Hacker News Reader

Comment by xhkkffbf | Hacker News Reader

A friend at Princeton told me they're having the opposite scandal. Apparently the student newspaper did some data analysis and found that legacy students had both higher SATs and better GPAs. That sort of breaks the narrative.

The feeling is that the trustees have capped the number of legacies at around 10-12% and so it's a bit tougher to get in if you're a child of an alumnus. Apparently they treat the pools fairly independently.

https://www.dailyprincetonian.com/article/2023/07/princeton-...

View on HN

> that legacy students had both higher SATs and better GPAs

Then they can get in without legacy admission.

Isn't admissions into the Ivies mostly a crapshoot after you reach a certain GPA/SAT threshold?

You have to show you are unique through essays and ECs, and that you will be an asset to the university community?

Or that you have chosen good parents?

Curious about this too. Sort of a popular logic I've never sincerely questioned that most people applying to these schools have the test scores/GPAs etc so you're sort of lost in a sea of 'excellence'. (Only put quotes because it would only be one narrow definition - I was excellent at Unreal Tournament, setting up LAN parties, and eating Jack tacos).

> I was excellent at Unreal Tournament, setting up LAN parties

Interestingly, with good enough scores and grades and a slight spin on this skill set (e.g., setting up large scale LAN tournaments, being world ranked in UT, etc.), you would distinguish yourself from the “sea of excellence” (you are correct that scores and grades are a narrow and incomplete definition by Ivy admissions standards).

> Isn't admissions into the Ivies mostly a crapshoot after you reach a certain GPA/SAT threshold?

In general, it is not a crapshoot.

There are some things that are very doable by any reasonable Ivy admit that will set them apart. Cal Newport has written quite a bit about this, and I agree with most (if not all) of what he says.

It may seem like a crap shoot, because some/many folks don’t realize what it is about a successful admit’s application that sets them apart. Fwiw, it’s not “better SAT and GPA” or “one more extracurricular”.

Why not offer them a fast track if they are actually qualified?

> Then they can get in without legacy admission.

Not necessarily. Apparently there are a bunch separate pools: one for the athletes, one for the professors' kids, one for the orchestra etc. If the kid is merely a legacy with good SATs, they're not eligible for the other pools.

Harvard has 23,000 students - I find it hard to believe the orchestra is more than 1% of that.

Aren’t we talking about undergrad admissions selection process? That’s a little under 2K per year.

To put a big fat disclaimer on this - N is <50% of the entering Freshman class, data is self-reported, the statistics (like self-reported race) do not line up with official Princeton statistics.

My guess is that if you're a legacy that had bad grades, you are certainly not going to admit that in your first month on campus to the student newspaper.

Harvard's admits that have legacy status also have better SAT/Grades I believe

> And a Harvard spokesperson told me that admitted legacies tend to have higher median test scores and grades than the rest of admitted students.

https://www.theatlantic.com/education/archive/2019/04/legacy...

Class of 2025 student profile:

> Legacy students also had a higher average SAT score than non-legacy students, at 1523 for legacy students and 1491 for non-legacy students.

https://features.thecrimson.com/2021/freshman-survey/academi...

Guess you didn't read the point underneath

"Just under a fifth of respondents — 19.5 percent — reported cheating in high school. The wealthiest students reported the highest rates of cheating, with 25.2 percent of students with families making more than $500,000 a year having engaged in academic dishonesty. "

The actual base rate of academic dishonesty is unknown. Maybe these students are especially honest about their dishonesty?

This could easily be skewed by self-reporting bias. Perhaps wealthier students feel they have less to potentially lose by answering the survey honestly.

How many of the legacy students get unlimited time on the SAT? 20% of Harvard students do through claiming a learning disability, though the odds of 20% of the top ~.05% of students (Harvard students) all happening to have learning disabilities are insanely low, it is a higher rate than state schools. They do it because they have admissions advisors etc. that coach them to getting a diagnosis from the doctor for ADHD to get unlimited time.

I had absolutely no feeling of time pressure on the SAT. If anything, I already had too much time and the boredom of waiting to start the next section was more negative than any other effect.

I think the same is true of the overwhelming majority of people scoring 1500+.

I don’t think unlimited time is much of a cheat code if you’re faking (or getting a “light” diagnosis) and trying to break into Harvard level of SAT scoring. More time isn’t going to transpose the middle digits on a 1450.

On the math section you can for example plug and chug each option rather than use elimination heuristics etc. that are mostly required to complete in the time limit. You can also go back and triple check all your work.

That's not a huge range - a handful of missed answers - and, I believe, lower than it used to be. I distinctly remember being, at 1490, in the bottom quartile of "Harvard-bound scores". Without any proof, I imagine that Harvard et al. dropping standardized test scores from their admissions requirements has skewed who actually submits them. Legacies will only do so if it doesn't hurt their otherwise venerable pedigree; non-legacies will say, "Well, every bit helps."

Everyone comparing SATs and GPAs has completely missed the point. Princeton would rather accept for example someone who became a nationally known political figure before graduating high school than someone with a 4.0 1600. The "what sets you apart" part of the application is much more important than your grades.

These ivies bank on 1% of their class doing something great and they optimize like 50% of their class to find the 1%, the rest of the 50% is just to preserve old power structures (children of rich people, politicians etc..).

> much more important than your grades.

.. once your grades and test scores are above the level that makes them reasonably confident you can handle the coursework.

Coursework at ivy's is a joke. 90% of my (gifted with affirmative action and thousands of students) highschool could've gotten a 3.0 at princeton

If you did engineering or math or sciences at Princeton, I doubt you would say that.

I was curious if that held once one accounted for means, and sure enough they have an income graph. Interesting article!

No surprise there. They probably grew up in wealthier households and had all the positive network effects

It was a big surprise to me. The standard story in all of the papers is that legacy admission is affirmative action for white people. In other words, they let in dolts with low SAT scores merely because they had parents who attended the school. The newspapers say this all the time. But the statistics say something different. The Princeton admissions office apparently has so many legacies that they can pick and choose the most amazing.

They actually did modelling of this during the trial. It was found that removing legacy and affirmative action causes negligible change in the white population.

Which the evidence suggests is at most a small effect compared to genetic intelligence, but of course we have to brush that under the carpet.

What’s genetic intelligence can you explain it to me?

A longer way of saying "intelligence," more or less.

I believe they're referring to the heritability of IQ. I think they're asserting that genetics have a much larger effect than environment on IQ scores. I haven't read any recent research on the subject, so I can't vouch for the accuracy of that assertion.

I meant intelligence (the underlying "g" factor) rather than IQ specifically, but other than that, yes.

Why would we have to brush something like that under the carpet?

Probably because they didn’t control for race

Reply on news.ycombinator.com