Statistical Analysis of Deer and Elk Pellet-Group Data
Gary C. White, Lester E. Eberhardt · Journal of Wildlife Management · 1980
Several authors have demonstrated empirically the fit of observed pellet-group data from mule deer, white-tailed deer, and elk to the negative binomial distribution. This distribution is described by 2 parameters: the mean, m, and the positive exponent, k. The parameter k is a measure of contagion. As over-dispersion increases, k ..-->.. 0; conversely, as the pellet groups approach a random distribution, k ..-->.. infinity. Past application of the negative binomial distribution to pellet-group data required a common value for the parameter k in order to test for differences in mean number of pellet groups for different populations. We developed a method of testing for differences in the means of pellet-group data for 2 or more populations that does not require a common k value. Interval and point estimators of k and m were obtained by the method of maximum likelihood for a series of models where either m or k can be assumed constant for all populations. A set of FORTRAN subroutines is available to analyze negative binomial data.