Date:         Mon, 29 Jul 2002 09:03:02 -0400
Reply-To:     Paige Miller <paige.miller@KODAK.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Paige Miller <paige.miller@KODAK.COM>
Organization: Eastman Kodak Company
Subject:      Re: Calculating confidence intervals for binomial proportions and
              specifying the event
Content-Type: text/plain; charset=us-ascii; format=flowed

Christoph Helwig wrote:

> Hi there,
>
> I want to calculate the confidence limits of the proportion of responders
> for different treatment groups and have a response variable with 0/1 values
> no response/response. The problem is that Proc Freq gives me the proportion
> I don't want to have for drug=3, i.e. "the proportion of observations for
> the first variable level, or class, that appears in the output.
> (OnlineDoc)". Is there any way to force SAS to give me the proportion and
> CI for response=1 (like in Proc Logistic in the Model statement: Model
> response (event='1') = drug) even if no subject has response=1?
>
> Here's my snip of code
>
> data a;
>      input subject drug response;
>      datalines;
> 1    1    1
> 2    1    0
> 3    1    0
> 4    2    1
> 5    2    1
> 6    2    1
> 7    3    0
> 8    3    0
> ;
> run;


<...snip...>


Why don't you add to your data step the following command:

response = 1 - response;

--
Paige Miller
paige.miller@kodak.com
http://www.kodak.com

"It's nothing until I call it!" -- Bill Klem, NL Umpire
"When you get the choice to sit it out or dance, I hope you dance" --
Lee Ann Womack

.
.
.
.
.
.
.
.
.
.
.