Fieldworker Experience and Single-Episode Screening as Sources of Data Recovery Bias in Archaeology: A Case Study from the Central Pacific Northwest Coast

Abstract

Despite increasing concern with the effects of archaeological data recovery methods on the types and quantity of objects extracted from the material record, archaeologists rarely discuss recovery biases attributable to the most basic excavation procedures. In this study I examine how several factors, including variable artifact identification skills, can affect artifact recovery rates in the field. Data from household-level investigations at the Stó:lō (Coast Salish) village of Welqámex (DiRi 15) are presented to show how interobserver variation can compromise interpretations of past behavior when opportunities for artifact recovery are limited to observations at the excavation unit and single-episode (field-only) sieving. Laboratory sorting of screen residue retained in 3.2-mm (one-eighth-inch) mesh sieves is shown to account for the recovery of as much as 87.5 percent of lithic artifacts and nearly 90 percent of archaeofaunal remains. Rates of artifact recovery in the field are highly variable among excavation crews working at Welqámex, and I argue that the application of correction factors is inappropriate unless the magnitude of recovery bias can be measured for specific excavation teams and particular depositional contexts. The results of this study further highlight the importance of documenting and mitigating the unintended effects of methodological decisions on archaeological assemblages.