Have an idea?

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

I want to conduct a CBC with two different treatments, how can I design that?

Hello community,

I am conducting my first CBC survey and have difficulties deciding on the design. The participants will all be presented with the same CBC exercise, with the same attribute&levels but there will be two different treatments ('products').
First, I planned on doing a standard CBC with skips and quotas but when I looked into the analysis it seemed, that would be some difficulties. I found a post in the forum,  where someone had done a similar design and was not able to do the analysis without a lot of extra work because the resulting data couldn't be properly processed by the analysis manager. If it is possible I want to avoid this approach. For reference, here the post: https://legacy.sawtoothsoftware.com/forum/18028/comparing-cbc-hb-results-of-two-treatment-groups.

To circumvent these troubles I thought of two solutions and would like some input if they are feasible and if I should prefer one.

1. I could let all participants do both CBC exercises. Then I would have to lower the number of choice sets per exercise but because I only have 4 attributes I think it wouldn't affect the validity of the results.

2. I could instead use a ACBC. I saw this approach in a published paper, where they essentially did a CBC exercise but instead of the extensive BYO and filters they only randomly assigned participants into one of the two treatment groups. Then they conducted the analysis normally.

Than you all in advance for the help. I hope I didn't overlook some easy solution, this is my first Conjoint Analysis and I don't have much experience yet :)

asked Jul 16 by SofieBils (120 points)

Your solution to the original question

Please only use this to answer the original question. Otherwise please use comments.
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
Anti-spam verification:

To avoid this verification in future, please log in or register.