Hi,
I conducted a short study which included two CBC conjoint sets with 12 questions each. So one survey, two conjoint sets included. Both conjoints sets were answered by each respondent of the survey and the order of the conjoint blocks was randomized. All attributes and all attribute levels were identical in both conjoint sets, the only difference was that the respondents were once asked to imagine that the underlying product was a television and once a tablet computer. The utilities were estimates with CBC/HB.
So here is my questions: Am I allowed to compare part-worth values across the two conjoint sets (within one attribute) and explain the differences, for example that one attribute level has a significantly higher part worth for the television than for the tablet, by the product difference? I want to run t-tests to find out if the differences are statistically significant. I have tried it and all results are plausible against the backdrop of the product difference.
Thanks a lot in advance.
Leonard