Calibrating paleodemography using quasi-stable population methods: a meta-analysis of ancient and modern populations

Robert McCaa, University of Minnesota

A meta-analysis of paleo and modern datasets reveals strengths and weaknesses of paleodemographic methods that go unnoticed when research is restricted to a few sites limited in time and space. This paper uses quasi-stable population methods to derive fertility estimates for populations in three large bodies of data: 1. the Health and Nutrition in the Western Hemisphere database containing more than 12,000 skeletons representing three ethno-racial populations and ranging over eight millennia; 2. eighteen published collections of skeletal data from Europe, Africa and North America (n=5,792); and 3. seven historical datasets compiled from written records (n=81,306). Where conventional paleodemographic methods yield point estimates of life expectancy, combining quasi-stable population methods and sampling theory leads to a range of estimates based on empirical distributions of the data as well as the effects of variations in mortality on populations with stable fertility.