Date:         Mon, 21 Mar 2005 15:24:17 -0500
Reply-To:     "Nick ." <ni14@MAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Nick ." <ni14@MAIL.COM>
Subject:      Stat. Question--Size of CONTROL group
Content-Type: text/plain; charset="iso-8859-1"

Hello,
I would like to get your thoughts on how large a control group should be when designing a study or a model, etc.
I am in the banking business where we sent out campaigns for various bank products. Some of you might be in the medical field where you deal with clinical trials, etc. What I would like to learn is if there is some statistical reasoning behind selecting a control group and the role this control group ought to serve in the study or modeling process. Some people say if you put aside 10% of your population as a control that would suffice. But I don't know how people pick up such percentages. I also don't quite understand how the control group is to be used in the study or modeling process. I know this is a general question about selecting control groups, but I still think several of you out there can give me some help with this issue or perhaps references where I can go and learn more about this subject. Thanks kindly.
NICK
--
___________________________________________________________
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm