> I would want to see this experiment repeated trans females who (assuming the presented hypothesis) should see the inverse relationship.

Yep, there are ancedotes about trans women just as you predict:

> Joan Roughgarden is a biologist at Stanford who lived and worked as Jonathan Roughgarden until her early fifties, and her experience was almost the mirror image of Barres’s. In her words, “men are assumed to be competent until proven otherwise, whereas a woman is assumed to be incompetent until she proves otherwise.” In an interview, Roughgarden also noted that if she questioned a mathematical idea, people assumed it was because she didn’t understand it.

http://www.newrepublic.com/article/119239/transgender-people...

> I suspect a better (and likely cheaper/easier) test is to take the same papers and present them under a male and female author name.

There have been published scientific papers looking at (sort of) this. Send out job applications for a lab manager, and randomly assign a male or female name. Same job application otherwise. Male applications get more job offers and more money.

http://blogs.scientificamerican.com/unofficial-prognosis/201... http://www.pnas.org/content/109/41/16474

We might give out to Creationists who don't want to accept reality about evolution, but well...