Debate Magazine

Correlation Does Not Imply Causation

By Stevemiranda

I had a conversation today with a friend who recently took a job working for a terrific non-profit organization doing education reform. She’s proud of her work, but acknowledges some inherent limitations. Everyone wants data to prove their program works, but when you’re dealing with human beings, how are you supposed to quantify personal growth?

She said, “It’s close to impossible to provide the opportunities that some of these programs do without receiving support from the government and other similar bureaucracies, and their support is reliant on this sort of cold hard data.”

* * *

This exchange reminded me of an article I read back in the fall. Here’s an excerpt, from the Associated Press:

“For many years, diversity in higher education has been measured by how many low-income students and students of color enroll in college. The Bill & Melinda Gates Foundation wants to make a dramatic change in that definition, by focusing instead on college graduation rates.

“We know that in today’s economic climate and labor market, a high school diploma is no longer enough,” said Allan Golston, president of the U.S. Program at the Gates Foundation. “We must not only ensure that young people have access to college; we must ensure that they go on to complete college and earn a degree or certificate with value in the workplace.”

* * *

Everyone wants to prove their program works, and the way we do that is through cold hard data. Figuring out what to measure, however, is not simple. The Gates Foundation used to measure how many kids were getting into college, until it realized that it was measuring the wrong thing. Now, getting kids through college with a degree is the primary focus.

I suspect that, over time, the Gates Foundation will realize that it’s still measuring the wrong thing.

* * *

The cold hard data on the matter of education and income is clear: the more you learn, the more you earn. At least, that’s what we’ve convinced ourselves. I suspect that the relationship between the two is one in which correlation does not prove causation. In other words, it’s not clear that college graduates earn more because of what they learned in college.

Just off the top of my head, I can think of three friends who never finished college.

  • “Allan” got through high school without doing much—”I was one of those students who never really gave the teacher much satisfaction,” he says—but managed to get into a high profile private college in Philadelphia. He smoked a lot of pot and eventually dropped out. His passions are building things with his hands, community service, travel, and God. Now, he works as a project manager for Habitat for Humanity, heads off to Mexico as much as possible, and is an active members of his church. He’s happily married, and is a great father to his little boy.
  • “Dan” never went to college. He found his passion working as a sports writer right out of high school. He was really good at it, so his editor kept giving him more responsibility and higher profile assignments. Now, he writes for He’s married with two kids, totally happy.
  • “Terry” drank too much in college and dropped out during his sophomore year. But he’s a really friendly guy, the kind of person everyone likes (in education parlance, he has “advanced interpersonal intelligence”). Ever since I’ve known him, he’s been a dedicated, loyal friend. Now he’s happily married with a couple kids, and sells mortgages. Of course he sells mortgages—with a personality like his, I’m sure he gets tons of referrals from happy clients.

This is why I think the relationship between a college degree and success is merely correlation and not causation. What you learn in college often has very little to do with the work you do at your job. Would Andy be a better project manager or a better father if he’d spent more time analyzing Hemingway’s prose? Would Terry deliver better customer service if he’d passed Oceanography 101?

* * *

I know, I know. To suggest that kids don’t necessarily need to go to college is heresy in our society. But the relationship between academic learning and life satisfaction is not clear.

The thing that kids get out of college—the reason they’re able to earn more money after graduation— is they’ve completed a process in which they had to set a goal that’s achievable over time, in which they had to develop the self-discipline to overcome obstacles and see it through to the end. That’s a valuable process that some people learn in college. Allan, Dan, and Terry learned it some other way.

* * *

In the end, the whole point of this stuff is to figure out how we can maximize our happiness, right? We want to figure out how to be successful, contributing members of society because we want to be happy.

Well, consider this quotation from Martin Seligman, past president of the American Psychological Association and the father of the Positive Psychology movement. He’s a scientist, which means he makes his living analyzing cold hard data. “As a professor, I don’t like this, but the cerebral virtues—curiosity, love of learning—are less strongly tied to happiness than interpersonal virtues like kindness, gratitude and capacity for love.”

So, what exactly should we be measuring?

(Join the discussion at Get updates at

Back to Featured Articles on Logo Paperblog