The predictions were dire. In the course of a bitterly contested trial six years ago, Harvard University said that if it were forced to stop considering race in admissions, the diversity of its undergraduate classes would be badly compromised.
Now, a year after the Supreme Court struck down the school’s admissions system, effectively ending affirmative action in college admissions everywhere, the numbers are in for the first class to be admitted, and the picture is more nuanced and complex than predicted.
The proportion of Black first-year students enrolled at Harvard this fall has declined to 14 percent from 18 percent last year, according to data released by the institution on Wednesday — a dip smaller than the school had predicted, but still significant.
Asian American representation in the class of 1,647 students remained the same as last year, at 37 percent. Hispanic enrollment has gone up, to 16 percent from 14 percent. Harvard did not report the share of white students in the class, consistent with past practice, and it is hard to make inferences because the percentage of students not disclosing race or ethnicity on their applications doubled to 8 percent this year from 4 percent last year.
The post-affirmative-action demographic breakdowns have been trickling out over the last three weeks, and overall Black students appear to have been most affected. The percentages of Black students declined sharply at some elite schools, although surprisingly, they held steady at others. The suit against Harvard had accused it of discriminating against Asian Americans to depress their numbers, while giving preferences to members of other minority groups.
Admissions experts suggested even before the new numbers came out that the most coveted schools, like Harvard, Yale and Princeton, would be best positioned to maintain their Black enrollment because the students who were admitted to them were very likely to accept. So in that view, they are unicorns, part of a highly selective ring of schools that scooped up the top students and remained relatively unaffected by the ban on race-conscious admissions.
Schools that were slightly less selective — like Amherst, Tufts and Brown — saw bigger changes to their demographics.
This admissions cycle is the first national test of race-blind college admissions since a 1978 Supreme Court ruling allowed universities to use race as one of many factors leading to a diverse student body, as long as they avoided racial quotas. That case concerned the University of California, but the decision embraced what became known as the Harvard Plan, the idea that a student body was stronger when selected for diverse talents, skills, ideas, geographic and family backgrounds and, yes, race and ethnicity. The inclusion of race and ethnicity has been under fire ever since.
Harvard did not offer any analysis of the numbers. But it did stress the steps it had taken to beef up recruitment, like sending admissions staff to more than 150 cities, joining a consortium of public and private universities recruiting in rural areas and increasing financial aid.
Hopi Hoekstra, dean of the faculty of arts and sciences, suggested in a message to the Harvard community on Wednesday that the effects of the Supreme Court decision were still playing out: “We anticipate its full impact on the class composition at Harvard College may not be felt for several more admissions cycles,” she wrote.
Experts in admissions say they are still studying the numbers to figure out what to make of them.
“It’s fascinating to see the enormous variation emerging,” said Richard Kahlenberg, an expert witness for the plaintiffs in the Harvard trial, Students for Fair Admissions, and a proponent of socioeconomic — rather than racial — selection.
Elyse Martin-Smith, co-chair for the Harvard Undergraduate Black Community Leaders, said the difference in this year’s first-year class is noticeable. “It’s definitely something that we can feel as we’ve been hosting events,” said Ms. Smith, a senior from Charlotte, Vt. “Even 50 or 60 students makes a big difference.”
“To actually see the decline in the post-affirmative action landscape that we’re in is something that is disappointing but expected,” she added.
Jaylen Cocklin, a Harvard senior, said he was saddened that the percentage of Black students at Harvard had dropped by 4 points, but stunned that it had not gone down by more. He had been bracing for Harvard’s numbers to be more like those of the Massachusetts Institute of Technology, where Black students fell to 5 percent of the class from 15 percent.
He said he found it “a little shady” that Harvard had not released the white share of the class, and he wondered whether white students were less likely to report their race. “I would be interested in diving deeper into how they calculated it to be honest, because it doesn’t seem to fit the reality,” he said.
At a recent student club fair, the number of new Black students appeared to him to be much smaller than in past years.
Kelli Higgins, a first-year student from Jackson, Mich., said she found the decline in Black students “concerning,” but added that she had already found a strong community of diverse students at Harvard.
“It’s hard to be upset about it,” said Ms. Higgins. “At the moment, we’re so excited about the fact that we were able to beat affirmative action being overruled, in a sense.”
Ms. Higgins, who is Black, said she had purposely avoided any mention of race in her admissions essay. “I wanted to show other aspects of myself.”
Some experts took a critical view of the discrepancy between what was predicted and the actual outcome of the court’s ruling, suggesting some schools might have not complied with the decision.
Schools like Princeton and Yale had argued in court briefs that “there was no way you could get racial diversity without taking race into account,” said Peter Arcidiacono, an economist at Duke who testified against Harvard at the trial and developed statistical models showing that Asian Americans were being penalized.
“So I think the burden is on them to show how they were able to do so because it sure looks like they were cheating,” he added.
The trajectory for Black students at Harvard roughly resembles that of their counterparts at its peer institutions, including at Yale, where the percentage of Black students remained about the same at 14 percent, and at Princeton, where it was stable at about 9 percent. At Duke, the share of Black students increased slightly. It stayed the same at Caltech, at 5 percent.
But at Brown University, the share of Black students dropped to 9 percent from 15 percent. Brown also had a fairly sharp decline in Hispanic first years, to 10 percent from 14 percent.
At Columbia University, the share of Asian students increased compared with the share admitted last year — to 39 percent from 30 percent — while the share of Black students dropped — to 12 percent from 20 percent, admissions data released on Wednesday showed.
The portion of Hispanic students also dipped slightly, to 19 percent from 22 percent, and the portion of white students declined to 49 percent from 51 percent.
The Asian American share of the first-year class grew at Caltech, Brown, the University of North Carolina at Chapel Hill and M.I.T.
But it dropped to 24 percent from 30 percent at Yale, to 23.8 percent from 26 percent at Princeton and to 29 percent from 35 percent at Duke.
“When you see that these schools have no change in their Black enrollment and their Asian enrollment goes down, that is not indicative of taking race out of the equation, in my view,” Dr. Arcidiacono said.
Officials at the schools have urged caution in comparing the numbers, saying they could be distorted by differences in the way schools — and students — reported them, such as whether transfer students were included, whether students identified with more than one race or whether they chose not to report their race. Like Harvard, Tufts, Duke and Brown saw roughly a doubling of the percentage of students who did not report their race or ethnicity.
“As for why certain groups went up or down, we’re continuing to examine the data,” Tufts’s admissions dean, Joseph (JT) Duck, said.
Michael A. Elliott, the president of Amherst, expressed concern for the way the sharp drop in Black students at his school would be interpreted.
“It’s a really sad moment for this institution to have that decline in numbers,” he said in an interview. But he added that it would take time to understand the impact of the court decision on higher education, especially at small colleges like his.
“I’m ultimately more interested in those people — the students and the families — than in the numbers,” he said.
The Supreme Court case was rooted in two coordinated lawsuits, against Harvard and the University of North Carolina at Chapel Hill. They were filed in 2014 by Students for Fair Admissions an organization founded by Edward Blum, a conservative legal activist who was also the architect of a similar lawsuit against the University of Texas at Austin.
One of the suits accused Harvard’s admissions system of trafficking in racial stereotyping, by scoring Asian American applicants lower on personality traits like likability, courage and kindness.
But Harvard took pride in the very system that the lawsuit was attacking, saying that diversity was critical to its mission of educating future leaders and that abandoning race-conscious admissions would reduce the excellence of a Harvard education.
In June of last year, the Supreme Court ruled 6 to 3, with its conservative members in the majority, that the admissions programs violated civil rights law and the 14th Amendment guarantee of equal protection.
While the Supreme Court banned consideration of race, it did not prohibit schools from giving preferences to other categories of students, like recruited athletes and the children of alumni, faculty and donors.
The court also allowed students to write about their race in their application essays, if their race was relevant to some life experience.
Jeremiah Quinlan, dean of undergraduate admissions at Yale, said admissions officers received training in how to comply with the new legal landscape. He also said his office had stepped up outreach efforts to high school students who showed promise despite disadvantages.
“I am especially excited that the class of 2028 includes the greatest representation of first-generation and low-income students on record,” Mr. Quinlan said in the Yale announcement.
Stephanie Saul and Sharon Otterman contributed reporting.