Yup, here we go again: US education is falling behind other countries.
This time the cause is bad teachers protected by the teachers union. I'm all for busting up the teachers union, but I don't for a nano second believe that bad teachers are really the cause of what is being observed.
Instead, what's being observed is deliberately made up, mostly nonsense, mostly to spend still more money on K-12 education which, with some irony, would help the teachers union!
We went through all this just 17 months ago with
McKinsey's report, The Economic Impact of the Achievement Gap in America's Schools, April, 2009,
Cover to cover it is clear that the main concern of this, apparently pro bono effort by some dedicated McKinsey staff members, was to scream that US K-12 educational achievement sucks while what they really meant was that US Black K-12 educational achievement sucks. So, that pro bono effort was to save the US Blacks.
Throughout the two long PDF files, they omitted any view of the elephant in the room until they gave a glimpse on page 26 of
supporting materials (PDF - 1.0 MB)
where they did a little cross-tabulation, that is, started on the main technique in looking for causality in social-economic data.
So on page 26 we finally get the US students, White, Latino, and Black, compared separately with students in many other countries in the world.
So, wonder of wonders, the ranking goes:
Finland
Hong Kong
Canada
Japan
Australia
US whites
Korea
Germany
United Kingdom
Switzerland
Ireland
Sweden
The US Latinos? Just above Chile.
The US Blacks? A little ahead of Indonesia and Argentina.
Is there any question how the US Blacks would do against, say, Ghana, Liberia, Nigeria, Kenya, Uganda, Zaire, Zimbabwe?
So wonder how US students of Finish descent would do against Finland, Asian descent, against Hong Kong, Canadian descent, against Canada, Japanese descent, against Japan, Australian descent, against Australia? The study didn't say. Ah, wonder why? Why oh why? Maybe they followed the old rule, "If you don't think you will like the answer, then don't ask the question.".
So the study mostly tried to obscure the obvious cause: For the data they considered, the main cause of good performance is country of origin. Right: In particular, parents are a still more important cause, and it might be possible to dig still deeper.
Then the main reason for the low average US ranking is that the US is a diverse country and the other countries are homogeneous.
So, Virgina, this statistical pattern has NOTHING to do with education or US education: Instead, take any 100 homogeneous populations, of anything -- kitty cats, puppy dogs, or billy goats. Then pick a measure, any nontrivial measure you want from length of hair, body weight, running speed, jumping ability, whisker length, nearly anything at all. Then from the 100 homogeneous populations, make a diverse population and apply the measure. Then rank all the 101 populations. Presto: Wonder of wonders, the diverse population will be ahead of some of the homogeneous populations and behind some of the others. Obvious. Trivial.
And that obvious, trivial observation is so far ALL the screaming is about. How 'bout that.
So, if want to tar and feather US education, which for other reasons I'm plenty eager to do, will build the fire to warm up the tar right away, then compare US education with apples against apples. So, compare US education on people from Finland with Finland education on people in Finland. In this way have compared US education with Finland education. Then rinse, repeat for Hong Kong, Canada, Japan, etc.
But the semi-, pseudo-, quasi-smart, partly objective McKinsey team last year and the movie people this year showed their grasp of education: All they showed was the same thing would see from 101 collections of kitty cats measured on length of whiskers.
There's another big point: Apparently actually there isn't much in K-12 in the public school systems of any of the countries so that at most not much is missed.
Another point: Let's compare college education and, there, GRE scores. Bet the top 100 US colleges and universities do quite well, thank you.
Another point: Let's compare graduate education and research progress -- as we know, the US by itself totally blows all the rest of the world off the court and out of the arena.
Another point: K-12 really should amount to something. So let's move to that.
Another point: It may be that the US should take vocational education as seriously as, say, Germany does. We should consider that.
Another point: In the classic Democracy and Education, John Dewey made a big point: He defined education as the passing from one generation to the next and mentioned that what gets passed is mostly just what was there, both good and bad, with maybe a little improvement at each passing. So, without some quite special efforts, tough to take teachers from the middle third of US educational accomplishment and have them pass on only the top 10% of educational content.
Another point: Once again we see that high accomplishment is mostly the responsibility of individuals and where their best help is from their own families. So, enter the promise of home schooling.
Here's my approach to improving US education (listen up billionaires):
First, set up educational certification in nearly all subjects of interest from math, physics, and chemistry to auto body repair, grass mowing, all the way down to, say, computer science, and from there way, way down to, say, programming in C++, if could find anyone so weak minded. The CEEB tests should be a good start.
Second, get most colleges to agree on the required subjects to be ready for college work.
Third, fund some efforts at educational materials and programs to prepare students for the tests. Of course should try to make heavy use of PDF files, video lectures on YouTUBE, Web fora, tutoring, etc.
Then, let free enterprise take over and develop still better educational materials.
For K-12, leave that to local school boards: Some will get it and let well motivated and guided students just pursue some of the better sets of materials and, then, be ready for college work at age 10 or 12.
Then have those 12 year olds compete with 12 year olds around the world.
Here we cut out a lot of nonsense: If someone wants to learn, then sit in a cubicle with a computer, work through some of the best educational materials, take the test, and see how well they do, and then all the credit and/or blame is just theirs.
For varsity athletics, f'get about it except for the cheerleaders -- wouldn't want not to have the cheerleaders, a crown jewel of the US educational system!
How does moving educational goal posts based on race raise the level of education in the US?
Furthermore, the educational system you describe already occurs throughout many parts of the US. Test prep and the private sector is not an answer to a societal/cultural problem.
You've got a mishmash of pretty good and execrably bad in this comment. Context: I'm not a teacher, but my wife is; we've seen what happens when assessment-happy libertarians (or pseudo-libertarians for those who want to get upset with my assessment) actually get to destroy a school system. For the fellow Ontarians in the virtual room, I'm talking about Mike Harris.
For those who want to follow my deconstruction at home, I'm going to refer to the original report[1] and the detailed findings[2] that the original poster pointed to but with live links.
Let's start with a couple of points that go to history. Fifteen years ago (not too long before I emigrated), Mike Harris was elected in Ontario and started on a massive program of social change from which Ontario has yet to recover. His schools program changes weren't all bad, but could have been achieved much less confrontationally and more smoothly, and could have been done without harming the quality programs that existed before. The basis for all of the changes was rooted in a politicized interpretation of test results that weren't meant for such comparisons by the people who put those tests together (at the time, it was largely TIMSS as the driver).
TIMSS is the "Trends in International Mathematics and Science Study" (http://timss.org/), and at the time, the results were interpreted by the Harris government to show that Canadian students (and most especially Ontario students) were performing badly. One problem: TIMSS was supposed to be a universal student study (that is, all students at the grade level in the school system were supposed to be taking it regardless of ability); some of the investigated countries[3] did selected populations (that is, they excluded their worst students) instead of universal.
[3] My recollection is that the above criticisms were laid against some of the PacRim countries and some of the former Soviet Bloc countries.
Canada, the U.S. and many other countries followed the rules and were subsequently ranked much lower than they would have been ranked otherwise (they still would have been middling-to-low ranking, but not as low as they were). The charges of population homogeneity that you make were also made in Canada at the time ("of course our diverse kids didn't do as well as the uniform kids in Korea").
This last makes one of your conclusions nonsense:
"Then the main reason for the low average US ranking is that the US is a diverse country and the other countries are homogeneous."
Canada isn't homogeneous and ranked 5th in science and 3rd in math on the PISA tests (page 4 of the detailed findings); New Zealand and Australia (both also immigrant countries) exhibited good results. The United Kingdom (a diverse country that has some immigration) exhibited average results. It's also worth noting (page 7) that Canada's performance declined by 5 points between 2003 and 2006; the U.S. declined by 9 points in the same period. (Most countries, in fact, declined, with France dropping by 15 and clustering around the U.K. for overall performance.)
It's also worth noting that the statements on page 8 essentially assume that only the U.S. has changed (e.g., it ignores the efforts that other countries have put into improving and focusses solely on the U.S. lag).
The ethnic make-up of Canada is similar to that of the U.S. as we're both immigrant countries. This isn't about race or diversity. It's about socioeconomic disadvantage (sorry, but it's true). Those two things tend to be highly correlated in the U.S. (and to some degree in Canada, too), but school funding in Canada is much more uniform than it is in the U.S. so that socioeconomic differences are smoothed out but not eliminated. See pages 12-13 of the detailed document.
You choose to focus on the ethnic achievement gaps pointed out by McKinsey, but they don't stop there and go on to point out the income achievement gap. Look at pages 42-45 for how this works. Compare the chart on page 26 (a 3 - 8 times difference in low achievement by race) with that on 43 (a 4 - 6 times difference in low achievement by income). There's strong correlation there.
The entire report is worth reading, not just the bit that you quoted or that I quoted.
You're right, though:
"This time the cause is bad teachers protected by the teachers union. I'm all for busting up the teachers union, but I don't for a nano second believe that bad teachers are really the cause of what is being observed."
The problem is a bad system, not bad teachers (although they exist and should be removed from the system). The problem is local funding that isn't smoothed out so that students with all of the advantages get even more advantages because their schools can afford more and better equipment on the same tax rates. The problem is an over-reliance on quantization of the problem space, when students aren't cogs but are individuals (and if you place performance bonuses, etc. on classroom performance of a test, you get "teaching to the test", not learning). The problem is hunger (hungry children cannot study as well as those who aren't).
There are things to be fixed with how teachers and teachers unions work, certainly, but they should be done with a deeper understanding and NOT by misunderstanding what we do know.
A couple of other points:
"Is there any question how the US Blacks would do against, say, Ghana, Liberia, Nigeria, Kenya, Uganda, Zaire, Zimbabwe?"
One might be surprised. I don't know of any particular test for this, but at least some of the countries you named (and other countries in sub-Saharan Africa) might outperform U.S. blacks; they will, at the very least, be better than your off-hand comment suggests.
"So wonder how US students of Finish descent would do against Finland, Asian descent, against Hong Kong, Canadian descent, against Canada, Japanese descent, against Japan, Australian descent, against Australia? The study didn't say. Ah, wonder why? Why oh why? Maybe they followed the old rule, "If you don't think you will like the answer, then don't ask the question.".
"So the study mostly tried to obscure the obvious cause: For the data they considered, the main cause of good performance is country of origin. Right: In particular, parents are a still more important cause, and it might be possible to dig still deeper."
Um, no. The study does say, after a fashion. Your comment is a little nonsensical, though, because you're simply taking the top countries on the first couple of pages and applying them as ethnicities. As I noted at the top, both Canada and Australia are immigrant countries with broad ethnic diversity (but a strong English/WASP contingent, just like the U.S.).
The rest of your comment is nonsense that's easily covered by the summary McKinsey did and the original data (see page 8 of the detailed report for a sample about post-secondary education). Home schooling is mostly a disaster (home schoolers in the U.S. on average perform worse than public school children on all subjects; this may partially be because most home schoolers in the U.S. are nutty Christians who don't want their children to be able to think critically).
What your prescription for education will do is pretty much put the U.S. results 100% in the toilet, as it ensures that only those people with money will have access to education. Education is not a simple subject, and the libertarian approach will fail--and has already (it's why we have public schooling and that the majority of economic growth has happened since public schooling started).
> The ethnic make-up of Canada is similar to that of the U.S. as we're both immigrant countries
According to http://en.wikipedia.org/wiki/Demographics_of_the_United_Stat..., the US is over 15% Latino and over 12% black (as of 2008). Canada is listed as 1% Latin American, according to http://en.wikipedia.org/wiki/Demographics_of_Canada, and blacks aren't even listed (implying less than 1%) (as of 2006). Since blacks and Latinos seem to be among the worst performers in US schools (citation needed?), your attempt to paint Canada as having a similar demographic to the US seems pretty bogus to me.
It has south asian: 4%, chinese: 3.7%, black: 2.5%, latin american : 1%, arab 0.9% and total visible minorities at 16.2%
Perhaps not as visibly diverse as the US, but obviously also mostly immigrant based and non-homogeneous as a result (unless by homogeneous you mean only skin color).
It's also worth pointing out that your figures for the US are out of 115%:
"These figures add up to more than 100% on this table because Hispanic and Latino Americans are distributed among all the races and are also listed as an ethnicity category, resulting in a double count."
I think you may have missed Hilbert Spaces' point (as halostatue may have). His thesis is not that "diversity" is a cause of low performance, but rather that genetics is. Holding up Canada as an example of another "diverse" country with better scores doesn't contradict that thesis if that diversity tends to be from the higher performing ethnicities.
Unforunately, in US politics, it is taken as an axiom that ethnicity (read: skin color) is not correlated with innate performance, and thus any difference in measured performance must be due solely to inequality of opportunity.
I didn't miss the point. I consider it deeply offensive and am baffled that anyone would consider it germane in 2010.
You mention "inequality of opportunity." The linked PDF overviews didn't ignore race; they (like I) indicated that in the U.S., race and poverty are tightly correlated. They also indicate (without touching the third rail of racial politics in the U.S.) that "racial" performance differences exist even within broader socioeconomic groups. This is also unsurprising, as there's a lot of subconscious racism in the U.S. (there's documented bias toward picking white or lighter coloured children to answer questions in class, even when the teacher tries to avoid it, as well as many other examples such as [1][2][3][4]).
Go back to Gladwell for a moment: advantages multiply. Whether you agree with his approach or not, on this part he's right. Canadian hockey players born in January are substantially more successful than those born in December because of the way that junior and senior hockey leagues are organized. This is because they are bigger, play better, have better coordination, etc. and then better coaches become more interested. They get more advantage because they started with a birth advantage. The inverse is also true: disadvantages multiply.
You also mention "measured performance"; many of these measures are unconsciously biased toward a middle-class to upper-class experience. In my wife's teaching experience, she has had students who are raised without religion who don't get (Christian) religious allusions that are present in some of the books that they read. If one doesn't have a particular experience, then one cannot be meaningfully tested on that experience. If your measure of "success" is based on those experiences, then your measure of success is by definition biased. That bias may be good or not, but the exposure to those experiences must be measured and controlled for before you start making sweeping (and wrong) statements as Hilbert Spaces was doing.
In 2010, Hilbert Spaces suggestion that ethnicity is the primary factor involved here is as nonsensical as the idea that girls should think that "math is hard." It's stupid, it's racist, and it's offensive.
--
[1] "One of the more upsetting discoveries is that children as young as three-years-old will associate positive traits with white people and negative traits with black people regardless of the race of the child or the attitudes of the children's parents and teachers." http://www.huffingtonpost.com/bob-samuels/unconscious-racism...
[2] "But inequalities extend beyond UD's African-American students and onto other campus minorities. An American born and raised student, junior Ed Hazboun, has faced discrimination multiple times due to his Arabic ethnicity. 'At one time my advisor for three years was going over my schedule and made a comment about the paper work I would have to fill out. You would think that after three years, that advisor would realize that I was a current student and not a foreign exchange student,' he said. 'Another time my philosophy teacher asked me if the Muslim religion viewed the topic we were discussing about ethics differently. Being born and raised Catholic, I was unable to answer.'" http://flyernews.com/articles/volume/57/issue/36/id/5718
[3] "A 2008 study published in the Journal of Personality and Social Psychology took Norton’s research a step further, examining the effects that whites’ attempts at colorblindness had on black participants. Ironically, the negative nonverbal behaviors exhibited by “colorblind” whites were interpreted by blacks as signs of prejudice, making them suspicious of their partners. It is hardly surprising that racial tensions increased among participants." http://greatergood.berkeley.edu/article/item/the_perils_of_c...
[4] "In many situations, from either the dominant or the oppressed, simple unconscious associations may drastically change outcomes. An example is Steele and Aaronson's (1995) work on stereotype threat, in which the performance of African-American students in a testing situation was cut in half by asking them to identify their race at the start of the test. This simple act unconsciously reminded students of the stereotypes connected with their race. Moreover, when asked at the end of the test, the students who were primed to remember their race were unable to identify the reminder as a factor in their poorer test score (Steele 1997)." http://www.pdkintl.org/kappan/k_v90/k0901mou.htm
Digging further into your link on [4] we find the following:
> To acquire such evidence, Joshua Aronson and I (following a procedure developed with Steven Spencer) designed an experiment to test whether the stereotype threat that black students might experience when taking a difficult standardized test could depress their performance on the test to a statistically reliable degree. In this experiment we asked black and white Stanford students into our laboratory and gave them, one at a time, a thirty-minute verbal test made up of items from the advanced Graduate Record Examination in literature. Most of these students were sophomores, which meant that the test was particularly hard for them—precisely the feature, we reasoned, that would make this simple testing situation different for our black participants than for our white participants.
Right, so the test was in literature and administered verbally. Gee, I wonder why these researchers hoping to find a causal link would want to choose such an easily gamed format for the test? Why do you think they didn't choose a paper-based multiple choice math test?
As for your response to my other post, your attempt to label the questioning of a possible link between race and intellect as "eugenics" is laughable. The person with the burden of proof in a given situation is the person making the claim. The claim being made in the original article is that US schools are falling behind. HilbertSpace pointed out that simply partitioning that data according to race paints a very different picture. If you want to make a claim about what this data means, then you're the one with the burden of proof.
> I consider it deeply offensive and am baffled that anyone would consider it germane in 2010.
What exactly do you find "deeply offensive"? Because I think your tone isn't helping your cause. If you wish to argue that race and/or genetics are not significantly correlated with innate ability, please feel free to do so. If this is an established fact that you can document, I would love to see links to research supporting that position (your citations establishing certain biases are much weaker). Your righteous indignation that someone would even bring this up (in 2010!!) is anti-intellectual and needs to go.
In my opinion, based mostly on common sense (and racism, no doubt you will claim), I don't think any of the assorted biases you cite are a major cause of performance differences. My intuition is that the major causes are much less subtle: genetics and family life (particularly early family life). I do agree that there is likely a snowball effect that causes early labels of "not good at school" to compound.
BTW. Your citation [4] looks deeply suspicious. Asking them to check which race they are cut performance in half? HALF? As in, they were able to, on average, supply 50% fewer correct answers if asked to check a box indicating their race?
I find it deeply offensive that anyone would take what amounts to eugenic nonsense seriously. If we've learned anything about genetics in the last decade of having the genome decoded, it's that we don't understand squat about how intelligence comes from genetic factors.
I don't have a cause here, by the way. I am not arguing a negative; it is up to the folks who which to establish a significant correlation to make their case. So far, all of the indicators are against them (genetics per se seems not to play a major role in success later in life).
You may think that the biases that I cite aren't major causes of performance differences, but that's no different than you saying that you think that green tea tastes good. I may think that they're major causes, but that's no different than me saying that black tea is better. What I know is that there are studies out there that indicate that there are many factors much stronger than genetics will ever be for success.
The top of these is, by the way, family life. These are measurable more in the negative than the positive: children of alcoholics and drug addicts tend to do worse in school than those with non-addicted parents; children of broken homes tend to do worse in school than those with unified families. One of the major positive correlations is reading: children whose family reads, even if they don't read together, tend to perform better in school than those who don't. What families read, especially together? Those who don't have to spend a lot of time working to make ends meet and put food on the table.
You may think that [4] is suspicious, but I've heard about this study a few times and I believe that it has been corroborated by other studies. Yeah, it's surprising, but that doesn't make it wrong. Again, step to Gladwell for a second and look at what Korean Air found out. When the subconscious social status indicators happened in language, accidents were more likely to happen. As soon as the entire cockpit switched to an informal English, accident rates dropped dramatically. So yes, reminding someone of their "social place" can significantly reduce their performance.
Blacks are listed. They make up 2.5% of Canada. Asians (a high performing minority) make up about 10% of Canada, whereas they make up only 4.4% of the US.
You did well finding URLs of the two parts of the McKinsey PDFs of April, 2009. Your knowledge of the content indicates that you are a fast reader or already knew about the materials.
We have at most trivial disagreement on facts. We likely have some unresolvable disagreements on politics.
In a post below, you went on and on with a lot of what I called "nonsense": Do such work as well as you please, and the results will still be nonsense. Sorry: That direction just gets deeper and deeper into stuff for which there is no reasonable solution.
You seem to have some quite broad educational objectives: Those promise to be very expensive and have highly mixed results, e.g., a gigantic waste of time for any very capable student.
We should agree: For the main academic subjects, such as considered in the McKinsey study, etc., it really is possible to have standardized tests that really are in practice essentially the best measure of learning. Indeed, that you take seriously the McKinsey analysis of the results of just such tests implies that you agree.
Are such tests perfect? No. Better than anything else widely available? Yes.
From my experience, the tests of the US Educational Testing Service are about the best measures of learning in K-12 and at least early college academic subjects. The test results are well respected for admissions to both college and graduate school. Net, these tests are a good resource for an effort to make progress in the US educational system.
Okay, have various paths to education. Have classrooms, special education, private schools, guided home schooling, etc. Fine. But, for all the paths, evaluate the learning with accepted, well designed tests, e.g., from ETS. So, the various paths have the same measure, the test scores, and much the same, real educational goals; so, the paths are in competition. GOOD.
Have all the paths you want, but for my kids I'd home school them with the goal of having them be ready for college work by age 10-12 and graduate work about three years later and having them do some research, publish some papers, take GREs, use those and the published papers to get accepted to grad school, take the qualifying exams, submit the published papers as the dissertation, and, thus, after less than two months on campus, one year later get a Ph.D. and to HECK with more in academic education. Only one degree -- Ph.D. Period. We're talking cutting out lots of expensive, time-wasting, psychologically wacko nonsense from nearly irrelevant down to totally upchuck-able.
E.g., some girl 13 gets to concentrate on learning calculus and not her hair style, doing well in her education and not gossip and social climbing.
For your main struggles and concerns, I outlined a simple, special path to the tests that should please you:
"Here we cut out a lot of nonsense: If someone wants to learn, then sit in a cubicle with a computer, work through some of the best educational materials, take the test, and see how well they do, and then all the credit and/or blame is just theirs."
The "nonsense" I'm talking about you described with lots of references: As I mentioned, in that direction the swamp just gets deeper thicker with no hope.
Here's how this simple, special path addresses many of your concerns: As we know, on the Internet, no one can tell if you are a dog, nerd, wearing pajamas, blond, brunette, red head, white, black, brown, yellow, Christian, Jew, Muslim, African, Northern European, Russian, Mideastern, Asian, Hispanic, etc., rich, poor, handicapped, handsome, ugly, too thin, too fat, popular, socially awkward or if you are living with 1, 2 parents, natural or not, previously were an academic success or failure, were good or bad at other academic subjects, were good or bad at athletics, what your sex or age is, are pregnant or have been pregnant, have a bad reputation from siblings that went before, have an address in a good or bad neighborhood, etc. So, study the materials via the Internet, take the tests, submit the scores, and totally avoid any issues about social this and that. Don't look for a simpler solution.
Yes, the cry will go up: "But, but, but, this is too narrow and omits oceans of adult face time just so totally CRUCIAL for social and civics maturations ....". Yup, and also even more crucial for the teachers unions, really high real estate taxes, nearly meaningless or wasteful and even devastating social and athletic, but not academic, competition, making a king and queen out of the quarterback and head cheerleader and social serfs out of everyone else, degenerate pop culture, bullying, etc. That system IS especially good for getting all the students started on drugs and getting the sixth and seventh grade girls sexually active.
Besides, here are some facts of life about learning, especially relevant on Hacker News: Mostly people, for their careers and everything else, have to keep learning all their lives. Nearly all of this learning is self-directed, self-taught, and pursued independently or largely so. In particular, for all of the history of the US computer industry, nearly all the learning in the labor force has been self-taught -- assembler, Fortran, Cobol, APL, PL/I, Knuth's The Art of Computer Programming, SQL, C, C++, Dijkstra semaphores, communicating sequential processes, the fast Fourier transform, Java, HTML, MemCache, .NET, etc. So, students who can, should get started on such learning ASAP.
Yes, next the cry will go up: "But, but, but, what about students who for any of a long list of reasons would fall through the cracks and, thus, create problems for decades for themselves and the rest of society? These students need the help of good teaching."
Fine: I did indicate that there should be multiple paths to the tests.
We can't hope to make much progress arguing with the teachers unions or by pursuing arguments about nature versus nurture or genetics. So, for the whole debate, the whole issue, we need to get some clarity, simplicity, alternatives, and competition. So, for these ends, start with some good tests, e.g., from ETS. Then have multiple, competitive paths to the tests. In particular, one of the paths should be very high quality learning materials available via the Internet.
For the teachers unions, let's get to see how well their students do on the tests against students with home schooling and the Internet!
Or, for an example, if want to take part of the movie at the intended face value, then consider the students and parents so eager to get the students into charter schools: Such eagerness should serve well for a student with a computer, using high quality materials on the Internet, working through the materials, doing the exercises, taking the tests right along, AND seeing right then how well they did.
Look, guys: How to learn such material just ain't too obscure: The learning is NOT a spectator sport; all that's required is just working through the material which should be easy enough for any eager student or for any student with determined parents! "Sandra, do well on the test on integration by parts and THEN you can gossip with your friends, go shopping for a new dress, and watch your chick flick. Do really well and you can get two new dresses and two chick flicks. No, you can't go park until 3 AM with that college boy you met at McDonald's.".
E.g., nearly everyone on Hacker News worked through Kernighan and Ritchie with its overly succinct writing, sparse content, the tortured, obscure, idiosyncratic syntax of C, no details on the internals of malloc(), way too little detail on stack usage and its storage usage, too little in exercises, no tests right along, no final test at the quality level of the ETS, and still got it. The required learning work just ain't too difficult. We know we can do such work, for C, high school and college math, science, languages, history, etc.
Let's cut the nonsense, see the tests, prepare for the tests, get on to more, and cut out all the social this and political that.
This time the cause is bad teachers protected by the teachers union. I'm all for busting up the teachers union, but I don't for a nano second believe that bad teachers are really the cause of what is being observed.
Instead, what's being observed is deliberately made up, mostly nonsense, mostly to spend still more money on K-12 education which, with some irony, would help the teachers union!
We went through all this just 17 months ago with
McKinsey's report, The Economic Impact of the Achievement Gap in America's Schools, April, 2009,
then but no longer at
http://www.mckinsey.com/clientservice/socialsector/achieveme...
and
supporting materials (PDF - 1.0 MB)
then but no longer at
http://www.mckinsey.com/clientservice/socialsector/detailed_...
Cover to cover it is clear that the main concern of this, apparently pro bono effort by some dedicated McKinsey staff members, was to scream that US K-12 educational achievement sucks while what they really meant was that US Black K-12 educational achievement sucks. So, that pro bono effort was to save the US Blacks.
Throughout the two long PDF files, they omitted any view of the elephant in the room until they gave a glimpse on page 26 of
supporting materials (PDF - 1.0 MB)
where they did a little cross-tabulation, that is, started on the main technique in looking for causality in social-economic data.
So on page 26 we finally get the US students, White, Latino, and Black, compared separately with students in many other countries in the world.
So, wonder of wonders, the ranking goes:
Finland
Hong Kong
Canada
Japan
Australia
US whites
Korea
Germany
United Kingdom
Switzerland
Ireland
Sweden
The US Latinos? Just above Chile.
The US Blacks? A little ahead of Indonesia and Argentina.
Is there any question how the US Blacks would do against, say, Ghana, Liberia, Nigeria, Kenya, Uganda, Zaire, Zimbabwe?
So wonder how US students of Finish descent would do against Finland, Asian descent, against Hong Kong, Canadian descent, against Canada, Japanese descent, against Japan, Australian descent, against Australia? The study didn't say. Ah, wonder why? Why oh why? Maybe they followed the old rule, "If you don't think you will like the answer, then don't ask the question.".
So the study mostly tried to obscure the obvious cause: For the data they considered, the main cause of good performance is country of origin. Right: In particular, parents are a still more important cause, and it might be possible to dig still deeper.
Then the main reason for the low average US ranking is that the US is a diverse country and the other countries are homogeneous.
So, Virgina, this statistical pattern has NOTHING to do with education or US education: Instead, take any 100 homogeneous populations, of anything -- kitty cats, puppy dogs, or billy goats. Then pick a measure, any nontrivial measure you want from length of hair, body weight, running speed, jumping ability, whisker length, nearly anything at all. Then from the 100 homogeneous populations, make a diverse population and apply the measure. Then rank all the 101 populations. Presto: Wonder of wonders, the diverse population will be ahead of some of the homogeneous populations and behind some of the others. Obvious. Trivial.
And that obvious, trivial observation is so far ALL the screaming is about. How 'bout that.
So, if want to tar and feather US education, which for other reasons I'm plenty eager to do, will build the fire to warm up the tar right away, then compare US education with apples against apples. So, compare US education on people from Finland with Finland education on people in Finland. In this way have compared US education with Finland education. Then rinse, repeat for Hong Kong, Canada, Japan, etc.
But the semi-, pseudo-, quasi-smart, partly objective McKinsey team last year and the movie people this year showed their grasp of education: All they showed was the same thing would see from 101 collections of kitty cats measured on length of whiskers.
There's another big point: Apparently actually there isn't much in K-12 in the public school systems of any of the countries so that at most not much is missed.
Another point: Let's compare college education and, there, GRE scores. Bet the top 100 US colleges and universities do quite well, thank you.
Another point: Let's compare graduate education and research progress -- as we know, the US by itself totally blows all the rest of the world off the court and out of the arena.
Another point: K-12 really should amount to something. So let's move to that.
Another point: It may be that the US should take vocational education as seriously as, say, Germany does. We should consider that.
Another point: In the classic Democracy and Education, John Dewey made a big point: He defined education as the passing from one generation to the next and mentioned that what gets passed is mostly just what was there, both good and bad, with maybe a little improvement at each passing. So, without some quite special efforts, tough to take teachers from the middle third of US educational accomplishment and have them pass on only the top 10% of educational content.
Another point: Once again we see that high accomplishment is mostly the responsibility of individuals and where their best help is from their own families. So, enter the promise of home schooling.
Here's my approach to improving US education (listen up billionaires):
First, set up educational certification in nearly all subjects of interest from math, physics, and chemistry to auto body repair, grass mowing, all the way down to, say, computer science, and from there way, way down to, say, programming in C++, if could find anyone so weak minded. The CEEB tests should be a good start.
Second, get most colleges to agree on the required subjects to be ready for college work.
Third, fund some efforts at educational materials and programs to prepare students for the tests. Of course should try to make heavy use of PDF files, video lectures on YouTUBE, Web fora, tutoring, etc.
Then, let free enterprise take over and develop still better educational materials.
For K-12, leave that to local school boards: Some will get it and let well motivated and guided students just pursue some of the better sets of materials and, then, be ready for college work at age 10 or 12.
Then have those 12 year olds compete with 12 year olds around the world.
Here we cut out a lot of nonsense: If someone wants to learn, then sit in a cubicle with a computer, work through some of the best educational materials, take the test, and see how well they do, and then all the credit and/or blame is just theirs.
For varsity athletics, f'get about it except for the cheerleaders -- wouldn't want not to have the cheerleaders, a crown jewel of the US educational system!