By DAVID LEONHARDT
Published: March 16, 2013
“Most low-income students who have top test scores and grades do not even apply to the nation’s best colleges, according to a new analysis of every high school student who took the SAT in a recent year.
The pattern contributes to widening economic inequality and low levels of mobility in this country, economists say, because college graduates earn so much more on average than nongraduates do. Low-income students who excel in high school often do not graduate from the less selective colleges they attend.
Only 34 percent of high-achieving high school seniors in the bottom fourth of income distribution attended any one of the country’s 238 most selective colleges, according to the analysis, conducted by Caroline M. Hoxby of Stanford and Christopher Avery of Harvard, two longtime education researchers. Among top students in the highest income quartile, that figure was 78 percent.
The findings underscore that elite public and private colleges, despite a stated desire to recruit an economically diverse group of students, have largely failed to do so.
Many top low-income students instead attend community colleges or four-year institutions closer to their homes, the study found. The students often are unaware of the amount of financial aid available or simply do not consider a top college because they have never met someone who attended one, according to the study’s authors, other experts and high school guidance counselors.
“A lot of low-income and middle-income students have the inclination to stay local, at known colleges, which is understandable when you think about it,” said George Moran, a guidance counselor at Central Magnet High School in Bridgeport, Conn. “They didn’t have any other examples, any models — who’s ever heard of Bowdoin College?”
Whatever the reasons, the choice frequently has major consequences. The colleges that most low-income students attend have fewer resources and lower graduation rates than selective colleges, and many students who attend a local college do not graduate. Those who do graduate can miss out on the career opportunities that top colleges offer.
The new study is beginning to receive attention among scholars and college officials because it is more comprehensive than other research on college choices. The study suggests that the problems, and the opportunities, for low-income students are larger than previously thought.
“It’s pretty close to unimpeachable — they’re drawing on a national sample,” said Tom Parker, the dean of admissions at Amherst College, which has aggressively recruited poor and middle-class students in recent years. That so many high-achieving, lower-income students exist “is a very important realization,” Mr. Parker said, and he suggested that colleges should become more creative in persuading them to apply.
Top low-income students in the nation’s 15 largest metropolitan areas do often apply to selective colleges, according to the study, which was based on test scores, self-reported data, and census and other data for the high school class of 2008. But such students from smaller metropolitan areas — like Bridgeport; Memphis; Sacramento; Toledo, Ohio; and Tulsa, Okla. — and rural areas typically do not.
These students, Ms. Hoxby said, “lack exposure to people who say there is a difference among colleges.”
Elite colleges may soon face more pressure to recruit poor and middle-class students, if the Supreme Court restricts race-based affirmative action. A ruling in the case, involving the University of Texas, is expected sometime before late June.
Colleges currently give little or no advantage in the admissions process to low-income students, compared with more affluent students of the same race, other research has found. A broad ruling against the University of Texas affirmative action program could cause colleges to take into account various socioeconomic measures, including income, neighborhood and family composition. Such a step would require an increase in these colleges’ financial aid spending but would help them enroll significant numbers of minority students.
Among high-achieving, low-income students, 6 percent were black, 8 percent Latino, 15 percent Asian-American and 69 percent white, the study found.
“If there are changes to how we define diversity,” said Greg W. Roberts, the dean of admission at the University of Virginia, referring to the court case, “then I expect schools will really work hard at identifying low-income students.”
Ms. Hoxby and Mr. Avery, both economists, compared the current approach of colleges to looking under a streetlight for a lost key. The institutions continue to focus their recruiting efforts on a small subset of high schools in cities like Boston, New York and Los Angeles that have strong low-income students.
The researchers defined high-achieving students as those very likely to gain admission to a selective college, which translated into roughly the top 4 percent nationwide. Students needed to have at least an A-minus average and a score in the top 10 percent among students who took the SAT or the ACT.
Of these high achievers, 34 percent came from families in the top fourth of earners, 27 percent from the second fourth, 22 percent from the third fourth and 17 percent from the bottom fourth. (The researchers based the income cutoffs on the population of families with a high school senior living at home, with $41,472 being the dividing line for the bottom quartile and $120,776 for the top.)
Winona Leon, a sophomore at the University of Southern California who grew up in West Texas, said she was not surprised by the study’s results. Ms. Leon was the valedictorian of her 17-member senior class in the ranch town of Fort Davis, where Advanced Placement classes and SAT preparation were rare.
“It was really on ourselves to create those resources,” she said.
She first assumed that faraway colleges would be too expensive, given their high list prices and the cost of plane tickets home. But after receiving a mailing from QuestBridge, an outreach program for low-income students, she came to realize that a top college might offer her enough financial aid to make it less expensive than a state university in Texas.
On average, private colleges and top state universities are substantially more expensivethan community colleges, even with financial aid. But some colleges, especially the most selective, offer enough aid to close or eliminate the gap for low-income students.
If they make it to top colleges, high-achieving, low-income students tend to thrive there, the paper found. Based on the most recent data, 89 percent of such students at selective colleges had graduated or were on pace to do so, compared with only 50 percent of top low-income students at nonselective colleges.
The study will be published in the Brookings Papers on Economic Activity.
The authors emphasized that their data did not prove that students not applying to top colleges would apply and excel if colleges recruited them more heavily. Ms. Hoxby andSarah Turner, a University of Virginia professor, are conducting follow-up research in which they perform random trials to evaluate which recruiting techniques work and how the students subsequently do.
For colleges, the potential recruiting techniques include mailed brochures, phone calls, e-mail, social media and outreach from alumni. Another recent study, cited in the Hoxby-Avery paper, suggests that very selective colleges have at least one graduate in the “vast majority of U.S. counties.”
Claire E. Jones 