ApolloChoiceModelling forum

Hi all,
We are estimating the results of a generic design, with two alternatives. that offers respondents washing machines under a circular business model.
One of the att. was whether the machine is offered for purchase or leasing (dummy coded).
The latent variable captures users' ownership preferences, was not significant.
One of the other questions in the survey was to rate on a 1-5 scale how much each of each washing machine attributes matters when deciding on a washing machine (e.g. load capacity, spin speed, energy efficiency, or the option to lease).
I thought of using the latest att. as a latent variable that can explain the preference toward the attribute (leasing). I know it is highly endogenous, the question is whether estimating both the choice and the motivation factor in a hybrid model overcomes this endogeneity?
Thanks for any advice
Anat

Anat

not sure I fully follow. The rating scale is an indicator, it's not a latent attitude. So if you use a latent variable to explain that indicator, and at the same time use it in the choice model, it would be a standard hybrid choice model.

Maybe show us your code to explain what you had in mind?

Best wishes

Stephane