Hello,
I am working on a Partial Least Squares approach between two matrixes. One contains community data (as abundance of the single taxa in different locations) and the other matrix contains environmental variables for the same locations. Regardless of the reasons that led me to use a PLS approach rather than other ordination methods (e.g. NMDS, RDA or CCA) I was wondering whether a Hellinger transformation of the community data was meaningful with a PLS approach or not. I have tried both with and without the transformation, the results do not change dramatically although the analysis with the transformation is marginally better. Anybody with a previous experience or opinion on this? Thanks.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size
More Michele Innangi's questions See All
Similar questions and discussions