What you are doing is saying "No, I don't think the conclusion should be what the data says", and then you're going out and finding ("cherry picking") different data about related but not identical subjects that appears to contradict it.
Sorry, but the burden of proof is on you if you want to make a numeric argument here. And you're doing it with extraordinarily bad analysis.
Summary with a link to the research paper, which is really good science as demonstrated with experimentation and doesn't try to sell a narrative, BTW. Worth the read. http://www.news.cornell.edu/stories/2016/03/ilr-school-resea...
"New research by ILR School professors Francine Blau and Lawrence Kahn finds an eight percent gender wage gap that cannot be accounted for, even after controlling for observable variables that influence workers’ pay.
Gender discrimination in the workplace COULD be a cause, they suggest."
The data shows there is a difference in pay.
Underpaid means less pay for the same work.
That's not the same thing. Not even close.
In fact it is intentionally misleading.
Believe what you want. Just like climate change denial, there's enough uncertainty in these numbers to make any level of ostrich-headery justifiable. I just want to know why no one ever manages to show data that is "wrong" in the other direction...
You on the other hand are trying to draw conclusions that aren't backed by the data.