The Awful Truth about Statistics in Archaeology

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

The archaeology of the past two decades has become increasingly quantitative, computerized, statistical, and this is as it should be. All right-thinking archaeologists begin with samples and attempt to generalize about the populations from which their samples were drawn. Statistical theory has evolved to assist investigators in making just this important inferential step and archaeologists have increasingly turned to statistics to square their research with the canons of Science. But the statistical revolution in archaeology is not without its price. We must now face the fact that all applications of statistics to archaeology can no longer be applauded. The archaeological literature is badly polluted with misuses and outright abuses of statistical method and theory. This paper discusses some of these faulty applications and makes some recommendations which, if heeded, should improve the quality of quantitative methods in archaeology.

Related collections

Most cited references 44

Record: found
Abstract: found
Article: not found

Archaeological Systematics and the Study of Culture Process

Lewis Binford (1965)

It is argued that the normative theory of culture, widely held among archaeologists, is inadequate for the generation of fruitful explanatory hypotheses of cultural process. One obvious shortcoming of this theoretical position has been the development of archaeological systematics that have obviated any possibility of measuring multivariate phenomena and permit only the measurement of unspecified “cultural differences and similarities,” as if these were univariate phenomena. As an alternative to this approach, it is proposed that culture be viewed as a system composed of subsystems, and it is suggested that differences and similarities between different classes of archaeological remains reflect different subsystems and hence may be expected to vary independently of each other in the normal operation of the system or during change in the system. A general discussion of ceramic classification and the classification of differences and similarities between assemblages is presented as an example of the multivariate approach to the study of cultural variability. It is suggested that a multivariate approach in systematics will encourage the study of cultural variability and its causes and thereby enhance the study of culture process.

0 comments Cited 52 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

Testing Contemporaneity and Averaging Radiocarbon Dates

Austin Long, Bruce Rippeteau (1974)

Techniques for pragmatically interpreting arrays of radiocarbon dates are given. These include weighted averaging, statistical rejection of data, and the evaluation of contemporaneity. Commonly encountered situations are discussed with several actual examples to illustrate these procedures. This paper is Contribution No. 70, Department of Geosciences, University of Arizona.