The science of love

Do science and love mix?  Many online dating sites provide users with daily or weekly suggested "matches," or simply pick your matches for you, claiming to have the magic scientific formula for success. eHarmony, for example, requires you to fill out a long (I'm talking at least an hour… if you're lucky) "personality profile." Chemistry.com has you take a personality test, similar to Myers-Briggs, that "reveal[s] your personality type [to] see what makes you tick."  For both of these sites, the premise is that similar people attract each other. Taking a slightly different approach, OKCupid provides you with a "match percentage," based on how your answers to certain questions compare to someone else's.  The questions range from the more serious, "Do you like to discuss politics?" to the absurd, "In a certain light, wouldn't nuclear war be exciting?" OKCupid designed the site so that you can use the match percentage if you want, but they leave it up to you. OkCupid's founder Sam Yagan agrees that similar people make the best matches, recently saying, "We [at OKCupid] say that opposites attract… and then attack."

Around this time last year, a group of psychology professors (one of whom happened to be in my a cappella group at Cornell – small world, right?) released a report questioning these algorithms' accuracy.  The authors found that, while they might be a good way to reduce the sheer number of potential partners on an online dating site, they are no better at creating success stories than two people put together at random. The authors don't say that online dating is bad.  In fact, they conclude that the best part of online dating is that the sites bring people together who wouldn't have met otherwise. It's the "science" behind the websites' claims, or their matching processes, that the authors refute. Eli Finkel, the lead researcher, said, "Eighty years of relationship science has reliably shown you can't predict whether a relationship succeeds based on information about people who are unaware of each other… The assumption is that the algorithms work. We reviewed the literature and feel safe to conclude they do not."

Sites like JDate and Match.com have come to a good balance.  In addition to providing you with "matches," the sites simply let you browse the database to find people on your own.  This allows, say, a man in Chicago to communicate with a woman in Philly because he's planning to move here in three months.  On a site like eHarmony, however, this guy will never even be shown profiles so far outside of his own location.

In the end, take the sites' suggested matches with a grain of salt.  Maybe someone in there will strike your fancy.  Maybe not.  No one is better able to predict whether you like someone than you, and it will still always take an in-person meeting to know if you have the right connection.

Do you think that online services add value to the dating game with their supposed algorithms, or should just they leave it to us to decide who might be the best fit?  (Feel free to use the comments section for your responses.)

Calling all single ladies.  If you're looking for something to do on V-Day, which happens to be a Friday this year, check out this event: https://www.eventbrite.com/e/men-sex-fashion-the-ultimate-single-girls-valentines-day-event-tickets-10313204071 I have not met the organizers personally, but it sounds like fun!