ABSTRACT
In the mapping of services, most of the 123 cities in the analysis were barely featured. The principal components analyses of 100 service firms’ office distributions highlighted the important cities for particular services, a few articulator cities and a few more primary field cities, but with the great majority of cities merely appearing in some positive fields and not in others. The reason for this is quite simple: it was the variation among service firms that was being analysed and not the variation among the cities. In this chapter I focus on the latter. I use the same interlocking model, the same large 100 × 123 service values matrix and, for the most part, the same multivariate technique, but I turn the analysis around to bring cities to the fore. Hence below I produce mappings of cities rather than of services.
