Decision trees for combining morphological traits and measurements of the skull for osteological sex estimation

Abstract

Forensic anthropologists commonly estimate osteological sex using separate metric and morphological analyses, without integrating both data types into a single statistical model. Combining data types into one classification model has the potential to increase sex classification accuracies for the skull. Therefore, the present study seeks to improve sex classification accuracies for the skull by combining morphological and metric variables using decision trees. The main objectives are to (1) generate multiple decision trees that combine metric and morphological variables, (2) compare the classification accuracies of the generated trees to current standard osteological sex estimation methods, and (3) compare the results of the combined data trees to separate morphological and metric trees. The sample included 212 European Americans (males = 106, females = 106) and 191 African Americans (males = 114, females = 77). Decision trees were trained on 80% of the sample and tested using a 20% holdout sample. Multiple trees were generated using 12 morphological and 14 metric variables. The skull (87.9%–100%) and cranium (90.9%–100%) models achieved higher accuracies compared to the mandible models (72.7%–92%). Additionally, the pooled, population-inclusive models performed as well as or better than the separate population models. Overall, the combined-data models attained higher classification accuracies than previous studies that integrated skull measurements and morphological traits, as well as compared to separate decision trees for both data types. Future research should continue to explore implementing decision trees for osteological sex estimation, including models combining metric and morphological variables from multiple skeletal regions.