Identifying Misidentifications: Fordisc 3.1 Application to Korean Crania

When cases are of unknown identity or race, determining the population of the individual significantly increases the chances of determining the identity of that victim. Forensic anthropologists have been using skeletal methods to estimate population affinity for decades. Fordisc (Jantz and Ousley 2005) is the primary method used by forensic anthropologists when building biological profiles for forensic cases. The results from Fordisc are assumed by users to be accurate, however, Fordisc can often yield misleading results or incorrect affinities. This means that the use of Fordisc to estimate population affinity is potentially misleading the biological profiles of forensic cases. Fordisc has only thirteen populations to compare cranial measurements to (five female and eight male), so if a case is outside of those populations, it is impossible for Fordisc (Jantz and Ousley 2005) to provide an accurate affinity. This study aims to find a pattern in the outputs produced when the cranial measurements are from a population not within the populations available. The measurements of 365 cases, 219 males and 146 females, were provided by the National Forensic Service in South Korea for use in this study. The craniometric data will be entered in the application following standard protocols. We expect the craniometric data to yield low probabilities or indecisive results. We hypothesize that the cranial measurements will affiliate with Japanese, Guatemalan, and Chinese crania due to the close regions and morphological history of the populations. We will conduct a statistical analysis of the typicality probabilities and posterior probabilities provided to find potential patterns in the output that indicate the affiliation could be incorrect. Even though we are unable to fix the programming of the application, if forensic anthropologists can begin to notice the possibility of misleading outputs, then the biological profiles will not be impeding the investigations.