Hacker News new | past | comments | ask | show | jobs | submit login
SAT Scores Fall as More Students Take the Test (wsj.com)
128 points by dpflan 18 days ago | hide | past | web | favorite | 153 comments



I missed 1 question when i took the SAT 6-7ish years ago, resulting in the second highest possible score. It made me a US Presidential Scholars Medal Program Candidate. Take what I say for what it’s worth, which probably isn’t a lot, but I’d like to share my experience and why I believe I scored well.

I never took any prep course. I always was of the mindset that the only value they provided was forced practice. I never bought that they taught secrets of test taking.

Instead I bought 2 books, and 2 books only. One was the official SAT Guide which was $20ish. Secondly I bought a Kaplan book which just had 12 tests in it for $20. Both of these books were also available in the reference section of the public library.

I worked through about a test a week the 4ish months prior to the SAT, and read articles online about the SAT if I came across them. I felt very in the know regarding the test and just flew through it when taking it.

I used that score to tutor others for pay, and my friends for free. I found out that just walking people through the test was about confidence, which really jumped people’s score up. Naturally most people I came across were between a 1500 and an 1800 out of 2400. For people below that, they tended to need remedial education to improve their scores. Between 1800 and 2150 usually was a result of understanding patterns in the test and what to look for. Those were learned through some rigor in prep using the two aforementioned books. Finally, for students who wanted to score above a 2150 to a 2400, the differences in those that did and didn’t were not educational content based, but rather big picture problem solving strategy in how to logic through problems and knowing what to look for.

The SAT, IMO, is a relatively good predictor of someone’s educational attainment the first time they try one prior to any prep. I scored exactly what I predicted and after $40 of self-study, which could’ve been done for free, I scored what I knew I could.


I did 2160 out of 2400 (1490 on the 1600 scale), which got me Merit Commended and an Achievement Scholars Candidacy. No test prep, save for basic instructions and having taken the PSAT twice. AFAIC I bombed the math portion and would have scored over 700 if I'd taken it a second time. I was a decent student, but nothing special. That was a decade ago. Currently working in a big chain retailer's warehouse for $17/hr.

Accordingly, I think the SAT is a bit of a farce. Like BMI, it's essentially worthless without context that, assessed prudently, could advise stakeholders on its own, without the test score. Likewise, it obfuscates the true nature of the strengths and weaknesses of the tested. With a 9xth percentile score, are you like me, possessing a high natural aptitude but dealing with the fallout of a mental health issues, a dysfunctional home life, and the fundamental difficulties of being black in American society? Are you a intellectually-middling child of wealthy parents who can afford to shove test prep down your throat until you "get it," and then provide for your professional needs post-college? Or are you a hard-worker who studied diligently, only to exit undergrad to find that your options are severely limited by your debt load?

This is all kind of beside the point, however. We get stuck in this quagmire of discussing how best to facilitate the sorting and tracking of our youth into will succeed/won't succeed boxes, when we should be working to assure success across the board. Why are some American colleges (entrance into which is the aim of taking the SAT in the first place) excellent, and others not? Why the acceptance of such disparities in educational quality?


>Why are some American colleges (entrance into which is the aim of taking the SAT in the first place) excellent, and others not? Why the acceptance of such disparities in educational quality?

Some people, such as Bryan Caplan, believe that college is mostly about signaling.[0] The idea is that most of the value college provides for individual students is to indicate to society that the student can jump through all the hoops necessary. It's much less about what the student actually learns in college. Essentially, college is mostly about being a filter - to weed out those that can't make it. This line of thinking would also explain college acceptance tests - they are the same type of filter, but applied before college. It's there to keep the college prestigious, so that more people will try getting into it, which will allow them to only pick the best of the best.

I think there's a lot of merit to the signaling idea, because a lot of what you learn in college doesn't have much practical value for the vast majority of students. College entrance tests are just one part of this whole process.

[0] https://blog.press.princeton.edu/2018/01/23/bryan-caplan-on-...


The Harvard Extension School is a good example of this. The school wants to make money by offering classes to more students, but doesn't want people to call them "Harvard students" because it would diminish the exclusivity in their branding (the same exclusivity that attracts people to the extension school).


I majored in EE, and can say it was only the non-STEM courses that were worthless. There is no way I’d gain the same amount of knowledge without attending college.


Sure, but a large number of majors consist of non-STEM courses. I don't think that non-STEM courses don't teach you useful things either, but I think that most people don't have a use for what those courses teach. (Some people do have use for those skills though.)


It definitively is about signalling but that doesn't mean you don't learn anything in college.


Of course you learn things in college. The point is that what we're doing as a society with college is inefficient. We require a certificate from job-seekers that says that they've gone through college. But college teaches a lot of people a lot of skills and knowledge that will not be useful for a large portion of the students. Essentially, our society requires a prospective employee to demonstrate capability in things that are not necessary for their job or even rest of their life.


Hi. This was very nicely written. I certainly understand that thinking regarding the big picture.

However, IMO, the SAT adds another variable for contextualization of these inequalities in the American education system, regardless of if the SAT itself is unequal.

Structural inequality in higher education is a whole different ballgame, and is much more complex. As someone who went to a public college, and then got to see what life was like at Ye Olde Ivy colleges and schools like Stanford/Berkeley, it's pretty apparent that there were some structural decisions from the top down that facilitated that success outside of even the economic differential between schools/programs.

Very tough things worth thinking about. I don't think that people treat standardized testing as some all telling oracle, and if they do, I believe that they're mistaken. I remember that our school offered something called the PLAN? which was like the PSAT for the ACT. The career advice it gave me based on my results was that I'd be good at landscape work.

I've never defined myself by my test taking capabilities but I will admit I always felt a little affronted upon reading everything from the movements that seek to de-legitimize all standardized testing. I felt that something I specifically chose to gamify and study for was being increasingly seen as a negative signal.

Life is tough man. I hope people succeed regardless of if they even go to college or not. With that in mind, I also hope that perhaps there can be a cultural shift that facilitates scoring better on aptitude testing because perhaps that may be a signal about our society, even if that is not a perfect signal.


If every college is excellent, none of them are.

I get that we should be raising standards across the board whenever possible. Education is extremely valuable and I'd like more people to be more educated. But there is always going to be a continuum of prestige, educational value, intelligence of students, etc. among various colleges.


Success and life satisfaction (shouldn't be) zero sum.


Probably not. But there is always going to be a spectrum of variously qualified college students, and they are going to sort into a hierarchy of colleges. That will be true even if every college graduate has a life that would be considered "successful." Of course, success and satisfaction tend to be moving scales as well, and are often judged against peers. That's not about being zero sum. More education probably grows the pie. But only a limited number of people can go to the best colleges, unless you lower standards such that they are all mediocre.


I agree, but rankings are by their very nature zero sum.

If one person goes up, another goes down unless they are exactly equal. That level of equality is something I think impossible to achieve and even if we did, humans would find some other way to rank people.

But compared to the past, we can make it better. There is no reason why the 50th percentile today cannot be equal to the 60th percentile of yesteryear.


I only did alright on the SAT, but I smoked the GRE by doing something similar.

The Kaplan book had a CD with a few practice tests. I think I did 2. If for no other reason, they are super useful for figuring out how to pace yourself.

The math section was (is?) easy for most people on this site. This seems the norm because the score/percentile relationship was incredibly high.

The English section kills people. My pro-tip is, in the same Kaplan book, they have a list of the most common words. I made flashcards for the top 100, and I'll be damned if like 80 of them didn't show up on the test. I imagine this is similar for the SATs.


In grad school, we took an unofficial survey of GRE verbal and quant scores. Far and away, the non English speaking students scored significantly higher than natives. Anecdotally, The non native speakers actually studied in that way you describe, and the native speakers assumed the fluency would carry its weight instead


Hi. Thanks for sharing this. I suspect that would be true for foreign students within programs across the board in the US, because you tend to learn as an outsider through an application of “rigor” and more raw memorization than higher order types of learning. I’m speaking this from experience as a foreigner who moved to the US early on. I’ve always doubted many higher order learning methods precisely because they didn’t ever distill the discipline to work better...


Weirdly, I scored completely average on the math for the GMAT despite an engineering and math degree and pretty much aced the verbal without any studying (figured out early on I needed to focus on studying quant, which I hated the test). But, I probably had a heavier English background than most people taking the test as I worked for newspapers and did a lot of personal writing for many years. Guess it's all about how you line up against the field in some cases.


This was my experience. I had an almost perfect score on the math section of the GRE, but that only landed me in the mid 80s percentile wise. On the other hand, my English score was quite a ways from perfect, but I was surprised to see that it was 98th percentile.


It definitely sucks that you can not miss a single question yet still end up with something like a 98th percentile because of the question set you happened to get.


Hi. Thank you for sharing your experience. I am about to take a GRE myself so this type of insight is very helpful. I too remember making a lot of flashcards for the SAT. I had a rather relaxed approach to studying but the consistency over time made it seem like everything I learned showed up one way or another on the test, especially vocal words.


As I'm sure you know (but maybe others don't), the main difference is that the GRE adapts to how you're doing, asking harder / easier questions when you get them right / wrong. This means the early questions have outweighed importance in your overall score and hence pacing yourself is very important. (I am very glad I did a few practice tests)

You want to leave yourself time for (hopefully) hard questions at the end, but a silly mistake early on hurts, so it pays to take your time in the first few minutes as well.

Fortunately it should be obvious how you are doing after the first group of questions, if they go from roll-your-eyes easy to ones that make you pause a moment. :)


You seem to think the amount of prep you did was minimal. I'd wager it's more than most students do. It's certainly well in excess of anything I did to prepare and while my score was lower, it was more than enough (along with grades, activities, etc) to get me into my state's top public university.

The test prep centers do basically what you did - they just cost more and hold the student accountable (in as much as the student spends time at the center and presumably works through the test exercises).


I'm 100% convinced there is no "secret" to test taking, however there are a hundred "tactics" and they apply variably to each person.

I was asked to take the SAT when I was 12 as part of the Johns Hopkins Gifted Talent Search (although back then I seem to recall it just being the Talent Search), this was back in 1990. I got a 790 on the verbal and 590 on math. When I took it "for real" in high school, I got an 800 on verbal and 690 on the math. I took one PSAT that I was required to take by the school, I never read any books, I never did practive tets at home, etc. I've ALWAYS been a "good test taker" in that they don't intimidate me at all, no nerves, etc. For me there's no trick to doing well on a test other than knowing the material.

I have friends who are very, VERY smart, and terrible test takers. My wife is very smart, and while she's good on tests, she needs to prep a fair amount to feel ready. Other people get nervous, or freeze, etc.

It's all about how a person deal with the situation of a test, and you have to come up with a psychological plan for that. Aside from that, you either know the information or you don't.


I did the same for JHU, and got good enough scores to get into their program. They wouldn’t let me in because I missed the age cutoff by 1 month. Screw JHU.


Seems like that rule should have been applied BEFORE you took the test, so as not to taunt you.


I think it was the issue of taking the SAT at 12, but turning 13 before the program. I don’t recall, it was almost 40 years ago. I just know my mother was pissed.


Heh, mine was 30 years ago, I hear ya!


Got an 1540/1600 on the SAT and a full score on the PSAT. I’m now in college on a National Merit Scholarship because of my PSAT score.

Today, it’s even easier to study for the SAT. Khan Academy offers free, official prep to every single student and 8 official practice exams to take. I also used free practice tests made by other companies to study, too. The only purchase I made for SAT prep was a $20 book from Kaplan that had 1000 practice questions and helpful (guessing) strategies.

The test nowadays seems to be a socioeconomic equalizer-those who can’t afford a book or a computer can always go to the public library and get the same resources. The only real reason anyone would pay for test prep classes is because it forces you to practice at regular intervals. Test prep classes also help students revise material that they may have been rusty on, which again you could do online nowadays. That’s why things like the Collegeboard adversity score seem absurd to me. Anyone, really, can afford to do well on the SAT.

Now, of course, the actual educational value of the test is another question altogether.


>I never took any prep course. I always was of the mindset that the only value they provided was forced practice. I never bought that they taught secrets of test taking.

...

>Secondly I bought a Kaplan book which just had 12 tests in it for $20.

Did the Kaplan book not have "secrets" of test taking? Mine did, and one of them was actually useful:

When you have an "analogy" question, where they give you an A:B, and then ask which of the five choices have the same analogy.

What do you do if you don't know the meaning of both A and B?

Well, of the five choices, you will find two that are identical as far as analogies go. As an example, if you have two choices:

cat:kitten

dog:puppy

You know neither of these two can be the answer because there would be no reason to pick one over the other.

Then you'll have another choice with absolutely no relationship (e.g. pen:donkey). You can eliminate that immediately. So now you're down to two choices. Pick one randomly.

But yeah - overall I didn't learn any other big secrets - other than the obvious one of working backwards from the answers for many quantitative problems (don't solve an equation for x - just start plugging them in!).


Hi. I think that the quotations around the secrets is totally warranted. I was trying to state that I always thought that the "secrets" that many test prep courses tried to sell to parents/students were just question design, and not actually some behind the scenes insight.

Do you really disagree with that? That's what a lot of the marketing seems to be focused on. I felt like question exposure alone got me those same insights, and in fact, the official SAT book rather explicitly lists those things out.

Most people don't read the annoying small text that accompanies the books that spells out a lot of those strategies. Instead they'd rather pay someone a lot of money to tell that to them :)


I myself did not take any SAT prep courses, so I can't disagree. My point was that some small tricks do exist, and most test takers are not aware of them. Knowing those tricks will give you a slight advantage.

For me, those prep courses would likely be a poor value. However, I likely am not in the demographic who would benefit most from them. There are plenty of people out there who haven't learned how to learn from a book, and so having someone teach them will result in better scores. I would have thought such people will likely struggle in college, yet my grad school had quite a few of them. I guess they found a way to compensate.

The trick I mentioned was not in the official SAT book.


I didn't do any prep for the SAT or ACT. Later I considered taking the GRE and read a prep book that had tips like above. I was astounded how much easier it made the verbal questions.


My story matches yours almost exactly. Didn't take a class, did a bunch of practice tests, got an almost perfect score.

I agree that confidence is very important, and aptitude matters too (otherwise everyone who took a class would get a 2400).

Commonly-given wrong advice is about guessing. You should never leave a question blank. Prior to 2016, guessing had an EV of 0, so if you had any hint that any of the answers are better or worse than chance, you should guess. I heard advice like "only guess if you can eliminate two answers", which is clearly wrong.

Now they got rid of the penalty for wrong answers, so guessing has positive EV even if you have no information at all.


I believe that the EV just changed recently though right? Because when I took the SAT, there was an incorrect answer penalty that now doesn't exist.


I ninja-edited my post. Previously the EV for uniform guessing was zero, because the points value for all the answers (1, -.25, -.25, -.25, -.25) summed to 0. Now the EV is positive because they got rid of the -.25 penalty.

To make things more complicated: if you were gunning for a high score in the old system, it was even more important to guess, because your partial points could get rounded up.


Thank you for the comment. Do you have any idea which question you missed?


Hi. Yes. The only section I did not get a perfect on was writing. I missed 1 of the multiple choice questions. I always believed that the grammar choices offered were a bit dubious but it is what it is. Totally possible and probably likely that I just got some grammar rule wrong, though I’d like to think that the test was wrong :)


Am I misremembering how the test is scored? My recollection was that score == writing score + math score + verbal score, with each section being worth 800. If you only missed a question on the writing, shouldn't your score have been at least 2300, and your 1600 scale score would have been 1600?


I believe so. The curve on the test, if that is the right word here, is a bit random. Sometimes one could miss 2 or 3 questions and still have an 800. My Math and Verbal Scores were 800 each. The one question I missed in Writing gave me a 790 out of 800, but in reality it was 590 out of the 600 allocated towards the multiple choice questions, with the "optional" writing section weighing in for the other 200 points. This is a gross oversimplification and could be totally wrong but that was what my score was.


Yes, I understand that, but that still doesn't seem to add up right to me. You seem to have gotten a perfect score on the optional writing section, so let's just ignore that since it would only increase your overall score. Your score was 800+800+590 out of 800+800+800, which is 2190/2200. Just applying the same ratio to 1600 gets 1592 on the 1600 scale. If I adjust for the fact that the 1600 scale is really "the same math test and writing+verbal combined into one score), I get 800 + (1390/1400)*800 = 1594. Right?

I know this isn't important, I'm just trying to understand where my math is breaking down.


I don't think that everything scales 1:1.

I took this from the College Board website:

"Your raw score is converted to a scaled score of 200 to 800 points, the score you see on your score report. We use a process that adjusts for slight differences in difficulty between various versions of the test (such as versions taken on different days).

We do this to make sure there’s no advantage in taking the test on a particular day. A score of 400, for instance, on one day’s test means the same thing as a 400 on a test taken on a different day—even though the questions are different."

https://collegereadiness.collegeboard.org/sat/scores/how-sat...


I definitely remember that it's scored, it's just really hard for me to imagine getting a single question wrong and having your overall score go down by over 100 points (almost 7%).


Not surprising. If only a small percentage of students take the test, it's only going to be the students on the top end of the bell curve. If everybody takes the test, you're going to get results that reflects the general population.

http://archive.is/6GW9b


This is the detail which needs to be added to almost all mainstream coverage of education stats — every year you'll see some hand-wringing about U.S. students falling behind internationally and the authors never mention that the effect is much weaker once you correct for the fact the U.S. has a much wider range of students taking most tests. Usually almost all of these metrics come back to the root cause of economic disparities along the spectrum from hard poverty to not being able to afford tutors and SAT prep courses.

(This is also why the cool curriculum ideas or charter school concepts we hear about never go anywhere: the TED Talk sounds great but inevitably once it scales up it turns out that the effects were due to some sort of selection bias (quite possibly unintentional) or simply lucky small sample sizes which peters out at larger scale)


> the U.S. has a much wider range of students taking most tests.

Citation? All countries have a mix of wealthy, educated students, working class students, students whose parents are drug addicts, immigrants who don't speak the local language, etc. In my experience the US isn't that different from other wealthy countries I've lived in.

SAT isn't really used for international comparison; a system like PISA does cross-sectional studies of countries, not elites. The results are not encouraging for the USA:

https://en.wikipedia.org/wiki/Programme_for_International_St...

https://nces.ed.gov/surveys/pisa/

http://www.oecd.org/pisa/


> Citation? All countries have a mix of wealthy, educated students, working class students, students whose parents are drug addicts, immigrants who don't speak the local language, etc. In my experience the US isn't that different from other wealthy countries I've lived in.

Yes, but many European countries engage in so-called tracking of students, putting them in separate tracks for tradeswork, skilled labor, and university such that the "dumb" ones aren't taking university entrance exams.

In the US this has become increasingly frowned upon by (perhaps) well-meaning administrators who feel that to do this is unconscionable. In my freshman year of high-school they had advanced academic and standard groupings in the same school, but removed this my sophmore year. They re-branded all the base-level courses as "College Prep," with "Honors" being the former advanced-track classes, AP and IB courses already being branded as such.

Of course, if you were in the advanced group the counselors and teachers would all push you into AP and IB courses but they also encouraged everyone to take the SAT and ACT tests, to apply to university, and discouraged enrolling in the trades program that the county offered (half-day academic, half-day trades classes).

This type of "everyone goes to college" mentality is why the SAT is a poor academic bellwether.


But in Europe this often goes hand-in-hand with mandatory schooling while in US people sort themselves by dropping out.


Mandatory schooling in most of Europe is up to the age of 15-16 and you can do w/e you want after that not unlike the US...

https://os.unil.cloud.switch.ch/tind-customer-edudoc/aa7b2aa...


In my home state, one cannot drop-out until 16 which is well past when one would be traditionally tracked.


>In the US this has become increasingly frowned upon by (perhaps) well-meaning administrators who feel that to do this is unconscionable.

Because it is unconscionable. I dropped out of high-school at 16. I was able to go back and earn a degree in computer science at the age of 28 at a Canadian school. Under the European system where your life path and career trajectory is decided for you at 14 years old, I wouldn't have been able to do this.


What about tracking students in high-school precludes your being able to attend university at the age of 28?

Insofar as I'm aware, one can change tracks in most tracking systems. It just isn't entirely common, likely because it puts most on a path where they should be.


> What about tracking students in high-school precludes your being able to attend university at the age of 28?

Because the "dumb" students are pushed into trades instead. And good luck going back to university when you don't have a long chain of prerequisite courses from age 14 onwards. In the states you can just get a GED and take the SAT (along with individual selection) to prove you belong in university.

Assigning a person a career based on a test they take at 14 seems awfully reductive and like it would segregate students based on the social standing of their parents. Wouldn't surprise me if the German system reduces social mobility.

Should students have their entire lives mapped out for them at 14? And if so, why at 14? Why not at 10, or 6, or at birth?


Nothing about a tracking system actively precludes your being able to jump tracks later in life. To be frank, you're blindly speculating based on this assumption.

> Assigning a person a career based on a test they take at 14 seems awfully reductive and like it would segregate students based on the social standing of their parents. Wouldn't surprise me if the German system reduces social mobility.

That's not how tracking works. They don't assign you a vocation.


>Nothing about a tracking system actively precludes your being able to jump tracks later in life

Nothing necessarily precludes it but does it really happen that often? I couldn't find any data on this. Is there any evidence that "changing careers" is easier and more likely under the German VET system as opposed to the US system? In fact, evidence suggests that the permeability between the academic and vocational tracks is quite low in actuality [0]. One would logically think that making long "on the track" prerequisite chains a requirement for a career change would add significant barriers to entry. And there is some evidence that supports my theory that VET-like systems increase social stratification and socioeconomic inequity [1][2]

>They don't assign you a vocation.

A difference without a distinction IMO. They assign you a track of schooling which will leave you qualified for only a certain subset of jobs.

[0] https://www.oecd-ilibrary.org/education/a-skills-beyond-scho... page 38

[1]: https://link.springer.com/article/10.1186/s40461-016-0033-0

[2]: https://www.semanticscholar.org/paper/The-social-stratificat...


> Nothing necessarily precludes it but does it really happen that often? I couldn't find any data on this.

Does it happen often in the alternative system? You started with saying that you left school at 14, then went to university at 28. No matter the system that's going to be an unusual situation.


You would just spend 1 year taking the requires courses in an adult education school and then getting a 3 year major. Seems pretty equivalent to doing a 4 year American major which had no prerequisites.


> Because it is unconscionable. I dropped out of high-school at 16. I was able to go back and earn a degree in computer science at the age of 28 at a Canadian school. Under the European system where your life path and career trajectory is decided for you at 14 years old, I wouldn't have been able to do this.

Would tracking you differently have perhaps resulted in you being able to graduate high school?


You can just redo the 2 years of high school you missed out on.


> Yes, but many European countries engage in so-called tracking of students, putting them in separate tracks for tradeswork, skilled labor, and university such that the "dumb" ones aren't taking university entrance exams.

Err, this is a positive thing? Personally I found the AP classes and SATs to be pretty useless beyond getting into college. We should just admit anyone who wants to to.

If the point of this is to justify high international test scores, there’s a much better argument there to test either nobody or everyone.


Community colleges admit anyone who wants to go. Students who do well there can then transfer to complete a Bachelor's degree. It's a great option for students who perform poorly on standardized tests.


This doesn’t change the fact that meritocratic admission is fundamentally broken at the other schools.


There are limited resources to provide high-end education. Effective use of those resources is key, and meritocratic admission is far preferable here to "just let everyone in".


> We should just admit anyone who wants to to.

Well, the problem was not only that anyone could enroll in the higher-track classes but teachers and counselors pushed many to take courses they weren't prepared or willing to take. I myself was heavily pressured into taking higher-level Spanish courses than I was capable of, or even willing to take. I was explicitly told I had to because I wouldn't get into any universities without it. They really wouldn't take no for an answer, and I ended my senior year with a 40 / 100 grade, but thankfully had a high enough GPA and enough credits to take the hit without issue.

In almost every class one could expect the population to be bi-modal, you would have the flunkies with no desire or hope to succeed and you would have the students who wanted to be there. This resulted in the teachers often having to meet the lowest-common-denominator.

We had a programming course (Visual Basic) that after the first month they switched to one of those "learn-to-program" visual building-block languages, but then myself and a few others raised hell so they allowed everyone to choose which they wanted to do in the same class of 30 or so. The reality was the counselors pushed the students to take the course, and most figured they would get a free 90 minutes of fucking around on the internet, and the teacher ended up giving the lower-end of students low-but-passing grades.

As it were, I'd had that same teacher several years in a row and ended up as something of his assistant because I was already a competent programmer. They're not able to do anything, it's all a metrics / numbers game.

To be clear, my point isn't about the benefits of tracking as it is a rebuke of a broken system of metrics-appeasing administrators cloaked in the lie that everyone is equally capable of the same things.


> Yes, but many European countries engage in so-called tracking of students, putting them in separate tracks for tradeswork, skilled labor, and university such that the "dumb" ones aren't taking university entrance exams.

Not at all, this is largely false.


Instead of downvoting, provide evidence to the contrary.

In various countries in Europe where I lived there are no laws preventing people from taking university entrance exams.


> Citation? All countries have a mix of wealthy, educated students, working class students, students whose parents are drug addicts, immigrants who don't speak the local language, etc. In my experience the US isn't that different from other wealthy countries I've lived in.

If you look at [0], It looks like the parent is right.

> However, the PISA’s results are drastically over simplified. For instance, as noted in a report by Dr. Martin Carnoy of Stanford and Richard Rothstein of the Economic Policy Institute, American students actually perform better than the much higher ranked Finland in algebra in general, but worse in fractions. More importantly, when you normalize the results between the countries, adjusting for the relative poverty of the students taking the PISA tests, the U.S performs significantly better, ranking 6th in reading and 13th in mathematics, a drastic jump in both categories.

> Dr. Carnoy and Rothstein further note in their report What Do International Tests Really Show About U.S. Student Performance? that when you divide the kids based on family wealth, the actual gap in performance isn’t so stark between any country, with a not insignificant portion of the ultimate ranking of each nation being based on how many impoverished vs. middle class vs. wealthy students are taking the tests.

[0] http://www.todayifoundout.com/index.php/2015/01/kids-get-sum...


> when you normalize the results between the countries, adjusting for the relative poverty of the students taking the PISA tests, the U.S performs significantly better, ranking 6th in reading and 13th in mathematics, a drastic jump in both categories.

We're doing better than that. When you break the results down by race, the US generally ranks second or first. (US whites do better than whites in Europe; US asians do better than asians in Asia, US blacks do better than blacks in the few countries with a large black population that aren't afraid to take these tests in the first place...)


Would you mind providing a citation for that? I've genuinely never heard that.


2012, reading: https://isteve.blogspot.com/2013/12/pisa-reading-scores-by-r...

2012, reading/science/math reported separately: https://isteve.blogspot.com/2013/12/overall-pisa-rankings-in...

2015, average of reading/science/math: http://www.unz.com/isteve/2015-pisa-mean-scores-in-perspecti...

This is not the stuff of an "educational crisis".


The US has a caste-like professional class and an underclass.

You can pretty easily observe the difference by looking at school performance in the x largest suburban districts in your state and comparing it to the x largest urban districts.


Wasn’t Shanghai famous for getting really good PISA scores, that being the richest city in China?


The Shanghai results are quite controversial, not only in terms of accuracy but whether they might skew Chinese results significantly.

Not being in the field (and not really caring much about it TBH) I don't know how to evaluate the controversy or even its legitimacy.


The US has far more ethnic diversity than any country that’s at a remotely similar level economically. If you compare the average scores for Americans by race to countries where that race is the majority the US looks pretty great considering it’s 1/20 of humanity, less than ten countries do better at educating White or Asian populations, one does better with Hispanics and US Blacks outscore the only majority Black country that takes part in PISA by a small amount. You see the same dynamic in the US by state.

http://www.unz.com/isteve/2015-pisa-mean-scores-in-perspecti...


>The US has far more ethnic diversity than any country that’s at a remotely similar level economically.

That sounds like a cherry picked criterion to make a point.

In any case, looking at GDP per capita, Singapore is similar to the US, and also similar in terms of racial/ethnic diversity. Yet Singapore (as a whole) tops your rankings, and the US as a whole is way down in the list.


>The same census also reports that about 74.1% of residents were of Chinese descent, 13.4% of Malay descent, 9.2% of Indian descent, and 3.3% of other (including Eurasian) descent

That's extremely different than the US.


Different in terms of the ethnicities involved, but not in terms of the distribution:

72.4% White, 16.3% Hispanic, 12.6% African American, 4.8% Asian


Honestly, I'd say that a good chunk of that is culture. Singapore has an insane focus of education, with enormous amounts of spending on supplementary education and such.

In the US, the Asian "tiger mom" is seen as an odd peculiarity. That alone speaks volumes.


>Different in terms of the ethnicities involved

That's the entire point though Africans and Hispanics score lower than Asians and Whites, as shown in his link.


Like it or not, different racial groups have significantly differential SAT mean scores. And the US has a growing number of minorities that do not do well on said tests.


> from hard poverty to not being able to afford tutors and SAT prep courses.

I went to a pretty fancy prep school with a lot of rich folks, but I don't think most did any special prep for the SAT. I think we had a class period where they talked about the sorts of questions that appear on the exam, told us about when/where it'd be offered, etc., along with ads for SAT-prep services, but I think a lot of folks ended up deciding it would be best to get a good-night's sleep before walking in on it. (We also debated stuff like if drinking coffee or/and having a chocolate bar beforehand would be helpful.)

Which isn't to downplay the advantages that affluent students enjoy, just, I don't think that SAT prep courses are really the big thing. (Plus I suspect that students who spend a lot of time doing test-specific prep are losing out on their actual education that they'll actually need later on.)


    > I went to a pretty fancy prep school with a lot of rich folks...
I suspect you might not be aware of what a vast number of public schools in the USA are actually like. SAT prep becomes more and more critical for students as they come from increasingly underperforming high schools.

There are schools where ~1/3 of the students just wander the halls aimlessly instead of going to class, where teachers turn over at a rate of 25%/year. In such schools even students who intend to go to college are going to be utterly unprepared for the SAT. These students actually need a prep course.


I'm not sure I follow your point.

I mean, hypothetically, let's say that a student never attends class, just aimlessly wandering the halls (which I'm skeptical about, but let's go with it). Then even if they somehow game the SAT and get into a good college, then what? If they're basically uneducated upon entering college, then it seems like, best-case scenario, they'll get like 4 years of education in college -- making them analogous to a student with a gradeschool education, right?

I mean, this sounds like a smoke-and-mirrors scheme.


I am not saying that the students aimlessly wandering the halls are taking the SAT-- they're in the jail or dead-end-job pipeline. They are NOT going to college.

But there are other students in such schools, however, that are trying to make the best of the situation they're in and these students want to go to college. Sadly, these school is so busy with discipline and attendance issues that it's almost impossible for serious students to get prepared for college.

And yes, there are many public schools that would appear to a visitor as chaos. Students do just roam the hallways if they don't feel like going to class. There can be so many doing this that it looks like they're changing class-- all the time. Not exaggerating.


Don't be skeptical. I know several teachers in the parts of big cities where many classes are just the teacher trying to keep the kids from hurting each other and any lectures aren't really listened to, where teachers are cussed at and assaulted with zero consequences. There are many schools where kids are effectively offered zero education because the environment is too chaotic and violent.


> just aimlessly wandering the halls (which I'm skeptical about, but let's go with it)

For some schools it is a real problem. My girlfriend works at a pretty bad high school and she has tons of students who keep failing their classes because they never show up and the principal literally wanders the halls himself to try and force kids to go back to class.


Same: in 1981 our SAT "prep" was literally two sentences one morning: 1 - there will be little bubbles you need to fill in and you must use a pencil and 2 - some statistical strategy concerning guessing. Everybody took the test only once; I don't remember any efforts to game the system.

But that in itself points out the advantage: the test was basically designed to reflect the level of instruction my school already gave. If you don't get that from your school (or believed you don't) you would have resorted to additional study and the like.

Likewise back then there weren't "AP" classes and the like: if you wanted to take the AP test you could but there were no special arrangements at my school.

I don't know how much things might have changed: my kid's SAT prep in 2017 was simply to purchase the book but not open it. On the other hand there's a whole apparatus of AP infrastructure with special "AP" classes, different scoring for those classes(!!) etc. AN enormous kabuki of infrastructure that as far as I can tell signifies nothing.


> I think we had a class period where they talked about the sorts of questions that appear on the exam, told us about when/where it'd be offered, etc., along with ads for SAT-prep services, but I think a lot of folks ended up deciding it would be best to get a good-night's sleep before walking in on it.

That's actually non-trivial amount of prep, but I personally don't know anybody that took the SAT "cold". Everyone prepped quite a bit - I prepped for months and got mediocre scores consistently ¯\_(ツ)_/¯


Yeah.. prep doesn't seem too effective. That's what they told us at prep school (heh). I mean, they basically just said that we ought to ensure that we have sharpened pencils; adequate batteries for our calculator; a good-night's sleep; and to show up early.

I can appreciate how some folks want to do better, and are willing to work for better scores by investing in test-prep. And I can understand how test-prep providers would encourage them to do so, talking up their own services, as any business would.

But, I don't think it's that big a thing in the end. If anything, I worry that students who spend a lot of time prepping are being cheated out of their time, money, and education.

---

PS: To be clear, students would definitely do well to spend time studying for their future. Just, instead of trying to game the SAT, they can learn something useful, whether that's math, physics, computer programming, philosophy, creative writing, or whatever.


> I personally don't know anybody that took the SAT "cold"

How do you define "cold"? IIRC, my prep consisted of just taking the PSAT. I ended up with a 1370 (700 verbal, 670 math), which was good enough for my purposes.


I also don't recall any particular preparation that I did for the SAT - I may have picked up a second-hand prep book somewhere.

But it was not like it was rocket science to know how one is supposed to take these kinds of tests. I graduated from a very unremarkable rural school system in New England. From 2nd grade on, we took a standardized scantron-type test (the CAT test, if I recall correctly). In 4th and 7th or 8th we took a state-wide standardized test that was more of the same multiple-choice sections, with an additional writing component that was a watered-down version of the now-discontinued writing portion of the SAT. Then I believe that they had us all take the PSAT in 10th or 11th grade, and they also encouraged people to take the ASVAB, as we had a large percentage that were planning on the military anyway.

By the point it was time to take the SATs, all of the strategies and gimmicks that are covered in test prep for the SAT were pretty old hat, and we'd been through that wringer at least a dozen times over by then.


> Usually almost all of these metrics come back to the root cause of economic disparities along the spectrum from hard poverty to not being able to afford tutors and SAT prep courses.

Do you have any resources for how much an SAT score can change from studying? In my experience offering free ACT tutoring at the local YMCA, I wasn't able to improve scores that much. Maybe I'm just a bad tutor. I did pretty well on the SAT when I took it, with little prep.


My experience with standardized testing is that there are some unprepared students who are not familiar with, or fail to use basic strategies like process of elimination, jumping between question and text, etc and these students will get a hugely significant boost, but beyond that progress from tutoring at best equal to self studying. I also did well with little prep but that was because of my highschool experience.


My experience working with outreach programs taught me that we would see a 1 or 2% increase from focused efforts on content mastery in SAT/ACT prep classes. When we focused on specifically the skills needed to do well on 'generic' standardized tests, we saw a 7-12% increase in individual student score.

These tests mostly test how well someone (a) responds to pressure and (b) takes a standardized test. They're not accurate measures of student learning.


This was my experience. In my SAT prep class 15 years ago, the first thing the instructor did was talk about the statistics of guessing and how even if you don’t know any answers at all, you can improve your score dramatically just by eliminating the obviously wrong answers before you guess. Then we went about learning ways to find the obviously wrong answers based on the format of the question that was asked. We also took practice tests, but it was the generic test taking rules that made the difference for me. I did very, very well on the SAT and then later, I applied the same rules to the LSAT and did much less prep and did very well on that too.


I appreciate your anecdote and that lines up with my experience. But a 10% increase is what I had in mind when I said I couldn't increase their scores much.

It doesn't seem to me to support the idea that the SAT tests mostly your (a) and (b). The average SAT at MIT is 1530 with the average at my local state school University of Indiana at 1250. Prep classes for the average UI student increasing their score 10% to 1370 wouldn't even get them to the 25th percentile MIT student.

It seems to me the SAT does capture something beyond simple ability to take a test.


If that were true then teaching how to take the test would have resulted in a larger jump on scores.


I have a 10th grader who is just starting out on his SAT/ACT journey. A combination of Khan Academy[1] and a $10 book is his prep. I'm interested in how much these resources and the deliberate practice that goes along with them will improve his baseline scores.

[1] https://www.khanacademy.org/test-prep/sat

[2] https://www.amazon.com/gp/product/0525567658/ref=ppx_yo_dt_b...


A good example of this was when Mississippi test takers routinely outscored those in New York taking the test. There were two reasons. The first was that New York required all students to take the SAT while in Mississippi only students interested in attending college took the entrance exams. The other factor was that the ACT was much more commonly used by admissions offices in Mississippi and surrounding states. So out of college bound students, the SAT typically was only taken by a subset planning on attending schools further away, which were much more likely to be Ivy League or other types of schools that attract students much more broadly than the average state school.


I think I took the ACT instead of the SAT, as the ACT was the more popular exam of my day.

The only people that took the SAT in addition to the ACT, were:

1. Applicants to "elite" schools that preferred the SAT.

2. Smart kids trying to get a leg up for large state schools.

3. Pretty smart kids trying to get a bump into a higher scholarship bracket.

If your were anything less than an 80th percentile student, it'd be extremely unusual for you to take both the ACT and SAT. This was also at a Catholic school - most parents pushed their kids to do well academically to make tuition worthwhile.


This should have been the most predictable result ever. If only the top 20% of students are taking the test, as used to be the case, then the average score is going to be quite high. If you expand that pool with weaker students, it's going to pull the average down. If you make everyone take the SAT, then you are going to wind up with a lot of really low scores dragging that average down hard.

They run into this all the time with mandated standardized testing. In smaller school districts, just the relative number of special ed students from class to class can create so much variance that year-to-year comparisons are useless.


The official numbers are wildly against you on this. Here are the numbers of testees, total and those scoring 700+ on the math section, for recent years and then in 5-year intervals back to 1999:

    year     takers      math 700+
    2018    2,136,539     202,088
    2017    1,715,481     120,043
    2016    1,637,589     117,067
    2015    1,698,521     121,057
    2014    1,672,395     123,781
    2009    1,530,128      97,296
    2004    1,519,870      94,985
    1999    1,302,903      71,214
(numbers pulled from https://research.collegeboard.org/programs/sat/data/archived )

In 1999, 1.3 million students took the SAT and 5.5% of them scored 700+ on the math section. Just last year, 2.1 million took the SAT and 9.5% scored 700+ on the math section! Or year-over-year, we can see that the marginal 421,058 students taking the SAT in 2018 compared to 2017 earned a marginal 82,045 700+ math scores, which at 19.5% is well over double the rate at which testees earned those scores in 2017. This makes it clear that the additional students are mostly long-overlooked geniuses.

(In fact, what's happening is that the college board is making it much easier to get a high score. A 720 math in 2018 is a much lower score than a 700 math in 2004. Sometime between 2004 and 2009, they stopped even providing the information of how many students scored 750+ on each section; 700+ is the finest resolution they're willing to provide now.)


Yeah that has been used so many times with the "old test questions prove we are getting stupider!" trope circulating.


There are easy statistical methods to still get useable data from something like this. Average is just the average.


Oh, of course. But statistical illiteracy hurts us badly here. Even just distinguishing between median and mean is beyond the abilities of most popular reporting.

As it is, this article is almost worthless. The raw data would be interesting to go through.


I remember seeing this in Texas a few years ago too. SAT scores fell statewide, which made for anti-Texas fodder, but what was lost was the fact that record numbers of black & hispanic students took it for the first time.

https://www.texastribune.org/2012/09/24/texas-sat-scores-dro...


For context, in the US there are 15.3 million students enrolled in high school, 3.7 million of which are expected to graduate this year (i.e. a little under a quarter, as one would expect) [1]. So the 2.2 million graduate test takers mentioned in the WSJ article represent about 60% of the pool of fresh high school graduates.

Also given the surrounding numbers sum to 100%, I'm guessing the author meant to say 56% percent of SAT test takers have parents with a college degree, rather than "56% of students whose parents have a degree took the test". Probably just a careless mistake, but I still find it ironic (or perhaps indicative) that no editors caught it even though it appears a few sentences after "students struggled most in math".

[1] https://nces.ed.gov/fastfacts/display.asp?id=372#PK12_enroll...


Interesting side effect of this:

Before - if 100 people took the test, there were 10 people in the top 10%.

Now - if 1000 people take the test, there are 100 people in the top 10%.

If we assume that most of the first 100 were already college track/high achievers/however you want to describe them, then you just bumped a significant chunk of those people into the top 10%. Obviously, some of the other 900 will also move into that group but - based on the overall average dropping - not a huge number.

Fundamentally, it causes another form of "grade inflation" but for the tests themselves. No wonder students apply to more and more schools. :(


A more interesting statistic would be the (percentage) change in the number of students whose scores indicate that they are college-ready. Increasing that number is, presumably, what these get-more-kids-to-take-the-SAT programs are about.


According to the article that percentage dropped to 45% from 47% last year, but since ~100,000 more students took the exam the absolute number of college-ready graduates rose by ~3000.


If I am reading this right, we added 100k students got 3k new college ready ones? So 3%? Seems like we are grasping at straws here.

Or is college readiness trending down for other reasons?


Thank you, the corresponding Washington Post article (which I have access to) does not seem to have that information.


Given only the above values, how many students took the SAT this year?


I work in higher ed tech, and the state I work in has been having showing this trend for more than a decade, losing 2 points or so each year on average.


It's been going on for over 4 decades. They recentered the scaled score in 1995 to account for the downward shift in mean due to expansion of access.


Related: US average marathon times have fallen in the past 20ish years, as more people run marathons.

It's a good thing.


As I point out elsewhere in the thread, the proportion of new testees getting very high scores is much higher than the proportion that were getting high scores before, so this would be like if US average marathon times had risen (falling times are good, not bad) in the past 20ish years as a lot of professional marathon runners started to participate. It doesn't make any sense given the raw data.


You mean they've increased?


A UK perspective - my son is now 10 and is in the age for taking two tests - an 11+ test and a UK SAT test. They are basically the same thing and roughly analogous I think to the US tests.

The 11+ score may affect which secondary school he attends (some schools in this part of the UK can be selective and so would choose to take only the higher scoring kids. However at 11 the predictive power is less good and there is sufficient competition that the selective cut off gets pushed down. anyway it helps him if he wants to go to a selective school

The SAT - meh. It will affect the school ranking - but has no affect positive or negative for the pupil and as such seems pointless for him to worry about or even take.


No. The SAT in the USA is a test people take in secondary school to get into university. Very few 10-year-olds would be taking it.


Hmm. I took the SAT when I was 11, not sure how common that is.

I think it was pretty commonly used for admission to gifted programs in the 80s when I took it.

When I learned that the PSAT existed, I was curious why that wasn’t used, instead, but... whatever.


Certain gifted programs, eg JHU's CTY would administer it slightly later.


Another interpretation is that young people are becoming dumber given their obsession with smartphones and videogames. The skills I needed to master before the "everything online" era where more mentally challenging and strengthening than the current era.

I would call this the 'reverse Flynn effect' after the sociologist who observed average IQ scores were rising a few points each decade for a century. That had been attributed to universal schooling, sophisticated new electronic media, white collar knowledge workers, etc. But now we have reached a turning point perhaps due mental dependence on wired devices.


(For perspective, I'm still in High School, and have scored pretty well on the SAT)

I think what a lot of people fail to realize is that the SAT being based on Aptitude is intentional. Doing well on the SAT means knowing how to study for the SAT (with an advantage given to those who do well in school). That's what the test is measuring, your ability to study.


yes and no. anecdotally, I did quite well on the SAT and never did a minute of study for it.

it is certainly possible to study for it, but it is also possible to just be naturally very good at standardized testing.


Yes, and those who are naturally good at standardized testing (such as myself) have an advantage on the SAT, which is one of its issues. But as far as College Board is concerned, the actual content is largely irrelevant.

SAT Math does not test number sense or ability to reason with abstsactions, instead how effective you are at applying a specific set of memorized rules.

SAT ERW does test comprehension, but only in a very specific context. The College Board specifically chooses passages that are meant to be boring, so the reader skims them. SAT Reading is mostly about the grit of paying attention to detail.

SAT Writing, of course is about applying specific prescriptivist grammar rulesets (no singular they comes to mind)

Fundamentally, SAT is designed to be a test of attention to detail and studying, not content mastery. Schools will look at transcripts, AP Tests, or SAT Subject Tests to determine that.


Seems like SAT scores skew in the direction of people who are capable of "giving a shit" about a test. If you don't "give a shit" about tests, then you'll likely be at most a C student and a ~1150 (just an example number) on the SAT regardless of your level of intelligence.


Regress to the mean.


Also, like it or not, the SAT does predict pretty well how a person would do in a typical college program.

Maybe too many people are going to college! And/or we need other types of post-high school education.


I used to not understand why the SAT was supposed to be a predictor of how someone would do in a college program, but after checking the schedule of my college Chinese class and studying to memorize the material everyday, I've realized that the SAT is just like any midterm or final for a class, albeit a bit broader in subject. If someone studies for the SAT then they'll probably study for their classes. Then again I think I spent a total of seven different days (not 168 hours, just some time spread across seven non-consecutive-days) studying for the SAT, which is nothing compared to what I have to do now. If only I could go back in time and retake the test with my newfound knowledge.


I sure didn’t study for the SATs, nor did any of my close friends.

Our college educated parents told us to not worry about it.



I am not sure how practical this sort of standardized testing is. I am in the high performance range on most aptitude and general knowledge tests, but I did horribly on the SAT. This didn't slow me down from graduating college and becoming a senior software developer. Likewise, I have seen people who scored supremely high on the SAT that simply cannot perform in the real world.


Let's not forget that many "great schools" are really cram academies for the SAT.


How do you cram for the SAT?

I've always understood the ACT as more about content and the SAT as more about aptitude, which means studying for the SAT is less useful than studying for the ACT.

You can teach test-taking strategies but those hardly require 2.5 years.

I agree some schools provide a better learning environment than others, and also engage in selective admissions either explicitly or via the socioeconomic status of the students' parents. Both of those could cause standardized test scores to improve but I wouldn't call it cramming.


> more about aptitude

Though this was some years ago, I took both ACT and SAT and found them almost identical.

I know SAT at least has added an essay section and otherwise moved away from multiple choice but still, to imagine you can't cram for it seems absurd.

In middle 70s, Federal Trade Commission investigated Stanley H. Kaplan, the big test prep company, over advertising that they could raise test scores. Well, investigation showed that yes, Kaplan could raise your SAT scores.

Like the College Board claims they can give approximately the same test for decades but somehow you can't study for it? Something doesn't seem right.


Source on this?

the SAT of the 70s were much more of an IQ test than now. I'd be surprised if Kaplan prep could raise more than 20 or 30 points on average.

Also, having used Kaplan's prep material... They're pretty bad. But that was in the early 2010s for myself so they might have fallen in quality.


Friend worked for Kaplan during the time period of the investigation.

IQ - you know IQ tests are renormed to reflect education levels, and raw scores mapped to give the bell curve?

So an IQ of 100 in 1900 would end up about 70 today. 70 oddly enough the number used to "prove" that people in the less developed world are inferior.


SAT is Math and English only, ACT includes Science.

I took both in the early 1990s, 1430 SAT and 32 on the ACT, didn't study for either with exam prep.

For high school coursework I had AP Biology, Chemistry, Physics, and Calculus so from a content perspective I took all the courses you'd need to know things for both tests.

My understanding on the 32/36 and 1430/1600 is those are equivalent scores for the exams. I'm just one anecdotal data point so take it for what it's worth.

P.S. Studying for SAT felt like it would have had a bigger impact on my personal score. The analogy vocabulary section I did very poorly on (480/800). In my day they showed you your English score in two sections, vocab and reading comprehension (which I scored 800/800). Those 2 combined / 2 + math gave you your total.


Is it all because more students take the test? Reports of falling IQ scores in many developed countries have been pretty well documented. Perhaps it is starting to happen in the US also.


Could you please provide references to these "pretty well documented" studies?

Thanks.


I believe they're referring to the reversal of the Flynn effect[0].

[0] https://en.wikipedia.org/wiki/Flynn_effect#Possible_end_of_p...


https://www.gwern.net/docs/iq/2016-dutton.pdf

https://www.nbcnews.com/think/opinion/iq-rates-are-dropping-...

This is fairly well documented. Average IQ scores in France have been declining a couple of points per decade if I recall correctly...

I believe mass immigration from low IQ countries and dysgenics are often sourced as two of the potential reasons. But I have no idea how true that is.


I believe IQ scores are rising still for lower scores, but falling for the top 1-2%?


The very most recent US Flynn effect study finds the opposite pattern, actually (falling for low end, rising for high end): https://www.gwern.net/docs/iq/2019-platt.pdf

Personally, I don't buy it. The Flynn effect is large enough, and the way the normal distribution works with tail effects, that if there were any Flynn effect in general (much less focused on the high end), that you should see dramatic effects like several-fold increases in the number of perfect SAT scores. You don't see anything remotely like that - even despite loads of countervailing factors like increases in SAT access and prep!


Intelligence isn’t normally distributed though.

E: And even if it was, and was capable of “shifting” left or right, which is a nonsensical notion about populations, a shift of a few IQ points wouldn’t have a dramatic many-fold increase in perfect scores.


You're right, intelligence, like anything else (eg height), is not exactly normally distributed: it has fatter tails than a normal distribution - so a shift in the mean would increase the density at a high threshold even more than you would calculate based on a normal.


No, a fat tail would do just the opposite. A fat tail is where more points are far from the threshold.

But it's a moot point because an increase of 2 points among people in the 99th percentile implies little about those in the 99.99 percenteth quantile.

Some description of a fat tail from 60-99 percentile doesn't say much about the top one percent. At a certain point, the dating market tops out.


No, it would. Simulate it at a threshold like 1600 scores. The steeper increase in that part of the range yields the same phenomenon as normals.

> (Also, intelligence/aptitude/performance scales like IQ, SAT, or ASVAB scores are arbitrarily numbered, so I'm don't know what a "fat tail" means in that respect. Maybe there's a standard definition that isn't completely arbitrary.)

Er, you can look at plenty of cardinal-scaled things like vocabulary or digit span. Doesn't change the non-normality at the tails.


I believe it is the same case with height.

https://www.johndcook.com/blog/2008/07/20/why-heights-are-no...

It's close to a normal distribution one or even two deviations away from the mean, but real-life has way more samples appearing +5 deviations away from the mean. I haven't really looked into it ever, but this could be because of a positive skew versus a distribution with purely fat tails?


The difference is, it's easy to make a long-boned freak or a blithering idiot. But there's no hormone (I know of) that makes you a genius.

(Also, intelligence/aptitude/performance scales like IQ, SAT, or ASVAB scores are arbitrarily numbered, so I'm don't know what a "fat tail" means in that respect. Maybe there's a standard definition that isn't completely arbitrary.)


When I took the test a perfect score was 1600. Without getting into detail, the day of the test I was late and didnt get enough sleep. I wound up getting an 1160 and I never studied. Though, my grades were shit so the good unis rejected me outright.

The interesting part was one of the principals A/A+ list students was next to me at the college fair and hung his head in shame when they asked him about his SAT score and he replied with 900. A fucked up D student with an 1160 and a studious A student with a 900 both turned down. Go figure.


I was confused until I realized that you think 1160 was a good score.


Depends on your definition of "good". The year I took the SAT (1997) the average score was 1017.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: