[R] Vegan package treatment of zeros

Gavin Simpson gavin.simpson at ucl.ac.uk
Wed May 15 16:49:48 CEST 2013


On Wed, 2013-05-15 at 09:04 -0500, Michael Rentz wrote:
> Hello:  I have a good size database of plant cover measurements I was
> running a Shannon Diversity on.  I have every cell as a default of "0" if
> it is not present.  The tutorials I have looked at all had blanks in cases
> of absence.  I just wanted some kind of verification that VEGAN will
> recognize 0s as absent....
> 
> I just wanted to make sure I was not inadvertently setting myself up for
> problems.

Blanks would be interpreted as `NA` in R when you read the data in.
Vegan would respect that and most likely fail because of the missing
data as not all methods are appropriate when there are `NA` values.

Hence a common first step is to replace `NA` with 0. As you have already
have this indicated in your file external to R then you won;t need to do
this extra step *in* R.

How zeros are then treated will depend largely on the method employed
and/or the dissimilarity coefficient used, but that is not an R
question. A recent posting to the R-Sig-Ecology list by Brian Cade
presents some interesting observations on the double zero issue:

https://stat.ethz.ch/pipermail/r-sig-ecology/attachments/20130419/114d1e5f/attachment.pl

HTH

G

-- 
Gavin Simpson, PhD                          [t] +1 306 337 8863
Adjunct Professor, Department of Biology    [f] +1 306 337 2410
Institute of Environmental Change & Society [e] gavin.simpson at uregina.ca
523 Research and Innovation Centre          [tw] @ucfagls
University of Regina
Regina, SK S4S 0A2, Canada



More information about the R-help mailing list