Hi, 
Im currently trying to run a generalised ordered logit regression. I understand how to do this however in the paper which i'm following it calculates in a way I don't understand. for example the y variable is happiness which has 3 categories. Very happy, pretty happy and not too happy. In the paper instead of just having one results for the regression it has Very happy versus pretty happy or not too happy and Very happy or pretty happy versus not too happy. I've done a lot of research around the ordered logit regression but haven't found anything that presents it's results like this. Any help would be much appreciated.

1 attachment