There's no data to back up the assertion though. At least not in this article - all it says is that we attribute more positive attributes to women, not that women exhibit those attributes more often than men.