![Sample our Global Development journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5929/TOC-Subject-Sample-2021-Global-Development.jpg"})
Abstract
This article reflects on the racial configuration of urban space. Previous research tends to posit racial segregation and diversity as either endpoints on a continuum of racial dominance or mirror images of one another. We argue that segregation and diversity must be jointly understood; they are necessarily related, although not inevitably as binary opposites. Our view is that the neighborhood geographies of U.S. metropolitan areas are simultaneously and increasingly marked by both racial segregation and racial diversity. We offer an approach that classifies neighborhoods based jointly on their compositional diversity and their racial dominance, illustrated by an examination of the neighborhood racial structure of several large metropolitan areas for 1990 and 2000. Compositional diversity increased in all metropolitan areas in ways rendered visible by our approach, including a sharp reduction in the number of highly segregated white neighborhoods, transitioning mostly into moderately diverse yet still white-dominated neighborhoods, and a fourfold increase in the number of highly diverse neighborhoods. Even so, many highly segregated spaces remain, especially for whites and blacks. Latino-dominated spaces show a mix of persistence and emergence. Although compositional diversity is increasing, highly diverse neighborhoods are still rare and are the least persistent of all racial configurations. Our approach clearly demonstrates the “both/and”-ness of segregation and diversity.
Este artículo versa sobre la configuración racial del espacio urbano. Las investigaciones previas tienden a situar la segregación racial y diversidad bien como puntos finales de un continuo de dominación racial, o como imágenes de espejo de unas y otras. Nuestro argumento es que la segregación y la diversidad deben entenderse en conjunto; estos fenómenos están necesariamente relacionados, aunque no inevitablemente como binarios opuestos. Nuestro punto de vista es que las geografías vecinales de las áreas metropolitanas de los EE.UU. están simultánea y crecientemente marcadas tanto por segregación racial como por diversidad racial. Presentamos un enfoque que clasifica los vecindarios basados conjuntamente en su diversidad composicional y predominio racial, ilustrados con un examen de la estructura racial vecinal de varias áreas metropolitanas grandes para 1990 y 2000. La diversidad composicional se incrementó en todas las áreas metropolitanas de maneras que se hacen más visibles con nuestro enfoque, incluyendo una pronunciada reducción del número de vecindarios de dominio blanco altamente segregados, en proceso de transición hacia vecindarios moderadamente diversos pero todavía con predominio blanco, y un incremento cuadruplicado en el número de vecindarios altamente diversificados. Con todo, todavía permanecen muchos espacios altamente segregados, especialmente de blancos y negros. Los espacios dominados por latinos muestran una mezcla de persistencia y emergencia. Aunque la diversidad composicional está aumentando, los vecindarios de alta diversidad todavía son raros y son los menos persistentes en todas las configuraciones raciales. Nuestro enfoque claramente demuestra la continuidad de condiciones de segregación y diversidad racial.
Palabras clave:
Notes
1. We follow the convention of scholars using Census data in defining as Latino all respondents reported as having Hispanic origin, regardless of their reported race. We understand well the conceptual inadequacy of this single category to capture the rich and varied histories and contemporary realities of racial identity for persons with Latin American heritage.
2. Allen and Turner (1997, 2002) are notable exceptions. They effectively demonstrate this link for Los Angeles with a pair of tract-level maps designed to be used together, one of diversity and the other of the predominant ethnic group. We share their conviction that maps provide an essential tool in understanding the complexities of urban racial configuration but wish to create a single system that jointly depicts the two constructs. Ethington (Citation2000) coined the phrase “segregated diversity” to describe the evolution of Los Angeles's neighborhood racial configuration as a combination of increased diversity linked to increased white isolation. His report includes many maps, many of which are also available online at http://www.usc.edu/schools/sppd/research/census2000/race_census/racecontours/la.htm. Brown and Sharma (Citation2010) also argued that compositional diversity and segregation, which they referred to as residential intermixing, are distinct constructs that intersect in meaningful ways. They focused, however, on entire metropolitan regions rather than the neighborhoods, which constitute our primary concern.
3. We agree with Peach's (2009) main critique of the Johnston et al. framework, which centers on their criteria for identifying and naming “ghettos”—practices that he argued artificially create, rather than discover, these contentious spaces.
4. The standard entropy diversity measure for a tract is:
| |||||
5. If any of the five minority groups constitutes less than 3 percent of tract population, scaled entropy drops rapidly.
6. We initially evaluated a 50 percent rule (i.e., no group constitutes an absolute majority) based on precedent in the literature, but that approach captured too many tracts and tracts with odd characteristics. The minimum scaled entropy for a tract satisfying the 45 percent rule would be a relatively low .53—but setting a threshold at .53 would allow tracts into the high-diversity category that we did not consider to be highly diverse.
7. The fact that the threshold for defining high-diversity tracts is twice the threshold for defining low-diversity tracts is purely coincidental.
8. It would be easy to also identify the dominant racial group in tracts defined as high diversity. We chose not to do so in this article because (1) we want to highlight that in high-diversity tracts, no single group is in the clearly visible majority; and (2) there were numerically few tracts that satisfied our criteria, and depicting variants of these tracts based on the dominant group introduced too much complexity into our analysis, especially the cartographic representations.
9. We define metropolitan areas as consolidated metropolitan statistical areas (CSMAs) that consist of multiple contiguous metropolitan areas (primary metropolitan statistical areas [PMSAs]) and freestanding MSAs as defined for the 1990 Census. For narrative convenience we refer to all areas as MSAs.