Moi!
I have an experimental design headache: Briefly, my question is whether
anyone else has come across the problem of designing experiments where
subjects are used repeatedly in trials and where in the trial they may
interact.
The details: I was asked last week about an experimental design with one
of those twists you never see in the textbook. Briefly, the biologist
wanted to look at the fitness of a female fish (i.e. number of eggs
produced) and how it is affected by the number of males she was given
access to. So, for each trial she wants to put a female in a with
several males and carry out trial for, say, 1, 2, 5, 7, males. She has
plenty of females, but only about 20 males, so she will have to re-use
them. Horrible problems of independence and estimating the male's
effects arise (e.g. what if the strength of the effect varies with
number of males?), and randomly assigning them to trials will mean that
they have a higher probability of being in a trial with more males.
Does anyone know of any literature on this sort of problem? I have some
thoughts I'm that there is far greater wisdom out there than I can come
up with, but I'm not sure where to look for it.
Please reply to me, and I'll send a summary of any replies to the list.
Thanks!
Bob
--
Bob O'Hara
Rolf Nevanlinna Institute
P.O. Box 4 (Yliopistonkatu 5)
FIN-00014 University of Helsinki
Finland
Telephone: +358-9-191 23743
Mobile: +358 50 599 0540
Fax: +358-9-191 22 779
WWW: http://www.RNI.Helsinki.FI/~boh/
|