"Alternatively, the results could be a statistical fluke, resulting from a small sample. 'This is very common among these kinds of studies because the genetic effects on cognition are so tiny that you need perhaps thousands of people to get a good estimate of the effect,' she says."

That was very good of the reporter to mention this possibility. That kind of issue always has to be evaluated when looking at new scientific studies.

http://norvig.com/experiment-design.html

It's an intriguing finding that should probably be looked at further in a different cultural context (is taking the SAT in America like taking the Taiwan college entrance exam in Taiwan?) and definitely should be looked at with a larger sample.