I am analyzing the microbial community composition of different soil samples and I am wondering if it makes sense to run adonis (a permanova test) with only 18 samples? Permanova are worth it when used with many samples because it's based on permutations. Is there another method that would fit better the nature of my dataset? ANOSIM for example? Or can I use a permanova test no matter the number of samples?
Thanks for your help!
Hi Malcolm,
Thanks for your answer :) I am still a bit confused about the results I get with the permanova. So the % of variance explained by depth and comp are both similar (around 30%), but one is significant, while the other is not... How am I suppose to understand that? I was wondering if it's because the number of samples within comp group is too low for the test to be robust (n=6), whereas is not the case for depth (n=9). Would it be the reason why I have a similar percentage of variance explained but one is significant and not the other? Thanks.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size