[hgdfhgfdhgdf]

The article doesn't mention the grades and test scores of legacy students are as strong as non-legacy admitted students.

"Among the admitted legacies, grades and test scores were indistinguishable from non-legacy students. Both groups had an average SAT score that surpassed 1430. Once on campus, legacy students tended to have slightly higher college grades, but their involvement in campus activities, merit awards, academic recognition and on-time graduation rates were indistinguishable from non-legacy students. In sum, legacy students, on average, were about as academically strong as non-legacy students, neither superior nor inferior."

https://hechingerreport.org/proof-points-why-elite-colleges-...

[polygamous_bat]

I can pretty confidently tell you that for every student admitted to Harvard with 1430 SAT there are at least 10 more students out there who were rejected with at least as much score. SAT score and grades are merely a necessary condition to get in, but they haven't been sufficient in a very long time.

As for grades, check out https://www.thecrimson.com/flyby/article/2013/12/5/the-harva... but in summary, the most common grade in Harvard is A, so don't think the data point says much.

[e1g]

The original study does address whether ALDC applicants are "as good" as typical applicants -

> Overall, our results show that only one-quarter of white ALDC admits would have been admitted if they had been treated as a typical applicant.

And on athletes -

> Being a recruited athlete essentially guarantees admission even for the least-qualified applicants. An athlete who has an 86% probability of admission—the average rate among athletes—would have only a 0.1% chance of admission absent the athlete tip

[unyttigfjelltol]

But the article doesn't deal with whether recruited athletes as a category are diverse. Probably, they are diverse and these statistics mean something remarkably different from the knee-jerk narrative in the article and many comments here.

[kemayo]

Article doesn't, but the study it's based on does. Recruited athletes are less-diverse than the general applicant pool:

> For example, recruited athletes, legacies, and dean’s interest list applicants are all over 68% white, yet the share of non-ALDC applicants who are white is less than 41%. All other racial groups see higher representation among non-ALDC applicants and admits than in any of the corresponding ALDC applicant and admit categories.

- https://www.nber.org/system/files/working_papers/w26316/w263... page 16

My anecdotal understanding is that "recruited athlete" often functions as a way for rich people to get their kids in. Get your kid on an expensive-and-niche sports team in high school -- lacrosse, water polo, etc -- and that puts them into a much-smaller pool of students that can be "recruited" by the college team.

[thrill]

"For example, recruited athletes, legacies, and dean’s interest list applicants are all over 68% white"

75% of the US population is white, so it sounds like that group is underrepresented in athletic admissions.

https://www.census.gov/quickfacts/fact/table/US/PST045222

[kemayo]

You're looking at a misleading data point on two levels.

First, note that the 75.5% figure in that table includes Hispanic/Latino people, which not all sources will -- importantly the study I was quoting doesn't include these under "white", so we can't compare those numbers. There's a different row in your table that excludes those to get 58.9% white. This alone gets us back to the ALDC group being disproportionately white.

Second, it doesn't matter if 75% of the US population is white, but rather what percentage of the college-admission aged US population is white. (Not that people outside of the 17-23ish bracket don't apply to Harvard's undergraduate program, but I suspect that they do so in insignificant numbers.)

Here's a source that breaks out US racial demographics by age: https://nces.ed.gov/programs/digest/d21/tables/dt21_101.20.a...

It says that in 2021 amongst 18-24 year olds, 52.8% were white. Thus the ALDC group is even more disproportionately white than the whole-population number would make you think.

[jgwil2]

The claim was made about the applicant pool, not the general population. The white population skews considerably older than other groups.

[pknomad]

I don't feel like that's too surprising. There is that saying "the apple doesn't fall far from the tree." It reminds me of this video: https://www.youtube.com/watch?v=Kqi_6v2RGB0

Essentially a high school student (now a college student @ MIT) solved a conjecture on distribution of Carmichael numbers. The parents of the high school student also happens to be first rate mathematicians.

I went to a southern ivy and many of the legacy students that I've met were children of doctors, lawyers, business executives, etc and grew up in a very fostering environment where they were exposed to how to behave in such an environment. I don't think it's too surprising that children would also excel like their parents.

[hgsgm]

There's not what's happening. What's happening is (made up numbers, relative proportion a true)

1000 ALDC students qualify. 200 admit.

1000 non-ALDC students qualify, with equivalent resumes. 50 admit.

Thus it's easier to get in at ALDC, all else equal, and les qualified ALDC displace "overqualified" non-ALDC, or if you believe their qualifications are truly equal and those 2000 students can't be compared to each other, ALDC get loaded dice for the random selection.

[pknomad]

What's ALDC? (a quick google search gets me Abby Lee Dance company which I presume is incorrect)

[deely3]

From other comment: [A]thlete, [L]egacy, [D]ean's interest list (i.e children of big donors), [C]hildren of faculty and staff

[atleastoptimal]

This is an important point. I do agree wealth and access to privileged resources are too concentrated, but it's silly to assume that the child of two Harvard graduates, due to education, genetics and preference, wouldn't be a better candidate for Harvard than a random person

[tchaffee]

Sounds like a good reason to eliminate legacy admissions if the claim is that these students can compete without the crutch.

[leetcrew]

I think the argument is the opposite... they start with one or more orders of magnitude of qualified applicants in excess of capacity. and to meet DEI goals, only X% of those accepted can be white anyway. how they actually pick from the qualified applicant pool is fairly arbitrary to begin with. so as long as whatever on-campus KPIs they track for the legacy-admit population are the same, why does it really matter?

I was not personally a legacy admit to my college (so no skin in the game really), but fwiw I don't think it's an entirely illegitimate thing to select for. most schools have their own distinct culture and consider that an important thing to preserve. I imagine families that try to send their children to a specific school over multiple generations probably feel a stronger connection to that school than the typical graduate, so putting a thumb on the scale to help them seems like an approach that could work.

if anything, I'm more upset about how recruiting for college sports works.

[tchaffee]

So these schools have goals to increase diversity but they want to preserve their culture? Those are competing goals. Eliminating legacy admissions would be a lasting increase in diversity.

[leetcrew]

I don't see those two as mutually exclusive, prima facie, unless you see "culture" as a euphemism for "perpetuating white supremacy". it certainly can be, but it can also be things like haverford's honor code, which only works when a critical mass of students find it meaningful. or it can be totally innocuous traditions like u chicago's annual scavenger hunt. elite schools can accept being each other's close competitors, but they can't accept being indistinguishable substitutes for their +/- 1 peers on the us news leaderboard.

[tchaffee]

Legacy admissions have a close to zero chance of increasing diversity. The entire point is that people who are just like former students get preference. By definition it's the opposite of a program to seek out those who have a different life experience than the status quo.

[runsWphotons]

No, it won't. Plenty of legacy admits are "diverse" now. And with increase of standards from banning affirmative action, legacy status will be the best chance of many from "diverse" backgrounds.

[tchaffee]

Why are you putting diverse in quotes?

Do you have data supporting your claim that "plenty of legacy admits are diverse"? By definition the legacy admissions are very similar to previous attendees. You need to explain how that would increase diversity.

[runsWphotons]

You brought up diversity, so what kind do you mean? Going forward, post affirmative action, just by test scores and gpa almost all students admitted at top schools will be asian or white. But there have been five decades where significant numbers from other races have attended these top schools and their kids and siblings can benefit from legacy admissions. (For example, this year ~15% of Harvard freshman were Black) Legacy admissions will absolutely increase this kind of diversity. If you get rid of it, it may further slightly benefit asians over whites, but I am not sure this increases "diversity".

[atleastoptimal]

"legacy admissions" don't exist as an official thing, it's not like any school explicitly states how many legacy students it will admit. It's a term to address the phenomenon of those related to alumni being more frequently admitted

[detaro]

Places like Harvard do have explicit policies to give such students advantages, so no, it's very much an official thing.

[Symbiote]

They have "college attended" for each parent on the first page of the PDF application form.

Removing that would be easy.

[tnecniv]

It totally is an official thing or was when I applied. Statistics on legacy vs non-legacy admission rates and how legacy status was factored (the nature of the advantage you got varied by school) in were readily available when I was applying to schools.

[snapplebobapple]

Whats the distribution look like though? I can generate you two datasets that have the same average but different distributions to the point that a small minority of one dataset would qualify for admission if they were combined.

[raincom]

"On average, LDC applicants (that is, excluding athletes) are stronger than typical applicants. However, the average LDC admit is weaker than the average typical admit, suggesting an admissions advantage for LDC applicants." (p.3 of http://public.econ.duke.edu/~psarcidi/legacyathlete.pdf )

[dirtyid]

>indistinguishable from non-legacy students.

If you break down non-legacy cohorts where Asian applicants have to score 50-100 points higher, then academic performance of both legacy and non-legacy (average dragged down by AA) is artificially low.

[CHSbeachbum420]

Of course the legacy student test well, they have had access to amazing resources their entire lives.

This isn't about test scores, it's about the opportunity for lower class advancement rather than perpetuating existing wealth