Analyzing Randomness in Point Patterns: An Algorithmic Approach

There are multiple methods available to determine if a set of points is distributed randomly, clustered, or dispersed. These methods usually involve comparing the expected distances between points to the actual distances, assuming no barriers are present. However, what qualifies as a "random" distribution can be affected by various socio-environmental factors like wetlands or transportation networks. This tool introduces a series of spatial analysis steps and statistical tests to explore the relationship between observed point patterns and potential spatial drivers (e.g., polygons). If any of these drivers influence the point distribution, the assumption of randomness must be reevaluated. The algorithm is implemented in Python as a QGIS script, with two primary phases: the first deals with overlay operations and preliminary calculations, and the second applies a chi-square goodness-of-fit test with Bonferroni correction..