I can't speak for STATA. Curiously enough,
the same question arises in Stata.
The key question is how you expect Stata to
know that "Green" was also a possible category.
I guess that there are some languages in which
the fact that "Green" was
defined as a possible value by way of a specification
of possible values would be enough.
Stata doesn't work that way, however. Even if
you define a set of value labels, Stata doesn't
automatically tabulate those which are not
represented in the data, or the subset of the
data you specify. If these data are string,
then why should Stata know that "Green" was
a possible answer any more than (say) "Aardvark"?
So you need a work-around in which you yourself spell
out the possible values. The only ones I know
are -tabcount- from SSC and the earlier -tabcond-
from SSC. Note that had you used
. findit zeros
or
. findit zero
you would have found out about -tabcount- for
yourself (together, naturally, with lots of
other stuff).
. findit tabulate zero
would have zeroed in further on what you need.
Here are a few simple examples of how -tabcount-
works:
. tabcount foreign, v(0 1 2 3 4)
----------------------
Car type | Freq.
----------+-----------
Domestic | 52
Foreign | 22
2 |
3 |
4 |
----------------------
. tabcount foreign, v(0 1 2 3 4) zero
----------------------
Car type | Freq.
----------+-----------
Domestic | 52
Foreign | 22
2 | 0
3 | 0
4 | 0
----------------------
. decode foreign, gen(FOREIGN)
. tabcount FOREIGN , v("Domestic" "Foreign" "Martian" "Venusian") zero
----------------------
Car type | Freq.
----------+-----------
Domestic | 52
Foreign | 22
Martian | 0
Venusian | 0
----------------------
Missings are not shown by default. I guess the reason
you have a puzzle about missing is that you think string
"." means missing. With strings in Stata only the empty string
"" means missing. If you have strings "." that you
regard as missing you should just exclude those if desired
by
... if COLOR != "."
or replace them with empty strings.
There was a more discursive account in the Stata Journal
SJ-3-4 pr0011 . . . . . . . . Speaking Stata: Problems with tables, Part II
Q4/03 SJ 3(4):420--439
reviews three user-written commands (tabcount, makematrix,
and groups) as different approaches to tabulation problems
Nick
[email protected]
buddyb
> I'm running STATA 8.
>
> If I had a multiple choice question, like, what is
> your favorite color with red, yellow, black, green as
> the given choices,and asked 7 people and got the
> following results....
>
> Red 1
> Yellow 2
> Black 3
> Green 0
> . (missing)
>
> When I do a Table COLOR (this is my vname),
> I get:
>
> Red 1
> Yellow 2
> Black 3
>
> The Green does not show up and I would like it too. I
> want to repress the missing. How do I do this?
*
* 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/
