Hi,
i need some help with component of variance( VARCOMP).
I analysed 16 varieties of plants.
I have a 4*4 lattice experimental design of plants with 4
replicates.Each of the 16 (4*4) plots contains 12 plants.
So i have a total of 768 plants. I want to do analysis of
components of variance but as they are so many plants
i could take only 1 measurement (for each character), in each plant.
So when i run varcomp, almost every variance appear in the error, and
the variance " among plants within plots" is very small.
And i don't have the possibility of take the variety-replicate
interaction. But the truth is that the error is absorbing the most
of the variance of among plants within plots. Beside, the statistic
error is not a real error, is the variance between the plants,
if some error of measurement can exist i think that it can be
neutralized because of the number of plants( is this correct?).
Finally, i don't know if i can to:
-assume the error variance don't exist,
-consider the labeled error component of variance as the among plant
within the plot, and the labeled among plant, as the interaction?
I hope i have been clear, in order to receive SOME HEEEELP!!!!
Sent via Deja.com http://www.deja.com/
Before you buy.
===========================================================================
This list is open to everyone. Occasionally, people lacking respect
for other members of the list send messages that are inappropriate
or unrelated to the list's discussion topics. Please just delete the
offensive email.
For information concerning the list, please see the following web page:
http://jse.stat.ncsu.edu/
===========================================================================
