This paper makes comparisons between the nearest-neighbor distance and quadrat methods, both of which are used for analyzing the pattern of the spatial distribution of a given set of points, in terms of the degree to which the results yielded by the methods are mutually dependent. First, we discuss the conditions that should be satisfied if the two methods always yield exactly the same results. Second, based on the discussion, we develop a Monte-Carlo-simulation-based method for examining whether the conditions hold. Third, by using the method, we show that the conditions do not hold and that the nearest-neighbor distance and quadrat methods quite frequently yield different results, which may be totally opposite. We however also show that the results are mutually dependent to a certain extent. Fourth, we develop indices for quantifying the extent, and show that the indices are very low irrespective of parameters used in analysis. Finally, we conclude with suggestions for future study.
View full abstract