Applying bootstrapped Correspondence Analysis to archaeological data

This paper examines the usefulness of bootstrapping in Correspondence Analysis when applied to archaeological data. By simulating and displaying possible variation within the data sets, bootstrapping provides us with a means to assess the stability of our CA maps and influences the interpretations we can place upon them. Five real data sets are examined and the results discussed. The paper concludes that bootstrapping is a useful and powerful way of examining the results of CA and should be employed on a regular basis.     

Refining correspondence analysis-based ceramic seriation of regional data sets

Ceramic seriation is one of the primary tools used by archaeologists to create chronologies, especially for surface survey data. In this article, we outline a bootstrapped approach to correspondence analysis-based seriation designed to help assess and improve the stability of relative orderings produced through such analyses. This procedure systematically identifies and removes small samples and sites with unusual samples, such as those with multiple components, which are not handled well by seriation and require secondary interpretation. Our approach combines data from multiple projects in the Zuni region of the American Southwest in order to gauge the effects of intraregional variation in ceramic distributions and reexamine prior interpretations of demographic change based on individual projects. This analysis highlights two previously unknown regional trends, including variation in the distribution of multicomponent sites and potential short-term depopulation of a significant part of the region. Based on this example, we further suggest that our method may be particularly useful in situations where little prior chronological information is available.     

Mapping the possible occurrence of archaeological sites by Bayesian inference

Regional archaeological prospections are often done by field walking, where the location of the sampled fields is often determined by factors like feared disturbance or recent plowing. The resulting data configuration can be suboptimal for spatial prediction of the archaeological potential by geostatistical methods like kriging. As an alternative, we propose a Bayesian method to map the possible occurrence of archaeological finds and compare this to indicator regression kriging. Three types of predictive models were implemented in the Bayesian context following deductive, inductive and mixed approaches to use auxiliary geographical information in the mapping. After prediction of a validation set, it was concluded that the mixed approach gave the best results in terms of map quality, and that the kriging method performed poorly. Usage of data on the presence and the absence of archaeological finds is to be preferred above usage of presence data only. Furthermore, a method is presented that filters those parts of a predictive map that are not strongly supported by evidence.