Dear All,
I'm trying to construct a design for a consumer research study in which
respondents will be presented with a subset of 4 products from the 10
available. In addition is is believed that there may be carryover so I'm
effectively trying to produce a 10 treatment, 4 period balanced incomplete
block design which is also balanced for carryover. I've been told that I
need to restrict the number of blocks to between 20 and 30 for practical
reasons, which I suspect means that I cannot achieve the balance for
carryover. I am aware that there is some literature around about efficient
change-over designs when the number of periods is less than the number of
treatments but I do not have the papers to hand, so any advice would be
gratefully received.
Thanks,
Steve Ferris.
Dr Steve Ferris
Statistician
MMR Food & Drink Research Worldwide
T: 00 44 (0) 1491 824999
F: 00 44 (0) 1491 824666
E: [log in to unmask]
________________________________________________________________________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________
|