Skip to content

Spatial analysis - Category Home

Spatial analysis - Topics

Is a law of geography a real law or just a strong empirical regularity? How does distance decay shape the results of a gravity model? How do you test whether a set of points is spatially random? Choropleth or proportional symbol: which shows a rate better? How do you weight criteria in multi-criteria site selection without making the result arbitrary? What does Moran's I actually tell you about a mapped variable? Spatial autocorrelation: why nearby places resemble each other more than chance allows Nearest neighbour analysis: testing whether a point pattern is clustered or random Kriging vs inverse distance weighting: which suits rainfall maps from sparse gauges? What the quantitative revolution of the 1950s changed about how geographers work Euclidean distance vs network distance in accessibility studies: when does the gap matter? When should you use geographically weighted regression instead of one global model? Why can changing the spatial weights matrix flip your conclusions? The quantitative revolution: when geography started running regressions instead of describing regions Why is population usually published as a gridded raster instead of per administrative district? What is the modifiable areal unit problem and why does it wreck comparisons between regions? Kernel density estimation: how do you choose the bandwidth honestly? How do you find spatial outliers rather than global clusters? Tobler's second law: why what happens outside a region shapes what happens inside What is the ecological fallacy in spatial analysis? Tobler's first law: everything is related to everything else but near things are more related Distance decay explains why your second nearest hospital may be your usual one Gravity models of interaction: how population and distance predict trips between two places First law of geography vs the death of distance — which one does the evidence support? Accessibility is not proximity: measuring what people can actually reach in thirty minutes Why does spatial autocorrelation break the assumptions of ordinary regression? The modifiable areal unit problem: how redrawing a boundary changes the answer you get