Much like Facebook, dating site OkCupid has been conducting secret social experiments on its users—not to influence their moods, but sometimes to influence their opinions of one another. In the latest post on OkCupid's data-driven chronicle of man's inhumanity to man, the OkTrends Blog, we learn that your picture is all that matters and you'll like whomever a fair-to-middling cupid tells you to like.
Unashamed of its experiments, OkCupid has come right out and explained the results—and all of them point to profile photos as the main driver of behavior on the site.
When OkCupid removed photos for one day, to promote its failed blind date app, traffic and interaction dropped dramatically. As soon as photos were turned back on, people abruptly bailed on the "blind" conversations they had started.
Even more telling, OkCupid revealed it had to remove its "personality" rating scale years ago, because it turned out to only measure looks.
"The two scores are within a half point of each other for 92% of the sample after just 25 votes (and that percentage approaches 100% as vote totals get higher). In short, according to our users, 'looks' and 'personality' were the same thing," wrote OkCupid's Christian Rudder.
The problem of hotties with literally no text on their profiles receiving high marks for "personality" inspired OkCupid to measure exactly how much your potential dates care about all those words you spend so much time writing. The results are bleak.
Displaying the same profiles with and without their text produced more or less the same responses.
"Essentially, the text is less than 10% of what people think of you," Rudder wrote.
Lesson learned: If you're hot enough, you can go ahead and skip the written exam. If you're not, give it a shot. You've got nothing to lose and 10 percent to gain.
OK, Cupid, what about the mind-control part of this experiment? How was a dating website influencing its users in nefarious, Facebook-like ways? That neat trick has to do with the site's compatibility ratings.
Basically, OkCupid decided to see how much influence those scores actually had on singles' opinions of one another by finding pairs with low compatibility—say, 30 percent—and telling them they'd be perfect for one another.
And, holy shit, it worked. Here's Rudder again:
Does the mere suggestion cause people to actually like each other? As far as we can measure, yes, it does.
When we tell people they are a good match, they act as if they are. Even when they should be wrong for each other.
He points out that this placebo effect seemed to work better when the two subjects were more compatible to begin with, but in the absence of actually hitting it off—or liking one another's photos, as we learned earlier—being mind-tricked together by a website is almost as good.
How is this any better than Facebook using our news feeds to see if it can make us miserable? OkCupid doesn't have a very thorough justification.
"[G]uess what, everybody: if you use the Internet, you're the subject of hundreds of experiments at any given time, on every site. That's how websites work," is about as close as they get to giving a fuck.
But perhaps we're more willing to accept this sort of thing from OkCupid because online dating already feels like consenting to participate in a social experiment. It's a game we play with virtual strangers, while Facebook is a place we trust with our "real friends," even when we know we probably shouldn't.