This is an advice column about getting into an elite college. And, like every column about the anxiety of getting into an elite college, it must begin with a massive caveat: If you and your parents are worried about getting into an elite school, odds are that you are already elite.
The raw numbers are instructive. There are more than 4 million 18-year-olds in the United States. About 3.5 million of them will go to college. And just 100,000 to 150,000 of those—somewhere around 3 percent of the entire age group—go each year to selective schools that admit fewer than half of their applicants. College-admissions mania is a crisis for the 3 percent.
That said, the 3 percent mania is real. Hence the annual April blossoming of op-eds promising, with comforting certainty, that "it doesn't matter where you go to college." That is a really nice message. It's also wrong.
It matters where you go to college, plain and simple. Graduates of the most-select colleges often earn more than graduates of less-select public universities, who are employed at higher rates than those of community colleges, who get more calls from potential employers than graduates of online universities. A world where "44.8% of billionaires, 55.9% of [Forbes's most] powerful women, and 85.2% of [Forbes's most] powerful men" attended elite schools is not a place where college doesn't matter.
One of the most reproduced statistics from Frank Bruni's new book on this subject, Where You Go Is Not Who You'll Be, is that just 30 percent of Fortune 500 CEOs went to elite colleges like the Ivies. (I have not read his book, and am not commenting on its quality outside of this statistic.) This factoid is being widely interpreted to prove that elite schools are overrated predictors of business success. But that's the wrong way to think about it. If a tiny share of college attendees account for a third of business leaders, that means graduates of elite schools are at least 10 times more likely than their peers to be Fortune 500 CEOs.
This is starting to feel like a downer column for the thousands of families who will open skinny rejection letters from Harvard, Stanford, and their kin. But I promise there's an inspirational message lurking here, and it starts with the question: Why is college, broadly speaking, a reliable indicator of earnings, employment, and success?
College acceptance and future success are both reflections of an obvious but often overlooked variable: the person you're becoming in your late teens. After all, elite schools aren't taking a random sample of high-school students and churning out success stories. They're accepting people who are already on the road to success, connecting them with peers and alumni in successful jobs, giving them a degree that signals to employers that this person has the potential to be successful, and then basking in their eventual success. (The fact that networks and signals are enough to make elite schools valuable doesn't prove that they don't also offer a great education; it just makes it all the harder to study exactly how much better an elite education is.)
But let's say you took a group of similarly gifted kids. Let's say some were randomly accepted to super-duper-elite schools (like Yale) while others were rejected and attended less-elite schools (like Penn State). What if the results showed that both groups turned out nearly the same? Wouldn't that offer sort of a nice message: that the habits you've built and the person you're becoming at 18 matter more than the exact school you attend?
In fact, that was broadly the finding of one of my favorite studies, a 2002 paper from economists Stacy Berg Dale and Alan B. Krueger. Here's the summary, in their words:
We matched students who applied to, and were accepted by, similar colleges to try to eliminate this bias. Using the College and Beyond data set and National Longitudinal Survey of the High School Class of 1972, we find that students who attended more selective colleges earned about the same as students of seemingly comparable ability who attended less selective schools.