[R] chaining & consensus in cluster analyses

[Penner, Johannes]

Dear R-helpers,

conducting different community ecology analyses my main aim is to find
groupings in the data and geographical borders between communities and
to prove them statistically.

So after conducting a global test (Mantel) I am running a NMS and
cluster analyses. These are followed by other tests. However, I have two
questions concerning the cluster analyses which I was not able to solve
so far:

1) I would like to calculate "% chaining" for specific clusters. How can
I do that in R?

2) I would like to calculate several clusters with different methods and
then merge them in a "consensus" cluster, similar to a consensus tree
used in genetic analyses. Is this possible and if yes how?

Any help or hint is greatly appreciated!

with kind regards
Johannes

--
Project Coordinator BIOTA West Amphibians

Museum of Natural History
Dep. of Research (Herpetology)
Invalidenstrasse 43
D-10115 Berlin
Tel: +49 (0)30 2093 8708
Fax: +49 (0)30 2093 8565

http://www.biota-africa.org
http://community-ecology.biozentrum.uni-wuerzburg.de