choice-metrics.com

Hi There,

We have designed a DCE with six attributes with 2 levels each and 2 attributes with four levels each (8 attributes in total) and an opt out option. Now we need to run the full factorial in Ngene. It took me awhile to work out the correct code, but I finally came back with something which runs, but I am still unsure if this is giving me the correct result:

Design
;alts = alt1, alt2, alt3
;rows = all
;fact
;model:
U(alt1) = b1 * A[0,1] + b2 *B[0,1,2,3] +b3 *C[0,1,2,3] + b4* D[0,1] + b5* E[0,1] /
U(alt2) = b1 * A + b2* B + b3* F[0,1] + b4* G[0,1] + b5* H[0,1]
$


I am hoping you can advise me at all if you think we are in the right direction?

Yes that syntax looks good.

There will be 128 profiles for alt1 (2*4*4*2*2) and 64 profiles for alt2 (2*4*2*2*2), so that means that the full factorial will have 128*64 = 8192 choice tasks. When you run the syntax, this generates indeed a design with 8192 rows.

Michiel