Dear Statalist,

I am using ordered logistic regression, depvar: 5-scale happiness level (women only), indepvars: employment status (employed, unemployed, housewife, other employment status) and educational level (low edu, mid edu, high edu) as well as their interactions. Is the below command correct? I obviously leave a reference category out for employment status and edu level as indep variables (ref categories: employed and high edu), should I do it for the interactions as well (employed and highedu not included in any interaction terms)?

ologit happiness unemployed housewife otherstatus lowedu midedu unemployed#lowedu unemployed#midedu housewife#lowedu housewife#midedu other#lowedu other#midedu if gender==2

As per interpretation, which one (or if neither what) is correct?

E.g regarding the coefficient for unemployed#lowedu interaction:

a) it is the impact of being unemployed AND low educated compared to all other employment statuses and educational levels on happiness level
b) if it is correct to keep employed and highedu out of the interaction terms as well, should I say being unemployed rather than employed and being low educated rather than highly educated is ...

Do you think I should follow up with "margins unemployed#lowedu unemployed#midedu housewife#lowedu housewife#midedu other#lowedu other#midedu" as well?

Thank you SO much in advance!

A.