News your connection to The Boston Globe

The Chosen Few

The elite colleges are more diverse than ever, but the students being admitted are still overwhelmingly wealthy. The author of a new book argues it's time that the Ivies welcomed more middle-class and disadvantaged students - and he shows them how.

HAD FRANKLIN DELANO ROOSEVELT returned to Harvard College in the fall of 2000, precisely a hundred years after he had patiently waited in line in front of Sever Hall to register as a freshman, he would have been shocked. As he looked around, he would have seen immediately that nonwhites (half of them Asian-Americans) made up over a third of the freshman class. Even more striking would have been the presence of women, now walking confidently through the Yard, where they constituted nearly half of the students.

Less noticeable, perhaps, would have been the extraordinary growth in the number of Jews, whose proportion of the freshman class had tripled from 7 percent in his time to more than 20 percent a century later. But what Roosevelt would have grasped above all was that Harvard was no longer the private property of his social group: white, Anglo-Saxon, Protestant men, largely from the upper classes of New England and the mid-Atlantic states. Once the unquestioned rulers not only of Harvard (where they made up approximately 85 percent of students when FDR was an undergraduate) but also of virtually all the major institutions of American life, WASP men were in 2000 a small and beleaguered minority at Harvard - no more than 20 percent of freshmen.

The transformation of Harvard, Yale, and Princeton from enclaves of the Protestant upper class into institutions with a striking degree of racial, ethnic, and religious diversity was by any standard historic. Yet beneath this dramatic and highly visible change in the physiognomy of the student body was a surprising degree of stability in one crucial regard - the privileged class origins of students at the Big Three. By 2000, the cost of a year at Harvard, Yale, and Princeton had reached the staggering sum of more than $35,000 for tuition and room and board - an amount that fewer than 10 percent of American families could afford. (In 2004, annual expenses had risen to well over $40,000 per year.) Yet at all three institutions, a majority of students were able to pay their expenses without scholarship assistance compelling testimony that, more than 30 years after the introduction of need-blind admissions, the Big Three continued to draw most of their students from the most affluent segments of American society.

While "paying customers" still constituted the majority of students in 2000, students from modest backgrounds continued to be vastly underrepresented not only at the Big Three but also at highly selective colleges nationwide. At Princeton, which offered the most generous financial-aid packages in the Ivy League, students from families in the bottom 50 percent of the income distribution made up just 10 percent of the freshman class. At Harvard, the number was a bit higher, at roughly 12 percent, but the median family income for 2004 freshmen was roughly $150,000. At Yale, though no statistics are available on the income distribution of the freshman class, the paucity of students from families with limited education suggests a similar pattern; in 2001, 8 percent of Yale freshmen came from homes where neither parent had attended college.

Despite their consistent emphasis on "diversity," highlighted in an amicus brief submitted to the US Supreme Court in the landmark University of Michigan case on affirmative action, the Big Three are notoriously lacking one of its most critical dimensions: class diversity. In a study of the percentage of low-income students in 2000 (as measured by the proportion of federal Pell Grants - need-based awards that do not have to be repaid and make up the bulk of many poorer students' aid) at the nation's leading universities, the Big Three were found to be among the nation's least economically diverse schools. Of the 40 universities studied, Harvard and Princeton ranked 39th and 38th respectively, with Yale at 25th. While the three top universities in economic diversity were all public institutions (the University of California at Los Angeles, UC-Berkeley, and UC-San Diego), the next two - the University of Southern California and New York University - were private. And one university in the top 10, California Institute of Technology, is among the most selective private institutions in the nation.

The Big Three and other highly selective private colleges have not been unaware of the paucity of students from poor and working-class backgrounds at their institutions. In their landmark study published in a 1998 book, The Shape of the River, William Bowen and Derek Bok reported that only 1 percent of white students at the most selective institutions came from backgrounds with low socioeconomic status.

While troubled about this finding, the authors - former presidents of Princeton and Harvard, respectively - made it clear that they believed there was little that could be done. Committed to maintaining "high academic qualifications," as measured by grades and test scores, they concluded that "the problem is not that poor but qualified candidates go undiscovered, but that there are simply very few of these candidates in the first place."

But such resignation is unwarranted, for there is, in fact, a significant pool of students from disadvantaged backgrounds with "high academic qualifications," at least as measured by the SAT. In 2004, the number of students with family incomes under $30,000 who scored 650 or higher on the 800-point math and verbal sections was 12,755 and 6,995, respectively; for students from homes where neither parent had more than a high school degree, the parallel figures were 22,477 and 14, 812. In a nation in which more than 1.4 million students took the SAT in 2004, this is a modest pool. For purposes of comparison, however, it is worth noting that it is far larger than the comparable pool of African-Americans with scores over 650: 2,962 (math) and 3,039 (verbal) in 2004.

Yet, despite the smaller pool of high-scoring black students, elite college administrators have shown no such resignation with respect to race, having noted with alarm that a race-neutral policy would reduce black enrollment at elite private colleges to perhaps 2 percent or less. Indeed, it was the prospect of such a reduction that led to a massive mobilization on the part of the nation's most prestigious universities to preserve race-based affirmative action in the face of serious legal and political challenges.

The point here is not that class-based affirmative action should replace race-based affirmative action; it is rather that the radical underrepresentation of students from modest socioeconomic backgrounds at the nation's elite colleges is a serious problem demanding immediate attention. Bowen himself has recently recognized this, reversing his position and issuing a compelling call for the adoption of class-based affirmative action - a "thumb on the scale" for applicants from lower socioeconomic categories, to be put in place alongside (and not as a substitute for) the race-attentive policies that have long existed at the selective colleges. Harvard's president, Lawrence Summers, has issued a similar appeal for preferences for the socioeconomically disadvantaged, noting in a major speech before the American Council on Education that "in the most selective colleges and universities, only 3 percent of students come from the bottom income quartile and only 10 percent from the bottom half of the income scale."

The heavy emphasis on race in the admissions decisions of the elite colleges and the relative lack of attention historically accorded class is hardly surprising. While there has been little political mobilization in recent decades on the issue of class inequality, the issue of racial injustice spawned the nation's greatest political mobilization of the 20th century - the civil rights movement. With the foundations of the social order seemingly at risk of crumbling in the late 1960s, important concessions were extracted by racial minorities and then institutionalized. The imprint of that pivotal period is with us still, inscribed in the definition of merit we now take for granted. That no parallel redefinition of merit occurred with respect to social class - an equally powerful source of inequality of opportunity - reflects the fact that subordinate social classes in the United States today lack the political power attained by racial and ethnic minorities after years of arduous struggle.

The new and more academic definition of "merit" that emerged at the Big Three in the decades after World War II did not raise the number of poor and working-class students in the freshman class. On the contrary, the increasingly rigorous academic requirements for admission - among them, the rising scores on the SAT - may well have been associated with a decline in opportunities for students from modest backgrounds. At Harvard, between 1952 and 1963 - a period when average combined SATs rose from 1181 to 1401 - the proportion of freshmen whose fathers had not attended college plummeted from 38 to 16 percent. Though part of this decline could surely be attributed to rising rates of college attendance during the interwar period (when most of the fathers of students entering Harvard between 1952 and 1963 would have attended college), the magnitude of the drop was far too large for it to have been the primary explanation for such a dramatic change.

If the growing emphasis on high grades and high test scores in the early decades of the Cold War further reduced the already meager chances of students of low socioeconomic status to enter the Big Three, it did promote a certain degree of social mobility. But the main beneficiaries were the children of families that, while lacking the wealth of the old upper class, were richly endowed with cultural capital. In 1956, the sons of business executives (22 percent of all freshmen) outnumbered the sons of professors (5 percent) by a ratio of more than 4 to 1; in 1976, the sons of professors - who constituted perhaps one-half of 1 percent of the American labor force - made up more than 12 percent of Harvard freshmen, compared with 14 percent for the sons of business executives. At Princeton, a similar shift was visible; whereas in 1954, the sons of businessmen outnumbered those of elite professionals by a ratio of 2.5 to 1 (50 percent to 20 percent), by 1976, the gap had narrowed to 1.4 to 1 (32 percent to 23 percent). Under the more academically demanding admissions regime that prevailed by the 1960s, the scholastically brilliant children of the middle and upper middle class enjoyed greater opportunities than ever before to enter Harvard, Yale, and Princeton. At the same time, the less academically talented children of the old elite were finding it increasingly difficult to gain access to colleges where admissions had until recently been taken for granted. To be sure, children of wealth and social standing remained vastly overrepresented. But unless they could demonstrate a high level of academic accomplishment, even the children of the old upper class faced the prospect of rejection. The specter of downward mobility now extended to the most privileged sectors of the old elite - a development that would enduringly transform the atmosphere of the Big Three.

THE STRANGE SYSTEM of elite college admissions that we have inherited from the past may prove surprisingly vulnerable to public scrutiny and debate. Many of the features we have long taken for granted were created at particular historical moments; the time has come to eliminate or modify several of them. It is with the goal of stimulating further scrutiny and debate that I offer several possible avenues of reform.

LEGACIES: Legacy preference flagrantly violates the core American principle of equality of opportunity, which requires that rewards be based on individual accomplishment. Historically, one of the main arguments for favoring legacies was that private institutions depended heavily on the largesse of their alumni; however, with the endowments of Harvard, Yale, and Princeton in 2004 at $22.6 billion, $12.7 billion, and $9.6 billion, respectively, this argument - which was never one of high principle - has become much less persuasive. The time has come to consider whether legacy preference has any place in universities publicly committed to equality of opportunity. As a first step, the passage of legislation - such as the bill drafted by Senator Edward Kennedy (himself a legacy, Harvard '56) - that would require universities "to publish data on the racial and socioeconomic composition of legatees" would seem warranted. So, too, would a requirement that universities disclose the admission rates of legacies and non-legacies to cast a spotlight on a policy that former senator John Edwards of North Carolina has rightly characterized as "a birthright out of 18th-century British aristocracy, not 21st-century American democracy."

EARLY ADMISSIONS: Early decision and early action programs, used by virtually all of the highly selective private colleges, also pose a threat to equality of opportunity. The problem with these programs is twofold: Students who use them enjoy an advantage in the admissions process, and students admitted on early decision (though not early action) are obliged to attend the accepting institution and hence cannot compare financial-aid offers. Both patterns further advantage the already advantaged. It is generally affluent and well-informed students who benefit from the boost given to early applicants; moreover, the inability to compare financial-aid packages precludes many students who need scholarship assistance from applying early. Students from less-affluent families - who often attend secondary schools where both information and counseling are in short supply - thus face a playing field slanted even more against them as a result of early admission.

From the viewpoint of equality of opportunity, there is no substitute for eliminating early admission programs and replacing them with a standard notification date for all applicants. In the absence of such action, sunshine legislation calling for greater transparency would be appropriate. Just as Congress requires colleges to release graduation rates by race, so it could require that institutions with early admission programs publicly disclose the racial and socioeconomic backgrounds of the students so admitted, the acceptance rate for early and regular candidates, and the proportion of places in the freshman class filled by early applicants.

ATHLETES: Viewed from a distance, one of the strangest features of admissions at the Big Three is the heavy preference for talented athletes. They are, after all, among the leading research universities in the entire world, and the ability to throw, catch, kick, hit, or bounce a ball would not at first glance seem to be a relevant qualification. Yet the policy has deep roots, its origins dating to the football mania that swept the Big Three in the late 19th century. Though the Ivy League has made vigorous efforts - including a ban on scholarships specifically for athletes - to keep the competition for outstanding athletes within bounds, the preference they get in the admissions process remains substantial. The stakes remain high, because athletes are apportioned a sizable slice of the admissions pie; at Princeton in 1997-1998, for example, varsity athletes constituted 22 percent of all the men.

Recognizing that the recruitment of athletes has compromised their core academic mission, the Ivy League has recently taken some measures to bring the situation under control. In June 2003, the Ivies voted to cap the number of athletes admitted each year to play in the 33 sports sponsored by the league, to raise the academic requirements for admission, and to reduce the number of places reserved for football players from 35 to 25 per college. But more drastic reform is needed, for too many applicants are admitted almost exclusively on the basis of their athletic ability. According to the Big Three's official ideology, skill in athletics is a legitimate form of "merit," but one that should be no more valued in the admissions process than exceptional ability in music, art, drama, dance, and debate. Putting this ideology into practice would not necessarily reduce the number of students participating in varsity athletics. It would, however, almost certainly reduce the number of recruited athletes, freeing up precious places in the freshman class for students more willing and able to take advantage of the intellectual opportunities available in great research universities.

CLASS DIVERSITY: The Big Three are conspicuously lacking in significant representation of the poor and the uneducated. Though hailing diversity as indispensable to their mission, the Big Three are, as noted earlier, among the least economically diverse of the nation's major research universities.

The paucity of working-class students at the Big Three and other highly selective institutions contradicts their professed commitment to both diversity and equality of opportunity and threatens to undermine their legitimacy by reinforcing their image as citadels of privilege. Nor do low-income candidates receive the kind of preference still accorded legacies and recruited athletes - groups that are, according to a study of 19 selective colleges and universities (three of which were Harvard, Yale, and Princeton) by William Bowen and colleagues, 30 and 20 percentage points, respectively, more likely to be admitted, controlling for SAT scores. Indeed, contrary to the repeated claims of elite colleges, says Bowen, "applicants from low socioeconomic backgrounds, whether defined by family income or parental education, get essentially no break in the admissions process."

But remedying the massive underrepresentation of poor and working-class students will not be easy. It will require, at the very least, a forthright acknowledgment of the problem as well as the adoption of specific measures to address it. Class-based affirmative action programs of the type developed at the University of California in the wake of Proposition 209 (which banned race-based affirmative action) would be a useful first step. But only a redefinition of merit that acknowledges the profound differences in educational opportunity holds a real possibility of bringing more than token class diversity to the Big Three.

Taken together, reforms in these four areas would bring the Big Three a bit more into conformity with their professed ideals but would not dramatically transform them. Yet the tendency of universities to place institutional interests over the interests of students and the broader society suggests that even such modest measures are unlikely to be implemented unless powerful pressure - whether internal, external, or both - is applied. Real change does not come without cost; it is possible, for example, that the elimination of early admission programs might place the Big Three at a competitive disadvantage in the "positional arms race" in higher education and that this disadvantage might even be reflected in a drop in the despised but feared national rankings. But is it too much to ask the leaders of our most prestigious institutions of higher education - institutions that constantly proclaim their commitment to the ideals of meritocracy and inclusion - that they exhibit the same integrity and firmness of character they demand of their applicants?

Excerpted from The Chosen: The Hidden History of Admission and Exclusion at Harvard, Yale and Princeton, by Jerome Karabel, to be published by Houghton Mifflin Co. in October. Copyright © 2005 by Jerome Karabel. Reprinted by permission. All rights reserved. E-mail comments to

Jerome Karabel is a professor of sociology at the University of California at Berkeley and a senior fellow at the Longview Institute.

(Illustration / Michael Witte)
top magazine articles
Today (free)
Yesterday (free)
Past 30 days
Last 12 months
 Advanced search / Historic Archives