Nonmetropolitan viability in the East South Central Census Division of the United States

The purpose of this study was to examine the patterns and characteristics of nonmetropolitan viability in a sample of towns in the East South Central Census Division of the United States. A total of 103 nonmetropolitan places were examined. In order to test for the significance of size of place, the 1o3 nonmetropolitan places were divided into two size categories; 4,000 to 6,000 and 6,000 to 13,500 population. In analyzing the characteristics of the nonmetropolitan places, eighteen social, economic, and retail sales variables were examined for each town. To further examine factors associated with nonmetropolitan viability, six towns in Tennessee were selected for indepth study.

The viability of places studied was examined using the traditional population change criterion and the less used retail sales change criterion. In addition, viability was measured by absolute change of population and retail sales in the towns, and by change relative to the change in the region. Correlation and discriminant analyses were employed in determining the relationships between population and retail sales change in the towns and eighteen selected socioeconomic variables. In order to account for and explain the spatial patterns of growth and decline, the study region was subdivided into physical provinces and economic areas.

The major findings of the research were that spatial patterns of viability in the study region differed according to whether the population or retail sales change definition was used. The differing spatial patterns of viability were further substantiated by the weak

iv relationship between population change and retail sales volume change. When the viability of the towns was determined by relative change, rather than absolute change, still different spatial patterns emerged. Caution should, therefore, be taken when determining how to measure and define nonmetropolitan viability.

In the section which correlated the eighteen socioeconomic variables to the viability of the towns, several dominant relationships emerged. Growing towns tended to have populations with low employment, high numbers of industrial workers, high median family income, and high sales volumes of medium and high order retail functions.

The discriminant analyses were used to determine if nonmetropolitan viability could be predicted given a selected set of socioeconomic variables. . The highest prediction success rate for the growing and declining towns occurred when viability was measured by absolute change of retail sales. It is believed that absolute retail sales change is the most efficient definitional criterion for viability because retail trade areas are not sensitive to corporate boundaries. . Because the Census Bureau uses the corporate boundary in determining nonmetropolitan growth and decline, suburbanization of population and retail sales establishments adjacent to the corporate boundary are not accounted for and false viability impressions emerge. On the other hand, if a town annexes an adjacent suburban area, false growth is implied. By breaking the corporate boundary, retail sales change gives a more accurate impression of nonmetropolitan growth and decline than population change.