Dear all,

I am testing the effect of a policy using a DID strategy, I get positive and significant effect with probit but I get positive and non-significant effect with margins, I want to know how can I exactly explain this in my paper, should I say for example, that there is an effect of the policy but we are unable to get the exact value of the effect?

With another variable I even get politive and significant effect with probit but negative non significant effect with margins. How should I interpret this?

I have read about this topic here: https://www.statalist.org/forums/for...eraction%C2%A0
and here https://www.statalist.org/forums/for...s-significance.
I understand that not the same hypotheses are being tested with probit and margins.

Many thanks.