It has been suggested that a substantial proportion of the excess stroke mortality among black Americans may be attributable to relatively lower socioeconomic status (SES) in this group. In this report we provide the first quantitative estimates of the proportion of excess black stroke mortality attributable to SES for a large population-based cohort. We used data from the National Longitudinal Mortality Study for persons 45 years and older (73,400 white men, 87,528 white women, 6522 black men, and 8816 black women). Sex-specific proportional hazards model were used to estimate excess black stroke mortality with and without adjustment for education and income (measures of SES). The contribution of SES to the excess black stroke risk was estimated from the difference in regression coefficients for race in these models. In men, low SES was associated with increased stroke mortality (P < or = .0001) and accounted for 14% to 46% of the excess black stroke risk (P < .05). However, we could find no association between SES and stroke mortality in women, and SES did not account for a significant proportion of the excess stroke mortality in black women. Although SES proved to account for a statistically significant proportion of excess male black stroke mortality, overall SES explained less than one quarter of the observed excess between ages 45 and 65. In women, SES did not significantly reduce the estimated excess black stroke mortality. Although SES may be playing a role in excess black stroke mortality, a substantial proportion of the excess appears attributable to other sources, including cerebrovascular risk factors that are unrelated to SES, unmeasured lifestyle influences, social resources, and genetic factors.