Background: Depression is a very important part of global mental health concerns. Many of the studies on correlates of depression stopped short of finding the predictors. Predictive models will empower preventative efforts by healthcare providers and policy makers. The purpose of this study was to determine the factors predicting depressive symptoms among a population of older men and women in rural South Africa. Methods: Data were obtained from "Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community in South Africa (HAALSI) in the INDEPTH Health and Demographic Surveillance System (HDSS) site of Agincourt" in rural Mpumalanga province, South Africa. Previously validated short-version Center for Epidemiologic Studies Depression Scale (CES-D 8) was used to assess for depressive symptoms. Multivariable logistic regression model with stepwise selection, and receiver operating curve were used to examine the predictors of depression. Results: Of the 4027 participants included in this study, 743 (18.5%) met the criterion for depression (CES-D 8 score ≥3). Older age (OR 1.025, CI 1.016-1.034), diabetes (OR 1.467, CI 1.152-1.868), and alcohol consumption (OR 1.536, CI 1.261-1.872) predicted depression. Being male (OR 0.734, CI 0.588-0.915) and homemaker rather than not working (OR 0.513, CI 0.372-0.707) were protective. Compared to those who were married, depressive symptoms were significantly higher among the separated/divorced (OR 1.372, CI 1.027-1.834) and the widowed (OR 1.468, CI 1.172-1.839). Conclusions: It is possible to predict the development of depression in this community, and findings are generalizable to other communities and countries. Healthcare workers and policy makers should use the findings for preventative care and policies.