Bismarck said that there are two things you should never watch being made: sausage and legislation. At a February conference in Washington, D.C., on the Third International Mathematics and Science Study (TIMSS), one of the study's directors, Albert Beaton of Boston College, suggested a third item whose creation one should strive to avoid witnessing: an international test. Hmmm.
Although the media continue to focus on test scores, much of the research community is looking at other aspects of TIMSS, and these appear promising. After all, as William Schmidt of Michigan State University, who is U.S. TIMSS project director, has said, the test scores provide only the context for the study. The real interest is in determining why the test scores turn out the way they do. In the March Research column, I noted that the differences among most developed countries are not large. After sitting through the two-day conference, I wonder if we will end up with more variables to explain the differences than we have points of difference to explain.
For instance, take those top-scoring students from Singapore. At least three variables appear to affect their scores, independent of anything that happens in school. First, while 97% of U.S. students who are age-eligible to be in secondary school are in fact in school, only 84% of those in Singapore are. Second, each day thousands of poor Malaysians commute into Singapore to sweep streets, pick up garbage, and do other low-level service jobs. They return to Malaysia each evening, sparing Singapore the task of educating the children of these workers. Third, some families in Singapore whose children are not making it in the Singapore schools send their children to schools in Malaysia.
From all accounts, Singapore (with about the population of metropolitan Washington, D.C.) is a nation intensely concerned with education. But any nation that can "outsource" its poverty and its low-achieving students certainly has a leg up in the race to get high test scores. Indeed, these two characteristics make Singapore sound like an American suburb - little poverty and few low-achieving youngsters. In this regard it is interesting that a consortium of 20 suburban Chicago school districts paid to have their students take the TIMSS tests. Their scores were second only to Singapore's in math - and second to no one's in science.
In all 41 nations that participated in TIMSS to the end, test scores were associated with both the socioeconomic status of the family and the number of books in the home. Neither finding is much of a shock, but it is interesting to see this common U.S. outcome replicated in 40 other nations.
TIMSS, whose first data were released only in October of 1996, has already produced a new cliche: namely, that our math curriculum (and, to a lesser extent, our science curriculum) is a mile wide and an inch deep. Cliche or no, the evidence for this conclusion is compelling. At the eighth-grade level, American textbooks are often three times as thick as those in other nations, and teachers try to cover everything in them. If one looks at the five most emphasized topics in other nations, these five topics will occupy most of the textbooks and most of the teaching. In the U.S., the five most emphasized topics in textbooks occupy just 35% of the textbooks and an even smaller proportion of the teaching. This is not true for the 15 % to 20% of American eighth-graders who are studying algebra. About 75% of their textbooks pertain directly to algebra, and the rest cover only a few topics.
I should emphasize that, by the end of 12th grade, most countries have covered about the same number of topics in mathematics. But in the U.S. many more topics are introduced each year and are repeated in subsequent years. This could reflect the pursuit of the oft-espoused goal of a "spiral curriculum," but the more descriptive figure appears to be a circle: repeating the same thing …