The first 50 you grab are likely to be the 50 slowest, most docile, etc.
animals. These would all end up in one group and these characteristics
would be confounded with the experimental treatment. We make randomization
a habit at every choice point (assignment of words to lists to be
remembered, assignment of subjects to conditions, determining the order of
running multiple conditions, etc.) because it is our only protection from
the subtle effects of the many things that can bias our choice when we make
them by the seat of our pants in the hopes that that practice will be
"random enough" - usually it isn't.
Claudia
At 05:06 PM 3/1/00 -0500, you wrote:
>
>how necessary is it to randomly select animals for the control and
>experimental groups in animal studies?
>so you got 100 rats,50 to go to the E group and 50 to go to the
>C group,can randomization really equalize for the two conditions?
>It appears that there is already a stability factor,since some of those
>animals may be bred from the same strain or litter?
>
>Send me something.
>
>Michael Sylvester
>Daytona Beach,Florida
>
>
>
>
>
>
>
>
________________________________________________________
Claudia J. Stanny, Ph.D. e-mail: [EMAIL PROTECTED]
Department of Psychology Phone: (850) 474 - 3163
University of West Florida FAX: (850) 857 - 6060
Pensacola, FL 32514 - 5751
Web: http://www.uwf.edu/psych/stanny.html
