choice-metrics.com

Hello everyone!

I tried the following foldover design with Ngene. It gives me a design including a folover block which divides the design in to 2. For my experiment this is too large. Would it be ok if I separately add blocking to the desigh below (for example ;block=3) and not use the foldover block generated by Ngene?

Design
;alts =alt1, alt2, alt3, alt4
;rows = 36
;orth= seq
;foldover
;model:
U(alt1) =b1 +b2*A[0,1,2]+b3*B[0,1,2]+b4*C[0,1]+b5*D[0,1,2]+b6*E[0,1,2]/
U(alt2)= b2*A+ b3*B+ b4*C+ b5*D+ b6*E/
U(alt3) = b2*A+ b3*B+ b4*C+ b5*D+ b6*E
$

Thank you

Thames

Hi Thames

I don't see any issue with what you propose. Indeed, I believe you can still use the ;block = command even with the foldover. When I tried the following, I got a design.

Design
;alts =alt1, alt2, alt3, alt4
;rows = 36
;orth= seq
;block = 6
;foldover
;model:
U(alt1) =b1 +b2*A[0,1,2]+b3*B[0,1,2]+b4*C[0,1]+b5*D[0,1,2]+b6*E[0,1,2]/
U(alt2)= b2*A+ b3*B+ b4*C+ b5*D+ b6*E/
U(alt3) = b2*A+ b3*B+ b4*C+ b5*D+ b6*E
$

In terms of the design properties, it won't affect it provided each block is equally replicated over the sample.

John

Dear John,

Many thanks for the quick response!

Thames