 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: My last word on strange world
At 10:36 AM 1/10/2008, Nick Cox wrote:
On (a), list members can test themselves. Suppose x is missing on some
observations. What do you consider appropriate Stata behaviour for
. list x if x > 42
Do not list the case if x is missing. Let us say x is age. If I
give a command like this, I probably want those older than 42, not
people whose age is missing.
. regress z y if x > 42
Do not include the case if x is missing. If i do this, I want to
restrict my sample by age, and I probably don't want to include those
whose age is unknown.
If by some chance I do want the missings included, I can always
change the ifs to
if x > 42 | missing(x)
I think I am much more likely to want the missings excluded rather
than included, hence I would make you go to the extra work when you
want them included, rather than make you work harder when you want
them excluded.
Intuitively, most people don't think of missing as being a really
really large number. For that matter, missing could just as easily
have been coded as a really really small number. To most people,
missing is just missing, and the fact that missing has to be assigned
a number is just a necessary evil. I assume SPSS assigns system
missing a number too, but I have no idea whether it is a big number
or a small number, nor does their seem to be much reason I would need
to know (because missing is one of the very few areas where SPSS is
more logical than Stata!)
That is how I would have Stata behave if I was designing it from
scratch. Changing its behavior after it has done something a
particular way for 20 years may be another matter.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/