Chris, Disclosure: I am a statistics user, not a statistician.
Not one time during my graduate coursework, or in my subsequent 5 years performing and reviewing applied research, has anyone mention transforming the ANOVA output from transformed data back into untransformed values. In fact, I had a hard time even drafting that sentence! I was instructed that one of the limitations of using transformed data in statistical analysis was you would not know the actual values, but the magnitude of difference(s) between(among) means would still be accurate - otherwise the ANOVA would not be accurate eh? I will have to ponder the idea of back-transforming (perhaps even fire off a question to my mentor). But I just wanted to state my recollections, since I would expect that if it was a valid operation I would have been instructed as such. I too would be interested what is acceptable practice as well as its theoretical basis. David Thomson -----Original Message----- From: Ecological Society of America: grants, jobs, news [mailto:[EMAIL PROTECTED] On Behalf Of Chris Caudill Sent: Tuesday, March 21, 2006 9:49 AM To: [email protected] Subject: transforms and CIs Hello all, Here's a question that I feel like I should know the answer to... I've conducted an ANOVA in sas on a large data set using log(e) transformed data. I'd like to plot means and 95% CI's using the sas output. Is it kosher to simply back-transform the CI's (I have a nagging feeling that it isn't). Thanks in advance. Chris Caudill Christopher C. Caudill Department of Fish and Wildlife Resources College of Natural Resources University of Idaho Moscow, ID 83844-1136 208-885-7614 (voice) 208-885-9080 (fax) http://www.cnr.uidaho.edu/UIFERL/Christopher_C._Caudill.htm NOTE NEW EMAIL: [EMAIL PROTECTED]
