Have an idea?

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

Counting analysis CBC


In your CBC technical paper i read:

"Counting analysis calculates a proportion for each level, based on how many times a concept including that level is chosen"

Would it be wrong to say that the proportions could be biased since whether or not a concept is chosen depends on ALL the attribute (levels) in the concept? It could be that a level is often chosen, but it was actually chosen because of an other attribute?

If so could i say that MBC counting analysis is less biased in predicting price demand curves since it is an item specific and not concept specific analysis?

asked May 16, 2013 by anonymous
retagged May 16, 2013 by Walter Williams

1 Answer

+2 votes
Counting analysis (isolating the effect of the levels of a single attribute on choice, or the joint effects of multiple attributes on choice) would indeed be biased if the other attributes within the same concept or across concepts (but within the same task) are somehow not balanced and orthogonal (independent in variation).

But, our CBC designs (assuming you don't do any prohibitions, and that you use enough versions of the design--and 4 to 10 versions are often enough) are near-balanced and near-orthogonal.  So, the presence of any level is (across all concepts and tasks for the sample) accompanied by essentially equal presence of all other levels of all other variables.  This makes it possible to isolate the effect of each level on choice under assumptions of main effects (for example) or first-order interaction effects (as another example).

MBC should also (asssuming you follow similar design principles) be equally unbiased in Counts.
answered May 16, 2013 by Bryan Orme Platinum Sawtooth Software, Inc. (134,015 points)