Wide Variations in Math, Science Teaching Found in 50-Nation Study
Preliminary findings from the largest international study of its kind to date suggest that there are wide differences between nations in the ways that students are taught science and mathematics.
"What all this variety says to me is that the only way you can make sense of cross-national comparisons of student achievement is in the context of understanding those differences," said William H. Schmidt, a Michigan State University professor who is directing part of the study. "Comparisons without such context are meaningless."
The Third International Mathematics and Science Study, or TIMSS, involves students in 50 countries. In addition to measuring students' achievement in mathematics and science, researchers are studying the curricula and textbooks used in most of those countries and observing classrooms and videotaping lessons in Japan, Germany, and the United States.
The effort is aimed at finding out what works in high-achieving countries, rather than concentrating, as international studies have done in the past, on whose students are "best."
The findings, released here last week at the annual meeting of the American Educational Research Association, come from the first phase of the study, an analysis of the official or "intended" curricula in 48 countries. Researchers presented those preliminary results, but they did not identify the countries studied pending a review of their findings by those nations.
Country-specific information will be included in the final report on the curricula analysis, slated to be released in the fall.
Content and Approach
Investigators at Michigan State University's college of education and elsewhere have thus far pored over data from 242 science curriculum guides, 323 science textbooks, 219 mathematics curriculum guides, and 267 mathematics textbooks.
In math, for example, the researchers found that the space that textbooks devoted to algebra ranges from 2 percent to 3 percent in one country to more than 40 percent in another.
"If you put algebra problems on a cross-national test, I think it is hardly likely that you could claim that all students have had an opportunity to do algebra problems," said Curtis McKnight, a mathematics professor at the University of Oklahoma who took part in the study.
In science, the percentage of textbook pages devoted to the study of rocks and soil at one level of schooling ranged from none to 10 percent to 15 percent. Other topics, such as reproduction of organisms, were almost universally included.
Besides content differences, the researchers also found that the instructional approaches varied.
Textbook analyses suggested that algebra instruction in many countries consisted of teaching students appropriate vocabulary and procedures. In one country, however, such instruction was almost entirely focused on problem-solving and reasoning.
Students are also taught different topics at varying points in their school careers, the researchers said. In one country, for example, students had finished with instruction on proportionality by the time they had reached grade 7 or 8.
In another, students in the same age group were in the midst of learning it. In a third, proportionality appeared in the textbooks for those grade levels but did not appear in the curriculum guide.
In much the same way, clusters of countries also differ in the way they approach math and science.
Gilberto Valverde, a research associate at Michigan State University, noted, for example, that Latin American nations appear to devote more time to teaching middle school-age students about ecology than do Eastern European nations or developed countries, "which are those countries that are proposing these changes in the science curricula," he said.
Researchers do not yet know, however, how much the content that is outlined in textbooks and curriculum guides reflects what is being taught in classrooms.
John Dossey, a former president of the National Council of Teachers of Mathematics, said that question is one of many that researchers may be able to answer in later phases of the study.
"This also may be a chance to talk about whether there is a world-class standard at all," he said. "Or do we have four or five models that are equally good at reaching the goals that you want to reach?"