A comparison of age estimation using discriminant function analysis with some other age estimations of unknown skulls

Available methods of estimating the ages of unknown skulls are notoriously inaccurate. Making use of a unique opportunity to view several hundreds of remains from two nineteenth century cemeteries, various methods of estimating age were tested. A statistical method which gives relative weighting to each age indicator before reducing the data to a single pooled age estimate was found to be the most reliable. The greater the number of criteria which were utilized for age estimation, the more highly correlated was the estimated age of remains with the actual age at death. The degree of accuracy in estimating the ages of skulls of under 20 years was relatively high, when only the stage of development of dental tissues was examined. With increasing age over 20 the accuracy decreased, particularly at ages over 45. The range of age distribution of skulls estimated from published data on suture closure tended to be more restricted than that which may acutally occur.