1Department of Evolutionary Anthropology, Duke University, 2Institute for Human Evolution, University of the Witwatersrand
Saturday All day, Clinch Concourse
Sex determination is a critical part of developing a biological profile for unidentified human skeletal remains. When more commonly used elements for estimating sex, such as the pelvis, are not available, methods utilizing other skeletal elements are needed. Previous studies have suggested a pattern of sexual dimorphism in the lumbar vertebrae, which may be due to biomechanical differences in the way males and females bear load through the vertebral column. In addition, as an important weight-bearing region, the lumbar vertebrae should reflect body size, which on average tends to be larger in males than in females. Using osteometric data from the lumbar vertebrae, statistical analyses (SPSS 20) were conducted on a sample from the Dart Collection (47 males, 51 females) to evaluate the usefulness of the lumbar vertebrae in estimating sex. Eleven characters—including linear dimensions as well as the vertebral body wedging angle—were examined at each lumbar level (L1 to L5). Multiple variables at each lumbar level showed significant sexual dimorphism (p<.01), and univariate discriminant function analysis (DFA) generated equations with accuracies ranging from 57.7% to 83.5%. The highest accuracies tend to be associated with dimensions of the vertebral body. Stepwise DFA generated discriminant functions with cross-validated classification rates of 88.7% for L1, 81.4% for L2, 85.9% for L3, 80.5% for L4, and 75.9% for L5. These results are comparable to other methods that use discriminant functions for sexing the skeleton, and indicate that measures of the lumbar vertebrae can be used as an effective tool for sex assessment.