A Family of Five is Not the Same as One Household: The Effects of Disaggregation on Demographic Outcomes in Archaeological Simulation Models

Summary

Many archaeological agent-based computer models (ABMs) use the household as the smallest unit of investigation but, in order to answer questions about how factors such as disease, social interaction, and population movement contributed to population dynamics in prehistory, there is a need for individual-level models. Our team has worked to disaggregate an early archaeological ABM, the Artificial Anasazi model, into an individual-level model, the Artificial Long House Valley model. The baseline version of the latter model has been designed to be as similar as possible to the original model and uses identical environmental data and demographic parameters. A new version of the model maintains the structure of the original, but includes updated environmental and demographic input data.

We compare the results of simulations using the household-level and both individual-level models, and discuss how factors required for disaggregation, such as individual level fertility and mortality, contribute to differing results. Most notably, individual-level simulations generate populations that are significantly smaller than those generated by the household-level model. Through this process, we have learned that disaggregation of a model from the household level to the individual level is not a straightforward process and can result in the emergence of unexpected differences.

This Resource is Part of the Following Collections

• Society for American Archaeology 81st Annual Meeting, Orlando, FL (2016)

Keywords

Spatial Coverage