choice-metrics.com

Dear all,

I design labelled DCE for pharmacy technician(2 years) and pharmacy technician(4 years)that attributes are

different levels/values depending on the type of alternative. Can I design by syntax
Design
;alts=cer, ba
;rows=14
;eff = (MNL, D)
;model:
U(cer) =b1.dummy[0 I 0]*i1 [1,2,3] +b2.dummy[0I 0I 0 ]*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] $
and Is it a little bit difficult to measure alternative-specific preferences for the attributes?
Best Wishes,
Sukunta Muadthong

Hi Sukunta

You can have alternative specific designs. These are more common in some literature than others, however there is nothing wrong with them. For example, in transportation, we are often interested in mode choice - car, bus and train. In these examples, the attributes might be different as unless you have really upset your partner, you are unlikely to be charged a fare to use your car, but you have a fare for bus and train, and the bus driver is not going to ask you to pay a toll on a toll road. You can also have different levels for similar attributes as you suggest, as the car might have shorter travel times than the bus or train (but be more expensive when you add together parking and toll costs).

So there is nothing wrong with your proposal. I haven't checked your syntax as that is what the other part of the forum is for.

John

Dear Professor John R.,

thank you very much for your reply
and when my pilot result finished.Can I post in this forum to report and ask you again or I post in a new topic?

Best Wishes.
Sukunta Muadthong

Hi Sukunta

You can post general questions about design theory here. the other part of the forum is for specific questions related to Ngene such as questions about syntax.

John

Dear Prof. John R.,
Thank you very much for your kindness.

Best Wishes,
Sukunta Muadthong