Have an idea?

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

ACBC calibration question

I got the raw data file from an ACBC study. In the calibration section, there are variable names like "ACBC_calibration1" and "ACBC_calibration1_shown". I understand these are the respondent's choice towards each calibration question and the product profile shown to the respondent. I notice that all the "ACBC_calibration1_shown" are 0s. Tech support told me they are the BYO profiles.  Yes, it makes sense to show the BYO profile as the first calibration question. However I found later that the respondents had the lowest willingness to buy towards this "ACBC_calibration1_shown", compared to other calibration profiles. It doesn't make sense.
asked Jun 15, 2017 by Xiaowen
It's not really intended to view the raw ACBC data and pull insights there.  If you think something might be going wrong, you can send in your survey file and data to support@sawtoothsoftware.com and we can take a look.

1 Answer

0 votes
Yes, look into whether you're interpreting the values properly in the data file.
 Aside from some strange error, sometimes respondents answer purchase intent questions with noise and error.

Also, please recognize that the BYO concept isn't necessarily the best concept shown to the respondent.  If you are using "summed pricing", then some products will have nearly the same composition as the BYO, but will have low price shocks, putting them on a much better price compared to the BYO concept.
answered Jun 15, 2017 by Bryan Orme Platinum Sawtooth Software, Inc. (142,815 points)
Is there other scenarios that the BYO concept is not the best concept, besides "summed pricing"? I didn't use "summed pricing" in this study. In the BYO concept, the respondents simply choose the levels of each attributes, including the price. But I found the BYO concept is not the concept with the highest purchase intent. Why is that?
If you are seeing that for a particular respondent, then perhaps that respondent: 1) made an error in the BYO, 2) Made an error in the purchase likelihood questions.  Even good respondents answer with some inconsistencies.
...