A common trend among bad writers seems to be to assume that any measurable sociological phenomenon is somehow inherent to the beings themselves.
An example: If you were to take 10 men, and 10 women, and isolate them, then painfully beat the 10 men daily, for a period of a year, you could bring someone in to do a study, and they would say "ahh, yes. it seems that men are definitely more timid and fearful. this muts be by their very nature".
In this example, the differences in behavior of the two groups are not because of the inherent differences between men and women - it is because of how society treats them differently. In this way, it is very difficult to determine anything inherent about the behavior of any group of people who are in a single society - the affects on that society can be much more subtle than the case provided here. The most powerful statement one can make in these sorts of stuide is only one of disproof - if enough exceptions exists, it may be a reasonable thing to assume that whatever behavioral delta being discusses is _not_ inherent but it societal.