Race as an Element in the Intra-City Regionalization of Atlanta's Population

RACE AS AN ELEMENT IN THE INTRA-CITY REGIONALIZATION OF ATLANTA'S POPULATION Donald R. Deskins, Jr.* INTRODUCTION. Begionalization and its related problems have long plagued geographers. The question of how to effectively determine homogeneous and nodal regions has to date consumed much professional time and energy and is yet to be satisfactorily resolved. Recently, factor analysis has been employed as a technique to regionalize socio-economic data aggregated on the census tract scale in metropolitan areas. (1) The employment of factor analysis in the process of regionalization has permitted geographers to reduce and simplify large numbers of spatially distributed variables into a few manageable factors, thus expediting interpretation. Although the efficacy of factor analysis as a regionalizing technique has been demonstrated, there are numerous techniques equal to the task of regionalization, among them are found the various geographical grouping or clustering algorithms. (2) STUDY PURPOSE AND RATIONALE. It is the purpose of this paper to examine the effectiveness of a single grouping algorithm in the determination of Atlanta's intra-city population regions. It is not the intent of this paper to argue that a geographical grouping technique is better adapted to regionalizing census tract data than is factor analysis. On the contrary, it is only the aim of this effort to demonstrate that a geographical grouping technique can also expedite the process of regionalizing the population found in a city. With this aim in mind, Atlanta's population was selected for examination. Atlanta, like most southern cities has not been as intensively studied as cities in the North (e.g., Chicago). (3) This point alone is rationale enough for examining the population patterns of Georgia's capital. However , Atlanta's population has several other attributes which justify its selection as the focal point of this study. Atlanta not only contains the seat of state government, but it is also a regional distribution center which has as its hinterland much of the southeastern United States. In addition to its commercial importance, it is also a major center of black culture. Atlanta has long been an intellectual center for black Americansnoted for the many traditionally black institutions of higher learning which comprise the Atlanta University Center. Moreover, the black residential community is well established. The longevity of Atlanta's black residential areas increases the chance that residential patterns between races as well as within racial categories will appear when *Dr. Deskins is assistant professor of geography at The University of Michigan, Ann Arbor, and Director of the AAG Commission on Geography and Afro-America (COMGA). The paper was accepted for publication in August 1971. Vol. XI, No. 2 91 geographical grouping techniques are applied to socio-economic data aggregated by census tract. Other studies of Atlanta have identified and verified residential regions stratified by race. (4) The regions resulting from these efforts are displayed in Figure 1 and will be compared with regionalization resulting from the forthcoming analysis. Sources: see footnote 4. Figure 1. Residential Regions, Atlanta, 1960 92 Southeastern Geographer White High Income 1.1Ansley Park 1.2Buckhead 1.3Northside FIGURE 1 LEGEND Residential Regions Black High Income White High-Medium Income 1.4Cascade Heights White Medium Income 1.5Ben Hill 1.6Bolton 1.7Center Hill 1.8Druid Hills 1.9Morningside 1.10Peachtree Hills 1.11Perkerson Park Black High-Medium Income 2.1Collier Heights Black Medium Income 2.2Adamsville 2.3Atlanta University Center White Medium-Low Income White Loto Income 1.12Cabbagetown 1.13Bellwood 1.14Tight Squeeze Black Medium-Low Income 2.4 West Atlanta Black Low Income 2.5Bedford Pines 2.6Butler Street 2.7East Atlanta 2.8Mechanicsville 2.9Perry Homes 2.10Plunkett Town 2.11Summer Hill 2.12Thomasville 2.13West End 2.14Vine City DATA AND TECHNIQUE. The data used in this examination consist of the 20 variables listed in Table 1. These data were drawn from the U. S. Census of Population: 1960 for the City of Atlanta and compiled by census tracts yielding a universe of 106 tracts. To control for variance among the 20 variables, the raw data comprising each variable set were standardized by employing a Z transformation. Actually the Z scores resulting from the transformation of the raw data comprising each variable...