Reliability Study of Methods for Scoring a Non-Metric Human Osteological Trait

Abstract

To gain meaningful insights from non-metric trait analysis in the field of human osteological study, issues of reliability and context need to be addressed, especially for the analysis of discrete cranial traits. A preliminary study tested the reliability of different methods of quantifying wormian bones, with the purpose of establishing a consistent method that would enable further applicability for this and other non-metric traits in mortuary analysis. The determination of reliability for both methods was made using Olsson and Janson’s (2001) iota statistic together with Pearson’s product-moment correlation. This study examines the reliability of scoring methods on an interobserver scale, an imperative step for the utility of these techniques for the wider archaeological community, as the majority of current data collection is a collaborative effort. The results of this study support the initial application in a case study involving a Bayesian probability analysis utilizing individuals from the Milwaukee County Institutional Grounds collection.