Dear All
I have to create a paired comparison design for 30 different
concepts in a product test.
The objective is to rank all 30 concepts in order of preference.
Obviously we cannot compare every concept with each other as this
will result in n(n-1) = 870 different cells which is not an option.
It seems that a round robin design would be suitable but I would be
interested in hearing what others would suggest?
Does anyone know if there is a quick way to work out the minimum
number of comparisons needed for a balanced design given there are 30
concepts, minimum sample size needed and then in fact how to create the
design?
Any help / advice would be much appreciated.
Many thanks
Natalie
|