Date:         Wed, 17 Jan 2007 12:39:03 -0500
Reply-To:     "data _null_;" <datanull@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "data _null_;" <datanull@GMAIL.COM>
Subject:      Re: proportions test in SAS
Comments: To: baogong jiang <bgjiang@gmail.com>
In-Reply-To:  <cabe11870701170819w3166582ejc51a274f3fa5b2a6@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Maybe you want CMH (Row Mean Scores Differ).

proc freq;
   tables region*race / cmh2;
   weight recip_no;
   run;

I am not a statistician but I have seen this test used and understand
it to be somewhat like a one-way ANOVA.  I think.



On 1/17/07, baogong jiang <bgjiang@gmail.com> wrote:
> Zack:
>
> Thank you for your suggestions. my problem is: I know that the race
> proportion is defferent in all regions. like we know in the population, it's
> about 60% are white, 35% are black and 5% are others.
> In real word, region 1 have 61%: 34%:5%
>                   regioln 2 have 63%:32%:5%
>                   region3   have 60%:38%:2%
>                   region4 have 55%:30%:15%
>
> I want to know if the race proportions is differrent in those 4 regions.
> like above sample, I think region 4 has different race propotions compare to
> other regions.
>