RLQ analysis ("Q & R non convienient weight" )

From: Byungok Ahn (byungok.ahn@uni-essen.de)
Date: Fri Aug 17 2001 - 02:04:31 MEST


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