I came across ANOSIM and I would like to know your opinion about the following issue. I am unsure about it, so the question (and the approach) could turn up to be erroneous and ingenuous. If it is the case, sorry
Once one has devised the existence of, say, two groups of items (let's say rows profile points) by means of simple Correspondence Analysis on a contingency table, does it make any sense to use 1-way ANOSIM in order to test the significance of the difference between the two groups?
Advertise on Talk Stats