Have an idea?

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

More clarification

Hi Bryan,

Say I have 10 features in my study and I would like to find out the WTP for say Printing speed with 5 levels in it (20, 24, 28, 35, 45, 70). I tried out the above way by giving 1 product the enhanced features and the other one the lowest features, but I am not able to get the result.

I am also doing the other way by dividing the $ range with the utility range and corresponding no. which i get, is being multiplied with the utility scores of each attribute. Say price range is 0 - 400 and utility range is -7 to 10. So diving 400-0/(10-(-7))=133. After which multiplying each utility score by 133. Please suggest if this is fine.

Kindly help as soon as possible.

related to an answer for: willingness-to-pay cbc-hb covariates
asked Mar 27, 2019 by anonymous

1 Answer

0 votes
In the first case, the simulation-based approach, make sure to keep all the other variables constant, and only compare a product with the one level of print speed with one that has another level, all else being equal (except price, which you're going to increase in the product with the better print speed until the two shares equalize).

In the second case, using the formula, if you've run respondent-level HB utilities, I would do this calculation for each individual respondent and then take the median value - this tends to reduce some of the WTP inflation we frequently see.
answered Mar 28, 2019 by Keith Chrzan Platinum Sawtooth Software, Inc. (83,450 points)