I am attempting a Difference in difference model with Probit/logit
Looking at the effect of a reform in education at the Junior Secondary school level in Ghana on employment outcomes, such as the probability of gaining employment in the top bracket of industries in Ghana,
the Treatment dummy would be whether or not the individual has completed Junior Secondary School (JSS)
the Pre/post reform dummy would be whether they studied before the reform was implemented or after,
the interaction term, therefore, will help determine the effect of those who have only completed JSS,

What kind of tests could I run in Stata to check if the model is accurate/robust?
How do you interpret the coefficient on the interactive term, regarding marginal effects?

Thank you,