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.