some detailed questions

[xiaojin]

Dear Ngene experts:
First of all, thank you very much for your help in the early stage of design. Now my design has entered the final stage. I want to confirm some details.
This is my syntax:
Design
;alts=alt1*,alt2*,alt3
;rows=20
;block=4
;eff=(mnl,d)
;model:
U(alt1)=b1.effects[0.001]*A[0,1]+b2.effects[0.003|0.002|0.001]*B[3,2,1,0]+b3.effects[0.001|0.001|0.001]*C[3,2,1,0]+b4.effects[0.003|0.002|0.001]*D[3,2,1,0]+b5.effects[-0.003|-0.002|-0.001]*E[3,2,1,0]/
U(alt2)=b1*A+b2*B+b3*C+b4*D+b5*E/
U(alt3)=b0[0]

1 As shown above, my “block=4”. My question is: There are lots of blocks, will it (the number of blocks) affect the efficiency of DCE ?

2 As shown above, the number of my attribute levels is 2,4 and I have 13 parameters to estimate in total, my “rows=20”. Is that ok? Can “rows=20” satisfy the data diversity ?

Thank you again for your help.

[Michiel Bliemer]

1. Blocking does not affect efficiency.

2. 20 rows satisfies the degrees of freedom and therefore you will be able to estimate all parameters. However, if you are including any interactions between your attributes and estimate more parameters you may run into trouble. If you do not intend to include interactions then 20 rows should be fine. You can also consider 24 rows with 3 blocks, i.e. 8 choice tasks each. Your design does not look too complicated so I think that respondents may be able to handle 8 choice tasks instead of only 5 (in the case of 20 rows and 4 blocks).

Michiel