independent ttest with multiply imptued data

[halladje]

What would be the code to run and pool a independent t-test after multiple imputation?

The code for an independent t-test already has a "by" specification - so how do you compute the t-test separately for each imputation? Is there a way to then pool results? Or is it just the average of the t-tests and SE's given there's no covariance structure to take into consideration since it is a univariate analysis? 

Thanks in advance, 

Jillian 

[Reeza]

You can have more than one variable in your BY statement and I think you use MIANALYZE to combine the results.

[halladje]

hmm I also have a categorical variable that I am trying to compare just particular categories for each t-test. For example, my chi-square code is as follows:

proc freq data=lpa;
by _imputation_;
tables other*class_mem/chisq nocol nopercent;
where class_mem in ( 2 4);
output out=ChiSqData n nmiss pchi lrchi;
run;

Can you do this in a t-test as well? How do you specify both the categories to be compared and the imputation statement?

Thanks so much, 

Jillian