Hello friends,

I am trying to run a model about ethnic pay gap and use heckman selection method to adjust for selection bias. I am bit confused about the method of using labour force variable fore the model. I initially thought that I could drop observation which doesn't earn salary. Then it occurred to me that I can't run heckman selection if I drop those who are unemployed. Should I keep all the the observations which participate in labour force and drop those are not ?

If i do this, does stata automatically adjust the wage equation by ignoring missing values ?