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

Agreement and error rates associated with standardized data collection protocols for skeletal and dental data on 3D virtual subadult crania

NCJ Number
305169
Journal
Forensic Science International Volume: 334 Dated: May 2022
Author(s)
Louise K. Corron ; Kristen A. Broehl ; Elaine Y. Chu ; Tatiana Vlemincq-Mendieta ; Christopher A. Wolfe; Marin A. Pilloud; G. Richard Scott; M. Kate Spradley; Kyra E. Stull
Date Published
May 2022
Annotation

Since observer error and agreement rates for craniometrics, odontometrics, and cranial and dental morphological traits have been inconsistently evaluated on three-dimensional cranial reconstructions and almost never assessed on subadult individuals, this study uses a computed tomography (CT) scan sample of 12 subadults aged between birth and 20 years from the Subadult Virtual Anthropology Database (SVAD) to evaluate intra- and inter-observer error and agreement rates associated to these four types of data on virtual crania.

 

Abstract

Forty-eight cranial landmarks, 33 standard inter-landmark distances (ILDs), 13 cranial macromorphoscopic traits, four permanent and four deciduous dental landmarks and measurements per tooth, and 21 permanent and 12 deciduous dental morphological traits were collected on each individual. Results matched or improved on published standards for dry bones, teeth, or dental casts. Technical Error of Measurement (TEM) associated with metric data ranged from 0.00 mm to 0.99 mm and relative TEM ranged from 0% to 5.76%. Cohen’s kappa coefficient values for agreement on morphological traits scores were above K = 0.5 for 90% of the traits. Type III cranial landmarks showed higher error rates than Type I and II cranial landmarks. Agreement on dental morphology scores seemed influenced by observer experience and rater agreement improved when using di- or tri-chotomized grades. Skeletal maturity did not significantly affect error rates, meaning most craniofacial and dental metrics and morphological traits can be reliably obtained from virtual subadult crania. (Publisher abstract provided)