How to apply conditions for preparing conjoint cards in Conjoint Analysis in R?

r
business_analytics
#1

I have 6 attributes in which there is a constraint that in one card in one option if we have given 2nd Level of the 1st attribute then few levels of another attribute can’t be there in the same card.

For example, there are 2 attributes: Colors and Offers. Say, Color has 2 levels Black and White; and offer has 4 levels 0, 10, 20, 30. So the constraint is that; Black has only 2 offers 0 or 10; and White has another 2 offers 20 and 30.

This is just an example; because the actual scenario has little more attributes and levels with constraints.

I want to know if this is a normal constraint, which happens to be there in all the conjoint studies or it’s only in my study and so I need to make some changes.

I am looking for a function within Conjoint library in R which “considers these constraints” unlike expand.grid and caFactorialDesign.

#2

Well, I understand your concern regarding the constraint on the choice of colors and Offers. The very nature of Conjoint Analysis is to give you discrete choices and identify your nature of choices and your trade-offs . Let me take a better example for instance:
You have the following three variables:
price : 100, 200 and 300.
mileage: 15, 20 ,25
Car: A, B, C

When you give a choice to the buyer, you give conjoint choices. Ex: 1) A, 15, 300 2) B, 20, 200 3) C, 25, 100
The reason you do this is to identify the nature of the buyer whether he is going for the price by trading off with mileage or if he is going with mileage trading off price or if he is neutral with both.

Now you may ask why are we giving just three choices, while we have 333 = 27 combinations. Again this is a business decision we try to determine the choice predictions with the minimum number of discrete choices.

So if I give a buyer a choice to select the attributes individually, we know that he will select the best car, the best mileage, and the best price. and the whole purpose of making choice predictions is gone.

The CA is well explained in this link, you can refer to this for deeper understanding:
Conjoint Analysis

Regarding the R scripts, I am also trying to find a good R script on Conjoint analysis. If any one has on git hub, I request you to kindly share.

#3

Thank you for the response and a better example.
I understand that we have to give minimum choices to the buyer in Conjoint Analysis, because some kind of combinations are not possible in real business world. The buyer would always want the best product with best features with minimum price, which is not possible. So We can not give such impossible combination of options to the buyer for our study.
That is the reason why I have constraints in my study, and I am also looking for R Scripts.