The Effect of Quadrat Size on the Estimation of the Parameters of Neyman's and Thomas's Distributions

Publisher Website
Google Scholar
Abstract
Models of Neyman''s type A and Thomas''s distributions, with the clusters occupying appreciable areas, were constructed and sampled with quadrats of various sizes. Results show that these distributions are indistinguishable by quadrat sampling and that the size of quadrat used greatly affects the form of observed frequency distributions obtained. In all cases the derived statistic which should give an estimate of number of points per cluster was too low. The highest and most nearly correct estimates of this parameter were obtained by sampling a population with quadrats of several sizes and calculating the regression of log percentage absence on density. The constant term in the regression equation gives some idea of the degree of diffuseness of the clusters constituting the population.

This publication has 3 references indexed in Scilit: