Date: Tue, 3 May 2005 09:51:40 -0400
Reply-To: "Gerstle, John" <yzg9@CDC.GOV>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "Gerstle, John" <yzg9@CDC.GOV>
Subject: Re: Weird statistical issue regarding correlations and
logisticregression
Content-Type: text/plain; charset="us-ascii"
Peter,
One thing comes to mind about your data where the means are the same,
but the correlations are not. At SESUG in Nashville last October, I took
David Cassell's Sampling class and he had an example where the means
were the same, but the variance differed when the population correction
factor was included in the calculations (the difference between using
PROC MEANS and PROC SURVEYMEANS). Don't know if this helps at all, but
it might suggest an area to look into.
John Gerstle, MS
Biostatistician
CDC Information Technological Support Contract (CITS)
NCHSTP \DHAP \HICSB \Research, Analysis, and Evaluation Section
Phone: 404-639-3980
Fax: 404-639-2980
Email: yzg9 at cdc dot gov
>> -----Original Message-----
>> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
>> Peter Flom
>> Sent: Tuesday, May 03, 2005 9:46 AM
>> To: SAS-L@LISTSERV.UGA.EDU
>> Subject: Re: Weird statistical issue regarding correlations and
>> logisticregression
>>
>> Kevin Viel wrote
>>
>> <<<
>> It sounds like you have a follow-up publication, likely a letter.
As
>> I see
>> it, this is necessary. Would you client change her or his mind if he
>> or she
>> knew that you or someone else will have to point your findings out in
a
>> letter to the journal?
>> >>>
>>
>> My client just wants to get his dissertation approved, and his
>> committee has (apparently) approved his current plan. If I write a
>> letter, it would be after he get his dissertation - but it isn't my
>> substantive area, so I am hesitant.
>>
>> Peter
>>
>> Peter L. Flom, PhD
>> Assistant Director, Statistics and Data Analysis Core
>> Center for Drug Use and HIV Research
>> National Development and Research Institutes
>> 71 W. 23rd St
>> www.peterflom.com
>> New York, NY 10010
>> (212) 845-4485 (voice)
>> (917) 438-0894 (fax)
| 
