significative adonis test, but no significance in paired adonis test?

Im comparing microbiota from different insect species, and i did by Permanova (or adonis) and i got a pvalue less than 0.05 but when i try to check which of them are different i got a P-paired value greather than 0.05 (which means: no significative).
I tried to change test to ANOSIM but i got the same results.

Does anybody know, what could be due this?, how should that results be interpreted? or even, how could i solve it?

aitchisonps.qzv (826.0 KB)

Parwise Permanova
pairwiseAdonis::pairwise.adonis(dis.mat, factors = metadata$insect,
perm = 999)

Hello again Kevin,

I noticed in the results for pairwiseAdonis::pairwise.adonis(), all your Df values (degrees of freedom) are 1. *

How many samples do you have in total?
How many replicates for each insect do you have?

P.S. *Those R2 values are super high... which is suspicious. :face_with_monocle:

I have 3 samples per group, 46 samples in total.
As i have understood, DF from pairwiseAdonis::pairwise.adonis(), DF is 1 because is comparing by pairs of groups and DF is n-1 groups

1 Like

Ah, I understand.

Having n=3 for each group may not be powerful enough to get small p-values. But, let's try increasing the permutations to perm = 9999 to make sure it's not a randomization issue.