|Top Previous Next|
Simpson's index for each selected sample is displayed with this option. A simple plot of the way the index changes between samples is displayed by clicking on the Graph tab on the output window.
A diversity index proposed by Simpson (1949), to describe the probability that a second individual drawn from a population should be of the same species as the first. A similar type of index had a few years earlier been proposed by G. Yule to compare an author's characteristic vocabulary (frequency of different words in his writings). The statistic, C (or Y) is given by:
where, SOBS is the number of observations and, strictly,
but is usually approximated as:
where Ni is the number of individuals in the ith species and NT the total individuals in the sample. The index is:
and the larger its value the greater the diversity.
The statistic 1 - C gives a measure of the probability of the next encounter (by the collector or any animal moving at random) being with another species (Hurlbert 1971). May (1975) showed that this index is strongly influenced for values of SOBS > 10 by the underlying distribution. As Magurran (2004) states "Simpson's Index is heavily weighted towards the most abundant species in the sample, while being less sensitive to species richness."
However Magurran (2004) also states "The Simpson index is one of the most meaningful and robust diversity measures available. In essence it captures the variance of the species abundance distribution.
To compare two indices see Testing for significant differences between indices.