Ah, the myriad uses for our voluntarily relinquished personal data. Data scientists have once again made use of Facebook's arsenal of billions of relationships to parse who's romancing who and how likely people are to break up.
We all know that Facebook released data on when you're most likely to get the axe in your relationship, but this study uses a different method. Researchers created a measure they called "dispersion," or how likely relationships between people's very different peer groups were to be intertwined. So, in the image below, the top clump of dots are this person's co-workers. The righthand clump of connections are her college friends.
AdvertisementADVERTISEMENT
And, that central dot in the lower left? That's this person's spouse. See how his connections fan out in a totally different way than the others? And, how they connect to people he might never know if he wasn't intimately acquainted with his wife, the central dot? It's a beautiful, visual depiction of how our romantic partners are literal bridges to other social worlds.
Now, the way that these relationships overlap (i.e. how many connections we have in common) leads to dispersion measurements. The more interwoven we are, the higher the dispersion measure. So important is this measure that researchers found that people in relationships with low levels of dispersion were 50% more likely to break up over the next two months than their high-dispersion counterparts.
For example, if you're Facebook friends with my childhood best friend back in Georgia AND my college roommate AND my current co-worker, then that's highly dispersed. There's probably no reason you would be friends with all those very different people other than the fact that we're in love.
Now, there could be many reasons for this. The longer you're in a relationship, the more common friends you'll develop, and the higher your dispersion will be. So, this is likely a factor. But, it's also an indication that sharing more of your life with someone — friends included — might be the special something that helps things work. (The New York Times)
![points](http://static1.refinery29.com/bin/entry/f5b/x,80/1116095/points.jpg)