As I did not get any answer on CrossValidated, I will try my luck here.
I am working on a project where I have collected ordinal outcome data at three different time points. I am interested in modeling the change over time with these ordinal outcomes. Initially, I considered running an ordinal logistic regression separately for each time point. However, I realized that this approach might not effectively capture the temporal dynamics of my data, such as changes over time and within-subject correlation.
Therefore, I am seeking advice on properly modeling this kind of data. Here are some specific questions I have:
Is it appropriate to use a time series ordinal logistic regression model in this case? If so, how should I handle the time aspect of the data?
How can I perform this analysis in R? Are there specific packages or functions that I should know about?
Should I check any assumptions before and after running the model?
If a time series ordinal logistic regression model is not suitable, what alternative models would you recommend for this kind of data?
