Ordinal regression (logit) with selection variable?


New Member
Hi all,
want to do an ordered logit regression. I have data from different sources and need to use the source as a selection variable for the dependent variable (only for the dependent variable, not for the regressors!).
I know how to do this for a linear regression. Is there a way to do this for ordinal regressions as well?

(I am researching whether past online ratings for retail stores influence future ratings.
I have a dataset of approx. 800.000 ratings. The unit of analysis is the rating given by person i to store s.)

Thanks for the help! :tup:
Last edited: