Ancestry estimates: Evaluating the reliability of Hefner's cranial morphoscopic method

Abstract

To identify unknown skeletal remains, forensic anthropologists provide police with information of who they might belong to, such as ancestry (someone’s familial lineage and geographic origin). The cranium has shape-based traits (morphoscopic traits) that can be scored using visual analysis, and these scores are used to estimate ancestry. The purpose of this study is to evaluate the reliability of Hefner’s (2009) morphoscopic trait scoring method, which assesses sixteen traits, as well as the impact of score disagreement among and within observers on the resulting ancestry estimates. Reliability is determined through intra-observer and inter-observer repeatability tests, whereas the impact of score disagreement is observed by comparing ancestry results generated by statistical programs from each observer’s scores. In general, most traits have high intra-observer agreement, most trait scores are in agreement, and lower inter-observer agreement. Each trait has their own pattern of disagreement, such as a score of 2 and 3 were always confused with each other for the trait anterior nasal spine, but never a score of 1. Score disagreements caused ancestry estimates to change between observers in most cases. Error causing lower inter-observer agreement included experience, tool use, method iteration, prevalence of traits within the study individuals, vague descriptions, and interpretation differences. This is the first study to thoroughly assess and identify sources of error, as well as provide recommendations for improved descriptions/pictorial representation of all sixteen of Hefner’s traits. Overall, Hefner’s method requires pictorial and description improvement for the majority of traits before it can be reliably used among practitioners.