'How do you interpret the output of an ordered logistic regression when one of the predictors has three levels (0,1,2)?
I fitted an ordered logistic regression model with two predictors and their interaction. One predictor (term) has three levels: 0,1,2. The other predictor (pred) has two levels: 0,1. I was wondering how to interpret the output of the model given that the first predictor has three levels. When I see term1, do I compare it to term0, or to term0 AND term2? What implications does this have on the interpretation of the interaction terms?
Thank you in advance for any help!
| Terms | Odds Ratios |
|---|---|
| term1 | 2.50 |
| term2 | 0.30 |
| pred1 | 1.20 |
| term1:pred1 | 0.35 |
| term2:pred1 | 1.15 |
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|