Have an idea?

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

Is there a formula for calculating the zero-centered diffs?

Yesterday a client ask me about the meaning of zero-centered diffs. I was able to explain what they mean, but the client wanted to know exactly how these utilities are calculated. I told him, I will give him the formula in the upcoming days. I tried to calculate the ZC-Diffs from the part-worths of a HB/CBC estimation of a recent study, but was not able to get the right values for the ZC-Diffs. I already read the article from Sawtooth Solutions, 1999, Issue 10 and the explanation in die SSI Web Manual.

Can anybody help me with this issue, please? I would really appreciate it!

Thank you in advance,
asked Jun 4, 2014 by Frank

1 Answer

0 votes
For each respondent...

1. Within each attribute, compute the mean utility.  Within each attribute, subtract the mean utility from each utility (this zero-centers the utilities within each attribute...which often doesn't have to be done since they are often already zero-centered in their raw form).

2. Then, for each attribute compute the difference between best and worst utilities.  Sum those across attributes.

3.  Take 100 x #attributes and divide it by the sum achieved in step 2.  This is a single multiplier that you use in step 4.

4.  Multiply all utilities from step 1 by the multiplier.  Now, the average difference between best and worst utilities per attribute is 100 utility points.
answered Jun 4, 2014 by Bryan Orme Platinum Sawtooth Software, Inc. (132,290 points)
Hi Bryan,

Couple of queries on the steps. Step 1 - is this within attribute across respondents or for each respondent across attributes? I have HB maxdiff raw utilites which already sums to 0 across attributes.

Step 2: Is this across respondents for each attribute or across attributes for reach respondent?
What are Diffs used for in ACBC?
What if preference order changes between raw values and zcd?

the above mentioned formula does not work for my estimation, when I try to reconstruct the calculation from raw to zcd values.

The following could play a role: What if I have
- an advanced design with primary and conditional attributes
- interactions
- dual-response none option
in my result table?

Can you please mention the exact formula for the multiplier which also applies for these cases?

Thank you in advance!