The State Soil Geographic Data Base (STATSGO) developed by the Natural Resources Conservation Service is a spatially explicit database of soil characteristics known to influence the distribution, abundance, or quality of vegetation, which ultimately may affect white-tailed deer (Odocoileus virginianus) population phenotype. The national availability of STATSGO makes it an attractive tool for wildlife ecologists to use for explaining variation in deer population morphometrics. We constructed linear regression models relating site-specific mean population 1.5-year female body mass and 2.5-year male antler size to soil pH, organic matter, bulk density, and available water capacity values from STATSGO to evaluate the explanatory power of the database. For comparison, we developed general linear models that related the same deer morphometrics to nominal classification variables that represented 4 soil regions in Mississippi, USA. At the statewide level, regressions using site-specific soil pH, organic matter, bulk density, and available water capacity values from STATSGO did not explain more variation than did regional classification models. Although STATSGO did not generate good models for predicting or explaining localized variation in white-tailed deer morphometrics in Mississippi, it did reveal differences in soil properties that may influence white-tailed deer populations at regional spatial scales. At the statewide level, soil resource region classification models may be a more effective tool because they explained similar variation in deer morphometrics, and model input is easier to obtain.