A Significance Test for Morisita’a Index of Dispersion and the Moments when the Population is Negative Binomial and Poisson
The moments of Morisita’s index of dispersion are derived assuming the observed counts follow negative binomial and Poisson distributions. The moments are expressed as truncated infinite series. Bounds are placed on the truncation error. The rate of convergence to normality of the index appears to be slow for populations with low mean density. A significance test for the comparison of the dispersions of two populations is suggested. Data from a census of the presence of the southern green stinkbug (Nezara Viridula) on three crops are used to examine the effect of sample size and quadrat size on the power of this test and on confidence interval coverages.
KeywordsMaximum Likelihood Estimator Negative Binomial Distribution Negative Binomial Confidence Interval Estimation Quadrat Size
Unable to display preview. Download preview PDF.
- Boutwell, J. R. 1982. Computation of moments of Morisita’s index of dispersion under binomial sampling. MAMS Thesis, University of Georgia, Athens, Georgia.Google Scholar
- Chiang, H. C. 1981. Moments of Morisita’s index of dispersion assuming negative binomial frequency counts. MAMS Thesis, University of Georgia, Athens, Georgia.Google Scholar
- Lloyd, M. 1967. Mean Crowding. J. Anim. Ecol. 36: 1 – 30.Google Scholar
- Morisita, M. 1959. Measuring of the dispersion of individuals and analysis of the distributional patterns. Mem. Fac. Sci. Kyusha Univ., Ser. E, 2: 215 – 35.Google Scholar
- Morisita, M. 1962. index as a measure of dispersion of individuals. Res. Pop. Ecol. 4: 1–7.Google Scholar
- Reed, W. J. 1983. Confidence estimation of ecological aggregation indices based on counts — a robust procedure. Biometrika39: 987 – 998.Google Scholar
- Stiteler, W. M. & G. P. Patii. 1969. Variance-to-mean ratio and Morisita’s index as measures of spatial patterns in ecological populations. Stat. Ecol. 1: 423–59, Penn State Press.Google Scholar
- Voramongkol, A. 1982. Moments of Morisita’s index of dispersion assuming a Poisson distribution of Counts. MAMS Thesis, University of Georgia, Athens, Georgia.Google Scholar