by

Transparency About Legacy Preferences

Universities jealously guard their right to make decisions about whom to admit as a fundamental element of academic freedom. Supreme Court Justice Felix Frankfurter long ago cited “four essential freedoms” of a university: “to determine for itself on academic grounds who may teach, what may be taught, how it shall be taught, and who may be admitted to study.”

These rights are not, however, unlimited. The Civil Rights Act, for example, curtails the ability of both public and private colleges to engage in racial discrimination; and the statute, along with the 14th Amendment, have been applied to limit the ability of universities to engage in affirmative action on behalf of under-represented minorities. Quotas were outlawed in the 1978 Bakke case, and a system awarding bonus points based on race was outlawed in the 2003 Gratz case.

In the area of legacy preference, too, legislative proposals have sought to bring greater transparency to admissions processes. About a decade ago, Senator Edward M. Kennedy introduced legislation to compel universities to make public data on legacy admissions—not to ban legacy preference, but simply to disclose the extent of its use. The higher-education community strongly opposed the idea (Kennedy called the reaction a “firestorm”) and the legislation was dropped.

The issue of transparency arose recently regarding the question of whether one institution—the Massachusetts Institute of Technology—engages in providing a preference to the children of alumni. As part of a debate in the Wall Street Journal with former George Washington University president Stephen Joel Trachtenberg on legacy preferences, I cited, in opposition, among other things, a comprehensive study finding that legacy preferences are not associated with increased university donations and illustrated the point with an anecdote suggesting that the California Institute of Technology, which does not use preferences has raised money at comparable rates to MIT, which does provide special treatment to the offspring of alumni.

MIT cried foul. Chris Peterson, Admissions Counselor for Web Communications at MIT denied that MIT provides, or in past years provided, special treatment of alumni children. “We do not do anything of the kind,” he wrote in a blog post. Peterson correctly traced my source to a 2009 law review article written by attorneys Steve Shadowen, Sozi Tulante, and Shara Alpern, in which the authors wrote: “we also found data showing that alumni of CalTech, which grants no preferences, donated $71-million in 2007 versus $77-million donated in 2006 by alumni of legacy-granting MIT.”

Peterson says the law-review article is mistaken and writes: “The idea that MIT granted legacy, in other words, appeared entirely out of thin air.” The Wall Street Journal ran a correction recently because the editors said they could not confirm definitively that MIT uses or has used legacy preferences in admissions at any point in its history. The controversy is a case study in how opaqueness in admissions plays to the advantage of universities.

Because Kennedy’s legislation never passed, ferreting out whether an institution such as MIT uses legacy preference requires an examination of circumstantial evidence and voluntary disclosures of information.

To begin with, MIT asks all applicants the following question: “If any family members have attended MIT, check the boxes below, give names, class years, and relationship to you” and then lists mother, father, grandparents, and siblings. Unlike virtually every other university which asks this question in order to provide a preference, Peterson says MIT seeks this information for “the opposite” reason. He explains: “One reason we ask students for this sort of information is that so we can give our development office a heads up so we don’t go and solicit folks for large gifts immediately after denying someone they supported for admission.”

Another source of information is the public statements that MIT officials have made over the years regarding the role of legacy status in admissions. The MIT student newspaper, The Tech, reported in 1999 and again in 2001, that some applicants receive “special attention.” On March 16, 1999, the paper reported on an interview with then-MIT admissions dean Marilee Jones, “Another special case the admissions office looks at involves the direct descendants of alumni. Jones said, ‘There is a bit of professional courtesy involved in the cases of the [children or grandchildren] of MIT alumni. If the Admissions Committee turns one of these students down because the student is not one of the top candidates, I will personally review the case to ensure the decision is a sound one. Rarely, though, do I actually change the decision made by the Committee. It’s just one more look.’” If we take her at her word that legacy “rarely” matters, that suggests, of course, that it sometimes does.

MIT has also consistently indicated in surveys that it considers legacy status in admissions. According to former Wall Street Journal reporter Daniel Golden’s 2006 book, The Price of Admission, for example, Caltech was “the only one of the country’s top twenty universities that does not take into account an applicant’s alumni affiliation, according to a U.S. News survey.” (p. 262). (MIT is consistently among the top twenty universities in the survey.) Likewise, research by Chad Coffmann and colleagues in a 2010 book I edited, entitled Affirmative Action for the Rich, counted MIT as a legacy-granting institution based on MIT’s consistent response to the Common Data Set (CDS) survey that legacy status is “considered” in admissions. (See, for example, a cached version of the 2008-2009 survey.)

According to a Wall Street Journal editor, MIT officials now say that the CDS form was filled out incorrectly by the provost’s office—a mistake that was repeated year after year after year (at least from 2003-2009). As Shadowen notes in the comments section to Peterson’s post, in recent days, someone went in and retroactively changed past years’ data to indicate that alumni status was “not considered” by MIT (see the current version of the 2008-2009 survey) though there is no notation that the information was altered.

Finally, we can look at whether admissions rates have differed substantially for legacy and non-legacy candidates at MIT. In the winter of 1997/1998, the Journal of Blacks in Higher Education reported in “Naked Hypocrisy: The Nationwide System of Affirmative Action for Whites,” that while 30 percent of all MIT applicants were accepted, 55 percent of MIT legacy applicants were accepted. The magazine also reported that in 1997, the number of legacies enrolled at MIT (325) outnumbered the number of black students (64) by 5 to 1. (The magazine acquired its information through telephone interviews.) The differential rate of legacy and nonlegacy admissions might reflect the fact that legacy candidates are, on average, 1.83 times as likely to be highly qualified, but that seems unlikely because that ratio falls into the middle of the pack for eight other universities listed in the article—all of which publicly acknowledge using legacy preference.

What is the admissions rate of legacies and non-legacies at MIT today? Peterson says “I don’t have the numbers, and I wouldn’t know how to calculate them from our database, since the information we capture about ‘relation to MIT’ contains not only ‘legacy’ but also a broad swath of things.” Even if admissions rates were higher for legacies, Peterson says, that may be because the children of alumni are more likely to “have had a lot of exposure to math and science than a random applicant.”

But a rigorous 2011 study conducted by researcher Michael Hurwitz of legacy admissions at the elite colleges that make up the Consortium on Financing Higher Education (COFHE) (of which MIT is a member) may be able to shed light on the issue. Analyzing the applications of almost 62,000 students to 30 institutions, Hurwitz found that on average, in 2007, being the child of an alum increased a student’s chance of admissions by 45.1 percentage points. What was the benefit, if any, at MIT? Would MIT be willing to release these data?

Because Senator Kennedy’s legislation never passed the Congress, MIT isn’t bound by law to disclose how legacy status factors into admissions. But if the university has in fact dropped the use of legacy preference, why not stop asking applicants whether mom or dad attended MIT, cease providing special reviews of legacy candidates, and disclose the data on differential rates of admission?

As American universities become increasingly competitive, the call for greater transparency is likely to grow louder. And if the U.S. Supreme Court curtails affirmative action in the coming term in the case of Fisher v. Texas, focus on legacy preferences in particular will likely intensify. Academic freedom—including the right to shape a university’s class—is an important value, but the public right to know whether universities receiving federal funds engage in discrimination based on lineage should not be ignored.

 

Return to Top