Abstract: Average life expectancy is an important measure of the physical and socio-economic conditions in modern demography. A common approach to the calculations of the average life expectancy is the life table. However, recent work calls into question the usefulness of this method in prehistoric demography because of poor burial and conservation conditions for infants and younger age groups, making the skeletal data incomplete and underrepresentative. Remarkable theoretical progress in demographic statistics sheds new light on skeletal remains from Neolithic graveyards in terms of survival analysis. If the age at death can be treated as a continuous random variable that follows a specific probability distribution function, then the mortality probability at any age of death can be derived within the framework of survival analysis, which would allow a calculation the average life expectancy. Although few studies on the average life expectancy of Neolithic man in East China have been conducted using the life table method, little is known about the characteristics of probability distribution of age at death for this prehistoric population.
In this study, we investigated 44 excavation reports on human skeletal remains from the Dawenkou Cultures, Haidai area. Among these sites, nine graveyards had a large sample size, and therefore were selected for statistical analysis of age at death. Histogram, P-P probability plots, skewness, and kurtosis were used to test the normality of the probability distribution of age at death of Dawenkou man. Our results show that age at death of these nine archaeological sites closely follow the normal distribution rather than the Weibull distribution. We also discussed potential applications of normality of age at death in archaeological demography.

Key words: Haidai; Dawenkou Cultures; Age at death; Normal distribution; Prehistoric demography