Date:         Mon, 23 Sep 2002 12:16:32 -0400
Reply-To:     Jay Weedon <jweedon@EARTHLINK.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Jay Weedon <jweedon@EARTHLINK.NET>
Organization: http://extra.newsguy.com
Subject:      Re: How to run Goodman-Kruskal Gamma test in SAS?
Content-Type: text/plain; charset=us-ascii

On Mon, 23 Sep 2002 13:44:33 +0800, "YCF" <yiu628fan@yahoo.com.hk>
wrote:

>Hi all,
>    I have a dataset which is about the relationship between the income and
>job satisfication:
>
>Income       dissatisfed   no opinion  satisfied   total
><6000                  20                24             80       124
>6000-15000        22                38             28         88
>15000-25000      13                28             81       122
>>25000                  7                18              54        79
>
>Total                    62              108           243      413
>
>    Now, I would like to apply the gamma test to see whether they are
>independent or not, also I would like to find out the asymptotic variance of
>the test statistics. How can I archieve this?

RTFM: In other words, look up GAMMA in the documentation and you'll
see that you can get it by running proc freq with the /measures option
in the tables statement. For your table gamma=0.067, asymptotic
SE=0.0633.

If the dataset actually resembles the above (i.e., it has only 4
records), you can remove the totals then transpose it:

proc transpose data=old out=new name=satisfaction prefix=freq;
  by notsorted income;

before running weighted frequencies:

proc freq data=new; tables income*satisfaction /measures;
  weight freq1;
run;

JW