Nick already gave the answer I would have given:
Your problem is clearer this way, but I can't see an alternative to
working with several variables simultaneously, or indeed why you think
that's so bad.
In other words: If you are doing
g generatedvar=""
levelsof wais, local(lev)
foreach l of lev {
replace generatedvar="`l' "+generatedvar
* plus whatever qualifiers etc.
}
ren generatedvar g
split g
g i=_n
reshape long g, i(i) j(j)
you need to instead
levelsof wais, local(lev)
loc j=1
foreach l of lev {
g g`j'="`l'"
* plus whatever qualifiers etc.
loc j=`j'+1
}
g i=_n
reshape long g, i(i) j(j)
i.e. make separate variables in the first place. You can also avoid
the -reshape- using a different approach. It's not clear why you want
to do any of this, though, so maybe there is a better global approach.
On Tue, Sep 8, 2009 at 8:15 PM, David Torres<[email protected]> wrote:
> I ran this one once and it looks like it may be useful. It's getting late
> here, though, so I'll respond more fully tomorrow if I should have any other
> questions.
>
> Thanks for -logout-.
>
> Give this guy a raise; it's people like him who make things easier for
> technological inepts like me.
>
> ;-D
>
> David Torres
*
* 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/
