Have an idea?

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

MaxDiff : some items filtered for some targets


I have a question really close to the former one.
I have a full list of 22 items and 4 "target populations".
One of the groups has a design with the full list of items. For the 3 other ones, I have different designs,  consisting of the same list with 2 to 3 items excluded.
How can I handle it in HB? (I use the version 5.2.8 and specified user specified and dummy coding).
I thought I could use the same cho file with all respondents - including choices and design - and then specify covariates (I would like to use the  information common across my 4 groups), But the utilities computed don't seem to be as described above :
For the people not having seen the item "a draw near the mean of those having seen it".
And then, how should I rescale the scores , just excluding the items is not enough, is it?
Am I missing something?

FYI, the link provided doesn't work anymore.

Many thanks in advance for your answer.
Best Regards,

asked Apr 1, 2014 by Sophie LOHEZIC

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.