That is a plausible interpretation.
My priors were that, for similar levels of maturity girls collectively do better on classwork and boys collectively do better on tests, that girls mature faster at young ages, and that the bulk of grades come from classwork in younger grades. So I would expect girls collectively to be routinely outperforming boys collectively at the 5th grade level. The statement that there are no differences between girls and boys thus struck me out of left field- the only explanation I could come up with was sex sensitivity (if girls are doing worse, there are differences, if boys are doing worse, there are no differences).
Thinking about it again, I came up with another interpretation: the "these" in front of "boys and girls" is referring to the bright ones. So bright girls routinely outperform average boys, just like bright boys outperform average boys. But that seems like a sloppy way to compare distribution tails.
The Trouble with Bright Girls (article @ the Huffington Post)
Excerpt:
The topic of this article seems to relate to several common Less Wrong issues: the nature of human intelligence, and the gender imbalance among LW readers.
I'm not sure how much credence I give to the proposed explanation of the difference in mindsets. It may well have to do with socialization and feedback, but the specific description of feedback that is presented seems a bit too much of a "just-so story" to me. The difference itself is fascinating, though, and I hope more is done to further our understanding of it.