Skeletal element independence and implications for quantification

Autor: Valerie Sgheiza
Rok vydání: 2021
Předmět:
Zdroj: Journal of Archaeological Science: Reports. 35:102796
ISSN: 2352-409X
DOI: 10.1016/j.jasrep.2021.102796
Popis: This work was undertaken to test skeletal element types for independence of recovery. This informs the accurate representation of uncertainty in existing estimators of number of individuals, such as Most Likely Number of Individuals (MLNI). Calculations based on elements that have dependent recovery probability will underestimate uncertainty. Appendicular skeletal inventories were collected from three forensic and three archaeological samples of individuated remains (N = 201). Analyses were performed in R on a distance matrix derived from correlation of 18 left and right skeletal elements. The effect of dependent element types and dependent antimeres on MLNI estimation was tested via simulation from a multivariate normal distribution with an initial population size of 100, 3 element types, and a 50% recovery rate. Hierarchical clustering showed a consistent division between the upper and lower extremities. The clavicle showed potential association of antimeres and buried remains showed a positive correlation between recovery rate and recovery relatedness. According to simulations, element type dependence increases MLNI variability without a corresponding increase in Highest Probability Density Region (HPD) width, resulting in reduced HPD accuracy. Antimere dependence causes a rapid decline in MLNI values, reducing HPD accuracy five times faster than type dependence. When calculating MLNI, avoidance of highly dependent antimeres should be prioritized. The analyst should then select taphonomically robust elements from both the upper and lower extremities.
Databáze: OpenAIRE
Externí odkaz: