Greetings all
I wonder if anyone has the time to help me with this problem.
I have been asked to comment on the model building and specification process for the following experiment.
It is a drug potentcy assay.
There are eight treatments, which have the following structure...4 drugs at two dose levels (low and high). These are the factors of real interest. Each treatment is applied to an animal. There are therefore eight sites in which to apply each treatment to each animal. These sites have a structure. The eight sites consist of 4 positions on the right hand side and four on the left directly opposite. It is believed that there should be no difference as regards these eight positions.
There are eight animals in a group.
The paper I have been asked to criticise simply includes the main effect of drug, dose, animal and site as fixed effects.
If I was to write out the model, I would start by regarding the animals as random factors. The treatment structure is 2x4, however how should I best incorporate the site structure.....have a Side and Position variable?
Any advaice would be appreciated
Kieran
|