Despite competition from platforms like Twitter, Instagram, and Linkedin, recent polls show that Facebook continues to dominate as the world’s leading social networking platform. With an estimated 1.23 billion monthly active users, Facebook is now frequented by about 71 percent of all online adults (about 40 percent of savvy socials check their streams multiple times per day). Armed with stats spanning ten years, the Facebook Data Science team has recently announced that the superpower site can actually predict just how long love will last.
What factors, if any, determine the life of a relationship? Bogdan State, a Stanford University sociologist-in-training and member of the Facebook Data Science team, decided to find out. The data, released as part of the site’s week-long tribute to Valentine’s Day love, focused on relationships that started between January of 2008 and December of 2011. The research team tossed out short-lived flings lasting less than three months, relationships where both parties were under 23 years old, and relationship statuses that started as “Married.”
After applying some super sweet statistical analysis, the team found that if Facebook lovebirds can keep their relationships alive for at least three months they are far more likely to be found celebrating their love four years later. Interestingly enough, the group also found that sentiment can vary by season. It seems that the summer months of May, June, and July were the most popular time to break someone’s heart. Another notable dip happened in February, leaving some to speculate whether Valentine’s Day pressure could be to blame. According to State, 2011 also seemed to be a year of accelerated breakups. State hypothesized that the recession-ridden U.S. economy could have played a role.
This is not the first time Facebook has attempted to predict how long love will last. Late last year, Lars Backstrom of Facebook partnered with Jon Kleinberg, a professor in the Department of Computer Sciences at Cornell University, to create an algorithm designed to determine a relationship’s likelihood of success. Many speculated that relationship compatibility could be predicted by the number of mutual friends two people share, but Backstrom and Kleinberg discovered something different. In fact, how close mutual friends, co-workers, acquaintances, and other connections were to each other actually made all the difference. According to the algorithm, sweethearts who do not share intermingling social circles can pretty much consider their relationship doomed from the start.
What about Valentine’s Day singletons? For those looking for love, Facebook suggests a trip to Colorado Springs. The sunny city poses the greatest likelihood of “relationship formation,” according to Mike Develin of Facebook’s Data Science Team. Notable runners-up include El Paso, Louisville, Forth Worth, and San Antonio. Singles who just want to mingle should head to San Francisco, Washington, New York, or Los Angeles. These cities, according to Develin, have loads of singles but low levels of relationship formation.
For those both lucky and unlucky in love, remember that Facebook’s ability to predict how long love will last is based on data collected from Facebook users only. Not everyone is on Facebook and, of those who are, some simply share more than others.
By Katie Bloomstrom