Although spatial epidemiology may at first appear to be very different from other areas of epidemiology, it has many features in common. In particular, a major goal of epidemiology—to infer etiological relationships from observed associations—applies equally well to geographical studies. The characteristic feature is that risk varies from place to place, and geographical location is an important explanatory variable for many reasons: it reflects environmentally determined elements of risk, spatial autocorrelation is very common in geography, distance-based relationships between people are an important risk factor for infectious diseases, and so on. The two-dimensional nature of geographical location means that standard statistical techniques for dealing with sets of essentially univariate variables must be supplemented by more sophisticated methods.