Since the Supreme Court ruling in Students for Fair Admissions (SFFA) v. Harvard and SFFA v. University of North Carolina in June 2023 that repealed race-conscious admissions policies, many observers have wondered what would happen to the racial makeup of elite universities. In the past, such schools have proudly advertised the data on the racial makeup of incoming freshmen. So far this year, most have remained strangely silent.
Last week, however, MIT broke the silence by reporting that the percentage of underrepresented minorities enrolling had precipitously dropped. Whereas black and Hispanic students, respectively, made up 15 percent and 16 percent of MIT’s Class of 2027, the Class of 2028 is just 5 percent black and 11 percent Hispanic. Meanwhile, Asian Americans have increased their enrollment from 40 percent to 47 percent, while the white share stayed essentially unchanged at 37 percent. That Asian Americans were the primary beneficiaries of the removal of racial preferences is consistent with the work we have published on the SFFA cases.
Both MIT and the media advertised the changes in the racial makeup of incoming freshmen as reducing diversity. The Chronicle of Higher Education described it as a 36 percent drop in racial diversity. Despite no change in the share of white enrollment, NBC News claimed that “The university’s white and Asian American student populations have increased, while all others have declined—some even down to zero, according to MIT.”
While MIT is to be lauded for actually releasing its numbers, the picture is more complicated than MIT and the media let on: it depends heavily on how one defines “diversity.” As MIT and the media are using it, the term seems to mean “representative of the national population.” Asian Americans are a diverse group, representing many different cultures and ethnicities. But MIT and the media treat them as a monolith. To them, the diversity they bring as individuals of particular cultures and ethnicities is less important than their representativeness of the U.S. Asian-American population as a whole.
The framing of the MIT numbers also neglects another component of diversity. Compared with the classes of 2024 through 2027, the number of first-generation college students rose from 18 percent to 20 percent, and the number of students eligible for Pell grants increased from 20 percent to 24 percent. MIT actually became more diverse based on socioeconomic measures, perhaps partly in response to the ruling.
We still don’t know whether other elite universities experienced similar demographic changes. The MIT numbers, however, put other top universities on notice. Given MIT’s strongly stated desire for a racially diverse class and its indication that the Supreme Court ruling has thwarted that desire, if we do not see similar shifts at other schools, then it is likely that they have either flouted the ruling or reweighted their admissions criteria to put less focus on academics.
While MIT’s numbers suggest that it has complied with the Supreme Court ruling, other factors are also at play. Student decisions about where to apply, university decisions about whom to recruit, financial aid, how the various admissions criteria get weighted, and the yield rates of admitted students all will affect the racial makeup of the class. For example, black and Hispanic students might be less likely to apply to an elite university if they believe that the school no longer practices race-conscious admissions.
Universities may also change their own behavior in response to the ban on race-conscious admissions. Universities could recruit more aggressively or offer more generous financial aid packages; indeed, MIT says that it has done both. Schools may also adjust their admissions policies to put less weight on test scores, which allow them to compare applicants across different high schools, and more weight on factors like grades.
Another factor is the yield rate: the likelihood that an admitted student chooses to enroll. Yield rates of different demographic groups tell us how sought-after those groups are in the broader college market. Historical data from Harvard indicate that black and Hispanic admits persistently enroll at much lower rates than Asian-American and white admits. This is because black and Hispanic admits, on average, have better options outside of Harvard compared with their Asian-American and white counterparts when schools can use race in their admissions processes. The driving factor is the relative scarcity of black and Hispanic students with exceptional grades and test scores. But if all schools are no longer using race, then yield rates for black and Hispanic students should increase.
Because MIT revealed no information about how its applicant pool or the yield rates of various demographic groups changed, it’s impossible to tell whether or to what extent MIT adjusted its admissions criteria after the SFFA ruling. One complicating factor is that schools like MIT have removed the race field from their applications in order to comply with the Supreme Court’s ruling. But surveying their admitted class about their other options would help us understand how the market for top students is changing along racial lines.
A further complication for monitoring universities’ compliance with the Supreme Court’s ruling is that no systematic or generally accepted way of classifying mixed-race students exists, and the U.S. population will only continue to become more mixed-race over time. MIT double-counts mixed-race students. But for MIT’s Class of 2028, the number of mixed-race students appears to have dropped substantially: the total for all race categories was 112 percent in 2027 but only 102 percent in 2028. This means that the share of MIT students who would be classified solely as black or Hispanic may not have dropped by as much as the numbers seem to indicate. The Integrated Postsecondary Educational Data System (IPEDS), a federally run survey, requires universities receiving federal funding to report on the characteristics, including race, of their student bodies. The effect of the Supreme Court ruling may not be fully revealed until 16 to 18 months from now, when IPEDS releases its updated numbers.
Photo by: Sergi Reboredo/VW Pics/Universal Images Group via Getty Images