Combined ANOVA

#1
Hi everybody,
My problem is as follows:

I designed three plots of 250 m2
On each plots there is three plant species with 3 replicates by species (3*3= 9 plants). The species are not randomized trough the plot. They are planted at the same place through the plots

On each plot a different fertilization was applied
Plot 1: Raw Pig slurry
Plot 2: Centrifugal pig slurry
Plot 3: Tap water (Control)

The treatments are:
The fertilization
The species (3 levels)

I measured the biomass for each species replicate.

I want to determine the effect of Pig slurry on the biomass yield compare with control.
My problem is that i have just one plot for one treatment so there is no replication.

I have read that i can proceed to an ANOVA for each plot separately as different experiments.
Then applying a combined anova.
But i don't know if this the right way to do and i don't know how to proceed to a combined anova.
 

bugman

Super Moderator
#2
It sounds more like a Latin Squares ANOVA design to me. Your main problem is that you dont have any replcaition of the plots, otherwise you could just add a block effect into the model to account for block to block variation.