05-18-2017 04:04 PM
I conducted an experiment containing three treatments over a two year period (the trial on the second year was a replicate of the first year: same treatments but different animals). I was looking at the effect of proper animal handling at early age on growth. In all these two years the animals after birth were randomly allocated to treatments (treatment 1- animals were gently picked at random and stroked while treatment 2- animals just experienced human presence in their enclousure and treatment 3- was control human presence was limited feed supply. Now since treatment was supplied on a group of animal not individual animal due to the fact that not all animals (N=60 treatment 1 first year and 70 second year) per day were picked/stroked for 2 months a split plot might be a good design to analyse the weight after 2 months with with year being your whole plot factor and treatment within year as your subplot factor.
My questions are:
1. How do i define the covariate (weight of animal after birth) in the glimmix model?
2. Can you check my random variiable statements i doubt they are alright, i am not that familiar with GLIMMIX.
3. Is the following commands appropriate to run my analysis for meaningful results?
PROC GLIMMIX DATA= growth;
CLASS YEAR CAMP GENOTYPE GENDER TREATMENT;
MODEL WEIGHT= YEAR|GENOTYPE|GENDER|TREATMENT/ SOLUTION DDFM= RES;
Any help will be apprecited.
05-25-2017 11:02 PM
With an appreciative nod to @Reeza who linked to
in another posting, you will need to provide more information to get a correct answer to the right question. Please include definitions of all variables in the model and a more complete description of the experimental protocol.