Reinstating our SAT/ACT requirement for future admissions cycles

100% agreed. I could also buy a test prep book on amazon plus get a few others from the school and public libraries for a grand total of $15 spent on test prep.

That's a lot cheaper than dedicating my working summer to smarmy volunteer/extra-curricular projects.

This thread is filled with people arguing false dichotomies: "test scores" vs. "fancy extra-curriculars" or "rich experiences" vs. "working class experiences."

I recently saw one child into college, and have another on the way. I've spoken to many admissions officers, counselors, and similar folks making admissions choices for public and private universities.

To a one, they all claim to be looking for two things: (1) students who will succeed at their university, and (2) a diverse range of students to avoid a monoculture.

This means they try to look at the "whole student." For some students, that means academics and test scores to show academic prowess. For others, it's working summers to pay for their college to show it's a personally meaningful step for them. For others, it's being a primary caregiver to siblings or parents while still meeting college requirements to show grit and determination. For others, it's demonstrating consistent progress in pursuing a passion (e.g., volunteering or music or art) with a trajectory that continues through college.

Admission officers say they want a student body comprised of all of the above, and that's how they get the diverse range of students to create a dynamic learning environment.

Some of the least interesting students are "box checkers" who have a bunch of disconnected extra-curriculars that don't show passion, direction, or a proclivity towards success. If a student defines "success" as "getting into college," then what will they do once they get there? Probably flounder.

Test scores can be a useful data point towards determining collegiate success, but it certainly won't be the only one used. Nor will extra-curriculars or any other single factor.

The single biggest piece of advice for any applying student: create a narrative that shows how your activities (whatever they are) demonstrate why you will be a successful student that achieves your major-related goals. Your grades, scores, work life, home life, interests, and personal experiences should all be connected to show you are thoughtful, motivated, driven, and have your shit together. Do that, and you'll have the best chance of success at most admissions offices.

Bingo. I came from an economically challenged background with a very chaotic family life. Grades were impossible to keep up let alone extra curricular activities. Tests were how I was able to demonstrate my academic potential.

I gamed the SAT by "studying". It took an SAT book checked out from the library and a few hours of practice. The SAT is not particularly hard for a bright student of any background that puts some effort into it. And these days there are more free resources available than ever before.

I get the feeling policies to remove standardized tests are set by those who have no idea what real intelligence and achievement is.

Seriously, is there no talk about how unfair the notion of extra-curricular activities is in the US? It is one of these things that gave me a major culture shock moving into the US. Things like giving an advantage to people that have so many hours of volunteer work feels a really blatant way to give an unfair advantage to folks that are already way privileged and over represented.

it's very likely that i would have never gone to college had there not been an option to test-in. instead i got offers from several top tier institutions based on test scores, essays, work experience and the equivalent of a ged alone.

as such, i'm immensely thankful that the route exists and believe that it should continue to exist as part of a variety of routes for prospective students to demonstrate their aptitude, grit and readiness for higher education.

just as exams and associated scores should not discourage anyone from applying, neither should grade averages or k-12 school experiences. the application process should provide those with all manner of experiences and strengths to make cases that suit their situations best.

I'm not against standardized tests. What I am against is a test ran by a private company as a barrier of entry into a public institution. Private schools who want to require SAT/ACT, fine, they are private institutions, I see no issue them requiring a test administered by a private company. University of North Carolina, eh, can't we have a test ran by a public institution as a barrier of entry into a public institution?

There are intelligent people though that don't do well on standardized tests (or at least don't get top tier results). I agree that standardized testing is a good metric to get a general idea of aptitude, but I think it hugely fails in appreciating people who are able to solve 'non-standardized' problems. And those are actually the people who should be placed in top institutions

I used to work for a place that did the summer fancy extra-curriculars.

Basically, think a very very fun summer camp, BUT we would package in X service features, one per week. This was a win win. Parent's would pay because kid could put the stories of painting the walls in X town on the application, or cleaning up Y pollution on a beach. And the rest of the times the kids could have fun and learn cool stuff (I'm being vague because I liked the program even though you had to have good money to go).

I can relate, and I appreciated the doors the PSAT and SAT opened for me.

OTOH, I test higher than my actual abilities in my opinion, and that was a relatively unfair advantage the test gave me.

what do you think about people who claim standardized test is systematic racism?

standardized test is the only way to measure somebody on their aptitude for doing well like the LSAT.

yet somehow asian americans being discriminated at ivy leagues even with good standardized test score is not seen as systematic racism.

Extra-curricular well-roundedness was basically added to separate the well off from the poor

I didn't see your point in the beginning, but as I read this thread I realized how much pressure we are putting on these kids. Even for kids whos parents can afford the extra curricular stuff, it just seems bonkers to me how much they have to do to "stand out". I don't know where I am going with this, but just wanted to say this is too much pressure.

May I ask on what basis you were against them? Was it just ignorance of the facts (like the canard that they just measure wealth), or did you change your opinion of something subjective?

I imagine that test preparation is far more accessible when you have well funded schools and parents who can cover any additional tutoring/preparation. For me the game changer was community college, it gave me a second chance at a price I could afford and with a clean slate I earned a 3.9 GPA and got into a top 10 engineering program.

They literally wrote again, it's a multiple factor process. Why is there this concept "test vs no test"? No one expects meritless acceptance to anything.

Using that as a metric for society is exactly why we have out of control poverty, drug abuse, and homelessness.

If you were able to get good grades in school it wasn't because of your efforts, as a child and teen your brain is largely programmed by the community and environment you grow in. There are no gifted children, only those with educated parents that started early. This is becoming an ever more common theme, especially of kids with teachers for parents whom make too little yet know the systems enough to know exactly how to get their kid past the trite systems created for class warfare.

The solution is to defund universities and stop this elitist nonsense. Just like COVID has shown we don't need to show up to the office, we don't need universities either. We are a better society without their control and political agendas. There is little of value that isn't better done by a YouTube video today.

>Of course the real tragedy is that we fixate so much on college, even prestigious ones, at all given how unnecessary they are for making people into productive and happy human beings. This change is significant but affects less than one percent of the population each year..

I'm ambivalent because I sympathize with both sides of the equation. One one hand, I have the hindsight now that I didn't at 18 to realize that the world didn't end because I couldn't get into Harvey Mudd. And in many ways it turned out 100x better; instead of staying in LA-ish areas, I explored a whole new area I never would have considered otherwise, still got a great education for a small fraction of what Mudd woulda cost me, and was exposed to a completely different flavor of computer science that helped me decide my career.

But on the other hand, college really does open a ton of doors that for many non-upper class people would never open otherwise. While I was still in tech and had many choices, I imagine people at MIT or Mudd would be fighting off recruitment at top companies with a stick, with many opportunities coming from certain companies who only recruit at such universities. It can accelerate your career on the order of 5-10 years if you stick it out. And you'd likely be unparalleled in resources and opportunities if your focus is on research. If your goal is as lofty as being the next household name or to pivot into creating your own business, there are oodles more oppurtunities there.

I wouldn't trade the education I got for an MIT one, but I can understand why 18YO me (and many others) do feel that way.

> Of course the real tragedy is that we fixate so much on college

This is what gets me. The whole concept of college, and especially selective, high profile colleges such as Harvard, MIT, etc., seems antiquated and almost entirely manufactured at this point. It's artificially crafted scarcity designed to create a market and brand. I feel the true, primary function of these institutions is to give the rich and powerful a means to identify each other and their respective pedigrees. Sure, they throw the unwashed a few bones and let them take home some degrees. But what do you actually get for your effort and money? I can't believe the competitiveness is justified by an actual difference in the quality of the experience or the person that results from it.

Also foreigners—a lot of these side activities and volunteer opportunities depend on having social capital and connections.

Same thing with essays. I’ve observed that working class Americans are reticent to talk about diversity and adversity in the way college admissions officers expect, and recall having a similar experience. My wife’s parents grew up so poor in rural America that “store bought meat” was a phrase they used when she was growing up. Meanwhile, my family left Bangladesh when I was 5 under political circumstances where one day my mom’s brother (a military officer) showed up to our house in uniform and my dad thought that he was coming to detain us (it was a social call). But she would have been embarrassed to write about how her father grew up poor or that she faced any adversity, and I would have been embarrassed to write about how I was a foreigner instead of a middle class kid from Virginia.

Kind of the same. I was too poor to participate in extracurriculars. I was lucky to have clothes on my back and food on the table and electricity to do my homework and run my hacked together computer. I still graduated valedictorian. No sports, no music, nothing. I was on the chess team as a senior because we moved closer to the school, and I was able to walk to school in only 15 minutes which was huge for me. I was already a senior at that point, and it felt silly to start joining extracurriculars at that point. I had more time than some of the rich kids who were in everything in addition to keeping their grades top-notch. Between my great SAT scores and being valedictorian, I was able to go to great university that I might not have been able to attend otherwise.

I live in slovenia, and the system is pretty much unchanged from the socialist times.... standardized testing + grades are used to get entry into colleges (and previously high schools). No resumes, no diversity/affirmative actions, clubs, volounteering etc... just grades+results.

I am not sure how it compares but India suffers through the same problem. It is essentially a rat race. A lot of those problems could be trick questions, but to be able to identify it, especially on a regular basis takes longer(when the question has less than a minute dedicated to it).

I have a hard time seeing US standardized testing as "better." All of my Korean peers who have studied for the CSAT or Suneung laugh at how easy the SAT/ACT math section is in comparison. K-12 education in the US is years behind at this point from many Asian countries. The fact that we are debating whether standardized testing for colleges and prestigious high schools (Lowell in SF, Stuyvesant in NYC, etc) should be banned is just laughable and only sets us further back.

As someone who uses to be very good at this kind of "math" in high school...it's not even real math. It's a bunch of teenagers being tested on how good they are at being poor man's computers. And I don't just mean standardized tests - almost the entirety of the high school math "education" is sad.

Yeah, a lot of tests like that are actually testing for memorization, and not problem solving ability.

i mean... integrals are weird. there's a whole slew of ways of doing them and usually there's a handful of techniques that are taught for a handful of integrals of specific forms and then those show up on tests with the forms sometimes slightly hidden.

how do you know it wasn't just forms and techniques you were unfamiliar with? any math you're unfamiliar with can appear as complexity for the sake of complexity.

I really don't understand how many tests still allow trick questions like that when we have so much data that indicates that it's an awful metric. To me it indicates incompetence of the test creators.

sounds painful, the only thing I could think of would be to test the precision of transcribing step by step. Sounds like the surface area for loss of precision is wider when you just add the noise of messy figures into a problem. A clever student could just replace the messy figures with constant variables that are shorter to write of course and at the very end substitute everything back to evaluate what comes out in the end.

In the software development world, we simply moved it from university admissions to employment interviews.

100%! As a parent of a child about to go through the college admissions process (with his heart set on MIT--of course), I want him to read this particularly for the later part of the article: "...you are also not your MIT application..." The acceptance rate is so low, that it should not be used as a measure of self-worth and accomplishment.

Yeah, when you're doing something that is politically/ideologically unpopular or in the minority you tend to need to have bulletproof justification if you want to get away with it.

Then again, with these sorts of things you can never be sure how much the tail wags the dog.

It’s also important to note that they try to make clear that they’re describing their own situation and not providing a blanket statement.

thanks!

> huge slap in the face to social mobility out of the middle class

No, it was an intentional move to restrict the number of Asians admitted because despite their best efforts to hide it, prestigious universities' data showed systemic discrimination against Asians in the admissions process (see: SFFA vs Harvard). Removing the SAT removes some of the data that opponents can use to sue them. One of the reasons the UC removed the SAT was to "diversify" their admit pool because Asians disproportionately did really well on the SAT and they aren't allowed to discriminate based on race by state law.

My hope is that the Supreme Court, now that it's a 6/3 conservative majority, finally strikes down race based "affirmative action".

Similar opinion and story. We live in Boise, Idaho and my daughter does a hybrid model of home school and high school. She knew she got top grades and 5's on all her APs but it didn't give her any idea of where she sat, nationally. She was planning on going to art school until a year ago, when she sat for her first standardized test, the PSAT. She saw the score, which was national merit range and realized that she should have higher ambitions. She took the SAT last October (she's a Junior) and got a great score. So now she's considering top schools. She wasn't before.

Without these types of tests, you could end up being biased toward a measure of "how hard did the parents push."

This wasn’t my path, but many of my peers where I went to college had a similar type of story. They came from very poor areas like rural Texas or rural Georgia, and they got into engineering after getting into things like Lego sets and video games. They got great at math because it intrigued them even in their subpar schools. Aced the math SAT and then doors opened for them. I’d hate to see that door close.

> Removing admission tests was a huge slap in the face to social mobility out of the middle class.

I agree, but arguably the benefits of the SAT with regard to class mobility is less than it was when it was introduced.

The heritability of IQ, combined with assortative mating, means that increasingly class divides are forming around differences in IQ. I think many people are rightly concerned that this will eventually result in simply a different entrenched class hierarchy.

I still think MIT should use the SAT, but the problem of maintaining class mobility is one we should continue to assess.

>Removing admission tests was a huge slap in the face to social mobility out of the middle class. To reiterate other commenters, admission tests are the hardest part of the process to game and the least biased towards things like having a tiger parent, while being the most predictive indicator of success in college. Wealthy and well connected people can easily game extracurriculars and essays, and HS grades are vastly inflated at this point across the country. That pretty much only leaves standardized tests for your average kid who isn’t being deliberately primed by their environment to stand out for selective college admissions.

I agree with most of this but Exams are absolutely gameable. There's a reason there is a multibillion dollar industry focused on them. There's a reason why Cram School exists across Asia and there's a reason why parents will pay thousands for ACT/SAT tutoring. Because it flat out works. You can increase your scores substantially by paying for these services. All else being equal I would expect that between 2 students with identical aptitudes where one has paid thousands of dollars for tutoring that the student with tutoring will score higher than the other. This can't be ignored. Can you mitigate this with self motivated preparation? Sure. But again, all else being equal and between 2 equally talented and motivated testers the one whose parents paid for a tutor that achieved a perfect score is almost always going to score higher.

I am also happy to see this, but requiring test scores and class building are not mutually exclusive. Test scores are a useful input to good class building. How else are you going to choose which student from a backwater school to put in the class if these students don't have access to the advanced classwork and research support that the rest of the class had?

They might be the least gameable, but this is kind of a “bare minimum” for diversity/equity practices. It is the cheapest way to get little bit of reliable signal.

Because let's be clear about the standardized test situation. Test takers had time to take the test, they weren't doing work on the side for their parents instead to make ends meet. They had transportation to the test location, they were able to pay the fees, and they were not being discouraged from college as “we can never afford that, I’m sorry.”

Standardized tests themselves have mostly tested not IQ or domain knowledge, but how anxious you are taking tests. This does predict later academic performance, sure, but to say that this anxiety is disconnected from demographics sounds like a rather strong claim. Instead you have that it is more disconnected than GPA and essays and extracurriculars which is not saying much.

If you want to be serious about diversity and equity, you invest some actual cash into it... The easiest way is talent scouting, you send people (trained, you can reduce bias) to underrepresented communities and allow folks there to interview with them. My wife worked at place that did this, it sounds financially intractable at first but it scales to whatever budget you want to put into it... Her place would send folks out to like Singapore as well as to inner city schools. But the point is that you have to leave the door open to the people who can come to you, but you also go to the people who cannot.

(This is also a startup idea... The reason the schools don't send out their best is that there are too many colleges, the reason the colleges don't canvas the schools that there are too many schools, these are obviously inverse problems that could cancel each other out in the appropriate sort of network, the recruiters just need to be common to both. The problem is getting people to pay for it—the schools who you want most are precisely the underfunded ones that cannot pay you, the colleges meanwhile are less willing to go in on these sorts of weird experiments, they, they have an admissions department already. So the idea has a dangerous scope creep where you want to also start a college so that you can dogfood... Not a degree-granting educational institution in itself, maybe, but just a “first year at college” school which sells its students to other top colleges. Obviously that's a much riskier investment for VCs.)

And young people on here imagine having 2 children 2 years apart. One of them doesn't get to go to college because of covid and the other doesn't get to go because ability didn't matter for a year.

EDIT:

I hope this never happens to your kids. Beware of being penalized for trying. Is that a lesson you will teach your kids?

I'm not "bummed". Read my post with anger.

U of MN was not his first choice. It was _fallback_.

Try on some shoes.... Imagine putting 30 years of high taxes into a state as a high earner and business owner (that's me). Then imagine that your almamater doesn't accept your kid when your kid is, in fact, a lot smarter than you and has accomplished a lot more than you did by his age.

Then imagine being that kid and being taught by the State that, while you are in the top 95th (99th) percentile nationally on test scores there's just something not right about you. Is it your GPA? It can't be. You're on the A honor roll every time.

In the end you'll never know and that is fine because the school you might have attended has changed into something it was never intended to be.

Land grant university or something else? The Great Emancipator rolls in his grave.

On the plus side the U of MN gets billions of dollars in state funding and keeps the washout money too. Win-win.

It's important in relating these horror stories to include whether or not he was applying to Computer Science and/or Engineering. The growth in demand for CS has outpaced any schools ability to handle the students. My daughter's college counselor had different sets of target schools depending on the major (in her case she was thinking about biomedical engineering, one of the most sought after).

I'm sorry this happened to your son. It can be really disheartening for kids who work so hard.

Similar thing happened to my kid at University of California this year where test scores are also not even considered.

Wasn't as bad - they got into a 2nd tier UC so we are at least thankful for that.

> If you’re worried about having the account in one state and attending school in another, don’t be. With most plans, your school choice is not affected by the state of your savings plan. You can be a resident of Minnesota and send your student to college in North Carolina.

https://www.mnsaves.org/plan/details.shtml

(This is generally the case for most 529 plans.)

I’m not too familiar with UMN admissions stats are, but a 3.8 weighted GPA might have hurt your son considering a lot of people applying for computer science will have an unweighted GPA that is around that or higher. You haven’t shared the unweighted GPA here but if it’s substantially lower colleges will not like that, especially state ones, since they think it shows you just take classes beyond your level.

definitely not over. kicking ass in community college for a year or two and transferring to a much better school with scholarship is a great option for anyone

Wait, your state 529 plans penalize you for applying the funds to out of state schools? That sounds like a highly unusual setup. Not all states do this.

It is crazy how competitive even state schools are now a days. Back in my day, it would be pretty easy to go to UVA or University of Minnesota...

35 ACT and he was rejected to University of Minnesota? Tell your kid that he’s perfectly fine and that their admissions are FUBAR.

Try UIUC (my alma). Though I was in-state. It’s gotten more selective but a top one for CS. They admitted me into CE but it was easy to transfer after doing well in my first few CS classes.

I'm curious. How many universities did he apply to ? How many accepts/ rejects from that pool?

> Why? Because U of MN doesn't consider weighted grades nor do they accept test scores anymore.

This is the worst. Do they do this for equity reasons or something else?

Asian?

> If the SAT is not predictive, as some claim, we've been selecting our elites and professionals the wrong way for three generations.

Yes, and?

The easiest way to understand the shortcomings of these tests (at least their older versions) is to realize that students who use paid SAT/ACT-prep materials and services get higher scores than students who don't. Yes, there are confounding factors, but this fact alone is fairly damning evidence that these tests can be "defeated" using techniques beyond simply learning the things taught in high school. A better-designed test would not yield higher scores to test-takers with more specific knowledge of how the test itself is constructed. (In software terms, think "property based testing" as opposed to "unit test cases written in a predictable manner", with the assumption that your implementation under test has adversarial motives to obtain passing builds.)

I would say standardized testing has been critical to higher education and the professions since it was established 1416 years ago and is a major reason that China and Japan (which began standardized testing more recently, some 1100 years ago) never went through the catastrophic colonial subjugation that did so much damage to India and the Middle East and obliterated most of the cultural heritage of Africa and America.

Moreover, the British adoption of the system some 200 years ago was crucial to their imperialist success, and Western academia began adopting it slightly earlier. Without this Sinicization of Europe the Enlightenment might have fizzled out; perhaps we would never have had an Industrial Revolution.

This is kinda how medical residency is filled in the US, med students rank their top choices, residency programs rank their’s. It’s called The Match, and generally results in a very fair equitable outcome.

You wants kids to grind for an exam instead of becoming passionate about something?

I would argue exactly otherwise... with the advent of the web platform and frameworks like .net, the vast majority (and I mean like 95%) of developers will never touch anything ML related in their careers. I mean, I get that this is MIT and many of their students will end up working with ML, but applying that globally to CS is nonsense. Back when I was studying CS (cze) more than a decade ago, we had to pass linear algebra, graph theory and calculus, but honestly, that was like in the first year and a half and then it completely tapered off (later years were all about projects, algorithmization, i.e. "doing the work" and very little about hardcore theory) and guess what, I never needed it again. A bit of statistics and some graph theory here and there, but that's about it.

Contrary to popular belief, there are NOT that many ML jobs out there and the ones that are there are more about data science and messing with model zoo type of shit than actually coding useful programs. Most programmers will be lucky if they get to integrate inference of a prepared model into the apps they work on.

Calculus is not just about gradients. Fundamentally, calculus is about calculation and symbolic logic, a much older concept that predates its usage in gradient-based machine learning. In this sense, calculus holds deep connections to logical reasoning, proof theory and the foundations of computer science. [1]

[1]: https://compcalc.github.io/

True, but the usefulness of calculus to a working developer in ML is pretty marginal. The difference between different ML algorithms for hill climbing or gradients, is orders of magnitude less than the effect of having the right training data, formatted the right way. Statistics or data science is far more applicable to nearly any field on real-world programming.

But, you know, like Latin in the 19th century was always still useful to one's education, calculus is still useful. It is also something a lot more people know how to teach, than know how to teach statistics (or other more useful topics). I think the latter is the primary reason it remains central to most engineering programs.

and this is why trade schools like coding bootcamps are relevant at all

people aren't going to school to learn computer science, they are going to school to get a job and be effective in that field, but the universities shouldn't feel obligated to adjust to that since they've been for the privileged folks who are actually there to pursue education for the sake of higher learning for nearly 200 years (or much longer). it is mere coincidence that they have to put up with a few decades of people needing the school for subsequent employment and the school will exist after this phase as well

so with that observation it really is useful to push for trade schools again, for the people that actually need it

for the people that are really going for that upper echelon of access to other privileged people whether they get a wage-slave job or not, yeah they should slog through MIT, but everyone else should consider other things that more closely match the lane they were born into

Honestly every educated adult should understand calculus, and certainly anyone with a technical degree should. Sure many people will be unlikely to need to do the raw computation of calculating derivatives and integrals (even people doing machine learning are typically letting computers do that work), but to understand the way these two concepts work together and describe the world is really essential to understanding so many problems.

If we need calculus for CS, I wish we'd teach it under the CS heading so that exposure to math didn't have to be so biased towards real analysis. Students spend years achieving this arbitrary (unless you're going to be an engineer) goal and end up with the erroneous (and often harmful) intuition that all spaces are continuous metric spaces.

90% of software engineers don't do algorithms, even in ML -- nor are they capable of, besides rote memorization of interview algorithms. Software engineering these days is mostly a job of complexity management and automation, requiring little math and more secretarial skills. That's just the ugly truth that nobody wants to hear.

I would argue that calculus is essential in discrete maths, but ML is not essential in Computer Science.

I’m not sure that elementary calculus is particularly related to machine learning (I have a similar opinion of linear algebra which is fundamentally geometric in a way that machine learning isn’t). That doesn’t mean that you don’t need to know anything about calculus or matrix multiplication to understand machine learning but knowing eg Green’s theorem won’t help and neither will understanding a tensor as a multi linear map or in the way a general relativity physicist might.

I also don’t particularly know what goes into calculus (in the U.K. we studied something called ‘calculus’ in high school which included integrating/differentiating polynomials, some trig functions, easy integration by parts and, in the ‘further maths’ course, some second order linear ODEs with forcing, first order ODEs via integrating factors, first order linear systems of ODEs via the eigenvectors method, and I think some integration by parts based recurrence relations. At university things were divided into ‘calculus’, which contained practical tools for applied maths like Green’s theorem or partial derivatives or contour integration or Sturm–Liouville theory, and ‘analysis’ which had foundational things like epsilon–delta stuff or Dedekind cuts or the definition of a limit or Riemann integration or the conformal mapping theorem and so on.

I think a first course in the thing I called analysis above is very useful for building mathematical maturity (ie the ability to not deduce false things but also playing with definitions and thinking about counter-examples) but the calculus knowledge can be useful for understanding the physical world. But I don’t know if that understanding should be required for e.g. computer scientists.

A few calculus examples I can think of in computer science:

- Some famous story of Feynman ‘interning’ at Thinking Machines and solving some capacity management problem using bizarre differential equations with terms representing e.g. ‘bits per second’. No sufficiently good solutions had been found using discrete methods.

- I was once asked an interview question which I suggested solving with differential equations but I was quickly directed towards not doing that.

- Honestly I can’t think of many more but maybe this is a lack of imagination. I think there are a few things that are really probability theory that you need some understanding of calculus for, e.g. emergent behaviour of distributed systems, reasons to prefer random cache eviction, some intuition to answer a question like ‘if the time X takes has some distribution, but sometimes we have a gc pause for y milliseconds, how would that affect the distribution?

Calculus underpins almost every scientific advancement since its discovery. To exclude it from the curriculum at a school focused on technology would be insane.

There are discrete forms of “gradients” which generalize classical CS concepts. Submodular optimization, for example, covers many algorithms that search for optimal configurations of discrete sets, and it does so by arguments that are analogous to convex optimization.

I'd like to see 1980s-1990s graphics programmers get away with just discrete math for all the innovations they did in that period.

Curiously, MIT does not have an university-wide computer requirement, though some departments do. They have been debating this requirement for decades.

Very few developers work in machine learning.

I don’t remember the details anymore, but my combinatorics class required calculus. Something to do with Taylor series.

How so? Things other than standardized tests are easier to game. Letters of recommendation, extracurriculars, GPA, etc.

this is a perfect example of why governmental services are so important - even if they are often run poorly. Without them, dis-equities inherently perpetuate due to external incentives.

The SAT is treated as a standardized test but it, as you note, is inaccessible to many. It is not government run or organized. I've found, from teaching undergraduates, that most of the students who go to college presume that everyone takes the SATs, that it is government administered, and that it is free.

compare this to the, still far from perfect and problematic in other ways, centralized university admissions testing system in many other countries.

It's really weird to see this as the top comment.

Is there evidence physical access to the test center still a problem today (small or large)?

Is SAT worse than other criteria (such as extra-curriculum achievements) when looking at those poor suburbun middle of nowhere students? i.e. even in this example, did SAT actually made thing worse for those people when comparing against the alternative.

It’s not just the getting to a testing center, it’s awareness (and cost) of the process.

Was a long time ago now, but parents weren’t really worried about or involved in my college prep process, so I had to figure it out for myself.

It wasn’t until later I realized it would have been better to start taking the test way before graduation. To say nothing of the benefits of test prep..

I took AP classes in high school, but couldn't afford to pay $65 per exam (at the time) to take the AP test for college credit.

> The real culprit here is the lack of public infrastructure.

What was your area like? I grew up similarly poor but in a wealthy area and would walk several miles to a public library for early computer access, biked to school, etc.

Even then, our school offered SAT, etc.

That's not the political argument being made against them though.

Have you ever tried using a bicycle?

This is the best outcome but then you don't get "prestigious" schools of the elite. I went to a public university and I'm much better off for it than my MIT peers.

I think the bigger question is whether we should have exclusive schools like MIT.

Screening for college admissions made sense a century ago, when 5% of people went to college. Today, not so much. From an undergrad class perspective, the difference between MIT and a better state school is mostly the brand stamp and the admissions department.

Why do we want to stamp people with a stamp like "MIT"?

MIT is not an Ivy League school, and does not give any preference to legacy applicants.

I agree, but MIT doesn't care about legacy admissions. They're one of the few schools that doesn't.

Without legacy and the prestige and the entire shebang of old English style college, the Ivy Leagues aren't the Ivy League.

I agree with you that eliminating legacy would solve the issue of making the school different - it's never gonna happen though, so there's no point in talking about it honestly.

Also yeah, MIT is not a legacy giving school

Here's how "former MIT admissions director" McGreggor Crowley justified providing preferences to children of alumni and wealthy donors:

"What about university donors, though? Don’t they have an unfair advantage in this process? In truth, for every office of admissions there is a development office that builds a university’s endowment through donations from alumni and wealthy individuals. And every year, regardless of what a college or university says publicly, a number of children of wealthy donors and alumni get a nod in their direction while other applicants are rejected.

The reality is, the money generated by admitting wealthy students often serves to subsidize the financial aid of those less fortunate. If one squints, one might see here a karmic balance enabling many students to attend a college they otherwise could never afford."

Note he said "every office of admissions" and "regardless of what a college or university says publicly." If MIT were an exception, presumably he would have mentioned it. The "regardless of what a college or university says publicly" implies that MIT may be not stating the entire truth when they claim "we don't do legacy"[2] or that MIT's internal behavior may have changed since Crowley worked there. I'm not sure what MIT has to say about providing admissions preferences to children of wealthy donors the way many (most?)[3] universities including Stanford[4] do.

[1] https://www.bostonglobe.com/opinion/editorials/2019/03/13/co...

[2] https://mitadmissions.org/blogs/entry/just-to-be-clear-we-do...

[3] https://www.latimes.com/local/lanow/la-me-ln-college-admissi...

[4] https://provost.stanford.edu/2020/06/26/admissions-considera...

P.S. Many people dropped out of Caltech because of the stress of exams.

I don't know that all stressors are alike. I don't feel stress in exam situations, most of the stress I feel at work is around social stressors. I don't think a university education should be easy persay, but I do think that the stressors of education should not be unique in the world. If success in the face of stresses of examination in school don't predict success in the face of stressors you face elsewhere in life, then maybe they should be re-evaluated.

Yeah.... no.

Number of stupid mistakes on a test is not indicative of success at MIT. I was lucky in that I test well from having competed in different math contests, but I don't confuse that with being better. If I didn't compete in those, I would have been no less qualified.

The article is spot-on. MIT ignores test scores above some threshold. That's a reasonable way to go.

While acceptance rates have gotten lower, some of it is due to more people applying.

Average scores are higher because more people are studying for exams than 20 years ago.

But I don't think the course work has necessarily gotten any more difficult.

So while it's true that 1480 wouldn't get you in MIT these days, if the same exact "You" were born amongst the kids applying to college these days, your score would likely be higher.

I doubt it - 1480 could easily be indicative of stress-related or attentional mistakes, rather than lack of mastery.

I had the same observation. There is no data. It's nonsensical if you think about it: MIT stops using SAT/ACT in 2020 and now in March 2022 they have "robust" data that those students don't perform as well academically.

In the case of students trying to get into MIT I wonder if there is actually much point in gaming the test?

The article says they are using the tests as a threshold. It sounds like they don't care by how much you pass the threshold, just that you have passed it.

I'd expect that most students who will be able to survive at MIT can make the SAT threshold with no gaming of the test and no test prep other than maybe doing one or two free sample tests.

Someone who could not easily make the threshold on their own who games their way in is just going to find that the coursework crushes them. All that gaming their way in gains them is the ability to in a year or two add "flunked out of MIT" to their bio.

Very good point, and explains exactly why there should be no extra time on these tests. Give everyone the same time, same test, see how they do on the curve, and then you can make accommodations for disabilities ie 'how do you do among students with your disability or disabilities.

isn't that already the case for all grades? and all extracurriculars? (i.e., picking teams that allow anyone to join, or giving students who couldn't make the cut a pity position on a team such as `team manager` so they can add it to their college application?)

> MIT will be accepting anyone over a (presumably secret) threshold

I don’t think it’s a hard threshold. Some people are bad standardised test takers. If the rest of their application shows they won’t flunk their math tests, a lower score could be fine. If, on the other hand, it looks like a pattern, a marginal score could be seen to not make the cut.

I think the parent means "a score above a threshold gets the rest of your application read."

Certainly the article would not support reading "accepting" as "admitting".

"We do not prefer people with perfect scores"

"our research shows students also need [...] the resilience to rebound from its challenges, and the initiative to make use of its resources. That’s why we don’t select students solely on how well they score on the tests, but only consider scores to the extent they help us feel more confident about an applicant’s preparedness ..."

I read this as saying two extremely important traits are:

* resilience

* initiative

Not test scores.

This seems a bit ridiculous. One would assume they could fill their entire class with 1600 SAT scorers but they don't and I think it's well known that isn't even sufficient.

It seems like a good approach but the funding of things like special needs education (FAPE) is federal in nature; cutting those would likely nuke special needs programs nationwide.

You would need to start with your 2nd or 3rd "approximation" of your iterative approach (ie, pre-calc which programs are already popular/effective and keep those) unless your goal is to cause maximum disruption and possibly jeopardize your ability to do the "make Federal recommendations" effectively.

I'm always mildly assumed that Switzerland has 26 widely-different (far more variety than found in the US) school systems with only ~8 million people. And a high-school degree that guarantees entrance into any university in any subject.

Their are federal standards for said high-school degree (also for earlier education stages). And it's a relatively hard degree to get (22% of students get it).

I had a high school teacher that liked to say the following:

You can be smart and lazy and do fine. You can be dim and hard-working and do fine. But smart and hard-working will always beat them both.

I was pretty clearly in the smart and lazy camp, and was OK with that lot. And, as he said, I've done fine.

EDIT: I should add, though, that I'm very happy the SAT did not have the writing section when I took it. I severely doubt it would have improved my relative score.

That opinion is unpopular because it is incomplete. SAT/ACT scores correlate with lots of things, not just IQ and work ethic.

Another huge correlation is with family income:

https://budgetmodel.wharton.upenn.edu/issues/2021/9/28/is-in....

Having a high IQ means you can probably retain information better with less effort but for the vast majority of students, its a question of consistent work ethic.

Important to differentiate places like South Korea has a massive privatized cram schools where it does a fantastic job of creating high scorers but because it is scaled, it means if you are not paying extra and spending ridiculous time in privatized cram schools, you are almost certainly going to fail unless you can rely on some innate genius. Here in this case, it is truly discriminating students based on the economic capacity of their parents and it has the highest houshold debt not only from real estate but from debt to finance the young into educational hell.

Another unpopular opinion: Unhappiness correlates with high IQ if a proper outlet is not met due to a variety of factors.

> A student can literally have a perfect GPA in the best courses offered by their school and be considered a very poor applicant academically.

Do you have a source for this?

> Harvard considers roughly 4 in 5 applicants to be academically capable of doing the work at Harvard

I have no doubt that this is true of Harvard. I mean, after all, you can pick your own classes! That said, I think there is a difference between admitting just those capable of doing the work vs. a set of some of the best of the best, in that that second group will be the one filling the advanced physics classes for first years or whatever.

You want to do a calc test instead? I wouldn't even complain. If the average leetcode is hard for someone they're probably going to suck at even just integration e^z over the unit circle showing your work.

I suspect the engineering schools will be the least affected / first to recover from this trend.

I can't comment on law, but back when I took the SAT and ACT (granted, it was a previous millenium) you could retake the test. The issue was, unless you were sick or otherwise in an abnormal state, you would usually get a pretty similar score the second time. They provide a few examples of old tests, so you know going in what your score is likely to be, thus you will know if it is likely to improve much if you take it again.

> Nothing about these high pressure situations really reflects anything important in the real world.

Then why do the rest scores correlate better with college performance than other factors?

If SAT scores also correlate well with, say, income, would you then accept that they actually reflect something important in the real world?

> I'm a law student now. In law school, your entire grade usually comes from one 3-4 hour essay exam. Much depends on luck, how well you slept the night before, and how frantically you can type. It's absurd. It's been that way for some 100 years.

Well, not quite that way for 100 years. When I was in law school 30 years ago (University of Washington) it would for most people be "how frantically you can handwrite" rather than "how frantically you can type".

There was a room set aside for people who wanted to bring and use typewriters but it was fairly hard to actually find a typewriter that they would allow. By the early '90s even low end typewriters often had several lines of buffer memory and an LCD display so that you could store and edit text before printing it, and higher end models were essentially specialized laptops that only ran a word processor and a printer driver. At my school if it had two or more lines of text storage it was considered to be a word processor or computer and not allowed.

I had to drive all over Seattle before finding a place that still sold typewriters plain enough to be allowed.

Still standardized tests are the most fair and least gameable way for admissions. Just because there is some issues, doesn't mean everything else isn't infinitely worse.

I think I'd have less of a problem with the tests if the time limit was removed and you could bring reference materials with you.

The "expensive preparatory courses" offer something like a 10-30 point benefit over checking out an SAT prep book from the library, if you control for the time spent on preparation, last I checked.

What the courses do offer, of course, is the structure to force indifferent students to sit through the preparation.

There's a lot of bad statistics, whether by accident or with an agenda, being done around the SAT....

But more important than all that, as the MIT post says, pretty much all the other measures elite colleges use for admitting students correlate even _more_ with income than the SAT does.

because certain demographics performed poorly at it, that was grounds to label it systematic racism and justifying affirmative action vs. meritocracy by discriminating other demographics in favor of ones that were perceived to be victims of systematic racism

if you are confused so am I. I still don't understand how they can discriminate against americans of asian descent who do well on standardized tests but not call it for what it is.

Feelings and "demographics" and surnames.

The SAT/ACT aren't IQ tests.

> Standardized testing is a joke. There are other, better forms of tests. See: Europe

Europe isn't a country.

France has BAC and BEPC and plenty of other forms of standardized tests at every step of its education system, as a requirement to integrate highschool or college. Same in Spain or Germany. If anything most countries in Europe have a more conservative approach when it comes to candidate selection: strict standardized tests.

All Nordic countries have (Standardized) national level testing and also have tests that have to do with the local school. All of them are top performers when it comes to education.

The no-test policy is espoused by two groups who find themselves to be unlikely allies: naive progressives and actual racists.

The naive progressives think what you’d expect: “Minorities and poor people can’t possibly be expected to do well on anything objective, so it’s unfair to test them”. It is bigotry, but at least it’s well-meaning.

The actual racists are more cynical: “I don’t want Yale to be 67% Asian.” Obviously, this is even worse.

I grew up not terribly well off and I knew somewhere in the back of my mind that college admission depended on extracurriculars and such - I never really thought much about that until I grew up, did relatively well for myself and had my own kids. Then I found out just how much these extracurriculars cost and how much parental guidance is involved in sticking with them. My kids did sports in high school - but the busses don't take kids to and from the school in the off-hours that sport practices take place, so I had to drive them in early and pick them up late. That was an option for me - it wouldn't have been for my dad. The only reason they made the teams in the first place, also, was because they had been doing rec league sports since they were little kids and were already competitive going into high school (we knew plenty of kids who tried out for the teams and didn't make it). We've sunk who knows how much money into private lessons/coaching/one-on-ones, etc. None of this would have been possible for my parents, even if we'd been the type of family that did that sort of thing.

You don't put out fires by disabling smoke alarms. And you don't solve socioeconomic / class-related barriers by disabling their indicators.

> When you have a test, you have something definite to prepare for. Even if you do not have dedicated mentors or well-wishing, caring teachers, you simply know there is a test that significantly improves your chances for MIT.

Totally tangential, but this is also why I’ve slowly come around to appreciating technical interviews. Yes, it’s annoying, but it’s also a pretty straightforward path to getting a $150k+/yr job.

It's far easier for rich students to game GPA, college essays and extracurriculars than it is for them to game the SAT.

For GPA, they can hire a private tutor in the subject they're struggling with. There are services out there that will basically write your english essay for you / do your math/science homework.

For college essays, they can hire college counselors to help them draft a compelling essay.

For extra curriculars, they can hire a private coach, etc.

However, there is very limited evidence that SAT coaching actually increases your SAT score. The NACAC did a study on this and they found that average gains of test-prep students is ~30 points (this was when the SAT was 2400 points).

A 30 point increase out of 2400 points is not material to college admissions.

Out of the college essay, gpa, extracurriculars, etc. the SAT is the least influenced by your socioeconomic status. There obviously is an influence, but removing the SAT means more reliance on even more skewed factors.

Here's the study -> https://files.eric.ed.gov/fulltext/ED505529.pdf

> So much for that common, popular notion that standardized tests do not predict anything of value.

To be fair, I don't think the debate was ever about the quality or predictive value of the tests. There is a small, but well-organized and vocal subset of the population that hates the idea of excellence and differentiation. They want, and have been quite successful in, the replacement of standards of excellence with vaguely defined (defined by them, of course) buzzwords like "equity" and "diversity".

> So much for that common, popular notion that standardized tests do not predict anything of value.

To people with a particular agenda -- that society will be improved if equal outcomes are mandated -- I suppose standardized tests aren't "valuable" to them.

I am very grateful for the SAT. I wasn't a good student. I was unable to do any work outside of school because of a bad home (my brother did hours long "hand-clapping/stimming" and chanting rituals, my mother drank, and my father was violent). But I did well on the SAT -- enough to get a national merit scholarship and a scholarship from Hofstra University where I got my BA in Math in the early 80s.

If it wasn't for the SAT I don't know what I would have done. Standardized tests are the only answer, and I was very upset when I read of schools getting rid of using them for admission.

SATs are also a great predictor of a person's ability to complete college. That was one of their original uses. So without SATs, you'll get more people doing poorly, and more people with no degrees and a lot of debt. And if these people are members of what's considered an "under-represented minority" then there will be even more remedial action required elsewhere to fix the problem of the high failure rate (like giving them degrees anyway, etc).

I guess the issue in their analysis is that they are trying to predict MIT grade success, which is possibly another flawed metric. If say exams at MIT resembled SAT, it seems more logical that you'd find a correlation between SAT success and MIT academic success.

What would be a more interesting measure of real value is to study for academic innovation and invention (valuable to society), as well as future success on the job market or at new business ventures (valuable to the student and economy).

I'm making the assumption here that since we have limited educational resources, we'd want to provide the best education to those most likely to advance an academic field through new discovery or insight or invention, as well as those who'd best innovate or provide for existing business and services to society.

And I'm curious if we've ever had any study looking into that? Or if this one did?

Just an interesting anecdote on the predictive power of standardized tests:

When I took the ACT in 2006, I scored around a 24 or 26 for composite score. Not very good. I didn't prep nor study for it because I was pretty apathetic about school and did my best to coast on whatever natural talent I could muster. Since my score wasn't very good, I retook it several weeks later. About halfway through I realized I was given the exact same question set as the first time I took it. I had of course not studied nor prepped for the second time, being the apathetic teenager I was. However this time I score around 32 or 34 or so. I don't remember exactly.

What was different? Why did my score go up about 8 points? Better mood that day? Which score was the "real" one that best represented my abilities?

Did I eat a better breakfast beforehand? I recognized some of the questions but of course I never expected to see the same questions a second time so I didn't prep for that. I didn't prep for anything about it.

I wonder how many kid's college admissions results are ultimately because of one bad or good day? I suppose the other lesson is if you get a bad score, try again. You only need one good score.

MIT and Caltech have always held a 'number-oriented universities' perception in my mind. This is in contrast to a 'prestige-oriented or identity-oriented' perception that I hold for Ivies.

It is nice to see MIT do justice to those priors.

What's surprising though, is that APs and similar exams are not enough. In the UK, I though they essentially looked at A Level results, which are much more representative of what you'll actually study at uni. But I guess both SAT/ACTs & APs must be a better measure that just APs. I just remember fucking hating studying for the SATs though. So boring. SAT IIs were somewhat fun to study for though. In France for instance, they mostly just look at the baccalaureat to get into prep schools / first year at uni. Then exams to get into engr/business/vet schools are actually very interesting topics and very close to what you'll actually study. Same with exams at the end of the first year of med school (which you get into right after 12th grade, unlike in the US where it's after your bachelors).

That being said, they seem to have backed up their numbers, and MIT knows how to count, so they must be right! I just always hoped SAT/ACTs weren't that conclusive so that we didn't have to go through them anymore and could focus on the funner AP/A Level stuff :)

It's important to keep in mind that the context here is the SAT/ACT versus the currently available alternatives. They're not saying that the SAT/ACT is good, but simply that it's better than other options. Some of those options, like alternative standard tests, may be significantly better, but in the world as it is today, it doesn't matter if they are, because lower-income candidates don't have access to them.

The current discussion I thought focuses around the idea that higher scores should just about guarantee acceptance due some fantasy notion of objective merit, and here they pretty much say that's not the case here:

> To be clear, performance on standardized tests is not the central focus of our holistic admissions process. We do not prefer people with perfect scores; indeed, despite what some people infer from our statistics, we do not consider an applicant’s scores at all beyond the point where preparedness has been established as part of a multifactor analysis

> So much for that common, popular notion that standardized tests do not predict anything of value.

It was always an example of people refusing to believe something because it would be nicer if it weren't true.

The University of California commissioned a study to determine the predictive value of SATs for college success. A strong positive correlation was found. The test requirement was dropped anyway.

The problem with testing isn't that it's inaccurate, but that it's a poor tool for the problem that it's being used to solve. You can't express "likelihood of success in college" in one number. Looking back on my time in college, the idea that my success could be predicted by my knowledge of geometry and vocabulary words is laughable.

If there was a test that could accurately predict someone's chances of "success", be it SAT or IQ test, it would be used by everyone for everything. Billion dollar companies would be giving CEOs a version of that test before hiring them. A near perfect SAT score is noteworthy personal trivia, but other than that it loses all meaning as soon as someone steps on campus.

Schools with strong brands use tests because they have way more applicants than they can properly review. Standardized testing is made necessary by scale, not predictive accuracy. I'm sure that anyone who has worked in college admissions for years has a very clear picture of what a successful student looks like. But there doesn't seem to be a way to quantify and codify that knowledge. And there likely wouldn't be time to apply it to tens of thousands of people in a few months anyway. So, standardized testing is what we have until someone comes up with something better.

Many years ago--business grad school--one of the professors (forecasting?) had done research into GMATs and various post-school success metrics. Like it or not, the standardized test scores were a better predictor (by far) than anything else.

In terms of gaming scores on standardized tests generally, yes people with more money can take prep classes and the like, but there are also test prep books available--presumably even from the library--that probably get you a lot of the way there.

> common, popular notion that standardized tests do not predict anything of value

It's backed up by research, for example: https://news.uchicago.edu/story/test-scores-dont-stack-gpas-...

I don't doubt that MIT's study showed otherwise for their needs. But UChicago is also a top-tier school and is not requiring standardized tests, for educational not political reasons.

Don't underestimate the value of the kind of conscientiousness it takes to do well on these tests! I found grit much more useful than cleverness when I was an academic researcher.

I wonder why we can't hire more teachers to grade word problems in tests like SAT, like Asian countries do. Those problems are much harder to game or cram.

> IQ science is well established, and it is well known that the SAT is an IQ test, plain and simple. To pretend otherwise is to dissemble, to be anti-science. Hopefully, MIT will save itself, but I'm sadly not expecting miracles.

The "woke mob" you called is also well established, and people in academia have been fired before for recognizing aspects of IQ tests (e.g. Ronald Sullivan, a dean at Harvard). I agree with your sentiment, but also want to recognize that it takes time for the policy makers at MIT to fight against this trend.

Using IQ as the sole metric for intelligence is bogus though, in my opinion. Hard working "low IQ" slow learners beat mentally fragile or lazy "high IQ" quick learners 9 times out of 10 in my book. Low(er) IQ people can still master advanced sciences, it just takes them longer to do so. Standardized tests filter for the quickest learners who can think the fastest but they don't do as well filtering for mental fortitude or work ethic. Completely anecdotal, but it seems like a lot of the smartest people I've ever known ("high IQ") turned out to be mentally fragile (paralyzed with nihilism, depression, purposelessness). But most of the hardest workers I know who required a few extra years of work to master their fields ("low IQ") are extremely dependable and have high mental resilience by contrast. Of course, you can't beat hard working, high IQ people, but they are very rare in my experience, at least in the fields I've worked in.

I read that sentence as they think SAT scores are correlated with but not causal for MIT student success. You would need to do a randomised controlled trial or a mendelian randomisation analysis (requires a lot of genotyping of students) to establish any kind of causality. Those experiments have not been done, so the relationship is correlated.

> a wealthy one who thought he would leave his fortune to MIT

A tangential question, so I apologize in advance but how did you build your fortune? I am curious if it was not through entrepreneurship.

So let's cut the shit then and just do IQ tests.

This is the best technical university in the world offering the best technical education in history. It's where the smartest minds learn foundational knowledge that will enable them to make amazing technological contributions to mankind.

Not everything needs to be a battleground for the boring diversity of skin color. It's actually important that we get the most qualified people in these seats.

IQ Tests aren't illegal though? If colleges wanted to administer IQ tests they absolutely could.

Why are IQ tests controversial?

> SAT scores correlate with IQ test scores.

Do you have a source on this?

A is for aptitude.

Speaking from personal experience, it does help in at least two ways:

(1) Being surrounded by other hard-working students pushes you to do better and exposes you to more advanced classes and research early on (a majority of my friends in CS started taking graduate courses by their sophomore year and did undergraduate research at least for a summer in one of research labs on campus).

(2) Recruiting/Ability to get interviews. It isn't a problem in getting interviews for software engineering internships/full-time positions if you have MIT, Stanford or Berkeley on your resume.

Compare a list of companies that attend MIT's career fair vs those that attend UMass. There will be some overlap, and it's not a closed door if you don't go to MIT/Stanford/etc. But, if you want to land an internship at "the best" companies (which can often lead to jobs), being recruited at a college career fair is the best option.

i grew up poor and SAT helped me escape poverty. are we forced to only accept negative outcomes when sharing experiences ?

I think you’re focusing your anger on the wrong target. Is it Google’s fault for performing blind objective skill tests? Or is it your society’s fault for not giving you the chance to grow up fed and safe enough to study?

If I were you I wouldn't apply right away to a 4 year college. Do community college in your chosen field and then transfer.

In California if can do 2 years in a community college you are guaranteed a spot in a school in the UC system. Not UC Berkeley, but still some place great. I think to qualify for this transfer you need to have a 'C' or 'B' average.

If you stay on the pity pot you won't get anywhere. Also, in California, community college is very cheap.

I grew up poor enough that I got a full need-based ride to a state school. The SAT helped me differentiate myself from all the kids with expensive summer camps, tutoring for competitions, and expensive sports lessons that didn't actually do that well in school.

> Google style hiring process

A couple leetcode questions is hardly a difficult task compared to getting matched for residency as a doctor or accruing enough volunteering experience as a poor college kid for law school. Don't even get me started on investment banking.

Sorry that reality has hit you I guess? My perspective has always been that if you fail you should get up, evaluate what went wrong, and try again.

> but I was rejected everywhere I applied

Did you try going to community college and transferring?

I do think the standard leetcode style interviews are often administered poorly.

But I also think that if you create a fairly complex mousetrap, the profile of the average person who gets to the other side will be decent enough.

So likely many false negatives, but works at scale.

I'd do it very differently for my own company. But I think some of the bigger cos have actually gotten better, e.g. not asking certain types of questions too far removed from 99% of programming work.