Hi all,

I am trying to use a probit model to analyze a bunch of interaction effects for a binary dependent variable. My dependent variable is callback, taking the value of 1 if a subject receives a job interview, 0 if not. At the moment, I have 4 independent variables, three of which are binary: black (1 if black, 0 if not), woman (1 if woman, 0 if not), parent (1 if parent, 0 if not), and occupation (which can take 6 values). What I want to know is mainly the interaction effects; for example, I want to know what the probability of receiving a callback is for a woman with kids that applies for occupation 3. At the moment, I have the following code:

Code:
probit callback black##woman##parent##i.occupation
To get the marginal effects, I have entered the following code, for, respectively, the marginal effects at the means and the average marginal effects:

Code:
margins, dydx(*) atmeans
margins, dydx(*)
As I'm sure will be known to those who have a better understanding of it than me, the latter two commands don't give me the marginal effects for the interaction terms, as marginal effects don't exist for interaction terms in probit models (or in general even?).

I came across some literature and statalist posts advising on the use of the - inteff - command, but it's not quite clear to me how it works.

My questions are the following: (i) Is there any way of working around the problem so as to get a placeholder for marginal effects of interaction terms in a probit model? (ii) Is it advisable to switch to a linear probability model to compute the coefficients for the interaction terms?

I hope I provided enough information and I would really appreciate if someone could help me out, thanks!

Sam