By universities, I would also include colleges of course. The claims about tertiary education and schools are generally well known, and backed by various rankings.
In regards to K-12 education, there are a couple of studies done on that (wherein most find the same results). This article from the Pew Research Center seems to be a good basis for the question.
The simple explanation is that virtually all K-12 schools are funded directly by governments (in other words they are politically-driven entities), while many colleges and universities are funded in no small part by tuition. Thus, if a public school system begins to fail, it may or may not get any reforms to stop it from failing in the future.
A simple Google search for "Harvard Tuition rates" gave me $50,420 for one year of tuition 2018-2019. Actual list prices might be higher and some students pay less than others based on financial assistance (the article quotes Harvard as saying students pay an average of $12,000). Harvard also has a net worth of $38B and pays professors about $225,000. It also spends $1.1B on research annually (with a fair amount of government-funded). I'm willing to bet most of the Ivy League schools like Yale would clock in about the same. For the sake of argument, I'm going to run with those numbers for a minute.
State schools, by contrast, spend about $10,000 per student (2016 numbers, with some as low as $7,000, and some as high as $22,000). That's not terribly surprising. States have to educate far more people and are obligated by law to do so. These schools are free to residents (in tuition, not taxation).
What's more important is that states have other things to do with that money beyond just funding education. Illinois, for instance, has a serious teacher pension problem
In the [2020] school year, 36% of the money the state allocates to education will be diverted away from teachers and students to meet required pension payments for retirees.
In other words, $1 of every $3 for education is being spent on a non-education cost. And that's just pensions. We're not even talking the cost of buildings, buses, school lunches, etc. As such, Illinois ranks 20th in graduations
New data from the U.S. Department of Education show that 86 percent of Illinois students graduate from high school. That's the 20th-highest graduation rate in the country [for 2015].
Furthermore, if I dislike the services that a university is providing, I can always change colleges. Thus it is in the interest of a university to ensure I can getting what I am paying for, lest I take my money elsewhere.
By contrast, if someone lives in a failing school district, they may not have any options at all to change schools. Worse is some districts are stuck in The Lemon Dance, where underforming teachers are shoved between schools because they cannot be fired (New York City actually has dedicated rooms for educators who cannot be in the classroom, but cannot be fired). This is why charter schools are contentious, because they are schools that can (and do) fail, but it offers some modicum of choice, even if that choice involves an education lottery. Since that system is so hard to rely on some people have become desperate enough to turn to fraud to get children better schooling
Ohio mother of two Kelley Williams-Bolar was released from jail on Wednesday after serving nine days for falsifying records so that her two daughters could attend a better school.
To get down to the heart of it, In the USA, K-12 education is considered 'populist' while upper-level education is considered 'elite', which brings into play a whole range of social and political biases that affect funding and resource allocations.
K-12 education, ever since Dewey started pushing to universalize it at the end of the 19th century, was seen as an equalizer. Its goal was to ensure that even the poorest of children would receive basic knowledge and skills that would help them increase their standing in life, and make them better, more competent citizens. The arguments in favor were obvious. An agricultural worker of the 19th century sort could get by perfectly well without being able to read, write, or do arithmetic, because his livelihood did not require much knowledge of the world outside of the (complex) details of running a farm. But a twentieth century technological worker could not afford that same lack. Technical work required technical skills, which implied studying and learning; advancement in a technological world hinged on skills that could only be mastered with a basic level of education. Unless the USA was content to have a permanent underclass of illiterate workers, some kind of public education system was needed.
However, in the early twentieth century (and even now) this met with a lot of resistance. Those in the Jim Crow South objected, because they resented the idea that blacks might be allowed to advance with respect to whites. Those in the industrial Northeast resisted the idea that they should pay to educate the children of ignorant immigrant laborers, because (to the blue-blood mind) those children didn't need education beyond that needed to effectively work in a factory or a mine. And most of the country, honestly, held the bias that the education of children was women's work. 'Teacher' was perhaps the most respectable job a woman could reach for, but it was hardly considered a job at all by men, who viewed it as a natural extension of child-care.
By contrast, college and graduate work were and are viewed as vehicles for success, where 'superior' people could advance themselves into coveted positions in society. Thus, colleges and universities are willingly funded by the wealthy through tuition, donations, grants, funding of chairs and departments, etc. College is perceived to boost status in society for the elite who can manage to gain entry — whether through wealth or talent — and anything that boosts status, gains status.
Even today, public schooling is beset by people who complain that their tax dollars are being used to educate other people's children. I believe it was JD Salinger who (about ten years ago) sued New Hampshire because New Hampshire had passed a state tax that allocated revenue to public schools across the state, where Salinger thought his tax dollars should be dedicated only to his school district. Likewise, conservatives across the nation have pushed for voucher systems which pull tax dollars out of public schools to help certain people send their own children to private K-12 schools. The net effect of all these issues is to strip money from public institutions into more private and exclusive ones, extending the 'elitist' model of higher education down into the K-12 system.
I would advise you to beware statistics!
Speaking as someone who works in the UK education sector, where we exhibit similar metrics to the US education system, when you start looking at world rankings you have to be aware that you are comparing societal educational values as much as educational outcomes.
OECD, who are the prime source for worldwide statistics base their PISA statistics on Mathematics, Science and Reading Comprehension, at age 15, because they are the only things that are important to a 15 year old, right?
The world rankings, as a result, ascribe a higher value to what I would call an Eastern-Style of education. The PISA rankings are topped by China, Singapore, Japan, etc and reflect the fact that these nations put a much higher value on a scientific education than many western nations do. In the UK, for example, it is a regular occurrence for 15 year old students to be spending 50+% of their time studying Arts & Humanities with the rest being a balance of Languages and the STEM subjects.
To summarise, I wouldn't pay to much attention to the "Average" rank of K-12 institutions, because it needs to be understood in the context of the relative breadth of the US curricula vs. other countries.
Funding issues have already been mentioned, but there are other major dynamics at play here, too: especially rigor and the standards used to measure success... and the stratification that those factors lead to.
Performance metrics for public primary and secondary schools in the U.S. are based largely on how many of their students graduate. They are constantly under heavy pressure to make sure the slowest students in their classes can keep up and the makeup of the classes are often more or less random relative to the population assigned to attend that school, with often little-to-no stratification based on ability, especially in the primary grades. This means that the entire classes are often, if not usually, slowed down to paces where the slowest students can at least come close to keeping up. Meanwhile, the students most proficient in a given field will be bored and mostly unchallenged (yeah... personal experience there.)
Tertiary education - especially at universities - is thought of entirely differently. Their standards for success are more along the lines of academic prestige (especially with regards to their research production,) how easily their graduates can find jobs in their desired fields, how much those jobs pay, etc. This leads to a dramatic difference in focus.
One of the most dramatic differences is that colleges can select which students they admit and students can select the colleges to which they apply. This leads to far more stratification in courses by ability level, both from student self-selection and selection from among the applicants by the colleges. This, in turn, allows the courses - especially at higher-ranking universities - to have much faster paces where even the most proficient students can be constantly challenged.
Another of the most dramatic differences is that college instructors/professors are also under far less pressure to make sure everyone passes than their primary and secondary counterparts. On the contrary, they're under more pressure to make sure that their courses are sufficiently rigorous that the students who pass them will be prepared for jobs and/or research in those fields. This dramatic difference in focus and rigor often comes as quite a shock to first-year students and, as a result, pass rates, especially in the first year, are far lower than in primary and secondary education. At the more technically-focused university I attended (known mostly for its engineering school,) it was not uncommon for first-year courses like Calculus, physics, chemistry, and introductory computer programming/software design courses to have pass rates around half. And that's after the aforementioned self-selection of the university by students and students by the university. These courses are often known as "weed-out" courses due to their propensity to result in a large number of change-of-major forms being filled out by first-year students and even more first-year students who drop out or are placed on academic suspension. They are also jokingly known among both upper-division students and faculty to be the courses that freed up parking spaces in the university's parking lots...
These factors then feed even further into stratification based on abilities and motivation. By the time you get to upper-division classes (i.e. third- and fourth-year undergraduate classes,) only the students who have actually survived the first-year and second-year courses remain, which then allows the courses to be far more focused and academically rigorous than anything you would find in a public secondary school.
Then along comes graduate school, adding yet another layer of stratification by ability and motivation, again by means of self-selection on the part of both the students and the university graduate programs' admissions processes. It's at this level where much of the "academic prestige" factor comes in for universities, as this is where the vast majority of the research is done. According to the U.S. Census, 13.1% of Americans aged 25 and over currently possess at least a Master's degree. If you remove the 41% of those in the mostly non-research-based MBA and M.Ed. programs, you're down to a highly-self-selected 7-8% of the population. By contrast, over 90% (and 93% of those aged 25-30) have at least a high-school level education.
While I think the above accounts for the vast majority of the difference, it's also important to point out the differences on the faculty side of the equation in universities, as well. Whereas the average primary or secondary teacher will have a Bachelor's and maybe M.Ed., virtually all university professors will have a Ph.D. or equivalent terminal degree. And that terminal degree will typically be in the specific field-of-study that they're teaching and/or researching, rather than a more general education degree as is typical for primary and secondary education teachers. The selection process for university faculty is also far more rigorous than for primary and secondary education, with the university faculty selection committees typically seeking out literally world-class experts in the specific fields that they will be teaching and researching. And, going back to the beginning of the answer, the standards of success that the faculty are measured by in universities are mostly their academic qualifications and research production, not so much how much they go out of their way to make sure every student who enters their door passes and standardized test scores like in primary and secondary schools.
This is actually a huge misconception, US K12 education is actually absolutely top-notch, but there are significant disparities among various ethnic groups that muddle the picture.
There is an organisation called PISA, that runs massive tests of high school students every 3 years(It's what the question's Pew article was referring to. PISA also provides demographic information in their data and here are the results for the US in 2018, in comparison to every other country in the test dataset, ranked from best to worst:
| Education system | Mean | Reading | Math | Science |
|------------------------- |------ |--------- |------ |--------- |
| B-S-J-Z (China) | 579 | 555 | 591 | 590 |
| Singapore | 556 | 549 | 569 | 551 |
| Asian Americans | 549 | 556 | 539 | 551 |
| Macau (China) | 542 | 525 | 558 | 544 |
| Hong Kong (China) | 531 | 524 | 551 | 517 |
| Estonia | 525 | 523 | 523 | 530 |
| European Americans | 521 | 531 | 503 | 529 |
| Japan | 520 | 504 | 527 | 529 |
| Korea, Republic of | 520 | 514 | 526 | 519 |
| Canada | 517 | 520 | 512 | 518 |
| Chinese Taipei | 517 | 503 | 531 | 516 |
| Finland | 516 | 520 | 507 | 522 |
| Poland | 513 | 512 | 516 | 511 |
| Ireland | 505 | 518 | 500 | 496 |
| Slovenia | 504 | 495 | 509 | 507 |
| United Kingdom | 504 | 504 | 502 | 505 |
| New Zealand | 503 | 506 | 494 | 508 |
| Sweden | 502 | 506 | 502 | 499 |
| Netherlands | 502 | 485 | 519 | 503 |
| Denmark | 501 | 501 | 509 | 493 |
| Germany | 500 | 498 | 500 | 503 |
| Belgium | 500 | 493 | 508 | 499 |
| Australia | 499 | 503 | 491 | 503 |
| Switzerland | 498 | 484 | 515 | 495 |
| Norway | 497 | 499 | 501 | 490 |
| Czech Republic | 495 | 490 | 499 | 497 |
| United States | 495 | 505 | 478 | 502 |
| France | 494 | 493 | 495 | 493 |
| Multi-racial Americans | 492 | 501 | 474 | 502 |
| Portugal | 492 | 492 | 492 | 492 |
| Austria | 491 | 484 | 499 | 490 |
| OECD average | 488 | 487 | 489 | 489 |
| Latvia | 487 | 479 | 496 | 487 |
| Spain | NA | NA | 481 | 483 |
| Russian Federation | 482 | 479 | 488 | 478 |
| Iceland | 481 | 474 | 495 | 475 |
| Lithuania | 480 | 476 | 481 | 482 |
| Hungary | 479 | 476 | 481 | 481 |
| Italy | 477 | 476 | 487 | 468 |
| Luxembourg | 477 | 470 | 483 | 477 |
| Belarus | 472 | 474 | 472 | 471 |
| Croatia | 472 | 479 | 464 | 472 |
| Hispanic Americans | 470 | 481 | 452 | 478 |
| Slovak Republic | 469 | 458 | 486 | 464 |
| Israel | 465 | 470 | 463 | 462 |
| Turkey | 463 | 466 | 454 | 468 |
| Ukraine | 463 | 466 | 453 | 469 |
| Malta | 459 | 448 | 472 | 457 |
| Greece | 453 | 457 | 451 | 452 |
| Serbia | 442 | 439 | 448 | 440 |
| Cyprus | 438 | 424 | 451 | 439 |
| Chile | 438 | 452 | 417 | 444 |
| African Americans | 436 | 448 | 419 | 440 |
| United Arab Emirates | 434 | 432 | 435 | 434 |
| Malaysia | 431 | 415 | 440 | 438 |
| Romania | 428 | 428 | 430 | 426 |
| Bulgaria | 427 | 420 | 436 | 424 |
| Moldova, Republic of | 424 | 424 | 421 | 428 |
| Uruguay | 424 | 427 | 418 | 426 |
| Brunei Darussalam | 423 | 408 | 430 | 431 |
| Montenegro, Republic of | 422 | 421 | 430 | 415 |
| Albania | 420 | 405 | 437 | 417 |
| Jordan | 416 | 419 | 400 | 429 |
| Mexico | 416 | 420 | 409 | 419 |
| Costa Rica | 415 | 426 | 402 | 416 |
| Qatar | 413 | 407 | 414 | 419 |
| Thailand | 413 | 393 | 419 | 426 |
| Colombia | 405 | 412 | 391 | 413 |
| Bosnia and Herzegovina | 402 | 403 | 406 | 398 |
| Kazakhstan | 402 | 387 | 423 | 397 |
| Baku (Azerbaijan) | 402 | 389 | 420 | 398 |
| Peru | 402 | 401 | 400 | 404 |
| Brazil | 400 | 413 | 384 | 404 |
| North Macedonia | 400 | 393 | 394 | 413 |
| Argentina | 395 | 402 | 379 | 404 |
| Georgia | 387 | 380 | 398 | 383 |
| Saudi Arabia | 386 | 399 | 373 | 386 |
| Indonesia | 382 | 371 | 379 | 396 |
| Lebanon | 377 | 353 | 393 | 384 |
| Morocco | 368 | 359 | 368 | 377 |
| Panama | 365 | 377 | 353 | 365 |
| Kosovo | 361 | 353 | 366 | 365 |
| Philippines | 350 | 340 | 353 | 357 |
| Dominican Republic | 334 | 342 | 325 | 336 |
As you can see, in terms of actually educating it's students, the US educates:
The only African country that participated in PISA is Morocco, but since it's near the top of African countries in HDI, it's safe to say that the other African countries are at least as bad as, or much worse at educating their students than the US is at educating African Americans.
I'm not speculating upon the reasons for the group differences, because it's a white hot, radioactive potato and not relevant to the question, I'm just pointing out that the US is actually doing great at educating whichever children it has.
This is actually a classic example of Simpson's Fallacy, but mentioning group differences is understandably avoided, so every time PISA results come out, there is the same set of predictable articles lamenting the US K12 education system and they'll be the same in 2021, when the next round of PISA tests takes place.
