Have an idea?

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

Is there a way to conduct a CBC study with large number of levels (150 or more)?

Is there a way to conduct a CBC study with large number of levels (150 or more)?

The task is to test 150–200 items , each of those may have 3 different price levels. As a result I need items rating and price elasticity for each item.

If there were less itemsm I would use plain CBC design with 2 attributes: «Price» and «Item», but it seems that this wouldn't work if the number of items is huge, even for large samples.

I'm also considering Maxdiff to rank the items, but it won't let me evaluate the price elasticity.

Maybe there is an alternative approach for such cases?
asked Jul 25 by Ahegor

1 Answer

0 votes
Ahegor,

Actually, there is a way to use MaxDiff  for price elasticity (http://www.sawtoothsoftware.com/support/technical-papers/maxdiff-best-worst-scaling/the-options-pricing-model-an-application-of-best-worst-measurement-2004).

The problem will be the huge number of items you have, but that may be something you can address with Express MaxDiff, which you can read about in the 2012 Conference Proceedings or here (http://www.sawtoothsoftware.com/support/technical-papers/175-support/technical-papers/maxdiff-best-worst-scaling/1493-a-parameter-recovery-experiment-for-two-methods-of-maxdiff-with-many-items).

If all of you items have about the same price ranges, then I may even have another idea for you.
answered Jul 25 by Keith Chrzan Gold Sawtooth Software, Inc. (48,525 points)
...