Re: Transforming ratios
- From: Bob Wheeler <rwheeler@xxxxxxxxx>
- Date: Wed, 26 Oct 2005 08:16:44 -0400
Geoff Hammond wrote:
I work with physiological signals where it's conventional to express the size of signals obtained in diffrent experimental conditions as ratios of signal size in a baseline condition. I routinely take log transforms of the ratios before analysis but scarcely anyone else does. Am I being too fussy?
Geoff Hammond
School of Psychology
University of Western Australia
Give an example of a "psychological signal." Is someone twisting a dial to indicate sensory strength? If the upper end of the signal can be very large, then you are doing the right thing.
-- Bob Wheeler --- http://www.bobwheeler.com/ ECHIP, Inc. --- Randomness comes in bunches. .
- Follow-Ups:
- Re: Transforming ratios
- From: Geoff Hammond
- Re: Transforming ratios
- References:
- Transforming ratios
- From: Geoff Hammond
- Transforming ratios
- Prev by Date: Goodness-of-Fit modifications of chi^2 : Haldane-Dawson, Craddock-Flood,....
- Next by Date: Re: Comparing two cross product tables
- Previous by thread: Transforming ratios
- Next by thread: Re: Transforming ratios
- Index(es):
Relevant Pages
|