Does neighborhood socioeconomic status (SES) have a significant positive effect on health over and above the personal and household socioeconomic status of the residents who live there, and, if it does, what is its relative importance compared to an individual's own SES? Resolution of this question has been impeded by lack of conceptual clarity in the definitions of socioeconomic status on the micro- and macro-levels, and unreliable and noncomprehensive adjustment for micro-level socioeconomic status. Using the Community, Crime and Health Survey (CCH), based on a representative sample of Illinois households with linked census tract information, we find that, with adjustment for personal socioeconomic status, residents of socioeconomically disadvantaged neighborhoods have significantly higher levels of physical impairment than do residents of more advantaged neighborhoods. The neighborhood effect is small compared to individual socioeconomic status, especially education, employment status, household income, and economic hardship, all of which have larger associations with health than does neighborhood socioeconomic status (measured as home ownership, college-educated adults, and poverty). In comparison, home ownership on the micro-level does not have a significant effect on physical functioning. We conclude that about 40 percent of the association between neighborhood socioeconomic status and individual health is contextual and about 60 percent is compositional.