Suppose I have an opinion survey on some topic. Both my dependent variable and independent variable are categorical variables. My question is, if I use the ordered probit model, how do I interpret the results? The independent variables are also categorical so I can't find the marginal effects either. Then, how do I interpret the coefficients and is there anything else I can do similar to finding the marginal effects?


Generally your are estimating probabilities for every category j of your dependent variable y. Similar to marginal effects, not as far as I know. You can estimate the probabilites for the response-categories with mfx in stata if I remember correctly.

Concerning the interpretation of the coefficients UCLA can help: "Standard interpretation of the ordered logit coefficient is that for a one unit increase in the predictor, the response variable level is expected to change by its respective regression coefficient in the ordered log-odds scale while the other variables in the model are held constant."

| cite | improve this answer | |
  • $\begingroup$ If you are interested into the theory I refer you to the Wooldridge from 2002. The big one with a lot of equations. In my experience the best companion to understand the math behind econometrics. $\endgroup$ – 5th Jul 21 '16 at 13:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.