FERRET-SEARCH Archives

Searchable FML archives

FERRET-SEARCH@LISTSERV.FERRETMAILINGLIST.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Bob Church <[log in to unmask]>
Date:
Sun, 1 Feb 1998 03:50:08 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (37 lines)
Ok.  folks, I'm running a bit behind on finishing the adrenal post, mainly
becuase the data conflicts so much with a couple of papers that I've read.
I hesitate to post something that may be obviously biased, so I am setting
on the stuff and pondering it.  I will post the result no later than next
weekend, although, like I said, the results were somewhat different than
what I expected.
 
Thats the problem with being a scientist; the ethics of reporting values
which are apparently biased without attempting to explain or correct for the
problem.  I've just been considering reasons for the difference, thats all.
I will report the findings, I'll probably just put a lot of disclaimers into
the post.
 
One other problem is the post will be quite long, so I'll probably have to
post it directly to Bill (and I know his middle name, naa naa naa) so he can
post it as space permits.  Its way longer than 125 lines so be prepared for
a long read (I can never figure out the 125-lines thing anyway because I use
a file size method that doesn't translate in my brain.  Sort of like I can
use metrics or miles, but I have a hard time translating them.  Well, I
*know* a pound is .45359237th of a kilogram.  Just stuck in my head, like
pv=nrt and stuff like that.  Just don't ask my phone number...)
 
As for *why* the adrenal reports were so biased?  The FML is *not* composed
of typical ferret owners, its that simple.  Rather I should say, the people
who reported the adrenal stuff are not typical ferret owners.  A very large
part was reported from ferret shelters or people who adopted ferrets that
had a history of illness, neglect or malnutrition.  That means the sample
was not random.  What I'm trying to do is see if I can correct for the skew,
but it's not looking favorable.  As for any correlations, well, if I can't
correct for the skew, then any correlations whould be as good as "96% of all
Americans who die in accidents are wearing clothing, therefor clothing
causes accidents."  The skew can hide the cause-effect relationship.
Everyone knows its nudity that causes accidents.
 
Bob C and 20 MO Accidental Biters.
[Posted in FML issue 2205]

ATOM RSS1 RSS2