The Earth’s surface displays almost incredible variety, from the landscapes of the Tibetan plateau to the deserts of Australia and the urban complexity of London or Tokyo. Nowhere can be reasonably described as an average place and it is difficult to imagine any subset of the Earth’s surface being a representative sample of the whole. The results of any analysis over a limited area can be expected to change as that limited area is relocated, and to be different from the results that would be obtained for the surface of the Earth as a whole. These concepts are collectively described as spatial heterogeneity, and they tend to affect almost any kind of spatial analysis conducted on geographic data. Many techniques such as Geographically Weighted Regression (Fotheringham, Brunsdon, and Charlton, 2002, discussed in Section 5.6.3, Geographically Weighted Regression (GWR), of this Guide) take spatial heterogeneity as given — as a universally observed property of the Earth’s surface — and focus on providing results that are specific to each area, and can be used as evidence in support of local policies. Such techniques are often termed place-based or local.