Spatial mismatch research in the 1990s: progress and potential *

. This article reviews recent research about the spatial mismatch hypothesis from a range of social science disciplines. Since 1990, researchers have tested the mismatch hypothesis in diverse metropolitan settings; devised more accurate measures of geographical access to employment; and developed models to address issues such as compensating variations, sample selection bias, and contextual effects. We argue for a broader conceptualization of spatial mismatch that considers how social and spatial relations affect employment outcomes for women, immigrants, and other ethnic minorities. This broader view will enhance the contribution of research to current theoretical and policy debates about urban poverty. The effects of metropolitan context and neighborhood‐level differences in services, resources, and social networks on spatial access and, independently, on wages and employment also warrant future research attention.