- a review and commentary by Paul MacFarlane, written in November 1999 of:

Learning from Asian Schools
by Harold W. Stevenson
published in Scientific American, December 1992
(the original article is not available online - only articles since 1993 are online

Everyone has heard that American students lag behind Asian students in math and science. You've probably wondered why. So did Harold Stevenson, a psychology professor at the University of Michigan. So he and his colleagues conducted a study to find out. And he found out. The answers are both simple and obvious, and clearly show the challenges America must meet to catch up, if indeed we decide to do so. It's not a matter of knowing how to do it - we actually do know, since many of the teaching styles and institutional structures used in Asia were pioneered in America - but it's a matter of setting priorities and hard decision making on the part of our society as a whole.

Rather than study middle schools and high schools (which usually are the focus of attention in America), Stevenson studied elementary schools, (which are more of a focus of attention in Asia) to find out where the foundations of Asian education and educational attitudes are laid.

If you look at Asian elementary schools, you'll find that they are not at all like the Asian stereotype that we often hear about in America. Such stereotypes come from Asian secondary education, and generally include rote learning, repeated drilling, and (we assume) frantic, stressed-out students. But the basis of Asian education, which enables the students to survive such pressures of secondary education, is laid in the lower grades, where, according to professor Stevenson, "Knowledge is not forced on children; instead the students are led to construct their own ways of representing ths knowledge. The long school days in Asia are broken up by extensive amounts of recess. The recess in turn fosters a positive attitude towards academics."

To reach their conclusions, Professor Stevenson and his colleagues studied several elementary schools from each of four locations:

In each of these 51 schools, they studied four classrooms, so 204 classrooms altogether.

They developed cross-cultural tests for math and reading, and found, indeed, what everyone knows - that American students, in general, fall further and further behind the pace set in Asia as they progress through the grades.

So why?

The answers can be grouped into five areas:

  Family Attitudes and Priorities

Asian families were not as satisfied with their child's neighborhood school as Americans were. At the same time, they were more likely to attribute success in school to a student's hard work, whereas Americans thought that innate talents and abilities, (which were therefore beyond a student's control) had a greater influence. Kids in America had much higher opinions of their own abilities than those in Asia. In America, childhood is thought to be a time for a variety of activities, whereas in Asia, school work is seen as a child's primary task.

Children in America and Asia were asked to fantasize about what they would like if they could have any wish. Over two thirds of Asian kids wished for something related to education and educational success. Only one in ten students in America wished for something related to education, choosing instead wealth, material objects, or fun fantasies, such as a trip to the moon.

Clearly, the life of the Asian child is much more centered on school than the life of an American child. Asians parents are much more involved and critical of their local schools than parents in America . This situation goes a long way to explain why Asians have more success in school than Americans. But these attitudes don't exist in a vacuum. There are structures in Asian societies and educational systems that support and encourage them. The rest of Stevenson's article deals with these structures.


In Asia, educational standards are set by the state. One can argue about the reasonableness or appropriateness of particular standards. One might argue that they test in a way which may not promote true education. One could argue that they stifle creativity in learning. One can even decry the lack of local control they represent. But they exist, and there's no getting around it.

Furthermore, they are clear and quantifiable. And herein lies their strength, which in Asia outweighs the above disadvantages. Every group that is part of the educational enterprise - teachers, parents, students, and the community - all understand exactly what is required to pass them. This has the rather powerful effect of putting all these groups on the "same team," so to speak, where the adversary is the test itself.

And that "teamwork" results in some very different social dynamics, particularly between teachers and students. In America, the teacher is the educational authority who challenges students directly, and is therefore not simply a source of knowledge, but also a source of confrontation and conflict. And students often treat them accordingly. In Asia, on the other hand, teachers are more like coaches, helping students to achieve high test scores the same way a track coach might help his team to break speed records.

The tests also provide the basis for effective parent and community involvement in the schools, since parents and community members all have access to the same understandable goals.

The School Day

The School Day in Asia also promotes this sense of teamwork, and makes school much more meaningful in the lives of individual students than it is in America.

It is well known that the school day in Asia is much longer than in America - it's about eight hours. But what makes it longer are mainly recesses and activities. To be sure, Asian students do a whole lot more homework than American students do, and they also spend a lot more time reading for pleasure, but instructional time is about the same. At school, they have much more time for socialization with their schoolmates.

Whereas American recesses and lunch breaks tend to be as few and as short as possible, Asian recesses are frequent, and long lunch breaks may allow the students a leisurely commute to home and back (since there may not be a cafeteria at school, anyway). Whereas American students generally go home as soon as the last bell is rung, Asian students stay for clubs and more activities. In general, about a fourth of an Asian student's school day is spent in non-academic activities. All this recreation makes school a more enjoyable place, and allows one lesson to digest before the next lesson takes place. Children are more rested for class, and pay attention to the teacher more often. In general, Asian students enjoy school more than American kids, and have fewer physical symptoms to indicate psychological stress or tension, and many fewer problems getting along with others.

Teacher Training and Support

Although the school day is very long, individual teachers are responsible for a much smaller part of it than their American colleagues. Indeed, when most Asian teachers were told how much an American teacher had to do during the day, and what resources were available to do it, their response was usually simple disbelief.

In Asia, teachers are usually responsible for about three hours of instruction a day. In general, teachers were in charge of classes for somewhat more than half (60%) of their school day. What did they do the rest of the time? Prepare lessons, grade papers, work with individual students and confer daily with their colleagues.

Indeed, most teachers in Asia share a workroom, separate from their classrooms, where they have a desk and office space. They spend most of their non-teaching time there, and frequently consult with one another and engage in group discussions over lesson plans and instructional techniques. It makes a lot of sense for them to do this because of the state-imposed standards and state-imposed curriculum, which ensure that all instructors at a particular grade are generally teaching the same thing at about the same time. Such workrooms do not exist in America, where each teacher's desk is in his/her own separate classrooms, and so consultation between teachers must be planned for outside the normal flow of daily activities. Generally American elementary school teachers spend their nights and weekends working while their Asian colleagues use this time for rest and recreation.

The collegial atmosphere of Asian teachers' workrooms also makes it quite natural for experienced teachers to mentor new ones, and, indeed, in Asia, most teacher training is done on the job. Again, in America, such mentoring, if it exists at all, must be carefully planned and arranged, and teachers are "trained" by putting them in university classes.

In Asia, then, the art of teaching is nurtured and developed by the system. The most important characteristic of an instructor is "clarity." Since students tend to be treated as a group, time and resources are devoted to improving instruction. In America, "sensitivity to the needs of individuals" is perceived as the most important characteristic of an instructor. Time and resources are devoted to evaluating the needs of individuals and meeting them as well as one can in the limited time allowed.

Teaching Methods

Those who hold to the Asian stereotype of backbreaking rote-learning for students will be most surprised when they learn that Asian elementary schools already practice most of the reforms in teaching methods that have been developed and pioneered in America and Europe.

Indeed, the land where children are most often engaged in boring, unaided "seat work" is America, not Asia. In fact, our system encourages this, because it's often the only way a teacher can get time during the instructional day to get a head start on planning and grading. This also means that American students are much less likely to have meaningful personal interactions with the teacher than their Asian counterparts, even though classes in Asia are larger. Asian "seat work" is frequent and brief, rather than long and sustained, and the teacher generally monitors closely what the students are doing while they're doing it.

In Asia, students interact much more frequently with real concrete objects (what Americans call "manipulatives" and what many kids call "toys") in learning, particularly in mathematics classes. In math, they are much more likely to use "word problems" which relate arithmetic to the real world, rather than rote drill. Indeed, math students in Asia often construct their own word problems, an exercise only rarely found in American schools.  They also interact with each other much more in traditional American classes.

Asian elementary school teachers are also more likely to teach through discussion rather than lectures, and to spend more time teaching students how to deal with academics (what we in America call "process"), such as how to make smooth transitions from one activity to another.

In America, reform movements of various kinds have been encouraging teachers to adopt such methods for decades, but because of the ways we allocate resources, it is often not practical for teachers to actually do so, even if they are willing (and I think most American instructors would agree that such methods are "reasonable and effective").

My own 2¢

Asian and American schools are different in culture and in allocation of resources, which makes it unlikely that the Asian system will be adopted in America any time soon, despite the fact that common sense and evidence from testing shows that the Asian system works better. Still, it's a system worth studying and considering.

We as a society should take an especially hard look at Asian elementary schools. The elementary school years are critical years. My own feeling is that kids make up their minds by age 11 or 12 whether education and academics are worthwhile pursuits. One of these days maybe I'll have to go to graduate school, make this my thesis study and find this out for sure, but it seems to me that by seventh grade, the decision to drop out of school is already made. And having made that decision, the student drops out mentally, if not physically, when the opportunity arises.

All this is to say that good foundations in the early years will carry students through difficulties later. As we look at American "test scores," we see the scores dropping further behind relative to Asian students as kids go through middle school and high school, but probably this is the inevitable result of Americans missing out on a good foundation. It is for this reason that I think that if we reformed our elementary schools, we would not need to reform middle and high schools. After all, I don't think we as a society want to put our young people through an Asian-style "exam hell," nor do I think we need to. Also, it's my experience that middle school and high schools are much less progressive than elementary schools generally (and there are structural reasons for this), so elementary school reform has a greater chance of success.

Reform especially needs to happen in two areas -- Standards and Teaching.


Any real reform of elementary school education requires a whole lot more than changing just the schools themselves. Each child in an elementary school is actually not an individual, but one representative of a family (however you define family). True reform is not possible without effective involvement of the rest of the family members. Happily, it is my experience, as I've observed through the years, that families want to be involved, and in fact, my own experience shows this desire increasing, not decreasing from its low point in the 1970s and 80s. The key, then, is to find ways to help them become involved.

To this end, the standardization of goals would help out immensely. It would also have the effect of putting everyone on the "same team," as noted above in Stevenson's article.

I think that we, as a society, already feel the need for more rigorous standards, but the problem is how to implement and assess them. Most teachers know that pen-and-pencil multiple-choice tests are inaccurate at best in showing how much education an individual has learned, or in predicting how well he or she will do in the job market. All they're good for, it seems, is predicting the success of grouped populations of students in college and graduate school.

On the other hand, such tests generally have clear and well-defined goals, which are understandable to all, even if they are a bit misguided. And perhaps for this reason, the benefits of using them (that is, giving parents tools to be more effectively involved) might outweigh the disadvantages (that they are inaccurate in showing an individual's real progress, and, in some sense, are destructive to true education - the skills and knowledge that can be used later in life)

Of course, it's possible to develop standards that do not depend so heavily on multiple choice testing, yet are still clear and understandable to all. However, there are forces arrayed against such development. First, most other ways of standardized assessment are much more expensive. Second, they are different enough from the "tried and thought to be true" that politicians out to make a career for themselves can shoot them down through raising fears in the voting populace. This has, at least, been the experience in California. Given these forces, then, it might be better to sharpen the #2 pencils, grin and bear it.


The second major reform we need in elementary schools a greater focus on teaching. And by "teaching," I mean the instruction provided by teachers in a classroom. It has been my opinion for at least a decade now that "teaching" is the most neglected part of our elementary school education system, except, perhaps, for the teachers themselves.

The problem is not with the teachers as individuals. Everyone knows that teachers don't pursue their careers in order to become rich, but because the personal rewards are so great. Although unmotivated teachers exist, as they do in all professions and trades, the vast majority are motivated self-starters. At least that's been my experience with elementary school teachers over many years.

The problem is that teachers are not given the education, training and resources that they already want in order to do their jobs.

This goes far beyond the well-known fact that most teachers spend lots of their own money on classroom and students. It goes far beyond the fact that teachers generally have to provide their own educational opportunities, whereas most other high-tech industries provide it for their employees. It also goes far beyond the dismal condition of the physical plant in many schools.

It starts with proper teacher education about the way children learn things. Although the brain's ability to learn is far from "figured out" yet, considerably more is known about it than most people realize, and much of that knowledge is directly applicable to instruction. As an analogy, one might say that the understanding of learning processes is about where medicine was in the early part of this century. There's a lot left to learn, but there's already enough known for making informed decisions about educational issues. Perhaps if more teachers were educated in developmental psychology and epistemology, they might successfully weather and/or resist the constant swinging back and forth of fads and styles that has characterized teaching methods in this country up until now.

But this kind of education doesn't come cheap, nor can one learn it in a year, and it can especially not be learned in the day or two of "inservice" instruction that school districts might provide to their teachers every year.

So on-the-job mentoring would be the only way to impart it to any significant number of teachers. Since it's best learned in a situation where teachers can study and research young children, elementary schools are the best setting to provide such education, anyway.

Teachers also need resources of time and space - a workspace that's orderly and separate from students, and enough time during the day to properly prepare lessons. They and their students need technology tools (i.e. laptops and computer networks, etc.) on the order of those available to most businesses. Finally, the teachers need a smaller teaching load (fewer hours teaching their class and more hours in preparation, evaluation, and helping students with difficulties). Students need longer hours in school -- not more instructional time, but more frequent recesses and activity times. That way, everyone will be rested enough for their best effort in teaching and learning.

Finally, elementary school teachers need the relative freedom from interruption enjoyed by their colleagues in secondary education. Many elementary school classrooms are "revolving doors" where students constantly enter and exit throughout the day to attend special programs. Such "pull-outs" (as they are called) can only be justified when students can't adequately function in their normal classroom, anyway. When random bunches of students constantly enter and leave to receive "enrichment" lessons, or to provide labor around the school (such as in the cafeteria), it not only disrupts their own learning, but it also disrupts any reform of teaching methods that involve group work, use of authentic materials, manipulatives, etc. It also demonstrates rather physically that a classroom teacher's teaching is not a valued commodity in our present system.

These reforms face a lot of stiff cultural resistance, but most importantly, they would probably cost one heck of a lot of money. That's the reason for my pessimism that they'll be implemented any time soon. (Paricularly in California - the richest state in the richest country in the world, which is usually last among the 50 states in how much money it spends on education). On the other hand, the day may come when our society may decide that it wants to either put up or shut up, and really do something about the educational systems that it so often complains about. Maybe then, it will happen.

Further Reading - Internet Links

A bit of searching around the Internet revealed more information about Professor Stevenson, his colleagues and their work.

Stevenson put together a video about all this (but I don't know where it's available) intitled The Polished Stones: I found some points about it in a Google cache from the University of Alberta. To see them click here.

The Third Internatinal Math and Science Study in 1996-1998 has continued. More recent results can be found here.

Colleague James Stigler made a videotape study comparing eighth-grade math teachers Asia, America and Europe. The Schafer Park webmaster has a copy of it if you'd like to see it. In another Google Cache I found a transcript of a talk James Stigler gave in 1999 and saved it on this site. Click here to read it. 

The article "Learning from Asian Schools" is summarized from a book that Stevenson and Stigler published in 1992. It's called:

Stevenson, Harold W. and James W. Stigler. The Learning Gap: Why Our Schools are Failing and what we can Learn from Japanese and Chinese Education. New York: Simon & Schuster, 1992. ISBN 0-671-88076-4. You can buy it at Amazon.com by clicking http://www.amazon.com/exec/obidos/ASIN/0671880764/qid=943020398/sr=1-2/002-7805482-3854627

Recently (1999), James Stigler published a book of lessons about teaching, which he distilled from the video study mentioned above:

Stigler, James W. and Hiebert, James, The Teaching Gap: The Best Ideas from the World's Teachers for Improving Education in the Classroom.Free Press, 1999 ISBN: 0684852748

You can buy it at Amazon.com by clicking:

A list of works (in pdf file format) by James Stigler can be found at: http://www.psych.ucla.edu/Faculty/Stigler/publications.pdf

There is a trend in the United States now to eliminate elementary school recesses completely.  Luckily, there is a newer trend to re-instate it where it have been lost. 

Some arguments in favor of recess for kids can be found at http://www.ipausa.org/recess.htm.and on this other page that's linked to it.

An essay based on a lecture given by Gail Benjamin on October 7, 1997, which draws on her book Japanese Lessons: A Year in a Japanese School through the Eyes of an American Anthropologist and Her Children (NYU Press, 1997). It's interesting that Ms. Benjamin noticed similar things to what Stevenson and Stigler noticed, even though there's no connection (that I know of) between their works.

Return to the Thoughts from the  Schafer Park webmaster page