I suspect it's accurate but tells us less than one might think. In that sense it may be a "bad thing" if the conclusions are taken at face value and applied out of context without being interrogated.

i.e.: saying "this shows people are predisposed to favour women, ergo the narrative of gender discrimination against women is disproven."

In actuality, this shows favourable sentiment toward women in quite narrow, specific ways: words quoted from the article are "good, happy, and sunshine" & not things like "competent, reliable, trustworthy, and cool". There's a hint of infantilisation within the words used. Yes they're positive, but it's worth looking at how those specific sentiments might exhibit in real world situations.

Try applying this to any other group. Truth is no defense for "bias".

Deers and lions?

Depends on your average. If you take mean then male averages are always skewed by outliers. The mean number of people murdered is going to be higher for men than for women. But the median and mode for both will be 0. Also note that such statistics have a lower limit (0) but no upper limit. It might be that there are men who are overwhelmingly "good" and prevent many murders, but this won't count as a negative contribution to male murder statistics.

> Is this a bad thing if it's accurate?

Yes. That's exactly how discrimination works. Stereotypes reinforce themselves through the ways we treat people.

> That's exactly how discrimination works.

Discrimination means treating things differently because they're different. You seem to be assuming that all discrimination is bad; but the "bad discrimination" is discriminating based on irrelevant differences.

> Stereotypes reinforce themselves through the ways we treat people.

Whatever; I don't understand the sentence. Perhaps I should have studied sociology! But a "stereotype" is like a cardboard cutout; it's a super-simplified (ands therefore false) representation of a difference. I don't know what stereotypes have to do with discrimination, unless somehow we're suddenly talking about racial discrimination, or gender discrimination, or some other unjustified discrimination.

People are individuals, not averages of everyone. To assume someone is a statistical extrapolation of other people is stereotyping, regardless of how frequently it is correct.

Not on average, but I believe men do have a flatter Gaussian curve. We have more mass murderers but also more notable leaders, scientists, etc.

At any rate, this is interesting in the large but useless at the individual level. We should not allow fuzzy impressions over the entire human race change our impressions of an individual person.

When the book "The Bell Curve" came out it caused an uproar for this reason. We cannot discriminate against anyone on the basis of their broad studies of race, as the authors themselves noted.

The flatter Gaussian curve has also been observed at school results.

Both genders have a similar average, but you will find more extreme results in make students (very weak or very strong).

And the flatter curve exists because males are essentially the experimental units of a species.

Because eggs are scarce but sperm is abundant, males are largely disposable in the context of a species' or society's collective survival. Therefore, it is advantageous for males to engage in high risk, high reward behavior to keep prodding the environment they live in for changes that might require an adaptation from the group. The result is a flatter curve with more extreme failures and more extreme successes.

Women are not better behaved on average, they're simply not prosecuted equally for their poor behaviour.

To get a full understanding of this, compare self reported spousal abuse data against police reported data. The difference is shocking.

My ex-wife used to physically abuse me (hitting, throwing things). An attorney friend told me to leave the house whenever that happened, because if anyone called the cops (including me), the police would arrest me instead of her. It's called the "Duluth Model" of domestic violence enforcement.

Here where I live, police are instructed to use a gendered lens in their responses, specifically to treat everything as though women are the victims, always, and never the aggressors.

It's grotesquely sexist, but lauded by the NGOs.

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.

What does it mean to be "well-behaved"?

Heh, another example of the bias in action. I'm sure people tend to more easily forget instances where a woman was not well-behaved. Take a look at /r/PublicFreakout. Also note the Karen meme is not about a man.

Also, similar to bullying in childhood, women's bad behavior often doesn't manifest in physical displays of aggression, but psychologically manipulative and emotional ones. Obviously men can be psychologically manipulative too, yada yada... I'm not a woman hater.