Change in total Vp when adding variance components

Dear all,
we have been trying to fit an animal model for some habitat preference (% of one habitat type within an buffer zone for each individual). We fitted 2 models, one with and one without ide(ID), and the total amount of variance differed between models.

More specifically we got for the model

Forest ~mu !r ID
 Source Model terms Gamma Component Comp/SE % C
 id 729 729 3.39665 254.001 8.95 0 P
 Variance 894 893 1.00000 74.7798 17.54 0 P

Va = 254.0
Vp = 328.8

And for
Forest ~mu !r ID ide(ID)
Source Model terms Gamma Component Comp/SE % C
 id 729 729 1.51355 109.761 4.04 0 P
 ide(id) 729 729 0.925964 67.1501 3.96 0 P
 Variance 894 893 1.00000 72.5191 17.73 0 P
Va = 109.8
Va + Vpe = 177
Vp = 249.5

So total phenotypic variance is greatly reduced.

We would be very grateful if anyone could help us to make sense out of this!

All the best

Celine

-------------------- m2f --------------------

Sent using Mail2Forum (http://www.mail2forum.com).

Read this topic online here:
http://www.vsni.co.uk/forum/viewtopic.php?p=405#405

-------------------- m2f --------------------
Received on Tue Feb 23 2009 - 18:47:08 EST