Dear Statalisters,
I am working in a random effects probit model in which I want to test whether scoring high on an index (high=1, low=0) changes the effect of being in either treatment 1 or 2. For this I was thinking to use an interaction term. I was trying to use the inteff command proposed by Norton and colleagues, which unfortunately does not work for me. Is there another way or a way at all to test the significance of this interaction in a random effects probit model?
Thank you,
Eva Straka
I am working in a random effects probit model in which I want to test whether scoring high on an index (high=1, low=0) changes the effect of being in either treatment 1 or 2. For this I was thinking to use an interaction term. I was trying to use the inteff command proposed by Norton and colleagues, which unfortunately does not work for me. Is there another way or a way at all to test the significance of this interaction in a random effects probit model?
Thank you,
Eva Straka
