Measuring archaeological diversity by comparison with simulated assemblages

A method for measuring diversity is proposed that uses an archaeologically derived underlying frequency distribution of classes of artifacts to generate theoretical expectations for the number of different classes of items that should be found in a collection of a given total size. Because sample size is controlled for, collections of different sizes can be directly compared in a simple graphical display. Because of its rigor and simplicity, this method serves to focus interpretive attention on issues of anthropological (rather than methodological) importance. Examples from the archaeological literature are used to illustrate the operation and potential applicability of this method to a wide range of archaeological problems.