U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Cranial Nonmetric Variation and Estimating Ancestry

NCJ Number
228500
Journal
Journal of Forensic Sciences Volume: 54 Issue: 5 Dated: September 2009 Pages: 985-995
Author(s)
Joseph T. Hefner, Ph.D.
Date Published
September 2009
Length
11 pages
Annotation
This research focused on the frequency distribution and inter-trait correlations of 11 common morphoscopic traits in order to demonstrate that the experience-based approach to ancestry prediction is an unscientific art, because it is unreplicable, unreliable, and invalid.
Abstract
Ten of the 11 traits examined had frequency distributions with significant differences between groups (African, Asian, European, and Native-American), but the range in variation of these traits far exceeded previous assumptions. These within-group variations show that extreme trait expressions are not reliable for estimating ancestry through visual observation alone. Instead, traits should be analyzed within a statistical framework. The morphoscopic approach to ancestry determination using traits pioneerd by Hooton has been tested within a statistical framework in order to assess the validity and reliability required to comply with the Daubert challenge. The problem of subjectivity in scoring traits (reliability) can be largely solved by using standard drawings, following Hooton and Walker. The optimal weighing of the traits seen in an individual to proudce the best prediction of ancestry (validity) can be achieved through statistical methods and reference group trait distributions. Only when these are used together can nonmetric traits be reliable, replicable, and valid indicators of ancestry. In order to examine trait variation among groups, 11 common morphoscopic traits were collected for 747 individuals. Following the typical forensic model, populations were grouped according to geographic ancestry and a pooled, four-group model was used for all subsequent analyses. 14 figures, 14 tables, and 38 references