As someone who takes comfort in reducing highly complex human experiences into single-line mathematical equations, let me share my latest shiny new toy: a formula that predicts the breakup percentage of a given sample of dating folk over a particular period of time.

Isn’t that intriguing?

Got that? All your love problems are solved? (J/k, I’ll explain later)

Before we delve into the nuts and bolts, let’s take a moment to appreciate the research that produced this nifty equation. It came out of a meta-analytic study[1], but let me just say–this ain’t yo mamma’s lit review.

Meta-analyses often look at a decade of research at a time, but after spotting some holes in earlier reviews, the team just decided to start at the beginning. (Incidentally, after coming across the study independently, I realized it was spear-headed by my internet-friend Dr. Ben Le, a professor at Haverford College. He’s one of the triumvirate founders of Science of Relationships, a pop-scholarly blog that I contribute to from time to time. He’s very kind and down-to-earth.)

Reaching back more than thirty years, the researchers hunted down each and read every study that had EVER looked at breakup patterns in dating. Combing through this enormous pile, they then enumerated each breakup variable that had floated through the brains of three decades of empirical thinkers. That’s basically anything that fits in this blank: “I’ll bet breakups are caused by or reduced by _______,” such as relationship duration, dependence, positive illusions, and alternative dating options. To qualify for the current meta-analysis, a variable had to show up in at least four different studies.

And what, you ask, was the fruit of their winnowing work? Thirty possible breakup predictors made the cut, dragging one-hundred-thirty-seven separate research papers and data sets along with them. Next, the team deconstructed and recoded all the data and fed it into a new database. In the statistical equivalent of a barn-raisin’, they erected a huge new mega-study that evaluated the breakup patterns of 37,000 participants over a thirty-three year period.

Those wily statisticians.

As you might expect, this new data set affords a pretty neat birds-eye view of the premarital relationship landscape. I’ll be devoting my next couple of posts to telling y’all more about what they found. (Sorry, the barn-raising imagery has me itching my coverall shoulder straps).

Today, though, I want to play with this breakup equation. Here it is again:

As Le et al says,* “Y** *is the anticipated breakup rate, *X*1 is the time lag between Time 1 and Time 2 (in weeks), and *X*2 is the average relationship duration of the sample at Time 1 (in weeks).”

Er, thanks guys.

In normal words, you start with a group of dating couples and determine how long (in weeks) everyone has been together. Some couples might have been dating for three weeks, others for seventy-three, but let’s say that the average for the group is thirty weeks. Thus, 30 is your *X*2.

Next, you have to decide how far into the future you want to look. Do you want to know how many couples are likely to be together two months later? Two years later? For the sake of easy numbers, let’s say the group agrees they want to see their breakup fortune sixty-six weeks out. In this case, their *X*1 would be 33 (66-33 = 33).

Feed it all into the equation and bingo! 36.8% of this group of couples, or about 1 in 3, will break up within sixty-six weeks.

Isn’t that wild?

To be fair, the 37,000 participants had been dating for several months on average, so the predictive value of the equation weakens in the early phase of dating. For example, it predicts that only 41% of couples who have been dating for 3 weeks will break up over the next 52 weeks. In reality, I would expect a higher rate–though future studies may be able to evaluate that assumption.

Anyway–find your friends, grab a pencil, and figure out your breakup fate…then try to overcome it.

And stay tuned—more insights about what triggers breaking up, and predicts commitment, coming soon!

[1] Le, B., Dove, N. L., Agnew, C. R., Korn, M. S., & Mutso, A. A. (2010). Predicting nonmarital romantic relationship dissolution: A meta-analytic synthesis. *Personal Relationships, 17*, 377-390.

Pingback: Top Three Predictors of Successful Relationships: PI.C.L. | Where Is This Going?

Our HS (gymnasium) math teacher way, way back (early 60s) in at that time communist Czechoslovakia showed us this heart/love equation in our regular class.

Really? How interesting!

Love and mathematics together is a great combination. Very rarely seen and understood by anyone. The equation given is very impressive. Really interesting class.