Almost two years have passed since the U.S. Supreme Court embraced the principle that our colleges can take race into account, as one of several factors, in admitting classes. We believe that the time has come for America’s wealthiest and most selective institutions to put another “thumb on the scale,” as a complement to race-sensitive admissions, for students from socioeconomically disadvantaged backgrounds.
Many highly selective colleges are officially “need-blind” in admissions (meaning that a student’s financial need is not weighed in the admissions process). However, in their amicus brief in the Michigan affirmative-action cases, the presidents of a number of those institutions also stated their commitment to do more by giving “special attention to applicants from economically and culturally disadvantaged backgrounds ... [and] those who would be the first in their families to attend any college.” Until now, no one has known whether that statement reflects actual practice. Thanks to the cooperation of 19 selective institutions and the assistance of the College Board, we now have the data to answer that question.
The answer is no. Although early-decision applicants, legacies, members of underrepresented minority groups, and recruited athletes are anywhere from 19 percentage points (early applicants) to 30 percent (recruited athletes) more likely to be admitted than otherwise similar applicants, candidates from low-income families and those who would be the first in their families to attend college have an admission rate that is virtually identical to that of more-privileged applicants with the same test scores, grades, race, and other characteristics. Those students are not receiving any kind of “admissions advantage”; nor are they being penalized for their disadvantaged backgrounds, as some observers have suggested.
That finding needs to be considered in the context of an applicant’s chances of having made it into the “credible admissions pool” in the first place. Our research demonstrates that the odds of achieving SAT scores that will make an individual competitive for admission to a selective institution are six times higher for an applicant from a high-income family than for one from a poor family, and seven times higher for an applicant whose relatives have attended college than for one who is a first-generation college-goer. Simply put, poor families have great difficulty investing sufficient personal and financial resources to prepare their children to attend college, do well, and graduate.
Today students whose families are in the bottom income quartile represent roughly 11 percent of all matriculants at the 19 institutions we studied, and first-generation college students represent a little more than 6 percent. When we combine the two measures and estimate the share who are both first-generation college-goers and from low-income families, we get a figure of about 3 percent. Needless to say, those proportions are significantly lower than the proportion of those groups in the national population of college-aged young people.
In the long run, the only way to ensure educational access is to attack the problem at its roots by improving the health, social environment, and precollegiate education of disadvantaged students. But in the meantime, colleges can do more to close the gap in both educational opportunities and life outcomes.
Why should that be a policy priority? Colleges have an interest in the educational value that diversity brings, and democratic principles demand an educated citizenry and leaders drawn from all walks of life. There is also a clear economic interest in expanding the share of our population with the skills that top-quality higher education affords. Beyond that, fairness and efficiency recommend giving extra weight to the accomplishments of qualified applicants who have “bucked the odds” by making it into the applicant pool. Indeed, the disadvantaged students who currently enroll do very well academically (as well as otherwise-similar students with comparable test scores, grades, and other characteristics). For all of those reasons, institutions that can afford to move beyond need-blind admissions when considering the applications of disadvantaged students should do so.
Some commentators have suggested that income-based or class-based preferences could, in fact, replace race-conscious admissions. When we simulated such a change, we found that the share of students from underrepresented minority groups fell by nearly half -- although African-Americans, Hispanics, and Native Americans are disproportionately represented among socioeconomically disadvantaged college applicants, the vast majority of disadvantaged applicants are white. Our simulation also showed that adding income-based preferences to current race-conscious admissions policies increased the share of students from the bottom income quartile by half and had little effect on the academic profile of the enrolled class. If current financial-aid policies were maintained, colleges would have to increase their financial-aid budgets by about 12 percent -- certainly not a small amount, but quite possibly affordable for the 30 or 40 selective institutions that are among the wealthiest.
If America’s leading colleges are to continue their progress toward becoming “engines of opportunity” rather than “bastions of privilege,” complementing affirmative action with a “thumb on the scale” for academically qualified but socioeconomically disadvantaged students seems the appropriate next step.
William G. Bowen is president of the Andrew W. Mellon Foundation and former president of Princeton University; Martin A. Kurzweil is research associate at the foundation and a student at Harvard Law School; and Eugene M. Tobin is program officer for the Liberal Arts Colleges Program at the foundation and former president of Hamilton College. They are the authors, in collaboration with Susanne C. Pichler, of Equity and Excellence in American Higher Education, to be published in April by the University of Virginia Press.
http://chronicle.com Section: Admissions & Student Aid Volume 51, Issue 25, Page B18