Hello Everyone, I have been trying to do some statistical analysis of soil data from 4 different locations. The experimental design is split-plot design and has three(3) factors; tillage (2 levels)d as main plot factor, and crop residue (2 levels)d and nitrogen dose (3 levels) as sub plot factor randomized within the main plots. The whole plot is replicated 4 times. Now I have some soil parameters , e.g., soil organic carbon (SOC) and this SOC is determined for 2 different soil depths (0-20cm and 20-40cm). The dataframe that I made for this approach has columns or headers as : replication, depth, tillage, CR, N, SOC. SO far I understand that I need to use proc mixed model and I used the following codes to to do it: proc mixed;
class tillage CR N replication ;
model SOC= tillage |CR |N ;
random replication replication (tillage);
lsmeans tillage|CR |N/ alpha=0.01 diff adjust=simulate ;
run; But I am struggling to fit the co variate "depth(2 levels)" in the model as well as to make a multi group comparison with letters. I would be grateful if anyone could take a time out of his/her busy schedule and suggest me the best solution. Thanks in advance.
... View more