Have an idea?

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

How to calculate zero-centered diffs out of SEM-part-worths

Hey hey,

I have a little problem with the results of two of my surveys. The first one is a normal SEM with a rank order for the attribute levels in stage 1 and with a constant sum (100 points) for the attribute importances in stage 2. The second survey is an ACBC that shows part-worths in zero-centered diffs.
Now I would like to normalize the simple part-worths of the SEM into zero-centered diffs to be able to compare the them within the attributes for the two surveys.
Is that somehow possible? And if it is how does it work? I was not able to find any academic article about the calculation of the zero-centered diffs.
Can you help me out please?

Regards
asked Jul 28, 2016 by Mo

1 Answer

0 votes
answered Jul 29, 2016 by Bryan Orme Platinum Sawtooth Software, Inc. (142,515 points)
...