Look at programming course. East Asian students don’t seem to have an edge over Americans.
Look at math. I believe East Asian students have an edge, but sometime during the 4Y in college (possibly starting in high school), in STEM courses American students somehow catch up to close the gap.
In the big picture, the math advantage is short-lived
Many say the Chinese students (including Singapore, HK …) are the smartest compared to Grade6 counterparts in other countries, largely due to their math standard. (More than other subjects, basic math is associated with IQ.) China students are traditionally known to be very strong in math “problem solving”, from primary school till pre-U level. If we take a random sample of 10,000 age-16 students from China to form a “China sample” and use a standardized test to grade them in A/A-/B+/B/C (not numerical score [1]), and repeat the same on a U.S. sample, and a Japan sample, and a German sample, … we get 150 national sample and a score for each country. Probably China team will be stronger than most teams. (I have reason to believe Russian team, Korean team, Singapore team will also be strong.)
[1] For the statistically inclined, the “score” is not a number, but rather a histogram. We can compare two histograms by their IQM or their median. We can even compare their mean since there can be no outlier in a letter-grade. Tthere’s no real difference between a score of 881 vs 882 — such granularity is completely misleading.
Now, how about programmers? If I take a random sample of 100 China-trained programmers (below age 28 [2]) and grade them on a standardized coding test, and repeat the same on a random U.S. sample or Australian sample… I believe China team will Not be so strong. Why? Below is my educated guesses.
(I will use “American students” as a proxy for students in developed western countries, where emphasis on math practice is much lower.)
My Answer — many American students before age 17 don’t bother to practice math problems as much as their China counterparts. (My son is this type. My sister’s Australian boyfriend Chris was this type when he was young.) They don’t see a purpose. If an American student and a China student are equally intelligent then the China students would end up practicing a lot more on math problems, and score higher. For a programmer, the American is equally motivated as the Chinese, so they put in similar effort.
My Answer — Also, China education system is academically more competitive. China’s top schools practice competitive enrollment. In American schools, before Grade 9, most exam scores don’t matter that much. Even at Grade 12 (last year of Pre-U), exam scores matter much less than in China. Most top colleges use national test scores as one of many selection criteria. Therefore, China students not only put up with a lot of practice, but also work hard to understand the concepts better, so they can cope with hard math exams. If we look at Google/Facebook coding interviews, competition is equally high for anyone, so the American programmer also feels motivated by the competition and works hard and competes effectively against Chinese programmers.
There’s a deep cultural factor that’s hard to describe fully — For centuries the Chinese family values education achievements more than other nationalities, esp. the visible signs of achievements like enrollment in a famous school, be it university, secondary school or a primary school. There’s enormous effort by parents, teachers and students to improve on standardized benchmark scores, something the American family don’t care that much and don’t put in so much effort. Therefore, at the middle-school level, the Chinese students have put in far more effort on math practice than the American students. As to programming skill …. well no such cultural factor… no prestige …
| xxxxxxxx |
By age 25, in the tech or scientific sectors, the Chinese young professionals are still strong, but not “head and shoulders above” the other nationalities.
Q: if indeed the China students were technically strong at age 16 but no longer so strong at age 25, then at what age did they start losing their lead? I know it’s a gradual process between age 18 and 25, but please pick one age.
A: I would say in college, around age 20, when other nationalities start seeing a purpose in working hard and /building technical competence/. That’s consistent with my own experience. My friend Jack He pointed out that in college, the Chinese parents can’t control them.
The math subject is the best example. (To a lesser extent, we can talk about science, too.) The Chinese students are strongest in these subjects. Note there’s no such advantage in performing arts, visual arts, sports etc.
Q: what if I grade those high school student samples on a reasonably tough programming test rather than math test?
A: I would think China team would show no visible superiority, since programming is not a subject they practice for. They don’t have 20 years of past exams to study. There’s no purpose in practicing on programming since this skill doesn’t help them get into top colleges.
Q: what if I grade the students on a physics test?
A: I would think China team is still strong if the test is mostly theoretical physics, but their advantage will be smaller than in math test. Math is more abstract than physics.
Q: what if I grade the student sample on a standardized chemistry exam?
A: I don’t know, but I doubt the China team would show superiority. Chemistry is even less abstract and theoretical.
Q: how about history?
A: Not a subject with an international benchmark
Observation — Many China schools have impressive track record in training students on theoretical, abstract subjects like math and theoretical physics.
Observation — For two equally intelligent students, motivation (and therefore effort) is the key differentiator
Observation — China students were motivated to study math at an earlier age. Other nationalities can catch up when they find the motivation.
Therefore, I now believe math (or any other subject) test scores in primary school, middle school, pre-U don’t matter as much as some Chinese parents believe. We have a cultural bias to place too much emphasis on test scores.
The statistical correlation between Grade-10 test score and college GPA is probably not very high. The mediocre American students can often catch up with Chinese students later on, perhaps in college (or after leaving college) and become equally strong technically. The American students’ talent is Not wasted in middle school or high school.
Last Observation — when the China student loses motivation, he would lose the competition. This is a marathon. 龟兔赛跑，可以后来居上。
What I want to focus on in my kids is not academic benchmark, but learning habits, motivation, self-discipline, desire to learn. These attributes had better get built-in at an early age.
[2] I put in this condition because among the U.S. programmers, there are many older ones.
— Why I pick programming as the contrast against math?
In this blogpost, I chose programming because programming skill is highly correlated with math skill. Among all the secondary school subjects, math skill has the highest correlation with programming skill. Is there another mainstream professional skill that’s more correlated with math? statistician? actuary? I’m unfamiliar with them, and programmers outnumber these professions by a factor of 500.
I chose “programming test” also because it’s easily standardized. All programmers are trained to solve the same classic computer science problems. Math and programming are both highly abstract and well-defined, so every programmer can understand a standardized coding question.
I chose “programming” also because it can be self-taught with little outside help. If you have a grasp of basic math, and plenty of motivation, you can read tons of free books and practice on hundreds of problems and improve effectively. I think millions of young programmers/students world-wide do precisely that on leetcode.