Princeton’s President makes bogus arguments that diversity and academic excellence are compatible

The article below, by the President of Princeton, just appeared in the Atlantic.  (Christopher Eisgruber has been Princeton’s President for 11 years.)  The title clearly implies that college diversity (and the implication is “racial diversity”) is not at all in conflict with excellence.

This is a message that, of course, the Progressive Left wants to hear, but when I read the article, I found it deeply misleading. It turns out that excellence at Princeton has not been maintained by admitting more racial minorities, but by allowing certain barred classes, like Asians, Jews, and the impecunious, into the school.  As far as the evidence goes for racial groups, yes, there is a tradeoff between excellence and diversity, and we know that for several reasons that I’ll mention below. One that we know well that colleges are omitting indices of merit, like SAT and ACT scores, as ways to increase equity, for racial minorities (save Asians) don’t do as well as whites (including Jews, which are seen as “white adjacent”).

This does not mean that colleges shouldn’t strive for more racial diversity, but I think they shouldn’t do it by substantially lowering the merit bar of admissions. There are other ways, like casting a wider net among prospective students, or, for equally qualified students, give the edge to minorities. But to imply that there’s no tradeoff between academic excellence and ethnic diversity (not including, of course, Jews and Asians, known to be overachievers) is to purvey a lie. But it is of course a lie in the service of “progressivism”.

It’s hard to imagine how the Atlantic could accept an article whose arguments are explained by the conflation of causation with correlation, as well as with cherry-picked examples or recent trends in grade inflation and selectivity. But let’s look at the argument.

You can click on the headline below, or find it archived here.

First, many American colleges either implicitly or explicitly have eliminated standardized tests (or made them optional) as criteria for admission, and yet, as I’ve written several times (e.g., here and here), SAT scores correlate better than anything else, including high school grade-point averages) with academic success in college. The reason they have done away with the tests, or made them optional, is to increase racial diversity, concentrating on blacks and Hispanics, who do worse on these tests.  Increasingly, medical schools are also ditching the once-required MCAT admissions tests for the same reasons, and Graduate Record Examinations, or GREs for graduate schools, are being deep-sixed for the same reasons.

But of course this isn’t mentioned by Eisgruber, nor the fact that Princeton itself did away with required SAT and ACT tests; apparently they’re now optional in the school’s “holistic admissions” process. And, as I’ve posted before, making them optional, or omitting them, actually hurts diversity! But misguided colleges don’t seem to realize that.

But here are the four main arguments Eisgruber giv; I’ve characterized them and put them in bold. First, though, his thesis:

A noxious and surprisingly commonplace myth has taken hold in recent years, alleging that elite universities have pursued diversity at the expense of scholarly excellence. Much the reverse is true: Efforts to grow and embrace diversity at America’s great research universities have made them better than ever. If you want excellence, you need to find, attract, and support talent from every sector of society, not just from privileged groups and social classes.

He’s right about how to achieve excellence—finding talent where you can—but this is not the same thing as saying that there’s no tradeoff between excellence and (ethnic) diversity and that you must reduce merit-based admissions if you want to increase diveristy. What the above says is that the more widely you look around, the more likely you are to find talented people. But again, that’s not people will read this article. Now, on to Eisgruber’s arguments:

He points to a few examples of ethnic minorities at Princeton who have been successful.

Not surprisingly, the first example is a Chinese-American, Fei-Fei Li.  But Asians, like Jews, are overachievers for what I think are largely cultural reasons, and that’s why Ivy League schools used to have Jewish quotas and why Harvard, until recently, had “Asian-American” quotas.

He then names one black person, one poor person, and one white but economically deprived person (Mark Milley) who became successes after attending Princeton.  Again, this proves nothing, for Eisgruber is making a general statement, and picking out one example from each of three minority groups proves nothing.

Princeton is academically better than it was in the middle of the last century because it began admitting public-school students and women.

But again, this proves nothing other than widening the pool of applicants that might contain meritorious students will allow more of those students to enrich Princeton. Once you begin at least considering public-school students and women, you suddenly have a whole large group of people from which to pluck the talented. But again, this says nothing about Eisgruber’s implied thesis: that admitting more minority students in general will not reduce “excellence” (presumably construed, though not defined, as graduation rates, grades in college, and success after college).

. . . Princeton’s history is illustrative, not because it is special but because—in this respect, at least—it isn’t. At the beginning of the 20th century, Princeton had a reputation as “the finest country club in America”—a place where privileged young men loafed rather than studied. When asked early in his Princeton presidency about the number of students there, Woodrow Wilson reportedly quipped, “about 10 percent.”

Half a century later, when the university began admitting public-high-school graduates in significant numbers, it sought to reassure alumni that the newcomers would not displace more privileged but marginally qualified children. The Alumni Council published a booklet declaring that Princeton would admit any alumni child likely to graduate. As evidence, it boasted that the sons of Princetonians were overrepresented not only in the bottom quartile of the class but among those who flunked out.

The Alumni Council’s brochure spoke about Princeton’s sons because, of course, the university did not admit women to the undergraduate program until 1969, thereby turning away roughly half the world’s excellence. That was only one of many unfair and discriminatory distinctions that American universities embraced at the expense of excellence.

Eisgruber also maintains that concerted efforts to obtain black students didn’t occur until the 1960s, but of course he doesn’t tell us how they fare at Princeton relative to Asian, white, or Jewish students (I guess the latter are counted as “white”). He also notes that Princeton had Jewish quoteas untyil the 1950s:

People who accuse universities of “social engineering” today seem to forget the social engineering that they did in the past—social engineering that was designed to protect class privilege rather than disrupt it. At Princeton and other Ivy League universities, anti-Semitic quotas persisted into the 1950s. Asian and Asian American students, who now form such an impressive part of the student body at Princeton and its peers, were virtually absent.

So now that there’s more “diversity” of Jewish and Asian-American students at Princeton, and the classes are doing better, does that prove that diversity is compatible with excellence? I don’t think that’s what Eisgruber means in his title. As New York Magazine says, and Harvard admitted, accepting Asians only by merit would result in “too many Asians”:

Harvard itself found in a 2013 internal study that, if it admitted applicants solely on the basis of academic merit, its share of Asian American students would explode from 19 percent to 43 percent.

No, no, we mustn’t have that! This is why, of course, Harvard discriminated against Asians by lowering their “personality scores,” and this is what the Supreme Court found when it banned race-based admissions. And, of course, blacks and Hispanics with the same indices of merit as Asians or whites are admitted much more often via affirmative action.  Again, this shows the conflict between merit and ethnic diversity.

Opening up admissions to poorer students increased excellence. 

With help from charitable endowments funded by grateful alumni and friends, Princeton and other leading research universities have also dismantled financial barriers that in the past discouraged brilliant students from attending. Contrary to what readers might infer from the endless stream of articles about debt-ridden college grads who become baristas, America’s elite research universities now offer financial-aid packages that make them among the country’s most affordable colleges. At Princeton, the percentage of students on aid has risen from about 40 percent in 2000 to 67 percent in the most recent entering class, covering low-, middle-, and even some upper-middle-income students. The average scholarship exceeds the tuition price.

The elimination of barriers to entry coincided with two other changes: students’ increased willingness to travel for an outstanding education and improved informational tools that colleges could use to assess the quality of students (and vice versa). The result, as documented by the Stanford University economist Caroline Hoxby in 2009, is that student bodies at America’s best colleges and universities are significantly stronger academically in the 21st century than they were in the 1980s or ’90s. By 2007, she reports, America’s leading colleges were “up against the ceiling of selectivity” defined in terms of academic credentials, not acceptance rate: Further improvements to the quality of the student body would be so refined as to be invisible.

Again, all that’s happening here is the advent of “need-blind admissions,” which we practice at the University of Chicago. If you don’t prevent impecunious students from attending Princeton—I was one of those, by the way; I couldn’t apply to Princeton, my first-choice school, because my family couldn’t afford it—then of course you increase the chances of finding students with good grades and high SAT scores.

Finally, over time, the degree of “excellence” of Princeton students has increased. This correlates, says Eisgruber, with an increase in diversity. 

Princeton’s internal data show striking changes consistent with Hoxby’s more general findings.  Princeton’s undergraduate-admission office has long assigned academic ratings to all applicants based on their scholarly accomplishments in high school, with 1 being the strongest and 5 being the weakest. In the late 1980s, Academic 1s made up less than 10 percent of the university’s applicant pool and less than 20 percent of our matriculated class. Indeed, if you plucked a student at random from the Princeton University student body in 1990, the student was as likely to be an Academic 4 as an Academic 1 (but unlikely to be either: Academic 2s and 3s made up half the class).

 

In recent years, by contrast, Academic 1s have constituted roughly 30 percent of the applicant pool and about 50 percent of the matriculated class. Princeton’s academic excellence has increased substantially across every segment of its undergraduate population.

Here we have the classic example of confusing correlation with causation. And there’s a double causation: standards for admissions have increased overall, which has raised the “rank” of admitted students, and the grade-point averages of students in college (presumably one index of “academic excellence” of Princeton students) has ballooned due to grade inflation.  At the same time, Princeton increased its ethnic diversity.  This is no evidence that the latter caused the former.

So there we have it, a pastiche of misguided or erroneous arguments, made by a guy who is a President of an academic powerhouse, to “prove” that you needn’t sacrifice academic merit for diversity. It’s all wrong, and it’s embarrassing—embarrassing for both the Atlantic and the hapless Eisgruber.

So how do we test whether diversity really is compatible with excellence?  There are two ways, and I’ve already mentioned them both:

1. See if lowering the bar for merit of admissions (i.e., eliminating SAT scores) affects academic excellence and achievement. We already know it does because of the correlation of SAT scores and other standardized tests with academic achievement.  If Eisgruber were right, why are schools like Princeton getting rid of mandatory test scores, or making them optional? There’s only one reason, and it shows that Eisgruber’s thesis is wrong.
2. Follow students of black or Hispanic ethnicity through college and see if their achievement (or post-college success) is negatively correlated with their minority status. I believe this is also the case, though I don’t have the data at hand. (I believe this is true for medical schools as well.) But if it is the case, it shows that there is a tradeoff between merit and diversity.  That is surely the case, and it’s one of the Great Lies of Wokism.

All the evidence I know of goes against Eisgruber’s contention. Why didn’t the Atlantic editors point out these simple problems? Because, of course, Eisgruber’s flawed conclusion happens to comport with the dominant narrative of “progressive” liberalism.

Two final points. I’m saying nothing about genetics or inherent abilities here, for I think that differences in achievement between racial/ethnic groups is cultural. (The genetic data simply aren’t in.) All I’m saying is that, given differences in qualifications and achievement among groups, Eisgruber’s thesis is wrong.

Second, I’m not saying that colleges should give merit 100% priority over diversity. That is a judgment call about whether, as Jon Haidt puts it, you want “Social Justice University” or “Truth-Finding University.”  But Haidt also notes that you can’t have both, and in this abysmal piece of analysis, Eisgruber takes issue with that. I have always said that I prefer some form of affirmative action, and I stick by that, but I’m not pretending that substantial increases in equity can be achieved without lowering overall “excellence.” There are other ways, though they’re slower. One of them is giving children from different groups equal opportunity at the outset. American doesn’t seem to have the dosh or the will to do that, but that’s what it will ultimately take to comport merit with diversity.