Now We’ve Got Computers Labeling Us
In the 90s, people were obsessed with their identity. It was important to us all to be known as individuals, not mindless sheep who accepted the status quo. Many of us were, in fact, sheep, but we weren’t your typical sheep.
We sheepishly followed the beat of the same different drummer that our other sheep brethren followed. And we didn’t just want to be unique individuals; we wanted everyone to know what kind of individual we were.
We wore the T-shirt of our favorite band. We plastered our political beliefs on the back of our cars via bumper sticker. We only hung out at coffee shops and bars with others like us. Using a few quick clues, any attentive individual could quickly identify a stranger’s likes and dislikes, political leanings, and sexual orientation. We still care about our individuality these days, though to a lesser extent. And instead of finding ways to craft and broadcast this individuality, we simply navigate to Facebook, click a thumbs up, and hope someone’s paying attention.
The point is, an attentive person with an elementary education can use context clues and a little assumption to piece together some details about a stranger’s life.
But you don’t have to take my word for it, just look to a recent study performed by the University of Cambridge which found that…yes, if given a set of things that a person “likes,” then someone can make a good assumption about that individual’s personality.
And they all said in unison: Yeah, no duh.
Since the computers of today are slightly more powerful than they were in the 90s, the Cambridge researchers were also able to create a model that they claim can determine gender, political leanings, race, religion, and sexual orientation, based solely on Facebook likes.
I imagine it’s an incredibly simple task.
Does a person “like” the bible? They’re probably a Christian.
Do they “like” President Obama? They’re probably a Democrat.
Do they “like” Maxim magazine? They’re probably a dude.
Essentially, the Cambridge team has made a judgmental computer model that can be used to make some (albeit statistically accurate) assumptions based on what people say they enjoy. Haven’t we been doing this all along? Sure, we try not to do this. We all try to not make assumptions based on hobbies or interests or favorite music and television shows, but in the deep recesses of our brain, little connections are being made.
Twilight reader? Dark and brooding romantic.
REI Shopper? Outdoor enthusiast.
Beatles fan? Clearly someone you want to know.
This Cambridge computer model calls upon some statistical data, stitches it together with Facebook likes, and has the audacity to make assumptions about people. Of course, it’s a computer and working only with numbers, so it’s hard to get too mad at it.
For instance, according to this computer, a person who “likes” Kathy Griffin, Juicy Couture and the musical “Wicked” is statistically likely to be a gay man.
Sports terms, Bruce Lee references, and admitting to being confused after waking up for a nap means you’re statistically a straight man.
The model can even predict intelligence levels.
For instance, if you like Bret Michaels, Harley Davidson and Clark Griswold, then statistically you are on the lower side of the intelligence scale. But you can’t get mad about it. It’s only a computer judging you based on your interests. If you happen to enjoy the same things as other dumb people, then the computer simply locks you into that crowd.
This is just another example of science attaching itself to things we’ve known all along. Though it may not be very “PC,” people who enjoy a certain thing tend to flock towards others who also enjoy that thing. I think the quote goes something like: “Birds of a feather flock together.”
It’s not a good idea to seek out only those like you or refuse to befriend those who like a different kind of movie than you, of course. Perhaps this computer model further proves that we naturally gravitate to people who are similar to us, even if we aren’t trying to.
Image Credit: Lasse Kristensen / Shutterstock