Hello!

I am trying to perform a Blinder Oaxaca Decomposition on the log of hourly wages for the gender wage gap.

The regression I have is:

Code:
 reg lghourpay female age yrsemployer married fulltime degree alevel Q2 Q3 Q4
Currently I have this code for the decomposition:

Code:
 oaxaca lghourpay  yrsemployer married fulltime degree alevel Q2 Q3 Q4 if inrange(SEX, 1, 2), by(SEX) detail pooled
My question is: I understand that one drawback of this decomposition is the index number problem, so the choice of the reference groups will affect the results of the decomposition. I want to go with the most standardised method in the literature by reporting results based on the male wage composition. Does my code account for this?

Hope that makes sense – let me know if not!

Thank you!

(Using StataIC 16)