Integrating Growth Variability of the Ilium, Fifth Lumbar Vertebra, and Clavicle with Multivariate Adaptive Regression Splines Models for Subadult Age Estimation

Abstract

Subadult age estimation should rely on sampling and statistical protocols capturing development variability for more accurate age estimates. In this perspective, measurements were taken on the fifth lumbar vertebrae and/or clavicles of 534 French males and females aged 0-19 years and the ilia of 244 males and females aged 0-12 years. These variables were fitted in nonparametric multivariate adaptive regression splines (MARS) models with 95% prediction intervals (PIs) of age. The models were tested on two independent samples from Marseille and the Luis Lopes reference collection from Lisbon. Models using ilium width and module, maximum clavicle length, and lateral vertebral body heights were more than 92% accurate. Precision was lower for postpubertal individuals. Integrating punctual nonlinearities of the relationship between age and the variables and dynamic prediction intervals incorporated the normal increase in interindividual growth variability (heteroscedasticity of variance) with age for more biologically accurate predictions.

Osteological evidence of short stature and parieto‐squamosal arch craniosynostosis in a non‐adult male from the 13th century Leiria, Portugal

“…Finally, the tibia is 30.6% smaller than the tibia of MB61‐336 (Figure d). Indeed, the shortness of the long bones is distinctive features observed for SM‐100 and is consistent with age ranges between 9.5 and 12.5 years when based on the clavicle (Primeau, Friis, Sejrsen, & Lynnerup, ), ilium (Corron et al, ), femur (Rissech, Schaefer, & Malgosa, ), and tibia (Maresh, ). The estimation of stature based on the regression equation for boys (femur) is 134.6 cm.…”

Mixed cumulative probit: a multivariate generalization of transition analysis that accommodates variation in the shape, spread and structure of data

“…Notable non-TA age estimation models can accommodate a mix of variable types, such as multivariate adaptive regression splines [43][44][45], but these are mostly regression models in a frequentist framework. Regression models have some inherent failings and, ultimately, may be more restrictive than useful in biological anthropology [41].…”

“…This is circular and counterproductive when used for age estimation since any division of an age estimation model into two or more 'sub-models' requires one to estimate (or guesstimate) an unknown's age prior to selecting the appropriate submodel and then estimating age. Other researchers have implemented nonlinear regression methods [43][44][45], but still face issues with heteroscedasticity and subsequent interpretation.…”

The over‐citation of Daubert in forensic anthropology

“…Examples of this misunderstanding within forensic anthropology are demonstrated by the use of language that exceeds that of the Daubert ruling. Approximately half of the articles that cited Daubert did so using constraining language, referring to the factors or guidelines as "criteria" or stating that one or more factors were "required," often in the introduction sections of the paper when providing a general background of the relevant legal framework (e.g., [15,[36][37][38][39][40][41]). Some authors went further in their interpretations, implying or expressly stating that purely objective or quantitative methods were required by Daubert (e.g., [36,[42][43][44]).…”