CBC - Two price models

Question

Hi everyone,

I got a questionnaire containing a CBC from a project manager that is set up like this:
- Respondents are shown a concept and are asked, before entering the CBC, whether they would prefer to pay for it a) at a monthly rate or b) the whole price at once.

- Then the CBC is conducted with a variety of attributes, one of which is price (5 attribute levels). The levels depend on the price model preference question that was asked before the CBC exercise begins. Distances between the five price points are proportional between the monthly-rate-model and the one-time-payment model.

My question is: Do I have any disadvantages incorporating both price models in the same Conjoint analysis and just using some unverified perl to show either a monthly rate or a one-time-payment price? Or is it advisable to set up two separate conjoint exercises, one per price model?

As I understand it, this would only be called conditional pricing if the attribute "price model" would be included in the cbc itself, rather than being asked outside of the conjoint exercise?

Note: reservation prices shall be calculated during the analysis.

Answer 1

If you're doing it as a single CBC exercise, and dynamically switching the text for price to be either one-time payment or monthly model, then you indeed could model them in a single HB estimation run.  But, I'd include whether they got the one-time payment or the monthly prices as a covariate (easy to do in our software interface for HB).

The challenge is that in the market simulator, the standard approach makes you pool everything together and express the prices as a single price attribute.

There is a power trick, with "Store shopping distribution" option to be able to assign respondents who pick the "monthly fee" to one "store" and the respondents who pick the one-time fee to a differnent "store".  And, one store can offer one set of products from the market simulator (set as monthly prices) and the other store can offer a different set of products from the simulator scenario (with the one-time prices).  

So, there is a way to keep the data all together, estimate with a covariate, and then use store-shopping option in our simulator to be able to offer and specify products that offer either monthly or one-time prices.

Comments

But, I should add that when you use the simulator in "store shopping" mode, the price attribute is still one price attribute, and you'll be expressing prices as relative prices on that single price continuum.

---

Thank you for the quick response, this information helps us a lot!