Using spatial statistics to assess aggregation effects

Recently there has been a resurgence of interest in the analytical problems referred to as aggregation effects. A number of recent papers have provided new insights into the nature and extent of aggregation effects attributed to scale and zoning. This paper presents results from exploratory analyses examining the correlation between aggregation effects and a set of spatial statistics. In particular, a modification of the G statistic introduced by Getis and Ord (1992) performs very well for a set of variables from the County of Lancashire in the U.K. We conclude that the modified G statistic holds promise as a diagnostic statistic, given our operational definition of aggregation.