Chapter 6 introduces geostatistical techniques that characterise the spatial structure of a dataset. Spatial dependence is a fundamental property of most data because of Tobler’s First Law of Geography: “everything is related to everything else, but near things are more related than distant things”. Positive spatial dependence means that geographically nearby values of a variable tend to be similar. Spatial autocorrelation is the correlation between paired values of a single variable related to their relative position. Practical reasons to measure spatial autocorrelation include evaluating the statistical validity of spatial analysis, incorporating autocorrelation into models and guiding the design of sampling schemes for data collection. Measures of spatial autocorrelation reveal the nature and dimensions of spatial structure in a dataset (e.g. the join-count test, the Geary’s C and Moran’s I statistics and the semi-variogram and spatial autocovariance functions). Spatial interpolation techniques use similar principles to estimate unknown values of a variable on the basis of known point values using methods such as inverse distance weighting, spline and kriging.