing the elderly: A new approach to the estimation of the age-at-death distribution from skeletal remains

This study reports on the use of a proportional measure to estimate the age-at-death distribution of an assemblage and, when combined with a seriation method, additionally estimate the age-at-death of individuals. Traditional methods of estimating age-at-death suffer from a number of issues, including decreasing accuracy with increasing age, age mimicry of the reference population, and difficulty balancing accuracy with precision. A new method is proposed for estimating the age-at-death distribution of middle and older adults. As the age-at-death distribution is significantly impacted by the fertility rate, it was hypothesised that the D0-14/D ratio (the number of individuals who died aged 0–14 years divided by the total population; an indicator of fertility) may be able to estimate the proportion of individuals that might be expected to die in each five-year age group over 35 years. The method permits the estimation of individual age when used in conjunction with seriation methods and the age-at-death distribution of a population. The method is tested on two samples of known age, the Spitalfields crypt and St Thomas' Church cemetery collections and found to provide greater accuracy over previously applied methods.