Dear Mark,
     Gower does seem to the correct dissimilarity metric based upon your 
variables.  This will return a dissimilarity matrix that I believe you can use 
to calculate the functional diversity indices, by base code and likely by FD 
package as well.  Using this matrix will ensure that you are actually 
calculating the indices based upon the functional differences of the actual 
replicates, and will give you a continuous dissimilarity measure.  However, I 
am unsure why you want to insert PCoA into this before using FD?  This will 
output a reduced dimensionality object with replicate scores on the PCo 
axes---is this what you are after?  In any case, you will not get a number to 
replace the classes of your categorical data---you will get a dissimilarity 
measure between all pairs of replicates.
     Hope this helps,
Jeff Holland

-----Original Message-----
From: R-sig-ecology <> On Behalf Of Mark 
Sent: Friday, August 10, 2018 7:07 AM
Subject: [R-sig-eco] Functional Diversity help!

Dear r-sig-ecology,
Please forgive me if this question has already been answered, I have perused 
the web and due to me being fairly new to R, I have had trouble interpreting 
the answers.

I wish to calculate the functional diversity indices (Functional Richness, 
divergence, evenness, Rao-quadratic).

My dataset consists of 14 species, 25 sites, and 8 categorical traits.

I have been using the FD package but from what I have read I need to convert 
the categorical data using the gowdist function and then run pcoa to generate 
continuous data?

However, when I run the gowdist and pcoa I thought I'd get a number to replace 
the categorical data so I could then run the dbFD function in the FD package to 
calculate the FD indices, but this isn't happening, and I am not sure how to 
take the output from the pcoa and run in dbFD.

Any help would be greatly appreciated (apologies for the beginner question) 

        [[alternative HTML version deleted]]

R-sig-ecology mailing list

R-sig-ecology mailing list

Reply via email to