Abstract
This study investigates the inconsistencies between two major public databases—NamUs and the Doe Network—that are used in the identification of unidentified human remains in the United States. Despite significant advances in forensic science and the establishment of these databases to aid in the identification process, thousands of individuals remain unidentified, raising questions about data consistency and reporting standards. This research analyzed 50 matched profiles from both databases using a mixed-methods approach, combining statistical analysis of quantitative variables with qualitative assessment of descriptive information. Results revealed that while most discrepancies between the databases were not statistically significant, race and location data were exceptions due to variations in terminology and specificity. Additionally, Doe Network was found to include more narrative detail, while NamUs prioritized concise, data-driven entries. These findings underscore the critical need for standardized, comprehensive documentation practices across platforms to improve identification efforts, reduce bias, and support future technological applications such as AI. The study calls for more uniform data reporting and mandatory case submission protocols to enhance database completeness and usability in forensic investigations.
