Dear all -
I have fielded a survey over the past months and have now received an adequate amount of responses to analyze the data.
I have run the HB and Logit and both results confirm each other in terms of directionality within attributes.
I am now wondering how to test for significance on two different levels:
On the aggregate level, i.e., full sample: Are the utilities computed significant in explaining the decision? At what level are they significant?
On the segment level, i.e., comparing separate subgroups: Are the computed utilities still significant when looking at smaller subsets of the sample? When there are differences, are they random or statistically significant (e.g., one segment prefers red, the other blue)?
Same goes for the relative importance analysis, but I guess the methods from the utilities apply as well.
I have not worked with conjoint data before and have not found concrete examples in the help guide. So if anyone could point me in the right direction for readings or methods I would be very grateful!