Hi everyone,
I am a beginner on quant and STATA, and I am working on a project in which we are required to find our preferred model.
We have a journal article, and we are asked to adjust the author's original model. He uses an ordinal logic regression, with the DV a five-scale ordinal variable and the IV several categorical, ordinal and nominal variables.
I was thinking about collapsing the ordinal dependent variable (5-scale) to a dichotomous one, because the original model violates the parallel assumption.
I know that if I am using multinomial logistic regressions, I can use "mlogtest, combine" to test whether the DV could be collapsed.
So my question is, is there any test for ordinal logistic regression that can help me determine if I can collapse the DV?
Thank you!
I am a beginner on quant and STATA, and I am working on a project in which we are required to find our preferred model.
We have a journal article, and we are asked to adjust the author's original model. He uses an ordinal logic regression, with the DV a five-scale ordinal variable and the IV several categorical, ordinal and nominal variables.
I was thinking about collapsing the ordinal dependent variable (5-scale) to a dichotomous one, because the original model violates the parallel assumption.
I know that if I am using multinomial logistic regressions, I can use "mlogtest, combine" to test whether the DV could be collapsed.
So my question is, is there any test for ordinal logistic regression that can help me determine if I can collapse the DV?
Thank you!
Comment