Hacker News Reader

Comment by ren_engineer | Hacker News Reader

I see this as another way to pull the ladder to success away from talented poor kids under the guise of somehow being more fair. This really only benefits rich kids who can pad their applications in other ways and makes admissions standards even more opaque.

Any kid can get a used copy of an SAT prep book for cheap and get similar results to a rich kid getting private test tutoring.

View on HN

Let’s say there’s two high schools, one is the ritzy private school (Eagleton Academy) and the public school that has a lot of poorer students (Pawnee High). Harvard needs a way to distinguish between students across schools. An A at Academy, after all, might be harder to achieve than an A at High, after all, Academy is a world-class institution. So, Harvard needs to tune its algorithm to factor that in.

So, the A+ prodigy at High might not stand out against a sea of B+ students at Academy, after all, Academy is much more rigorous.

Enter standardized testing: the prodigy gets a 2400 and the B+ students at High get ~2000s. Suddenly the prodigy can actually catch the attention of the Harvard admissions.

Yes, but the prodigy at Pawnee High happens to be Asian and we're full up on them so we take the B+ students at Academy whose parents went to Harvard.

Sorry, we've been advised by legal to say that we're not really "full up" because that would imply we have quotas and also that we only consider race peripherally and only ever as a plus factor. It just so happens to not be a plus factor here, while it's a default plus factor for some other candidates.

I'm glad I live in a country where only grades and standardized tests are taken in account, and colleges don't even know the race or background of anyone.

Where? Seems like DEI is turning up in most western universities.

Slovenia. But the system is pretty much the same in other former socialist countries (atleast the end results... some colleges have specialised entry exams in some countries or even a combination of both, but in the end, the students are sorted by whatever score they get (based on grades and exam scores, not race or whatever), top X get acceppted, and only the cutoff score is published - everyone with that score or more is accepted, everyone below is not).

This may be a leftover of socialism though, because in yugoslavia everyone was considered equal to everyone else.

equality > equity

If it was literally just grades & test scores it would be easy, but once you include personal essays & extracurriculars, inevitably inferences can be made about the applicant.

Your hypothecation doesn't match my experience.

The prodigy at High will come across as a prodigy: not only will they have the better GPA, they're also more likely to have the nice "extra factors" that play well in admissions--class valedictorian, leadership in student council or maybe other clubs, etc. By contrast, the mediocrity at Academy has a worse GPA, and is unlikely to get nice statuses in those other extracurricular activities (who will be snatched up by the prodigies at Academy). So for the mediocrity at Academy to be marked as better, the admissions officer has to know of Academy and its reputation... and that's less common than you might think, even for the most-highly-ranked exemplars of Academy.

The other factor to bring up is that mediocrities at Academy are often competing against the prodigies at Academy; many schools don't want to admit too many people from Academy and will have a cap on the number of admissions.

[Yes, I did go to such an Academy, so I have some personal experience on the matter.]

The prodigy might be valedictorian, but might not. Maybe this student is gifted at math and gets all perfect math SAT scores and subject tests and mails in the rest of their classes for easy A’s.

Yep, defnly matches my experience as a dirt-poor high-school dropout. SAT saved me. It's very sad to me to see these developments.

Same story here. I graduated with my GPA in the bottom 10th percentile of my class. My SAT scores were in the top 95th percentile across the nation though. Seeing this change just makes me reflect on how much it will harm the disadvantaged kids, while giving a free pass to the people who learned how to Google the answers to their homework.

It's been said many times, many ways, but the education system is totally broken.

Nothing in this change says that a kid like you wouldn't be able to get into Harvard under new policy. They won't REQUIRE SAT scores, they aren't saying they won't use them if you submit them.

> they aren't saying they won't use them if you submit them.

gp is wants others won't be rejected because of low sat scores

right, harvard reserves the right to admit someone who may potentially have really bad SAT scores. But that's fine, if admission officers see that person as Harvard material, why shouldn't they? Most candidates will still submit ACT/SAT scores, and average ACT/SAT score of the incoming class will likely still be high.

Look at what happened in San Francisco when they dumbed down the math for these same reasons. The rich kids just supplemented it with private tutoring. Which is what they will always do and you aren't ever going to be able to stop people with money from spending it on their children, at least not in a semi-free society. I was poor growing up and got into Pomona College because of my test scores.

If the rich kids are also getting SAT prep books AND tutoring on top, and everyone gets maxed out scores, seems like it's a waste of everyones's time and we're better off just skipping it.

I don’t think you should be downvoted for asking an honest question. As someone who tutored in SAT test prep a couple decades ago, I’m aware of the issues with standardized testing. However, the solution you propose is not so simple. The parent had a good point: high test scores may have given many students a boost in the eyes of admissions teams, especially if they did not attend a high school that is notable in the eyes of those admissions teams. They may not see top grades as equitable to those from a top prep school if they don’t know much about the school and don’t have high test scores to confirm their view of the student. Again, it’s not fair, but for these students, it’s a way to stand apart. If you take that away, it could be harder for them to be noticed until we have something else that’s sufficient.

SAT proposed an adversity score to address this, but it was extremely controversial and abandonded: https://www.nytimes.com/2019/08/27/us/sat-adversity-score-co...

Ah yes the. We are going to modify your score based on the color of your skin. I think that is a good way of doing it. I am sure no one would lie about that just to get a few extra points on the SAT, besides treating people different based on their race is totally a good idea an never results in unintended consequences. I like the idea basically we are making sure each side is equal but slightly separate from each other.

here are the criteria they used in the adversity score:

Crime rate, Poverty rate, Housing values, Vacancy rate, Family environment, Median income, Single parent, Adversity score, Education level, ESL, High school environment, Undermatching, Curricular rigor, Free lunch rate, AP opportunity

Each school got a score.

https://www.wsj.com/articles/sat-to-give-students-adversity-...

Oxford does something a bit like this, but AFAICT only for getting interviewed; you still have to get the same exam results as anyone else, and you aren't given any advantage in the interviews:

https://www.ox.ac.uk/admissions/undergraduate/applying-to-ox...

The former head of Green Templeton College, Oxford, did an analysis of the final results achieved by people from different backgrounds who had got into Oxford, and the conclusion was that those who came from fancier backgrounds did less well in their degrees than their poorer colleagues with similar entry scores. The hypothesis was that the private schools were very good at hothousing borderline students, but that this obviously stopped happening once they were admitted and got the same environment as everyone else.

This is one of the reasons that Oxford is beginning to seriously concern itself with background - it turns out that if you're already expecting top entry grades it's a good way of getting better students.

That, and its long-standing gross overrepresentation of children from posh schools is getting embarrassing.

Looks like I am too late to edit my comment but I will acknowledge I was wrong about that those indicators seem to be reasonable and I could see myself possibly being in support of something like that not in the SAT itself but as a factor in admissions.

The alternative is to make it completely subjective and based on the whims of the committee. I can't see that going wrong at all, like because there are no committee members who might be racist or sexist or have a bias against religious people or anything like that. No the college acceptance committee stands as paragons of virtues moral giants in this land of ethical pygmys they alone are incorruptible unswayable and perfectly and completely fair in all things.

So standardized merit based metrics serve only as a barrier to their ineffable judgement and pure deductions let it be them alone and their powerful "algorithm" that is too pure and holy for mortal eyes to sully with their gaze be the new standard by which we determine if a student can go to Harvard, for clearly this is the one true way forward and will never be manipulated by self serving or short sighted beauracrats to manipulate and politic in.

Are the rich kids going to stop being able to afford prep books and tutoring once they get in?

Had this, did this, got 99th percentile score -> ivy league.

It's happens. I think I'm pretty smart but obviously very privileged.

>seems like it's a waste of everyones's time and we're better off just skipping it.

Agree, but what are we replacing it with if we're "just skipping it"? Extracurriculars? Are those also not "waste of everyone's time"?

The real question is: how much does tutoring really helps as compared to simply prepping with free ressources?

Is it 5%? Does that mean you can't tie-break between two scores within 5% of one another? That's called a measurement error. No, this steel beam isn't 8 feet up to the micron.

I strongly suspect there are diminishing returns. You get the biggest benefit from knowing the material from school. After that, familiarity with the test (practice tests every weekend for a few months) will help the most (and tutoring is no substitute for actually practice the test). After that, yes, tutoring will help some, but again, diminishing returns.

It certainly targets "asians" students who are deemed too successful at school, despite their diverse economic background, as they are already discriminated against in Ivy league colleges right now. It's just obfuscate discrimination. Yes, when you are told "you're too successful as a community" to get in, it's not only discrimination, it's straight out racism.

It will change absolutely nothing for actual rich kids, no matter their race, who can buy in admission through donations, unlike what some people are claiming here. For these kids, SAT scores already don't matter.

Right, and won’t the fancy private school count for more comparatively now?

Relative performance compared to peers at the same school will likely be weighed much more heavily.

if you have a shitty class rank and you submit a stellar ACT score, it will still count in your favor, that has not changed. They just don't require you to submit it.

Interesting. I see it as another way to pull another ladder up so that the kid who didn't get to go to the private prep school to train to take a test isn't being discriminated against for his parents' financial position. There are literally entire private schools where I grew up whose average student performs better than the best public school student.

That's probably selection bias.

The SAT is actually pretty tough test to game. Anxiety drives parents to pay a lot for test prep, but there isn't much indication that it actually improves scores.

Do you have any verifiable information to back up the claim that test prep doesn't improve scores?

I have a handful of sources that disagree.

https://www.act.org/content/dam/act/unsecured/documents/R171...

https://collegefinance.com/college-admissions/does-sat-prep-...

https://www.nytimes.com/roomfordebate/2011/12/04/why-should-...

Your second link clicks thru to Slate which links to this: https://nepc.colorado.edu/sites/default/files/Briggs_Theeffe...

"Powers and Rock concluded that the combined effect of coaching on the SAT I is between 21 and 34 points."

clearly parent commenter was hoping no one would actually read the "handful of sources" with the "verified information" that the scores are improved. you literally could improve your score more by getting a good night's sleep. i really cannot empathize with the level of self-delusion necessary to be that intentionally misleading.

Your third source summarizes the common knowledge that has already been stated here: anyone can practice on their own with cheap or free materials to achieve the same benefit of those expensive classes.

"Anyone can practice" is pretty meaningless. The largest component of benefit of expensive classes is how they motivate students. Motivating students (unless they are already self-motivated) is very difficult and skill and resource intensive, that's why the service is expensive.

In a class what is it that finally motivates the unmotivated? Fear of failing the exam. You already got this scary motivator with standardized testing. If you're talking about self-study versus a taking a college class to learn something, then I'd agree with you. Also of course, preparing for a general test is easy review material you don't need a skilled teacher to coach you through.

I used to teach a free SAT prep for underprivileged high schools in Southern Cali. I found it amazing how much I could improve scores of kids who were engaged. Unfortunately, only about a third of the kids were really engaged. But going from 800 to 1100 I felt like I could do for almost any engaged student. And 1200 to 1400 was harder, but I think if you give me a year -- I could do that for a lot of students too.

If you're at 1550 I probably can't help you -- at least if you're like me -- you sometimes just read a question wrong.

I don't know about your SATs, but here in Eastern EU I've been taking advanced math as school leaving exam and I had some money that allowed me to visit tutor like thrice.

I had problems with geometry and decided to get some help with one of its branches. Tutor explained me stuff, gave exercises that I did and so on.

At the real exam there was one task from this geometry branch and I managed to do it. I don't know whether I'd be able to do it without going to those lessons, hard to say.

But at the end of the day it gave me 5 of 50 points at this exam, so I basically in very simplified version I received 10% on hard exam for less than 10% of minimal wage.

Of course it'd be naive to think that it scales like that, but you get the point.

The poor kid will have more motivation than "Tim nice but dim" who would need pushing by his parents.

> Any kid can get a used copy of an SAT prep book for cheap and get similar results to a rich kid getting private test tutoring.

I would say the environment, social/parental expectations, exemplars within family/friends, peers ect matter more than a book.

Talented poor kids acing SAT by reading a cheap book is stuff from the movies.

only way to effectively end that would be to have government seize all children upon birth so every child has an identical environment growing up.

The fact is a standardized test is far less biased than the metrics they will use to replace them. That's the reason they were created in the first place

My sibling and I are were like this and we were definitely not the only ones. You’re out of touch if you think this just stuff from the movies.

There are many such examples in the real world. That's not stuff from the movies. I don't think these things get reported in US but in India we see this in the news all the time. You can make a generic google search[1] and find tons of such people.

[1] https://www.google.com/search?q=poor+family+child+tops+exam+...

If it's being reported in international media I would guess it's still newsworthy enough to not be a regular occurrence. Like India is huge so you'll find a steady stream of events that are common enough to be frequent in absolute terms, but still unusual in proportional terms.

This doesn't counter my point through. From the first link[1]

> "I could not study because of financial problems, so I thought I must send my children to school and wipe out the darkness from their lives,"

> "I got tremendous support from my school teachers," he said.

This is exactly what i said. Environment of achievement is more important. "poor" doesn't mean just lack of financial resources. you cannot simply overcome lack of parental and social investment and care by reading a SAT book.

> I would say the environment, social expectations, exemplars within family/friends, peers ect matter more than a book.

1. https://www.bbc.com/news/world-asia-india-32791175

There aren't enough to be statistically significant, which is why they are being reported in the media.

I think every school, has its "smart kids" who define their identities by their academic achievement, and who seek to ace every test. It's their way of standing out and being special so they try extra hard to keep it that way.

Unfortunately they often tend to be introverts which colleges want less than overachievers in other areas. Except for the science/technology departments, who were probably overruled by other departments if decisions like abandoning test scores are made.

Yeah, no. I took the SAT back in the day and scored horribly. Bought a SAT prep book, worked through it during lunch and brought my score up 300 points in 6 months.

while rare, thats exactly how I got in a good school

Add me to the list of counterexamples to your claim. While my family was more lower-middle-class than poor, I was (evidently, according to the financial aid office) the poorest person in my graduating class. And my SAT score was likely the reason for that. (I increased my score by 200 points to a perfect score using practice books.) I was also admitted to Harvard but didn’t attend because of the insufficient (at the time) financial aid that was offered.

> I increased my score by 200 points to a perfect score using practice books.

what motivated you to do this at the time ?

Also, I never said its not possible to do it with practice books.

> I would say the environment, social/parental expectations, exemplars within family/friends, peers ect matter more than a book.

my point was the word "more" in that sentence.

What motive would you need beside "a higher score gets me into better schools"? Surely someone smart enough to get a perfect SAT score can figure that out pretty easily.

I guess what i meant was why gp had this motivation while their peers didn't, what is the difference.

Clearly nature. No reason to keep this as an elephant in the room.

Late replying, but I readily admit that supportive parents and a school subculture that valued academics were both factors. I didn’t mean to suggest that I single-handedly pulled myself up by my bootstraps. But the SAT was the main path I had to stand out in applications as a public-school-educated person from a rural place you’ve never heard of.

I did it

Reply on news.ycombinator.com