[R] matrix for diversity functions?

Sarah Goslee sarah.goslee at gmail.com
Tue Nov 18 18:32:50 CET 2008


If all you have is species richness at each site, then you can't
calculate diversity at all. You need to have the raw species abundance
data. The data you provided is not a "community data matrix".

The diversity function is doing exactly what it is supposed to: calculating
the Shannon diversity for each row of the matrix. It assumes that sites
are rows and columns are species.

If the data frame contains all of the data you have available, then what
you want to do is choose the appropriate test to compare species
richnesses between paired disturbed and undisturbed habitats, and
that problem has nothing to do with calculating diversities.

Sarah

On Tue, Nov 18, 2008 at 11:27 AM, Sharma, Manju
<m.sharma05 at imperial.ac.uk> wrote:
> Hi,
>
> I have a small simple data frame (attached) -  to compare diversity of
> insects encountered in disturbed and unditurbed site.  What i have is
> the count of insects -  the total number of times they were encountered
> over 30 monitoring slots.
>
> Can someone please check for me to make sure how the 'community data
> matrix' for the diversity function needs to be oriented so that i'm
> comparing the right sets.
>
> I know that community data matrices mustn't carry characters that aren't
> numbers. I replaced my sps. names sp1, sp2, sp3, etc., with just 1, 2,
> 3.... but not sure what I can replace 'dist' and 'undist' with!
>
> I try this to start with --
> insects.div<-diversity(insects, index="shannon")
>> insects.div
> [1] 0.7242788 0.7485246 0.7298712 0.9012085 1.0366280 0.9470281
> 1.0466542
> [8] 1.0133127 0.6450332
>
> & thats not what i want. Any advice on the matrix format or commands
> would be big help!
>
> Cheers,
>
>
>
>
> Manju V. Sharma
> Gardenwood East 3.1, Division of Biology,
> Imperial College London, Silwood Park Campus
> Ascot SL5 7PY
> UK
> Tel (O): 0044 207 5942360
>      (R): 0044 207 8520808
>

-- 
Sarah Goslee
http://www.functionaldiversity.org



More information about the R-help mailing list