It is a truth universally acknowledged…

The setting: An experiment using speed-dates to determine what people want in a first date. These are brief, four minute interactions, after which you write down whether or not you’d be willing to go on a date with that person.

The subjects: Columbia University grad students

The results:

Both men and women like attractive mates, but men show a slightly stronger preference.

Both like smarter mates, though it is more important to women. Women’s desire for a smarty-pants does not atrophy as he gets smarter than her. Men like women who are much smarter than them about the same as those who are just as smart as them. Both are preferred to women dumber than him. Using objective data (SAT scores) instead of subjective assessments, men don’t care at all about the intelligence conveyed in the SAT score. Women do.

Men like women who are about as ambitious as them, but don’t like them to be higher or lower in ambition. Women don’t care about ambition.

Women, unlike men, prefer men with higher incomes (proxied by the wealth of people in their home town zip code).

Women show a strong preference for men of their own race. Men show a slight preference.

As the number of potential dates go up, men like about the same percentage of available women. Women get more selective as the number of daters rises.

The paper: available here.

21 comments for “It is a truth universally acknowledged…

  1. January 10, 2006 at 8:57 am

    “Men of sense do not want silly wives.”–Mr. Knightley, Emma

  2. BS
    January 10, 2006 at 9:32 am


  3. ed
    January 10, 2006 at 10:35 am

    Pretty interesting. They really should mention in the abstract that the subjects are Columbia grad students, since this is a highly atypical population in many ways.

  4. A Nonny Mouse
    January 10, 2006 at 11:05 am

    I agree with ed. Also, I wonder if those preferences can be used as universal indicators, since this was a speed dating round. Maybe these are the things that men and women prefer in a short-term companion rather than a long term companion?

  5. Adam Greenwood
    January 10, 2006 at 11:36 am

    “since this was a speed dating round. Maybe these are the things that men and women prefer in a short-term companion rather than a long term companion?”

    But the whole point of speed-dating is to see if you want to follow up and try to get to know the person more. So even if other factors come to the fore in the long run, these are the initial filters.

  6. January 10, 2006 at 11:38 am

    I also agree with Ed. Grad students aren’t going to have large disparities of intelligence, or more especially prestige that accompanies their intelligence.
    As a female physics undergrad, men at church would often stop talking to me after they asked what my major was (They’d just say “oh, well, uh, see you later.” and walk off). And more annoyingly, of the 4 men who did like that I was smart(er than them) 3 of them didn’t really like that I was smart, they just liked that they could tell people they were dating a physics major. (If you’re wondering about the 4th man who liked me being smart, I married him.)

  7. Frank McIntyre
    January 10, 2006 at 11:46 am

    I agree, as do the authors, that the population is atypical. The same critique can and should be raised about a huge number of lab experiments that use college students. One’s expectation is that the gender stereotypes found here would be at least as large in the general population, but, then again, maybe not.

    As for the use of the “speed dating format”, I’d probably put it on par with the research on job interviews. It tells about the opening filters, as Adam put it, but it is not obvious how much it tells us about final outcomes. More work could be done, but this is one of the few pieces of quantitative work on the subject that I’d seen, so I thought I’d share.

  8. gst
    January 10, 2006 at 11:50 am

    Are physics majors smart?

    What I find surprising is the finding that “Women don’t care about ambition.”

  9. January 10, 2006 at 1:31 pm


    I would assume that women do not equate ambition with wealth (which I think you are doing). The women in the study were looking for someone who had already arrived more than they were looking for someone who might arrive at some unspecified future point in time.

  10. January 10, 2006 at 1:36 pm

    Let me put it this way, successful physics majors aren’t dumb.
    One normally assumes that a person picks a course of study they could be good at, which mean that most physics majors will have proved to themselves that they are of adequate intelligence.

  11. greenfrog
    January 10, 2006 at 2:39 pm

    Definitionally, successful anythings aren’t dumb within the area of their successes, right?

  12. greenfrog
    January 10, 2006 at 2:40 pm

    Oops — one occurs to me: the dumb lucky.

    But besides them…

  13. January 10, 2006 at 3:26 pm

    There are certainly significant differences in intelligence among graduate students. There probably isn’t anyone that’s dumb as a rock, but anyone in a given program that interacts with his or her peers will be able to tell you who’s bright and who isn’t. This might be harder to perceive in a four minute mini-date, but you’d better believe there are differences. There are also prestige issues, although I don’t know that you’d get that in four minutes except from the truly pretentious or the real wackos.

  14. gst
    January 10, 2006 at 4:40 pm

    Paul, I do not equate ambition with wealth. I can think of many examples, both historical figures and personal acquaintences, who are ambitious but also uninterested in wealth, and even some that are ambitious in ways which preclude or ignore relatively easy paths to wealth. Consider those that distinguish themselves (or seek to) in government or military service. I should have thought that women valued the trait of ambition, and not just because it may collaterally result in wealth.

    Also, as to your point about the responding women looking for men that have already arrived rather than those still gunning for their ambitions, I don’t think that ambition is a characteristic that goes away when one reaches a level of success. That is, the successful who reached their success through ambition are not all of a sudden no longer ambitious.

  15. Adam Greenwood
    January 10, 2006 at 4:43 pm

    I’m pretty surprised too, GST. It could be that my way of thinking about things is wrong or, far, far, far, more likely, the study is flawed. But only in that respect.

  16. January 10, 2006 at 4:47 pm

    I just love this stuff.

  17. Frank McIntyre
    January 10, 2006 at 5:10 pm

    If a variable is measured with (typical kinds of) error, this tends to cause it to look like it doesn’t matter even if it does. But that is a little tricky in this case, because the story is then that:

    1. Individuals perceive ambition to which they respond in wanting dates but they cannot or do not convey that perception when directly asked about ambition.

    2. This effect is, apparently, stronger for women than for men, and particularly strong for measures of ambition, as opposed to intellect.

    3. Since the results are the same when using the group averages, not even those can convey the perceived ambition.

    That is certainly possible. It’s a big world.

  18. El Jefe
    January 11, 2006 at 4:00 am

    I read somewhere that men make up their minds within the first 10 minutes of a date whether they’d care to date that person again. I don’t know what it is for women, but I suspect it is not quite that fast.

  19. January 11, 2006 at 10:55 am

    I usually have to spend at least 5-10 hours with a person before I know whether I’d want to date them at all. Maybe that number would go down if it was one-on-one, in a setting where all parties were open about their intention (i.e. of trying to determine whether they want to date each other at all,) there were lots of other people around to compare them to, and there was a structured set of information given about them (i.e., I didn’t know any of them casually before, but now I know their first and last names and maybe a bit more, like “they’re all fellow grad students” or “they’re all professionals.”)

    Still, I think I’d be saying either “yeah, sure” to nearly every candidate, or “no way” to nearly every candidate, largely on the basis of how many candidates there were to judge at one time (if it was a small group, I’d say yes to nearly all of them, and if it was a large group, I’d say no to nearly all of them.)

    And I know I’m a unique individual and all that, but I care about ambition a lot. It doesn’t have to be ambition for money, but I broke up with a guy whose main goal in life was to live till he was 35, so he could inherit the money his parents had been saving (they deliberately lived far below their means so he’d have a nice inheritance.) Oh, and he liked to watch cartoons. Even though he was going to be an engineering major… yeah. (from what I understand, he’s gotten a lot more focused since then; neither of us was 18 yet, so.)

    In a speed-dating situation, I’d probably be willing to accept SAT scores as a proxy for intelligence: in a large group I think I’d exclude anyone with a score lower than a) 500 on either section or b) 1100 overall (my own score was rather higher than that; I’m going for an “average to above average” cutoff.) In a small group, those numbers would go down to a) 400 and b) 1000 (roughly, assuming a normal distribution of scores within the group — if it was a large room full of Columbia grad students, they’d probably be higher.) In real life, I’d laugh at the idea of dating someone on the basis of 5-8 year old SAT scores, but speed-dating enhances the value of otherwise essentially useless pieces of information: the cologne used, the judgment shown in picking the clothes they’re wearing tonight, how much hair gel was applied, whether their shoes squeak…

    Speed dating overall puts a great deal of emphasis on the factors impacting a first impression. People end out making choices that they might not have, if they’d spent a few hours with each candidate instead. I don’t think I’ve yet dated a person I would have selected at a speed dating service meeting, and I doubt very much my ex-boyfriends would have picked me, either.

  20. Jay
    January 30, 2006 at 12:41 am

    The study is certainly interesting, but surely one would have to be careful about drawing too great conclusions from it, because of two rather obvious biases.

    The first has already been mentioned: Limited to grad students. It would surely be plausible to speculate, for example, that college grad students might put a higher premium on intelligence than the general population would. I’m reminded of a study I saw years ago on women’s attitudes about the relationship of career and family. And about halfway through they casually mentioned that they had done the study by questioning married female college students, and they had eliminated any from the study whose husbands were not also college students. No it would seem to me that there were many obvious potential biases here, like: Surely it is likely that women who go to college are more interested in careers than women who don’t go to college. By only counting women whose husbands were also college students, they eliminated older women who were relying on their husbands for financial support but decided to go to college for intellectual enrichment. Etc. How much of that applies here? Difficult to say. Surely college students are different from the general population in many ways.

    2. Doesn’t the very idea of asking people to decide who they want to date based on a 4-minute “interview” automatically bias the results to being superficial things? Frankly, I’m surprised that physical attractiveness didn’t come out as the overriding concern, because that would be easy to see in 4 minutes, while more complex things could not. It wasn’t clear to me from the description what information they gave the participants about the other person. They mentioned SAT scores. Did this mean they told the participants each other’s SAT scores, or that they used these scores to analyze results after the fact? My thought being, if you give someone a piece of paper listing five things about another person and then ask them to decide whether or not they would date that person and why, it would then be pretty lame to say that what was important to people in choosing potential dates is these five things. That’s what’s called a “setup”. Anyway, my original point being, if they just threw people together for 4 minutes each to discuss whatever they wanted, with no additional information, that would be less biased, but it would still limit the person’s conclusions to what they could learn about someone else in 4 minutes. Personally, before I married someone, I’d want to know things about them a lot deeper than what I could find out in a 4 minutes conversation.

  21. Frank McIntyre
    January 30, 2006 at 11:35 am

    Jay, FYI, they just used the SAT scores after the fact. I think all of the concerns you mention are going to come into play. That doesn’t mean the results aren’t interesting, just that they don’t answer all the questions we have about marriage and dating.

Comments are closed.