Hello,
first of all I would like to say that you are doing an excellent job in this forum. I plan conduct to an academic study with your software and the forum did help me a lot during preparation!
Now I have two questions concerning my choice based conjoint study. I plan to estimate utilities with HB approach. A friend of mine gave me a HB estimation report file and I am not sure if I understand the values right.
In the first tab of the Excel file there is a summary tab (average utilities zero centered diffs). In the third tab I can see the individual ZC diffs and in the fourth tab the individual raw utilities are reported.
I read that the individual raw utilities are already zero-centered (effect coding) and that the individual ZC diffs are the normalized zero-centered raw utilities. They are normalized by multiplication with a unique constant per respondent to set all respondents on the same scale.
In my study I want to be able to say how the attributes included differ in their impact on preference. In the summary tab I see the average importances. Can I say that the attribute brand for example does has a 40% impact on choice while style does have a 25% impact?
If style is another attribute in my study, can I say which styles are more preferred than others? What is the difference between the average utilities given in tab 1 (average of zero-centered diffs) and if I calculated the average of the individual raw utilities from tab 4? I read that "only differences in utility matter" and multiplying part-worths by a constant has no consequence on the meaning of the data. So I am confused which average I should use and why.
Finally I will need to sum up part-worth utilities to obtain the total utilities for different product concepts to compare them. I would like to say that one product concept is more preferred than another concept for example. In a Marketing Handbook I read that for that purpose one simply sums up part-worths but they do not report which values they summed up, the average zero-centered diffs or the average raw utilities.
Thank you very much. Have a nice day.
--Thomas