Subject: | |
From: | |
Reply To: | |
Date: | Fri, 8 Mar 2013 16:58:09 -0800 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
>> Then I like to normalize all mite infestation rates to 100% (or 1),
>> so that controls and treated hives all start at the same place.
> for the sake of clarity, perhaps you > could explain exactly what you mean by "normalize".
Allen, I'm no statistician. I simply divide each mite count by the
starting count.
If there is no change in mite count, then you get a result of 1. If
the count doubles, you get 2 (or 200% of the original).
If the count goes down to half of the original, you get 0.5 (or 50%).
When you plot these out, no change in mite levels results in a level
line. It is then easy to compare the (typical) upward slope of mite
infestation rate of the control group to the rates for the treatment
groups.
If there are any statisticians out there, I'd be happy to hear
constructive suggestions.
Randy Oliver
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
|
|
|