An indication that it is cultural is from looking at other cultures:
If you look globally, there are countries where that isn’t the image, and in fact, their numbers are dramatically better. I was recently speaking with some of our Oracle engineers from China and they pretty much have a fifty-fifty split of men and women. And they think it’s sort of odd that we don’t.
That's interesting, although I'm not inclined to take hearsay of hearsay at face value. It'd be great to see a report with some hard numbers proving that this is the case, and examining the cultural differences that could explain why their numbers are so drastically better than numbers elsewhere in the world.
Also, from a scientific perspective, this is a disastrously imperfect experiment. You cannot rule out the effect of the genes unless you control for that. Looking at a group of people who are physically different AND culturally different , then arbitrarily concluding that any differences are the result of cultures is about as unscientific as it gets.
I actually agree with your first statement - I saw that a few days ago, and I'm keeping it in mind provisionally.
As to the second point, humans are not very genetically diverse. Of course there are enough differences that we can sometimes conclude that a particular person's genes come from an ethnic group, but it's not as nearly as varied as even other primates. Basically, we hit a genetic bottleneck sometime in the past 100,000 years. So it's far more likely to be a result of culture than genes.
You're right about the lack of genetic diversity, but even that limited amount of diversity can result in drastically different personalities, tendencies, etc. For some truly shocking material on just how much of a role genes can play, look into the studies done on identical twins raised apart vs adoptive siblings raised together.