This study examines the process of older adults' driving retirement within the framework of the Selection, Optimization, and Compensation (SOC) model. Data for this study are from the Naturally Occurring Retirement Community Demonstration Project (Department of Health and Human Services, Administration on Aging, Award 90AM2612) that examined the needs of elderly individuals dwelling in the community in a geographically bound section of Saint Louis County, Missouri. Participants (N=281) were included in this study if they fell into one of our three comparison groups: those who still drive (n=212), those who now drive less (n=33), and those who have ceased driving (n=36). Findings indicate that as compared to those who still drive or those who have given up driving, the group who drive less have the lowest self-rated health, highest severity of chronic medical conditions, the greatest amount of depression and stress, the least amount of social resources such as informal help, and the greatest level of unmet need for formal transportation services. Furthermore, the group who drive less has given up the greatest proportion of social activities yet have not given up as many instrumental activities as the group who have given up driving. This supports the SOC framework by providing a cross-sectional depiction of those drivers who are in the process of selecting and optimizing their driving behaviors. Implications for practice include a need for more interdisciplinary collaboration to identify elders who are in the process of driving retirement and to intervene earlier to help ease this transition.