Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Procedure for CBC with dual response None and paper/pencil data

I am considering using LightHouse Studio to do CBC with Dual Response None - using an external programming vendor as well as paper/pencil data import (using the built in accumulated data template file) followed by HB estimation. I have not used dual response none before.

Is the process in LightHouse Studio essentially the same as for CBC with a traditional none option? Specifically:
(1) You generate and export the design for programming from LightHouse Studio - the only difference is that (on the Design tab) instead of choosing "Traditional None" you choose "Dual Response None". Is this correct?
(2) For paper-and-pencil data import - does LightHouse Studio generate an appropriate/correct accumulated data template file for dual response none data AND what needs to be entered in this file that is different from CBC with traditional None?
(3) Once the paper/pencil data is imported into LightHouse Studio, does HB analysis proceed the same as CBC with traditional none (in other words, you can run the default HB estimation settings without any special selections or additional data formatting to accommodate dual response none)?
(4) Is HB outputs for dual response None the same as CBC with traditional None (that is, attribute importance, level preferences, raw utilities that are all interpreted/used the same way for analysis and simulation as is done for CBC with traditional none)?

Thank you.
asked Apr 12, 2019 by anonymous

1 Answer

0 votes
Answers to your questions, in order:
(1) Yes, but if you're having the survey programmed in another software, they could just add the DR none question; choosing "Dual Response None" in Lighthouse Studio tells it add the DR question but it has no effect on the design (but you will need it checked for your analysis).
(2) Yes and the only difference is that after every CBC response you'll have the DR response (1 for yes, 2 for no)
(3) Yes, everything is the same
(4) Yes, your assumption here is correct.
answered Apr 14, 2019 by Keith Chrzan Platinum Sawtooth Software, Inc. (102,700 points)