Striving for Educational Equity

WASHINGTON -- Six years after they were first published, the data that Anthony Carnevale and Stephen J. Rose produced showing that students from the lowest socioeconomic quartile of Americans were 25 times less likely than wealthy Americans to enroll in the most selective colleges have helped to reshape public policy around higher education.

June 18, 2010

WASHINGTON -- Six years after they were first published, the data that Anthony Carnevale and Stephen J. Rose produced showing that students from the lowest socioeconomic quartile of Americans were 25 times less likely than wealthy Americans to enroll in the most selective colleges have helped to reshape public policy around higher education. In addition to building the case for more federal and state financial support for students from low-income backgrounds, the numbers also helped prompt a group of highly selective public and private institutions to alter their admissions and financial aid policies and practices to focus more on low-income students.

One of those programs, the University of North Carolina at Chapel Hill's Carolina Covenant program, was celebrated Thursday at an event here at which the Century Foundation released a followup to the 2004 book -- America's Untapped Resource: Low-Income Students in Higher Education -- in which Carnevale's and Rose's original analysis appeared.

The new book, Rewarding Strivers: Helping Low-Income Students Succeed in College, includes one chapter on the Carolina Covenant, describing the progress that one highly selective university has made in transforming itself. As described by Edward B. Fiske, the program's initial results, in terms of low-income students' access to and success at North Carolina, are promising.

But that upbeat assessment is more or less overwhelmed by the book's new analysis from Carnevale, this time co-written with Jeff Strohl, who works with the economist at Georgetown University's Center on Education and the Workforce. Their contribution to the volume edited by Century's Richard D. Kahlenberg argues that social, racial and ethnic stratification in higher education has actually increased in recent years, despite the fledgling efforts by the most elite colleges.

That's in large part because in the last decade and a half, more colleges have ramped up their selectivity, driven by "check-writing parents sending their kids to test-prep programs chasing selective colleges that become selective by raising test score requirements and investing mightily in their faculty and equipment," Carnevale said Thursday. Selectivity in this case is defined heavily by the traditional measure of standardized test scores, on which underrepresented minority students and those from low-income backgrounds historically score much lower.

Since 1994, the numbers of four-year colleges in the top two tiers of selectivity as measured by Barron's have increased sharply, while the third tier has stayed flat and the fourth is "melting," as Carnevale described it, seen in the table below:

Number of Four-Year Colleges, by Selectivity, 1994-2006

Selectivity Level 1994 2006
Most and Highly Competitive Colleges 146 193
Very Competitive Colleges 253 279
Competitive Colleges 578 572
Less and Non-Competitive Colleges 429 299

As more colleges have moved up the "quality" scale, as measured by Barron's, the research by Carnevale and Strohl shows, the institutions have become slightly more racially and ethnically diverse, but students from lower-income backgrounds have made virtually no progress in gaining access to more selective colleges. By 2006, students from the lowest socioeconomic quartile made up 5 percent of students at the most competitive colleges, 7 percent of students at highly competitive colleges, and 8 percent of students at very competitive colleges, up from 3, 4 and 8 percent, respectively, in 1982. They end up disproportionately in nonselective four-year colleges and in open-access two-year institutions.

When one considers the differences between the inputs and outcomes at the more selective institutions and at nonselective ones -- per-student spending that is 4-5 times as great, and far higher graduation rates and entry-level earnings of students -- the stratification by socioeconomic income (and to a lesser but still meaningful extent by race and ethnicity) means that the higher education system operates as an "engine of inequality," Carnevale said Thursday. "If this were K-12 education, we'd be in court" over the differences in how low-income and other students are treated in what he called a "dual system" of higher education, Carnevale said, noting the major lawsuits that have been filed (and often won) in many states over inequality of access to elementary and secondary education.

Potential Changes

How might the stratification be eased? Programs like those adopted at North Carolina and other public universities (including Colorado State University, which announced a similar plan of its own Thursday) and at private elite colleges like Princeton and Harvard Universities certainly help a little, but such "boutique" approaches "will probably not change the underlying systems trend toward greater postsecondary stratification," Carnevale and Strohl write. At Kahlenberg's request, their contribution to Rewarding Strivers examines the impact that affirmative action (perhaps of a different sort) might have in ameliorating the inequity of access and outcomes.

Using regression analysis, the researchers sought to calculate the relative weight of the impact of various traits and types of disadvantage on SAT scores, to try to measure how much effect class and demographic obstacles might pose in the admissions process at selective institutions. A student loses 48 points if one's father is a laborer (compared to being a physician), for example, and 41 points if his or her family has no college savings compared to a comparably qualified student from a family with $40,000 in accumulated wealth.

Given the political and legal vulnerability of race-based affirmative action, Carnevale said that he and Strohl "tried mightily to find a way to replace race with various measures of income class," to see if they could end up arguing that purely class-based affirmative action (which is likely to be more palatable to many critics) would suffice. While they managed to whittle down the effects attributable to race -- being black (as opposed to white) amounted to an SAT penalty of 56 points, they found -- "no matter what we did, we couldn't erase the stain of race" entirely. "It was incredibly stubborn."

Taken together, the researchers argue that the cumulative effects on SAT scores for students from highly economically disadvantaged backgrounds are huge -- hundreds of points, in many cases, enough to make them unlikely to be successful in selective college admissions processes that often lean heavily on standardized scores. Instituting a system of socioeconomic-based affirmative action that took such disadvantage into account (on top of race- and ethnicity-based preferences, which they conclude are still necessary) could certainly begin to erase the class stratification in higher education.

But given the tendency of colleges to pursue status (often defined by high test scores, etc.), and the desire of parents to get their children into the selective institutions, trying to use that sort of affirmative action would be a "bit like spitting in the wind against that tide," Carnevale said. "It would be a huge undertaking, and probably unimaginable," to change institutional behavior in that way.

So "if you can't move low-income and minority kids en masse into the high-quality systems" of colleges, Carnevale said, the likelier alternative to improving the lot of students ill-served by higher education is to strengthen the quality of the institutions they do attend -- "two-year schools and lower-end four-year colleges." The Obama administration (which Carnevale has advised in both formal and informal ways) took steps in this direction with its proposed American Graduation Initiative, which would have poured $10 billion into community colleges, but had to be scaled back significantly.

But Carnevale said most of the action will be up to the states, as they consider rewriting funding formulas to reward institutions based on performance (enrolling and graduating low-income students, etc.), bolstering student services programs at community colleges, and encouraging students to get educated at institutions that cost less, but still have high quality.

"Instead of continuing to struggle to move more students into selective colleges where the high-priced quality programs reside, we may be more successful moving money and quality programs to the community colleges where most of our students reside," he and Strohl write at the conclusion of their long chapter of Rewarding Strivers.


Back to Top