I understand what you are saying but still think non-gender varables are probably dominant in the process. Don't get me wrong here I actually do subscribe to the idea that personality is a LOT more genetically determined than we assume. Unfortunately as I said before, there are no real ways of measuring this at the moment so the whole field remains an area of speculation.

Also, as with any population study, it's almost impossible to apply it to an individual so the usual caveats apply when interpretting the research.

As you very astutely point out, I suspect the main strength of the article is in it's ease of misrepresentation so people can have sensational copy for their crappy magazines. The issue of irreponsible journalism is a seperate one ( don't get me started! ).