I'm a PhD student at Caltech. There's also a push to eliminate the GRE as a requirement for graduate admissions, in much the same way as some people want to eliminate the SAT as a requirement for undergraduate admissions. In my opinion both proposals are well-intentioned but ultimately bad ideas.
I went to Caltech in the 70's. Caltech did a pretty good job of only admitting people who could hack the program. But they did make mistakes here and there, and those would suffer the humiliation of flunking out. This did nobody any good.
BTW, I found out years later that they took a chance on admitting me, and I very nearly flunked out my freshman year before I found my "sea legs". I was rejected by MIT, Harvard, and Stanford.
My run-of-the-mill state school had quite lax admission standards. It was understood that most of the freshman class each year would just party and flunk out after one or two semesters. As long as grading standards are maintained, I suppose that's another way to go.
Some people self study and pass GRE subject tests without having done the relevant undergraduate degree. Especially in fields such as mathematics and physics where testing results are easily quantified, this is a gradually growing long-term threat to the credentialing power of universities.
If an independent company comes along that can reliably get students to equivalent GRE subject test scores in a hard science most quickly than the university itself does, the result could be many of the best students stop applying. Some would just take the nontraditional route, save a fortune, ace their GREs and apply as a graduate student. Others would take their tests as evidence of their skills and and go directly into industry.
Making this move now, before such a challenger exists seems like a "good" idea from the selfish perspective of universities.
I had a GRE score in the 50th percentile and was still accepted because of my research focus. My PhD was fairly successful with a track record that was better than most of my peers. So, my disdain for these general-purpose tests is quite personal.
In my opinion, the GRE mostly checks how good you are as a test taker. None of that is important to academic success. It invariably filters for a fairly streamlined and optimized class of people, who have the money and time to invest into such a test.
The fact that the ETS seems to have a clear profit motive really rubs me in the wrong way. It is a private organization attaching itself to education and siphoning the money off students by selling their tests and training materials. Universities should design their own admissions tests and not rely on such parasitic entities.
I don't agree with the idea that one can "optimize" for the GRE without knowing anything relevant to the field the GRE is in.
I did not optimize for the GRE, I did not prep for it, I did not look at a study guide for it, I simply went and took the test cold. The questions were on material one would expect to cover in basic engineering and science courses.
I don't buy the notion that anyone who does well on the GRE is doing it with tricks and fake/useless knowledge and/or cheated.
But you are kind of proving the point with your example... you are in the minority, and most people do not take it "cold". If everyone took it cold, then what you said makes more sense. But most people, unlike you, are trying to game the system for a higher score, through extensive prep work and additional test-specific training (else, why would there be so many test prep books and courses and tutors out there?). The fact that you didn't utilize those methods doesn't mean others won't.
In a sense, you're like an athlete who doesn't take steroids or doesn't push the limits of doping policy by trying to game the countermeasures, and instead of feeling aggrieved that others are achieving similar/better results as you via their just-within-the-letter-of-the-law tactics (or sometimes wildly outside the law tactics, a la Varsity Blues), instead you still feel some allegiance to an idealized view of a fair system that simply doesn't exist.
> why would there be so many test prep books and courses and tutors out there?
People are always looking for an edge, and entrepreneurs are always going to be trying to sell them one, whether it works or not. (Look at all the diet books out there, it's a billion dollar industry.)
> idealized view of a fair system that simply doesn't exist.
I'm not suggesting it is ideal or perfectly fair. I'm suggesting it isn't nearly as irrelevant as many argue.
For those who want to prep or game the system, why not do it right? Take the math classes, and actually learn math.
Most of Caltech's revenue is from government and private research grants, not tuition. They only raised it because they don't have economies of scale for their undergrad program and admissions wanted to keep up with the Joneses at Stanford and Princeton,
Why are they bad ideas? I took the SAT and GRE for undergraduate and graduate schools respectively and neither was worthwhile preparation for anything I studied. If we want to put up barriers to entry then I think we can find better and more predictive ways than measuring general pattern matching skills with the SAT and GRE.
Admissions to the top US universities—and hence, most chances for social advancement in the US—will henceforth be based entirely on shifting and nebulous criteria that rich, well-connected kids and their parents spend most of their lives figuring out, rather than merely mostly based on such criteria. The last side door for smart noncomformist kids is now being slammed shut.
It reminds me of a section of Thomas Sowell's book Discrimination and Disparities.
It was talking about automatic background checks for applicants at businesses. Some people wanted such processes to be illegal, on the grounds that they were supposedly racist.
In reality, automatic (i.e. indiscriminate) background checks actually resulted in more blacks being hired. Even if the local black population had a higher or much higher percentage of people with bad marks on background checks, being able to screen them out and allow those blacks with no bad marks to apply was made easy.
Without automatic background checks, employers fell back on their own biases, resulting in fewer blacks being hired.
The SAT is like an automatic background check. Without this indiscriminate screening, the schools have more ability to discriminate based on biases.
Credit checks had a similar impact for minority and disadvantaged communities. Before, if you wanted to get a mortgage or a car loan, the bank would assess you based on a number of factors. Many banks explicitly used race as a criteria for denial. Now, with near instant credit checks, race has been removed as an explicit criteria, rightfully leading to much higher rates of loans and mortgages for minority and disadvantaged communities. Are credit checks anywhere close to reliable or fair? Can you correlate race based on zip code or other factors? Yes, these methods are not anywhere close to 'good enough'. But they are miles ahead of the explicit racism that preceded these form of financial product applications.
Eh, this kind of "the sky is falling" argument doesn't hold much water to me. So, some "smart nonconformist kid" doesn't get into Harvard. Is that the end of the world? Larry Page went to Michigan, Brin to Maryland - they did fine. There are a lot of great public schools out there. Talented people like that aren't getting locked out of anything.
A much bigger problem in education starts much earlier. If your parents don't provide a stable and education-leaning home life, or you live in a school district where safety is a bigger concern than educational quality, well, social mobility becomes a lot harder then. These are some hard problems, but reach way beyond education, into the social safety net and beyond. Guaranteeing that a bright kid who is not born into the best circumstances gets a fair shot at life is still a work in progress, but has little to do with whether they get into Harvard or MIT.
That's really not the argument though is it? Saying "so what if someone are left out of X, Y is a comparable alternative" doesn't change the fact that X is being shut out. If you remove these "objective" measures, it starts being reliant on "holistic factors", and feels like this inherently gives an advantage to well off kids whose parents can send them on different summer schools or labs or whatever. Harvard (and the Ivy league) still maintains a reputation and the industry still does care about it (especially in liberal arts / banking etc), so these fields will reduce in accessibility to those who can't develop "holistic skills".
The other problem you point to is very real, but that's independent of the issue currently being discussed in this post.
Colleges, as insitutions, care about progressive, equality of opportunity values. Those nebulous criteria will coalesce to produce roughly the same outcomes that they currently do.
They might adjust their admission picks until e.g. aggregate ethnic/racial admission stats look the same as before, but at the same time picking more of the rich, well-connected black/other minority kids and fewer of the poor but hardworking ones.
This reminds me of an article about tuition cost going up by just the right amount to make sure 55% of people qualify for financial aid. I don't remember where I read this but I remember the argument making sense. What numbers do we get when we look at what percentage of people qualify for financial aid at Harvard and how has this number changed? I wonder if they keep track of and make this data available?
Some? I think the assumption implicit in this thread of the conversation is that scoring well on the SATs is a way for brilliant poor kids to compete with rich well-connected kids.
When I attended Caltech, I didn't notice any bias towards rich kids. The ones whose families I knew were all economically ordinary. None of the kids had a Porsche or Ferrari or BMW or anything expensive. Hal Finney had an ancient VW bug, and I had my 67 Mustang. Another friend bought a Chevette upon graduation with his own money.
Yes. I understand this is the assumption but why is this the assumption? I can imagine poor kids doing just as well if not better without artificial barriers to entry like the SAT and GRE. For some reason everyone is assuming these tests are "good" and "meritocratic" but I don't see it. Everyone seems to assume there need to be tests and no one is wondering what happens when there are no tests.
I suspect no one is wondering because without tests people have to face some harsh truths about economic and social class structures that most people in the US like to pretend do not exist. If these tests were about meritocracy then they would be enforced without exceptions and rich kids wouldn't be able to buy their way into their favorite schools like they do now after their parents make a generous donation. The existence of the test does not matter if you're rich enough so the argument about meritocracy rings hollow.
Poor kids aren't going to have the resources to compete as equitably on extracurriculars. A kid in a wealthy school might be able to host a fundraiser for charity and raise thousands of dollars easily, while a kid at a poor school would be able to raise only a fraction of that. That poor student is also going to have more trouble showing excellence in art/sports/intellectual hobbies if their parents aren't able to pay or take time off to send them off to competitions. The schools might be adjusting for the difficulty, but they'll be doing it in a very non-transparent way.
> Poor kids aren't going to have the resources to compete as equitably on extracurriculars.
Yes, that's my point. Given fewer learning opportunities poor kids end up at a disadvantage even though the next Einstein is more than likely somewhere in a poor neighborhood. The SAT might be a good way to find this person but it would be better if we required a standardized minimum from everyone and helped them as much as possible to achieve those minimum standards. That way the next Einstein wouldn't be like finding a needle in haystack with very poor objective measures of learning potential (Einstein himself being famously unfit for his own educational system based on standard measures of the time).
I am hard pressed to think of better methods to fairly and uniformly measure if person A is more qualified for college than person B. There has to be something to differentiate, as objectively as possible, between candidates. And, while not perfect, standardized tests fit the bill.
I, for one, find standardized tests far more stomach-able as barriers to entry to a university than the exclusive whims of administrators attempting to gauge the qualities of applicants.
Universities just need to do a good job of weighting the exams - I had lackluster LSAT scores, though not awful, but wrote an amazing law school entry essay and had a unique background: AmeriCorps, ski bum, tutor, etc... that catapulted me in to law school.
Entirely removing any type of exam from the college entry process just seems antithetical to what college is about.
Why aren't high school class grades enough? If we want predictive measures then why not make the preparation for college the actual measure instead of tacking it on at the end as another exam. I can see it being useful for people who don't go through the traditional path and need to prove that they have the equivalent knowledge of a typical American high schooler but otherwise I really think there is no value in these tests.
There are 24,000 high schools in the US (not even considering international students) with wildly varying grading standards. You have one applicant from Podunk City High School, valedictorian with a 4.0. Is that ... good? How does it compare to the thousands of other kids with 4.0s you have applying?
I'm completely sympathetic to the idea that grades should be a better predictor of performance than one day of testing, but in practice the grades have a lot of flaws too.
> There are 24,000 high schools in the US (not even considering international students) with wildly varying grading standards. You have one applicant from Podunk City High School, valedictorian with a 4.0. Is that ... good? How does it compare to the thousands of other kids with 4.0s you have applying?
Some other places in the world admit based on the Z-score for each subject, with the average and standard deviation being specific to each school. While the university might not know how one would compare to other students based on an equivalent test, they'd at least know that a particular student is in the top decile for their school in a given subject.
I don't know if it's good or bad. I don't think grades mean much either other than a feedback/signal for the student. Using them as measuring sticks for admission into exclusive clubs seems pretty broken. We should be lowering the barriers to education and the exclusive clubs that elite educational institutions demarcate.
I want more people that study calculus and higher level mathematics because it is a standard part of their education and not because it is a requirement on some test. I don't care what grade they get on their exams because if I end up working with them I'm not gonna care about the grade but the fact that they learned how to think rigorously (or at least were given the opportunity to do so without having to waste effort and energy on playing status games).
Because there's nothing even remotely standardized about high school education. A kid growing up in Seattle learns about evolution, while a kid growing up in small-town Oklahoma learns that Jesus rode a dinosaur to work. Cultural differences aside, a student who graduated #1 in a class of 50 probably enjoyed a fraction of the learning opportunities available to a child who graduated #12 of 500 at a larger school.
These are the problems that standardized tests were supposed to solve. Eliminate the tests, fine, but you'd better have an adequate replacement for them in mind.
I'm not arguing for their elimination. I'm sharing anecdotal evidence that they weren't useful for me and probably people like me who got an adequate education and had access to a local community college and a library.
Re: uneven distribution of education. This is the fundamental problem and standardized tests don't actually address the root cause. We really shouldn't be confused about what skills and concepts people need to learn up until high school and the rigor with which they need to learn them. Addressing this confusion is a hard problem so post high school standardized tests end up being the band aid. We can keep the tests and work to address the underlying root causes for uneven educational outcomes. Test preparation industry is essentially a form of tax on something that we should have democratized by now.
The SAT is an objective measurement. By dispensing with it, all the other measures are subjective. This means that it leaves the administration free to admit whoever they please, based on any agenda.
Are there any laws about SAT score requirements for school admission? Because if there aren't then the administrators are not obligated by any rules/laws for objective measures. If that's the case then they are free currently as well. No law means no actual requirement to abide by any rules that reduces freedom in their decision making process.
