The help file for -mrdum- (a user-written
program by Lee Sieswerda) explains its
syntax. The syntax for its -stub()- option
is that you must supply an argument. Just
specifying -stub- is illegal (and useless
anyway).
I am not sure that you can always
obtain dummies for multiple responses
which sum to 1. It depends on your data
structure. If I answer "yes" to
"use Stata" and "yes" to "use R" _and_
those are separate variables, then
dummies for me will add to 2. This
is the whole point of the multiple
responses problem, is it not?
Nick
[email protected]
peter harper
>
> Nick
> Thanks for your reply. I am new at usimg -mrtab- and
> -mrdum- . May I ask for example, -stub(q4) precisely
> means. I did look at their help for these things, but
> it was not clear to me. Perhaps I have overlooked
> something.
> I wanted to find out how to generate dummy variables
> of multiple responses which will sum to 1. then obtain
> means and standard deviations.
> Thanks
>
> --- Nick Cox <[email protected]> wrote:
> > Both the error message and the help for -mrdum- tell
> > you
> > what is required, an argument for the -stub()-
> > option,
> > e.g. -stub(q4)- not -stub-.
> >
> > Nick
> > [email protected]
> >
> > peter harper
> >
> > > dummy variables for multiple responses to a survey
> > > question. I have tried -mrtab-
> > > I have tried
> > > .mrdum x1 x2 x3, stub
> > > option stub() required
> > >
> > > This is the message I got. Can anyone please tell
> > me
> > > what is precisely required here.
> >
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
