Dear members,
I need to carry out a repeated measures analysis where there are
unequal group sizes, as per Winer 'Statistical Principles in
experimental design', example 7.38. (Briefly, there are an unequal
number of subjects for each level of factor A, then each subject has a
result for each level of a second factor B)
Winer distinguishes between 'unweighted means' and 'least squares
analysis'. SAS GLM and MIXED give me the latter....is there any
direct way of getting the former, other than writing my own program?
(I'm not a SAS expert...so I'm sorry if the answer to this is trivial!)
Can anyone give me some hand waving advice?
Please reply to me, not to the list,
Many thanks,
Linda Hunt
[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|