I will be soonly conducting a CBC study to estimate share of volume. I read some tips in SSIWeb help file but still have some concerns. I would be very grateful if someone could guide me how to set up this kind of study.
The general background is as follows. I'd like to estimate share of volume for 30 SKUs. Each SKU has its name and price range. Therefore, I initially plan to set up 31 attributes, with the first attribute is SKU list, and the rest 30 are each SKU's price range. I wish to design it as 15 SKUs per task via shelf-displaying format, and 15 tasks would be taken in total.
Now, I have two plans.
A. "Allocation design". I will change response type for each SKU from single-choice to constant sum / allocation format. There is no need to sum figures to the specific value, and would exclude none option. The value is the purchase quantity within next one month which asks respondents to fill in.
B. "Choice based plus allocation design". I still ask respondent to select his most preferred SKU (including none option), and in the next page, I will ask him to fill the value of purchase quantity within next one month for his selection.
My questions are ...
1. Which plan would be more suitable, feasible and realistic for the analysis and simulation?
2. If Plan A is adopted, how to deal with those boxes without any value filled in (I think respondents will leave some boxes blank for those they would not buy it within next one month)? Leave them as blank or fill 0 for them?
3. If Plan A is adopted, are there any specific settings or must-have procedures I need to pay attention to during estimating the utilities?
4. If Plan A is adopted, could share of preference calculated by utilities treat as share of volume for each SKU?
5. If Plan A is adopted, can I still use your online simulator or I need to develop one by myself?
6. If Plan A is adopted, does the simulator which developed by myself need to do more convertion or special settings?
7. If Plan B is adopted, how to combine choiced answer and quantity in one data file?
8. If Plan B is adopted, can I still estimate utilities as general CBC analysis, then apply quantity as weighting? If so, should I set up the quantity under the specific SKU+price combination? If not, how can I apply the quantity?
9. If Plan B is adopted, how to set up simulator integrating the quantity?
I know this is a long post, and seems to be a complex design. But it is really confused me for some days. Since this is my first time to conduct volumetric CBC study, thus, any suggestions and kind guidelines and supports are highly appreciated.
p.s. I will use the latest SSIWeb (8.0.2) with advanced modules to set up this study.