[csa]

Hmmm... Im not sure I should respond given your straw man response, but just so others can keep the record straight...

1. PISA is a widely administered test for 15 year olds.

2. TIMSS is a widely administered test for 4th and 8th graders (Advanced TIMMS does "last year of secondary").

3. We are talking about university -- specifically elite universities.

4. I am extremely familiar with education in East Asia (most familiar with Japan), and their teaching of the fundamentals of math in junior high school and high school are in fact excellent. Sadly, this does not translate well into creative thinking. The best math people in Japan are largely limited to a small number of schools (3 or 4 elite ones and maybe a handful of others that have some players).

5. While you are correct that the average quality of education in the US is lower than other places (there are lots of US-specific reasons for this), the education at US elite universities and the US high schools that feed into elite universities is quite a bit higher than the US average. It's tough to find hard statistics on this because it would reveal a class bias in the US (gasp, there is one!), but spending some time in these classrooms and talking to and working with the students from these schools is fairly convincing anecdata.

I'm not sure what you have against US schools and/or US elite universities, but I encourage you to open your mind -- there is a reason why we have a lot of world-class top-rated (by any measure other than cost) universities here.

[bjourne]

1-2. I know the tests are administered to school children. But academic success in school translates to academic success later in life.

3. What makes you think that MIT would be more elite than elite universities in other countries? Talent follows a normal distribution and if PISA and TIMMS show that the American mean is below the average, the American elite must also be below the average. So, if the most gifted American students are admitted to MIT and the most gifted Singaporean students are admitted to the Nanyang Technological University (NTU), the latter university will contain the more gifted students. And if it is true, that you claim, that the quality of the education is dependent on the skill level of your peers, then NTU must be (much) better than MIT.

4. You can't measure creative thinking and the "Asians may be good at math, but they aren't creative" thing is a cliche.

I don't have anything against US schools or elite universities. Rather it would be you who have something against non-elite universities because you are claiming that you can't acquire the same education at other non-brand name universities. My claim is merely that one can excel at any university in any country as long as one puts in the time and effort. Especially when talking about math. You need a book, pen and paper and solitude and you're set.

I would be happy to put my money where my mouth is and run down to my local university with your hard calculus final. :)

[jcranmer]

You make a big deal about how secondary school rankings show that Americans are idiots and use that to guess that university rankings should similarly show that American universities are poor quality.

Which is obviously why only 12 of the top 25 schools, and 4 of the top 4, are American. I'd cite the logic failures in your argument as well, but I think letting actual statistics speak for themselves works better. And FWIW, I'm citing the university rankings that's published by a non-US publisher lest you think it's merely patriotic bias.

[bjourne]

Secondary school rankings show Americans are below the average. I never claimed that the quality of the alumni correlated with the quality of the school. The GP did. My point is that if you want to rank schools by how smart their students are, then no American institution would rank in the top 10. So you need to find a better metric if you want to claim that MIT is best of the best.