choice-metrics.com

Dear Michiel,

because my model was block . I want to calculate minimum sample size for proper my model by Ngene.
and How?
thank you very much
Best Wishes,
Sukunta

If you have 2 blocks, you simply multiply sample size by 2. If you have 3 blocks, you multiply the sample size by 3, etc.
Please note that sample sizes reported by Ngene only make sense when your priors are sufficiently accurate, i.e. come from a pilot study.

Dear Michiel,
thank you very much and i will read more in John M. Rose • Michiel C. J. Bliemer,2013 again.

Best wishes,
Sukunta

Ok. In that paper we provide the equations, in Ngene you find the sample sizes reported as S-estimates, which are the ones you need to multiply with the number of blocks.

Dear Michiel,

and I will add , s in syntax in my model that
Design
;alts=cer, ba
;rows=14
;eff = (MNL, D, S)
;block=2
;model:
U(cer) =b1.dummy[0 I 0]*i1 [1,2,3] +b2.dummy[0 I 0 I0 ]*L1 [1,2,3,4] +b3[0]*salarycer[8985,13800] /
U(ba) = b0+b4.dummy[0I 0I 0 ]*i2 [4,5,6,7] +b5.dummy[0I 0I 0 ] *L2 [5,6,7,8] +b6[0]*salaryba[11000,15000,18000] $

Or not?
Best wishes,
sukunta

No you just keep ;eff = (mnl,d) to optimise on the D-error (recommended) and when you open a design, Ngene simply reports the S-estimates so you can inspect them. If you would like to optimise on sample size, then you specify ;eff = (mnl,s).