[R-sig-eco] vegan: goodness() and pRDA

[Jari Oksanen]

This seems to be a bug in vegan. In fact, there are both bugs and bad design decisions.

I have pushed a pull request to https://github.com/vegandevs/vegan which fixes the issues in this message plus a couple of other quirks. I'll test that fix a bit, and if everything looks OK, it will be merged and included in the next vegan release.

Thanks for reporting this.

Best wishes, Jari Oksanen
________________________________________
From: R-sig-ecology <r-sig-ecology-bounces at r-project.org> on behalf of Christoph Eberhard Freiherr von Redwitz <christoph.redwitz at uni-rostock.de>
Sent: 05 March 2015 15:29
To: R-sig-ecology at r-project.org
Subject: [R-sig-eco] vegan: goodness() and pRDA

Hi all!
I am using rda() in vegan and goodness() as a diagnostic tool. I want to know how good the species fit to the first two ordination axes.
But I got problems combining partial RDA and goodness(). The species sum of CA and CCA don't add up to 1 (as it does in RDA without Condition term). Furthermore, I get warnings when only one axis is left over in the result of goodness.
Here an example:


#########----------------------------------------------#############
data(dune)
data(dune.env)

mod1 <- rda(dune~A1 + Moisture + Management, data=dune.env)
mod2  <- rda(dune~ Manure + Condition(A1 + Moisture + Management), data=dune.env)
mod3 <- rda(dune~A1 + Condition(Manure + Moisture + Management), data=dune.env)

goodness(mod1, mod= "CA ",sum=T) + goodness(mod1, mod= "CCA ",sum=T)#result as expected
goodness(mod2, mod= "CA ",sum=T) + goodness(mod2, mod= "CCA ",sum=T)#only 4 RDA axes left over, they don't end up with 1

#here comes the warning:
goodness(mod3,mod= "CA ",sum=T)#works
goodness(mod3,mod= "CCA",sum=T) #only 1 RDA axis left. Warning messages. The "sum" option seems to be affected.
#########----------------------------------------------#############

Two questions:
Is the warning because of wrong usage of the method? Can I use goodness with partial RDA?
1 - goodness(mod3,mod= "CA ",sum=T) should do it in this case?

I hope the mail is clear and not an old story.

Greetings!
Christoph


        [[alternative HTML version deleted]]

_______________________________________________
R-sig-ecology mailing list
R-sig-ecology at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology