Gentrification has inspired considerable debate, but direct examination of its uneven evolution across time and space is rare. We address this gap by developing a conceptual framework on the social pathways of gentrification and introducing a method of systematic social observation using Google Street View to detect visible cues of neighborhood change. We argue that a durable racial hierarchy governs residential selection and, in turn, gentrifying neighborhoods. Integrating census data, police records, prior street-level observations, community surveys, proximity to amenities, and city budget data on capital investments, we find that the pace of gentrification in Chicago from 2007 to 2009 was negatively associated with the concentration of blacks and Latinos in neighborhoods that either showed signs of gentrification or were adjacent and still disinvested in 1995. Racial composition has a threshold effect, however, attenuating gentrification when the share of blacks in a neighborhood is greater than 40 percent. Consistent with theories of neighborhood stigma, we also find that collective perceptions of disorder, which are higher in poor minority neighborhoods, deter gentrification, while observed disorder does not. These results help explain the reproduction of neighborhood racial inequality amid urban transformation.

We propose a theoretical framework on the structural sources and spatially embedded nature of three mechanisms that produce collective efficacy for children. Using survey data collected in 1995 from 8,782 Chicago residents, we examine variations in intergenerational closure, reciprocal local exchange, and shared expectations for informal social control across 342 neighborhoods. Adjusting for respondents’ attributes, we assess the effects of neighborhood characteristics measured in the 1990 census and the role of spatial interdependence. The results show that residential stability and concentrated affluence, more so than poverty and racial/ethnic composition, predict intergenerational closure and reciprocal exchange. Concentrated disadvantage, by contrast, is associated with sharply lower expectations for shared child control. The importance of spatial dynamics in generating collective efficacy for children is highlighted—proximity to areas high in closure, exchange, and control bestows an advantage above and beyond the structural characteristics of a given neighborhood. Moreover, spatial advantages are much more likely to accrue to white neighborhoods than to black neighborhoods.

This article develops and assesses the concept of triple neighborhood disadvantage. We argue that a neighborhood’s well-being depends not only on its own socioeconomic conditions but also on the conditions of neighborhoods its residents visit and are visited by, connections that form through networks of everyday urban mobility. We construct measures of mobility-based disadvantage using geocoded patterns of movement estimated from hundreds of millions of tweets sent by nearly 400,000 Twitter users over 18 months. Analyzing nearly 32,000 neighborhoods and 9,700 homicides in 37 of the largest U.S. cities, we show that neighborhood triple disadvantage independently predicts homicides, adjusting for traditional neighborhood correlates of violence, spatial proximity to disadvantage, prior homicides, and city fixed effects. Not only is triple disadvantage a stronger predictor than traditional measures, it accounts for a sizable portion of the association between residential neighborhood disadvantage and homicides. In turn, potential mechanisms such as neighborhood drug activity, interpersonal friction, and gun crime prevalence account for much of the association between triple disadvantage and homicides. These findings implicate structural mobility patterns as an important source of triple (dis)advantage for neighborhoods and have implications for a broad range of phenomena beyond crime, including community capacity, gentrification, transmission in a pandemic, and racial inequality.