Thread by @jmlarkin

At last set up for Day 6 of the #HarvardTrial, where @RickKahlenberg of the left-leaning @TCFdotorg is on the stand. Kahlenberg has argued for @Fair_Admissions that Harvard could achieve real diversity without considering race in admissions: …2i553a2t1s05i-wpengine.netdna-ssl.com/wp-content/upl…

Kahlenberg says he considers it extremely important for a college like Harvard to promote diversity—of all kinds. He adds: "the socioeconomic diversity at Harvard is deeply lacking... Raj Chetty found that there have been 23 times as many rich kids on campus as poor kids."

You can look at some of what Raj Chetty (formerly a Harvard economist) found out about Harvard's economic effects in this well-designed #infoviz from @UpshotNYT: nytimes.com/interactive/pr…

One of those findings is that Harvard's median parent income is $168,000 per year. That's the fifth highest of 79 Massachusetts colleges — Roxbury Community College is lowest, at $33,000 a year — but third-lowest of the eight Ivy League schools.

When asked why he supports using "socioeconomic status" in admissions, rather than race, Kahlenberg says the obstacles students overcome are "most strongly associated with socioeconomic status." He points to @BarackObama saying his daughters didn't deserve an admissions "tip."

Here's the quote, from then-candidate Obama, from this piece by @rachelswarns:

RK sides with the courts, too, saying: "The goal of racial diversity is compelling, but narrow means should be used... Stigma can be associated with beneficiaries of racial preferences... If you can get the benefits of racial diversity without using race, that's to be preferred."

(Ahem, the actual piece is here.) nytimes.com/2008/08/03/us/…

"Some of the world's greatest universities — Oxford, Cambridge, Berkeley, Caltech — none of them use legacy preferences," Kahlenberg says. That removes an "impediment" to racial diversity that doesn't have directly to do with race.

Kahlenberg denies that he opposes racial preferences in all circumstances. "In an instance where a race-neutral alternative does not work... I'd approve of the use of race as a last resort." And he says he disagrees with @Fair_Admissions in some respects.

Kahlenberg is not an economist or a statistical-modeling expert. But he is now being asked to explain the basic findings of competing models of the Harvard admissions process.

First, we look at the status quo. Kahlenberg says in @Harvard's actual admitted class of 2019, there was a "vibrant" level of racial diversity: 24 percent Asian-American admits, 14% African-American, and 14% Hispanic.

The average SAT score of that class was 2244 points out of 2400. But just 18 percent of that class was tagged "disadvantaged," which @RickKahlenberg calls a sign that economic diversity is a "mixed bag" at @Harvard.

Now we'll compare simulations to that reality. Simulation (A) goes first. Its ground rules:
• It allows no preference based on race, relations to alumni or faculty/staff, or on early action.
• It does include athlete preference and boosts applications based on SES/geography.

Kahlenberg explains that doing away with a preference for recruited athletes would be seen as "radical," given that it's such a fixture in American higher ed. But this first simulation does use a bigger boost for poor students than @Harvard currently gives.

I know this is a lot of buildup and abstraction. But you could reasonably call this the *core* of the case: could @Harvard arrive at an equally, or even more, diverse class without considering applicant race?

Now we're seeing Simulation A's results. Take away all those existing preferences and favor low-income students: the Asian-American enrollment would rise by 4 pct. points, while Af.-Am. enrollment would drop by same. The class would now be 54% "disadvantaged" (compared to 18%).

Kahlenberg is saying that @Harvard did not allow experts to consider family wealth data. A "'wealth variable' would better capture the history of slavery and segregation," and its effects on African-American families. A true race-neutral approach would use that data, he says.

Remember this, from @Harvard's opening statement a week ago:

https://twitter.com/jmlarkin/status/1051853463364218881

Kahlenberg is arguing that these simulations show that under a race-neutral admissions regime, Harvard could admit a class that is: (1) substantially racially diverse, (2) unprecedentedly economically diverse, and (3) almost as academically prepared as its actual class of 2019.

They're now reviewing a second simulation conducted by Harvard's expert, David Card. This found, Kahlenberg argues, much the same thing as he found for the plaintiffs: getting rid of racial/legacy/athletic preferences allows the school to meet all 3 goals above.

Another simulation: (C) for those counting at home. This one doesn't favor racial minorities, early action applicants, legacy/"dean's list" students, or children of faculty/staff.

This model would also endeavor to pull from the bottom third "census tracts" (in terms of economic strength). Under that model, the SAT scores would only drop one percentile (to 98th), and Hispanic enrollment would rise six percentage points. (Af-Am. enrollment still down.)

Kahlenberg was asked if it was "feasible" to pursue these models at Harvard. He says yes. "This would essentially replicate [the current] type of system, with the major exception that what counts in admissions would shift." In other words, this would be a tweak—not a sea change.

Kahlenberg says @Harvard's "disadvantaged" tag is "a little bit amorphous... it's basically up to the admissions officer." He wishes Harvard had shared more granular data about family income, and especially wealth.

Kahlenberg is now detailing how top Harvard officials responded to his data and models. In a report, they said his proposed alternatives would result "in a 19% drop in the proportion of admitted students with the highest academic ratings," which they called unacceptable.

Kahlenberg disagrees: "The students who would replace academic 1s and 2s [top scores]" under his model would presumably be "academic 3s... who overcame [great] obstacles." He says Harvard would be "enriched" by those new students.

Here's another of Harvard's arguments against weighting socioeconomic status highly in admissions decisions:

And Kahlenberg has been asked whether he's persuaded that @Harvard would incur a financial disadvantage by going to race-neutral, poverty-focused admissions. He says it's "unpersuasive" for the richest university in the world to make that argument.

As Kahlenberg is enumerating how the $9.62bn @Harvard raised in its latest capital campaign, Bill Lee objects. Judge Burroughs says, "Let's leave it at 'Harvard is rich.'" A laugh line. More on campaign: harvardmagazine.com/2018/09/harvar…

Now Bill Lee takes over to cross-examine Kahlenberg. He's pointing out that @Fair_Admissions is only calling two non-Harvard witnesses: himself and Peter Arcidiacono, both experts paid by the group.

Oops, sorry. Chetty is back at Harvard! He was at Stanford for a few years.

"You and Mr. [Edward] Blum have a mutual interest, correct?," Lee asks Kahlenberg, as he pulls up SFFA's mission statement. It reads, "A student’s race and ethnicity should not be factors that either harm or help that student to gain admission to a competitive university."

This is Lee's first bite at an SFFA witness—and he's going hard. Of Kahlenberg, he asks: "You've never worked as a college admissions officer, have you?... You've never served in the administration of a university, correct?"

"Never implemented a financial-aid program." "Never designed a recruitment program." The questions are still coming: "never implemented...," "never asked...," "not an expert on Harvard's institutional goals, correct?"

Kahlenberg pushes back on that last one. He's read up on Harvard adequately, and, he says, "I reference Harvard's goals when proposing my own."

"Mr. Kahlenberg, isn't it true that you were paid to help SFFA prepare its complaint in this very case?" He says not to draft it, but he admits he *was* paid to consult on the complaint, filed 11/17/2014 (link below): studentsforfairadmissions.org/wp-content/upl…

Lee is now pointing to a quote Kahlenberg gave to @FoxNews that appeared the day after the complaint was filed. Here's the article in question, and Kahlenberg's argument therein: foxnews.com/us/rejected-as…

Lee's purpose, of course, is to impeach Kahlenberg's findings, to show that he had formed and publicly expressed an opinion well in advance of preparing his expert report.

Lee's asking Kahlenberg to imagine a Chinese-American kid growing up in a completely white neighborhood in Lincoln, Nebraska. He doesn't disclose his race on an application, but speaks of overcoming discrimination.

"Can a college admissions officer consider that or not, when not considering race?," Lee asks. Kahlenberg says they could consider the applicant's stories of discrimination without necessarily considering his race on its own.

Lee is now quoting from this footnote from Kahlenberg's report, which alludes to certain public universities' inability to regain prior racial diversity after abandoning racial preferences in admissions:

Kahlenberg is saying that race-neutral admissions practices wouldn't necessarily have to result in lower racial diversity, and that @Harvard's decision to block his use of some economic data meant he couldn't fine-tune his model for maximally-diverse effect.

Lee is citing Kahlenberg's historic approval for Harvard dean of admissions Bill Fitzsimmons, and his efforts to bring in more low-income students. Lee pulled up this 2013 @HarvardPolitics article as an example: harvardpolitics.com/united-states/…

That was just a year before @Fair_Admissions filed its complaint alleging exclusion, with Kahlenberg consulting. It's a strange dichotomy of opinion, but Kahlenberg says he stands by his praise for Fitzsimmons and for certain aspects of Harvard's approach.

We're going to break in five or so. If you have *any* questions about this trial or any of the broader issues it raises, feel free to send me a DM or email me here: wbur.org/inside/staff/m…

And we're back. Lee is questioning @RickKahlenberg about Harvard's almost-best-in-class generosity in terms of financial aid to students from working- and middle-class households. Lee used this index as evidence: nytimes.com/interactive/20…

"Your suggested race-neutral alternative is to admit the same number of students from each of the 33 @CollegeBoard clusters," Lee says, and Kahlenberg agrees. The social maps of these "clusters" is laid out in this document: media.collegeboard.com/digitalService…

Lee goes on: isn't race a major organizing feature of these 'clusters'? (See three examples below.)

Kahlenberg says he sees his proposal as akin to Texas's taking top students from all of its high schools. That's seen as race-neutral, even though some such schools are predominantly one race or another, because it doesn't consider individual racial backgrounds.

Now Bill Lee is asking @RickKahlenberg whether he agrees with the 2012 Unz article—which criticized the work ethic of Jewish students and made other stereotypical assertions about students. Kahlenberg denies agreeing with any such claims.

Kahlenberg is one of the leading educational experts on the political left. So right now, you can feel the inherent tension of his siding, in this case, with the conservative Edward Blum.

The tone of attorney Lee's questions is still what I would call "genteel-hectoring." A parade of "Corrects?," to which Kahlenberg has mostly agreed.

Now Lee is asking Kahlenberg about projected dips in black enrollment under a variety of simulations, including some of his own. Kahlenberg is protesting that he doesn't support that as a goal and hasn't advocated for the policies being modeled.

"In every single one of your simulations, the racial group that bears the burden... is African-American students—correct?" Lee asks, then asks again over Kahlenberg's protestations. This is the page Lee is pointing to, from Kahlenberg's rebuttal report: ……2i553a2t1s05i-wpengine.netdna-ssl.com/wp-content/upl…

Lee is pseudo-grilling Kahlenberg over whether he spoke to any @Harvard students or faculty about the prospect of effectively shrinking African-American enrollment by 40 percent. (The answer is no; that wasn't part of his approach to the research question, Kahlenberg protests.)

Lee has pulled out a quote from Anthony Kennedy's decision in "Fisher v. Texas"—so far, Edward Blum's great defeat. Read that quote below:

Now Lee is trying to establish whether Kahlenberg agrees that @Harvard is pursuing "its educational goals" by its current admissions system — again, he's largely agreeing.

Lee just wrapped on cross. No more questions for @Fair_Admissions. @RickKahlenberg is excused, and dismounts the witness stand.

Now Lee is questioning Marlyn McGrath, @Harvard's director of admissions. With Dean Fitzsimmons, McGrath is among the top figures overseeing the admissions system under dispute.

McGrath is married to Harry Lewis, who was dean of Harvard College from 1995-2003. Here's the story of their marriage, as reported by @thecrimson with their son-in-law, Pulitzer Prize winner @Fahrenthold: thecrimson.com/article/2018/5…

From the stand, McGrath is running the courtroom through the admissions office's "casebook," which contains example applications to @Harvard. She makes clear that her office does not distribute "written guidelines" to evaluate integrity, charisma, or other intangible qualities.

They started out considering an example of a "clear admit": the apparently pseudonymous "Grace Chen," whose application showed strong personal attributes. Now we're considering "Peter Duran," a Guatemalan American student with better grades than Grace who was *not* accepted.

Under "appeal" section of the Duran entry, the casebook notes that his "mixed background"—half Hispanic, half-Caucasian—is underrepresented at Harvard. But under "pause factors," it says his academic record is "not unusually strong in Harvard's applicant pool." Not admitted!

Now SFFA attorney Adam Mortara is questioning director McGrath. He's calling attention to a sequence of numbers ("2 3+ 4 3+") next to Peter Duran's name. They reflect the ratings Duran was given.

He is again seeking to confirm whether an applicant's race is systematically factored into those numbers.

Sorry—tied up on radio duty for a moment there. I'd say that was well-worn territory at the end.

I'm back in the courtroom. Harvard College dean Rakesh Khurana is on the stand. Adam Mortara is asking Khurana whether he is now or ever has been "aware" that Harvard's admissions process "disadvantaged Asians." Dean Khurana has been equivocal, here and in an earlier deposition.

Khurana—who studies leadership at @HarvardHBS — starts out every meeting reading the mission statement of @Harvard College as a way of focusing attention on it. Here that is: college.harvard.edu/about/mission-…

Khurana, Harvard's outgoing dean of its faculty of arts and sciences Michael Smith, and admissions dean Bill Fitzsimmons sat on the so-called Smith Committee, which looked into "race-neutral alternatives" in admissions, first convened in June 2017.

The Smith Committee followed on the much larger Ryan Committee, which included experts and other voices. The Ryan Committee was suspended as the SFFA suit was first made public in 2014.

SFFA disputes that the Smith Committee—a hastily-convened meeting of top officials—met the standard of the "serious, good faith consideration of workable race-neutral alternatives" that the Supreme Court requested in Grutter v. Bollinger (2003).

The Smith Committee concluded with a report stating that Harvard should not abandon its consideration of race in admissions. SFFA calls it a "dilatory, highly scripted committee of three officials pre-committed to an outcome."

Mortara is now going back a century, saying that the origin story of "holistic admissions" included the attempted identification and exclusion of Jewish students in the 1920s. Khurana agrees that that's part of the story.

"The court is here to determine whether Harvard did its best to consider race-neutral alternatives," Mortara says. Why a three-man committee? Khurana says: "The Ryan Committee was a university committee... [concerned] with multiple schools and tens of thousands of students."

The smaller Smith Committee considered striking legacy preferences, and bumping up preference for poor students. Mortara is asking Khurana why they decided not to recommend those changes.

"The simulations show that Harvard could not *both* achieve its diversity interests *and* achieve other equally important educational objectives, such as academic excellence," reads the Smith Cmte. report.

More: "if Harvard afforded [SES] weight sufficient to produce a [status quo] proportion of Af.-Am., Hispanic, and Other students... the proportion of admitted students with the highest academic ratings (as assigned by admissions officers) would be expected to drop from 76 to 66%"

Mortara is grilling Khurana. Being rich doesn't have anything to do with being qualified, right? "Don't you think Harvard should have a socio-economic makeup" that better resembles that of America at large? Khurana is dismissing "hypotheticals."

Mortara: "Your socio-economic status isn't related to your ability to pursue the important mission of Harvard College, right?"
Khurana: "No."

Mortara: "What is special about wealthy people that Harvard has to have them overrepresented?"
Khurana: "...We're not trying to mirror the socio-economic, or income, [makeup] of the United States... We're looking for talent."

Mortara asks a last question — whether it's fair that America's wealthiest families have so much space in college seats. Doesn't that perpetuate inequality? Khurana replies: "That's not how the admissions process works."

And we're done.