This paper discusses the use of cranial and dental data in a multifactor and integrated approach to estimating population affinity.
By assessing the utility of a multifactor and integrated approach to estimating population affinity, the following research sought to treat the skull of modern human populations as an integrated biological unit, while considering the potential impact to forensic methodology. Results indicate the combination of multiple data types into a unified model consistently identified four and five latent classes or groups within the sample population. This suggests that an integrated approach to understanding human variation has utility in forensic anthropology. By accounting for more within-group variation, this combined approach adds to a growing body of work that indicates skeletally derived data can provide fine-grained group membership estimates (i.e., population affinity), beyond the broad, continental origins associated with ancestry estimation. Craniometrics, cranial non-metrics, dental metrics, and dental morphology data were collected from skeletal material representing American Black (n=15), American White (n=140), and Latin American Migrant (n=178) populations. Data were analyzed using latent class analysis, a type of finite mixture modeling used to identify unobserved subgroups or classes within a population. This type of mixture model assumes that within the heterogeneous data, there is a set of unobserved, or latent, variables whose pattern can predict class membership. (Published Abstract Provided)