Association analysis

The association between species (rows) in the working data is shown by selecting Association: Chi squared from the drop-down menu. For each species pair, the contingency table of presence/absence is used to calculate a Chi-squared value. For measures of similarity between samples based on species presence-absence, the observations can be summarised in a simple frequency table:

where the number of species present in both samples is a, the number of species present in sample 1 but missing from sample 2 is b, the number of species missing in sample 1 but present in sample 2 is c and the number of species missing from both samples is d. The total number of species, N, is therefore a+b+c+d.

The Chi-squared value for each contingency table is calculated in the normal fashion using the Yate's correction for small numbers of observations.

The Chi-squared values are presented in a grid displayed by clicking on the Chi-squared tab, with the following display conventions:

1. Chi-squared values which indicate a significant degree of positive or negative association between the species at the 5% level are shown in bold.

2. Positive associations are displayed in blue, and negative associations as negative numbers in red.

3. If either of the two species is present in every sample the contingency table is meaningless; N/A is displayed.

Please note that association analysis will give misleading results if the data comprise many zeros caused by low sampling effort, rather than true species presence/absence.