Keywords

sex estimation, talus, linear and geometric morphometric analyses, ancient Maya, ancient Egyptian, modern white American

Abstract

Estimating biological sex is an essential first step when analyzing human remains in both forensic and archaeological contexts. While the pelvis and skull are traditionally used for this purpose, they are not always present or sufficiently preserved to permit analysis. The talus is sexually dimorphic and can yield highly accurate sex estimates, making it a promising alternative skeletal element for sex estimation. As there is currently no accepted protocol for sex estimation using the talus, the present study attempts to replicate and validate existing methods using linear and three-dimensional landmark data, while comparing data from ancient Maya (n=16), ancient Egyptian (n=80), and modern American (n=60) samples. Based on Wilcoxon Signed-Rank tests and Procrustes ANOVAs, side asymmetry in the talus was not identified. For the modern American sample, Linear Discriminant Analyses (LDA) achieved up to 87% accuracy in sex estimation and set points achieved up to 90% combined accuracy in sex estimation. Canonical Variate Analysis (CVA) yielded 100% accuracy in shape space and 88% accuracy in form space for the Lamanai sample, and 58% accuracy in shape space and 75% accuracy in form space for the NMDID sample; the Lamanai sample is small (n=8) and the NMDID sample is larger (n=60) and therefore more reliable. It appears that size is a significant factor in the overall sexual dimorphism of the talus, and therefore it must be taken into consideration. Analyses of linear data may be ideal in field contexts with constrained time and resources, and analyses of three-dimensional data may be ideal in archival contexts with a significant amount of missing data. These results carry important implications for studies in paleoanthropology, bioarchaeology, forensic anthropology, and other studies predicated on accurate sex estimation.

Completion Date

2025

Semester

Summer

Committee Chair

Williams, Lana

Degree

Master of Arts (M.A.)

College

College of Sciences

Department

Anthropology

Format

PDF

Identifier

DP0029586

Language

English

Document Type

Thesis

Campus Location

Orlando (Main) Campus

Download

Share

COinS