A snapshot of the data I have is below where x is the count of the species recorded at the sites. The data is not replicated. When I try to analyse this matrix in Primer for ANOSIM I get a failure because the assigned factors are not more than 1 level (i.e. Site 1 is group 1, Site 2 is group 2, etc) My question is how do I 'group' the samples for an ANOSIM test? I'm using Primer 5. The null hypothesis is that there is no difference in species composition between sites.
SpA SpB SpC SpD
Site 1 x
Site 2 x x x
Site 3 x
Site 4 x x
Site 5 x x x