Statement by Dr. Michael O. Martin and Dr. Ina V.S. Mullis.
Co-Directors, the International Study Center at Boston College

Presentation of TIMSS 1999 International Results

Dr. Ina Mullis

It is our pleasure to release the results from TIMSS 1999, also known as TIMSS-R. I am going to begin by providing an overview of the achievement results in mathematics and science, and Mick is going to present highlights of the extensive information we collected about students' attitudes towards mathematics and science as well as their curricula and instruction. As you have heard Hans explain, 38 countries participated in TIMSS 1999 and have mathematics and science achievement data for their eighth-grade students.

Exhibit 1 shows the TIMSS 1999 results for mathematics achievement.

Five Asian countries were the top performers in mathematics. Singapore, the Republic of Korea, Chinese Taipei, and Hong Kong had the highest average achievement. Japan also performed very well.

To further help understand students' performance, TIMSS identified points on the scale to use as international benchmarks, including the Upper Quarter Benchmark. This is the point above which the top 25 percent of all the students in the assessment scored. As you can also see from Exhibit 1, Singapore, Chinese Taipei, Korea, Hong Kong, and Japan had about two-thirds or more of their students reaching the Upper Quarter Benchmark. Students reaching the Upper Quarter Benchmark did very well on the TIMSS mathematics test, demonstrating that they could apply their mathematical understanding and knowledge in a variety of relatively complex situations involving fractions, decimals, geometric properties, and algebraic expressions.

Exhibit 2

In science, Chinese Taipei and Singapore had the highest average performance, closely followed by Hungary, Japan, and the Republic of Korea.

Singapore and Chinese Taipei had more than half their students reaching the Upper Quarter Benchmark. Remember this is the level reached by just 25 percent of all the students assessed internationally. Students reaching this benchmark showed conceptual understanding of various science cycles, systems, and principles.

Exhibit 3 shows the trends in mathematics achievement.

Of the 38 countries participating in TIMSS 1999, 26 also participated in the original TIMSS in 1995 and have trend data between 1995 and 1999. The black bars show the changes that were statistically significant. The countries that showed a significant increase over time in average mathematics achievement at the eighth grade were Latvia, Canada, and Cyprus. Only the Czech Republic showed a significant decrease.

Exhibit 4

Between 1995 and 1999, countries that showed a significant increase in average science achievement were Latvia , Lithuania, Canada, and Hungary. Bulgaria was the only country showing a significant decrease. IEA has a long-standing policy of providing information about gender differences in achievement. The fact that boys historically have had higher average achievement than girls in mathematics and science assessments makes this subject very near and dear to my heart.

The good news for TIMSS 1999 is that in mathematics, most gender differences were negligible. As you can see from the chart, only four countries had significantly higher average achievement for boys -- Israel, the Czech Republic, Iran, and Tunisia. Korea even showed a decrease in the gender difference between 1995 and 1999. Fortunately, no country showed a significant increase in the gender difference in mathematics performance.

The bad news was in science which had significant differences in average achievement favoring boys in 16 of the 38 countries Ð nearly half. Interestingly, these differences were more apparent among high-performing students. Three countries, however, did have a significant reduction in the gender difference between 1995 and 1999 Ð Hong Kong, Slovenia, and Israel. So, there was some progress towards greater equity in science achievement.

Dr. Michael O. Martin

Presentation of TIMSS 1999 International Results In her remarks, Ina has focused on the mathematics and science achievement of the students that took part in TIMSS in 1999. However, to provide a context to interpret these achievement data, TIMSS also collected a rich array of information from national experts, from school principals, from mathematics and science teachers, and from the students themselves. In the few minutes available to me today, I want to share with you just a few of the highlights from this collection.

Exhibit 5

1. Given the central importance of teachers and instruction in the educational enterprise, let me begin with some TIMSS findings in this area. Exhibit 5 summarizes teachers' reports of their confidence in their preparation to teach mathematics and science to their eighth-grade students. The figure shows, for both mathematics and science, the percentage of students in each country taught by teachers who reported that they had a high level of confidence in their preparation to teach their subject. The data reveal some interesting differences, both between subjects and across countries.

The results indicate that eighth-grade mathematics teachers have more confidence in their teaching preparation than science teachers. Generally, mathematics teachers reported relatively high levels of confidence in their preparation, with 63 percent of students on average taught by teachers who believed they were very well prepared. However, national levels of teacher confidence varied considerably around this international average, from a high of more than 85% in Macedonia, the Unites States, the Slovak Republic, and Cyprus, to a low of just 8% in Japan.

In contrast, eighth-grade science teachers reported only a moderate level of confidence in their preparation. Only one-fifth of students, on average, had science teachers who believed they were very well prepared. Almost 40 percent of students, across countries, were taught science by teachers who reported a low level of confidence in their preparation to teach the subject. Teachers from Macedonia again reported the highest level of confidence in their preparation, and teachers from Japan the lowest.

There are many such data displays to be found in the TIMSS 1999 international reports, but time does not permit any more detailed presentations. Instead, let me briefly touch on some of the other findings from the 1999 assessment, including more on teaching and instruction, on the mathematics and science curriculum, on student attitudes, and on school resources and environment.

Despite efforts in a number of countries towards more student centered learning, teachers across the TIMSS countries reported that the two predominant activities in mathematics classes are teacher lectures and teacher-guided student practice. Between them, these activities accounted for nearly half of the time in mathematics class. Similarly, science teachers reported spending almost one-quarter of their class time on lecture-style presentations. They reported spending 15 percent of their class time on student experiments, and 14 percent on teacher-guided student practice.

Videotapes of mathematics classes in the United States and Japan in the original TIMSS, in 1995, revealed that outside interruptions can affect the flow of the lesson, and detract from instructional time. Internationally in 1999 for both mathematics and science, about one-fifth of the students reported that their classes were interrupted almost always, or pretty often.

2. While teachers are, of course, the agents of instruction in all countries, it is the curriculum that specifies what mathematics and science should be taught in the classroom and the laboratory. In almost all of the 38 TIMSS countries, curriculum goals in mathematics and science were specified in a national curriculum. The exceptions were Australia, Canada, and the United States, where curriculum decisions are made at the state or provincial level.

There are two major approaches to the organization of science instruction for students in the eighth grade and above. Just over half of the TIMSS countries follow a general science curriculum at this level, but 17 of the countries, mainly in central and eastern Europe, offered separate courses in the different science subjects Ð earth science, biology, chemistry, and physics.

Testing and assessment were widely used methods to support curriculum implementation. About two-thirds of the countries conduct regular system-wide assessment at two or three grades, primarily to inform policy makers about achievement of curriculum goals.

In countries generally, the percentage of time specified in the official curriculum for mathematics instruction remains about the same from grade 4 to grade 6 but then decreases by grade 8 (17, 16, and 13 percent, respectively). In contrast, the instructional time specified for science increases from grade 4 to grade 8 (from 11 to 16 percent).

3. A major curricular goal in most countries is to develop a positive affect among students for the subjects they study. Across the TIMSS countries, students generally had positive attitudes towards both mathematics and science, although attitudes were less positive in countries where science is taught as separate subjects at the eighth grade.

n every country, a positive self-concept in mathematics and science was associated with higher achievement. In general, boys had a more positive self-concept than girls in both subjects. However, in separate-science countries, girls had more positive self-concept than boys in biology. This, however, was outweighed by a more favorable self-concept for boys in physics, and to a lesser extent in earth science and chemistry.

Eighth-grade students internationally had high expectations for further education. On average across countries, more than half the students reported that they expect to finish university. In almost every country, students with high educational expectations also had high achievement in mathematics and science.

4. Well-equipped schools that provide safe and supportive environments are generally accepted as a basic requirement for effective student learning. In TIMSS, students in schools that reported being well resourced generally had higher average mathematics and science achievement than those in schools where across-the-board shortages affected the school's capacity to provide instruction.

Internationally, the overwhelming majority of eighth-grade students attended schools judged by principals to have few serious problems threatening an orderly or safe school environment. Sixty percent, however, were in schools where principals reported moderate attendance problems, and 19 percent were in schools with some serious attendance problems.

In summary, TIMSS is truly a rich resource for information about mathematics and science teaching and learning around the world. The reports provide considerable grist for the conversation about what we want schools to accomplish, and how we can go about improving the teaching and learning of mathematics and science.

Ð END Ð

TIMSS 1999 is a project of the International Study Center
Boston College, Lynch School of Education