Bad friends lead to lower academic performance. I wouldn't be shocked if people tend to make better friends (and much less commonly deal with the idea of being more than friends, which ruins friendships more often than not) within their own gender.

I prefer hanging around with girls, but that's because girls are much easier to have an interesting/serious/useful/fun conversation with. With guys, an hour is a lot, with girls I can talk all day long when they know they can trust me, and that's mutual. There's no romance involved for me as far as I can tell.

I suppose what I'm trying to say is that a) it doesn't have to be because of romance and b) the reason why girls would get my grades down is because they're too fun to be around, making me not want to study instead. It would be because I consider them better friends, not worse.

No romance involved for me in most cases, anyway :-)

I had low grades in high school and mostly female friends. So there's a data point. In fact I haven't made a really good friend with a guy since I was like 14. I find it much easier to get along with women.

I just think its a curious piece of information. Girls who hang with girls get better grades than girls who hang with guys. Guys who hang with guys get better grades than guys who hang with girls. This has nothing to do with the overall composition of the class (so not all girls school vs mixed) but I think its interesting nevertheless.

Also, a pretty good economics article to read.

Also, I never hung out with girls in high school (not by my own choice, I'm ugly af), and I had a 3.75 cumulative...

I totally agree with this article. My gpa isn't bad but it's low compared to others at my school and my friends consisted of like 80% girls.

I had a mixed group of friends in high school and my gpa was above average. I think a lot of my friends now are the opposite gender and my grades are still above average... though only very slightly.

all you youngin's. If you all only knew the truth, it would set you free

Here's another theory:

You spent more time on social skills because the opposite gender does demand a bit more of that, meaning you're not so occupied with learning (skills).

I suppose there's also more tension between boys and girls. Almost certainly..

@bo: Is that low or high?

My grades aren't low btw, but that doesn't neccessarily discredit the theory.

That's high. It's also unweighted.

All these things are measured in terms of standard deviation for a reason, guys :) Becoming friends with Jessica won't make you stupider.

It mentions some interesting things though as to why this might be the case. One of which is that academically unsuccessful people might surround themselves with members of the other sex to make them appear more popular/compensate.

So you noticed I didn't read the article, huh? :P

I understand standard deviation, but even though my personal experience deviates, I can still see how this result fits into my experience.

Seemed like an analyisis worth mentioning.

You think this might have an introversion/extroversion correlation? Extroverts being more socially inclined, more likely to have opposite gender friends, and often not as academically rigorous as introverts?

So... Be homosexual, you'll get better grades!

...

Yeah. You know what my pet hate is? Statistical "studies" that conflate correlation and causation. That is annoying as heck.

That is stats 101 level shit - if they're getting that wrong, I'm just going to assume the paper is complete bogus rubbish.

Their data collection methodology is probably fine, the correlations are probably real, but their conclusions that the gender balance is causative of the grade difference is ridiculous. Zero proof of that whatsoever. Causation could well be in the other direction, or from other factors which influence the correlation.

Which is it? Zero evidence presented.

Before I respond - thorfi did you read the article?

Let me rephrase that. Did you read the *peer-reviewed journal article from one of the most respected economic journals out there*

Yeah, I read it. At no point do they appear to actually prove that the link is causal.

And, to be clear, my problem with that is that they *simply say it is causal in that direction*. I'm not doubting that their data is probably quite real, and that the correlation is real. But they really do several times make the statement that the causality is in a particular direction, when there is no proof of that being the actual causal direction at all.

*pokes paper again* Hrm. "These estimates have a causal interpretation because the gender composition of same-school neighbors is considered exogenous."

... Ah. So the assumption is, because neighbour gender composition is "external", it must be the causal factor? Dodgy assumption.

Ah, okay. Just making sure you weren't like steephie :)

Been a while since I took stats, but I think this is the basis of their underlying argument

" Students with more opposite gender schoolmates in their close neighborhoods have more opposite gender school friends, and, given that the gender composition of schoolmates in a student’s neighborhood is essentially random, provides plausibly exogenous variation in the share of opposite gender friends from which a causal effect on the outcomes of interest can be estimated."

I haven't read the article but I'll speculate on an idea... you have the choice of people skills or brainy skills but not both... if you have people skills then you are more likely to "make" friends of the opposite gender and you don't have the skills to get good grades. If you have the brainy skills then you are more likely to "perform" better in school. So here is my question is having friends (of the opposite gender) the cause of lower grades or is a lack of friends (of the opposite gender) the effect of having higher grades?

Yeah, I would think that so long as its an exogenous variable that moves with an endogenous variable, the causality there is one way - especially if the exogenous variable is random. I don't think that's too dodgy an assumption.

@CB - the article says that in large part having friends of the opposite gender is a factor of who lives closest to you in your neighborhood. Are the kids in the neighborhood mostly girls or mostly guys?

And they take the popularity bit out of it by limiting the number of people kids could list as friends

@goldfinger: Yeah, I think that's the dodgy assumption. For one, despite the *neighbour* composition being "exogenous", that doesn't prove at all that the *friend* relation is exogenous.

I think it's a very interesting correlation - don't get me wrong. :-) And the most likely truth is that there are causal links in *both* directions of that correlation too, as well as other factors that influence the correlation.

But I really don't think that paper proves the causality well enough to actually be making that solid a claim.

Well, I don't know enough about stats to argue against you. So this is just another reason why its always a good reason to have mathematicians check economists....though PhD economists are essentially PhD mathematicians.

Yeah, economists tend to do a lot of model simplification, though. The physics classical example is "assume a spherical cow", but economists tend to often just assume lots of those sorts of things and not even point out they're assumptions.

To be fair, at least in this case they did point out that that is an assumption, even if it was buried in a one liner deep in the midst of a block of text.

It's entirely possible I'm not correct about the causality level implied - I'm not a professional statistician. But nevertheless I don't much like the look of that particular assumption. :-)

I'll buy it. Girls don't talk to me, and I'm a fucking genius.

How many times does this have to be reiterated??? Judge people on an individual basis, and outside correlations will have no substance to them.

@JY - what on earth does that mean?

If you look at a picture in the right way, the picture will be more than what the painter intended of it. You will pour more time into something with less meaning in it, and the ecess goes into a void. You don't find a place for yourself, you lose yourself.

Wars are fought by those who are either mad enough or desperate enough to fight in them, but they never start the wars. Those who start the wars are more mad and desperate but in a way different from any other. The people you look to in war are those who seek the end, for they are only desperate.

Damn it I can't find any more quotes, tell me what 2 movies those are collectively from and i'll give you 100 +1s

Weekend at Bernie's. Both of them.

Silent Night Deadly Night 2