Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: Table with medians of a variable with many categorical variables

From   "Nick Cox" <>
To   <>
Subject   RE: st: RE: Table with medians of a variable with many categorical variables
Date   Thu, 15 Mar 2007 22:45:37 -0000

My comment was one old-timer's advice, and manifestly not an
echo of list policy. 

My advice, which I repeat, is that calling your question
"quick" is not helpful and is indeed likely to create the wrong
impression. After all, no one says things like "I have
a question one sentence long" or "... one paragraph long". 
We can see how long a short question is! 

Naturally, I am sure that often the intent is totally innocent 
and that the poster is just "warming up", as it were. Many
people don't seem able or willing to ask a question
without a preamble or a postamble, and no harm in that 
as such. 

But my reaction, and perhaps it is shared by some others, is why 
are you telling me this? Are you soft-selling or grovelling? 
You needn't do that. A good question can stand up for itself.

Anyway, view this statistically, or decision theoretically. 
The distribution has two categories. There may be some 
like me who find it irritating, and some like Jeph, who
aren't irritated. The best strategy is still not to
do it. 

"Quick" questions are not as bad as "easy" questions, 


Jeph Herrin
> I don't think "quick question" deserves quite this
> much discredit. In particular:
> 1. I would agree with you that just as simple question
>     may have unsimple answers, quick questions may have
>     unquick answers; but I don't see why one would
>     therefore refrain from framing either as what
>     they are.
> 2. "quick question" doesn't indicate to me either
>     triviality (see 1 above) or apology.
> Nick Cox wrote:
> > Never say "I have a quick question". 
> > 
> > 1. If the question's quick, the answer may not be!
> > 
> > 2. If you think your question is trivial, then don't ask 
> > it. If you think your question is interesting, it needs no
> > apology. 

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index