Abstract

In this paper we present results of a screening study which sought to investigate the effects of using different sized meshes for the recovery of archaeological data. The implications of using various sizes of screen mesh on types and frequencies of archaeological material are discussed within the confines of statistical sampling procedures. That the quality of results from archaeological excavations and the direction of interpretations of prehistory are closely linked with the methodology is difficult to question, yet the variety of opinions regarding which of the data recovery techniques is most accurate is considerable. Statistical sampling serves to elucidate this situation and provide some structure to the kind of archaeological interpretation that is based on retrieval systems. The goal of proper sampling centers on obtaining statistically valid samples of sites, features, or artifacts existing within a defined sampling universe. In all cases, properly executed sampling schemes assume that information recovery is complete. This working assumption can be accepted if the item in question can be recovered in total. This study examines the practical validity of these highly structured approaches to archaeological interpretation.