Forensic Anthropological Research in Indonesia: Developing a Sex Estimation Standard for the Contemporary Population

Throughout its history, Indonesia has experienced a high frequency of mass fatality events, both natural and the result of human action. Despite this, there is a lack of forensic anthropological research in the country. Skeletal remains recovered are often disarticulated or fragmentary in nature. As such, the merits of morphological assessment were considered to address this gap in the literature: they are easy to deploy, can be applied to fragmentary remains, and are commonly used in modern forensic practice. One of the most utilised morphoscopic standards for the estimation of adult skeletal sex is that of Walker (2008).

Independent testing of the reliability and accuracy of the Walker method has demonstrated reduced classification accuracy and varied trait frequency distributions when applied to populations outside the United States (e.g., Greece, Italy, Japan, South Africa, and Thailand). Secular variation, the environment, lifestyle, and socioeconomic status are known to influence sex-based differences between population groups. Therefore, the application of standards on a population different to which it was developed is not acceptable; appropriate modification to the Walker method to suit an Indonesian population would thus serve to improve the identification process of unknown skeletal remains in both DVI and routine medicolegal casework.

There are no known documented physical skeletal collections available in Indonesia. Due to religious restrictions on invasive physical examinations of decedents, digital imaging modalities, such as CT scanning, represent an appropriate proxy to physical skeletal material. As such, the present study statistically quantifies the applicability of the Walker method in an Indonesian population based on the analysis of MSCT scans. In total, 200 cranial MSCT scans are analysed; scoring of cranial traits was performed following the written descriptions and illustrations in Walker (2008). Intra-observer agreement was high for all traits, ICC ≥ .730. Bilateral asymmetry was not statistically significant across all paired traits. Binary logistic regression predictive models using all five traits correctly classified 85.71% of females and 93.10% of males. These results provide forensic practitioners in Indonesia a more appropriate forensic standard, strengthening their capabilities and improving judicial outcomes.