29 June 2023

Affirmative Action In Higher Education Struck Down. Now What?

The U.S. Supreme Court has held to be illegal discrimination on the basis of race in higher educational college admissions at Harvard and the University of North Carolina, in a decision that effectively ends all race based affirmative action in higher education admissions. The six conservatives on the court were in the majority, the three liberals were in the minority.

In a previous post, I took data from the case to quantify the effect of race based affirmative action at these universities. The money chart was this one:

The undergraduate admissions rate at Harvard University is 4%. The University of North Carolina's admissions rate is 16.8% (8.2% for out of state applicants, 43.1% for in state applicants).

I noted in a previous post that NYT Piece mentions that: 

Harvard’s class of 2021 is 14.6 percent African-American, 22.2 percent Asian-American, 11.6 percent Hispanic and 2.5 percent Native American or Pacific Islander[.]" . . . . In a pending lawsuit: "A Princeton study found that students who identify as Asian need to score 140 points higher on the SAT than whites to have the same chance of admission to private colleges . . . Harvard’s Asian-American enrollment at 18 percent in 2013, and notes very similar numbers ranging from 14 to 18 percent at other Ivy League colleges, like Brown, Columbia, Cornell, Princeton and Yale" suggest a quota. "In contrast, [the lawsuit] says, in the same year, Asian-Americans made up 34.8 percent of the student body at the University of California, Los Angeles, 32.4 percent at Berkeley and 42.5 percent at Caltech. It attributes the higher numbers in the state university system to the fact that California banned racial preferences by popular referendum in 1996, though California also has a large number of Asian-Americans.

This analysis neglects, however, the fact that California has more Asian-Americans (16.1%) than the nation as a whole (6.3%), so its public universities which reflect the local demographic makeup of the state should also have more Asian-Americans. 

The downside of affirmative action mostly comes from Asian American and upper middle class academically strong white students who are denied admission at the expense of black and Hispanic applicants, with the edge given to black applicants being greatest.

The impact is greatest at the most selective institutions and is greatly diminished as you move down the selectivity ladder. Almost no one denied admission at a top school due to affirmative action is prevented from attending another college or university that is somewhat less prestigious but still very solid.

There is a big impact at top law schools that use affirmative action, although it has been gradually falling without much notice over time, after big changes, for example, when UC Berkley's law school ended it under a state mandate.

The larger patterns are also true in medical school, but the magnitude of the affirmative action effects is much smaller, because medical school has requirements that weed out weaker students much more strongly.

In U.S. medical school admissions, controlling for undergraduate grades and MCAT scores, the big losers from affirmative action are Asian Americans, and the winners are black and Hispanic applicants, with black applicants benefitting more.

The impact of affirmative action on white medical school applicants is very modest. The raw numbers of black and Hispanic affirmative action beneficiaries relative to total enrollments aren't huge. The number of white and Asians who suffer from it are smaller, proportionately, since those groups in the student body have larger base numbers.

Nationally, the racial and ethnic breakdown of medical school graduates is as follows: 

The impact at the academic graduate school level is minimal since that process is far more individualized and far less numbers driven.

Essentially, this ruling is more about an appearance of individual fairness than anything else, which isn't an illegitimate issue.

I've been ambivalent about affirmative action, in part, because I think that the costs of it are overstated. I can see legitimate arguments on both sides, and a more moderate solution (which two previous SCOTUS decisions tried to point universities towards) might have been better.

But taking it off the political table will probably help Democrats in the long run by depriving them of an option that is unpopular with many voters (just as Republicans were helped by Roe v. Wade since voters knew that when it was in force they couldn't deliver on their extreme anti-choice agenda, and just as Democrats are helped in Colorado by TABOR since voters know that state legislators can't increase taxes without a vote of the people undermining tax and spend attacks on them).

It is likely that some colleges and universities will try to react to this decision by establishing non-race based priorities with racial impact to mitigate its effects, for example, by (1) favoring first generation college students, (2) favoring students from lower income families or socio-economic status, (3) giving greater weight to class rank than to grades or test scores (which is one thing that was done in Texas when it abolished higher educational affirmative action), (4) addressing the financial need of lower and middle income students more, and (5) giving greater weight to having worked for money in high school.

The difficulty for elite institutions that try to take this path is that the black and Hispanic students with the grades and test scores and curricular rigor closest to the rest of their student body tend at elite colleges and universities to be the most affluent members of their ethnic communities, to have college educated parents, and to have attended schools full of similarly advantaged peers, just like white and Asian American students who get into these elite institutions. 

So, if these elite institutions focus on affirmative action for poor, first generation college students, and students who went to schools with less academically outstanding fellow students, the students who benefit from these admissions programs will need even bigger preferences in admissions than students admitted with a racial preference now, potentially making "mismatch" concerns more serious. I suspect that when push comes to shove, elite colleges and universities won't be willing to do this and will only give lip service to genuine income and social class based affirmative action. 

Conservatives might welcome those approaches, as they would help working class white students who have no shot at making it into these colleges, just as it would help working class black and Hispanic students who are also (less obviously) not in the running at these colleges in the current status quo. But, I doubt that they will get their wish.

Affirmative action has salience in conservative circles because conservatives think that people on the losing end of affirmative action are working class whites, such as culturally Appalachian men like J.D. Vance. And, there are a few people in that situation, but mostly, their wrong. The people affected by affirmative action are upper middle class white and Asian American high school seniors with excellent test scores, excellent grades, demanding classes, and "boring" non-academic lives. Those kids, incidentally, very disproportionately, come from families that vote Democratic and not Republican.

The aftermath of this decision will leave Republicans with a "moral victory" but few genuine benefits for the high school senior children of their rank and file voters.

The clean conservative-liberal break of the decision also means that less favorable results for African-American and Hispanic high school students from families that also overwhelmingly vote Democratic, will be blamed on conservatives and Republicans, so it will do little internal damage to the Democratic party coalition's unity, despite the fact that in substance, affirmative action pits on Democratic party constituency against another.

Campuses at elite colleges and law schools will have a lot less black students and a discernibly smaller share of Hispanic students, and a very noticeably larger contingent of Asian American students. 

At colleges any less selective than flagship state universities, the differences will be practically invisible, and med school and academic graduate schools, the differences will also be slight and not very noticeable.

The source of these ethnic differences in grades and tests scores also deserves a little attention. It has little to do with anything inherent in race, and a lot to do with recent immigration history. Recent Hispanic immigration to the U.S. has been predominantly driven by laborers and less skilled working class workers. Recent Asian American immigration has been predominantly driven by high end medical and tech industry professionals which as imposed brain drain on their homelands (this is also true of many recent African immigrants who often end up in elite institutions in slots envisioned as benefitting their long time Americans with significant African descent). Ancestors of African Americans were brought to the U.S. as slaves, remained slaves for a couple of centuries, and entered free life after the Civil War from a position below the bottom of the existing social class system with zero wealth into a Jim Crow segregated world that persisted for another century after that - but climbing the social class ladder is a matter of many generations. The nearly invisible ancestral divides between whites of different ancestral origins aren't quite as sharp, but are still very much there. Elite colleges have lots of white students with upper class New England and Mid-Atlantic ancestors, and few from Appalachia or the deep South. Elite institutions have few white, U.S. born, Evangelicals, and lots of secular and high caste Protestant students.

Some notes on the Colorado impact:
The court’s majority opinion did include an exception though: Military academies funded by the federal government are exempt from the ban and can continue using race-conscious admissions, such as the Air Force Academy in Colorado Springs. . . .

CU Boulder as a “moderately-selective institution,” a public university that could be at least partially impacted by the decision in the UNC case — the university’s acceptance rate in 2021 was almost 80%. . . .

In Colorado, private schools would be affected by the Harvard case, such as Colorado College — which had a 14.3% acceptance rate in 2021 — and even the University of Denver — which had a 63.6% acceptance rate — preventing them from using race as a factor in the process. . . .

MSU Denver is the only institution of higher education in the state that admits any prospective student that applies who is 20 years old or older and has a GED or high school diploma. Of its student body of nearly 5,500, 54% of the students identify as people of color and nearly 60% as first-generation students.

In Fort Collins where Colorado State University’s main campus is located, the court’s decision will not affect the undergraduate admissions process, according to President Amy Parsons. CSU had a nearly 90% acceptance rate in 2021, and Parsons said in a statement that the university will continue its commitment of admitting all qualified students from all backgrounds. Of the 5,700 students admitted last fall, 28% are from diverse backgrounds and 25% are first-generation students.

Other notable admissions statistics are from the Colorado School of Mines (56.9%), the University of Colorado Law School (29.9%), the University of Colorado Medical School (1.29% - not a typo - 180 new students per year), and the University of Denver Law School (46% admissions rate).

Colorado College, the University of Colorado Medical School, the University of Colorado Law School, and the University of Denver Law School are really the only Colorado institutions whose admissions are likely to be significantly affected by this ruling. 

Colorado College might have significant impact due to its selectivity, but its student demographics don't suggest all that strong an affirmative action program. Colorado College has 2.9% black students, 6.1% Hispanic students, 7.8% Asian American students, 6% international students, 1.2% Native American students, and 0.5% Native Hawaiian and Pacific islanders students. 

This may be a product of the fact that Colorado College is not known for its generous financial aid packages which gives it one of the most affluent student bodies in the nation. 

Colorado College's low percentage of Asian Americans for such a selective college may reflect the college's perceived weakness in STEM due to its block program which squeezes a full semester long class worth of material into a month, which can be good in some subjects, but is too intense for many advanced math and science classes - calculus and freshman college chemistry already have 50% fail rates nationwide at the usual pace - since Asian Americans are more likely than average to major in STEM fields.

7 comments:

neo said...

it's about time

Dave Barnes said...

I think that universities can take advantage of the military academies loophole via ROTC.

Ellie K said...

Why do you say that working class black and Hispanic students are "less obviously not in the running [for admission to] these colleges" compared to working class white students? I'm not suggesting that what you said is unfair, biased, partisan, or wrong. I just didn't understand what you meant. Might you clarify please?

Next, GOOD CATCH on the first chart, comparing admission rates (with and and without affirmative action) at Harvard, UNC out-of-state, and UNC in-state. I clicked through to your May 2022 post where you confirmed that the study had reversed one of the lower UNC middle columns. I do bank risk management (capital, liquidity, and never running afoul of Fed SR 11-7 supervisory guidance!) so I get all worked up over things like that.

In the next paragraph, you said:
"The people affected by affirmative action are upper middle class white and Asian American high school seniors with excellent test scores, excellent grades, demanding classes, and "boring" non-academic lives."
BORING? That is unkind; also, there's no reason to believe it is true. Or is there? Why do you think bright, diligent Asian and white high school students would have boring non-academic lives, especially if they are from upper middle class homes with the sort of privilege and access conferred?

Here's an idea that would alleviate some of the population-disproportionate representation for Asian and white students at Harvard and other Ivy League schools: Get rid of the sizable (but undisclosed) advantage given to legacy applicants. Alternatively, aspiring physicians need not go to Harvard Med. Instead, apply to and attend University of North Carolina's medical school, or any other land grant university; there's at least one in every state of the Union. I think you suggested that too. After graduation, med school diplomas don't make much difference, as long as they were earned in the United States. As time passes, it becomes even less important for doctors.

P.S. Sorry for my over-lengthy comment.

andrew said...

@Dave

I don't think so. I think that its a function of being part of the federal government.

andrew said...

***"Why do you say that working class black and Hispanic students are "less obviously not in the running [for admission to] these colleges" compared to working class white students? I'm not suggesting that what you said is unfair, biased, partisan, or wrong. I just didn't understand what you meant. Might you clarify please?"***

The black and Hispanic students who get into ultra-selective Ivy League programs are coming from upper middle class families (some of whom from families who are relatively recent immigrants from upper middle class or wealth families in Africa or Latin America), albeit often not from families quite as affluent families as the white and Asian students who attend them. Many went to private residential prep schools like Andover and Exeter, even if some of them are from lower income/wealth but high socio-economic status families (e.g. children of teachers at Exeter like my sister-in-law, as opposed to children to wealthy families who sent their kids there on the usual track). Plenty are legacies.

Few black and Hispanic students in Ivy League schools are genuinely working class kids who went to ordinary small town and inner city schools. But, this reality doesn't show up in the top line statistics and most people don't realize that this is the case. Most people think that black and Hispanic students are predominantly coming from comparable places in terms of socio-economic status and income and "privilege" to first generation white Appalachian/Great Plains/rural Southern kids who went to rural small town public schools.

My half-Asian, half-white kids attended two of the twenty most selective schools in the country - Brown University, and Colby College. But despite the fact that both their parents have graduate degrees, and all four of their grandparents have doctoral degrees (two M.D.'s and two PhDs), and they knew their way around a country club, had their fill of opera and ballet performances as children, and had a father who has had a long career, mostly as a small and medium sized law firm lawyer, they were still "low income", lower socio-economic status scholarship kids in that context. They went to genuine inner city schools, they grew up in a world where few people had second homes or yachts, they never met an investment banker or management consultant or Big Law lawyer or publicly held company executive, etc. They didn't have someone they could spend the night with or a pied-a-terre they could use in Manhattan for a department field trip to Wall Street like the lion's share of their peers did.

andrew said...

***"BORING? That is unkind; also, there's no reason to believe it is true. Or is there? Why do you think bright, diligent Asian and white high school students would have boring non-academic lives, especially if they are from upper middle class homes with the sort of privilege and access conferred?"***

You need "excellent test scores, excellent grades, demanding classes" just to prevent your application from being immediately tossed into the trash bin at an Ivy League school or similarly selective institution. But in order to convert that facially sufficient application into an admission you need a "plus factor" at a school like that which distinguishes you from the many similarly qualified people. By "boring" I mean things like participating in "Model U.N." or "Mathletics" (without winning a top national award) or being the President of an existing extracurricular club aligned with your academics like the robotics club, that you didn't found or do something dramatic while serving in, or playing a classical music instrument in the school orchestra, or having leading roles in high school drama productions.

You need something more than showing up and resume padding, something that shows that you have traits that your strong academic performance doesn't already reflect. Plus factors that are non-boring are things like achieving all of that while being blind, climbing Mount Everest, publishing a paper or at least co-authoring one in a peer reviewed journal, getting a patent, founding a viable business, getting a role in a Hollywood or NYC professional dramatic production, having a million followers on Spotify, or just doing something that is unusual for someone who gets straight A's and takes the most advanced available math classes.

For my son, who got into Brown, some of those "plus factors" were playing two varsity contact sports (lacrosse - also captain of the team- and football, which isn't that rare but is uncommon for Asian American math nerds), having half a dozen songs on Spotify that he produced himself as a singer-songwriter-guitar player (even though he was classically trained in violin as a kid like most Asian American kids), serving on the board of directors of the Young Americans Bank, getting 501(c)(3) status for the food bank charity he was the President of, and providing extra help at home for his legally blind mom, while also attending inner city public schools and flipping burgers for money that other people (certainly super-advanced in math Asian American suburban kids with straight A's), usually didn't. Proclaiming an intent to major in philosophy based upon the two philosophy classes he was required to take in the International Baccalaureate program he was in, instead of math or computer science, didn't hurt either (even though he ended up majoring in CS).

All of it was counter to stereotypes for Asian American boys who were good at math, and went beyond being a "grinder" who was excellent at school and participated in extracurriculars especially those related to academics. Those are the kinds of things that distinguish Asian-American and white kids who get into Ivy League schools from those who don't and instead go to very solid colleges and universities a notch or two down the selectivity list.

My daughter had similar plus factors, although not quite so "wow". So did my son's four Ivy League/Stanford girlfriends during college (bilingual Latina Mormon valedictorian at his inner city public school; bilingual nationally recognized flamenco dancer; trilingual Algerian, Parisian born Muslim premed with a peer reviewed medical research paper co-authorship; lengthy U.S. Senate internship, founder of a standup comedy troop, and leading role in other political campaigns).

andrew said...

"Get rid of the sizable (but undisclosed) advantage given to legacy applicants."

Where do you think those huge endowments that pay for everyone else come from?

Also, the boost in these cases while real is often fairly modest, it is the kind of boost that SCOTUS envisioned for race based affirmative action, even though race based affirmative action turned out to always be a much bigger boost.

"Alternatively, aspiring physicians need not go to Harvard Med. Instead, apply to and attend University of North Carolina's medical school, or any other land grant university; there's at least one in every state of the Union."

Every medical school in the United States is more selective than attending Harvard as an undergraduate. The University of Colorado Medical School's ***1.29% admissions rate*** is a bit more selective than average but not atypical. The average med school admissions rate is under 4% and most med school applicants apply to the med school of the state where they grew up. We just need more med schools period in the U.S.