Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# Re: st: prtest and survey weights

 From Steve Samuels To statalist@hsphsun2.harvard.edu Subject Re: st: prtest and survey weights Date Sat, 13 Mar 2010 09:49:12 -0500

I should have clarified that the first example tests the hypothesis
that the row and  column marginal proportions are equal. (These are
the "correlated proportions" I referred to).  This is the same
hypothesis addressed by McNemar's test in non-survey data.

On Sat, Mar 13, 2010 at 9:35 AM, Steve Samuels <sjsamuels@gmail.com> wrote:
> It's not clear which two proportions Omar wants to compare .  If he
> means correlated proportions (two attributes measured on the same
> person), here's a way that uses -svy: tab-
>
> **************************CODE BEGINS**************************
> sysuse auto, clear
> svyset _n
> set seed 2000
> gen u=uniform()
> sort u
> gen cat = _n<39   // new 2-level category
> svy: tab foreign cat
> test _b[p12] = _b[p21]
> lincom _b[p12] - _b[p21]
> ***************************CODE ENDS***************************
>
> If he means: test the equality of two cell proportions of the same
> attribute,  he can use -svy: prop- as Mary stated.
>
> ***************************CODE BEGINS**************************
> sysuse auto, clear
> svyset _n
> svy: prop rep78
> test _b[1] = _b[5]
> lincom _b[1] - _b[5]
>
> svy: prop foreign
> test _b[Foreign] = _b[Domestic]    //Use the category values shown in
> the results
> test _b[1] = _b[0]                // Unlabeled values don't work
> ***************************CODE ENDS***************************
>
> Steve
>
>
>
> On Fri, Mar 12, 2010 at 2:20 PM, Mary E. Mackesy-Amiti <mmamiti@uic.edu> wrote:
>> I believe you can use svy: proportion followed by test (post-estimation
>> command).
>>
>> see: help proportion postestimation
>>
>>
>> Omar McDoom wrote:
>>>
>>> Dear all,
>>>
>>> Does anyone know how I may compare two proportions drawn from a variable
>>> in
>>> a weighted survey?  The svy command I understand is not available for
>>> prtest.  Is there a work-around that allows one to incorporate survey
>>> design
>>> effects/weights?
>>>
>>>
>>>
>>>
>>
>> --
>> Mary Ellen Mackesy-Amiti, Ph.D.
>> Research Assistant Professor
>> Community Outreach Intervention Projects (COIP)
>> School of Public Health m/c 923
>> Division of Epidemiology and Biostatistics
>> University of Illinois at Chicago
>> ph. 312-355-4892
>> fax: 312-996-1450
>>
>> *
>> *   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/
>>
>
>
>
> --
> Steven Samuels
> sjsamuels@gmail.com
> 18 Cantine's Island
> Saugerties NY 12477
> USA
> 845-246-0774
>

--
Steven Samuels
sjsamuels@gmail.com
18 Cantine's Island
Saugerties NY 12477
USA
845-246-0774

*
*   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/