Number of Random Selections

When calculating species richness or fitting a species accumulation curve, the order in which the samples are listed can have a large impact on the result. To avoid this problem Species Diversity and Richness can randomise the sample order many times and produce an average for the number of species observed after 1, 2 3 .. n samples.

On the top bar there is a drop-down menu that allows the selection of the number of random selections of sample order (R).

If you do not want the order of the samples randomised then set a value of 1.

For the species richness estimators, R determines the number of randomisations of the samples that will be undertaken. The order in which the samples are analysed can have a large effect on richness estimators. For example, consider a situation in which one sample was unusually species-rich and held many species represented by only one or two individuals. If this were the first sample in the series then it could result in exaggerated species richness estimates. However, if it were the last it would probably have little effect, as most of the species would already have been recorded in other samples. By randomising sample order and averaging the estimates obtained over all the randomisations we remove this sample order effect.

The number of randomisations should be chosen with care. A large data set with a large number of runs can take a few minutes to calculate. If your data set is large (above 50 species by 50 samples) try a small number of randomisations (about 10) first. If the calculations are undertaken rapidly then increase the number and consider if it has made the estimates more consistent. There is little point in choosing a large value of R if you have a small number of samples.

The number of randomisations of sample order applied to the data is shown at the base of the program window. The image below shows runs = 10, which is a typical default when a data set is opened in Species Diversity and Richness.