The Measurement of Ecological Segregation

T HE PURPOSE of this paper is to devise objective and mensurative referents for the ecological concept of "segregation." The basic data utilized in this presentation were taken from the series of census tract bulletins on Population and Housing prepared by the United States Bureau of the Census in connection with the sixteenth decennial census. Out of a total of 6i tracted cities in I940, two samples of 44 and 25 cities, respectively, were selected for analysis. Although the concept of ecological segregation can be applied to any population group or class, data for the Negro population were selected as being particularly appropriate and significant for a report of this kind. Theoretically, there is virtually no limit to the number and variety of indexes of ecological segregation which might be constructed. More than twenty logically sound and computationally feasible indexes have been formulated. The definitions of "complete segregation" and "no segregation" are identical in each case. It must not, therefore, be assumed that any one index is necessarily superior to all others or that all are equally good for all purposes. The basic criterion for selecting a particular