Masters Theses

Date of Award

5-2017

Degree Type

Thesis

Degree Name

Master of Arts

Major

Anthropology

Major Professor

Dawnie W. Steadman

Committee Members

James A. Fordyce, Lee M. Jantz

Abstract

The use of Bayesian theory has been gaining popularity within the forensic anthropology community for its ability to model the way in which decisions are made based upon varying levels of confidence. However, many forensic anthropologists have been reticent to adapt Bayesian approaches given the general lack of knowledge in regards to this approach (Konigsberg and Frankenburg, 2013:153). The purpose of this thesis is to demonstrate how the application of an establish Bayesian framework can be used to determine likelihood ratios representing the probative value of skeletal lesions consistent with cancer for use in forensic personal identification. To do this, a sample of adult individuals from the William M. Bass Donated Skeletal Collection (BDSC) at the University of Tennessee who self-reported as having cancer near the time of death or having had a previous cancer diagnosis (n=302) was used to create likelihood ratios representing the weight of macroscopic lesions consistent with cancer in determining a correct identification of a skeleton. A random sample of these individuals (n=149) were analyzed for the presence/absence of macroscopic lesions and is the focus of this study. The sample size was then used to represent the “population at large” using Steadman, et al.’s (2006) likelihood ratio for pathological conditions of the skeleton as a model. These derived likelihood ratios represent the relative “weight” of evidence regarding identification when lesions consistent with cancer are present. The likelihood ratios can then be utilized within the likelihood ratio approach to identification established by Steadman, et al. (2006) to potentially support or refute a presumed identification of skeletal remains using antemortem records and postmortem skeletal data.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS