Hello everyone,
I was wondering if anyone could help me with a question about using
-mi- to do multiple imputation for multiple ordinal items for a scale.
I have read the previous discussion about this issue from the
following link(http://www.stata.com/statalist/archive/2009-07/msg01146.html)
I have 20 items taking the values {1,2,3,4,5} and I would like to
impute the items first and then compute the scale score by calculating
the average of the items after MI.
****My first question:
Can I impute items one by one using --mi impute ologit-- ?
-----------
forv i=1/20 {
mi impute ologit item`i' : x1-x10, add(5)
}
-----------
***Second question:
Alternatively, is it OK to use --mi impute mvn item1-10 x1-x10--? I
know --mvn--should be used when the items are continuous normal
distributed variables. But, if using it for the ordinal variables, is
it still OK or is it wrong?
Thanks for your help!
Sincerely,
Mandy
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
