Prev: National characters and export to SPSS
Next: MDB files compare
From: breezezhaoeei on 23 Mar 2010 14:25
Hi SAS experts,

I have a three level model like this:

proc mixed data=test;
class schoolid teacherid ;
model post = istx prec / solution outp=r outpm=mr covb influence
ddfm=bw ;
random intercept / sub=teacherid(schoolid) solution G;
random intercept / sub=schoolid solution G;
run;

I know outp r can give me student level residual and outpm can give me
marginal residual (student+all the random level). But is there a ways
to get teacher level residual and school level residual separately so
that I can draw plot to check whether they meet the equal variance
assumption of HLM model? Or is there any other way to check the equal
variance assumption?

Thank you very much,

Qingfeng
 | 
Pages: 1
Prev: National characters and export to SPSS
Next: MDB files compare