I am trying to determine how fish predator assemblage (species composition - diversity, abundance - MaxN and individual size (S,M,L) change over a distances gradient and sampling period (day, dusk, night, dawn).
My independent variables are time (fixed with four levels), tide (fixed with four levels), distance, and depth.
Would it be better to use a Permanova or a CCA?
Thank you
Dear Pat,
For your type of data, both analyses could be used, but CCA would be more appropriate. It's commonly used in environmental studies to assess relationships between species assemblage and measured environmental variables. In CCA, you can also use variables which are not continuous - dummy variables related to fixed time and tide levels as in your case.
Best regards,
Aleksandar
I believe that RDA are more appropriate than CCA in this instance.
I don't think that Permanova will work with your continuous variables (ex, depth and distance).
This being said, have you considered the used of mixed models where you could control for both fixed and random effects of you sampling design?
Hope it helps!
- Badges
- Science topic
- Similar topics
- Agriculture
- Horticulture
- Seeds