Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: st: RE: automatic disclosure in tables
From
Nick Cox <[email protected]>
To
"'[email protected]'" <[email protected]>
Subject
RE: st: RE: automatic disclosure in tables
Date
Wed, 16 Nov 2011 18:06:05 +0000
You could include zeros too:
sysuse auto, clear
gen edu=round(mpg,5)
ren rep78 sex
gen one=!mi(sex)
contract sex edu, zero
la de c 0 "\" 1 "\" 2 "\"
la val _freq c
tabdisp sex edu, c(_freq)
Nick
[email protected]
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Nick Winter
Sent: 16 November 2011 16:42
To: [email protected]
Subject: Re: st: RE: automatic disclosure in tables
A solution that does not show 0 and >0 the same, but does allow
replacement with a character:
sysuse auto, clear
gen edu=round(mpg,5)
ren rep78 sex
gen one=!mi(sex)
collapse (count) count=one , by(sex edu)
la de c 1 "\" 2 "\"
la val c c
tabdisp sex edu, c(count)
Nick Winter
On 11/16/2011 11:09 AM, Nick Cox wrote:
> This is a good idea.
>
> It shows cells with zeros and ones identically, which may well be a feature for Percy.
> Similarly the threshold is easy to tune.
>
> What I meant by not trivial was that inserting other characters in the table would be difficult, but Austin is clearly right that blanking out cells is more straightforward.
>
> Nick
> [email protected]
>
> Austin Nichols
>
> Percy Scheller--
>
> Is the kind of thing that is wanted?
>
> sysuse auto, clear
> gen edu=round(mpg,5)
> ren rep78 sex
> table sex edu
> egen ct=count(sex), by(sex edu)
> table sex edu if ct>1
>
>
>
> On Wed, Nov 16, 2011 at 5:20 AM, Nick Cox<[email protected]> wrote:
>> I don't recall any such command. It doesn't sound at all trivial to program. It seems easier to edit output, or even to write programs to edit output.
>>
>> In this particular case, suppression seems counter-productive!
>>
>> Nick
>> [email protected]
>>
>> percy
>>
>> I once discovered an ado, which helps to automatically disclosure/suppress
>> cell content if it makes a reidentification of single persons possible.
>> Unfortunatelly this is years ago, I can't remember it or find it in my
>> ado-files. Has anybody a hint for me?!
>>
>> Thanks in advance
>>
>> Percy Scheller
>>
>> Example (see low educated male, infact you might need to suppress a second
>> information, too):
>>
>>
>> . table sex edu
>> ----------------------------------
>> | Education
>> Sex | low medium high
>> ----------+-----------------------
>> male | 1 6 4
>> female | 8 53 26
>> ----------------------------------
>>
>> . table sex edu
>> ----------------------------------
>> | Education
>> Sex | low medium high
>> ----------+-----------------------
>> male | / 6 4
>> female | 8 53 26
>> ----------------------------------
>
*
* 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/