Abstract
The conventional mode of factor analysis, known as the R-mode, yields a typology or classification of study areas by factor weighting characteristics, but does not hold much promise for the delineation of regions without the introduction of additional constraints for regional contiguity and compactness. Q-mode analysis, on the other hand, offers greater potential for regionalization by emphasizing correlations between study areas rather than between variables. By introducing the contiguity constraint into the mathematical procedure (in the form of a binary adjacency matrix), the author derives a set of agricultural regions for the Crimea that bears some similarity to type-of-farming regions obtained by traditional techniques. The author suggests that even better results might be obtained if the compactness constraint could also be introduced into the mathematical procedure.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
