This analysis reexamines factors affecting farm change during the Farm Crisis using spatial analysis techniques to identify important spatial factors and correct for spatial autocorrelation. Results indicate the importance of indicators of farm structure, percent of prime farmland, and state-level processes in predicting changes in the number of farms. The findings also suggest a spatially-dependent process: counties near each other in space experienced similar types and rates of change in the number of farms. Reasons for the effects of geographic proximity are explored. (Contains 1 figure, 4 tables and 16 footnotes.)