This article from the New York Times discusses several online dating services that try to use scientific methods to help people find the best long-term romantic partner. eHarmony uses a lengthy personality questionnaire designed by a psychologist. Chemistry.com (a spinoff from Match.com) and Perfectmatch.com each use an algorithm designed by an anthropologist and a sociologist, respectively. You could consider the use of these systems by large numbers of people as something of an experiment; the algorithms have not been published or peer-reviewed, but the companies involved must have some data about how well they work, and evidently academic researchers are interested in that. Helen Fisher, the anthropologist who developed the system used by Chemistry.com, is hoping to publish not only information about how the algorithm works but data from Chemistry.com users that will validate the algorithm, which will be useful.
Personally I’m skeptical about all of the attempts to come up with a widely applicable, science-based method for helping people find a mate. I believe very strongly in science, of course, and I’m fascinated by the ways scientific research can be applied to the process of mate selection. However, I don’t think we’re anywhere near far enough along in understanding personality and relationships to be able to devise an algorithm that will work across the board. I wonder if people are selling these mate selection systems in part because it’s a huge (and somewhat vulnerable) market to be tapped. It’s like any other product that caters to a widely felt need (diets, baldness cures, wrinkle removers): Science certainly may have something to say about the problem, and maybe some of the products out there really do work, but there are so many remedies being pushed on the gullible that it seems to me you have to be extremely cautious about spending your money on any of them.
I will be interested in seeing Fisher’s work when she publishes it, because her method uses sociological and psychological data and also “indicators linked to chemical systems in the brain”, and I’m very curious about how that works. Some of the things discussed in the NYT article have to do with attraction and falling in love, and brain chemistry certainly mediates how that happens. But it seems to me that as hard as it might seem to find someone with whom to fall in love, what’s really hard is to find someone you will still be in love with, or at least contentedly married to, 30 years down the road. The hormonal rush of falling in love doesn’t always have all that much to do with how well you’ll be able to get along with a person over the long haul. In fact, if it did, a rich subject for literature would disappear. It seems to me that if you were to try to use any indicators from brain chemistry to help people find a good match, you’d need to have looked at the brains of people who were falling in love 30 years ago, and see which of them stayed together, and then use their brain chemistry as the model. I’m curious about whether there’s even a difference in the initial stages between those who stay together and those who don’t, and if there is, I’m not sure that’s what Fisher’s algorithm is using.
One reason I’m skeptical about the emergence any time soon of a widely applicable system for matching people up is the complex nature of some of the concepts involved. Take the idea of similarity, for example. Some similarities make for compatibility, but others make for conflict, and still others don’t matter all that much. If one of you votes for Clinton and one for Obama next Tuesday (or even if one of you is blue and the other red), that might not make much of a difference to your relationship. If one of you likes to hike rugged trails and the other prefers mall-walking, at the very least you’ll have a hard time figuring out what to do together on a Saturday afternoon. If you’re middle-aged and in a relationship with someone who is just as passionately attached to his house as you are to yours, you could be in for a rough ride unless the houses are similar and one or the other of you eventually gives his or hers up. (Which means one of you has to be less stubborn than the other.) It’s a complicated concept, any way you look at it. Some scientific papers are listed in a sidebar on the left of the NYT article; if you have a way to access the one by Ruth Gaunt, check that one out for a good summary of the difficulties involved in analyzing whether similarities in a couple are related to marital happiness. The one just above it is interesting as well (and available for free) but it describes research on dating couples and newlyweds, not couples that have been together for a long time.
Another sort of similar service is OKcupid. Here’s the URL to their explanation of their matching algorithm:
http://www.okcupid.com/faaaq.html
(I sound like an ad for their site, but it’s a fun place, their services are free and the matching algorithm works fairly decently, the only caveat being most users never answer enough questions, or don’t answer them honestly, in order to make it work really spectactularly).
Wow, I was beat to it.
The matching system described in that okcupid FAQ is very common sense, and I’ve found it to work well (at least, it is a definite improvement over an unsorted list). Okcupid has a different analysis which is available if you click the “compare me to this user” button. They have a bunch of personality characteristics (like adventuresomeness) and somehow (don’t ask me) correlate the characteristics to answers to the questions. So they will come up with a list of statements like “Fred is WAAAAY more adventuresome than Wilma.” The characterization is only indirectly related to the match percentage, but it is still a very useful information resource. In other words, rather than telling you who you match, these sites also can give you a lot of extraneous information you can use to make your own decisions (or even *sigh* use to make tactical decisions).
Online dating definitely improved when the quantification nuts got their fingers into it, even if there is no fundamental “match formula.”
We had a singles event at my office a few years ago where everyone signed up with eharmony. I thought their questionaire was interesting, but I turned quickly skeptical a few days later when every male in the office matched with every female and vice versa. Makes me think geographic location weighs heavily in their matching process. Also, no marriages or even dates from this excerise as far as I know, but a couple of people hooked up after the Christmas party.