An analysis of spatial patterns of polarization in the Brandon and Dauphin areas : the testing of a growth centre hypothesis
This thesis represents a methodological investigation of the validity of the growth centre concept against the backdrop of S.W. Manitoba. Theoretically, the growth centre is very efficient planning instrument for regional development. In this respect, the grwoth centre can be treated as the cornerstone of an investment strategy involving the spatial concentration of investment and the accompanying reorganization or restructuring of geographic space. Furthermore, growth centre programmes can be used as means for realizing specific policy objectives relating to social welfare servicing. There is, however, a marked discrepancy between theory and actuality. This is manifested in the following queries: Do spontaneous growth centres exist? Do these growth centres really affect, either positively or negatively, the development of surrounding areas? If they do, then, what spatial patterns, if any, will be formed? The principal purposes of this thesis will be an attempt to unravel these questions in the aforementioned regional context, namely: (1) to examine whether or not 'growth centres' exist at Brandon and Dauphin respectively, (2) to examine the presumed repercussions of two opposing spill-over effects and, then, to consider the degree of growth transmission from Brandon and Dauphin to the surrounding rural areas, (3) to examine the spatial form of polarized growth around Brandon and Dauphin. In order to compare the effects of different analytic techniques and to get a more precise picture of the development surface in S.W. Manitoba, twelve different techniques, including unrotated and rotated factoring methods, were employed in this study to analyze the same set of data for S.W. Manitoba. The outputs from these factoring techniques were, in turn, regarded as development indexes for a series of cubic trend-surface analyses which were established to test the polarization hypothesis. Finding provided only equivocal support for the existence of development-response surfaces based on Manitoban spontaneous growth centres.