Hi,
I have some survey data which contains likert scale questions from two different countries.I would like to see if the estimates obtained from the first model (using data from the first country) would be a good fit if I were to use these estimates in the second country. However, I'm not 100% sure on how to proceed in a rigorous manner.
My original thought was to use crossfold validation from the first country to refine the estimates, do the same for the second country and then use the first model's coefficients in the second country and vice versa and compare the model fit (r2, RMSE and F test). However, I don't believe that cross validation/lasso works with ordinal variables (https://www.youtube.com/watch?v=efYBzFcKWn8).
Thank you for your help
I have some survey data which contains likert scale questions from two different countries.I would like to see if the estimates obtained from the first model (using data from the first country) would be a good fit if I were to use these estimates in the second country. However, I'm not 100% sure on how to proceed in a rigorous manner.
My original thought was to use crossfold validation from the first country to refine the estimates, do the same for the second country and then use the first model's coefficients in the second country and vice versa and compare the model fit (r2, RMSE and F test). However, I don't believe that cross validation/lasso works with ordinal variables (https://www.youtube.com/watch?v=efYBzFcKWn8).
Thank you for your help
Comment