Stereotype-Confirming Science

"Women 'choosier' over partners" reports the BBC. What's more, "Men look for beauty, while women go for wealth when it comes to assessing future partners, researchers say." Shocking stuff. Kay Steiger points out that these conclusions are based on . . . a study of "the behaviour of 46 people taking part in a speed-dating session." That's pathetic.

But it's also emblematic of what's wrong with so much research in this vein. Even the flimsiest of experimental results will get pursued and widely publicized if it just so happens to have the virtue of re-enforcing our traditional stereotypes about gender behavior, and then get swiftly pronounced as providing confirmation for "evolutionary theories in psychology." Now, there are so many stereotypes about gender difference that it's almost a mathematical certainty that some of them are grounded in reality. But the way proper science normally works is that it turns out to confound many of our expectations (heavy objects fall at the same speed as light ones; time changes when you speed up) while also explaining why it is that things seem to be the way they are (air resistance; you need to move really, really fast to notice it). So much research in this vein, however, is just incredibly sketchy and obviously designed to confirm what's already conventional wisdom.

At any rate, unlike a lot of my political fellow travelers I don't think this kind of inquiry into the evolutionary basis of human behavior and the potential to discover meaningful, innate differences in the average distribution of mental traits between men and women to be inherently wrongheaded or absurd, but I think people need to be much more careful about this stuff. To have an entire research program that seems dedicated to upholding old-timey folk wisdom is odd and an awful lot of the specific empirical research turns out to be incredibly hollow. I'd highly recommend David Buller's Adapting Minds: Evolutionary Psychology and the Persistent Quest for Human Nature for more in this vein.

Comments

So much research in this vein, however, is just incredibly sketchy and obviously designed to confirm what's already conventional wisdom.

This seems like a fairly shaky assertion--or do you know of any "meta-studies" that try to assess the correlation of research results with popular opinion? You seem to be angling toward a claim that unexpected results are somehow more likely to be valid than predictable results, which seems on its face ludicrous.

That said--I always roll my eyes when I read about study of this sort. It just seems like there are far too many variables involved for this kind of research to produce meaningful results.

Posted by James Gary | September 5, 2007 5:28 PM

In other news, women like to cuddle and men prefer to fall asleep after sex.

Posted by blah | September 5, 2007 5:29 PM

I think in general it's useful to read scientific papers before commenting on them, rather than relying on second-hand reports in the media. For all we know, the thrust of the paper may be methodological.

Posted by RSA | September 5, 2007 5:45 PM

I love how you just call some reviewed scientific paper "pathetic" just because, ..., well just because.

There is nothing unscientific about 46 test cases. What makes it scientific is how the data is treated and analyzed. I can't find the actual paper, but I'm guessing they used real math and all that fancy stuff scientists do.

I did find the researcher though: http://www.cogs.indiana.edu/pmtodd.html

Why don't you ask him why he does such pathetic work.

Posted by Mark | September 5, 2007 5:58 PM

It's generally true that the media distort scientific studies. However, it seems incredibly arrogant to want to shut down a whole area of scientific research because you don't like the political implications. I'm not sure whether you really read Buller's book, but his arguments with the official "Evolutionary Psychology" paradigm do not generally have to do with the "sexist stereotyping" issues you have in mind. In fact, Buller's view that men are daddyish (when they are) to attract women is probably more problematic from your perspective than the view that they are daddyish to promote the welfare of their descendents.

Finally, if you accuse people of making studies "designed" to confirm conventional wisdom, you are actually accusing them of sceintific misconduct, so you should be careful of this sort of thing.

Posted by Pithlord | September 5, 2007 6:00 PM

Don't ya just love it when Matt makes an obvious blunder so we can all pile on him?

Seriously, I love this blog because of Matt's willingness to consistently go out on a limb, even if it does occasionally result in posts like the one above.

Posted by James Gary | September 5, 2007 6:08 PM

Matt writes:

"Kay Steiger points out that these conclusions are based on . . . a study of "the behaviour of 46 people taking part in a speed-dating session." That's pathetic."

There are hundreds and hundreds of studies like this, virtually all saying the same thing.

Posted by Steve Sailer | September 5, 2007 6:15 PM

Matt writes:

"But the way proper science normally works is that it turns out to confound many of our expectations (heavy objects fall at the same speed as light ones; time changes when you speed up) while also explaining why it is that things seem to be the way they are (air resistance; you need to move really, really fast to notice it)."

No, the way proper science normally works is that it turns out to get closer and closer to accurately describing how the world actually works. It's hardly surprising that human beings down through the millennia came up with reasonably accurate descriptions of what people are looking for in the opposite sex, which modern science confirms.

Of course, this vein of research _has_ confounded many of the expectations of academics who bought hook, line, and sinker into the feminist ideology. So, what Matt is actually mad about is that science has confirmed popular wisdom and disproved elite delusions.

Posted by Steve Sailer | September 5, 2007 6:20 PM

Kay Steiger points out that these conclusions are based on . . . a study of "the behaviour of 46 people taking part in a speed-dating session." That's pathetic.

Why is that pathetic? Is there a specific reason why Matthew believes the particular method used here is invalid?

Yglesias sounds like Stephen Moore when he criticized the Lancet Iraq casualties study, saying "In their 2006 report, 'Mortality after the 2003 invasion of Iraq: a cross-sectional sample survey,' the Johns Hopkins team says it used 47 cluster points for their sample of 1,849 interviews. This is astonishing: I wouldn't survey a junior high school, no less an entire country, using only 47 cluster points."

As it turned out, 47 cluster points was fine. If Matthew has some basis for saying that a study 46 persons is invalid or insufficient, let's hear it. Otherwise, I'll chalk this up to Matthew writing off the cuff on a subject he knows nothing about.

Posted by Al | September 5, 2007 6:23 PM

Not, I admit, Yglesias's best work.

Posted by SomeCallMeTim | September 5, 2007 6:24 PM

how about Matt instead takes a look at this study saying the same thing, with, oh, a sample size of 22,000 people (is that statistically significant Matt?):

http://fmwww.bc.edu/ec-j/semf2006/Hort.pdf

Posted by Nathan | September 5, 2007 6:24 PM

One of the problems with our society's general science illiteracy (of which I, sadly, suffer to an extent) is the inability of the average person to evaluate the scientific rigor (and thus the merit) of any given study. Particularly when saying "studies show" tends to make people think something is iron-clad.

Posted by Freddie | September 5, 2007 6:25 PM

Jimmy Soul..over 40 years ago.

If you wanna be happy
For the rest of your life,
Never make a pretty woman your wife,
So from my personal point of view,
Get an ugly girl to marry you.

A pretty woman makes her husband look small
And very often causes his downfall.
As soon as he marries her
Then she starts to do
The things that will break his heart.
But if you make an ugly woman your wife,
You'll be happy for the rest of your life,
An ugly woman cooks her meals on time,
She'll always give you peace of mind.

Don't let your friends say
You have no taste,
Go ahead and marry anyway,
Though her face is ugly,
Her eyes don't match,
Take it from me she's a better catch.

Posted by Mudge | September 5, 2007 6:28 PM

[U]nlike a lot of my political fellow travelers I don't think this kind of inquiry into the relationship between global temperature and carbon dioxide emissions to be inherently wrongheaded or absurd, but I think people need to be much more careful about this stuff. To have an entire research program that seems dedicated to upholding enviro folk wisdom is odd and an awful lot of the specific empirical research turns out to be incredibly hollow.

Posted by Pithlord | September 5, 2007 6:29 PM

Somebody ought to a scientific study with a large sample size of Matt's posts to check out my hunch that most of the time when he says something really dumb like this (which is rare), he's trying to be nice to some young lady. Such as, in this case, this former American Prospect staffer:

http://kaysteiger.blogspot.com/2007/09/research-please.html

Posted by Steve Sailer | September 5, 2007 6:31 PM

The more I think about this post, the angrier I get. Yglesias admits that he is a co-thinker of people who think it is "inhernetly wrongheaded" and "absurd" even to investigate an empirical relationship. He questions the integrity of scientists who dare to study this stuff, and adds a thuggish warning that they need to be "careful" if they are going to say things inconvenient for his politics.

The existence of differences in mating strategies between the sexes is surely the single-most clearly attested to fact in animal biology. It makes more sense to deny evolution itself. The fact that the folk knowledge of every culture in the world --along with everyday experience -- also supports this notion is Yglesias's sole argument against it.

Yglesias is well below the Intelligent Design people in terms of connection to scientific reality. They don't deny micro-evolution or engage in threats.

Posted by Pithlord | September 5, 2007 6:38 PM

The problem with this type of news story is that the peer-reviewed article in question isn't actually available yet (I just went to the Proceedings of the National Academy of Sciences website and checked).

Evolutionary psychology (esp. that hyped in the popular press) needs to taken with a grain of salt in large part because we don't have data on the relative success rates of our predecessors who did have a particular trait versus those who did not. Evolutionary psychologists can only measure traits of the population that did survive, and then (all too frequently) make huge assumptions about causal processes claimed to underly their findings.

This too frequently results in very bad (but popular) science. Exhibit A: the story from just a few weeks ago that women Prefer pink because being better able to perceive red colors would help them find berries in the forest when men were out hunting. Fortunately, that article was actually available at the time it was being written about, so anyone with even a passing education in research methods could tear it apart. (Short version: Bad statistical inferences + failure to address more likely causes than that which was proposed = Bad social science.) More on that here: http://scienceblogs.com/mixingmemory/2007/08/do_women_have_an_evolved_prefe_1.php

In short, this new study's one small sample of German daters and unclear evidence for a causal pathway together leave me unimpressed about the generalizability of their findings. So, yeah, I think I'll wait to see the whole article before I take this new piece of cultural-stereotype-confirming article of evolutionary psychology seriously. But I won't be holding my breath.

Posted by KC | September 5, 2007 6:42 PM

Buller's critique of Evolutionary Psychology as codified by John Tooby and Leda Cosmides makes a couple of very good points (i.e., Buller agrees with me):

- When Tooby and Cosmides emphasize multiple mental modules, that's all very well and good, but they are going out of their way to ignore the vast amount of psychometric research over the last century on IQ, specifically the general factor of intelligence ("g").

- When Tooby and Cosmides emphasize the Era of Evolutionary Adaptation (EEA), that's all very fine, but they go out of their way to ignore the fact that humans have continued to evolve since they emerged from Africa 50,000 or more years ago, and have evolved in different directions under different selection pressures on different continents.

In summary, Evolutionary Psychology successfully snuck the study of sex differences onto politically correct college campuses, a historic feat of academic entrepreneurship (one that even as smooth an operator as Edward O. Wilson had failed at under his title "sociobiology.") The price of this great accomplishment, unfortunately, was that Evolutionary Psychologists claim to be the truest politically correct true believers on everything except sex, especially on race and IQ.

Posted by Steve Sailer | September 5, 2007 6:42 PM

To be fair, the study of revealed preferences of online daters linked to above by Hitsch et al. (http://fmwww.bc.edu/ec-j/semf2006/Hort.pdf) is in fact really good. I'm not opposed to finding one conclusion over another, just opposed to using lousy methodology and large assumptive leaps to arrive at them.

Posted by KC | September 5, 2007 6:51 PM

yeah, I posted that one cause it's considered a landmark study....I'm not aware of any serious rebuttals...it was simply too rigorous and utilized too much data.

there's a lot of simplistic shit being pedaled under the title of "evolutionary psychology"...on the other hand, there's a lot of serious work being done...

it's a simple fact that there are twice as many mothers represented in the human gene pool as fathers....it shouldn't be hard to figure out why or to deduce that varied mating strategies will result.

Posted by Nathan | September 5, 2007 7:02 PM

Here's a theory that should be investigated scientifically, although the sample size would have to be one:

That even the single most logical young male blogger's brain can be turned to mush by a pretty girl.

Posted by Steve Sailer | September 5, 2007 7:17 PM

I suppose the main flaw with a study like this is the tendency to view a cultural trait as an evolutionary one. There have been societies in which women owned and administered all household property, while men fought or hunted. In societies like that, a man's wealth is not a consideration. I guess you could make a "women nurture, men protect" argument out of that, but it is likely that a strong warrior type would have many of the attributes associated with male beauty (muscles, poise, cool scars, etc.). In that case, a similar study might conclude that women value beauty and men value wealth in prospective partners.

Posted by Kiril | September 5, 2007 7:26 PM

Kiril, Kiril,

to vastly oversimplify things, it's not wealth, but status. the two heavily overlap of course...but they're not identical.

and in the few tribal societies with female control and management of household assets (often matrilocal in scope), males who are good at hunting and killing amass multiple wives and large amounts of property...so they end up wealthy too (thus preserving their status with women in their old age)

Posted by Nathan | September 5, 2007 7:41 PM

But the way proper science normally works is that it turns out to confound many of our expectations (heavy objects fall at the same speed as light ones; time changes when you speed up) while also explaining why it is that things seem to be the way they are (air resistance; you need to move really, really fast to notice it)

What the fuck? Ever read any of that Thomas Kuhn stuff? Or did you just ignore it because your agenda writers told you it was a patriarchal view of science?

Sailer's description is much closer to "how science works" than yours.

And I agree with Pithlord, even if your comment was tongue in cheek, you should be embarrassed to associate with anti-science, political agenda shapes the results feminists. You are in fact closer to intelligent design than you are to the way progressives, liberal, and even conservative scientists think about science.

Posted by jerry | September 5, 2007 7:41 PM

Nathan,

I like the study you posted and reading it through quickly it seems sound. However, I didn't see any conclusion in there that these traits result from evolutionary forces. How about a simpler economic explanation: for whatever reason, women on average are still paid less for the same work. It may then be that to obtain the same standard of living, women are more likely than men to prioritize income as a desirable trait in a partner. Men, less socially advantaged in this way, are able to put greater emphasis on other characteristics in selecting a potential mate.

A couple of tests of this theory spring to mind. First, one could use the same data set to look for interactions between personal income and importance of income in a potential partner. If my explanation is correct, you might expect to see a lower difference between men and women in importance of potential-partner-income at higher levels of personal income. It is possible the authors already did this and found nothing, as I admittedly didn't read all of the footnotes. If not, it seems worth examining.

Second, I would be curious to see how lesbian and gay male online daters compare to heterosexual online daters examined in the study. If two women are equally disadvantaged by a gender gap in pay -- and assume potential partners to suffer the same social disadvantage -- then it is possible that income would be a smaller factor for women seeking women than for women seeking men.

If this is actually a sex-linked evolved trait rather than a function of cultural or economic forces, compelling evidence for me would come in the form of some genetic research rather than observation of present-day behavior.

BTW, for the other readers on this board -- why not try some constructive criticism for a change rather than attacking the character and motives of this blog's author?

Posted by KC | September 5, 2007 8:01 PM

BTW, for the other readers on this board -- why not try some constructive criticism for a change rather than attacking the character and motives of this blog's author?

Oh, because that doesn't work. See, he's a Harvard Man!

Posted by jerry | September 5, 2007 8:15 PM

I really enjoyed that post and generally agree!

At the end of the day - we all wanna fuck - man and women - that is nature's way. As humans we have vast potential for all kinds of expression (chimpanzee or bonoboo?).

When it comes to behaviour and what we know about genetics so far.. nature accounts for about 1/3 and the majority is nurture? Among apes one can already see differences in culture. Some chimpanzee groups tend to rape more than others and accounting for resources, etc. this seems to be a purely cultural phenomenon. Bonoboos where females have learned to work together (more so than human females) do not know rape at all. Men have to be charming to get it as females are also happy pleasing each other.

And I feel like quoting some Epicurus - 2300 years old:

Men usually fall on the side
on which they are wounded;
The blood flows in the direction
the blow comes from,
and straight at the enemy
if he stands in the way.
.
It is the same with a man
wounded by Venus’ arrows,
whether they come at him
from a girlish boy or from
a woman whose whole body hurls
love at him;

He runs at the person who shot
him and wants to copulate
and to plant in that body the
fluid from his own body;

His dumb desire suggests it
will give him pleasure.
That is Venus for you, it is
that which we call love
which is the source of sweetness
which Venus pours drop by drop
in our hearts: and then we are
worried.

Posted by Hugo Pottisch | September 5, 2007 8:32 PM

Matt did say some very silly things in his post but those of you defending the sample size in this study are being silly.

Lots of psych studies use very small and homogenous N's, but that's because they are either:

a) interested in very basic phenomenon that we're confident is generalizable (e.g. low level visual processing) or

b) isn't interested in making strong claims about how well the results generalize to the larger population

A study on dating preferences meant to generalize to the entire species is about as far away from either of those as is possible. The results may be true, but it's still a very silly study.

I'm sure it was just a poster at the conference, and those are typically barely reviewed.

Posted by orangatan | September 5, 2007 9:21 PM

Lots of evolutionary psychology studies are done on UC Santa Barbara students.

But, there are two things to keep in mind:

- There are hundreds and hundreds of such studies, all pointing in the same directions.

- The feminists don't have much of any data at all. Feminist science is an oxymoron.

Posted by Steve Sailer | September 5, 2007 10:09 PM

The really interesting question here is whether a young fellow will get farther with a pretty girl by pretending to take her dumb feminist emotional non-ideas seriously or by showing some masculine self-respect.

Posted by Steve Sailer | September 5, 2007 10:14 PM

Sailer might be on to something here:

Somebody ought to a scientific study with a large sample size of Matt's posts to check out my hunch that most of the time when he says something really dumb like this (which is rare), he's trying to be nice to some young lady.

But this raises a question: does anyone have a photo of Megan McArdle? I don't know what she looks like, but if she easy on the eyes, Matt might be more tolerant of her opinions, no? Or is Matt only willing to indulge arguments he thinks are flimsy if they are related to an unimportant subject matter (e.g., the self-styled feminist musings of twenty-something white girls)?

Posted by Juan | September 5, 2007 10:25 PM

The study represents perhaps what people who go in for speed dating do. How to generalize outside that group is beyond me.

Posted by Arun | September 5, 2007 10:36 PM

Steve.

We get it. You think that feminists are dumb and Matt linked to the story because Kay Steiger is attractive. You've made your point, multiple times over. Now move on. Please. For your own sake -- you're starting to embarrass yourself. After all, this is only a blog comment thread and you've clearly gotten waaaay too invested in trashing the author.

Posted by KC | September 5, 2007 10:38 PM

KC,

Saying that Matt might relax his standards of intellectual rigor a little when passing along a fairly weak argument by a pretty girl is hardly trashing him. He is a young, heterosexual man, after all. It's clear from Sailer's comments that he considers these feminist posts, and the naive posts about the effects of liberalism in decades before Yglesias was born to be low-quality exceptions; if Sailer didn't respect the quality of Matt's other writing (i.e., the bulk of it), I doubt he would spend so much time here.

Posted by Juan | September 5, 2007 10:55 PM

those of you defending the sample size in this study are being silly.

The point isn't that small samples size don't have problems, of course they do. The point is small sample sizes have known problems that, depending on the results, can be dealt with.

For example, to the first approximation a sample size is good to plus or minus the square root of the size. A sample of 46 is good to plus or minus 7 or about 15%. Thus, if the results were 7 more people made one choice than another, the results are meaningless. However, if 40 people went one way and only 6 the other, the results have real statistical support.

Matt is acting like a ID supporter with this post and it really reflects poorly on his judgment. Matt is one of my favorite bloggers and I generally agree with him, but this is just stupid. I'm glad so many have piled up on him tonight; he deserves it. Hopefully, he learns.

Posted by Mark | September 5, 2007 11:46 PM

Wait, did Al take the side of the Lancet study against its critics? How did this happen? Will he be required to surrender his hack license?

More on topic, it's amazing that Matt thinks there's some societal urge that encourages reporting of studies which confirm what everyone believes, while somehow sweeping counterintuitive results under the rug. To the contrary, it's generally the counterintuitive stuff that sucks up all the oxygen.

If you read a hundred stories about studies that conclude X and only one story about a study that concluded Y, it might be because the media is hostile to studies that say Y... or it might just be because virtually no studies ever conclude Y. Does Matt really believe that if a study concluded that men don't really care about women's looks after all, that study would get no attention?

Posted by Steve | September 6, 2007 2:54 AM

Further, to show any connection with evolution, you'd have to show that said behavior correlates with having more surviving offspring.

If physical beauty was a strong selection criterion then the human race would look a lot better than it does.

Posted by Arun | September 6, 2007 7:15 AM

if Sailer didn't respect the quality of Matt's other writing (i.e., the bulk of it), I doubt he would spend so much time here.

If you know him, that may be a legitimate conclusion. Otherwise, it could be S wants to address the same audience that MY does.

Posted by Arun | September 6, 2007 7:18 AM

Re: "If physical beauty was a strong selection criterion then the human race would look a lot better than it does."

Isn't that sort of like saying, if yellowness was really an indicator of bananas being ripe, then they would be more yellow than they already are?
I mean, exactly how much more yellow could a banana be?

Posted by Jim W | September 6, 2007 9:15 AM

I second the idea that science, ideally is not out to prove any agenda, especially a political one (even a "progressive" political one). While this is often not the case, either overtly or covertly, that's still the goal.

I do get a little squirmy when considering how many social science experiments seem to be based on studying college students, who, while they may be plentiful and accessible as subjects for university based researchers, really don't represent a very wide variety of people. One wouldn't want to draw hard and fast conclusions about the general population based on such research, but it seems to happen too often.

Still, most stereotypes got to be the way they are for a reason. The people who tend to complain most loudly about stereotypes are exceptions. So, again, if research disproves a cherished theory it's fair to ask for a "do-over", but if the evidence continues to mount, perhaps an adjustment of thinking is needed. The data, after all, is the data, like it or not, whether it's that the earth revolves around the sun, or that hormone treatment in post-menopausal women can increase the very health risks they were thought to decrease.

Posted by JMS | September 6, 2007 10:10 AM

Aww, I was all set to jump on Yglesias' comment suggesting that good science is inherently revolutionary. But everyone else already did! I don't have anything new to add! You guys are taking the fun out of the internet for me.

Posted by Patrick | September 6, 2007 11:52 AM

actually, the best book on how the media jumps on all these gender stereotype-confirming studies is Carol Tavris' "The Mismeasure of Woman". I highly recommend it.

Posted by Dilan Esper | September 6, 2007 12:37 PM

Mark - statistics can't fix problems with your sample itself, and it's not just a numbers issue.

If you tested 10,000 people and 9,999 all did the same thing you'd have uber significance on any test you cared to run. But it wouldn't mean anything if you hand picked those 10,000 in a non-random way that was biased in favor of the outcome you were hoping for.

46 single, American, speed-dating friendly people just aren't ever going to tell you anything meaningful about somethings as complex as "attraction" on an evolutionary scale. Statistics aren't magic.

On another note, the complaints about studies using undergrads are typically unfounded for the reasons mentioned in my above post. But when you're specifically hoping to study things where it IS worth worrying about, well, you need to worry about it.

On still another note, give the "feminists are dumb" bit a rest Steven. It gets old fast and hurts your arguments.

Posted by orangatan | September 6, 2007 1:10 PM

The odd thing is the way the BBC spun the results. When I read the text, the women come across as more shallow than selective.

Posted by AlanC9 | September 6, 2007 1:50 PM

Sailer meet Juan. Juan meet Sailer.

Oh? You already met in the bathroom mirror this morning?

Nevermind then.

Posted by AwComeOn | September 6, 2007 2:32 PM

Re: "If physical beauty was a strong selection criterion then the human race would look a lot better than it does."

Similarly, if linguistic ability was a strong selection criterion, then people would talk a lot more than they do.

Posted by Jim W | September 6, 2007 3:52 PM

Given the wide variety in humans, the only conclusions that one can draw from this study is that this is the data for these 46 people....at this time.

Plus the fact that they were all from a particular college that people VOLUNTARILY decided to attend and a particular age group. (can we say self-selected?)

And you want to draw grandious conclusions from this about all male-female interactions? Hello?

Go repeat the experiment 10,000 times around the world, with 10,000 different source groups (different ages, different locations, different religions, different cultures.) Show me that you get the same results in all cases. THEN I'll believe you might have something. Otherwise, your error bars are running the length of the page.

Posted by grumpy realist | September 6, 2007 4:01 PM

Grumpy realist,

If you study statistics, you might be surprised about how large a sample size needs to be to accurately measure the tendencies of a much larger population. An accurate sample size is often much smaller than you might expect.

Posted by Fred | September 6, 2007 4:15 PM

I agree with Matt that it sounds sketchy. Picking the 46 people who showed up at a particular speed-dating session certainly isn't a random sample. In fact, a lot of the subjects would seem likely to be ... single and speed-dating. Maybe such people are much more likely to be shallow. Who knows? (without doing an actual study ...) Anyway, I doubt such data would be accepted into a very competitive journal. If it's in a "proceedings" there's a wide range of quality possible.

But it's also emblematic of what's wrong with so much research in this vein.

Not just the research of course, but the journalism. I know it's too much to hope for that people will ignore these things or report them intelligently, but it's worth saying. Moreover, others are right to point out that Matt ought to wait for the actual paper. On the surface, yeah, it sounds like they've chosen things you can't really measure, and their data looks weak, but certainly it's meaningful in some small way. Of course, when you're relying on both a psychologist and a journalist understanding statistics, things can get depressing.

Posted by Mike | September 6, 2007 4:50 PM

Coming into this late, but ...

This study is essentially about snap judgements (as opposed to considered decisions).

There's a ton o' evidence that women make snap judgments about partner suitability according to riches (resources) and that (straight) mens' snap judgements are all about youthful and fertile good looks (not to mention realistic chances of scoring).

There isn't much research (AFAIK) on just exactly how people end up with the long term partners they do (when choice is a factor). In other words, there's not so much research (that I know of) on what factors actually influence people in the choice of long term partners.

Just looking around, it's pretty obvious that lots of women choose guys without very much in the way of conspicuous resources and lots of men end up with women who are not on the A-list in terms of fertile-youth signals.

IME I've known lots of guys who choose women who aren't the most attractive they get but who they find interesting. To paraphrase a sentence I've heard a lot "I'm never bored around her" is every bit as potent an attractor of long-term male attention as big boobs and a narrow waist.

Posted by Michael Farris | September 6, 2007 5:21 PM

The best part of this study is reading the guys in this comment thread who insist that THEY aren't shallow like this study suggests, THEY find women who are witty and politically active and good writers attractive.

Now lets think about this. This is a blog where young, snarky politically active people hang out and write to one another. These can probably be characterized as the "interests" of the people on this blog.

Meanwhile, the study suggests that people SAY they are interested in persons with shared interests, but in REALITY men actually select on looks.

You can almost see the guys writing these things glancing sideways to see if any women are listening, and to check out if they're hot.

Posted by Patrick | September 6, 2007 10:48 PM

I don't know what Yglesias ever did to attract the creepy crush of a third rate Charles Murray wannabe like Sailer or his little friends but I'm sure he regrets it.

Posted by rory | September 7, 2007 11:27 AM