site banner
PaperclipPerfector  4mo ago (text post)   29060 thread views

Culture War Roundup for the week of December 18, 2023

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

6
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

1460 SAT and rejected at Cornell has been trending on Twitter the last few days.

https://twitter.com/maiab/status/1736766407348814091?s=46&t=aQ6ajj220jubjU7-o3SuWQ

A lot of the takes were about him being rejected because he is white. The thing I find interesting is the condensing of the top 1-5% of scores into a smaller score range over time. My guess since the score differences look smaller it lets schools select more for other characteristics rather than pure mental horsepower. Getting a perfect score today or something that looks similar 1550 plus will not differentiate people as much.

Elon Musks apparently had a 1400 SAT. Bill Gates a 1590. Obviously they are both smart but I feel fairly confident Bill Gates is significantly higher pure IQ. With the way normal distributions operate I feel confident saying there is a big intelligence difference between the two but on the current system Musks would probably get 1580 and Gates 1600.

Digging thru SAT history there have been a few key years where the test had significant changes.

1993/1995 - some test changes but the big thing was a recentering to get scores back to about 1000 from 900. Before this update a median score at HYPS would have been 1370-1400 area. Bill Gates 1590 would have really stood out and guaranteed alone admittance to Harvard.

2005 - attempts to move the test closer to high school curriculum and eliminated analogies and quant comparisons. My guess is this made the test less of a pure intelligence test and closed gaps between highest performers and mid range.

2016 - more I guess dumbing down and trying to make the test more like what they did in high school. Multiple choice questions went from 5 options to 4 options and wrong answers no longer carried a penalty. This would make educated guessing far better.

Here is the current percentile for different scores.

https://blog.prepscholar.com/sat-percentiles-and-score-rankings

1500 is now solidly top 2%. 1450 is top 4%.

Here is the data from 2003

https://blog.prepscholar.com/sat-historical-percentiles-for-2005-2004-2003

1490-1600 was solidly differentiating between the top 1%.

I believe the new scoring significantly hurts the outliers at standing out from the test. And likely hurts the highest performing white, Asian, and Jewish males at getting into the most selective schools since the difference between a 1530 and 1600 SAT score just doesn’t seem that big statistically. It feels to me that studying for the new exam and learning test taking skills are more important today. Perhaps, you think this isn’t a big deal that the raw mental abilities of the top 1.2% and .3% of the population isn’t important and allowing schools to select more on other criteria is more important. My opinion for the very top programs finding the Bill Gates level intelligence matters. Men also have different intelligence bell curves (more people on the extremes) therefore on net I believe it hurts males.

I am also curious how someone who is really good at math could stand out in today’s environment. The SAT and a few good AP math scores wouldn’t seem to be enough. Do you need to have the opportunity to compete in high-end math tournaments?

Personally, the new testing I believe would have significantly effected my life. Coming from a lower class white family being able to crush the SAT gave me a way to stand out for a relatively cheap costs.

I am seeing a median SAT score of 1520 at Harvard and a median of 1440 at UMICH. My guess is back in the day that gap was much higher.

Too many students tie their sense of worth to what college they get admitted to. As others have said, a 1460 SAT score, especially nowadays, isn't something you'd expect SHOULD get you into an Ivy League or even the next-tier down colleges. It's impressive, but not good enough. There are plenty of good schools the next tier down that still have extremely high brand recognition and provide a good education.

There's certainly something to be said about the networking access you get and brand recognition if you go to a school like Harvard, but a lot of these students go from being the top or smartest kid in their town/school to being mediocre or below average. This is a huge blow to their ego and while it certainly is a humbling and valuable lesson a lot of these students end up switching from a difficult STEM track to something more manageable such as liberal arts. I think this is a net loss to humanity, while I can acknowledge there is some value to the liberal arts the world needs more doctors/scientists/engineers instead of another person writing papers nobody cares about. 82 percent of papers in humanities don't get a single citation 5 years after they are published. (I was unable to find a source with more recent data, but my gut feeling is that the work coming out of the humanities now is even worse on average than they were 20 years ago).

You can see this happen on a statistical level with students admitted via affirmative action. A decently smart black kid who's always wanted to be a scientist gets into Harvard, falls into the bottom 10% of students, and since he's human gets discouraged and switches majors to something else instead, where he has a much easier time because the coursework is not as difficult. The black community lost a future scientist or doctor to affirmative action. If that kid went to some state university instead, he may have graduated top of his class and proceeded to produce valuable work for humanity as a scientist or engineer.

Maclolm Gladwell makes a similar argument in his book David and Goliath. He points out how the top third of students, no matter the university, around 45-55% get a STEM degree, while the bottom third only 15-20% get a STEM degree. The top third of students at a place like Hartwick is equivalent in average SAT scores to the bottom third of students at a place like Harvard. It's the bottom third of Harvard students switching majors, even though they are likely as smart or smarter as the top third of students at Hartwick. In addition, Gladwell argues that it's better to be a big fish in a small pond, and points out how the top students at mid-tier universities publish papers at a higher rate than middle-tier students at elite universities. In terms of their SAT scores and academics, they are equivalent and Harvard should have a superior education, so you should expect the middle-tier Harvard students to perform better, but in reality, it's the opposite. Essentially, the relative position in their local environment mattered more than the absolute position nationwide. (Here is a link to an 8-minute video where he also talks about this idea)

That being said, for the individual, it's still probably better for their career to go to a school like Harvard and be a middle-tier or bottom-tier student than be a top student at some state university. It's a net loss for humanity on average, but a huge opportunity for the individual. If you can get past the ego loss and instead grow as a person just accepting that you're mediocre amongst the geniuses you'll gain a lot from an Ivy League environment.

82 percent of papers in humanities don't get a single citation 5 years after they are published.

Are we talking about regular papers published in journals, conferences and such and not some university internal reports?

Even my (engineering) masters thesis got 10 citations in Google scholar in the first 5 years and it's a common saying here that likely nobody beyond your professor will ever read your masters thesis. Just how pointless are those humanities papers if they get no citations at all?

Are we talking about regular papers published in journals, conferences and such and not some university internal reports?

Yes, published papers specifically. The exact method is explained in the method section of the source:

Data for this paper are drawn from Thomson Scientific’s Web of Science, which comprises the Science Citation Index Expanded (SCIE), Social Sciences Citation Index (SSCI) and Arts and Humanities Citation Index (AHCI), for the 1900–2007 period. Each journal was classified based on the taxonomy used by the U.S. National Science Foundation. For the Humanities, the NSF classification was completed using in-house classification results. NSF subject headings where grouped into four broad categories: natural sciences and engineering (NSE), medical fields (MED), social sciences (SS), and the humanities (HUM). Data for NSE and MED start in 1900, data for the SS start in 1956 and for HUM in 1975.

The matching of article citations was made using Thomson’s reference identifier provided with the data, as well as using the author, publication year, volume number and page numbers. Only citations received by articles, notes and review articles were included in the study and first author self-citations were excluded.

Also had excluded online data at the time:

The data reported in this paper do not take into account the “online availability” variable.

Note that this is data from studies published in the early 2000s, i haven't found a more recent analysis, but I find that things that studies/analyses that can put leftist doctrine and ideology into question don't get produced out of the universities and are quite rare. This leads me to assume that analysis would prove the numbers are even worse, as I imagine the Humanities sectors would be incentivized to disprove this statistic to justify their existence in the universities, and the fact that I could not find a detailed analytic reputation from within the last 16 years implies the truth of the scenario.

The quality of the majority of papers being produced is extremely questionable and the methodology has been in question. Back in 2018 three professors deliberately created 20 fake studies with the most outlandish claims, of which "seven of their articles had been accepted for publication by ostensibly serious peer-reviewed journals. Seven more were still going through various stages of the review process. Only six had been rejected." A similar stunt was performed in 1996, known as the Sokal Hoax. It is a fact that people can submit fake, bullshit papers into the humanities and have them published for the world to see. It's also a fact that nobody is reading these papers.

What was the content of these bullshit studies? Sokal submitted his paper proposing that quantum gravity is a social and linguistic construct. According to the Atlantic article I linked above, one of the published papers from the more recent 2018 example argued that "western astrology" was sexist and imperialist, and that physics departments should study feminist astrology and practice interpretative dance. Another asked if “dogs suffer oppression based upon (perceived) gender?" Even another argued that "men who masturbate while thinking about a woman without her consent are perpetrators of sexual violence." These were the ones that got published into supposedly reputable journals that publish works from professors from distinguished universities like UCLA, Penn State, etc. (There is a section at the bottom of the Atlantic article that provides some criticism/counterargument to what Sokal and the three professors are trying to prove about the state of Academia, for those interested, go look at the article).

The question then is why is this allowed to happen in the humanities? There is the common explanation that one must publish or perish in order to have a successful academic career, which drives people to publish whatever they can to succeed in the Academia rat race.

Jordan Peterson provided another explanation on the humanities papers fiasco.

The question is, why do these papers get published since no one reads them and they have nothing to offer? And the answer to that is very straightforward. The journals are extremely expensive. Way more expensive than they should be. So just to buy a single paper online for the ordinary person is like $40 which is more than a hardcover book. That's just to download the pdf. And so the journal itself - libraries are full of them - are very expensive and the subscriptions are very expensive. And so what happens is the professors pressure the university libraries to buy the journals, and the library funds the publisher, and so the publishers will publish anything - Routledge is a good example of that much to my chagrin because they published my first book - but and they used to be a great publishing house but they'll publish damn near anything and the reason for that is that the libraries are forced to pay radially inflated prices for the publications that no one ever reads and so people write, to publish in journals that libraries have to purchase at inflated prices, to produce knowledge that no one will ever read and that's the little scandal that plagues the humanities. I think it characterizes the humanities more than plagues them.

It seems like Jordan Peterson is arguing the humanities in the universities have either set up or taken advantage of a system that allows financial gain for the professors in the humanities so there is no incentive to publish good studies. It's possible the money generated from this system can be used to justify the existence of these humanities departments to the university. But essentially Jordan Peterson is saying the humanities are a scam.

I think the biggest problems with the SAT are both the range compression at the upper end, as well as the reduction in g loading as you discussed.

In contrast, a similarly equivalent exam in India, the NEET, is designed explicitly to avoid those problems. It is difficult (though the claims that some questions can stump Western physics or engineering majors are overblown), but more importantly, the sheer number of questions and the difficulty ensures that it's basically impossible to get a perfect score. You have to be both very good at the subjects (all STEM related), and also excellent at time management. That implies a far longer tail of results, which makes it easier to tease out "mere" 99th percentile nerds from the 99.999th percentile geniuses, and not even the latter can hope to get a perfect score.

Which is as it should be, I have a jaundiced opinion on extracurriculars being defacto mandatory for the purposes of college entry, it's largely a waste of time that both biases for the rich and wealthy (who on top of being genetically smarter) are also capable of gaming those systems by sending their kids off to a nice summer of helping dig wells in Africa or similar bullshit.

On the other hand, a smart but poor student in India is on a much more level playing field. All they have to do is focus on making it in the NEET, and everything else becomes moot. This laudable system is of course undermined by our own AA, but I still consider it an improvement.

That implies a far fatter tail of results

Longer, not fatter.

Thanks, fixed.

I will take a stand here: if they are under the impression that a 1460 SAT on its own is impressive enough that it is notable that it did not get them into Cornell, they are not an ivy league caliber student. Just flatly, they do not understand the system.

-- A 1460 SAT isn't really that impressive. It's just...not that big a deal. Even looking at the medians at schools isn't enough, because most of the students at those schools will have a whole pile of other stuff in their resume. Good High School GPA, good extracurriculars, good essays. Any given student might have lacking extracurriculars, a weak GPA, or have written a meandering poorly reasoned essay about how superior he thinks he is to the hoi polloi. He might have put down Stormfront Juniors as his extracurricular and written his essay about his admiration for Rudolf Hess. You just don't know.

-- Admissions are pretty random anyway. Any individual student getting rejected from any individual school isn't notable. At all. Personal story: I applied to all the T14 law schools. I only got into one, waitlisted at the rest. That one offered me a full tuition scholarship. Which makes no sense, because I didn't even get into the schools they were trying to buy me out of. My point being not only was my admissions result random, the admissions team at my school (who presumably know a lot about that kind of thing) didn't expect that result and tried to bribe me not to go to the schools that didn't admit me. Further, HYS all waitlisted me, effectively indicating that I was marginal as a candidate but on balance I was "good enough" for HYS, I was of the caliber of student they were looking for. Georgetown flat rejected me! Georgetown! You never know where you will or won't be admitted on an individual basis, at best it's a probability.

-- Not knowing the above indicates to me that the people involved aren't plugged into the gunner universe of students who put together Ivy League resumes in high school, and therefore probably didn't put together an ivy league caliber resume, and therefore didn't "deserve" to get in. Whether that is the system we want is irrelevant, it's the system we have. It's not about being white, it's about not being a gunner.

I didn’t apply to all T14 but applied to 12/14. Was accepted at one; waitlisted at pretty much all of them. It is to a certain extent yield protection. I focused on the Chicago - Columbia - NYU tier (or at least the tier at the time — I understand USN heavily changed their rankings recently) to get off the waitlist as those appeared to be reasonably the most likely schools I could get into given my grades / LSAT. You generally have a good shot of getting off waitlists if you show you are really interested in that school and ultimately I had success with that.

This is what you would expect. Let's assume among qualified applicants (qualified as in good grades, scores, etc.) the odds of getting in is 10%. So if you apply to 14, you have a 77% chance of getting into at least one.

It's what you expect if the result is essentially random, or contains a random element.

That's not how most people view it, most assume that it's like lifting weights. If one can press 225 there's a 100% chance one can press 205, and a near 100% chance one can squat 225. People broadly view college admin that way, if you get into 15 you definitely get into 20 etc.

College admissions are really more like dating: you might be hot or you might not be, and that helps your odds, but you can never look at a girl and say "x guy should definitely fall in love with you." You can say it's odd if a hot girl can't get any dates at all, but even then it often has something to do with her. If many hot girls can't get dates, you can say something about the system, but nothing about individual girls or boys.

IMO I am not sure how much the kid in question blew this up. I think someone else just claimed it and it got trending. So he may have known he was a reach.

My primary reaction is probably not exactly fair to the kid, but is still the first thing I think of - who gives a fuck? Seriously, how did getting into Cornell become such an important part of this kid's identity that it's upsetting to fail? Cornell is, of course, a very good school and people that go there are apt to go onto excellent careers, but the same is largely true for whatever slightly lower rung school he gets into. Nothing all that important is lost, his college future is still on the table, and the prospects after that remain bright. His SAT was good, but nothing special, and if it turns out that a black kid with a slightly lower score got in, I just don't really see the gigantic fuss. I don't like affirmative-action style policies, particularly in genuine tournament jobs, but I just can't get myself all that worked up that a bright, but unspectacular kid will wind up going to Pittsburgh instead of Cornell.

Cornell is, of course, a very good school and people that go there are apt to go onto excellent careers,

you answered the question. These schools are highly competitive for a reason.

The difference between Cornell and a slightly lower rank is just trivial. No one's poor because they got stuck going to UNC-Chapel Hill instead.

Eh, Cornell is ok. Do you know what Cornell Engineering students and MIT students have in common?

They both applied to MIT.

Wow, just wow. I thought promoting hate upon vulnerable groups was against the rules here.

I am confused at how the above comment is promoting hate upon vulnerable groups and would like to see this claim elaborated on.

I think the "vulnerable group" comment is referring to Cornell students' propensity to jump into the nearest gorge when undergoing stress.

Are you perhaps a Cornell student?

I don't give out details like that so I'll answer no.

I think it was just a joke about calling Cornell Engineering students so dumb that they're a vulnerable group. Obviously a joke because they're of course still quite smart, just not as smart as other top programs like MIT engineering.

What if a black kid with say a 1100 gets in? How big of a difference in score does it have to be for the white kid to complain?

Seems obviously stupid, like a waste of time to admit the not-very-bright kid, but I just don't care very much. Life isn't going to be fair in every individual example, but it's fair enough that I'm not really going to get wound up that a 1460 kid wound up one rung lower than they "should" be.

Honestly does that even hurt the white kid? White kid doesn’t get in over black kid. He then goes to next highest school on lists. Employer needing to find talent isn’t going to hire black 1100 SAT so the employer then hires at next school down the list they wouldn’t have gone to before. Kid ends up in same job.

The only thing is sort of effects is the excess hiring at Fang. He would have gotten the 250k a year at Facebook to do a light work versus the black kid.

Employer needing to find talent isn’t going to hire black 1100 SAT so the employer then hires at next school down the list they wouldn’t have gone to before.

How would the employer do this? Employers don't typically ask for your SAT score, and trying to do may result in a disparate impact claim.

Ideally grades but ivies have equalized those to so everyone gets a 3.8.

That does open the ability of somewhere like Georgia Tech to boost their prestige. Grade hard so employers can differentiate between candidates and tell the difference between candidates.

On a related note, has anyone else noticed how weird the scoring systems on graduate level admissions tests are? The GRE, for example, has 2 sections verbal and quantitative each section scores out of 170, except the minimum possible score is 130. So the score range is only 40 wide. Why didn't they just make it 0-40? Why 130-170? The MCAT is even weirder, each section scores from 118-132. Why not 0-14 or 1-15?

My hypothesis is that adding a constant to everyone's scores makes the scores appear artificially closer together which reduces the perceived unfairness of affirmative action admissions. I wonder if we'll see a reversal of this trend now that the Supreme Court has come down on AA.

Pretty sure the tests were scored like this decades before people cared about affirmative action.

If they're trying to artificially boost anyone it's the people who pay them for practice tests and tutoring.

But more likely it's just some obscure thing about how the scoring works. It's definitely not just a number of points for each question, some computerized versions have adaptive questions based on how well you're doing, etc.

The MCAT sections are balanced to add up to 500, which is just a nice round number I guess.

But more likely it's just some obscure thing about how the scoring works. It's definitely not just a number of points for each question, some computerized versions have adaptive questions based on how well you're doing, etc.

This doesn't make sense. It doesn't matter what the score curve is like. You can always transform a scoring system with a 130 min score and 170 max score to one with a 0 min score and 40 max score and the same curve by just subtracting 130 from each score.

You can literally make the numbers do that, yes, but they won't necessarily mean the same thing.

40 is twice as much 20. 170 is not twice as much as 150.

A scale may cut off the tails in either direction after enough standard deviations because it is not able to accurately measure those extremes, or because the testmaker doesn't have any use for measures beyond those limits. In that case, cutting off your scale at 130 properly implies that there's lots of lower numbers you could get but are not measuring; anchoring the bottom of your scale to 0 improperly implies that this is the lower bound of possible performance.

Etc.

Again, I don't know how these scales are designed or what they're supposed to mean. But it's very possible that relationships like these (probably not these specific ones, but other things that do not survive translation) may be intended in their scale.

I feel like people underestimate how small Ivy League universities are compared to how many people are out there with high standardized test scores.

Here are the seven Ivy League universities with their 2023 incoming class undergraduate numbers (plus MIT for fun):

  • Harvard: 1966

  • Yale: 1554

  • Princeton: 1782

  • Columbia: 1464

  • Cornell: 3218

  • Brown: 1730

  • University of Pennsylvania: 2420

  • Dartmouth: 1209

  • MIT: 1092

That's 16k total student admitted across all those universities. According to the college board, 1.9 million students in the class of 2023 had taken the SAT. That means there are 19k students with scores in the top 1% on the SAT. Getting a score above 1400 puts you in the 93rd percentile according to the college board's statistics for 2023, so 133k people. Even if Ivy League universities admitted students solely on SAT test score this guy would be nowhere close. Indeed, you could staff every incoming Ivy League class (and then some) with students who had a score in top 1%.

This is why it's been expanded to the top-20 or the Ivy+ . These are schools that confer almost similar career prospects and prestige as top schools (in some cases, like University of Chicago for economics are better). Ivy League admissions has not kept up with population growth either.

Why did the US end up with this system? In Canada, there is no equivalent to the SAT. To apply to university, you usually just submit a high school transcript.

If you want a full scholarship, that's another story, but you will usually get something just for having good grades and the tuition is less than half what it would be at a state school in the US.

We also don't have anything like the Ivy league. Our top universities have huge student bodies, and the gap between them and the lower ranked universities is not as big. The university of Toronto has 76,000 undergraduate students. Nor do we have the same problem with grade inflation. U of T notoriously grades its undergrads in a curve, which is not popular for students who all excelled academically in high school. Other schools, such as my alma mater don't do this, but it's common for courses to have C averages (which are stated on the transcript). Does it have something to do with the US having private universities?

The result is that, unlike the US, the UK, China, and probably many other countries, it's not so difficult to get into a good university, but it's hard to do well once you're in, and depending on the program, hard to even pass. But in these other countries where it's really hard to get into the top universities, once you're in, you don't have to work very hard from what I've heard.

I assume it has a lot to do with the private universities, I mean Berkeley is basically a public ivy and they have 45k students.

The US has 50 different education systems and each of those systems is subject to 3,200 systems of local oversight and quality control. That means a diploma means very, very different things across the 25,000 different high schools in the nation. A national university can never wrap enough factors to adjust for what a given class rank means from each school.

A national standardized test provides a sanity check/talent screen.

We have state-wide standardized tests, and almost every high school's average score is public knowledge. Graduation rates are also typically a matter of public record. So, it should be pretty easy for a college to know how impressive a diploma (or specific rank) is from a given school.

No idea if college admissions people do this ¯_(ツ)_/¯

Edit: Idk why you think a national test would add much more value over state tests. There are plenty of proxies you could use to compute average state-smartness

school_smartness = avg_state_smartness + school_smartness_on_state_tests

And this is feasible for a country of 40 million but not for one of 330 million? Because I think universities in Canada do exactly that. They adjust grades based on the specific high school. A standardized test would help with this. We even have provincial standardized math exams that every grade 12 math student takes, but I've never heard of a university asking for students' results on this test.

Canada's universities draw far more provincial students than similar US universities. McGill gets 47% of it's student body from Quebec where the Frasier institute ranks 468 high schools (with more than 10 students).

Harvard has only 15% of it's student body from any state and they pull from almost all states.

If Harvard and other top US universities were mostly regional schools, a national assessment would likely be unnecessary in the US, because the admissions officers would know the 2500 schools in their region well and they'd only take top students from other regions, but because most US universities are trying to pull the best student body they can from as broad a geographic brush as they possibly can a national standardized test makes things far easier.

Why did the US end up with this system?

As per usual, capitalism. These are all private companies with high profits, including entire ancillary markets for practice tests and tutoring and cram schools and etc.

Why did the US end up with this system? In Canada, there is no equivalent to the SAT. To apply to university, you usually just submit a high school transcript.

It helps that the stakes tend to be much lower. Lower paying jobs, less winner-take-all, fewer billion-dollar tech start-ups, no hedge funds, etc. Reduce the returns to elite education and you consequently reduce the competition too. More competition means you need more testing to differentiate applicants. Overseas, there is a lot of competition despite the economies being weaker than the US, but it's implied that top-scorers who get admitted to top universities, like in India, China, etc. will be able to land good-paying jobs in the US.

I’ve been thinking about a related question.

Germany certainly has universities of varying prestige, but they don’t vary nearly as much as the gap between the average mediocre college and Stanford, and the best at undergraduate level are also often very large. In several European countries (Italy is another example) course entry thresholds are pretty modest, but the first year gauntlet is extremely hard.

But in all of them - even where there are large gaps in prestige, like in the UK - the boarding school feel of most elite American colleges just isn’t a thing. Even at Oxford, as I understand it. The whole nature of the private liberal arts school is designed to increase the value of elite colleges. You could go to Cambridge or Heidelberg and never encounter the ruling class of that country except maybe very vaguely in passing. Classes are often very large, social circles small, and the few hours you spend a week in education are dwarfed by the rest of life in which all the usual divisions exist. There may be tutorials but hours per week are still much lower than in American colleges. The few elite US colleges that are actual ‘city universities’ (Columbia is probably the best example) are also somewhat like this.

If you go to Harvard or Stanford at undergrad you’re going to meet the kids of the American ruling class. Yeah you’re not going to be best buddies and people still stick to their own kind but it’s an order of magnitude more than in Europe.

You could go to Cambridge or Heidelberg and never encounter the ruling class of that country except maybe very vaguely in passing.

Can confirm. I met a shit ton of people far smarter than I could ever be, but the vast vast majority of them were less "elite" than me, in fact, because of my schooling background and family station back home, I was the "child of elites" person in most of my groups (fat load of good it did me though).

I met plenty of people who had studied at Eton and Harrow and Westminster etc. but even they by and large behaved little different from ordinary people (and most of them were sons and daughters of parents who had saved up to give their child a good education rather than anything more); sure the conservative association held their yearly dinner in a grand room at the houses of parliment (with price tag to boot) but the afterparty was a trip to the nightclub where everyone would slum it out with the common populace of London.

To be honest with you I was negatively surprised by how vulgar (used in the original sense of the word) all of it seemed to be, when I matriculated I expected things to be a lot more rarified than they actually were once you stripped away the institutional traditions.

The core problem with the SAT is that it’s much more common to get high scores on the math section than the verbal section.

780-800 on math is like 97-99th percentile, some places online suggest you need a perfect 800 for 99th percentile. Meanwhile you can hit like 755 or 760 on verbal and that’s 99th percentile for that section. For the most elite colleges, this essentially means that verbal scores matter much more at the ultra high end, and colleges know this. Someone with a 1500 composite score is almost always just another kid with a perfect or near perfect math section and good verbal. Someone with a 1600 is someone with truly exceptional verbal IQ.

The problem lies in inherent issues with spatial IQ as measured by math questions. The SAT math section can’t have math that every high schooler doesn’t know for reasons of fairness. This aside, even adding harder math questions would measure math knowledge as much as pure IQ (a lot of AP math is methods anyway). Raven’s matrices and literal shape rotation problems solve this - the idea is immediately obvious and they can scale in difficulty to pretty insane levels - but that would drop the act about the SAT not being a pure IQ test.

(If you believe theories of group-level verbal tilt, this arguably gives Jews a minor ‘unfair’ advantage for elite college admissions because smart Jews, per that theory, would be more likely to have great verbal and merely good quant scores than other groups with more even ability; it’s pretty unproven though afaik).

That's a good point about the math section being easier, because it definitely is (at least for the super high levels). But isn't the verbal section skewed towards native speakers with native-speaker parents? how the heck is someone who immigrated as a teenager, with non-English speaking parents, ever supposed to do well on the verbal section no matter how smart he is? It's not like they offer a Chinese language version of the SAT. (there is the AP Chinese test, which is something I guess)

Why is the verbal section so much more heavily g-loaded? Does it contain logic puzzles or something? I have never taken an sat (and took the act decades ago)

It's just that they chose easier math problems.

Some of the verbal sections are relatively difficult interpretive questions, with subtle reasons why one of the answers was right, and the other wrong.

I did about equally well between math and verbal on both the SAT and PSAT, but verbal always felt like it had several questions that I was not certain between two of the choices and going with the answer that felt slightly more likely than the other, whereas the math I always knew what I was doing, had no problems working anything out and just occasionally made stupid mistakes.

The SAT has significant clipping, particularly for math (though math scores and verbal scores are correlated; scaling the verbal score would give you a meaningful signal of math capabilities >800). You could add harder questions and get a more meaningful metric, and these questions aren't particularly difficult to construct, even ignoring exposure to advanced math topics in school. The issue is that elite universities don't want a bunch of exceptional shape rotators: they want to provide a reproduction ground for the elite, who need to be smart but not potential Von Neumanns.

There is no problem with the SATs, institutionally: they give universities what they want.

A lot of this feels like the idea that the left holds all the institutional power. They recreated the SAT to make it hard for math people to stand out but verbal to be the big separator. It also seems like the new test favors females versus boys. The changes overall seem to be better for blue tribe versus red tribe. It may only be a small shift but it’s another advantage they have built into the system.

I understand that its not straightforward to write difficult math questions that do not require any advanced knowledge but with how easy the SAT/ACT math sections are it feels like they're not even trying. The AMC/AIME contain multiple questions each year that require no special knowledge and are harder than the hardest question on the SAT. Here's one nice example https://artofproblemsolving.com/wiki/index.php/1985_AIME_Problems/Problem_6 solving it requires no knowledge beyond area of a triangle = 1/2 base * height and basic equation solving but I bet many 800 math scorers would still struggle, especially with time pressure.

Questions with combinatorics and counting are highly g loaded because you need to know the 'trick' or shortcut , which involves certain symmetries or patterns, instead of brute forcing it , but at the same time the actual math is not conceptually advanced.

The problem lies in inherent issues with spatial IQ as measured by math questions. The SAT math section can’t have math that every high schooler doesn’t know for reasons of fairness.

For what it's worth, I always chalked the far higher variance in the verbal section to the fact that math questions (almost always) have exactly one correct answer, while the verbal section plays a lot more into social expectations of how you interpret texts and questions, and doesn't quantify well.

I scored pretty well, but not 800, on verbal (and like many, higher on math) and felt that many of the multiple-choice questions had ambiguous answers in which at least two could be, arguably -- and I probably could have written a few paragraphs of justification -- interpreted as correct. I'm not going to claim that I deserved a higher score, but at some level it felt like the meta-game was determining which answer a College Board test writer would interpret as correct.

at some level it felt like the meta-game was determining which answer a College Board test writer would interpret as correct.

Absolutely; this was the bane of my existance when I was in my standardised testing years. With the quant sections I knew how well I had done pretty much as soon as I had checked over my answers during the test itself, with the verbal section there was always a lot more inherent variance based on which option was the supposedly "correct" one.

That just means you’re not as good at verbal though. I very consistently do extremely well on verbal sections and there are people who can consistently hit 800 on the SAT’s variant. They’re not inconsistent and there is pretty much always an obvious best answer. And I never studied examiner dictionaries or grading guidelines, it’s not scrabble or a spelling bee.

I think it's trickier than that, though. I've got this fairly strong memory of a multiple choice verbal question along the lines of

'How does Alcohol impact driving skills

A) Strongly negatively B) Negatively C) Significantly Negatively D) Positively '

Where I'm sure they wanted the exact wording of the pamphlet in hindsight, but it's also somewhat absurd to mark any of A-C as incorrect.

Fair enough, I definitely agree I'm less good at verbal compared to quant. The other thing that always ground my gears was the fact that we got more time per question on the quant section compared to the verbal section, they were giving 1:15 or so per quant question and like 0:50 per verbal question, even though the quant sections were stuff you could spit out answers to in 0:30 while the verbal stuff often required reading a passage, which used up the 0:50 you had per question.

Where does that theory come from anyway? Jews are clearly way overrepresented in high quant ability too. Are they all that skewed toward being wordcels?

Jews are extremely overrepresented in both, but there’s some evidence of relatively higher representation in verbal fields, eg. Jews are overrepresented by 700% among math professors but 1300% among law professors. As with any Kierkegaard theory it’s highly questionable and shouldn’t be taken as fact.

Some have argued, separately, that physics is really more of a hybrid or perhaps even primarily verbal field than a quant field.

Couldn't that have to do with network effects counting for more in Verbal pursuits, since it's less objective who the 'best' law professor is whilst mathematics can be tangibly measured far more easily?

Some have argued, separately, that physics is really more of a hybrid or perhaps even primarily verbal field than a quant field.

But physics is applied math. Theoretical physics is not uncommonly on forefront of math or indistinguishable from pure maths.

You might say that some pure maths is also a verbal field, I've taken semester long courses in pure mathematics where there was not a single number at any point (other than 0 or 1, but even then they didn't represent the numbers 0 and 1 as commonly used).

No indexed sequences, even? Was it all category theory?

Always has been.

Ah, there were indexed sequences, you are right about that. It was a highly complex Abstract Algebra course, but there were sequences of Ideals.

Maybe I should change my claim to "did not use numbers for any purpose beyond kindergarten counting".

Still sounds suspect. No explicit examples worked out all semester?

More comments

I think part of the issue is that a verbal/spatial split is not the same as a verbal/quantitative split. Something like theoretical math or physics might actually be more verbal than it is spatial even though it falls on the quant side of the verbal/quant split.

Harvard mostly boils down your smartness into their Academics rating as described here:

  1. Summa potential. Genuine scholar; near-perfect scores and grades (in most cases) combined with unusual creativity and possible evidence of original scholarship.

  2. Magna potential: Excellent student with superb grades and mid-to high-700 scores (33+ ACT).

  3. Cum laude potential: Very good student with excellent grades and mid-600 to low-700 scores (29 to 32 ACT).

Near-perfect test scores and grades will only ever get you the second-highest rating. I remember when I was looking at colleges 10 years ago that I noted that Brown only admitted ~25% of people with perfect ACT scores.

When you combine this with now-public data on Harvard's admissions, it becomes pretty clear that, with no change to the ACT/SAT, Harvard could pretty straightforwardly choose the next incoming class to have an average IQ of at least half a standard deviation higher than previous classes.

I think that's the rub: even if the ACT/SAT were redesigned to better discriminate among the top of the distribution, Harvard et al's current behavior makes me pretty skeptical that this would result in smarter people being admitted.

That being said, if you have amazing test scores and grades, you should probably really consider Caltech - they're have no legacy or affirmative action, and they place a huge emphasis on those exact factors.

This is neither here-nor-there, but there is good evidence that top schools under-weigh test scores if their goal is to predict who will be most successful. Who knows to what extent this is because (a) intelligence is super important at accomplishing things or (b) nearly all selective institutions [edit: including med school, law school, FAANG companies, consulting firms, etc] use intelligence filters since they're easy to evaluate - for instance, grit is hard to figure out in a test or interview.

Does cal tech still do this? I had heard that they had jumped on the social justice wagon and weren’t even excepting standardized testing scores for a while.

I’m actually having a hard time finding any concrete documentation from the last year. However, the Asian:Black ratio at Caltech is 10. At MIT it’s 4.4; at Harvard it’s 2.4; at Stanford it’s 3.5. So, at the very least it’s far lower in practice.

This is neither here-nor-there, but there is good evidence that top schools under-weigh test scores if their goal is to predict who will be most successful. Who knows to what extent this is because (a) intelligence is super important at accomplishing things or (b) nearly all selective institutions use intelligence filters since they're easy to evaluate - for instance, grit is hard to figure out in a test or interview.

I think this is it. They not only want smart students, but those who have potential to be leaders of industries, law, politics, etc. that will bring donations and repute. But interestingly, purely selecting for merit does optimize for donations though. MIT's endowment is also very big, about half that of Harvard, and bigger than Harvard on a per-student basis. High-IQ hedge fund managers and CEOs donate a lot.

1460 is around the 25% of Cornell's students. So he's worse than average, but it's still a score that plenty of people get in with.

But, like, didn't everyone know that these schools discriminate?

As to the overall point being made, I agree that they could have more discrimination. That said, getting very high scores is still pretty hard and indicates that people are pretty capable, but yes, there is information lost.

The other cost to having the top percentile having all scores right near the top is that it means that random mistakes cause more of an impact, as occasional random mistakes somewhere or another make more of a difference when the difference is only a few questions.

Sure but the rub of the averages is the people getting in with bottom half of the distribution scores are qualifying for admission elsewhere like athletics, legacy, large donor, or unique non-scholastic achievements (like being a nationally famous high schooler).

I am also curious how someone who is really good at math could stand out in today’s environment. The SAT and a few good AP math scores wouldn’t seem to be enough. Do you need to have the opportunity to compete in high-end math tournaments?

The AMC/AIME/USAMO series of tests is widely available and is the most prestigious high school math competition in the country. I went to a thoroughly mediocre public school and even my school offered it. Which allowed me to become a USAMO winner. Of course Harvard still rejected me because they're a bunch of pretentious assholes.

Which allowed me to become a USAMO winner. Of course Harvard still rejected me

Looks like it's IMO or bust for these sort of schools (tbf I actually know a fair amount if IMO silver medallists and even a gold medallist who was rejected by all of these schools, although all these people were internationals).

pretensious

Wordcels had the last laugh, huh.

I suppose lol

1993/1995 - some test changes but the big thing was a recentering to get scores back to about 1000 from 900. Before this update a median score at HYPS would have been 1370-1400 area. Bill Gates 1590 would have really stood out and guaranteed alone admittance to Harvard.

In Gates' era, perfect scores were really rare and a pretty big deal, signifying top-tier IQ. Nowadays, not as much but still impressive.

I am also curious how someone who is really good at math could stand out in today’s environment. The SAT and a few good AP math scores wouldn’t seem to be enough. Do you need to have the opportunity to compete in high-end math tournaments?

Many ways, although people of lower-SES may be excluded: original research (such as collaborating with a professor to co-author a paper during summer), letters of recommendation, math competitions, publishing code to github (not pure math but still correlated with ability). One of the advantages of the internet is it makes it easier to do independent research.

A major criticism is that the lower ceiling of the new SATs makes it harder to identify top talent from more disadvantaged or lower-middle class backgrounds.

1460 SAT

This really isn't very impressive. Looks like a Verbal 660 and a Math 800 (or close enough), anyone who takes even a cursory interest in reading/writing in the English language and isn't failing at mathematics should be able to match it. If it was 1560 maybe this dude would have a point, with his current scores it looks like he just got filtered due to there being better candidates.

Looks like a Verbal 660 and a Math 800 (or close enough), anyone who takes even a cursory interest in reading/writing in the English language and isn't failing at mathematics should be able to match it.

That's absurd. A 660 in English is 87th percentile. An 800 in Math is 98th percentile. And that's from the population of students who bothered to take the SAT, not from the general population. 98% of students are sure as fuck not passing algebra. More than 13% of college-bound seniors have taken a passing interest in reading and writing.

I agree that a 1460 SAT is not very impressive for an Ivy League application if you are white, though it's above the average for black students accepted to Harvard. But it's still a WAY above average score. It's just not enough to stand out when you are applying to the most competitive schools in a country of 300 million people.

This thread is super funny to me because I got a 1460 and went to community college and then a decent state school from which I went right into the workforce and never graduated.

I was a decent student, 3.5 GPA, took no sat prep.

I got a perfect score on the Verbal portion which does give me a twinge of pride.

Depending on the IQ scale I’m in or around the 99th percentile, but I had a bunch of other issues which hobbled my academic performance.

Learning not to mix up my sense of self worth with my academic performance or intellectual ability was an important aspect of reaching maturity in my young adulthood, and I’m now very happily living a working class-ish lifestyle far away from the influence of the PMC peer group I had in university.

Although my high intelligence is usually the most obvious attribute people notice upon meeting me, I don’t even think it’s near the top of my best qualities.

I’m not even sure it’s all that important in isolation anymore, only as a support to other good qualities a person can have.

Yes, and all US colleges weight against math and in favor of verbal scores because perfect math scores are so much more common.

Look at the GMAT. Tons of Indians and Chinese hit perfect scores on the quant section (50/51 is like 85th percentile, a perfect 51 is barely 95th percentile). Meanwhile, even 45/51 on the verbal section is 99th percentile.

So the GMAT - at the top level - is essentially a verbal IQ competition and HSW etc will even let you drop a few points on quant if you have an extremely impressive verbal score.

Tons of Indians and Chinese hit perfect scores on the quant section (50/51 is like 85th percentile, a perfect 51 is barely 95th percentile). Meanwhile, even 45/51 on the verbal section is 99th percentile.

I honestly think this is bad test design though. The scaled scores should correspond to roughly the same percentiles for both quant and verbal, doing otherwise just means the same score on Quant vs Verbal doesn't mean the same thing in relation to your peers also applying for the same programs, e.g. currently a 48/51 verbal is amazing, while a 48/51 quant is very meh.

It's not like this is hard to fix, if necessary you can just increase the number of questions on the quant section while keeping the same time limit for the sections, that way you can tell the super good quant people from the merely good quant people (the former will solve questions faster, so get more of them done in the allotted time) while right now basically everyone with a command of the materials gets close to full marks and you can't distinguish the "amazing" from the "merely very good".

I'm glad I never had to do the GMAT, I think I'd do well but it's just ugh, like why...

Does the relationship between problem solving speed and g scale infinitely? Can you really just take a dozen easy matrices, give everyone 4 minutes and get the same ranking as if you take a bunch of hard problems and give the same people a couple hours to do them? I’m skeptical.

I guess it seems likely they’ve considered that solution.

Does the relationship between problem solving speed and g scale infinitely

Probably not infinitely, but it probably scales farther than the current tests are pushing it. I remember when i took the SAT I thought they gave you way too much time on the math section. I finished every question and double checked all my answers in less than half the time allotted so i just put my head down at that point.

Fair enough, it doesn't scale infinitely but it does a lot, and it definitely scales in the region where SAT/GMAT tops out.

See e.g. the Wonderlic: https://en.wikipedia.org/wiki/Wonderlic_test which is just 50 easy questions that any IQ 100 person should be able to solve given unlimited time, the catch is that you just have 8 minutes for everything, and so the average score is 20/50, making it effectively a test of speed on "easy" questions. All the evidence I've seen is that high wonderlic scores are very highly correlated to high IQ scores as done with a certified test like WAIS, and this correlation doesn't disappear even close to the 50/50 upper limit.

Yes, that’s a good example. I’d be interested to see how efforts to raise the ceiling for the quant GMAT section would pan out.

Just what is the cut off for Cornell anyway? Top 4% seems very low. I would have assumed that the cutoff for ivy league was at least top 2%.

As part of the overall student population the ivies only account for some 0.4%, but there are obviously other desirable universities as well.

Yeah, 1460 isn't so high with the new SAT. This completely credible article from a super-reliable source about a student with a 1580 and a high GPA being rejected from all the top schools is a bit more damning towards the schools. Or would be if the "completely credible" bit above wasn't sarcasm.

Why waste time going to college if you already have a business making 200K a year? Especially since that business is in the digital marketing space, nothing you learn in college is going to help with that. Yes, I'm aware of the possibility that the TikTok user just made up the scenario for clicks and views. I also know people value a degree and that there may be family/social pressure to attend college.

In the Blue Tribe, not going to college is Just Not Done. College is a rite of passage, not merely a venal way to increase your salary; you are not a real adult until you have a degree.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats