Decoding Love - Andrew Trees [56]
TO VIEW YOUR PROFILE, CLICK HERE
Let’s turn our attention for a few moments to Internet dating sites. Almost all of them are engaged in a race to come up with a system to figure out the mating market, and while they aspire to science, they are still closer to medieval alchemists than modern chemists. Seemingly every site has hired its own “love guru” and developed some sort of top secret algorithm for matching couples. Chemistry.com has Dr. Helen Fisher, who has come up with a modified version of the Myers-Briggs personality test (if you ever spent time in a career counseling office trying to figure out what to do with your life, you probably took some version of this). She has written a number of wonderful books on love, but as she would be the first to admit, she isn’t even trained in this area—her degree is in anthropology. When Chemistry.com’s Web site tried to claim that its algorithm was based on the latest science of attraction, eHarmony complained to the Better Business Bureau and forced them to remove the claim. Match.com has created Perfect Match with its attendant guru Dr. Pepper Schwartz (also the writer of a number of excellent books), who has come up with the “Duet Total Compatibility System,” a more complicated reworking of Myers-Briggs than Fisher’s. The most complicated of all is on eHarmony, where hundreds of questions measure people across twenty-nine core traits. And the mathematical formula that makes sense of all those traits? They guard that as if it is the secret formula for Coca-Cola.
The first problem with all these sites is that while they may claim to be scientifically based, none of them has yet to pass the real test of science—peer review. In other words, it’s not enough to say that you have come up with a magic formula. You have to submit your research in a forum where other scientists can judge the validity of your claims. This may sound nitpicky, but it is the bedrock of scientific inquiry. While many of these dating services say that they intend to publish their results, talk is cheap. I guarantee you that if any of these sites had clear evidence of success, they would rush to publicize it.
If you start examining these sites using mathematics, the problems run even deeper. Lori Gottlieb wrote a very funny article for The Atlantic Monthly a few years ago in which she complained, among other things, that she wasn’t matched with a single person on eHarmony. Neil Clark, eHarmony’s founder, cheerfully explained that her problem was that she was too exceptional and that eHarmony did a much better job of matching average people. Of course, that answer makes one wonder what help, if any, these dating sites provide. The vast majority of people fall within one standard deviation of the statistical mean for virtually every trait, so if you are average, it would be almost impossible not to match you with similar people and also virtually impossible to figure out if there was any validity to the matching system.
You might think that the way around this is to measure more traits. From this point of view, eHarmony’s twenty-nine core traits look pretty good. It is reassuring to have a long list of questions about yourself and what you are looking for in a partner. You can carefully calibrate just how important ambition or sense of humor or kindness is. When you finally stop clicking with your mouse, it probably feels as if all you have to do is sit back and wait for the computer to spit out your true love. But a little mathematical analysis reveals a fatal flaw with multiplying the criteria. Unfortunately, the more qualities a dating questionnaire includes, the more unlikely it is that you will find anyone who matches you. Even if you limit the survey to include only six possible attributes, you only have a one in twenty-eight