Hi all,
Sorry for bothering you. I have found similar question has been
discussed previously.
I should have done more research in checking the archive in advance.
Best wishes,
Mandy
On 12/2/09, Mandy fu <[email protected]> wrote:
> Hi all,
>
> I would like to impute for the missing items for a scale measuring
> stress. The stress scale is calculated by adding 18 items all
> together, where each item is about one respect about stress. An
> example of one of the items is like this:
>
> item1 : question: How often did you feel no interest in things in last 30 days?
> 0: never
> 1: occasionally
> 2: often
> 3: very often
>
> In the data set, many respondents have a couple of missing items out
> of the 18 items. So, I am thinking to use multiple imputation to
> impute the missing items in order to get the scale measurement.
>
> My question is, using multiple imputation (command -ice- ) to impute
> the missing items, is there anything different than , say, any other
> 18 usual variables? Or is there anything I need to be especially
> cautious when imputing these items? Thanks in advance!
*
* 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/
