Date: Mon, 5 Feb 2001 12:13:24 -0700
Reply-To: Carl.Elliott@UCHSC.EDU
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Carl Elliott <Carl.Elliott@UCHSC.EDU>
Subject: Re: 95% C.I.
Content-Type: text/plain; charset="iso-8859-1"
Kevin,
Just for the record, I am running SAS Version 8.0 and I get the same output
you
listed below for a t-test procedure.
Carl
email:carl.elliott@uchsc.edu
-----Original Message-----
From: kviel [mailto:kviel@GMCF.ORG]
Sent: Monday, February 05, 2001 11:31 AM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Re: 95% C.I.
Dale, Chelsea, and SAS-L,
My apologies once again. I should create a signature for exclusive
use for submissions to SAS-L. The whole time I was writing my last
submission, I kept wondering why I had done the exact same problem "by hand"
for my classes. I am using v8.01. Here is a teaser for those not yet there
(note the third line):
The TTEST Procedure
Statistics
Lower CL Upper CL Lower CL
Upper CL
Variable c N Mean Mean Mean Std Dev Std Dev
Std Dev Std Err
x 1 30 9.7132 10.098 10.483 0.8207 1.0305
1.3853 0.1881
x 2 30 -0.352 0.0123 0.3769 0.7777 0.9765
1.3128 0.1783
x Diff (1-2) 9.5668 10.086 10.605 0.8498 1.0039
1.2267 0.259
Kevin
-----Original Message-----
From: Dale McLerran [mailto:dmclerra@my-deja.com]
Sent: Monday, February 05, 2001 12:25 PM
To: kviel@gmcf.org; hunchcc@HOTMAIL.COM; sas-l@listserv.uga.edu
Subject: Re: 95% C.I.
Kevin,
I sure don't see confidence intervals for the difference between
means in PROC TTEST. The only confidence interval I see generated
in TTEST is for the estimate of the estimate of sigma. I know
that one of the goals for the SAS STAT folks is to add confidence
intervals wherever appropriate. Do you by any chance have version
8.1? Perhaps confidence intervals were added to TTEST there.
With version 8.0 of SAS, specifying options S and CLPARM on the
model statement in PROC GLM will return confidence intervals for
the parameter estimates. This would be one way to get the
required confidence interval. If Chelsea is still in version
6.xx, then check out the procedure MIXED, with options S and CL
on the model statement. PROC MIXED may seem like overkill for
performing a simple t-test.
As an editorial note, let me say that I have found MIXED to be a
very useful procedure for many simple analyses simply because it
does have built in such things as confidence intervals. The
procedures TTEST and GLM were basically written in ancient days,
when folks were interested only in testing hypotheses. PROC MIXED
was developed much more recently, and reflects the different
approach to statistical analysis which is predominant today in
which confidence intervals for our test statistics are usually
reported.
>Date: Mon, 5 Feb 2001 11:28:12 -0500
>Reply-To: kviel <kviel@GMCF.ORG>
>From: kviel <kviel@GMCF.ORG>
>Subject: Re: 95% C.I.
>To: SAS-L@LISTSERV.UGA.EDU
>
>Chelsea,
>
> Yes, PROC TTEST does perform this calculation.
>
>Kevin
>
>-----Original Message-----
>From: Chelsea Lee [mailto:hunchcc@HOTMAIL.COM]
>Sent: Monday, February 05, 2001 10:16 AM
>To: SAS-L@LISTSERV.UGA.EDU
>Subject: 95% C.I.
>
>
>This is a simple newbie question.
>
>Can I obtain 95% confidence interval for the difference between the means
of
>the two groups by using any PROC, like PROC TTEST? Assume that two groups
>are indepedent and have normal distribution. Or I have to compute by hand.
>
>Thanks!
>
>- Chelsea
>_________________________________________________________________
>Get your FREE download of MSN Explorer at http://explorer.msn.com
>------------------------
>CONFIDENTIALITY NOTICE: This e-mail transmission, and any documents, files
>or previous e-mail messages attached to it may contain proprietary,
>privileged or confidential information. If you are not an intended
>recipient, or a person responsible for delivering it to the intended
>recipient, you are hereby notified that any disclosure, copying,
>distribution or use of any of the information contained in or
>attached to this transmission is STRICTLY PROHIBITED. If you have
>received this transmission in error, please immediately notify me by
>reply e-mail and destroy the original transmission and its
>attachments without saving them in any manner.
>-----------------------
Dale
---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra@fhcrc.org
Ph: (206) 667-2926
Fax: (206) 667-5977
---------------------------------------
------------------------------------------------------------
--== Sent via Deja.com ==--
http://www.deja.com/
------------------------
CONFIDENTIALITY NOTICE: This e-mail transmission, and any documents, files
or previous e-mail messages attached to it may contain proprietary,
privileged or confidential information. If you are not an intended
recipient, or a person responsible for delivering it to the intended
recipient, you are hereby notified that any disclosure, copying,
distribution or use of any of the information contained in or
attached to this transmission is STRICTLY PROHIBITED. If you have
received this transmission in error, please immediately notify me by
reply e-mail and destroy the original transmission and its
attachments without saving them in any manner.
-----------------------
