I proposed means for controlling for that variable.
[ol]
[li]Compare for productivity within a specialty and within an organization on a productivity basis. Pediatrician to pediatrician within an organization on a total salary and benefits compared to unit of gross revenue billed basis. (Or any other specialty for that matter.) I suspect that the average male pediatrician takes home more per year but is paid the same per unit of gross charges (ie productivity basis). But it is testable and I may be wrong.[/li][li]Get some historic payment figures for fields that have changed from being male dominated to female dominated over the past several decades (Pediatrics and OB/GYNE are the two big ones in medicine) and see if payment schedules for the same codes have changed at a different rate than for those specialties that have stayed male dominated over that period of time.[/li][/ol]Now again, my proposed studies are limited in scope and as I am familiar with medicine that profession has supplied my examples. I suspect that other individuals in other fields can think of ways to test the hypotheses within their own disciplines as well. One would of course generalize the results of such studies to all professions with great caution and with many caveats.
Separately I like to take issue with another point that the op made: the claim that it would be illogical for an employer to not hire disproportionately more women if indeed a woman of the same productivity as a man in that position was paid less on average. It would not necessarily be illogical (even if it would be wrong). An employer is looking forward not knowing how productive an prospective or new employee is going to be. Statistically it is true that a new 25 to 30 yo woman employee is going to be at greater risk of having less productivity over the next seven or so years of employment than a man is, is at greater risk of having the employer’s initial training investment fail to pay off as handily, as the women is at greater “risk” of wanting to balance work and (?future?) family than the man is. Whether or not that turns out to be true later for the individual the employer may be rationally deciding that the training investment (which leads to advancement) is more wisely spent on the greater possibility of more full pay-off. This is especially true in a field in which the females employees are more likely (statistically) to become part-timers - costing full benefits such as healthcare with less production.
It is of course by definition a pre-judgment (ie prejudice) based on gender (and one that I believe should be illegal) but it is not necessarily an irrational prejudice.
If you think that women will support changes in law or practice that will punish women just as much as men for taking time to care for their families instead of their careers, you are living in la la land. Most women’s idea of advancing women’s rights is that they should have more flexibility AND get the same pay and advancement as men do. The best case is that everybody gets more flexibility and it all comes out of the hide of the employer, or we all just get less productive and get paid less. Because there is no escaping it. When I stay in the office and do work that should have been done by a woman who left to pick up her kid from day care, I am more productive than she is, as far as my employer is concerned. And that is the only productivity my employer should be concerned with.
There is a social contract, where men work more outside the home, and women work more inside the home. Neither men nor women want to change this, by and large, except for a tiny minority. This is the feminist paradox. I can see the problem very acutely, because my wife would like to work more, and I would like to be with my daughter more. But the damage to her career from the opposite arrangement, is tiny compared to what my career would suffer if we switched roles.
But I don’t see how you can fix this legislatively, except by giving men more rights in the workplace, not women.