I have a morphological dataset with continuous and discrete characters, which I want to perform ordination analyses (PCA, PCo, LDA) on the whole dataset, as it was made for this study:
Article The ecological diversification and evolution of Teleosauroid...
I have already performed a z-transformation to standardize the continuous characters. However, I don't completely understand how the data are transformed into a Gower distance matrix in order to include both continuous and disrete characters in the same analyses.
It is not clear to me how much of the data set must be transformed (only the categorical data? Only the discrete characters? Both types?), and if the resulting matrix should be rearranged (e.g., deleting the diagonal of self comparisons typical of distance matrices -see image) after the transformation.
If you have any information on the details of these procedures, I would be very grateful.
Many thanks in advance,
Jorge.
- Badges
- Science topic
- Similar topics
- Mathematics