yes, always impute using all the variables. Amelia II has some facilities
to deal with these types of variables.
Gary
On Mon, 5 Nov 2007, ian.a.robinson(a)studentmail.newcastle.edu.au wrote:
Hi,
I am performing MI on a dataset in R and then performing my analysis in Mplus. I have a
nominal variable in my data set (with say three levels) but to perform the analysis in
Mplus, I have to tranform the varible into two ordinal variables: one for level 2 and one
for 3 which are both compared with the reference level. My question is what are peolpe
out there doing when dealing with a similar situation? Should I be performing the MI on
the nominal or ordinal variables?
Regards,
Ian
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia