The spatial distribution of economic activity has been the subject of much theoretical study during the last 150 years. The two-state study which provides primary evidence for this article is, to the authors' knowledge, the first attempt to analyze statistically the determinants of industrial location in rural communities with an explicit objective of more enlightened public action at the local, state, and federal levels.

Approaches for analyzing employment stability with aggregated data for SICS in large regions or major metropolitan areas are misleading indicators of the impact of manufacturing growth in rural areas. Performance of moderate-sized individual establishments seriously impact total employment variation in small-employment-sized rural communities, requiring analysis of the determinants of employment stability of these establishments. Aggregate SIC performance and most conventional criteria for judging probable stability appear to provide very limited predictability for individual firm performance. However, manufacturing development appears generally to have desirable effects on community-wide employment stability.S THE TITLE SUGGESTS, THIS PAPER addresses the special problems of A employment stability and labor utilization in rural communities and the effects of employment creation on them. More specifically, it focuses on the effects of growth in manufacturing industry on communities whose historical economic base had been coal mining, small-farm agriculture, and some forestry. The study Eldon D. Smith is a professor of agricultural economics at the University of Kentucky, Lexington KY 40546-0276. Funds to finance data collection and partial personnel support for analysis were provided by the Economic Development Administration under cooperative agreement with the Economic Research Service (ERS), US. Department of Agriculture. Dr. Donald Larson was principal representative of ERS and contributed greatty to design and later analysis. Mr. David Peters did most of the preliminary analyses, which were presented in an M.S. thesis, but he is not responsible for modified analyses presented here. Invaluable supplementary data was provided by the Human Resources Cabinet, Commonwealth of Kentucky. project of the Kentucky Agricultural Experiment Station and is published with the approval of its Director.The investigation reported in this paper (90-1 -191) is connected with a

A linear probability function permits the estimation of the probability of the occurrence or non-occurrence of a discrete event. Nerlove and Press (p. 3–9) outline several statistical problems that arise if such a function is estimated via OLS. In particular, heteroskedasticity inherent in such a regression model leads to inefficient estimates of parameters (Amemiya 1973, Horn and Horn). Moreover, without restrictions on the conventional OLS model, probability estimates lying outside the unit (0–1) interval are possible (Nerlove and Press). Goldberger and Kmenta suggest two approaches for alleviating the heteroskedasticity problems inherent in the OLS regression model. Logit analysis will also alleviate heteroskedasticity problems and ensure that estimated probabilities will lie within the unit interval (Amemiya 1974, Hauck and Donner, Hill and Kau, Horn and Horn, Horn, Horn, and Duncan, Theil 1970).