Comparing polluted vs non-polluted sites - Statistical approach?


New Member

In a forest area I want to compare two groups: polluted sites vs non-polluted sites. This is done by collecting species from three different sites for each group.

What is the best statistical approach to check if there are any significant differences between the two groups?

If I average the species abundance for the three sites for each group I can run a simple f-test for variance and t-test for means, but I`m not sure if that is the best way to approach the problem. I also am planning running multivariate analysis with Bray-curtis similarity index and visualize the data in a dendrogram or mds-ordination to check for any significant groupings.

Attached is an example of what my data will look like (the actual data will have more species).

Any help is much appreciated!
Last edited: