Skip Navigation

International Mathematics and Science Achievement from a U.S. Perspective: Results from TIMSS-R

Chat Host Good afternoon! My name is Gary Phillips and I am the Acting Commissioner of the National Center for Education Statistics. With me today is Dr. Judith Sunley, the Interim Assistant Director of the Directorate for Education and Human Resources at the National Science Foundation. We are here to chat about the results that were released this morning from the Third International Mathematics and Science Study-Repeat (TIMSS-R). We are sure there are a lot of questions out there, so let's get started…

Leigh from Whitehall, PA asked:
I noticed the panel that constructed the study only has one public school educator. Why weren't more private and public school educators included on the panel? Also, are homeschooled students taken into consideration in this study? Were any private schools included in the study?
Dr. Gary Phillips & Dr. Judith Sunley: There were many people involved in the development process including many private schools. The devlopement of TIMSS involved many more people than the panel. We had a representative sample of U.S. private schools in the study....Gary Phillips

Mary Ellen from Phoenix, Arizona asked:
I was privileged to attend the sessions when the TIMSS report was released in Kansas City. At that time it was evident that one of the reasons for our poor showing in math was that the focus of instruction in the U.S. schools was "drill the skills" instead of application of concepts. I am very interested to learn if instruction in our schools has changed sufficiently to cause an upsurge in the standing of U.S. youth. Are the countries showing a high level of achievement still using a concept/application model for instruction?
Dr. Gary Phillips & Dr. Judith Sunley: Mary Ellen..According to our most recent data, American teachers still use a lot of drill and practice....Gary Phillips

Joseph from San Francisco, California asked:
Will the data files being released be of the same format as the 1995 data? If not, what differences will there be? Is there any plan to compare the TIMSS-R results with the recent NAEP science test administration?
Dr. Gary Phillips & Dr. Judith Sunley: Joseph...The data files will be released in about 6 months and they will be in the same format as the 1995 data. We will also have a report next year on a link between NAEP and TIMSS-R...Gary Phillips

Bridget from San Antonio, TX asked:
What do you feel is the most significant finding in the study?
Dr. Gary Phillips & Dr. Judith Sunley: Bridget...The most significant finding was that after 4 years there was no improvement in U.S. performance. We may need to wait for more than 4 years to be able to detect change...Gary Phillips

When will results for participating states be available?
Dr. Gary Phillips & Dr. Judith Sunley: Julie...The results from the 13 States and 14 school districts will be available April 4, 2001...Gary Phillips

Paul Trafton from Cedar falls, IA asked:
Could you highlight some of the key findings from the report?
Dr. Gary Phillips & Dr. Judith Sunley: Paul.. The U.S. was above the international average in both math and science in the comparison of 38 nations. There was no improvement across the 4 years from 1995 and 1999 in the U.S., and for most nations among the 23 nations that participated in both years....Gary Phillips

Billy from Montclair New Jersey asked:
There has been little change in 8th grade performance from 1995. Are reforms being implemented that would lead to improvement and when would you expect to see that improvement?
Dr. Gary Phillips & Dr. Judith Sunley: Billy...Yes reforms have been implemented. We hope to see changes when TIMSS is repeated in 2003....Gary Phillips

Steve from Hartford, CT asked:
Gary and Judy - One simplified summary statement for TIMSS-R is that nothing has changed in 4 years. Is that an accurate summary and if so, what is a good answer to the question "Why haven't we improved?"
Dr. Gary Phillips & Dr. Judith Sunley: Steve-That is a good summary. We do not have definitive answers to your question. We will know more when we analyze the data from the Video Study....Gary Phillips

Carole from Arlington, VA asked:
What were the sampling procedures used in the United States?
Dr. Gary Phillips & Dr. Judith Sunley: Carole...In the U.S. we used a 3 stage sampling procedure in which we first sampled geographic regions, school districts, then schools. The U.S. sample was a nationally representative sample as it was in all of the nations....Gary Phillips

Peter from Hackensack, NJ asked:
Hi. I am a 6th-grade teacher who believes in and implements standards-based instruction. Accountability across the mathematics curriculum from K-12, however, is, in my opinion sorely lacking. Many teachers are either completely unfamiliar or just find the activities "too difficult." In order to achieve, students need to be immersed in activities which are part of a cohesive curriculum guided by skilled "coaches." My question deals with systemic accountability. How can we best create a system that embraces the standards in a cohesive, "spiralling" curriculum from K-12?
Dr. Gary Phillips & Dr. Judith Sunley: We agree that a systemic approach is very important. As states and districts review their curriculum, they need to think in terms of a comprehensive program of mathematics and science that strategically builds on learning across the grades. As the earlier TIMSS study pointed out, the curriculum in the U.S. may cover too many topics without enough depth. Curriculum, assessment, teacher professional development as well as state and district policies need to be aligned...NSF

David from Texarkana, Texas asked:
What accounts for the drop in performance from 4th to 8th grade by these students?
Dr. Gary Phillips & Dr. Judith Sunley: David....There wasn't a drop in performance. In the U.S., students learned a lot from the 4th to the 8th grade. However, in other countries they learned at a faster rate....Gary Phillips

Mr. K. from Oradell asked:
I'm a sixth-grade teacher who believes students need to come to class prepared with an arsenal of "tools" in order to achieve problem-solving success (as opposed to success at mere "exercises"). Because of the lack of accountability in our system, students come to grade six without many of these tools (frac/dec/perc, ratio & prop, integer usage, et. al.) Mathematics, as the language of science, thrives on problem-solving. The standards are obviously wonderful and, if implemented systemwide in each and every district, will yield outstanding results. Can we hold teachers accountable for standards-based teaching?
Dr. Gary Phillips & Dr. Judith Sunley: We agree that the standards, developed by the mathematical and science communities, have gone a long way in identifying important mathematical skills and problem solving abilities that all students need to know and be able to do. To achieve these standards, teachers must have an understanding of content, pedagogical skills, and appropriate curriculum. They need time to carefully plan their lessons. Assessments need to be aligned with content and focus on both basic skills and problem solving abilities. States, districts, and teachers can all be held accountable for ensuring that all students achieve in mathematics and science.

Bridget from San Antonio, TX asked:
Following up on the last question, did other countries improve their scores or were they also static?
Dr. Gary Phillips & Dr. Judith Sunley: Bridget...Thank you for the follow up on an earlier question. There was no improvement in almost all of the other countries (18-19 countries)....Gary Phillips

Marion from Providence, Ri asked:
How was it determined what was a statistically significant change? I noticed, for example, that in the 1995-1999 change in math achievement, Canada, with a 10-point difference, and Cyprus, with a 9-point difference, were considered to have "significantly higher" scores, but not Hong Kong (with 13 pts) or the U.S. Also, why is there a warning in the study against comparing the relative change in different countries?
Dr. Gary Phillips & Dr. Judith Sunley: Marion...A statistically significant change is one in which we have a 95% confidence level. The reason why a 10 point difference might be significant in some countries and not others is because the margin of error is different in each country.....Gary Phillips

Penny from Weston, MA asked:
Do you have any thoughts about why the performance of black students in mathematics has increased so substantially? Do you attribute any of the change to the USI/SSI programs? Also, it looks as if the performance of students whose parents have a college education has improved. Is this so?
Dr. Gary Phillips & Dr. Judith Sunley: Penny...The data do not tell us "why". However, we have been tracking a reduction in the black/white gap for some time in our domestic NAEP program. Yes, the students of parents with college degrees have improved over the 4 years....Gary Phillips

Darnella from Bethesda, Maryland asked:
What can you say about the range of scores? Is the achievement gap narrowing?
Dr. Gary Phillips & Dr. Judith Sunley: Darnella...Yes, the achievement gap is narrowing for blacks and whites, but not for hispanics.....Gary Phillips

Melissa from Atlanta, GA asked:
Has there been any work done to cross-examine the findings of the TIMSS-R and the Glenn Commission report?
Dr. Gary Phillips & Dr. Judith Sunley: Melissa-Yes. The Glenn Commission used the TIMSS data extensively....Gary Phillips

Jill from Maryland asked:
In what sub-areas did African-African students improve the most dramatically in mathematics?
Dr. Gary Phillips & Dr. Judith Sunley: Jill...African American students improved across the board in math.....Gary Phillips

kathryn from Ft. Collins, Colorado asked:
1. How can we ensure conceptual as well as skills knowledge? 2. What are perceived as the primary reasons for the differences in the TIMSS results? 3. Is there a math program you could recommend for American junior high school?
Dr. Gary Phillips & Dr. Judith Sunley: A rigorous curriculum taught by well qualified teachers in a system that supports them are the best strategies. There are a number of mathematics programs developed with support from NSF that emphasize conceptual and skills knowledge. Among these are Connected Mathematics (Prentice Hall); Mathematics in Context (Encylopedia Britannica) and MATHematics (McDouglal Littell). -NSF

Gerald from Washington, DC asked:
How much variation do you anticipate among the state and local school achievement reports?
Dr. Gary Phillips & Dr. Judith Sunley: Gerald....We expect as much variation among our States and districts as we found across the countries in TIMSS-R.....Gary Phillips

Fred from Potomac, MD asked:
Will you gather the data on the curriculum/training changes made by each participating school system?
Dr. Gary Phillips & Dr. Judith Sunley: Fred...We will have some information on this among States and districts, but not from the countries....Gary Phillips

Karla from Arlington, VA asked:
When will the NAEP correlation be available?
Dr. Gary Phillips & Dr. Judith Sunley: Karla...The results from the NAEP/TIMSS-R linking study will be available next summer.....Gary Phillips

Lane from Columbia, SC asked:
I'm sure the TIMSS-R results will spur many reform activities. However, until the next assessment, (FIMSS maybe?), how will we gauge the effectiveness of what we are doing? When is the next international assessment planned?
Dr. Gary Phillips & Dr. Judith Sunley: The most helpful strategy is to constantly evaluate your programs in order to gauge effectiveness of science and mathematics education. This is what NSF does. Further, many districts are using TIMSS findings to plan programs to assess their progress. The IEA is planning to conduct another assessment in 2003. We think trend data are very important, but the decision for the U.S. to participate has not yet been made. -NSF

Miriam from Pennsylvania asked:
I know my school participated in TIMSS-R as part of the US cohort. Will we receive our scores?
Dr. Gary Phillips & Dr. Judith Sunley: Miriam...Unfortunately we are not giving back results to schools this time. We do not have representative data for each school. Instead the study is nationally representative....Gary Phillips

Ellen R. Delisio from Educationworld, Wallingford, CT asked:
How will these results affect curriculum?
Dr. Gary Phillips & Dr. Judith Sunley: TIMSS findings as well as those of TIMSS-R reinforce the need for curricula solidly based in important concepts that are taught by inquiry, problem solving and other standards-based techniques. -NSF

Bridget from San Antonio, TX asked:
Just one last question...If there was no change in performance in the majority of the countries studied, is that why you said 4 years may not have been long enough to see a change? And, if that is the case, then, in your opinion, what is the value or benefit of the study?
Dr. Gary Phillips & Dr. Judith Sunley: Bridget. Thanks for your questions. They were good ones. We do not really know how long to wait to be able to detect a change. We only know by actually doing the study. This is the first international study that explicitly tracked progress over time....Gary Phillips

Mary Ellen from Phoenix, Arizona asked:
It is absolutely incredible that no change has been shown in improvement. Is there anyway that this information can be disseminated to states and school districts? I may be mistaken, but the last report did not seem to have a wide circulation.
Dr. Gary Phillips & Dr. Judith Sunley: Mary Ellen...We are working to make the information more available to States and districts. You can do your part by alerting people to the TIMSS website at Phillips

Billy from Montclair New Jersey asked:
What factors do you think contribute to the superiority of several Asian countries on TIMSS and TIMSS-R?
Dr. Gary Phillips & Dr. Judith Sunley: Billy...Our data show that U.S. teachers teach how to do math and science, but Asian teachers teach students to understand math and science....Gary Phillips

Patti from Ft. Sumner, New Mexico asked:
The study points to the fact that students in other countries have teachers with a degree in either math or science while US teachers usually have degrees in education. What provisions are being made to raise teacher salaries in order to allow people with subject degrees to even contemplate education as a viable job?
Dr. Gary Phillips & Dr. Judith Sunley: As you know, teacher assignments as well as their salaries are determined locally. However, NSF can affect the quality of teacher preparation and practice in mathematics and science. Specifically, programs at NSF encourage undergraduate degrees or substantive coursework in a content major and content based professional development. -NSF

Thomas from Delaware asked:
Nationally, only 25% of American 8th graders take Algebra. Have you examined the math achievement of this cohort of students for TIMSS?
Dr. Gary Phillips & Dr. Judith Sunley: Thomas...No we do not have data broken down in this way....Gary Phillips

Frank Cole from Oakhurst New Jersey asked:
If no improvement has occurred in the participating countries, could it be that it is not due to a "lack of data" but rather that a particular method (ie. drill & practice or concept based) is reaching its maximum potential.
Dr. Gary Phillips & Dr. Judith Sunley: Frank...We will know more about this issue when we analyze the Video data next year.....Gary Phillips

Paul D Boyer PhD from Kenosha, WI asked:
Although I'm rather shocked at the general lack of improvement, is there any sign that the science education standards are 1) being incorporated into learning outcomes and 2) making any difference in terms of conceptual learning vs rote memorization? Thanks.
Dr. Gary Phillips & Dr. Judith Sunley: NSF is supporting the development of new curricula on learning outcomes that is grounded in science standards. Studies in classes of teachers who are implementing science standards indicate that their students have increased their conceptual understanding. With the dissemination of new materials, it is hoped that this effect will increase. Where standards based curricula are implemented appropriately by well educated teachers with assessments that are in line with the materials, students learn at high levels. -NSF

Marion from Providence, RI asked:
A very technical little question: is the country listed as "Chinese Taipei" what we generally call Taiwan, or is it a more limited sample?
Dr. Gary Phillips & Dr. Judith Sunley: Marion. Yes, that is Taiwan. The sample covered the full Island.....Gary Phillips

Cindy from Pittsburgh, Pennsylvania asked:
As a followup to Billy's question - You summarized the differences between how our teachers and their Asian counterparts teach. I assume that came from a review of the TIMSS videos, but isn't it mostly the case that our unofficial "national" curriculum (as represented in MOST textbooks we use) presents a curriculum that is too fragmented and unfocused to support meaningful instruction, hence focused achievement?
Dr. Gary Phillips & Dr. Judith Sunley: Cindy...The answer is yes. In general the Asian curriculum is more focused and textbooks are shorter....Gary Phillips

Billy from Montclair, New Jersey asked:
Our data show that US teachers teach how to do math and science, but Asian teachers teach students to understand math and science....Gary Phillips SO, SHOULDN'T TEACHER EDUCATION AND PROFESSIONAL DEVELOPMENT BE CHANGING--SIGNIFICANTLY AND RAPIDLY--TO REFLECT THAT?
Dr. Gary Phillips & Dr. Judith Sunley: Billy...YES....Gary Phillips

Billy from Montclair New Jersey asked:
What steps will NSF take that can lead to concrete reforms in the nation's schools?
Dr. Gary Phillips & Dr. Judith Sunley: NSF has developed many programs that lead to changes in teaching practices and student learning. For example, our systemic initiatives as well as our local systemic change program specifically address reform in schools. Our new Centers for Learning and Teaching will rebuild connections between schools and higher education to collaboratively address reform. -NSF

Dale from Philadelphia asked:
What is the most troubling trend in this report?
Dr. Gary Phillips & Dr. Judith Sunley: The most troubling aspect of the report is that the U.S. continues to lag well behind the top performing countries. However, when comparing the results of 1995 to those of 1999, we must keep in mind that improvement in a large diverse educational system will be slow. We have time for one more question -NSF

Cindy from Pittsburgh, Pennsylvania asked:
What efforts are underway to better understand the context of curriculum and instruction in our middle schools that might make a difference on how our 8th graders perform? Are there any chances of being able to look at subsets of data from schools that have adopted/used exemplary curricula?
Dr. Gary Phillips & Dr. Judith Sunley: NSF has funded several projects to study the impact of innovative instructional materials on student achievement. Results from these studies will not be available for another year or two. However, TIMSS-R results for a number of states and consortia of schools will be released in April 2001. -NSF

Thank you for taking the time to ask these questions. They were all very good. If you have additional questions, please feel free to contact NCES or NSF. At NCES, you can contact Patrick Gonzales (, and at NSF you can contact Elizabeth VanderPutten ( Today's release of Pursuing Excellence: Comparisons of International Eighth-Grade Mathematics and Science Achievement from a U.S. Perspective, 1995 and 1999 is the first in a series of several reports focusing on mathematics and science achievement, teaching, and learning in an international context that will be released over the next several years. We look forward to continuing this dialogue with you.

Back to StatChat Home