crunching log data with same % but different GM

From: Maciej Simm (simmmmer@yahoo.com)
Date: Thu May 31 2001 - 15:35:56 EST


Dear group,

After the recent exchange on the topic of means and ratios in log vs.
linear space, I decided to read up on the topic and found some
background on all the terms.

In particular, thanks to Drs. Shapiro and Roederer for their emails.

My question:

when fluorescent signal on two groups of patients yields the same
frequency (% events relative to isotype control) but different  level
of fluorescence (log scale) how does one go about showing that
they're different?

I can put the GM's in excel and get a good P value but with
publication in mind I wanted to get some advice from the pro's on the
proper ways to do this.

This is a study where some cell markers were measured on fixed days
over about 2 years.

Is it enough to say "we ran FACSComp beads and our normal control
data for these markers do not change from lot to lot or month to
month so our GM differences in the patient groups are really really
true?"

I would appreciate any feedback.

Maciej


=====
`---------------------------------------------`
|    Maciej S. Simm      | 525 E 68th Street  |
|  Research Technician   | Room N-805         |
| Cornell Medical Center | Tel. 212.746.3428  |
`---------------------------------------------`
|www.cd4cd8.com | 917-734-6280 | AIM - XsimmX |
`---------------------------------------------`

__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail - only $35
a year!  http://personal.mail.yahoo.com/



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:01:21 EST