Testing simulated positive spatial autocorrelation by Getis-Ord general G

Abstract

Getis-Ord general G provides researchers with a straightforward way to assess the degree of spatial association in an entire sample. It measures how concentrated the high or low values are for a given study area. In order to check the ability at detecting global spatial autocorrelation pattern, several clustering arrangements were designed and G values were calculated. The results of the standardized G value and p-value indicate that G index is sensitive to the size of the clustering region as well as the location, a larger size and a near-center location is preferred. The selection of threshold distance will also affect the testing results, sometimes contradictive patterns are reported at different threshold distances in a same study area.