Why Asian Students Still Outdistance Americans

Harold W. Stevenson

Asian/American competition has spread from the marketplace to the classroom. A decade of comparative education research suggests why American students aren’t holding their own.

Members of our research group recently asked several hundred fathers of elementary school children in the United States and in China the following question: “Let’s say that your child took a math test worth 100 points. The average score was 70. What score do you think your child would get?” Both groups were confident in their children’s ability. The averages given by the fathers of children at grades 1, 3, and 5 were much above the hypothetical average of 70. Chinese and American fathers alike gave estimates in the mid-80s.

When we asked, “What is the least number of points you would be satisfied with?”, American and Chinese fathers diverged. Chinese fathers would be satisfied with a score that averaged 8 points higher than their first estimates; American fathers would be content with scores 3 points lower than the ones they expected their child would receive.

This phenomenon is not limited to fathers nor to mathematics. We have also asked about reading, and we have sought the opinions of mothers and of high school students. Our respondents have been representative samples of parents and students in Chicago and Minneapolis, in the Chinese cities of Beijing and Taipei, and in the Japanese city of Sendai. Results have been consistent. Americans will be satisfied if the scores are similar to or lower than those they expect; Asians will be satisfied if the students do better than what is expected.

We posed the following question to 11th graders in the three countries: Several recent studies have compared students’ school achievement in different countries. One study compared the performance in math of high school students from eight industrialized countries. Among the eight countries where do you think students in (the respondent’s country) ranked in math?

Seventy percent of the Japanese students, 48 percent of the Taiwan students, but only 5 percent of the Americans thought their country would be in first or second place. Their teachers responded similarly: 85 percent in Sendai, 56 percent in Taipei, but only 16 percent in Minneapolis placed their students in the top two countries.

Clearly, American students are aware of their peers’ poor showing in international studies. Nevertheless, when compared to students in Japan and Taiwan, Americans expressed the greatest confidence that they were performing as well in school as their parents and teachers wanted them to—and up to their expectations as well. In short, they believed they were meeting the standards of their society.

Parents’ attitudes were in line with students’ beliefs. During the past decade we’ve asked hundreds of parents about their children’s academic achievement (Stevenson and Lee 1990). Were they “very satisfied,” “satisfied,” or “not satisfied”? The results have been nearly identical. Over 40 percent of American parents say they are very satisfied. Less than 10 percent of the Chinese and Japanese express this view.

The results of our interviews all point to one conclusion: Despite the emerging awareness that American students are not competitive with their peers in other industrialized countries, American parents, teachers, and students hold markedly lower standards for academic achievement than do their counterparts in Asia.

Why the Discrepancy in Standards?

American parents can consult growth charts to decide whether their children’s physical development is proceeding normally; they can refer to behavioral norms to evaluate where their child stands in terms of language, social, and intellectual development. No such norms or guidelines exist in the area of academic achievement.

Compounding the difficulty is the fact that American grading systems convey only a vague message of how to judge academic progress. A star or a smiling face on the paper of an elementary school student, or the “S” or “S+” on the paper of a high school student fails to convey a clear message of how the student is performing. Parents, deprived of norms and clear evaluations, find it difficult to establish standards. Should their 5th grader be able to add fractions with unequal denominators? Should a 7th grader understand the grammatical classification of words? Parents on their own cannot answer questions such as these. Nor do they have accurate information about how their child is doing compared to other children.

Asian parents, in contrast, have ready access to such information. National curriculums define what is expected of children at each grade level, and textbooks conform to these standards. Ministries of education set high standards, but not so high that the average student, with appropriate instruction and practice, is unable to meet them. Also, grading systems provide numerical scores by which children’s status relative to these standards and to other children in the classroom can be readily discerned (Stevenson 1991).

The U.S. does not have a national curriculum and is unlikely to have one soon. States, districts, and even individual schools establish their own curriculums. They decide the way subjects are organized, the rate at which topics are presented, and the frequency with which they are discussed. Innovation is prized, and year-old curriculums may be rapidly displaced by still newer ones.

Perhaps in reaction to this disarray, the current movement to design national guidelines for education is meeting strong approval. Guidelines define what the nation’s students should accomplish at each grade level. Such guidelines, with their implicit standards, are not being designed by a governmental agency, such as the Department of Education, but by national organizations. For example, the guidelines of the National Council of Teachers of Mathematics are often praised as a model for other organizations. Once the guidelines have been established, local governments and school districts must decide how to devise curriculums that will enable students to meet these standards.

Where Do American Students Stand?

Is all this concern about standards and the need for educational reform merited? Some recent critics do not acknowledge that American schools are in trouble. They dismiss the results of studies comparing the performance of American students with those in other countries because of what they consider to be defects in methodology and interpretation. The tests used in cross-national studies are not fair, critics suggest, because they include material the students have not studied and because the sampling of students is not identical in all of the countries included. Others question the fairness of comparing scores when the amount of time children spend in school varies widely among countries. The last point may be less a criticism of the comparative studies than an explanation of why American students do not perform well. The appropriate response to the other two criticisms is to construct tests tapping material covered in the curriculums of all the countries involved and to follow identical procedures in sampling the students who will be tested.

Our research group has attempted to avoid these methodological traps in the studies we’ve conducted during the past decade. We devised our own tests in reading and mathematics on the basis of detailed analyses of the textbooks used by the children we studied (Stigler et al. 1990). To obtain comparable samples of students, we chose locations where there is universal elementary school education and where we could apply the same sampling procedures. We could do this because we included only a small number of locations and purposefully chose those with cultures and languages with which members of our research team were familiar.

The results are in line with those from the larger studies. For example, in spring 1990 we tested several hundred 5th graders in Minneapolis, Taipei, and Sendai in the same schools we had visited in 1980 and again in 1984. Experienced examiners administered the same reading and mathematics tests in one-on-one fashion during each of these visits.

Asian superiority in mathematics was evident in 1980 and continued in 1984. Would the subsequent years—ones of heightened interest in improving education in the United States—be accompanied by improved scores by American students? The answer was negative. The American 5th graders answered only one more question correctly in 1990 than their counterparts had answered 10 years earlier. Japanese students retained their superiority; students in Taiwan actually widened the gap between their performance and that of the American students. In reading, the results were also dismal. In 1980, American students outperformed their Japanese peers, but Chinese students received the highest scores. By 1990, Japanese students showed a marked gain in their scores; American students had slipped to third place.

In the intervening years, we tested representative samples of kindergarten children and students at grades 1, 3, 5, and 11. Chinese and Japanese students excelled in all studies of mathematics achievement—whether they were tested with routine problems in computation and geometry or with novel problems requiring creative solutions, and in areas such as measurement, estimation, and mathematical concepts and operations.

Comparing reading scores is more difficult when the writing systems, grammar, and content of textbooks differ. Nevertheless, when students were tested with words that were at-grade or below (words that they had studied in their readers), Chinese and Japanese students outscored Americans. When the words were above-grade, an important characteristic of writing systems became evident. Words written in an alphabet can be sounded out, but the characters used in Chinese and Japanese cannot. As a result, more American students were among the best readers, but they were also over-represented among the worst.

Clearly, an achievement gap exists between American and Asian students (Stevenson and Stigler 1992). Part of the reason for this gap is that American students, their parents, and their teachers maintain unnecessarily low standards for performance. But raising standards to improve academic achievement is only a first step. More difficult is the problem of putting the guidelines into practice.

American Teachers Speak

The ultimate success of any effort to raise standards depends upon teachers and the support they receive from parents. The teachers we have recently interviewed appear to be seeking fundamental changes in their duties. They are not opposed to establishing competency examinations for promoting students or to giving more homework. They believe that more time is needed for mathematics, science, and language arts—the core subjects of any curriculum. Their opinions are mixed about lengthening the school day and year. The four frequent complaints American teachers expressed are that they are overworked, have too many responsibilities, lack knowledge in the basic fields, and are not taught how to teach and manage students. These complaints have some validity, especially when their jobs are compared with those of the typical teacher in Asian schools.

Workload. American teachers spend most of their time at school teaching. Among the Chicago elementary teachers we interviewed, just 7 percent of the sample of 112 teachers have only one or two preparations each day. None of the 65 Beijing teachers has more than two classes to prepare for each day. Classes might differ on different days, but generally the Beijing teachers in grades 1 to 3 teach reading, mathematics, and one other subject. After grade 4, they teach reading or mathematics and a second subject. High school teachers in Beijing typically teach only one subject two hours a day. The rest of their time is spent preparing for class, working with individual students, consulting with colleagues, and correcting papers. Even though they spend longer hours at school than American teachers, the schedules of Chinese and Japanese teachers allow them to enter the classroom with a level of energy and a degree of preparation seldom possible for American teachers.

Responsibilities. American schools face an ever-growing set of commitments arising from the expectations of parents and society. Teachers are expected to counsel students about avoiding disease, preventing fires, and choosing careers; to coach winners in everything from sports to musical comedy; and to serve as parent-substitutes, disciplinarians, and psychotherapists. Academic standards may be higher in Asia, but the definition of teachers’ roles and obligations is more limited. In Asia, teachers’ primary function is to teach effectively and produce high levels of achievement among their students. Other functions are the responsibility of parents or professionals outside the school setting.

Basic training. Brief inquiry supports the American teachers’ complaint that they lack knowledge in basic subject areas. For example, I recently asked the head of the mathematics department in a large Eastern district about the qualifications of elementary school teachers. The typical teacher, I was told, had two college courses in mathematics—only slightly more than the average number of courses taken in mathematics education.

Teaching skills. The typical model followed in American schools of education is for the graduating senior to spend a semester as a student teacher. The aspiring teacher is assigned to an already overburdened teacher who receives little compensation or recognition for taking on the novice. A graduate teaching assistant, or even the professor of education, may visit the class a few times to observe the student’s progress. After graduation, new teachers are left on their own to sink or swim.

Compare this experience with what happens in Japan. The major function of university classes is to provide an education in basic academic disciplines and to introduce the history and practice of the profession of education. Learning to teach occurs in an actual classroom. Master teachers, relieved of other responsibilities for a year, devote full attention to working with new teachers. In addition to having a mentor, the new teacher joins others who are teaching at the same grade level. Participating in group discussions about developing and using effective instruction further enhances the new teacher’s skills. Supplementing these small-group activities are frequent workshops, both within the school and in the city and prefecture. In Japan, learning how to teach interesting, effective lessons to motivated students is expected to continue throughout a teacher’s career.

Redefining the Teacher’s Role

The United States can introduce higher standards, longer school days, greater choice of schools, new curriculums, and other innovations, but dramatic improvements in academic achievement will not likely result without a restructuring of the teacher’s role.

Effective teaching requires time for preparation and knowledge of what is to be taught and how to teach it. Teachers need the time, as well as the training and freedom from other responsibilities, to help their students master the subject matter. They need opportunities to observe and discuss excellent teaching and to share their knowledge with other teachers. These are not unreasonable goals. But accomplishing them may make higher standards something we can realistically expect to meet, rather than being a source of further disappointment.


  • Stevenson, H. W. (1991). “Japanese Elementary School Education.” The Elementary School Journal 92: 109–120.
  • Stevenson, H. W., and S. Y. Lee (1990). “Contexts of Achievement: A Study of American, Chinese, and Japanese Children.” Monographs of the Society for Research in Child Development 221, 55: 1–2.
  • Stevenson, H. W., and J.W. Stigler (1992). The Learning Gap. New York: Summit.
  • Stigler, J. W., S. Y. Lee, and H. W. Stevenson (1990). Mathematical Knowledge of Japanese, Chinese, and American Children. Reston, Va.: National Council of Teachers of Mathematics.