(Originally published in Inside Higher Ed’s “Admissions Insider” on December 9, 2024)
What explains the fascination, perhaps even obsession, with attending an elite college or university? It may be our culture’s fixation on luxury brands of all kinds, exemplified at this time of year by television commercials telling us that the best way to get in the spirit of the holidays is buying or leasing a Lexus or two. Or it may lie in the fact that almost all news coverage of college admission focuses on a small group of highly-selective institutions as if they represent the entire universe of higher education. When’s the last time you read an article in The New York Times or The Wall Street Journal about Rhodes College or Southeastern Oklahoma State University?
The most likely answer, however, is the widespread belief that attending an elite college is an express lane to economic success and happiness (with the two often seen as being connected or even the same thing).
A new working paper published by the National Bureau of Economic Research suggests that elite colleges remain enclaves of economic privilege rather than engines of economic and social mobility, with levels of socioeconomic diversity relatively unchanged from where they were 100 years ago. The paper, “The G.I. Bill, Standardized Testing, and Socioeconomic Origins of the U.S. Educational Elite Over a Century,” was authored by researchers Ran Abramitzky and Jennifer K. Kowalski of Stanford University along with Santiago Pérez from the University of California at Davis and Joseph Price from Brigham Young University.
The researchers compiled a dataset of records for 2.5 million students attending 65 elite colleges between 1915 and 2013. The 65 colleges include the 12 “Ivy-Plus” institutions (the eight Ivies plus Duke University, Massachusetts Institute of Technology, Stanford and the University of Chicago), 15 Northeastern liberal arts colleges, eight historically women’s colleges, and 31 public universities that have been listed as “public Ivies.” (If you are counting, that adds up to 66, but Vassar is included in both the liberal arts and women’s colleges groups.) The researchers tracked down college yearbooks and institutional records to determine enrollment for individual students, and then used Census data to identify family backgrounds for those students.
The paper concludes that for all the changes in American society and access to higher education over the past century, including increased opportunities for women and students of color, the representation of lower-income students at elite colleges has remained essentially unchanged. Elite colleges have become more racially and geographically diverse, but not more economically diverse. The representation of Black students at elite colleges has increased 4-5 times since the 1920s, and the percentage of students attending an elite college in a different part of the country has about doubled, from 30 percent to 60 percent at elite private colleges since 1950, and from 10 to 20 percent at elite public colleges since the 1960s. Meanwhile the percentage of low-income students attending elite colleges has been stable–and low.
In the early 1920s, 8 percent of those attending college (any college) came from the bottom 20 percent of family incomes. Today the percentage of college students coming from the bottom income quintile is 20 percent for women and 13 percent for men. The percentage for the elite college group, however, has remained essentially flat, at between 3-5 percent for elite private institutions and 5 percent for elite publics (the proportion is higher for the women’s colleges, which saw the most change over time).
This trend holds even at the wealthiest universities, with the paper reporting that “the share of lower-income students at Harvard, Yale, and Princeton has been less than 5 percent through the entirety of the last century.” Duke’s percentage dropped from 8 to 3 from the 1920s to the early 2000s, while liberal arts colleges Bates and Colby dropped from 8 to 2. Those numbers are consistent with the conclusions of a 2017 study by Raj Chetty and colleagues, which reported that 38 elite colleges enroll more students from the top one percent of incomes than from the bottom 60 percent.
As the title of the paper indicates, the study gives special attention to two developments that are widely believed to have broadened socioeconomic diversity in colleges. The first is the G.I. Bill providing educational and other benefits to World War II veterans.
One of the interesting nuggets in the research is that there is a dramatic difference in the likelihood of serving in World War II between those born before the last quarter of 1927 and those born after, in 1928, who would have come of age after the end of the war. Those born before the last quarter of 1927 were 40 percentage points more likely to have served. That hit home for me because my father was born in December of 1927.
The researchers used the gap between the two groups to determine if there is a significant difference in the socioeconomic status of those attending elite colleges, and they concluded that the difference was almost negligible. The G.I. Bill did not produce a cohort of veterans from lower- or even middle-income families attending elite colleges through their veteran’s benefits. The report offers two possible considerations. One is that the percentage of WWII veterans from the bottom socioeconomic groups had a much lower high school graduation rate (35 percent) than those from the top 20 percent (closer to 65 percent), meaning that they lacked the academic credentials to attend an elite college. The authors also posit that the very experience of being in the war could have interrupted an individual’s educational plans.
The second development discussed in the paper is whether standardized tests help elite colleges identify talented students they otherwise wouldn’t find, the so-called “diamond in the rough” hypothesis. That was one of the original justifications for the SAT, and it was used last spring by places like MIT and Dartmouth College as a reason for reinstating testing requirements.
The paper looks at the socioeconomic composition of the student bodies at elite colleges both prior and following their adoption of the SAT, and concludes that the introduction of standardized testing “was not associated with persistent changes in the socioeconomic mix of students at elite colleges in general, with even more limited effects when focusing on private colleges.” The researchers admit that they can’t assess how the introduction of standardized testing may have influenced applicant pools, giving students from non-traditional educational or low socioeconomic backgrounds the confidence to apply to an elite college, but the evidence in this study doesn’t back the idea that testing helps broaden economic diversity at the Ivies and other elite places. If there is evidence to the contrary that drove a number of elite colleges to return to requiring testing, I’d love to see it.
So what conclusions should we draw from this study?
First of all, the idea that any institution or industry is largely unchanged and serving the same clientele as 100 years ago is shocking, and something to be embarrassed about. That is particularly true for higher education, which is predicated on giving individuals the tools to transform their lives. The potential good news is that the most recent data used in the study is from 2013. Recently we have seen initiatives to make attending elite colleges more affordable for the middle class. Could those make a difference?
Second, there is a conflict here between self-interest and the public interest. Elite private colleges certainly have a right to admit whomever they choose, but given their outsized role in American society they have a duty to serve the public interest as well. At a time when higher education and other institutions and norms are already under attack, failure to do so will increase scrutiny from forces skeptical of the value of college.
Finally, we need an ongoing conversation about what we mean by calling a college or university “elite.” Are colleges considered elite because they offer a superior education or because they are socially and economically exclusive? Should elite colleges reflect the culture or strive to be counter-cultural?
Legendary football coach Bill Parcells said, “You are what your record says you are.” The record of elite colleges in promoting economic and social mobility over the past 100 years isn’t pretty. It’s time for that to change.