Need for attitudes survey for latent class model

[Rachel Murray-Watson]

Dear Professor Bierlaire (and others), 

I was wondering if it was possible to directly use the choices from a DCE instead of having some sort of psychometric indicator that measures attitudes and using this as an indicator in a latent class model. So, instead of measuring e.g. reported attitudes towards something, you are using their responses to the survey. 

Best wishes, 

Rachel 

[Michel Bierlaire]

> On 23 Apr 2024, at 15:01, Rachel Murray-Watson <r.mu...@gmail.com> wrote:
>
> Dear Professor Bierlaire (and others),
>
> I was wondering if it was possible to directly use the choices from a DCE instead of having some sort of psychometric indicator that measures attitudes and using this as an indicator in a latent class model. So, instead of measuring e.g. reported attitudes towards something, you are using their responses to the survey.

You should never do that. For two reasons:
- responses to surveys are "polluted" by potentially severe measurement errors that generate econometric issues such as endogeneity. This is why you need measurement equations.
- responses to surveys are not variables that can be used for forecasting. Basically, the model you would obtain is completely useless.

>
>
> Best wishes,
>
> Rachel
>
> --
> You received this message because you are subscribed to the Google Groups "Biogeme" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to biogeme+u...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/biogeme/58d81bc9-c74d-4cb5-b203-92017e60b0c9n%40googlegroups.com.

Michel Bierlaire
Transport and Mobility Laboratory
School of Architecture, Civil and Environmental Engineering
EPFL - Ecole Polytechnique Fédérale de Lausanne
http://transp-or.epfl.ch
http://people.epfl.ch/michel.bierlaire

[Rachel Murray-Watson]

Thank you for your quick response! So a survey of attitudes is always needed? 

[Michel Bierlaire]

If you want to include a latent variable in your model, yes.
And you need both structural and measurement equations.

> To view this discussion on the web visit https://groups.google.com/d/msgid/biogeme/e52a9e35-5766-40b8-bed4-70bbe06f4018n%40googlegroups.com.

[Rachel Murray-Watson]

Thanks so much for the feedback!