Spatial analysis

Spatial analysis covers analytical techniques that use where things are, their location, distance, adjacency or arrangement, to find patterns that plain tabular analysis cannot see. Typical operations include buffering (finding everything within a distance of a feature), overlay (combining layers to find intersections), and clustering of point events.

It relies on a geographic information system to store and query the underlying data, and on geocoding to place addresses or place names onto the map in the first place. A key technique, hotspot analysis, tests statistically whether events cluster more than random chance would predict, rather than just eyeballing a dense-looking area on a choropleth map, which can be visually misleading because of uneven region sizes or population.

Spatial analysis underpins site selection, logistics routing, environmental monitoring, epidemiology and crime analysis, wherever the relationship between location and an outcome matters. Common pitfalls include the modifiable areal unit problem, where results change depending on how boundaries are drawn or aggregated, and spatial autocorrelation, where nearby observations are not independent, which violates the assumptions of ordinary statistical tests and requires spatially aware methods instead.