Harvard University has a $25 billion endowment and in 2003-4, only 6 percent of its undergraduates were of sufficiently modest means to qualify for Pell Grants. While Pell eligibility varies based on a number of factors, only 5 percent of Harvard undergraduates that year came from families with incomes less than $30,000.
At Trinity University, in Washington, there's a lot less money in the bank -- but a much larger share of students are getting Pell Grants. The endowment is about $9 million. In 2003-4, fully half of Trinity's students were poor enough for Pell Grants, and 26 percent came from families with incomes less than $30,000.
If the comparison makes anyone in Cambridge squeamish (or just has someone objecting to the comparison's fairness), that's precisely the point of a new Web site, Economic Diversity of Colleges,  which is being unveiled today. The Web site features data from about 3,000 colleges -- taken from reports that the institutions file with federal agencies. Tools on the Web site allow users to browse institutions or to set up groups -- by geography or institution type, for example -- for comparison purposes.
In addition to information about Pell Grant recipients and family income levels, the site also features data on levels of student indebtedness and tuition costs. The Web site doesn't rank colleges or suggest what an appropriate level might be for percentages of Pell recipients or families with incomes at certain levels. "We decided not to say that this is good or this is bad. These are the kinds of discussions institutions should be having," said Robert Shireman, director of The Institute for College Access and Success,  a research organization that created the new Web site.
There are some obvious limitations to the site that Shireman acknowledges. While Pell Grants are frequently used as a proxy in measuring economic diversity, they aren't the only measure. And some institutions, such as low-cost community colleges, may have low percentages of Pell recipients because tuition just isn't that much and students therefore might not need the federal grants to afford to attend.
Likewise, the data don't reflect problems in public education (many low income students may attend high schools that don't put them in the pool for competitive colleges) and the data don't reflect a number of changes adopted by colleges (including Harvard) in the past two years to make aid packages more generous to low-income students.
Elite colleges continue to adopt new aid policies -- Stanford University  announced on Wednesday that families with incomes of less than $45,000 would no longer need to contribute anything toward tuition. The Massachusetts Institute of Technology  announced last week that it would match Pell Grants. But the new data drive home that many of the most prestigious and most wealthy colleges in the country -- public and private -- aren't necessarily the leaders when it comes to educating low-income students. And institutions with limited financial resources are managing to attract such students.
Patricia McGuire, the president of Trinity, says that she thinks "it's shameful" that the wealthiest institutions in the United States aren't educating more low-income students. "They make the big headlines when they drop the loan requirements for students, but that's a drop in the bucket, considering their condition," she said.
McGuire noted that Trinity, a Roman Catholic institution, once educated a much wealthier -- and overwhelmingly white -- student body. As the area around Trinity diversified and as it became clear that there were not enough colleges reaching out to low-income minority women, McGuire said that the college shifted focus and started providing more of its own aid to such students, on top of the Pell Grants that they receive. Of low-income students, she said, "this is who we are called to serve today," adding that more colleges should be reaching out to such students.
Elite colleges frequently note that with their highly competitive admissions, many in their applicants pools are in fact wealthy. But Shireman said that there were clearly many more low-income students who could succeed at top institutions that currently enroll relatively few such students. And his data show that other competitive institutions were managing to get significantly larger shares of Pell recipients than Harvard was two years ago. Cornell University's Pell percentage was 17 percent, nearly three times that of Harvard.
There's also evidence from Harvard itself that policy changes can make a difference. A study of the impact  of recent aid changes at the university found that they were resulting in increases in the numbers of low-income students enrolling.
Not surprisingly, the data show that community colleges -- which cost much less than four-year institutions -- are serving many of the nation's poorest students. Alabama Southern Community College has half of its undergraduates receiving Pell Grants and 38 percent come from families with incomes less than $30,000.
Many historically black colleges also show high enrollments of students with few resources. To take two that have been in the news because of the damage that they suffered from Hurricane Katrina, 48 percent of the undergraduates at Xavier University and 69 percent of those at Dillard University received Pell Grants. Similarly, many Hispanic-serving institutions are enrolling large numbers of low-income students. At the University of Texas at Brownsville, for example, the Pell percentage is 53 percent and 36 percent of students come from families with incomes below $30,000.
Not all public institutions will like the comparisons that jump out. In Wisconsin, for example, Beloit College's tuition and fees are five times that of the flagship campus in the state, the University of Wisconsin at Madison. But Beloit graduates leave with only marginally more debt than Madison graduates ($20,339 to $17,528). And Beloit outpaces Madison in the percentage of families with incomes under $30,000 (8 percent to 4 percent) and the percentage of undergraduates receiving Pell Grants (16 percent to 11 percent).
"It looks like we cost five times as much, but we don't," said John E. Burris, Beloit's president. He suggested that public institutions would be better able to reach a broader range of students by increasing tuition and putting substantially more money into financial aid.
Madison is hardly unique. In many states, similar comparisons could be made, with flagship public universities appearing to have smaller shares of low-income students than those enrolled in other sectors. At the University of Georgia, only 12 percent of undergraduates receive Pell Grants; the figure is 14 percent at the University of Minnesota-Twin Cities and 13 percent at the University of Colorado at Boulder.
Shireman said that he hoped such comparisons would prompt state legislators and educators to ask whether flagships were focused too much on competitiveness as opposed to serving state populations broadly. David Edelson, a spokesman for the National Association of State Universities and Land Grant Colleges, said officials there hadn't yet seen the data being released. But he questioned whether it was fair to compare figures from large research universities and small liberal arts colleges. He said that there was no doubt that flagship publics are serving many low-income students and working to serve still more. "Percentages don't tell the whole story," he said.
While many flagships don't fare well in the new database, there are some notable exceptions -- especially from campuses in the University of California and the State University of New York systems. SUNY's university centers at Albany, Binghamton, Buffalo and Stony Brook have Pell enrollment percentages of 28, 28, 34 and 43 percent, respectively. Berkeley's percentage is 29 and UCLA's is 30.
Shireman said that he hoped colleges would ask a series of questions about their data:
- Are they enrolling as many low-income students as might be possible?
- If they are enrolling many low-income students, are those students able to find and obtain federal student aid they should be eligible for?
- Are students borrowing too much money, based on total costs, and borrowing patterns are institutions with similar expenses.
"There should be different goals for different kinds of institutions," Shireman said.