Dear Byungok Ahn
Huge number of variables and not so huge number of samples ! Anyway, in the
version offered by ADE-4, the central table (abundance table) must be
performed by correspondence analysis as you did. This implies that both the
rows (samples) and the columns (species) are provided with COA weights. So
your point 2. is OK (MCA of environment with ".fcpl" weights for
rows-samples). By contrast your point 3. should use values of ".fcpc" as row
weights for the analysis of the species trait table (FCA with ".fcpc"
weights for rows-species).
Best wishes,
Sylvain Dolédec
----------------------------------------
UCB Lyon 1
CNRS - Ecologie des Hydrosystèmes Fluviaux
Bât 401C
43 Bd du 11 novembre 1918
69622 Villeurbanne
phone: +33 (0) 472431363
fax: +33 (0) 472431141
----------------------------------------
http://ecolstat.univ-lyon1.fr/equipe/accueil/
http://pbil.univ-lyon1.fr/ADE-4/ADE-4.html
Dear list,
I am currently trying to match abundance data (71 species of Trichoptera)
and species traits (72 traits; 20 categories) to environmental data (53
variables ; 16 categories) based on 25 sites using RLQ analysis. Following
the procedures in ADE-4/Fiche thématique 4.4 (Triplet d´analyses des
correspondances) three tables were separately processed as follows:
1. An abundance table (25 rows, 71 columns) was processed by COA.
2. A table with environmental variables (25 rows, 16 columns) was processed
by MCA (option: Row weighting file .fcpl).
3. A species traits table (71 rows, 72 columns) was processed by FCA without
any row weghting.
4. matching two statistical triplets was successfully done (.cmta & .fcta,
.flta & .fcta (after ´transpose´)).
But I always have problems with the message "Q & L: non convenient weight"
when using the module ´Prepare RLQ analysis´ by:
R: .cmta
L: .fcta
Q: .flta
Can anyone offer an explanation of how to solve this problem?
Regards,
Byungok Ahn
This archive was generated by hypermail 2b30 : Thu Jul 04 2002 - 09:39:13 MEST