No, sorry, but I think you are missing
two points here.
Easier one first: the expression "passing the buck"
was, as you probably realise, my
attempt to leaven an otherwise dull
posting. More seriously, the division
of labour whereby -tabstat- organises
and displays what is produced by
-summarize- internally is excellent
from all points of view. There is
no reason at all to change it. If
-tabstat- doesn't do something,
and a different program is needed,
that's a different matter. The
rationale of -tabstat- is to act
as a wrapper for, and thus to be
totally consistent with, what
-summarize- calculates. To get
-tabstat- to do something different,
or extra, would in my view be bad
program design, modulo a few arguable
details (e.g. cv is just a minor
extension of what -summarize- does).
On the original question, I maintain
that Stata is working consistently
with the documentation, but I do
agree that the latter could be
tweaked to make it clearer. A key
principle is that whenever a sum
of weights is being used, what that
sum of weights is should be accessible
in some way, and Stata satisfies this
criterion.
Nick
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Richard
> Williams
>
> At 03:34 PM 8/3/2004 +0100, Nick Cox wrote:
> >This isn't a -tabstat- issue as such. -tabstat-
> >just passes the buck to -summarize-, which
> >behaves in the same way.
> >
> >The issue is delicate, but hinges, I surmise,
> >on this distinction. When -summarize- (e.g.) _uses_
> >the sum of the weights, it rescales first.
> >As quoted, [U] 14.1.6 includes the expression
> >"when it uses them", which thus appears not
> >ornamental, but crucial.
> >
> >When -summarize- _displays_ the sum of the
> >weights, it displays the unscaled sum.
>
> Good point, although it still leaves open the question of why
> it behaves
> this way. A long-standing oversight or bug? Sound
> statistical practice?
> (If so, it wouldn't be the first time Stata has surprised me
> by not doing
> what I expected, and then having it turn out that Stata was
> right.) Should
> -tabstat- be modified so it doesn't just pass the buck to -summarize-?
*
* 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/
