We applied a public health approach to the study of mental retardation by providing a basic descriptive epidemiological analysis using a large statewide linked birth and public school record database (N = 327,831). Sociodemographic factors played a key role across all levels of mental retardation. Birthweight less than 1000 g was associated with the highest individual-level risk, but the impact varied considerably, depending on maternal educational level. Low maternal education was associated with the largest effects at the population level for mild and moderate/severe mental retardation. Focusing exclusively on specific biomedical causes is of little use in developing public health plans; a broader biosocial perspective reflecting the interactive complexity of the risk factors comprising the various etiological patterns is needed.