1. Department of Statistics and Actuarial Science , University of Waterloo , Waterloo, Ontario, N2L 3G1, Canada;2. Department of Mathematics and Statistics , University of Guelph , Guelph, Ontario, NIG 2W1, Canada
Abstract:
A simple modification of a T-square sampling procedure for studying unmapped spatial distributions allows for the collection of several distance measurements at each randomly selected sampling location. A test of the null hypothesis of a completely random distribution of point items using these data is found to have power comparable to a related test based on T-square sampling if the number of items of data is held fixed, and to have greater power if the number of sampling locations is held fixed.