The best answer to this question so far came in a 2009 book titled Crossing the Finish Line: Completing College at America’s Public Universities, a collaboration between two former college presidents, both economists—William G. Bowen, the president of Princeton University from 1972 to 1988, and Michael S. McPherson, who served for almost a decade as the president of Macalester College in Minnesota. Because of their standing in the education establishment, Bowen and McPherson—along with a third coauthor, a researcher named Matthew Chingos—were able to persuade sixty-eight public colleges, as well as the College Board and ACT, to give them access to detailed academic data covering about two hundred thousand students. They found in the data some surprising facts about which students successfully complete college, which ones drop out, and why.
In certain quarters, the college-dropout phenomenon has been explained as a problem of excessive and unrealistic ambition on the part of many students, especially low-income students. The conservative author Charles Murray argued in his 2008 book Real Education that the true crisis in American higher education is not that too few young Americans are getting a college education; it is that too many are. Because of Americans’ natural tendency toward “educational romanticism,” Murray wrote, we push students to go to college who are simply not smart enough to be there. High-school guidance counselors and college-admissions officers, lost in “a fog of wishful thinking, euphemisms, and well- intentioned egalitarianism,” encourage low-IQ, low-income students to attend colleges that are too intellectually demanding; when those students discover that they don’t possess the intelligence necessary to do the work, they drop out. Murray, the coauthor of The Bell Curve, is perhaps the country’s best-known cognitive determinist, and his thesis in Real Education is a pure expression of the cognitive hypothesis: what matters in success is IQ, which is fixed quite early in life; education is not so much about providing skills as it is about sorting people and giving those with the highest IQs the opportunity to reach their full potential.
But when Bowen, McPherson, and Chingos took a close look at their data, they found that low-income students generally weren’t overreaching their abilities when they chose their colleges; many of them, in fact, were attending schools well below what their GPAs and standardized-test scores qualified them for. This phenomenon, which the authors labeled undermatching, didn’t happen much with well-off students; it was a problem that almost exclusively affected disadvantaged teenagers. In North Carolina, the state for which the researchers were able to gather the most complete data, three out of four high-income students with the GPAs and test scores needed to gain admission to one of the state’s highly selective public colleges went on to attend a highly selective school. For them, the system worked. But among students who had those same lofty academic credentials but didn’t have parents who had attended college themselves, only a third chose to go to a highly selective school. And choosing a less challenging college didn’t make it more likely that those highly qualified students would graduate—it had the opposite effect. Undermatching, the authors found, was almost always a big mistake.
But the information on undermatching, important though it was, was not the most surprising or significant finding in Crossing the Finish Line. The authors also discovered that the most accurate predictor of whether a student would successfully complete college was not his or her score on the SAT or the ACT, the two standardized college-admissions tests. In fact, it turned out that, except at the most highly selective public universities, ACT scores revealed very little about whether or not a student would graduate from college. The far better predictor of college completion was a student’s high-school GPA.
To people involved in the college-admissions process, this finding came as something of a shock; it was essentially a repudiation of one of the founding tenets of the late-twentieth-century American meritocracy. In The Big Test, Nicholas Lemann’s history of standardized college-admissions testing, he explains that the SAT was invented, in the years after World War II, because of growing skepticism about the predictive power of high-school grades. How were college-admissions officials supposed to compare a 3.5 student at a suburban high school in California with a 3.5 student at a rural high school in the Pennsylvania countryside or at an urban school in the South Bronx? The SAT was designed to correct that problem, to provide an objective tool that would distill a student’s ability to thrive in college down to a single, indisputable number. But at the colleges that Bowen and Chingos and McPherson examined, high-school grades turned out to be excellent predictors of college graduation—no matter where the student attended high school. It was true that a student with a 3.5 GPA from a high-quality high school was somewhat more likely to graduate from college than a student with a 3.5 GPA from a low-quality high school, but the difference was surprisingly modest. As the authors put it, “Students with very good high school grades who attended not-very-strong high schools nonetheless graduated in large numbers from whatever university they attended.”
And when Angela Duckworth, the guru of self-control and grit at the University of Pennsylvania, analyzed GPA and standardized-test scores among middle-school and high-school students, she found that standardized-test scores were predicted by scores on pure IQ tests and that GPA was predicted by scores on tests of self-control. Put Duckworth’s findings together with the discoveries in Crossing the Finish Line, and you reach a rather remarkable conclusion: whether or not a student is able to graduate from a decent American college doesn’t necessarily have all that much to do with how smart he or she is. It has to do, instead, with that same list of character strengths that produce high GPAs in middle school and high school. “In our view,” Bowen, Chingos, and McPherson wrote, “high school grades reveal much more than mastery of content. They reveal qualities of motivation and perseverance—as well as the presence of good study habits and time management skills—that tell us a great deal about the chances that a student will complete a college program.”
It’s possible, of course, that once a student reaches adolescence, those skills and habits are no longer teachable. It may be that at that point, either you have them or you don’t, and if you have them, you’re likely to graduate from college, and if you don’t, you’re not. But consider Elizabeth Spiegel’s ability to reconstruct the thinking skills of her middle-school chess players. Think of the way Lanita Reed helped Keitha Jones change her whole outlook on life—essentially helping her rewire her personality—at the advanced age of seventeen. In each case, a teacher or mentor found a way to help a student achieve a rapid and unexpected transformation by using what James Heckman would call noncognitive skills and David Levin would call character strengths. What if we could do that for large numbers of teenagers—not to help them attain chess mastery or persuade them to quit fighting in school but to help them develop precisely those mental skills and character strengths they would need to graduate from college?
Jeff Nelson, the CEO of OneGoal, doesn’t seem like a revolutionary when you first meet him. He’s fresh-faced and clean-cut and midwesternly polite, with a tuft of blond hair sticking up over his forehead that makes him look a little like the comic-book character Tintin. He wears button-down shirts and keeps to a button-down schedule; once, when I made arrangements to talk to him on the phone, he e-mailed me in advance with a point-by-point agenda for our call that included three “objectives” and allocated ten minutes for “wrap up.” He seems most at home when surrounded by the typical tools of the modern education reformer—PowerPoint presentations, management consultants, strategic plans, venti lattes—and yet his vision of education reform is a profoundly unorthodox one, a thorough challenge to the cognitive hypothesis.
Nelson grew up in Wilmette, an affluent bedroom community that is part of the comfortable, Caucasian suburban enclave north of Chicago where John Hughes set Home Alone and The Breakfast Club. It is a mostly Democratic town, a reliable haven for progressive causes and notions of social justice, though those notions are often expressed in an abstract, distant way, through donations to Amnesty International or Habitat for Humanity or petitions supporting the refugees of Darfur. From an early age, though, Nelson was drawn to an issue closer to home: the challenges faced by children growing up in the metropolis fifteen miles to his south. In eighth grade, Nelson read Alex Kotlowitz’s book There Are No Children Here, the harrowing story of two African American boys living in the Henry Horner Homes, a dismal and dangerous high-rise housing project on Chicago’s West Side. The book, Nelson told me, “crushed my view of the world a little bit. It sparked something in me.”
Nelson went on to attend New Trier Township High School, which is legendary in the Chicago area for its lush campus and lavish facilities, all underwritten by the property taxes assessed on the luxurious homes of
