DOCUMENT RESUME
ED 433 243 SE 062 784
AUTHOR Kimmelman, Paul; Kroeze, David; Schmidt, William; van derPloeg, Arie; McNeely, Maggie; Tan, Alexandra
TITLE A First Look at What We Can Learn from High PerformingSchool Districts: An Analysis of TIMSS Data from the Firstin the World Consortium.
INSTITUTION National Inst. on Student Achievement, Curriculum, andAssessment (ED/OERI), Washington, DC.
REPORT NO SAI-1999-3011PUB DATE 1999-08-00NOTE 96p.
AVAILABLE FROM ED Pubs, P.O. Box 1398, Jessup, MD 20794-1398; Tel:877-433-7827 (Toll Free); TTY/TTD: 877-57677734; Fax:301-470-1244; Web site: http://www.ed.gov/pubs/edpubs.html
PUB TYPE Reports - Research (143)EDRS PRICE MF01/PC04 Plus Postage.DESCRIPTORS *Academic Achievement; *Cross Cultural Studies; Elementary
Secondary Education; Foreign Countries; *MathematicsAchievement; Mathematics Education; *School Districts;Science Education; *Socioeconomic Influences; TeachingMethods
IDENTIFIERS *Third International Mathematics and Science Study
ABSTRACTInterest in the First in the World (FiW) Consortium's
activities has been intense for a number of reasons. Initial results showthat the FiW Consortium is well on its way to meeting its goal, with the FiWperforming at or near the top of the world on the international benchmarkchosen by the Consortium--the Third International Math and Science Study(TIMSS). A status report is provided on the FiW's initial activities. Thefirst section describes the districts that make up the FiW Consortium andoutlines the Consortium's history, purpose, goals, and plan of action tobecome the first in the world in math and science. The second sectiondocuments how the students from the FiW Consortium performed when benchmarkedagainst an international comparison, the TIMSS assessment. The next sectiondiscusses the impact that socio-economic variables could have on theperformance of FiW students. The fourth section presents data on the contextsfor teaching and learning mathematics in the FiW, exploring differencesbetween the FiW and the U.S. and, where data are readily available,differences between the contexts for teaching in the FiW and countries withhigh mathematics achievement. The fifth section describes some of theactivities being undertaken by the FiW Consortium to improve math and scienceeducation, highlighting a recent project that is using data from TIMSS andother sources to improve science instruction. The final section summarizesthe whole report and offers some possible questions for future research. Twoappendices provide: (1) contact information and characteristics of FiWConsortium Districts; and (2) 12 tables presenting FiW TIMSS AchievementResults by country. (Contains 19 references and 32 endnotes.) (ASK)
E
cnC1
res
4.1
4
a
add
D aQ'CiC>
f?,-C-Cji2E1 cz.i>Qpow7. 0-0-6:71
D rOAV ED
hac3 171-- 3
c'0 DB' \o
J1
is----( 1 rGL (\-1j
IQ
UM rartrilrgi0 ee
0 LSVlS u 0
U S DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement
UCATIONAL RESOURCES INFORMATION
CENTER (ERIC)Th s document has been reproduced as
eived from the person or organizationoriginating itMinor changes have been made to
improve reproduction quality
Points of view or opinions stated in thisdocument do not necessarily representofficial OERI position or policy
L
COPY AVAILABLE
- e- I
e-s. Si
To help us improve future editions of this publication and giveyou better customer service, we would appreciate yourcomments on this survey form. Please check the appropriateboxes below for each question. Responses will be keptcompletely confidential. You may return the survey by mail orFAX. It can be folded and taped closed to allow mailing to theaddress listed on the reverse side of this form, or it can bereturned by FAX to 202-219-1321. Many thanks for yourcustomer feedbackft is very Important to us!
1. Name of publication
2. How did you receive a copy of this publication?
Bought it
Borrowed it
Mailing list membership
Telephone request
Internet request
Other (please describe)
3. Was this publication easy to get?
Very Somewhat Not at all
4. How did you find out about this and other OERIpublications? (Check all that apply.)
Conferences
Journal articles
Teacher/educator
Professional associations
0 Internet (WWW)
Publication announcement
Received in mail
OERI staff contact
5. For what purposes did you use this OERI publication?(Check all that apply.)
Planning
Policy or legislation
Administrative decisions
Teaching, class material
Research/analysis
General information
Writing news articles, TV or radio material
Marketing, sales, or promotion
Other (please describe)
6. Did the publication help you accomplish whatever youneeded ft for?
Yes No Partially
7. What is your occupation?
Parent Teacher Administrator
Librarian Researcher Statistician
Journalist/writer Policy Analyst Student
Program Planner
Other (please specify)
8. How could this OERI publication (or other OERIpublications) better meet your needs?(Check all that apply).
0 More important topics in education
More timely release of data
More text introductions to each section
More research statistics
Shorter reports (less than 10 pages)
Other (please describe)
9. Overall, how satisfied are you with the following aspects of this publication?
Very Satisfied
a. Comprehensiveness of information
b. Clarity of writing (readability, interpretability)
c. Clarity of presentation (e.g., tables, charts)
d. Timeliness of information
e. Accuracy of information
f. Clarity of technical notes
g. Usefulness of resources and bibliography
h. Organization
i. Length
j. Format
Satisfied Dissatisfied
O
0
PAPERWORK BURDEN STATEMENTOffice of Educational Research and Improvement (OERI)
Publication Customer Survey
According to the Paperwork Reduction Act of 1995, no persons are required to respond to a collection of Information unless it displays a valid OMBcontrol number. The validOMB control number for this Information collection Is 1880-0529. The time required to complete this information collection is estimated to average 10 minutes per response,Including the time to review Instructions. search existing data resources, gather the data needed, and complete and review the Information collection. If you have any comments
concerning the accuracy of the time estimate(s), suggestions for improving this form, or concerns regarding the status of your Individual submission of this form, writedirectly to:
P. Quinn, Room 204, Media and Information Services, OERI, U.S. Department of Education, 555 New Jersey Avenue NW, Washington, DC 20208-5570.
3 BEST COPY AVAICABLE
OERI Publication Customer SurveyMedia and Information Services
555 New Jersey Avenue NWRm. 202Washington DC 20208-5570
OFFICIAL BUSINESSPENALTY FOR PRIVATE USE, $300
0 I N
BUSINESS REPLY MAILFIRST-CLASS MAIL PERMIT NO. 012935 WASHINGTON DC
POSTAGE WILL BE PAID BY US DEPARTMENT OF EDUCATION
A. ReedRoom 202Media and Information ServicesOffice of Educational Research and ImprovementUS Department of Education555 New Jersey Avenue NWWashington DC 20208-5570
NO POSTAGENECESSARY
IF MAILEDIN THE
UNITED STATES
Fold on line TAPE CLOSEDDO NOT STAPLE
10. Do you have any suggestions regarding the content or format of future editions of this publication or other comments?
4
MIS 1999-6532
A FIRST LOOK AT WHAT WE CAN LEARN FROMHIGH PERFORMING SCHOOL DISTRICTS:
An Analysis of TIMSS Data From the First in theWorld Consortium
Paul Kimmelman, President, First in the World ConsortiumDavid Kroeze, Superintendent, Northbrook School District 27
William Schmidt, Michigan State UniversityArie van der Ploeg, North Central Regional Educational Laboratory
Maggie McNeely, U.S. Department of EducationAlexandra Tan, Education Statistics Services Institute
U.S. Department of EducationRichard W. RileySecretary
Office of Educational Research and ImprovementC. Kent McGuireAssistant Secretary
National Institute on Student Achievement, Curriculum, and AssessmentIvor A. PritchardActing Director
Media and Information ServicesCynthia Hearn DorfmanDirector
August 1999
This study was funded by the Office of Educational Research and Improvement, U.S.content does not necessarily reflect the views of the Department or any other agency
To obtain copies of this report (while supplies last) or ordering information on other U.products, call toll free 1-877-4ED-Pubs (877-433-7827), or write:
U.S. Department of EducationED PubsP.O. Box 1398Jessup, MD 20794-1398
Copies are available in alternate formats upon request.
TTY/TTD: 1-877-576-7734FAX: 301-470-1244
Online ordering via the Internet: http://www.edgov/pubeedpubs.html
Content Contact:Maggie McNeely202-219-1568
Department of Education. Theof the U.S. Government.
S. Department of Education
Acknowledgments
The authors of this report represent a unique collaboration among local school districts,the U.S. Department of Education (ED), the Education Statistics Services Institute(ES SI), and the North Central Regional Educational Laboratory. We wish toacknowledge all those who contributed to this report, especially the teachers, students,administrators, and parents of the First in the World Consortium. Without theirdedication and commitment to improving math and science achievement, this reportwould not exist.
We would also like to thank Erin C. Massie, Joanna Wertheimer, and Mark White ofESSI and Blase Masini of NCREL who provided valuable research assistance. J.W.McConnell, Eugene H. Owen of NCES, Kitty Mak of ESSI, Patrick Gonzales of ESSI,and John Luczak of ED all gave their time and expertise by reviewing draft reports andproviding comments and suggestions. We are particularly grateful to Ron Varnum, fromColgate University, who did the final formatting and quality control work on the report.Christopher Sheedy, of ED, provided editorial services.
7
Contents
Introduction 1
What is the FiW Consortium? 3
History and Description 3
Goals and Objectives 4
Plan of Action 5
How Well Did the FiW Consortium Perform When Benchmarked Againstan International Measure of Math and Science Achievement? 7
Fourth-Grade Results 7
Eighth-Grade Results 8
Twelfth-Grade Results 8
General Knowledge 9
Advanced Math and Physics 10
AP Calculus and AP Physics Results 10
Summary 11
Do Home Factors Explain the High Achievement of the FiW Consortium? 13
What is the Context for Teaching and Learning Math in the FiW Consortium? 14
Curriculum 15
Instructional Topic Coverage 15
Textbook Use and Topic Coverage 20
Fourth Grade 21
Eighth Grade 21
Summary 22
Classroom Instructional Practices 22
Teachers' Reports on Class Activities 22
Teachers' Reports on Classroom Organization 26
Students' Reports on Classroom Organization and Class Activities 28
Summary 31
Teacher Engagement 32
Teacher Involvement in School-related Activities Outside the School Day 32
Frequency of Teacher Meetings 35
Teacher Influence Over Key School Decisions 36
Teacher Familiarity with Key Curriculum and Assessment Documents 39
Summary 40
Teaching Environment 41
Summary 43
What is the FiW Consortium Doing to Improve Math and Science? 45
Teacher Learning Networks: Collaborative Learning Communities 45
Teacher Network Grants 48
vii
Example of a FiW Teacher Learning Network: Analysis of FiW Physics Achievement 49Project Activities 50Initial Results 50Topics for Further Research Identified by the Physics Teachers' Learning Network 51Summary 52
Summary and Conclusion 53
Bibliography 57
Endnotes 59
Appendix A: Contact Information and Characteristics of First in the WorldConsortium Districts A-1
Appendix B: FiW TIMSS Achievement Results B-1
viii
List of Exhibits
Exhibit 1: Fourth-Grade Math Teachers' Reports on Instructional Topic Coverage 18
Exhibit 2: Eighth-Grade Math Teachers' Reports on Instructional Topic Coverage 19
Exhibit 3: Teachers' Reports on Use of Textbooks and Other Teaching Materials and Percent
of Mathematics Teaching Time based on Textbook 20
Exhibit 4: Teachers' Reports on How Frequently Students Are Asked to Complete Specific
Tasks 25
Exhibit 5: Teachers' Reports on Classroom Organization During Mathematics Lessons 27
Exhibit 6: Students' Reports on the Frequency of Math Class Activities 30
Exhibit 7: Teachers' Reports on Hours Spent Per Week on Activities Outside the Formal
School Day 34
Exhibit 8: Teachers' Reports on the Frequency of Meetings with Other Teachers in Their
Subject Area to Discuss and Plan Curriculum or Teaching Approaches 36
Exhibit 9: Teachers' Reports on Their Influence over School Decisions 37
Exhibit 10: Teachers' Reports on Their Familiarity with Key Curriculum Documents 38
Exhibit 11: Teachers' Reports on the Factors that Limit How They Teach Mathematics Class 42
Exhibit 12: First in the World Teacher Learning Networks 47
Appendix A: Contact Information and Characteristics of First in the World Consortium Districts
Exhibit A-1: List of First in the World Districts A-2
Exhibit A-2: First in the World District Characteristics A-4
Appendix B: FiW TIMSS Achievement Results
Exhibit B-1: Distributions of Mathematics Achievement in the Fourth Grade B-2
Exhibit B-2: Distributions of Science Achievement in the Fourth Grade B-3
Exhibit B-3: Fourth-Grade Mathematics Performance, by Subtopic B-4
Exhibit B-4: Distributions of Mathematics Achievement in the Eighth Grade B-5
Exhibit B-5: Distributions of Science Achievement in the Eighth Grade B-6
ix
10
Exhibit B-6: Eighth-Grade Mathematics Performance, by Subtopic B-7
Exhibit B-7: Distributions of General Mathematics Achievement in the Twelfth Grade B-8
Exhibit B-8: Distributions of General Science Achievement in the Twelfth Grade B-9
Exhibit B-9: Distributions of Advanced Mathematics Achievement in the Twelfth Grade B-10
Exhibit B-10: Distributions of Advanced Science (Physics) Achievement in the Twelfth Grade B-11
Exhibit B-11: Distributions of Advanced Mathematics Achievement for Twelfth-Grade FiW
AP Students B-12
Exhibit B-12: Distributions of Advanced Science (Physics) Achievement for Twelfth-Grade FiW
AP Students B-13
11x
Introduction
The efforts of a group of small schooldistricts north of Chicago to become first inthe world in math and science achievementhave recently captured the interest ofeducators, researchers, and policymakersacross the country.
Working collaboratively, rather thancompetitively, this unique consortium ofdistricts has begun to benchmark its studentsagainst an international standard ofachievement, identify and implement bestpractices for improving math and scienceachievement, and establish learningnetworks among the educators within itsdistricts.
Interest in the First in the World (FiW)Consortium's activities has been particularlyintense for a number of reasons. First, initialresults show that the FiW Consortium iswell on its way to meeting its goal, with theFiW performing at or near the top of theworld on the international benchmarkchosen by the Consortiumthe ThirdInternational Math and Science Study(TIMSS).'
Given the disappointing performance of U.S.students on the same international math andscience assessments, many believe that theapproaches taken by the FiW towards mathand science may offer some importantlessons for other schools and districts in theU.S.
Second, the results obtained by the FiWConsortium on this international benchmarkalso provide the first U.S. multi-district leveldata available from TIMSS. As such, theymay provide valuable insights into thecontexts for learning math and science inhigh-performing districts, and how they
might relate to world class standards, notonly in achievement but also ininstructional, curricular, and assessmentstandards. Finally, the Consortium'sactivities offer a rare opportunity to learnfrom this unique cross-district, multi-partnercollaborative effort to obtain world-classstandards.
This report provides a status report on theFiW's initial activities. It examines five keyquestions:
What is the FiW Consortium?
How well did the FiW Consortiumperform when benchmarked against an ,
international measure of math andscience achievement?
Do home factors explain the highachievement of the FiW Consortium?
What is the context for teaching andlearning in the FiW Consortium?
What is the FiW Consortium doing toimprove math and science?
It should be noted that this report constitutesone of the first comprehensive examinationsof how FiW students performed on TIMSSand an initial exploration of some of thereasons for their performance.
This paper is not based on an exhaustivereview of the possible reasons for the FiWperformance, but rather examines possiblefactors that might have had an importantrole. In addition, the paper does notinvestigate causal links between FiWachievement and the different topicsdiscussed later.
The remainder of the report is organized intomajor sections, which roughly correspond tothe key questions listed above.
The first section describes the districts thatmake up the FiW Consortium and outlinesthe Consortium's history, purpose, goals,and plan of action to become the first in theworld in math and science.
The second section documents how thestudents from the FiW Consortiumperformed when benchmarked against aninternational comparison, the TIMSSassessment.
The next section discusses the impact thatsocio-economic variables could have on theperformance of FiW students.
The fourth section presents data on thecontexts for teaching and learningmathematics in the FiW, exploringdifferences between the FiW and the U.S.and, where data are readily available,differences between the contexts forteaching in the FiW and countries with highmath achievement.
The fifth section describes some of theactivities being undertaken by the FiWConsortium to improve math and scienceeducation, highlighting a recent project thatis using data from TIMSS and other sourcesto improve science instruction.
The final section summarizes the report andoffers some possible questions for futureresearch.
13
What is the FiW Consortium?
This section describes the First in the WorldConsortium. It answers questions such as:
How and why was the Consortiumlaunched?
How many students are in theConsortium?
What are its goals?
How does it hope to achieve these goals?
It begins with a description of theConsortium's history and origins, and thendescribes its goal and objectives, as well asthe Consortium's plan of action forachieving these objectives.
History and Description
The FiW Consortium is currently made upof 17 school districts located in the northsuburbs of Chicago and the Illinois Mathand Science Academy. Together, they arepursuing a common goal of becoming firstin the world in math and scienceachievement.
The Consortium grew out of a study groupof superintendents which was formed tofulfill an administrative re-certificationrequirement. Members of this study groupmet regularly over the course of severalmonths. Discussions at these meetingscentered on contemporary education reformissues facing the administrators in theirvarious districts.
The FiW Consortium was launched at thefinal meeting of the superintendents' group,
3 14
during which the National Education Goalswere discussed.
Determined to take the national goalsseriously, the superintendents decided toform a consortium of districts committed toproviding a world class education for theirstudents. The group agreed to first focuscollectively on obtaining Goal 5, to be firstin the world in math and science by the year2000.
Reflecting their goal, the group calledthemselves the First in the WorldConsortium, a title with which they feltuncomfortable, but which accuratelycaptured their goals and aspirations. InMarch 1995, the Consortium entered intopartnership with the U.S. Department ofEducation and the North Central RegionalEducation Laboratory in its efforts to obtainthis goal.
As of winter 1999, the Consortium included13 elementary districts (grades K-8), 3 highschool districts (grades 9-12), the NorthSuburban Special Education District (whichserves most of the districts' specialeducation students), and the IllinoisMathematics and Science Academy(IMSA).2 IMSA is a publicly fundededucational laboratory and three-yearresidential secondary program for Illinoisstudents gifted in math and science. WhileIMSA is a state agency, the term "districts"is used throughout the report as aconvenience to the reader.
Together, these districts contain 49elementary or middle schools, 6 highschools, and one special education school.During the 1996-97 school year, totalstudent enrollment in these schools was
approximately 36,000 students. Nearly fourout of five students (78 percent) within theConsortium were white, non-Hispanic.Fourteen percent of the students wereAsian/Pacific Islanders, while seven percentwere Hispanic and two percent were black,non-Hispanic.
The Consortium contains relatively highwealth districts, and thus may have differentcharacteristics from many individual U.S.districts and the U.S. as a whole. In the1995-96 school year, average per-pupilexpenditures across the Consortium wasapproximately $8,9583 compared to $5,774in the U.S .4
Yet, not all of the Consortium's studentscome from high-income families. In 1996-97, seven percent of the Consortium'sstudents were classified as coming fromlow-income families.' Six percent ofConsortium students had limited-Englishproficiency (LEP).
Approximately 2,600 classroom and specialeducation teachers teach within theConsortium.6 In general, FiW teachers tendto have higher education levels than theirU.S. counterparts: Sixty-three percent ofFiW teachers have earned at least a master'sdegree versus 56 percent of U.S. teachers.'The average number of years of teachingexperience for FiW teachers is 14 years.'
Perhaps reflecting the education andexperience levels of FiW teachers, averageteacher salaries in Consortium districts arerelatively high. In 1995-96, the averagesalary for FiW elementary school teacherswas $47,339. The average salary for FiWhigh school teachers was $65,263.9
U.S. teacher salaries for the same timeperiod are lower. The average salary in the
U.S. during the 1995-96 school year was$39,976 for elementary school teachers and$38,423 for secondary school teachers."'See appendix A for a current list of FiWdistricts and more information on FiWdistrict characteristics.
Goals and Objectives
As noted earlier, the FiW Consortiumdecided to focus first on obtaining Goal 5 ofthe National Education Goals: to becomefirst in the world in math and science. TheConsortium's leadership set three objectivesto help them obtain this distinction. Theywere:
Benchmark Consortium schools'performance against an internationalmeasure of student achievement;
Create a forum to clarify world-classstandards for business leaders,policymakers, educators, and communitymembers; and
Establish networks of learningcommunities that actively involveeducators, parents, and communityleaders.
These objectives were chosen to provideConsortium leaders and educators with abaseline against which to measure theirprogress, as well as a better understandingand knowledge of the instructional,curricular, and assessment practices neededto obtain world-class achievement.
Recognizing that obtaining this ambitiousgoal would involve input, advice, andsupport from all members of theConsortium's community, they also soughtto actively involve educators, parents, andcommunity leaders.
15
Plan of Action
To obtain the consortium's goal andobjectives, the FiW leadership developedand embarked upon a three-step plan ofaction. These steps were:
(1) developing partnerships at the national,regional, and local levels;
(2) identifying and defining world classstandards in math and science; and
(3) working with the Consortium's partnersto implement exemplary math andscience programs.
Each of these steps is described brieflybelow.
First, the FiW Consortium sought to formpartnerships with key organizations in theeducation and business communities toobtain technical, administrative, andresearch support in achieving its goals.Accordingly, the FiW has establishedworking partnerships with numerousorganizations, including the U.S.Department of Education, the North CentralRegional Education Lab (NCREL), andpolicymakers at the national level, includingmembers of Congress.
Under its agreement with the U.S.Department of Education, the Consortiumcommitted to work with the Department toexplore general outcome and specific mathand science competencies in a study ofglobal competition. It also promised todevelop world class standards and alignthese standards with their local curricularand instructional programs and to supportthe acceleration of technologyimplementation as it pertains to math andscience achievement.
5.1 6
In addition, the Consortium agreed todevelop a math and science resource centeron the World Wide Web, as well ascollaborate with the Department todisseminate the Consortium's findings. Italso pledged to include all students in itsefforts, including students with disabilities.
The Consortium resolved to develop andimplement assessment instruments todetermine student achievement, and toimplement staff development training toassist teachers in mastering new content andinstructional strategies.
Further, the Consortium committed toentering into school and businesspartnerships to foster the identification ofthe needed skills and knowledge to achieveworld class standards.
Finally, the Consortium agreed to committhe needed resources to ensure that the jointeffort did not fail due to a lack of resources.Taken together, these commitments form aunique partnership between a group of localschool districts and the Department ofEducation.
Second, the FiVleadership is working withits partners to identify and define world-class standards in the areas of math andscience. As part of this effort, FiW studentsparticipated in TIMSS, the most ambitious,comprehensive, and rigorous internationalassessment of math and science everundertaken. With financial support from theindividual boards of education that make upthe Consortium, FiW fourth, eighth, andtwelfth grade students, their teachers, andtheir schools participated in the TIMSSstudy.
Although an invitation to participate inTIMSS was extended to all school districts
in the nation, the FiW Consortium was theonly group that took advantage of thisopportunity.
Third, FiW districts are working together todesign and implement exemplary programsin mathematics and science. To achieve thisgoal, FiW leaders have established teacherlearning networks which bring together stafffrom across the Consortium to identify,develop, and enact model programs in mathand science instruction.
Teams are self-identified and includeteachers, principals, superintendents, andother educational staff. Relying on datafrom TIMSS, as well as research on bestinstructional and curricular practices, theConsortium hopes to improve achievementby strengthening instruction, using moreeffective assessment tools, learning aboutnew curricula materials and techniques, andidentifying and addressing topics or areas
6
where their students demonstrateweaknesses.
In sum, the districts that make up the FiWConsortium have sparked the attention of theeducation community by agreeing to worktogether to become first in the world in mathand science. The following sectionsdescribe the efforts to enact their plan, aswell as some of the their preliminary results.
In particular, the next section describes theresults of the Consortium's effort to measureits performance against an internationalbenchmark. The following two sectionslook at what might account for the highachievement levels obtained, exploring firstthe effect of socio-economic factors, andthen the Consortium-wide context forteaching and learning. The subsequentsection describes the Consortium's efforts todevelop learning communities of educators,policymakers, and community leaders.
.17
How Well Did the FiW Consortium Perform When Benchmarked Against anInternational Measure of Math and Science Achievement?
As mentioned earlier, one of the FiWConsortium's three primary goals was tobenchmark its achievement against aninternational measure of studentachievement. The FiW chose TIMSS as itsmeasure. FiW student assessments wereadministered during 1996, and preliminaryresults were made available in January 1997.
This section discusses the results of theseassessments.
In general, FiW students did exceedinglywell on TIMSS, particularly in the fourthand eighth grades. In the twelfthgradeadvanced math and physics tests, the FiWAdvanced Placement (AP) students alsoscored among the top performing countries.FiW students performed among the highestperforming nations on the twelfth gradegeneral knowledge achievement tests andnear the international average on theadvanced math and physics tests.
Fourth, eighth, and twelfthgrade resultsare discussed below and reported in moredetail in appendix B.
FoUrthGrade Results
Only students in Singapore had scoressignificantly above those of FiW students onthe fourth grade math assessment. The FiWConsortium had average scores that were notsignificantly different from four othercountries (Korea, Japan, Hong Kong, andNetherlands). FiW students outperformedtheir counterparts in 21 of 26 countries.
In science, no nations outperformed fourthgrade students in the FiW. The FiW fourthgrade science score was not significantlydifferent from that of one other country
(Korea). FiW students outperformed theircounterparts in 25 of 26 countries. Seeexhibits B-1 and B-2 for more detail onscores and distributions.
Although FiW students did well on thefourth grade math test, they had strongerperformance in some topic areas than inothers. FiW fo. urthgrade math studentsperformed among the best in the world in 8of 14 content areas." They were:
(1) integers and whole number operations;(2) common fractions;(3) rounding and estimating computations;(4) geometry: position and shapes;(5) symmetry, congruence and similarity;(6) proportionality;(7) patterns, relations, and functions; and(8) data and statistics.
Topics where students from other nationsscored higher were:
(1) meaning of whole numbers;(2) decimal fractions;(3) estimating quantity and size;(4) measurement units;(5) perimeter, area, and volume; and(6) equations and formulas.
For more detail on how the FiW performedrelative to other TIMSS countries, seeexhibit B-3.
EighthGrade Results
Eighthgrade FiW students also performedvery well on the TIMSS assessmentcompared to students from the 41 countriesthat participated in this part of the study. Asin the fourth grade, only students inSingapore outperformed eighthgrade FiWstudents in math and no nationsoutperformed FiW students in science.
In math, only students in Singapore scoredsignificantly above FiW students on theeighthgrade assessment. FiW had anaverage score that was not significantlydifferent from six other countries (Korea,Japan, Hong Kong, Belgium-Flemish, CzechRepublic, and Slovak Republic). FiWeighthgrade math students outperformedtheir counterparts in the remaining 34 of 41countries.
In science, no nation outperformed FiW onthe eighthgrade assessment. The FiWscore was not significantly different fromeight other high-performing countries(Singapore, Czech Republic, Japan, Korea,Bulgaria, Netherlands, Slovenia, andAustria). See exhibits B-4 and B-5 for moredetail.
As in the fourth grade, eighthgrade mathachievement varied by topic. FiW eighthgrade math students performed among thebest in the world in 9 of 20 content areas.'2They were:
(1) decimal fractions and percentages;(2) relationships of fractions;(3) estimations of quantity and size;(4) rounding;(5) estimating computations;(6) three-dimensional geometry and
transformations;(7) patterns, relations, and functions;
8
(8) data representation and analysis; and(9) statistics and probability.
Students from other nations scored higher onthe following topics:
(1) whole numbers;(2) common fractions;(3) measurement units;(4) perimeter, area, and volume;(5) measurement estimations and errors;(6) two-dimensional geometry basics;(7) polygons and circles;(8) congruence and similarity;(9) proportionality concepts;(10) proportionality problems; and(11) equations and formulas.
See exhibit B-6 for more detail.
TwelfthGrade Results
In the twelfth grade, two sets of assessmentswere administered to the countriesparticipating in TIMSS. The first setmeasured student achievement in generalmath and science knowledge. The secondwas designed to measure achievement of themost advanced students in their final year ofsecondary school. Accordingly, theadvanced math exam covered advancedmath topics, including geometry, numbersand equations, and calculus. The advancedexam in science focused on physics.
In the United States, the advanced mathassessment was given to students who hadtaken, or were taking, a full year of a highschool course that included calculus in thetitle, including calculus, pre-calculus, APCalculus, and calculus and analyticgeometry.
Using the U.S. definition for advanced math,approximately 14 percent of the U.S. school-
19
leaving age cohort was covered by theTIMSS sample of advanced math students.Internationally, 19 percent of the school-leaving age cohort was covered, under thevarious definitions used by differentcountries to identify their most advancedstudents.
By comparison, the FiW advanced mathsample covered approximately 65 percent ofthe FiW school-leaving age cohort under thesame definition used by the U.S., clearly amuch larger percentage of students than theU.S. or its international peers."
To take the physics assessment, U.S.students had to be enrolled in, or have takenat least one year-long class of physics (thisincludes physics and AP Physics). Underthis definition, approximately 15 percent ofthe U.S. school-leaving age cohort wascovered by the TIMSS sample of physicsstudents. Internationally, approximately 13percent of the school-leaving age cohortwere covered by this assessment using thedifferent definitions of eligibility developedacross countries.
As with mathematics, a much largerpercentage of FiW twelfthgrade studentswere exposed to physics than U.S. orinternational students. In the FiW,approximately 67 percent of the school-leaving age cohort were covered by thephysics sample."
The large difference between the percent ofthe school-leaving age cohort covered in theU.S. and FiW samples is explained primarilyby differences in course taking patterns,rather than differences in the number ofstudents in this age cohort who are still inschool or differences in which componentsof the system they may have excluded fromtheir sample.
9
Eighty-three percent of FiW students takemathematics and 74 percent take science intheir last year of schooling. In the U.S.,however, less than two-thirds of all seniorsenroll in a math class and less than one-halfof U.S. seniors take a science class.
Furthermore, nearly all FiW students take atleast one of the following classes: pre-calculus, calculus, AP Calculus, physics, orAP Physics:5 Results for each set of examsare discussed below.
General Knowledge
In general mathematics knowledge, FiWtwelfth grade students' performance was not'significantly different from students in theseven highest performing countries(Netherlands, Sweden, Denmark,Switzerland, Iceland, Norway, andAustralia).
FiW students outperformed theircounterparts in the remaining 14 countries.It should be noted that no Asian countriesparticipated in the end of secondary schoolassessments. See exhibit B-7 for more detailon the countries participating in thisassessment and the distribution of theirscores.
In general science knowledge, theachievement of FiW twelfthgraders wassimilar to their achievement in mathematics.FiW twelfthgrade students' performanceon general science knowledge was notsignificantly different from students in theseven highest performing countries(Sweden, Netherlands, Iceland, Norway,Canada, New Zealand, and Australia). FiWstudents outperformed their counterparts inthe remaining 14 countries. See exhibit B-8for more detail.
Advanced Math and Physics
The TIMSS advanced math and physicsexams were designed to be administered tothe highest performing students in the worldin math and science as they are about toleave secondary school. Each countrydefined the groups of students that theythought most appropriate to be included inthese assessments, based on the generalcontent of the tests as well as practicalconsiderations.
In order to have large enough samples ofstudents taking the advanced math andscience exams, the U.S. included pre-calculus and general physics students in theadvanced groups. Other countries limitedtheir testing to calculus and advancedphysics students.
The Consortium followed the U.S. samplingparameters, even though it had a largeproportion of students who would qualifyunder the more restrictive internationalcriteria, as discussed below. We have,therefore, presented two different views ofthe advanced math and physics test data.
In advanced mathematics, for example, wepresent the FiW Advanced PlacementCalculus student score on the advancedexamination to provide comparison withother countries. We also provide the scorefor all students who took the advanced testfor comparison to the U.S. national score.
As discussed in the rest of the section, theConsortium's AP Calculus and AP Physicsstudents were first in the world. However,when the pre-calculus and general physicsstudents are incorporated into the FiWscores, the FiW averages drop below theinternational average. The section considersthe general results first, then provides the
10
results for the AP Calculus and AP Physicsstudents.
On the advanced math assessment, FiWstudents scored near the internationalaverage. Advanced FiW twelfthgrade mathstudents were outperformed by students inseven countries (France, Russian Federation,Switzerland, Denmark, Cyprus, Lithuania,and Sweden). FiW scores were notsignificantly different from those of sixcountries. FiW placed significantly abovethree countries. See exhibit B-9 for moredetail.
FiW physics students were significantlybelow students in twelve nations (Norway,Sweden, Russian Federation, Denmark,Slovenia, Germany, Australia, Cyprus,Switzerland, Greece, Canada, and France).FiW performance did not differ significantlyfrom three nations, outperforming only theUnited States in the TIMSS physicsassessment. See exhibit B-10 for moredetail.
AP Calculus and AP Physics Results
Many have wondered whether it wasappropriate to be comparing two-thirds ofthe students in FiW against one-fifth orfewer students in the U.S. or internationallybecause such large percentages of FiWtwelfth grade students were included in theadvanced math and science samples. It hasbeen suggested that AP Calculus or APPhysics students might make a bettercomparison group for the advanced mathand science assessments because thesestudents are enrolled in the most advancedcourses and similar percentages of FiW andinternational students would be coveredunder such a comparison.
21
AP courses are offered in all FiW highschools and are generally considered to bethe most advanced classes offered inmathematics or physics, since studentsparticipate in national AP exams and mayreceive college credit if they score well onthese exams. FiW AP Calculus studentsrepresent 28 percent of the FiW school-leaving population. Under this comparison,then, 28 percent of the FiW students arebeing compared against 19 percent ofinternational students and 14 percent of theirU.S. counterparts.
In AP Physics, 7 percent of the FiW school-leaving cohort is compared against 13percent internationally and 15 percent ofU.S. students.I6
FiW twelfthgrade AP Calculus studentswere first in the world, as no nationsoutperformed FiW AP Calculus students onthe advanced math assessment. The FiWscore was significantly above that of sixteennations. See exhibit B-11 for more detail.
Performance of AP Physics students in FiWConsortium was also first in the world. Nonations scored significantly above FiW APPhysics students. Five nations (Norway,Sweden, Russian Federation, Slovenia, andGermany) had scores that were notsignificantly different from FiW AP Physicsstudents. FiW AP Physics studentsoutperformed their counterparts in 11countries. See exhibit B-12 for more detail.
Summary
When benchmarked against an internationalmeasure of math and science achievement,FiW students performed exceptionally wellin all grades tested.
FiW students excelled on the fourth, eighth,and twelfth grade general knowledge tests,and scored among, or just below, the highestperforming countries worldwide. Althoughall Consortium students tested on theadvanced math and physics tests did notperform as well as expected (they scoredaround the international average), theperformance of AP students was
11
exceedingly high, with their scores placingthem in first place internationally.
The outstanding performance of the FiWstudents, particularly given thedisappointing results of their U.S.counterparts, has generated a lot of interestin examining what factors might havecontributed to this world-class performance.
The next two sections look at the influencethat home factors and differences in thecontexts for teaching and learning mighthave on these gaps.
22
Do Home Factors Explain the High Achievement of the FiW Consortium?
As mentioned earlier, the districts that makeup the FiW Consortium are high wealthdistricts. The question that natuarlly arisesis: Could student and family backgroundcharacteristics explain the differences inachievement between the FiW Consortiumand the U.S.?
To examine this question, an exploratoryanalysis was conducted using TIMSSachievement and student and familycharacteristics data." The analysisidentified a set of student and familyvariables included in the TIMSS'questionnaires that were found to be highlycorrelated with student math achievement.These variables included parents' educationlevel, whether the student's parents wereborn in the U.S., language spoken at home,and number of books in the home.
A set of regression analyses were run toestimate the difference between the FiW andU.S. scores not attributable to home andfamily characteristics (in scale points) forthe fourth and eighth grades in math andscience, and the twelfth grade for generalknowledge of math and science.
These exploratory analyses showed that thepoint differences between the FiW and U.S.students' scores could not be fully attributedto students' home and family characteristics.
In fourthgrade math and eighthgrade mathand science, socio-economic factorsexplained approximately 20 to 25 percent ofthe difference in scale scores, but left 75 to80 percent unexplained.
13 23
Family and parental characteristics couldaccount for slightly more of the difference ineighthgrade science, and twelfthgrademath and science. In these cases, half of thedifference is attributable to family and homefactors.
There are, however, some immeasurableeffects of resource rich districts that may notbe well measured by this analysisa morestable teaching force, high levels ofinvolvement from parents, and highexpectations for students.
These factors may play as important a roleas that of high financial resource levels inpromoting high achievement. In fact, somebelieve that one of the explanations for thehigh achievement levels in FiW districts ishow they use their wealth to supportteaching and learning, not the wealth itself
While further analysis may give moreprecise estimates of the relationship betweenachievement and socio-economic status inthe FiW Consortium, this analysis andresults from other preliminary analyses'indicate that other factors, such ascurriculum, classroom instructionalpractices, and teacher engagement, play animportant role in their high achievementlevels.
The next section explores some of the otherfactors that may have contributed to theConsortium's success.
What is the Context for Teaching and Learning Math in the FiWConsortium?
This section offers some possibleexplanations that might account for thedifferences in math achievement betweenthe FiW and the U.S. that cannot beattributed to home factors.
Drawing on data from the TIMSS teachersurveys, an analysis of textbooks done byresearchers at Michigan State University,and anecdotal information based on visits toFiW schools, this section discussesdifferences in the contexts for teaching andlearning between the FiW Consortium andthe U.S.
Where international data are readilyavailable, compariions are also madebetween the contexts for teaching andlearning in the FiW and those in countrieswith high math achievement to point outsimilarities or differences where they exist.
In order to simplify the analysis, the sectionconcentrates only on the differences in thecontexts for learning math. An examinationof differences in the contexts for learningscience may show different results.
In addition, the section only looks at thecontexts for learning math in the fourth andeighth grades. These grades were selectedbecause of the rich data available fromTIMSS for analyzing math instruction atthese levels:9
The findings presented in this section shouldbe viewed as an exploration of somepossible reasons for the high achievement inthe FiW Consortium. This report is a firstattempt to identify factors that may play animportant role.
14
Further analysis may reveal other factorsthat might explain the differences inachievement between the FiW and the U.S.In addition, the paper does not investigatepossible causal links between the factorsdiscussed and the achievement gaps.
Finally, because of data limitations, thispaper points only to interesting differencesinpatterns between the contexts for learningin the FiW, U.S. and, when data areavailable, other high performing countries.20
Drawing on data from TIMSS, this sectionlooks at differences in four key areas that arecommonly thought to have a large impact onachievement levels:
Curriculum. In particular, differences ininstructional topic coverage and thecoverage of different topics by textbooksare explored.
Classroom instructional practices. Thereport looks at both student and teacherreports on the most frequently used classactivities and classroom organizationalmethods.
Teacher engagement. Four factors areexamined: the time spent by teachers onschool-related activities during personaltime, frequency of teacher meetings,teacher influence over key schooldecisions, and teacher familiarity withkey curricular and assessmentdocuments.
Teaching environment. Finally, teacherreports on the factors that limit theirteaching ability are compared.
24
Each of these areas is discussed in moredetail in the next section.
Curriculum
It is a common sense conclusion that what istaught in classrooms around the world hasan impact on what is learned. Accordingly,differences in the math curriculum coveredin the FiW and the U.S. may account forsome of the variations in achievementbetween the FiW and U.S. students.
In particular, differences in the organizationand sequence of topics, the level and depthof coverage, and the timing of topicintroduction may reflect importantdifferences in the opportunities for FiW andU.S. students to master math and sciencematerial.
The textbooks and other curricular materialsused may also play a potentially importantrole, if they are found to be more closelyaligned with the curriculum of highachievement nations. This section presentsdata on teacher reports on instructional topiccoverage and textbook content and use.
Instructional Topic Coverage
One of the first insights from the TIMSSdata has been the lack of rigor in the U.S.math curriculum. Many educationalresearchers and policymakers haveexpressed concern that U.S. students are notbeing taught the concepts and skills neededto achieve world class standards.
Accordingly, one possible explanation fordifferences in math achievement betweenthe FiW and the U.S. might be variations ininstructional topic coverage, i.e., whatcontent teachers cover in class.
15
Differences in both the number and type ofmath topics covered throughout the schoolyear may indicate that fourthand eighthgrade FiW students are exposed to moreadvanced material than their U.S.counterparts.
According to TIMSS data, FiW students aremore likely than their U.S. peers to beintroduced to relatively more advanced mathtopics in earlier grades. This pattern isreflected in data reported by math teachersin the fourth and eighth grade.
Exhibit.1 presents detailed data oninstructional topic coverage for 21 mathtopics in fourth-grade math.21 In a numberof key subject areas, all or nearly all FiWand U.S. students are introduced to thesesubjects by the end of the fourth grade,according to their math teachers.
Some of these students cover these topics intheir entirety in earlier grades (indicated onthe exhibit as "learned earlier"), others begintopics in earlier grades and receivereinforcement on the material during fourthgrade (indicated as "reinforcedinformation"), while still others are firstintroduced to the material in fourth grade(indicated as "new information").
In topic areas where these three categoriessum to 100, teachers report that all fourthgrade math students will have beenintroduced to these subjects prior to, orduring, the current year.
All, or nearly all, FiW and U.S. fourth grademath students are introduced to thefollowing five basic concepts prior to, orduring, fourth grade, according to teacherreports (that is, the percentage of students in"learned earlier", "reinforced information"
and "new information" sum to 100 or near100):
(1) whole numbers;(2) common and decimal fractions;(3) estimation and number sense;(4) measurement units and processes; and(5) data representation and statistics.
In all of these subjects, FiW students weremore likely to have been exposed to thetopic earlier than their U.S. counterparts.For example, 75 percent of FiW studentshad either learned whole numbers prior tofourth grade (e.g., covered it in grade threeor earlier and were no longer spending timeon this concept) or were receivingreinforcement.
In contrast, less than half of U.S. students(43 percent) had been exposed to thismaterial earlier, according to their teachers.
FiW students are also more likely than theirU.S. fourthgrade counterparts to beintroduced to two more difficult mathconcepts during or prior to fourth grade:
(1) percentages, and(2) number sets and concepts.
Seventy-seven percent of FiW students havebeen introduced to percentages prior to orduring fourth grade (calculated by addingthe three categories indicating when topicsare introduced: e.g. "learned earlier" +"reinforced information" + "newinformation").
This compares to only 47 percent of theirU.S. peers. Sixty-three percent of FiWstudents have been introduced to numbersets and concepts prior to or during fourthgrade, compared to only 32 percent of theirU.S. peers.
16
As illustrated in exhibit 2, according toteacher reports, all or nearly all eighth gradeFiW and U.S. students had been introducedto the following eleven math topics:22
(1)(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
whole numbers;common and decimal fractions;percentages;number sets and concepts;number theory;estimation and number sense;measurement units and processes;perimeter, area, and volume;basics of one- and two- dimensionalgeometry;
(10) ratio and proportion; and(11) equations, inequalities and formulas.
However, as in the fourth grade, FiWstudents are more likely than their U.S.counterparts to have been exposed to mostof these basic math concepts in precedinggrades. As a result, FiW students are morelikely to either not cover these topics againin the eighth grade or spend time reinforcingthese concepts.
For example, according to their teachers,eighthgrade FiW math students were muchmore likely than their U.S. counterparts tohave covered whole number operations andmeaning in earlier grades and moved beyondthis material. Forty percent of FiW studentscovered this material in earlier grades; whileonly 16 percent of U.S. students covered thismaterial earlier and were not covering it ineighth grade according to their teachers.
For the other ten subjects, FiW studentswere much more likely than U.S. students tohave been exposed to the material in priorgrades and to spend time in their eighthgrade math classes reinforcing this material.
26
Eighthgrade students in the FiW were alsomore likely than their U.S. counterparts tohave been introduced to new, moreadvanced material during the eighth grade.The more advanced material includes:
(1) geometric congruence and similarity;(2) geometric transformations and
symmetry;(3) constructions and three-dimensional
geometry;(4) proportionality: slope, trigonometry, and
interpolation;(5) functions, relations and patterns; and(6) sets and logic.
17
While the data on instructional coveragegive us some clues as to when FiW and U.S.fourthand eighthgraders are introduced todifferent topics, the data do not providemuch insight into the level or depth ofcoverage. Further research in this area couldpotentially give a fuller understanding of thedifferences in implemented curriculumbetween the FiW and the U.S.
27
00
Exh
ibit
1: F
ourt
h-G
rade
Mat
h T
each
ers'
Rep
orts
on
Inst
ruct
iona
l Top
ic C
over
age
FiW
Who
le N
umbe
rsu.
s.
Com
mon
& D
ecim
alFi
W
Frac
tions
u.s.
FiW
Perc
enta
ges
U.S
.
Num
ber
Sets
& F
iWC
once
pts
u.s.
FiW
Num
ber
The
ory
u.s.
Est
imat
ion
& F
iWN
umbe
r Se
nse
u.s.
Mea
sure
men
t Uni
ts &
FiW
Proc
esse
su.
s.
Est
imat
ion
& E
rror
of
FiW
Mea
sure
men
tu.
s.
Peri
met
er, A
rea,
& F
iWV
olum
e u.
s.
Bas
ics
of O
ne &
Tw
o Fi
WD
imen
sion
al G
eom
etry
u.s.
28C
ongr
uenc
e &
FiW
Sim
ilari
tyu.
s.
0
Perc
ent o
f Fo
urth
-Gra
de S
tude
nts
2040
6080
100
45,7
1111
-,%
;;;
Tra
nsfo
rmat
ions
& F
iWSy
mm
etry
u.s
.
0
Perc
ent o
f Fo
urth
-Gra
de S
tude
nts
2040
6080
Thr
ee D
imen
sion
al F
igur
esFi
Vi
& C
onst
ruct
ions
u.s.
g
FiW
Rat
io &
Pro
port
ion
us
Func
tions
, Rel
atio
ns, &
FaV
Patte
rns
u.s.
FiW
Equ
atio
ns &
for
mul
asU
.S.
Dat
a R
epre
sent
atio
n &
Fw
In
Stat
istic
su.
s
FiW
Prob
abili
tyu.
s.
Sets
& L
ogic
FiW
1
u.s.
b
FiW
Prob
lem
Sol
ving
Str
ateg
ies
FiW
Oth
er M
athe
mat
ics
Con
tent
u.s.
Lea
rned
Ear
lier
El R
einf
orce
d In
form
atio
nN
ew I
nfor
mat
ion
SOU
RC
E: N
CR
EL
Ana
lysi
s of
Tea
cher
Que
stio
nnai
re D
ata.
29
Exh
ibit
2: E
ight
h-G
rade
Mat
h T
each
ers'
Rep
orts
on
Inst
ruct
iona
l Top
ic C
over
age
Who
le N
umbe
rsF
iW
U.S
.
Com
mon
& D
ecim
alF
iW
Frac
tions
u.s.
Perc
enta
ges
rox
U.S
.
Num
ber
Sets
& C
once
pts
riw
U.S
.
Num
ber
The
ory
FiW
U.S
.
Est
imat
ion
& N
umbe
rF
iW
Sens
eu.
s.
Mea
sure
men
t Uni
ts &
FN
Proc
esse
su.
s.
Est
imat
ion
& E
rror
of
EN
Mea
sure
men
tsu.
s.
Peri
met
er, A
rea,
& F
wV
olum
eu.
s.
Bas
ics
of O
ne &
Tw
oF
N
Dim
ensi
onal
Geo
met
ryu.
s.
Geo
met
ric
Con
grue
nce
&F
iW
Sim
ilari
tyu.
s.
30
0
Perc
ent o
f E
ight
h-G
rade
Stu
dent
s
2040
6080
100
EM
IL
j
1
1
1
1
Geo
met
ric
Tra
nsfo
rmat
ions
FiW
0
Perc
ent o
f E
ight
h-G
rade
Stu
dent
s
2040
6080
& S
ymm
etry
U.S
. Mr7
1
Con
stru
ctio
ns &
Thr
eeD
imen
sion
al G
eom
etry
Rat
io &
Pro
port
ion
FiW
U.S
.W
.;
FiW
U.S
.
Prop
ortio
nalit
y: S
lope
,F
iW
Tri
gono
met
ry &
Int
erpo
latio
nu.
s.
Func
tions
, Rel
atio
ns,
EN
& P
atte
rns
U.S
.
Equ
atio
ns, I
nequ
aliti
es, &
FiW
Form
ulas
u.s.
Stat
istic
s &
Dat
aF
iW
Prob
abili
ty &
Unc
erta
inty
U.S
.
FiW
U.S
.
Sets
& L
ogic
FiW
U.S
.
Prob
lem
Sol
ving
Str
ateg
ies
FiW
U.S
.
Oth
er M
athe
mat
ics
Con
tent
FiW us
.
I
IN L
earn
ed E
arlie
rE
a R
einf
orce
d In
form
atio
nN
ew I
nfor
mat
ion
SOU
RC
E: N
CR
EL
Ana
lysi
s of
Tea
cher
Que
stio
nnai
re D
ata.
3-0
Textbook Use and Topic Coverage
Could the difference in achievement beinfluenced by variations in the waytextbooks are used in the classroom or thetextbooks chosen? This section looks atTIMSS questionnaire data and a specialstudy commissioned by FiW to examine thedegree to which teachers rely on theirtextbooks to structure their teaching time, aswell as the types of topics covered in thesetextbooks and the relative emphasis given toeach topic.
TIMSS teacher questionnaires indicate thatnearly all students use a textbook in FiWand U.S. eighthgrade math classes (100percent of FiW students and 97 percent ofU.S. students). The degree to which eighthgrade math teachers use these books tostructure their teaching time varies
considerably, however, according to teacherreports. FiW students are more likely thanU.S. students to be in classes where a largepercentage of the teaching time is based onmaterial in the text. As seen in exhibit 3, 55percent of eighthgrade students are inclasses where more than three-fourths of theteaching time is based on material in thetextbook. In contrast, only 36 percent ofU.S. students are in classes that rely thisheavily on material in the textbook.
FiW teachers also report using aconsiderable amount of supplementarymaterial in addition to their main textbook.As shown in exhibit 3, 91 percent of FiWand all U.S. students have teachers who useother materials in the place of or in additionto their main textbook. In both cases,textbooks appear to be used as one resourceout of many rather than the sole resource.
Exhibit 3: Teachers' Reports on Use of Textbooks and Other TeachingMaterials and Percent of Mathematics Teaching Time Based on Textbook
Textbook UsePercent of Eighth
Grade StudentsFiW U.S.
Use a textbook at all 100 97Use supplementary materials 91 100
Percent of teaching time based on textbook1-25 percent 6 16
26-50 percent 4 17
51-75 percent 35 31
76-100 percent 55 36SOURCE: NCREL analysis of TIMSS Data; FiW Teacher Questionnaire results.
3220
Data from the TIMSS curriculum study doneby researchers at Michigan State Universitycatalogued the topics covered by differenttextbooks and the relative emphasis given toeach topic. Their analysis of math textbooksused by the FiW, U.S. and TIMSS23countries report the following conclusions:
"The number of topics in Consortiumtextbooks is similar to that in textbooksof the US composite at all populations.[grade levels]. "24
Fourth Grade
At the fourthgrade level, "theconsortium's textbooks are essentiallyno different in their content profiles fromthe US as a whole and [they also have] agreat deal of overlap in content withJapan."
The major exception is "the prominentpresence of decimals in the Japanesetextbooks, which is not the case in U.S.or FiW textbooks."25
Eighth Grade
"At the eighthgrade level, there arenotable differences in the topicsemphasized in the Consortium...with thetopic of 'Equations and Formulas' forexample, more emphasized in theConsortium than is the case for the mostcommonly used textbooks in the
"The emphasis of Consortium textbookson 'Equations and Formulas' in Grade 8is similar to that of TIMSS countries inwhich mean student achievement wassignificantly higher than mean studentachievement in the U.S."'
21 33
"This greater focus and emphasis onalgebra is further reflected by the factthat the Consortium's books for the non-algebra tracks do not have any of thestandard arithmetic topics among the top5 contained in the book. This is inmarked contrast to U.S. non-algebrabooks."
"The Consortium teachers clearlyemphasize algebra (linear equations) andgeometry (2D geometry basics) morethan is common for their U.S. peers, anemphasis that appears to be aided by thefact that the textbooks themselvesprovide more material in these areas."28
In sum, some differences in curriculum maycontribute to the differences in achievementbetween the FiW and the U.S. At both gradelevels, these differences do not seem to bedriven by differences in the number of topicsfound in textbooks used by FiW and U.S.students, as the number of topics covered byFiW math textbooks is very similar to thenumber covered by other U.S. mathtextbooks.
However, FiW eighthgrade textbooks tendto emphasize algebra and geometry moreheavily, while U.S. textbooks emphasizearithmetic. This is true even for the studentsin the non-algebra tracks. In addition, at theeighthgrade level, a higher percentage ofFiW students than U.S. students are likely tobe in classrooms where most of the teachingtime is centered on material in the book.
However, there is some anecdotalinformation that indicates FiW teachers aremore apt to customize the textbooks used,e.g. carefully selecting chapters andexercises to be completed. This might be incontrast to the typical use of textbooks and
could be an important component inunderstanding how FiW teachers can usetextbooks with a large number of topicseffectively.
Summary
Differences in achievement do not appear tobe driven by the number of topics covered inthe classroom or in the textbooks.According to teacher reports, FiW studentsseem to cover just as many topics as theirU.S. peers. Likewise, FiW textbooks coverthe same number of topics as U.S.textbooks.
However, FiW eighth graders are morelikely than their U.S. peers to use textbooksthat emphasize algebra and geometry. Inaddition, FiW students may be introduced tomore advanced topics earlier than their U.S.peers.
Additional analysis is necessary todetermine whether more advanced conceptsare actually covered or just presented in asimplistic way.
Nevertheless, according to teacher reports,FiW students receive more reinforcement oftopics introduced in earlier grades and covermore new material.
This pattern is found in both the fourthandeighthgrades, and as might be expected, ismore pronounced in the eighth grade.
The next section explores whether there arenotable differences between the instructionalstrategies used by FiW and U.S. teachersthat may also contribute to the differences inachievement.
22
Classroom Instructional Practices
There is a rich research base regarding whathappens in most classrooms in the U.S.While much of the literature documentsestablished patterns of teacher-studentinteractions, some studies have examinedthe relationship between classroominstructional practices and studentachievement. The indicators selected foranalysis in this section are based on thatliterature and some of the unique practices inFiW districts.
As discussed in the following three sections,data from the TIMSS teacher and studentsurveys suggest that some of the explanationfor the differences in achievement may bedue to differences between the methods thatFiW and U.S. teachers use to presentmaterial to their students.
In particular, data on teachers' reports onclass activities and classroom organization,as well as students' reports on the same,indicate that notable differences may existbetween the FiW and the U.S. in the contextfor learning in student classrooms.
Teachers' Reports on Class Activities
Differences in classroom activities, that is,the actual tasks that teachers require theirstudents to do in class, may account forsome of the variations in achievement.
For example, some teachers may ask theirstudents to tackle more challenging tasks orcomplex math problems that requirestudents to apply their skills to differentproblems. Other teachers may challengestudents to explain the reasoning behind newconcepts more frequently.
34
Some students may spend more time usingcomputers to apply new math concepts tosolve exercises or problems. Other studentsmay spend their time completing lesschallenging tasks, like completing drills orpracticing basic computational skills.
This section explores whether differencesexist between the types of activities usedfrequently in the FiW and U.S. math classes.
Data on the types of activities that teachersreport they ask their students to do in classindicate that important differences may existbetween the classroom activities in FiW andU.S. classes. However, the data also showmany similarities. Both the differences andsimilarities are discussed below.
According to teachers, FiW math students inboth the fourth and eighth grades are askedto perform reasoning tasks more frequentlyand complete drills less frequently than U.S.students.
As shown in exhibit 4, FiW students aremore likely than U.S. students to have mathteachers who ask them to explain thereasoning behind an idea during "everylesson." Thirty percent of FiW fourthgrade students have math teachers who askthem to explain their reasoning during"every lesson," 21 percent of U.S. fourthgraders fall into this category.
At the eighthgrade level, the difference ismore pronounced: 48 percent of FiWstudents have teachers who ask theirstudents to explain the reasoning behind anidea during "every lesson," while roughlyhalf as many (23 percent) of U.S. studentshave teachers that do.
FiW students are also more likely than U.S.students to be asked by their teachers to
write equations during "most lessons" or"every lesson." In eighth grade, threefourths of FiW students have math teacherswho ask their students to write equationsduring "most lessons" or "every lesson."
The percentage of U.S. students being askedto express relationships in equations in"most lessons" or "every class" is farloweronly 38 percent of eighthgraders.
While the difference between FiW and U.S.eighthgrade students may be a function ofthe higher percent of FiW students takingalgebra classes, the difference is notable.
In the fourth grade, FiW and U.S. studentsare asked to use equations to expressrelationships less frequently, however FiWstudents are more likely than their U.S.counterparts to be asked to write equations(94 percent of FiW students, 83 percent ofU.S. students).
Most fourthand eighthgrade students, inboth the FiW and the U.S., are only asked toundertake more complex or challengingactivitiessuch as representing andanalyzing relationships using tables, charts,or graphs or working on problems for whichthere is no immediate solutionduring"some lessons."
Roughly 80 percent of FiW and U.S. fourthgrade math students and 70 percent of FiWand U.S. eighth grade students spend timeduring "some lessons" representing andanalyzing relationships using tables, charts,or graphs.
In both the fourth and eighth grades,teachers reported roughly 65 percent of FiWand U.S. students are asked by their teachersto work on problems for which there is noimmediate solution during "some lessons."
23 35
According to teacher reports, FiW studentsalso practice computational skills in classless frequently than their U.S. counterparts.Teachers report that 58 percent of FiWfourth graders practice their computationskills during "most lessons" or "everylesson" compared to 70 percent of their U.S.counterparts.
In the eighth grade, students in both FiWschools and schools across the U.S. are lesslikely to practice computational skillsfrequently. However, the differencebetween the FiW and the U.S. eighth gradersis even more striking: 25 percent of FiWmath students practice computational skillsduring "most lessons" or "every lesson",compared to 59 percent of U.S. students.
The differences between how often FiW andU.S. students are asked to practicecomputational skills could be a function ofhigher expectations of mastery in earliergrades without the need for teaching and re-teaching the same topics year after year.
Classroom computer usage is another areawhere notable differences exist betweenFiW and the U.S. Neither FiW nor U.S.students use computers routinely to solveexercises or problems. However, in both thefourthand eighthgrades, FiW teachersreport that over half of the students usecomputers during "some lessons," whileonly a little over one-third of U.S. fourthgraders, and less than a quarter (21 percent)of U.S. eighthgraders, use computers.
While these data point out interestingcontrasts, the differences in computer usemay be a function of the availability oftechnology. A further look needs to betaken at how computers are actually used inFiW classrooms, as compared to the U.S.
24
Finally, one additional key point that can bemade after reviewing exhibit 4 is the pattern,or lack of pattern, that emerges in FiW andU.S. classrooms.
It is interesting to note that the mostpredominant activities in FiW fourth gradeclassroomsexplaining the reasoningbehind an idea, practicing computationalskills, and writing equationsare consistentwith the U.S. patterns.
At the eighthgrade level, however, there isa striking difference between FiW and U.S.reports in the areas of writing equations andpracticing computational skills.
Taken together, these data suggest that FiWand U.S. teachers ask their students to dosimilar things in their math classes.However, there are a number of noteworthydifferences.
In general, FiW students are challenged toperform reasoning tasks more frequently andcomplete drills less often than their U.S.peers. As noted above, the differences areparticularly evident at the eighthgradelevel, where FiW students are morefrequently required to write equations, ratherthan practice their computational skills.
One could reasonably conclude that studentsin FiW eighthgrade classrooms areexperiencing very different content andinstruction from students in U.S. classrooms.
36
Exhibit 4: Teachers' Reports on How Frequently StudentsAre Asked to Complete Specific Tasks
Response
Percent ofFourth
_Grade
Students
Percent ofEighthGrade
StudentsFiW U.S. FiW U.S.
Explain the reasoning behind an idea
Every lesson 30 21 48 23
Most lessons 56 50 28 44
Some lessons 14 28 24 32
Never or almost never 0 1 0 1
Represent and analyze relationships usingtables, charts, or graphs
Every lesson 0 1 0 2
Most lessons 5 7 21 10
Some lessons 86 81 72 73
Never or almost never 9 10 7 15
Work on problems for which there is noimmediate...solution
Every lesson 1 1 0 2
Most lessons 12 6 17 10
Some lessons 64 59 68 65
Never or almost never 22 35 14 24
Use computers to solve exercises orproblems
Every lesson 0 1 0 2
Most lessons 1 1 0 2
Some lessons 64 39 56 21
Never or almost never 36 60 44 76
Write equations to represent relationships
Every lesson 4 6 23 6
Most lessons 25 22 54 32
Some lessons 65 55 23 58
Never or almost never 5 18 0 5
Practice computational skills
Every lesson 20 24 7 21
Most lessons 38 46 18 38
Some lessons 40 29 56 31
Never or almost never 2 1 18 11
SOURCE: NCREL analysis of lEA's Third International Mathematics and Science Study (TIMSS) data 1994-95;FiW Teacher Questionnaire results, NCREL; tables 5.10 and 5.17 in Mullis, I.V.S., et al. (1997). MathematicsAchievement in the Primary School. Years: lEA's Third International Mathematics and Science Study. Chestnut Hill,MA: Center for the Study of Testing, Evaluation, and Educational Policy, Boston College; tables 5.10 and 5.17 inBeaton, A.E., et al. (1996). Mathematics Achievement in the Middle School Years: lEA's Third InternationalMathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and EducationalPolicy, Boston College.
NOTE: Totals may not add to 100 due to rounding.
25 Q I
Teachers' Reports on ClassroomOrganization
The different methods that math teachersemploy to organize their classrooms and thedifferent pedagogical approaches they usemay have an impact on the achievement oftheir students.
In particular, the amount of time teachersspend standing up in front of their classintroducing new material, explaining newconcepts, and answering student questions,as opposed to having students work throughexercises on their own, may play animportant role in explaining the differencesin achievement between FiW and U.S.students.
Similarly, whether students work together asa class or break off into small groups mayaffect student achievement. Furthermore,the choice of organizational or pedagogicalapproach that may have the strongest impacton the achievement of younger students maybe different than the most effective choicefor older students.
Accordingly, this section examines TIMSSteacher data on classroom organization forany differences in patterns between the FiWand the U.S. for both fourthand eighthgraders.
Teachers' reports on classroom organizationshow some differences between the FiW andthe U.S., and some similarities between FiWand countries with high math achievement.
As shown in exhibit 5, FiW fourth grademath teachers reported that FiW students areless likely than their counterparts in both theU.S. and high performing countries to betaught by teachers that rely either heavily or
26
predominantly on any one classroomorganizational approach.
In fact, the method that is the most heavilyrelied upon for teaching FiW fourthgrademath studentsstudents working together asa class with their math teacher leading thewhole classis used during "most or everylesson" for less than half of fourthgradeFiW students.
FiW fourth graders are considerably lesslikely than both U.S. students and theircounterparts in Japan, Korea, and Singaporeto spend "most or every lesson" workingindividually with assistance from their mathteacher, according to teacher reports.
This is the one area where there are notabledifferences between FiW and the U.S.:twentyfour percent of FiW students areasked to take this approach, compared to 55percent of U.S. students, according to theirteachers.
In the eighth grade, classroomorganizational patterns are different fromthose in the fourth grade. FiW teachersreported that three quarters of their studentswork together as a class with the mathteacher teaching the whole class during"most or every lesson."
The strong emphasis on whole classinstruction was also reported in othercountries with high achievement. Accordingto teacher reports, over 60 percent of eighthgrade math students in Singapore, Korea,and Japan spend "most or every lesson"working together with the math teacherteaching the whole class.
By contrast, less than half of eighthgrademath students in the U.S. spend "most orevery lesson" receiving instruction in largegroups. U.S. teachers also report that half of
38
eighthgraders spend "most or every lesson"working individually with assistance fromteachers.
This is a much higher percentage than in theFiW Consortium, where only 35 percent arein classes where this technique is used asfrequently.
To summarize, TIMSS teacher data suggestthe existence of important differencesbetween FiW and U.S. and similaritiesbetween the FiW and high mathachievement countries in the context forteaching math.
According to these data, fourthgrade FiWstudents are more likely than both their U.S.
and international peers to be taught bymultiple methods of instruction.
FiW eighthgraders, however, spend moretime receiving whole class instructioninthis case, the pattern is similar to theirinternational peers but differs considerablyfrom that found in the U.S.
Again, one might conclude that the FiWstudents are being exposed to more newcontent in the eighth grade than U.S.students and that higher expectations areimposed on them to learn more complexmathematics.
Exhibit 5: Teachers' Reports on Classroom Organization During MathematicsLessons
Country
Percent of Students Whose Teachers Report Using Each OrganizationalApproach "Most or Every Lesson"
WorkTogether as a
Class withStudents
Respondingto One
Another
Workas
Class withTeacher
Teaching theWhole Class
WorkIndividually
withAssistance
from Teacher
WorkIndividually
withoutAssistance
from Teacher
Work inPairs orSmall
Groupswith
Assistancefrom Teacher
Work inPairs orSmall
Groupswithout
Assistancefrom Teacher
Fourth Grade *FiW 37 48 24 26 36 23
United States 32 54 55 15 20 11
Korea 50 77 57 37 30 20
Japan 50 78 34 25 7 2
Singapore 23 68 37 41 25 10
Eighth Grade *FiW 42 75 35 22 20 16
United States 22 49 50 19 26 12
Korea 39 89 41 30 12 11
Japan 22 78 27 15 7 1
Singapore 15 61 48 27 20 6
SOURCE: FiW Teacher Questionnaire results, NCREL; figure 5.5 in Mullis, I.V. ., et al. (1997). MathematicsAchievement in the Primary School Years: lEA's Third International Mathematics and Science Study. Chestnut Hill,MA: Center for the Study of Testing, Evaluation, and Educational Policy, Boston College; figure 5.3 in Beaton, A.E.,et al. (1996). Mathematics Achievement in the Middle School Years: lEA's Third International Mathematics andScience Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and Educational Policy, BostonCollege.
* Fourth/eighth grade in most countries. 27
39
Students' Reports on ClassroomOrganization and Class Activities
TIMSS also collected data from students onwhat happens in the classroom. As in theteacher questionnaires, the studentquestionnaires asked students to report onhow frequently different classroomorganization methods were used and howoften varying class activities wereundertaken
This section looks at FiW and U.S. studentreports on what happens in their mathclassrooms in order to get a fullerunderstanding of the varying classroomorganizational and activity patternscommonly used, as well as some insight intothe differences between teachers' andstudents' reports.
FiW and U.S. students' reports on classroomorganization and class activities provide aninteresting picture of the similaritiesbetween what goes on in FiW and U.S. mathclasses. Exhibit 6 presents data on students'reports on the frequency of math classactivities. As shown in the exhibit, fourthand eighth grade students in FiW and U.S.schools report similar patterns for the fourmost frequently emphasized activities intheir math classes.
These activities are:
(1) teacher demonstrations of how to domath problems;
(2) distribution of homework;(3) teacher checking of homework (fourth
grade)/class discussions of completedhomework (eighth grade); and
(4) students working from worksheets ortextbooks on their own.
28
The majority of students in both the FiWand the U.S. report that they perform theseactivities during "most lessons."
Although important similarities emerge,students also reported differences betweenthe FiW and U.S., and between the fourthand eighth grades.
The relative importance placed on twocommon activities, copying notes from theboard and taking quizzes or tests, illustratessome of these differences. In the fourthgrade, 32 percent of U.S. students reportedthat they copy notes from the board during"most lessons" compared to 22 percent ofFiW students.
In the eighth grade, this technique is reliedupon more frequently than in fourthgradeclassrooms, in both FiW and U.S. mathclasses. Interestingly, unlike in the fourthgrade, eighthgrade FiW math students aremore likely than their U.S. counterparts toreport that they copy notes from the boardduring "most lessons" (47 percent of FiWstudents versus 42 percent of U.S. students).
As for taking quizzes or tests, differencesalso exist between the FiW and the U.S. inboth the fourth and eighth grades. In thefourth grade, U.S. students are more likelythan FiW students to report that they aretested during "most lessons" (48 percent ofU.S. students compared to 32 percent ofFiW students).
In the eighth grade, however, FiW studentsreported that they are more likely to betested during "most lessons" than their U.S.counterparts (46 percent of FiW studentscompared to 39 percent of U.S. students).
FiW fourthgrade math students reportedthat they used calculators more frequentlythan their U.S. counterparts, however, most
40
FiW students (75 percent) reported thatcalculators were only used during "somelessons." For the U.S., 46 percent ofstudents reported that calculators are usedduring "some lessons," while almost fourout of ten U.S. students (39 percent)reported that they are never used.
Calculator usage was reported morefrequently in the eighth grade than in thefourth grade by both FiW and U.S. students,although the gap between FiW and the U.S.remains. Sixty-nine percent of eighthgradeFiW students reported that calculators wereused during "most lessons," while only 38percent of U.S. students reported thatcalculators were used "most lessons." Thiscould either be a resource issue, (i.e.availability of calculators), or a function ofthe content and topic coverage in mathclassrooms.
Other differences between the FiW and theU.S. exist in how homework is assigned andused. In both the fourth and eighth grades,FiW students are more likely than U.S.students to report that they have homeworkassigned regularly.
In the fourth grade, approximately three-fourths of FiW students, versus two-thirds ofU.S. students, report that their math teachergives them homework during "mostlessons." They are also more likely thantheir U.S. counterparts to be able to starttheir homework in class during "mostlessons" or "some lessons" (83 percent ofFiW fourthgraders versus 69 percent ofU.S. fourthgraders) and to discuss theircompleted homework in class during "mostlessons" or "some lessons" (87 percent ofFiW fourthgraders versus 72 percent ofU.S. fourthgraders).
In the eighth grade, differences persist inhow homework is handled in class. Exhibit6 also shows a higher percentage of FiWeighthgrade math students reported thathomework is assigned during "most lessons"(86 percent of FiW eighthgrade studentsversus 72 percent of U.S. eighth graders).
Classroom time spent on homework-relatedactivities follows a slightly different patternin eighth grade than in fourth grade. In theeighth grade, FiW students are much morelikely than U.S. students to report that theydiscuss their completed homework in classduring "most lessons" (72 percent of FiWeighthgraders versus 54 percent of U.S.eighthgraders).
In both FiW and U.S. math classes, eighthgrade math students reported that they areless likely to spend time in class working onhomework than in the fourth grade. But,unlike in the fourth grade, FiW eighthgradestudents are less likely than their U.S.counterparts to report that they can start theirhomework in class during "most lessons"(32 percent of FiW eighthgraders versus 50percent of U.S. eighthgraders).
As discussed earlier in this section, itappears as if most FiW fourthand eighthgrade students have the same types ofexperiences in their math classes as theirfellow U.S. students.
According to student accounts, the mostfrequently emphasized activities in the FiWand the U.S., in both fourth and eighthgrades, are:
(1) teacher demonstrations;(2) homework distribution, checking or
discussion of homework; and(3) independent work on worksheets or on
material in the textbook.
294 1
The three areas in eighth grade wherestudents report differences between the U.S.and the FiW are:
(1) calculator use;(2) beginning hoMework in class; and(3) discussing completed homework.
These reports support the findings reportedin the two preceding sections, and suggestthat important differences between whatFiW and U.S. students do in class may exist.
Exhibit 6: Students' Reports on the Frequency of Math Class Activities
Activity Frequency
Percent ofFourth-Grade ,
Students
Percent ofEighth-Grade
StudentsFiW U.S. FiW U.S.
The teacher shows us how to do math problemsMost Lessons 72 73 83 78
Some Lessons 27 25 16 21
Never 1 2 2 1
We copy notes from the boardMost Lessons 22 32 47 42Some Lessons 61 48 43 49Never 17 20 10 .9
We have a quiz or testMost Lessons - 32 48 46 39Some Lessons 65 47 53 61
Never 2 5 1 1
We work from worksheets or textbooks on our ownMost Lessons 56 55 69 59Some Lessons 40 35 29 38Never 4 10. 2 3
We use calculatorsMost Lessons 16 15 69 38Some Lessons 75 46 28 50Never . 9 39 2 11
We use computersMost Lessons 6 16 3 4
Some Lessons 35 29 34 28Never 58 55 63 67
The teacher gives us homeworkMost Lessons 74 66 86 72
Some Lessons 26 30 14 27Never 1 4 1 2
We can begin our homework in classMost Lessons 40 36 32 50Some Lessons 43 33 64 44Never 16 31 4 7
The teacher checks our homeworkMost Lessons 64 67 58 56
Some Lessons 30 25 35 37
Never 6 8 7 7
We discuss our completed homeworkMost Lessons 40 35 72 54
Some Lessons 47 37 24 37
Never 13 28 4 9
SOURCE: NCREL analysis of TIMSS data; FiW Student Questionnaire results.NOTE: Totals may not add to 100 due to rounding.
*The fourth grade and eighth grade student questionnaires contained a set of slightly different possible responses tothis question. To make comparisons across grades, some categories from the grade 8 survey were combined. Thesenew categories, along with the remaining responses were then matched to similar items on the grade 4 survey.Accordingly, "Some Lessons" is the sum of responses to "Pretty Often" and "Once in a While" on the grade 8 survey.Also, "Most Lessons" is "Almost Always" on the grade 8 survey.
30 42
Summary
In summary, data on instructional practicesindicate that there are differences betweenFiW and U.S. fourth and eighthgrademath classes. According to students, themost common activities in both the FiW andthe U.S. show similarities, with studentsreporting that the four most frequently usedactivities were:
(1) teacher demonstrations of how to domath problems;
(2) teacher assigning of homework;(3) teacher checking of homework (grade
4)/class discussions of homework (grade8); and
(4) students working from worksheets ortextbooks on their own.
Teacher and student reports indicate thatFiW and U.S. math teachers rely on differentmethods when demonstrating how to domath problems.
In the fourth grade, FiW teachers rely on avariety of approaches for teaching, so thatno one method dominates. Perhapsreflecting the different way in whichyounger students learn and process material,some instructional time is spent in largegroups, some in small groups, and someindividually.
In the eighth grade, however, FiW teachersreport that the most frequently used methodinvolves the math teacher instructing thewhole class. This approach is used far morefrequently than in U.S.eighth grade mathclasses.
In addition, in the FiW, both fourthandeighthgrade students are more likely thanU.S. students to be asked to explain the
reasoning behind an idea, or write anequation to represent a relationship.
According to teacher reports, FiW studentsreceive instruction in large and small groupsmore frequently than their U.S. counterparts,with the form of instruction varyingaccording to the grade level. In all cases,FiW math teachers are more likely than U.S.math teachers to challenge their students todemonstrate their mastery of more difficultideas or concepts.
The data also indicate that differences existin how homework is assigned and used.FiW students are more likely than U.S.students to have homework assigned everyday and to discuss their completedhomework in class.
This pattern of homework assignment anduse may mean that FiW students spend moretime outside of class reinforcing newconcepts. Classroom discussions ofhomework may clarify common difficultiesand serve to solve outstanding problems.
In the fourthgrade, FiW math students arealso more likely to spend class time startingtheir homework in class. Until the contentof the homework is analyzed, however, it isdifficult to determine whether this timeenhances instruction by allowing teachers towork more directly with their students on aregular basis.
31 43
Teacher Engagement
Teacher engagement and involvement in theinstructional process as well as in generalschool activities may also have an impact onstudent achievement. Although it can bedifficult to measure teacher engagementdirectly, a number of factors can give usinsight into the overall level of engagementand commitment.
For example, the amount of time and effortthat math teachers put into preparing andplanning for their classes, both during theregular school day and outside of regularclassroom hours, may provide a goodindication of teacher engagement.
Similarly, the amount of influence thatteachers have over basic school budgetallocation and curriculum decisions provideanother useful measure.
Teacher familiarity with key curriculum andassessment documents may offer anindication of the level of knowledge ofsignificant reform efforts.
Data from the TIMSS teacher surveyssuggest that teachers in the FiW may bemore engaged in school activities than U.S.teachers. Accordingly, this section presentsdata on four measures of teacherengagement:
teacher involvement in school-relatedactivities outside the school day;
frequency of teacher meetings;
teacher influence over key schooldecisions; and
teacher familiarity with key curriculumand assessment documents.
32
Together, thee measures provide someinsight into the relative levels of teacherengagement in the FiW and the U.S.
Teacher Involvement in School-relatedActivities Outside the School Day
Many teachers spend time outside the schoolday involved in school-related activities.The types of activities that they undertakeare varied and range from preparing for classactivities (e.g., planning lessons and gradinghomework) to helping, teaching, or workingwith individual students (e.g., tutoringstudents, meeting with parents, or consultingwith other educational personnel on theprogress of a particular student).
Teachers may also spend time outside theschool day attending to administrative orother record-keeping tasks (e.g., attendingstaff meetings, updating class grade books).Teacher involvement in any of theseactivities are used as a proxy for the level ofteacher engagement.
FiW and U.S. teachers spend similaramounts of their own time outside theformal school day working on nearly alltypes of school-related activities, with FiWfourth and eighthgrade math teachersmore likely than U.S. teachers to spend theirown time on a few key activities.
Exhibit 7 contains data on teachers' reportsof hours spent per week on activities outsidethe formal school day. Larger percentagesof FiW students than U.S. studentsat boththe fourthand eighthgrade levelshaveteachers who report spending more thanthree hours per week planning lessons bythemselves.
In the fourth grade, 67 percent of FiWstudents have teachers who spend more than
44
three hours a week preparing lessons outsidethe formal school day, compared to 46percent in the U.S., according to teacher'sreports.
In the eighth grade, 48 percent of FiWstudents have teachers who spend over threehours per week preparing for classes outsideof the formal school day, as compared to 34percent of U.S. students.
In the eighth grade, larger percentages ofFiW students than U.S. students haveteachers who devote more than three hoursper week to preparing or grading studenttests or exams outside the classroom.
Seventy-two percent of FiW students haveteachers who spend over three hours a weekof their own time preparing or gradingstudent tests or exams, compared to 47percent of U.S. students. This difference isnot found in the fourth grade.
Differences also exist between FiW and U.S.eighthgrade students with regard to howmuch time their teachers spend meeting withstudents outside the classroom. On average,eighthgrade FiW teachers spend more timemeeting with their students on their owntime than U.S. teachers (3.2 hours/week inFiW versus 2.0 hours/week in U.S.).Teachers of 72 percent of FiW eighthgradestudents report that they spend over threehours a week meeting with their students, ascompared to teachers of 29 percent of U.S.students.
To summarize, TIMSS teacher questionnairedata indicate that FiW and U.S. fourthandeighthgrade math students have teacherswho spend similar amounts of time outsidethe classroom on many school-relatedactivities.
33
Differences exist in the amount of time spenton a number of activities, however.
According to the data, FiW fourthandeighthgraders are more likely than U.S.students to have teachers who spend theirspare time preparing lessons.
In addition, eighthgrade FiW students aremore likely to have teachers who spendmore of their spare time preparing orgrading tests and meeting with studentsoutside of their class, perhaps suggestinghigher levels of teacher engagement amongFiW teachers than among U.S. teachers.
45
Exhibit 7: Teachers' Reports on Hours Spent Per Week onActivities Outside the Formal School Day
Activity Hours SpentPer Week
Percent ofFourth Grade
Students
Percent ofEighth Grade
StudentsFiW U.S. FiW U.S.
Preparing or grading student tests or examsMore than 3 hours 26 25 72 47Average (in hours) 2.0 2.2 3.4 2.7
Reading and grading other student workMore than 3 hours 76 65 48 46Average (in hours) 3.6 3.1 2.6 2.7
Planning lessons by yourselfMore than 3 hours 67 46 48 34Average (in hours) 3.2 2.5 2.8 2.4
Meeting with students outside of classroomtime
More than 3 hours 6 6 72 29Average (in hours) 1.1 0.9 3.2 2.0
Meeting with parents More than 3 hours 0 0 0
Average (in hours) 0.6 0.7 0.7 0.7Professional reading and developmentactivity
More than 3 hours 15 12 0 6Average (in hours) 1.2 1.3 0.9 0.9
Keeping students' records up to dateMore than 3 hours 19 16 34 16
Average (in hours) 1.9 1.4 2.1 1.6
Administrative tasks including staffmeetings
More than 3 hours 45 37 37 27Average (in hours) 2.5 2.2 2.4 2.0
SOURCE: NCREL analysis of LEA's Third International Mathematics and Science Study (TIMSS) data, 1994-95; FiW Teacher Questionnaire results, NCREL; table 5.5 in Mullis, I.V.S., et al. (1997). MathematicsAchievement in the Primary School Years: lEA's Third International Mathematics and Sciences Study. ChestnutHill, MA: Center for the Study of Testing, Evaluation, and Educational Policy, Boston College; table 5.6 inBeaton, A. E., et al. (1996). Mathematics Achievement in the Middle School Year: lEA's Third InternationalMathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, andEducational Policy, Boston College.
4634
Frequency of Teacher Meetings
Another indicator of teacher engagement ishow actively teachers seek feedback onways to improve instruction. Teachermeetings to plan and discuss curriculum andinstructional approaches allow teachers toget feedback from their colleagues on thebest methods to present different types oftopics, or the relative ease or difficulty thatother classes are having covering similarmaterial.
Teacher meetings not only allow teachers tolearn about more effective strategies forteaching their respective subjects, but alsopermit teachers to keep abreast of changes inmajor national, state, and local curricularstandards and assessments. Further, theymay provide important opportunities forteachers to share resources and ideas forinstructing and motivating their students.
As shown in exhibit 8, FiW students aremore likely than U.S. students to haveteachers who report meeting frequently withother teachers in their subject area to discussand plan curriculum or teaching approaches.
In the fourth grade, 26 percent of FiWstudents have teachers who meet with theircolleagues daily to discuss curriculum. Incontrast, 10 percent of U.S. students haveteachers who meet with their colleagues thisfrequently.
This gap between FiW and U.S. is evenmore pronounced for weekly teacherencounters. Eighty-one percent of FiWfourthgrade students have teachers whomeet with other teachers at least once aweek, compared to 59 percent in the U.S.
(These percentages are calculated bycombining the following categories:"almost every day" + "two or three times aweek" + "once a week").
In both FiW and U.S. schools, eighthgrademath teachers meet with their colleaguesless frequently than fourthgrade mathteachers. However, notable differences existbetween FiW and the U.S.
Approximately half (51 percent) of FiWeighthgrade students have teachers whoreport that they meet with other teachers toplan at least once a week, compared to aboutonethird of U.S. students (34 percent).(Again this is calculated by combining thecategories noted above).
At the other end of the spectrum, a notablegap also exists between the FiW and theU.S. Onethird of U.S. eighthgradestudents have teachers who meet with theircolleagues two or fewer occasions (never,once, or twice) over the course of a fullschool year. Three percent of FiW studentsfall into this category.
As with the TIMSS data on the amount ofout-of-school time that teachers devote totheir work, data on the frequency teachermeetings also suggest that FiW fourthandeighthgrade math teachers may be moreengaged than their U.S. counterparts inplanning curriculum and seeking feedbackfrom their colleagues.
Disparities between the number of studentswhose teachers participate in weeklyplanning sessions are most noteworthy, andexist at both the fourthand eighthgradelevels.
354 7
Exhibit 8: Teachers' Reports on the Frequency of Meetings withOther Teachers in Their Subject Area to Discuss and Plan
Curriculum or Teaching Approaches
Frequency
Percent of FourthGrade Students
Percent of EighthGrade Students
First in theWorld
UnitedStates
First in theWorld
UnitedStates
Almost every day 26 10 5
Two or three times a week 17 14 25 9Once a week 38 35 21 16
Once a month 7 18 29 23Every other month 7 5 15 12
Once or twice a year 5 16 1 25Never 0 2 2 8
SOURCE: NCREL analysis of TIMSS data; FiW Teacher Questionnaire results, NCREL.NOTE: Totals may not add to 100 due to rounding.
Teacher Influence Over Key SchoolDecisions
The level of teacher involvement in schoolbudget and curricular decisions can alsoprovide yet another indication of the level ofteacher engagement. Although the degree towhich states and districts allow theirteachers to be involved in these discussionsvaries considerably, actively engagedteachers may exert a lot of influence overthese decisions, while teachers who are lessengaged in improving curriculum,instruction, or the overall schoolenvironment may not.
This section looks at data on the influencethat eighthgrade math teachers in the FiWand the U.S. have over basic schooldecisions. (Data are not available for fourthgraders).
As shown in exhibit 9, eighthgrade mathstudents in the FiW Consortium were morelikely than U.S. students to have mathteachers who report they have a lot ofinfluence over key school decisions.
36
Nearly half (47 percent) of FiW eighthgrade students had teachers who reportedhaving "a lot" of influence over the subjectmatter to be taught.
Ninety-two percent of FiW eighthgrademath students had teachers who reportedthat they had at least "some" (if not "a lot")of influence over the subject matter to betaught.
By contrast, U.S. students tended to haveteachers who reported that they had lesscontrol. Thirty-eight percent of U.S.students had teachers who reported that theyhad "a lot" of control over the subjectmatter, and 73 percent had teachers who feltthey had at least "some" control. Seventy-four percent of FiW eighthgrade mathstudents had math teachers who claimedthey have "some" or "a lot" of control overwhich textbooks are used, compared to only63 percent of U.S. eighthgraders.
This difference may reflect the fact that, insome schools, these choices are made at thestate or district level.
48
FiW eighthgrade students were also morelikely than U.S. students to have mathteachers who reported that they have "a lot"of control over what supplies are purchased(47 percent of FiW teachers compared toonly 23 percent of U.S. teachers).
However, FiW eighthgrade students aremore likely than U.S. students to have mathteachers who felt they had no control overthe amount of money to be spent on supplies(46 percent in FiW compared to 35 percentin the U.S.).
Since not all schools or districts allowteachers to have a say in budgetary matters,it is not clear that these data indicate higher
levels of teacher involvement in the FiWthan in the U.S. However, they do indicatethat FiW teachers have a greater influenceon these decisions than their U.S.counterparts.
In summary, data indicate that eighthgradeFiW students are more likely to be taught byteachers who have control over somecurricular and budget decisions.
Again, these findings reinforce the datareported earlier in the section that FiWteachers may be more engaged than theirU.S. counterparts (to the extent that thesedata capture not only teacher influence, butalso teacher engagement)
Exhibit 9: Teachers' Reports on Their Influence over School Decisions
School DecisionAmount
ofInfluence
Percent of Eighth GradeStudents
FiW U.S.
Subject matter to be taught
A lot 47 38
Some 45 35
Little 6 18
None 2 9
Specific textbooks to be used
A lot 39 27
Some 35 36
Little 22 18
None 4 19
.
The amount of money to be spent on supplies
A lot 13 4
Some 16 27
Little 25 35
None 46 35
What supplies are purchased
A lot 47 23
Some 43 41
Little 10 29
None 0 7
SOURCE: NCREL analysis of TIMSS data, FiW Teacher Questionnaire results.NOTE: Totals may not add to 100 due to rounding.
37 49
Exh
ibit
10: T
each
ers'
Rep
orts
on
The
ir F
amili
arity
with
Key
Cur
ricu
lum
Doc
umen
ts
Doc
umen
tFa
mili
arity
Perc
ent o
f E
ight
hGra
de*
Stud
ents
FiW
U.S
.Ja
pan
Kor
eaS
inga
pore
Nat
iona
l Cur
ricul
um G
uide
for
Mat
hem
atic
s(U
S: N
atio
nal C
ounc
il of
. Tea
cher
s of
Mat
hem
atic
s (N
CT
M)
Pro
fess
iona
l Sta
ndar
dsfo
r T
each
ing
Mat
hem
atic
s)
Ver
y F
amili
ar69
383
475
Fai
rly F
amili
ar31
4871
4123
Not
Fam
iliar
013
1944
2
No
Suc
h D
ocum
ent
00
711
0
Reg
iona
l (S
tate
) C
urric
ulum
Gui
de fo
rM
athe
mat
ics
Ver
y F
amili
ar10
' 27
11
0
Fai
rly F
amili
ar21
3637
260
Not
Fam
iliar
6735
3648
0
No
Suc
h D
ocum
ent
22
2625
100
Nat
iona
l Exa
min
atio
n S
peci
ficat
ions
(US
: The
Nat
iona
l Ass
essm
ent f
orE
duca
tiona
l Pro
gres
s (N
AE
P)
Ass
essm
ent
Fra
mew
orks
/Spe
cific
atio
ns)
Ver
y F
amili
ar0
161
462
Fai
rly F
amili
ar16
2429
2436
Not
Fam
iliar
8258
4148
2
No
Suc
h D
ocum
ent
23
2924
0
Reg
iona
l (S
tate
) E
xam
inat
ion
Spe
cific
atio
ns
Ver
y F
amili
ar11
42
10
Fai
rly F
amili
ar33
2867
280
Not
Fam
iliar
5668
17.4
70
No
Suc
h D
ocum
ent
0.0
1424
100
SO
UR
CE
: NC
RE
L an
alys
is o
f lE
A's
Thi
rd In
tern
atio
nal M
athe
mat
ics
and
Sci
ence
Stu
dy (
TIM
SS
) da
ta, 1
994-
95; F
iW T
each
er Q
uest
ionn
aire
resu
lts, N
CR
EL.
NO
TE
: Tot
als
may
not
add
to 1
00 d
ue to
rou
ndin
g.*E
ight
h gr
ade
in m
ost c
ount
ries.
5051
Teacher Familiarity with Key Curriculumand Assessment Documents
As scientists, mathematicians, andresearchers make advances in math andscience, school textbooks and otherclassroom materials must be periodicallyupdated to reflect new knowledge and waysof thinking. Standards, curriculumguidelines, and student assessmentinstruments also undergo regular revisionsand updates so that they can accuratelyreflect current and emerging research andbest practices.
Thus, key local, state, and nationalstandards, curriculum, and assessmentdocuments typically reflect the most currentmath and science knowledge. Teacherfamiliarity with, and knowledge of, thesedocuments may provide another indicationof how engaged teachers are in keepingabreast of the latest advances in math andscience curriculum and assessment.
This section looks at FiW teacher familiaritywith major curriculum and assessmentdocuments. Comparative FiW and U.S.data, and those from other nations are onlypresented at the eighthgrade level becauseof limited data availability. It should also benoted that there are some definitional issueswith the international comparisons, thereforethese data should be interpreted withcaution.
As illustrated in exhibit 10, FiW eighthgrade students have teachers who showvarying degrees of familiarity with keycurriculum documents. Most show thegreatest familiarity with national standards.Sixty-nine percent of FiW students haveteachers who are "very familiar" and 31percent have teachers who are "fairlyfamiliar" with the National Council of
39
Teachers of Mathematics (NCTM)standards. In contrast, 38 percent of U.S.students have teachers who are "veryfamiliar" and 48 percent who are "fairlyfamiliar" with the NCTM standards.
In high math achievement countries, thepattern varies across countries. InSingapore, teachers of most students were"very familiar" with equivalent documents,while in Japan and Korea, students weremore likely to have teachers who report thatthey are "fairly familiar" but not "veryfamiliar."
The comparison to U.S. teachers' familiaritymay indicate that FiW teachers have agreater involvement with professionalassociations, place more emphasis onprofessional knowledge, or have moreopportunities to pursue outside interests.
The responses of the high performing Asianteachers are much less clear. While teachersin Singapore indicate a similar familiaritywith their national curriculum, those inJapan and Korea do not. Some believe thedata from Japan and Korea reflect anunderstated familiarity, as opposed to a lackof knowledge.
FiW teachers are less familiar with statecurriculum guides than national guidelines.Math teachers of less than half of FiWeighthgrade students are "very familiar" or"fairly familiar" with state curriculumguides. U.S. math teachers are more familiarwith these curriculum guides. Sixty-threepercent of U.S. students have teachers whoreport they are "very familiar" or "fairlyfamiliar" with these guides.
This lack of familiarity could also be afunction of when, and if, states had
52
completed development of their statestandards.'
As for familiarity with exam specificationsfor mathematics, math teachers of most FiWfourth and eighth grade students report thatthey are "not familiar" with the U.S.equivalent of national exam specificationsthe National Assessment of EducationalProgress (NAEP). This is probably becauseIllinois does not participate in the NAEP.
In the U.S., 16 percent of students haveteachers who are "very familiar" withNAEP, and 24 percent have teachers whoare "fairly familiar" with NAEP.
While FiW teachers report more familiaritywith state exam specifications than withNAEP, they do not report particularly highlevels of familiarity with state exams. Infact, teachers of at least half of the FiWeighth grade math students report that theyare "not familiar" with state examspecifications.
This pattern differs slightly for the U.S. Inthe U.S., more eighth grade math studentshave teachers who are "very familiar" or"fairly familiar" with the NAEP than withtheir state exams.
As was the case with state curriculumframeworks or standards, this could be morea function of state policy than teacherfamiliarity.
In summary, TIMSS questionnaire data onteacher familiarity with key curriculumdocuments and exam specificationsreinforce some of the earlier findings onteacher engagement. Although data are notavailable at the fourthgrade level, FiWeighthgrade teachers report high levels offamiliarity with the NCTM Standards,unlike U.S. teachers.
40
Data on state teaching guides and examspecifications show lower levels of teacherfamiliarity, in both the FiW and the U.S.High degrees of familiarity with nationalcurriculum standards are also found in somehigh math achievement countries, but notall. It is unclear, however, whether thispattern reflects actual differences infamiliarity or in understated familiarity(particularly in the case of Japan and Korea).
Summary
In summary, the TIMSS data indicate thatFiW students have teachers that indicatemore engagement than U.S. teachers in abroad array of school-related activities.
These activities include participating inschool-related activities outside the schoolday, meeting with their colleagues,identifying and selecting textbooks, buyingsupplies, and keeping up with newcurriculum and instructional developmentsand techniques.
Other research has posited that one result ofgreater teacher engagement is morestimulating, organized, and/or tailoredinstruction. While certainly not conclusive,these data may suggest that a key componentto delivering better math instruction may beencouraging more active teacherparticipation in classroom planning, schooldecision-making, and keeping abreast of keychanges in curriculum and assessments.
Teaching Environment
In addition to the key factors alreadydiscussed, the environment for teaching mayalso have an important impact oninstruction, and in turn, studentachievement. Teachers who must dealfrequently with non-academic issues withinthe classroom may not have as much time todevote to instruction.
For example, many teachers must routinelyspend class time dealing with disciplineproblems or disruptive students. Externalfactors may also contribute to, or detractfrom, teaching environments in the FiW andthe U.S.
In particular, state-, school district-, orschool-based decisions and policy prioritieswhich affect either the availability ofequipment, the adequacy of physicalfacilities, or the student/teacher ratio mayhave an impact on the teaching environment.
Exhibit 11 presents data on teachers' reportson the factors that limit their ability to teachtheir classes. These reports show similarpatterns exist between FiW and the U.S. andhigh performing countries (Japan, Korea andSingapore). In both the fourthand eighthgrades, across all countries, the three mostfrequently cited factors limiting teachers'abilities to teach by "quite a lot" or "a greatdeal" were:
(1) students with different academicabilities;
(2) high student/teacher ratios; and(3) disruptive students.
While different percentages of teachers fromthese countries reported that these factorsplaced "quite a lot" or "a great deal" oflimitations on their ability to teach, in nearly
all cases, students with different academicabilities were reported to be one of the mostimportant limiting factors.
The only exception was for eighthgrademath students from Singapore, wherestudents were more likely to be taught byteachers who report that high student/teacherratios were the most important limitationplaced on their ability to teach their class.
Across nearly all categories, FiW studentswere less likely than their counterparts in theU.S. and in highachieving countries to haveteachers who report that critical factors limittheir ability to teach their class.
At the other extreme, students in Korea werethe most likely to have teachers whoreported that the various factors placed"quite a lot" or a "great deal" of limitationson them.
Fewer students had teachers who reportedshortages of equipment as limiting factors.In the fourth grade, teachers of four percentof FiW math students reported thatequipment shortages limited their ability toteach by "quite a lot" or "a great deal,"compared to teachers of approximately 25percent of students in the U.S., Japan, andSingapore.
As they did with all factors, Korean teachersreported that equipment shortages placedmore limitations on their ability to teach,with teachers of 54 percent of studentsindicating that this limited their teachingability "quite a lot" or "a great deal."
A similar pattern was found on theequipment shortages in the eighthgradedata from the FiW, U.S. and highachievement countries. The relative wealth
41 5
of FiW districts probably has an impact onthe differences in these results.
In summary, the TIMSS teacher dataindicate that FiW, U.S., and high mathachieving countries report similar patterns inthe factors that affect their ability to teachmath. The limitation at the top of the list in
the FiW, U.S. and all highmath achievingmath countries except Singapore is dealingwith students with a range of academicabilities. Students in Singapore were morelikely to have teachers who reported highstudent/teacher ratios as their mostimportant limitation. Concerns overfacilities and supplies were less importantlimitations for all countries.
Exhibit 11: Teachers' Reports on the Factors that Limit HowThey Teach Mathematics Class
Country
Percent of students whose teachers report each factor limiting how theyteach class as "quite a lot" or "a great deal"
Students withDifferentAcademicAbilities
DisruptiveStudents
Shortage ofEquipment for
Use inDemonstrations
and OtherExercises
HighStudent/Teacher
Ratio
Fourth Grade *FiW 36 15 4 18
United States 41 r 31 r 25 r 38 rJapan 60 -- 28 41
Korea 69 64 54 62Singapore 66 42 25 60
Eighth Grade *FiW 32 21 7 25
United States 44 r 39 r 20 r 29 rJapan 63 -- 12 42Korea 77 60 31 67Singapore 55 44 25 60
SOURCE: Figure 5.4 in Mullis, I.V.S., et al. (1997). Mathematics Achievement in the Primary School Years: lEA'sThird International Mathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing,Evaluation, and Educational Policy, Boston College; figure 5.3 in Mullis, I.V.S., et al. (1998). MathematicsAchievement in Missouri and Oregon in an International Context: 1997 TIMSS Benchmarking. Chestnut Hill, MA:Center for the Study of Testing, Evaluation, and Educational Policy, Boston College; FiW Teacher Questionnaireresults.
*Fourth/Eighth grade in most countries.A double dash (--) indicates data are not available. This question was not included on questionnaires for teachers ofJapanese students.An "r" indicates teacher response data available for 70-84 percent of students.
42 55
Summary
In sum, differences in the contexts forteaching and learning between the FiW andthe U.S. may offer some insight intopossible explanations for the gap inachievement levels between these twogroups.
Accordingly, the preceding sectionsexamined four broad areas that help definethe context for teaching and learning forpossible clues as to the factors that mightdrive these differences: curriculum,instructional practices, teacher engagement,and the teaching environment.
Differences in curriculum may contribute tothe differences in achievement between theFiW and the U.S. These achievement gapsdo not seem to be driven by differences inthe number of topics covered by thetextbooks used by the FiW and U.S.students, as the numbers of topics addressedby U.S. and FiW textbooks are similar.
FiW eighthgrade textbooks, however, tendto focus on algebra and geometry moreheavily than U.S. books do, perhapsreflecting the difference in course-takingbehavior.
Nor do the differences seem to be driven bya more focused coverage of topics in theclassroom, since FiW students spend classtime on just as many topics as their U.S.peers. However, FiW students seem to beintroduced to more advanced topics earlierthan U.S. students. This pattern is found inboth the fourth and eighthgrades, and, asmight be expected, is even more pronouncedin the eighth grade.
Data on instructional practices indicate thatdifferences exist between FiW and U.S.
fourth and eighth grade math classes.According to students, the FiW and the U.S.show similar patterns with respect to thefour most frequently used activities.However, TIMSS data suggest that FiW andU.S. math teachers rely on different methodswhen demonstrating how to do mathproblems.
In the fourth grade, FiW teachers rely on avariety of approaches for teaching; no onemethod dominatesome instructional timeis spent in large groups, some in smallgroups, some working individually.
In the eighth grade, however, the mostfrequently used classroom organizationalmethodin the FiW and in high mathachievement countriesinvolves the mathteacher teaching the whole class; thisapproach is used far more frequently than inU.S. eighth grade math classes. These datasuggest that FiW students may have mathteachers who use direct teaching styles morefrequently than their U.S. counterparts, withthe form of instruction varying according tothe grade level.
In addition, both FiW fourth and eighthgrade students are more likely to be asked todo reasoning tasks than to spend timepracticing computational skills.
The data also indicate important differencesin how homework is assigned and used.FiW students are more likely than U.S.students to have homework assigned everyday and to discuss their completedhomework in class. Classroom discussionsof homework may help to clarify commondifficulties and serve to solve outstandingproblems that their students encounter.
Together, these results suggest that FiWmath teachers are more likely than U.S.
43
5 6
math teachers to challenge their students todemonstrate their mastery of more advancedideas or concepts.
The TIMSS data also suggest that FiWstudents may have teachers that are moreengaged than U.S. teachers in a broad arrayof school-related activities. These includeparticipating in school-related activitiesoutside the regular work day, meeting withtheir colleagues, identifying and selectingtextbooks, buying supplies, and keeping upwith new curriculum and instructionaldevelopments and techniques.
Finally, similar patterns were found inteacher reports on the type of factors thatlimit their teaching abilities. Across theFiW, the U.S., and high achieving mathcountries, teachers reported similar patterns:student factors most limited their ability to
44
teach, while the adequacy of class was lesslimiting.
FiW teachers reported that the adequacy ofsupplies hindered their ability to teach verylittle, no doubt reflecting the relatively highwealth of the districts.
While certainly not conclusive, these datasuggest that key components of delivering aworld-class math education may beencouraging the earlier introduction ofadvanced math topics into the curriculumand spending more time in the classroomconcentrating on instruction.
Efforts to encourage active teacherparticipation in classroom planning, schooldecision-making, and keeping abreast of keychanges in curriculum and assessments arealso likely to be beneficial.
What is the FiW Consortium Doing to Improve Math and Science?
The Consortium is launching numerousactivities as part of its effort to improvemath and science programs and to identifyworld-class standards in instruction,assessment, and curriculum development.
The Consortium also works hard todisseminate its findings to educators,researchers, and policymakers byparticipating in numerous presentations andseminars. Working closely with its partners,it has also taken full advantage of advancesin technology to disseminate materialsdocumenting the FiW's progress to othersvia the World Wide Web.(http://www.ncrel.org /fitw/homepage.htm)
Across the spectrum of activities undertakenby the FiW, educators have maintained theircommitment to including all students in theachievement of math and science.Reflecting this commitment, students withdisabilities were included in the sample ofstudents taking TIMSS, and specialeducation teachers participate in each of theConsortium's Teacher Learning Networks(TLNs).
This section focuses on one of the most fullydeveloped of these activities, the FiW'sefforts to establish TLNs, networks oflearning communities involving educators,parents, and community leaders.
This section also describes the FiW effortsin this area, as well as one TLN's efforts toimprove science instruction by examiningthe TIMSS results in light of current FiWinstructional practices.
455-8
Teacher Learning Networks:Collaborative Learning Communities
The Consortium's TLNs grew out of a cross-district planning effort that involved teachersfrom all districts and all levels of education,Consortium administrators, and outsideadvisors. After the Consortium was formed,planning teams were established toconceptualize a structure for developinglearning communities that would extendbeyond district boundaries.
These learning communities build upon theprofessional development efforts of theindividual districts in order to promotesystemic change. The planning teamsincluded teachers from each of the highschools in the Consortium, as well-asteachers from the elementary schools.
Working in collaboration with theprofessional development specialists fromNorth Central Regional EducationLaboratory, the planning teams developed astructure to engage teachers across four keydomains. These areas are:
Curriculum models. This networkexamines the Consortium's curriculumusing techniques similar to those used bythe International and National TIMSSCenters. It allows teachers to promotecross-district and cross-gradecoordination of curriculum, as well asgreater alignment with national andinternational standards. Networkmembers also explore differences andsimilarities of different districts on thesequence, emphasis, and content coveredin math and science courses.
Instructional practices and models. Thisnetwork focuses on encouraging theimplementation of instructional practicesthat promote engaged learning forstudents. Network members explorevarious instructional methods, includingproblem-based learning, hands-onscience, and activity-centered teaching.
Assessment strategies. This networkassists the Consortium in usingassessments to support decisions andestablish school improvement plans. Itplaces particular emphasis on the use ofperformance-based assessments in placeof traditional testing formats. Membersalso look at the importance of integratingteaching with assessment and methodsof alternative instruction that informinstructional practices.
Technology. This network explores thepotential for using different forms oftechnology to support and augment mathand science education.
As shown in exhibit 12, the structure of theTeacher Learning Networks relies on theorganizational, intellectual, and creative
46
resources of the FiW Consortium, itspartners, and the education community.Approximately 75 teachers are involved inthe learning network activities."
The TLNs are supported by the InstructionalSupport Network (ISN), a group ofcurricular and instructional directors whoprovide technical support to the TLNs. TheISN collects and assists in the interpretationof data and assesses the systemicfunctioning of the networks.
In addition, FiW teachers and staff can drawupon data and expertise available from theDepartment of Education and/or NCREL.They can also make use of the research onthe best practices for teaching math andscience being undertaken by theseorganizations.
Finally, the learning communitiesthemselves also function as a form ofintellectual capital to other teachers andadministrators for improving math andscience achievement.
59
A
Exh
ibit
12: F
irst
in th
e W
orld
Tea
cher
Lea
rnin
g N
etw
orks
Lea
rnin
g C
omm
uniti
es I
nitia
te a
nd S
uppo
rtE
duca
tiona
l Im
prov
emen
t and
Ref
orm
Cur
ricu
lum
Mod
els
Inst
ruct
iona
lPr
actic
e/M
odel
sT
echn
olog
yA
sses
smen
t
Org
aniz
atio
nal
Lea
ders
hip
(Cor
e C
apac
ities
for
Cha
nge)
Vis
ion
Bui
ldin
gIn
quir
yM
aste
ryC
olla
bora
tion
Proc
ess/
Act
iviti
esFo
rmul
ate
Prob
lem
s
lisG
ener
ate
Hyp
othe
ses/
Res
earc
h Q
uest
ions
Cre
ate
Met
hodo
logi
es f
or S
tudi
es
Dis
sem
inat
e Fi
ndin
gs
Prov
ide
Prof
essi
onal
Dev
elop
men
t
Inte
llect
ual
Cap
ital
Dat
aR
esea
rch
Bes
t Pra
ctic
esSt
anda
rds
The
orie
sL
earn
ing
Com
mun
ities
Dis
tric
t/Sch
ool I
mpr
ovem
ent
Eff
orts
Lea
ding
to W
orld
Cla
ss P
erfo
rman
ce a
nd S
choo
lsSO
UR
CE
: Kim
mel
man
, Pau
l., K
roez
e, D
avid
., (A
ugus
t 199
8). P
rese
nted
mat
eria
ls a
nd h
ando
uts,
Sem
inar
on
the
Firs
t in
the
Wor
ld C
onso
rtiu
m. S
pons
ered
by
The
Nat
iona
l Ins
titut
e on
Stu
dent
Ach
ieve
men
t, C
urri
culu
m, a
nd A
sses
smen
t. O
ffic
e of
Edu
catio
nal R
esea
rch
and
Impr
ovem
ent.
Sym
posi
um c
ondu
cted
at t
heU
.S. D
epar
tmen
t of
Edu
catio
n, W
ashi
ngto
n, D
C.
,061
Teacher Network Grants
Teacher network grants are small grantsprovided to Consortium staff interested inpursuing staff development activities,locally initiated action research, orcurriculum development initiatives in mathand science. The FiW leadership identifiedthe two primary objectives of the teachernetwork grants as follows:
To improve the teaching of math andscience through self-identified andConsortium-identified areas of study.The teacher network grants allowteachers to identify areas of study thathave direct links to their abilities toteach math and science.
To allow practitioners to take an activerole in network activities while bothproviding and receiving services. Thesecross-district activities allow teachersand others to learn from and interactwith their colleagues at other schoolswithin their district, as well as theircolleagues in other FiW districts.
To receive a grant, a network member ordistrict must submit a proposal to theConsortium's Grant Review Committee on atopic for study in one of the four networkareas (curriculum, instruction, assessment,or technology).
The proposal must outline the project'sgoals, planned activities, budget, and themethods by which its outcome will beshared with other districts within theConsortium. Grant applications arereviewed by a Grant Review Committee,which is made up of staff and administrativerepresentatives.
BEST COPY AVAILABLE 4
Proposals are assessed according to criteriaestablished by the FiW Board of Directors.Funding is provided for proposals that meetthese four criteria:
(1) congruence with the Consortium'sannual working plan;
(2) direct teacher or staff input;(3) the availability of activities and
outcomes for all Consortium districts;and
(4) focus on math and science.
During the 1997-98 school year,approximately 20 grants were awarded.'Grants were awarded for work on thefollowing general topics in the four networkareas:
Curriculum Models
Curriculum continuity for math andscience from kindergarten through grade12;
Analyzing physics concepts and thesystematic introduction of concepts tostudents; and
Mentoring program for females in mathand science.
Instructional Practices and Models
Implications of brain research in theteaching and learning process;
Staff development to improve instructionfor students with moderate to severelearning disabilities; and
Training staff in problem-based learning.
62
Assessment Strategies
Determining teacher knowledge aboutassessment;
Aligning instruction and assessmentpractices for students; and
Retooling science activities andassessments for students withdisabilities.
Technology
Designing web pages to further math andscience learning experiences.
As they pursue their projects, the TLNs areresponsible for organizing themselves topromote their own learning.
In particular, the networks are intended tofoster exploration of the extent to whichteachers' classroom practices are consistentwith both
(1) their stated beliefs about teaching andlearning, and
(2) current and emerging research and bestpractice.
Furthermore, the networks are exploringways to enhance consistency betweencurrent classroom practice and research onbest practices.
In addition, network members are expectedto lay the groundwork for the expansion ofthese networks into larger, more inclusivelearning communities.
Example of a FiW Teacher LearningNetwork: Analysis of FiW PhysicsAchievement
To provide an illustration of the types ofactivities undertaken by a TLN, this sectionpresents an example of work beingundertaken a group of teachers to use theTIMSS results to improve the Consortium'sscience curriculum and instruction.
With assistance from researchers at NCREL,high school physics teachers decided to takea close look at the performance of FiWstudents on physics-related questions onTIMSS, FiW physics-related curriculum,and instructional practices and beliefs ofFiW physics teachers.
Although much of this report focuses ondifferences in math achievement, thisexample was chosen because it illustrateshow FiW educators are using the TIMSSbenchmark, along with supplemental data, toimprove science achievement.
This section describes the project's goalsand participants, its initial results, and thepotential areas initially identified for futureresearch by the project's participants.
As noted earlier, the FiW results on thetwelfth grade physics exam were lower thananticipated, with twelve nations scoringsignificantly above the FiW, three nationsobtaining scores not significantly differentfrom the FiW, and one nation scoringsignificantly below the FiW.
To gain fuller insight into what it takes tobecome first in the world in physicsachievement, six current and former physicsteachers32 met over the summer of 1998. Atthese meetings, they reviewed the FiWphysics results from TIMSS, as well as their
49 6 3
instructional and curricular practices anddiscussed what they could do to improvephysics achievement.
This cross-district effort built on theteachers' considerable experience andcommitment to improve science instruction.Between them, the six teachers have morethan 100 years of combined physics teachingexperience, and all are extremely active inprofessional development activities andorganizations.
Project Activities
After initial meetings with NCREL staff toreview the TIMSS physics questions anddiscuss the high school results, teammembers identified three project activities.These activities were designed not only togive the network members a betterunderstanding of the FiW students' relativestrengths and weaknesses in solving physicsproblems, but also to give them a betterunderstanding of the context for learningand teaching physics in FiW and the varietyof instructional approaches used across theConsortium.
The project activities are:
Analyzing FiW students' performance onphysics exams using groups of similarTIMSS test questions. This analysis waslimited to the use of released TIMSS testquestions, a relatively small sample ofquestions. To conduct this analysis, thereleased physics questions from theeighth grade science exam and thetwelfth grade physics exam weregrouped according to conceptual modelsin physics (e.g., particles, matter, light,ideal gas, systems, relativistic physics,and force laws). FiW student responsepatterns were then compared to the
response patterns of U.S. students andthe international average.
Extending the TIMSS teacher survey to asample of FiW high school scienceteachers. The TLN administered one ofthe two TIMSS teacher surveys that hadbeen prepared for TIMSS but not usedduring the study. (Unlike in the fourthand eighth grades, TIMSS did not collectdata from twelfthgrade teachers). TheTLN collected data on teacher beliefsand attitudes using one of these surveys.
Creating and administering a pilotteacher survey to collect data on values,style, and "rigor" in physics classrooms.This task involved creating andadministering the Teacher Survey ofRigor, a new survey to a sample ofphysics teachers. The new survey wasintended to collect additional data onhow physics courses are similar ordifferent in style of delivery,expectations for students and whatteachers value as important. Theanticipated survey results also areexpected to complement the results fromthe TIMSS teacher survey.
As these activities demonstrate, the FiWteacher learning networks provide a uniqueopportunity for their members to work with,and learn from, their peers.
Furthermore, they allow FiW teachers to tapinto NCREL analysts and researchers, aswell as their colleagues in other districts, intheir efforts to become first in the world inmath and science.
Initial Results
At the time of this writing, the analysesundertaken by this network team are at
64
different stages of completion. Initial resultsare available from the TIMSS questionnaireanalysis, while the two additional surveysare in the very early stages of analysis.
Nevertheless, the preliminary results havealready pointed to areas where the FiWmight work toward improving its physicsinstruction.
The initial results from the analysis ofTIMSS physics questions done by NCRELpoint to areas of relative strength andweakness in FiW physics achievement. Onearea of interest was in the different topicalareas and different achievement levels.
For example, FiW results for eighthgradescience show strength in the physicalsciences with few exceptions. One of theseexceptions was questions associated with theatomic model, on which FiW studentsshowed their lowest performance.
Twelfth grade FiW students demonstratedstrength on questions related to theNewtonian concept of force. Questions inmodern physics and mechanical wavesshowed the lowest performance levels.
One unexpected result highlighted by theanalysis was that FiW students scored betteron questions dealing with constantgravitational force than constant electric ormagnetic force, despite the fact that thesame general concepts apply to both areas.
The analysis of TIMSS results on the twelfthgrade physics exam also gave the FiWteachers a fuller understanding of the typesof problem-solving skills needed to achieveworld class standards in physics. Forexample, the teachers discovered that fewquestionnaire items could be answered usingrote memory.
In addition, they discovered that all of theTIMSS countries obtained relatively lowperformance levels on the physicsassessment. On average, only 31 percent ofthe items were answered correctly.
Also, FiW physics students performed betterwhen tackling certain types of test questions.In particular, they had higher performanceon multiple choice items (as opposed to freeresponse items) than the internationalsample.
Topics for Further Research Identified bythe Physics Teachers' Learning Network
Based on their preliminary research, theteam has already identified several questionsthat they feel may deserve future attention:
(1) Are similar trends repeated in theTIMSS questions that were not released?
(2) Why do all students (International, U.S.,and FiW) perform poorly on the TIMSSphysics assessment?
(3) Can conceptual models be tracedthrough the fourth, eighth, and twelfthgrades?
(4) How do students perform onexperimental design and scientificprocess items?
To address these questions, the TLN hasidentified some opportunities for expandingtheir investigation that may be particularlyfruitful. In particular, they recommendbroadening the analysis to include all FiWschools, as well as an examination of studentresults on the fourthgrade TIMSS scienceassessment.
They also hope to work on identifyinggroups that might yield richer comparisonson what it means to be first in the world(e.g., identifying an appropriate comparison
51 6 5
group of FiW physics students or group ofnations that might be present for all threetest populations).
Finally, based on the results of the surveyanalysis, they plan to look for appropriatephysics content that might strengthen theircurriculum in areas where FiW studentachievement was not as high.
Summary
The Consortium has begun to embark on ahost of activities to define and clarify worldclass standards and establish learningcommunities. The establishment of TLNs;networks of learning communities involvingeducators, parents, and community leaders,represents one of the most fully developedof these activities.
As a result of a cross-district planning effortinvolving teachers from all education levels,Consortium administrators, and outsideadvisors, the networks engage participants infour key domains:
(1) curriculum models,(2) instructional practices and models,(3) assessment strategies, and(4) technology.
52
These learning communities build upon andcontribute to the professional developmentefforts of individual districts and leveragethe organizational, intellectual, and creativeresources of the FiW Consortium, itspartners, and the education community.
To facilitate the work of the TLNs, smallgrants are available to Consortium facultyinterested in pursuing staff developmentactivities, locally initiated action research, orcurriculum development initiatives in mathand science.
During the 1997-98 school year,approximately 20 grants were awarded.Throughout the upcoming year, thenetworks and grant recipients will exploreways to enhance consistency among currentclassroom practice, current and emergingresearch, and best practice, as illustrated bythe ambitious agenda of activitiesundertaken by a team of physics teachers.
Working with the Consortium's partners,these teachers are successfully using theTIMSS benchmark, along with supplementaldata, to examine physics curriculum andinstruction across districts and grade levels.
66
Summary and Conclusion
In conclusion, the effort begun by a group ofsmall school districts north of Chicago hasalready begun to show some promisingresults. Motivated to take the NationalEducation Goals seriously, this consortiumembarked upon a detailed plan of action to"become first in the world in math andscience by the year 2000."
As'a first step in their plan, Consortiumstudents became the only school districts totake part in TIMSS, the most ambitious,comprehensive, and rigorous internationalassessment of math and science yetundertaken.
In contrast to the U.S., the FiW performedexceptionally well on the Consortium'sinitial benchmark, indicating that they arewell on their way to achieving their goal. Infact, TIMSS results indicate that fourth andeighth grade students performed at, or near,the top of the world in both math andscience.
In the twelfth grade, results were moremixed. Although students taking the generalknowledge assessments achieved worldclass standards, FiW students taking theadvanced math and physics examsperformed near the international average.However, FiW AP students taking theadvanced math and physics exams, perhapsa better group to use for internationalcomparisons, performed at the top of theworld.
Given the Consortium's performance, thisreport explored some of the possible reasonswhy they did so well compared to the U.S.by focusing on math. Initial analyses of therelationship between FiW and U.S. mathachievement and student and family socio-
economic background characteristics foundthat home and family characteristics couldexplain less than half of the difference inscores.
Accordingly, differences in the contexts forteaching and learning math between the FiWand the U.S. were examined as possibleexplanations of the remaining gaps betweenthese two groups.
Four broad areas were explored:
(1) curriculum,(2) instructional practices,(3) teacher engagement, and the(4) teaching environment.
Although similar patterns were reported inall of these areas, important differences didemerge.
The review of curriculum and textbook datafound that the number of topics addressed byU.S. and FiW math textbooks is similar andFiW students spend class time on just asmany topics as their U.S. peers.
Nevertheless, some differences do existbetween FiW and U.S eighth grade mathtextbooks, with FiW eighthgrade textbooksmore focused on algebra and geometry thanU.S. books.
In addition, FiW students seem to beintroduced to more advanced topics earlierthan U.S. students. This pattern is found inboth the fourthand eighthgrades, and, asmight be expected, is even more pronouncedin the eighth grade.
53 67
TIMSS data on instructional practices alsosuggest additional differences between FiWand U.S. fourth and eighth grade mathclasses. Although students report thatsimilar patterns in the four most frequentlyused activities, the data indicate that FiWand U.S. math teachers rely on differentmethods when demonstrating how to domath problems.
In the fourth grade, FiW teachers rely on avariety of approaches for teaching; no onemethod dominates. In the eighth grade, bycontrast, group instruction of the whole classis reported as the most frequent classroomorganizational approach in the FiW and highmathachievement countries.
This approach is used far more frequentlythan in U.S. eighthgrade math classes,suggesting that FiW students may have mathteachers who use direct teaching styles morefrequently than their U.S. counterparts, withthe form of instruction varying according tothe grade level.
The TIMSS data also suggested differencesbetween the types of math activitiesperformed by FiW and U.S. in class. Inparticular, FiW fourthand eighthgrademath students are more likely than U.S.students to be asked to perform reasoningtasks than to spend time practicingcomputational skills.
In addition, important differences exist inhow homework is assigned and used. FiWstudents are more likely than U.S. studentsto have daily homework and to discuss thesecompleted assignments in class. Together,these results suggest that FiW math studentsmay be more challenged than U.S. studentsto show their mastery of more advancedideas or concepts.
54
As for teacher engagement, FiW studentsmay have teachers that are more engagedthan U.S. teachers in a wide assortment ofschool-related activities. Examples includeparticipating in school-related activitiesoutside the regular work day, meeting withtheir colleagues, identifying and selectingtextbooks, buying supplies, and keeping upwith new curriculum and instructionaldevelopments and techniques.
Similar patterns were also found in teacherreports on teaching environments. Acrossthe FiW, the U.S., and high achieving mathcountries, teachers reported similar patterns:student factors most hindered their ability toteach, while the adequacy of class supplieswas less of a limitation. FiW teachersreported that the adequacy of supplieslimited their ability to teach very little, nodoubt reflecting the relatively high wealth ofthe districts.
These data suggest that key components ofdelivering a top notch math education maybe introducing advanced math topics into thecurriculum earlier and spending more timein the classroom concentrating oninstruction.
Efforts to foster active teacher participationin classroom planning, school decision-making, and to allow teachers opportunitiesto learn about key changes in curriculum andassessments are also likely to be positive.
The FiW Consortium knows that theseinternational achievement benchmarks arenot static. While FiW students have donewell in 1996, this success does not guaranteecontinued success since the achievementbenchmark may be set at a different point in1999 and in coming years.
Accordingly, the FiW Consortium has also
68
begun its efforts to define and clarify world-class standards in instruction, assessment,and curriculum. Working with its partnersat the regional and national level, theConsortium is identifying current andemerging research and best practices in allof these areas.
Recognizing that the current context forteaching and learning within the FiWconsortium may also provide some clues asto what it takes to do well in math andscience, the Consortium is also exploring theTIMSS data for suggestions as to whichinstructional, curricular, and assessmentpractices may work well in the U.S.
Finally, the Consortium has worked hard tocreate a structure for developing a cross-district community of learners that would
55
involve educators, parents, and communityleaders.
It has established teacher learning networksin four areas: curriculum, assessment,instruction, and technology; and awardedgrants to groups of teachers pursuingprojects in these areas. These projects,along with the other efforts of learningnetworks, will lay the groundwork for theexpansion of these networks into larger,more inclusive learning communities.
Taken together, these results and activitiesprovide exciting news. They illustrate notonly that U.S. students have the potential tobecome the first in the world in math andscience, but also that districts can work in acollaborative, cooperative manner to strivetowards this goal.
69
Bibliography
Beaton, A.E., Mullis, I.V.S., Martin, M.O., Gonzales, E.J., Kelly, D.L., and Smith, T.A. (1996).Mathematics Achievement in the Middle School Years: IEA's Third International Mathematicsand Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, andEducational Policy, Boston College.
Beaton, A.E., Mullis, I.V.S., Martin, M.O., Gonzales, E.J., Kelly, D.L., and Smith, T.A. (1996).Science Achievement in the Middle School Years: IEA's Third International Mathematics andScience Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and EducationalPolicy, Boston College.
First in the World Consortium. Achieving Excellence: Initial Findings of Fourth GradePerformance from The Third International Mathematics and Science Study. (Grade 4).
First in the World Consortium. Achieving Excellence: Initial Findings of Twelfth Grade Studentsin The Third International Mathematics and Science Study.
Kroeze, David J., Johnson, Daniel P., and Zalewski, Eugene. Achieving Excellence: A Report ofInitial Findings of Eight Grade Performance form the Third International Mathematics andScience Study. First in the World Consortium.
Martin, Michael 0., Mullis, I.V.S., Beaton, A.E., Gonzales, E.J., Smith, T.A., and Kelly, D.L.,(1997). Science Achievement in the Primary School Years: IEA's Third InternationalMathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation,and Educational Policy, Boston College.
Mullis, I.V.S., Martin, M.O., Beaton, A.E., Gonzales, E.J., Kelly, D.L., and Smith, T.A. (1998).Mathematics Achievement in the Primary School Years: IEA's Third International Mathematicsand Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, andEducational Policy, Boston College.
Mullis, I.V.S., Martin, M.O., Beaton, A.E., Gonzales, E.J., Kelly, D.L., and Smith, T.A. (1997).Mathematics and Science Achievement in the Final Year of Secondary School: lEA's ThirdInternational Mathematics and Science Study. Chestnut Hill, MA: Center for the Study ofTesting, Evaluation, and Educational Policy, Boston College.
Mullis, I.V.S., Martin, M.O., Beaton, A.E., Gonzales, E.J., Kelly, D.L., and Smith, T.A. (1998).Mathematics Achievement in Missouri and Oregon in an International Context: 1997 TIMSSBenchmarking. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and EducationalPolicy, Boston College.
National Center for Education Statistics. (1997). Pursuing Excellence: A Study of U.S. Fourth-Grade Mathematics and Science Achievement in International Context. NCES 97-255, Office of
57 /0
Educational Research and Improvement. U.S. Department of Education. Washington, DC: U.S.Government Printing Office.
National Center for Education Statistics. (1997). Pursuing Excellence: A Study of U.S. Eighth-Grade Mathematics and Science Teaching, Learning, Curriculum, and Achievement inInternational Context. NCES 97-198, Office of Educational Research and Improvement. U.S.Department of Education. Washington, DC: U.S. Government Printing Office.
National Center for Education Statistics. (1998). Pursuing Excellence: A Study of U.S. Twelfth-Grade Mathematics and Science Achievement in International Context. NCES 98-049, Office ofEducational Research and Improvement. U.S. Department of Education. Washington, DC: U.S.Government Printing Office.
National Research Council. (1996). Mathematics and Science Education Around the World.What Can We Learn? From the Survey of Mathematics and Science Opportunities (SMSO) andthe Third International Mathematics and Science Study (TIMSS). Center for Science,Mathematics, and Engineering Education National Research Council. Washington, DC: NationalAcademy Presi
National Research Council. Third International Mathematics and Science Study (TIMSS) andSurvey of Mathematics and Science Opportunities (SMSO): First in the World. (June 1998)Meeting at Michigan State University.
North Central Regional Educational Laboratory, Materials from the First in the World Web Site:www.ncrel.org.
Robitaille, David F., Schmidt, William H., Raizen, Senta A., McKnight, Curtis, Britton, Edward,Niol, Cynthia. (1993). The Third International Mathematics and Science Study: TIMSSMonograph No. .1 Curriculum Frameworks of Mathematics and Science. Pacific EducationalPress. Vancouver, Canada.
Schmidt, William H., McKnight, Curtis C., Raizen, Senta A. (1996). Splintered Vision: AnInvestigation of U.S. Science and Mathematics Education: (Executive Summary). Boston, MA:Kluwer Academic Publishers.
van der Ploeg, Arie., Hager, Maureen., Kimmelman, Paul., Kroeze, David., Lamaster, Pat., &McNelly, Maggie., (August 1998). Presented materials and handouts, Seminar on the First in theWorld Consortium. Sponsored by The National Institute on Students Achievement, Curriculum,and Assessment. Office of Educational Research and Improvement. Symposium conducted atthe U.S. Department of Education, Washington, DC.
71
58
Endnotes
' The Third International Math and Science Study is the largest, most comprehensive, and mostrigorous international comparison of math and science achievement ever undertaken.
2 Consortium membership has changed over the past several years. This section presents data onthe districts that currently make up the Consortium's membership. Exhibit A-1 in appendix Alists the districts that currently make up the Consortium.
3 Average includes all districts except the Northern Suburban Special Education District.Average expenditures in the Northern Suburban Special Education District range from $8,000 to$25,000 per student, depending on the type of student disability.
Current expenditure per pupil in fall enrollment public and elementary schools. NationalCenter for Education Statistics, Digest of Education Statistics, table 169.
'First in the World Consortium. Low-income students include those who receive public aid, livein institutions for neglected or abandoned children, are supported in foster homes, or are eligibleto receive free or reduced price lunch. This figure does not include data on students from theNorth Suburban Special Education District.
6 See exhibit A-2 in appendix A for more detail.
'National Center for Education Statistics, Digest of Education Statistics, table 69. Data are from1996.
8 North Central Regional Educational Lab, FiW Web Site, Purpose and History. These figuresdo not include data on students or faculty from the Illinois Mathematics and Science Academyand the North Suburban Special Education District.
9 See exhibit A-2 in appendix A for additional data on average teacher salaries in FiW districts.
I° National Center for Education Statistics, Digest of Education Statistics, table 77. Data arefrom 1995-96.
" The topic areas discussed here and presented in exhibit B-3 are those which were determinedby Michigan State University to more closely relate to those included in the TIMSS textbookanalysis and teacher questionnaires. They are different than the topic areas used by theInternational TIMSS Center at Boston College and reported in the international comparisons.
12 The topic areas presented in exhibit B-6 are those which were determined by Michigan StateUniversity to more closely relate to those included in the TIMSS textbook analysis and teacherquestionnaires. They are different than the topic areas used by the International TIMSS Center at
7259
Boston College and reported in the international comparisons.
'3 First in the World Consortium. Achieving Excellence: Initial Findings from Twelfth-GradeStudents in the Third International Mathematics and Science Study. p. 3. Mullis, I.V.S., Martin,M.O., Beaton, A.E., Gonzales, E.J., Kelly, D.L., and Smith, T.A. (1997). Mathematics andScience Achievement in the Final Year of Secondary School: LEA's Third InternationalMathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation,and Educational Policy, Boston College. p. 19.
14 First in the World Consortium. Achieving Excellence: Initial Findings from Twelfth-GradeStudents in the Third International Mathematics and Science Study. p. 3. Mullis, I.V.S., Martin,M.O., Beaton, A.E., Gonzales, E.J., Kelly, D.L., and Smith, T.A. (1997). Mathematics andScience Achievement in the Final Year of Secondary School: LEA's Third InternationalMathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation,and Educational Policy, Boston College. p. 19.
15 First in the World Consortium. Achieving Excellence: Initial Findings from Twelfth-GradeStudents in the Third International Mathematics and Science Study. p. 4.
16 First in the World Consortium. Achieving Excellence: Initial Findings from Twelfth-GradeStudents in the Third International Mathematics and Science Study. p. 3.
"Internal NCREL memo from Bill Quinn to Jeri Nowakowski, July 17, 1998.
Is Preliminary analyses undertaken by NCES have found similar effects.
19 TIMSS administered teacher surveys to teachers of fourth- and eighth-grade students but didnot administer teacher surveys to teachers of students in twelfth grade.
20 The authors of this report looked at TIMSS data items that might point to areas that couldexplain achievement differences based on knowledge of FiW districts and educational researchfindings. No claims are being made as to the statistical significance of these findings.
21 Note that the instructional topics shown in exhibit 1 differ slightly from those presented inexhibit B-3. The categories used in exhibit B-3 are those used by the U.S. TIMSS Center atMichigan State University, while those presented in exhibit 1 correspond to the categories usedin the TIMSS teacher survey.
22 Note that the instructional topics shown in exhibit 2 differ slightly from those presented inexhibit B-6. The categories used in exhibit B-6 are those used by the U.S. TIMSS Center, whilethose presented in exhibit 2 correspond to the categories used in the TIMSS teacher survey.
23 The group of U.S. textbooks used for this analysis is described in Schmidt, William H.,McKnight, Curtis C., Raizen, Senta A. (1996). Splintered Vision: An Investigation of U.S.
60
Science and Mathematics Education: (Executive Summary). Boston, MA: Kluwer AcademicPublishers.
24 United States National Research Center, Third International Mathematics and Science Study.First in the World: Curriculum Analysis Final Report. East Lansing: Michigan State University.1998.
'United States National Research Center, Third International Mathematics and Science Study.First in the World: Curriculum Analysis Final Report. East Lansing: Michigan State University.1998, p. 6.
26 United States National Research Center, Third International Mathematics and Science Study.First in the World: Curriculum Analysis Final Report. East Lansing: Michigan State University.1998, p. 6.
"United States National Research Center, Third International Mathematics and Science Study.First in the World: Curriculum Analysis Final Report. East Lansing: Michigan State University.1998, p. 6.
28 United States National Research Center, Third International Mathematics and Science Study.First in the World: Curriculum Analysis Final Report. East Lansing: Michigan State University.1998, p. 9.
29 The U.S. TIMSS questionnaires were completed in spring 1995, at a time when many stateswere still developing their mathematics standards.
" Kroeze, David, and Daniel Johnson. Achieving Excellence: A report of initial findings ofeighth grade performance from the Third International Math and Science Study, p. 2.
31 First in the World. First in the World Consortium Science and Mathematics Grant Program1998-98. Material from FiW Web Site, www.ncreLorg/fitw.
32 The teachers had taught or were currently teaching at Glenbrook North High School,Glenbrook South High School, New Trier High School and Illinois Mathematics and ScienceAcademy.
7 461
Appendix A:Contact Information
and Characteristics ofFirst in the World Consortium Districts
75 A-1
Exhibit A-1: List of First in the World Districts
K-8 School Districts
Avoca School District No. 37Dr. John W. Sloan, Superintendent2921 Illinois Rd.Wilmette, IL 60091847-251-3587Web: www.avoca.k12.il.us
Glenview Community Consolidated No. 34Dr. Thomas Rich, Interim Superintendent1401 Greenwood Ave.Glenview, IL 60025847-998-5000Web: www.ncook.k12.il.us
Lincolnwood School District No. 74Dr. Steve Lake, Superintendent6950 E. Prairie Rd.Lincolnwood, IL 60645847-675-8234
Niles Elementary School District No. 71Dr. Eugene Zalewski, Superintendent6935 W. Touhy AvenueNiles, IL 60714847-647-9752
Northbrook School District No. 28Dr. James Kucienski, Superintendent1475 Maple Ave.Northbrook, IL 60062847-498-7900Web: www.district28.k12.il.us
Sunset Ridge School District No. 29Dr. Howard Bultinck, Superintendent525 Sunset Ridge RoadNorthfield, IL 60093847-446-6383
Wheeling School District No. 21Dr. Lloyd "Bud" DesCarpentrie,Superintendent999 W. Dundee Rd.Wheeling, IL 60090847-537-8270
Frankfort Community Consolidated SchoolDistrict 157-CDr. Pamela Witt, Superintendent10482 West Nebraska St.Frankfort, IL 60423-2235815-469-5922
Golf School District 67Dr. Linda Marks, Superintendent9401 Waukegan Rd.Morton Grove, IL 60053-1353847-966-8200
Mount Prospect District No. 57Dr. Maureen L. Hager, Superintendent701 W. Gregory StreetMt. Prospect, IL 60056-2220847-394-7300Web: www.ncesc.org/dist57
Northbrook School District No. 27Dr. David J. Kroeze, Superintendent1250 Sanders Rd.Northbrook, IL 60062847-498-2610Web: www.northbrook27.k12.il.us
Northbrook/Glenview School District No. 30Dr. Harry Rossi, Superintendent2374 Shermer RoadNorthbrook, IL 60062847-498-4190
West Northfield School District No. 31Dr. Paul L. Kimmelman, Superintendent3131 Techny Rd.Northbrook, IL 60062847-272-6880, ext. 223
A-2
High School Districts
Glenbrook School District No. 225Dr. David Hales, Superintendent1835 Landwehr Rd.Glenview, IL 60025847-998-6100Web: www.glenbrook.k12.il.us
Niles Township School District No. 219Dr. Grif Powell, Superintendent7700 Gross Point Rd.Skokie, IL 60077847-568-3590Web: www.niles-hs.k12.il.us
Special Education District
Northern Suburban Special Education DistrictSEJA-804Mr. David Peterson, Superintendent760 Red Oak Ln.Highland Park, IL 60035847-831-5100
Residential School
Illinois Mathematics & Science AcademyDr. Stephanie Pace Marshall, President1500 W. Sullivan Rd.Aurora, IL 60506630-907-5037Web: mocha.imsa.edu
New Trier School District No. 203Dr. Henry Bangser, Superintendent385 Winnetka Rd.Winnetka, IL 60093847-501-6310Web: nths.newtrier.k12.il.us
SOURCE: http://www.ncreLorefitw/ 1 stpagesfinembers.htm
NOTE: This list is the First in the World Consortium membership as of November 1998.
77A-3
Exh
ibit
A-2
: Fir
st in
The
Wor
ld D
istr
ict C
hara
cter
istic
s
Scho
ol D
istr
icts
# of
Schs
Enr
ollm
ent
Perc
enta
ge (
%)
# of
.
Tch
rs
Tea
cher
Sala
ries
Per
Pupi
l
Exp
endi
ture
Eth
nici
tyL
ow
Inco
me
LE
P
Tea
cher
s
Wh
Afr
Am
His
pA
/PI
B.A
.M
.S.+
K-8
Avo
ca S
choo
l Dis
tric
t #37
265
380
.80.
61.
017
.61.
93.
849
.051
.0. 4
9$
47,0
57$
9,49
1
Fran
kfor
t Sch
ool D
istr
ict #
I57
C3
1,39
396
.00.
81.
02.
10.
10.
080
.219
.868
$35
,575
$4,
492
Gle
nvie
w S
choo
l Dis
tric
t # 3
47
3,59
277
.12.
88.
011
.712
.07.
926
.473
.623
6$
46,8
91$
7,31
3G
olf
Scho
ol D
istr
ict #
67
253
470
.00.
44.
824
.79.
719
.159
.640
.443
$44
,135
$9,
483
Lin
coln
woo
d Sc
hool
Dis
tric
t #74
31,
266
99.1
0.0
0.0
0.9
3.7
0.0
30.5
69.5
106
$53
,188
$8,
769
Mou
nt P
rosp
ect S
choo
l Dis
tric
t # 5
73
1,77
091
.71.
42.
74.
11.
53.
647
.952
.110
1$
46,4
39$
7,89
6N
iles
Ele
men
tary
Sch
ool D
istr
ict #
712
533
70.0
0.9
2.3
26.4
19.1
8.4
29.7
70.3
39$
53,5
86$
9,09
8
Nor
thbr
ook
Scho
ol D
istr
ict #
274
1,44
886
.10.
41.
412
.10.
72.
734
.465
.612
648
,569
$9,
191
Nor
thbr
ook
Scho
ol D
istr
ict #
284
1,70
691
.30.
10.
77.
90.
65.
359
.140
.914
5$
47,2
58$
10,3
34
Nor
thbr
ook/
Gle
nvie
w S
choo
l Dis
tric
t #30
31,
195
81.0
0.1
2.1
16.8
0.6
2.5
28.8
71.2
85$
51,0
25$
8,22
9
Suns
et. R
idge
Sch
ool D
istr
ict #
292
512
91.0
0.0
0.6
8.4
0.4
2.7
41.8
58.2
41$
43,9
42$
9,70
6W
est N
orth
fiel
d Sc
hool
Dis
tric
t # 3
12
939
66.5
1.0
4.2
28.4
3.0
7.4
41.4
58.6
68$
51,1
28S.
9,10
0W
heel
ing
Scho
ol D
istr
ict #
2112
7,15
268
.52.
721
.96.
917
.814
.444
.355
.745
3$
46,6
97$
6,79
2
Sum
4922
,693
1,56
0
Ave
rage
3.77
1,74
679
.41.
69.
19.
88.
87.
942
.257
.812
0$
47,3
39S
7,78
1
Hig
h Sc
hool
Gle
nbro
ok H
igh
Scho
ol D
istr
ict #
225
24,
091
77.0
1.0
3.3
18.6
4.4
3.7
18.0
82.0
284
$65
,000
$12
,100
New
Tri
er H
igh
Scho
ol D
istr
ict #
203
13,
006
85.6
1.0
2.3
11.0
1.0
2.5
27.7
72.3
249
$66
,762
$13
,786
Nile
s T
owns
hip
Hig
h Sc
hool
Dis
tric
t #21
92
4,23
060
.31.
95.
831
.97.
44.
925
.474
.627
0$
64,1
58$
11,6
41
Sum
511
,327
803
Ave
rage
1.67
2,98
973
.01.
34.
021
.54.
63.
823
.576
.526
8$
65,2
63$
12,3
76
Spec
ial E
duca
tion
Nor
ther
n Su
burb
an S
peci
al E
d D
istr
ict
I1,
602
84.4
5.8
3.4
6.8
----
44.0
56.0
--$
42,3
11$
8,00
0-25
,000
Res
iden
tial S
choo
lIl
linoi
s M
athe
mat
ics/
Scie
nce
Aca
dem
y1
628
52.0
9.0
6.0
28.0
8.0
0.0
2.0
98.0
--$
55,1
12$
12,7
00
Stat
e
Stat
e--
--63
.320
.612
.83.
135
.76.
154
.245
.6--
$42
,429
$6,
158
Firs
t in
the
Wor
ld56
36,2
5077
.21.
87.
213
.7*
7.1
*6.
1--
----
--$
*8,
958
SO
UR
CE
: Firs
t in
the
Wor
ld C
onso
rtiu
m.
A d
oubl
e da
sh (
--)
indi
cate
s no
dat
a.
Fig
ure
was
cal
cula
ted
from
K-8
and
Hig
h S
choo
l num
bers
onl
y. S
peci
al E
duca
tion
figur
es w
ere
not i
nclu
ded.
NO
TE
S: I
. Thi
s ta
ble
pres
ents
dat
a on
dis
tric
ts in
the
Firs
t in
the
Wor
ld C
onso
rtiu
m a
s of
win
ter
1999
.2.
Dis
tric
t dat
a fr
om 1
996-
97 s
choo
l sta
te r
epor
t car
d.
3. A
ttend
ance
was
of S
epte
mbe
r 19
96.
4. L
ow in
com
e is
def
ined
as
fam
ilies
elig
ible
to r
ecei
ve fr
ee o
r re
duce
d pr
ice
lunc
h.5.
Ope
ratin
g ex
pend
iture
per
pup
il is
bas
ed o
n 19
95-9
6 da
ta.
BE
ST
CO
PY
AV
AIL
AB
LE7
Appendix B:FiW TIMSS Achievement Results
80
Appendix B: FiW TIMSS Achievement Results
Exhibits B-1 through B-12 present data on the average achievement of each country thatparticipated in the different TIMSS assessments, as well as visual representations of thedistribution of each country's scores.1 The distributions show student achievement at the5th, 25th, 75th and --th percentiles. The dark black band in the middle of each country'sdistribution is the mean plus or minus two standard errors; this band is intended toemphasize the point that each country's average score is only an estimate of the truescore.
The exhibits in this section should be interpreted with caution. Because these data weregenerated using statistical sampling procedures, the average scores are represented withtheir appropriate error bands. Therefore, average scores for countries overlap one anotherin many cases. Thus, it is incorrect to state that the FiW ranked x out of y countries.Rather, countries have been grouped according to whether their scores are significantlyabove, not significantly different from, or significantly below the scores for the FiW.Nevertheless, as the data in these exhibits illustrate, there is considerable variation inscores both within and across countries.
Since the TIMSS assessment was administered, two school districts have not continued toparticipate in FiW activities (Wilmette School District #39 and Glencoe School District#35). The results reported here include students from these districts.
' Note that different groups of countries participated in each of the different assessments. See figure B-1 ofMullis, I.V.S., et al., Mathematics and Science Achievement in the Final Year of Secondary School for asummary of countries that participated in different assessments.
81.B -1
Exhibit B-1: Distributions of Mathematics Achievement in the Fourth Grade
Country Meant Mathematics Achievement Scale Score
Higher than FiW 1
SingaporeSame FiW
625 (5.3)
l
I v...-zzi.,.:,,,,,,,./ 1
asKoreaJapanFirst in the WorldHong Kong(Netherlands)
Lower than FiW
611
597591
587577
(2.1)(2.1)(9.1)(4.3)(3.4)
I i ll:r./41111V,W2A I
..w...-m..4 i1
I v---....,,,,, I
v- ..e.e. 'WA I
I ve zma..- All I
Czech Republic(Austria)(Slovenia)Ireland(Hungary)(Australia)United StatesCanada(Israel)(Latvia {LSS})Scotland^England* A
CyprusNorwayNew ZealandGreece(Thailand)PortugalIcelandIran, Islamic Republic(Kuwait)
567559552550548546545
532531
525520513
502
502499492490475474429400
(3.3)(3.1)(3.2)(3.4)(3.7)(3.1)(3.0)(3.3)(3.5)(4.8)(3.9)(3.2)(3.1)(3.0)(4.3)(4.4)(4.7)(3.5)(2.7)(4.0)(2.8)
t v.., ,an..-...-, ZIA 1
.
I v-zm iz... I
1 zz.,,, ...e. A I
."......(v.",
I IT-m- zezz
w......,,,,-..wwI
V eZZZAWW.F.,.." II
1,/./r ZeM44/1,1
leVie, 'MA I
t v-zziz.,,,,,,,,a 1
I vm..., .......-Ar,...A
L,-...A.,I z.,/ .2 1
I IT.F.r.r., ./Z/Z4,11 I
Llvz.eze1 ,,,A I
II 17Z! 1/Zr A I
I v ........:.w..w I
I ..AS..,./IAI
I WV Zy=e," JI II
I ZA I
i,,,... zz I
Percentiles ofr Performance5th 25th 75th 95th
Mean and Confidence Interval (+/-2SE)
200 250 300 350 400 450 500 550 600 650 700 750 800 850
SOURCE: Student Achievement Data, NCREL; tables 1.1, C.1, and C.3 in Mullis, I.V.S., et al. (1997). Mathematics Achievement in the PrimarySchool Years: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and
Educational Policy, Boston College.
Nations in which more than 10 percent of the population was excluded from testing. Latvia is designated LSS because only Latvian-speaking schoo
were tested, which represents less than 65 percent of the population.
^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
B-2
Exhibit B-2: Distributions of Science Achievement in the Fourth Grade
Country Meant Science Achievement Scale Score
FiWSame as
First in the World 611 (9.0)1 1 1 1
Korea 597 (1.9) 1==12=211=21=1Lower than FiW I-
I
Japan 574 (1.8) 1 1/./MM144/-/A I
I I
United States 565 (3.1) 1L.w.m....m.,.....,,,....6,1
1
1
(Austria) 565 (3.3) 1 wz,4,w, " :":".1 I
1 j(Australia) 562 (2.9) 1,e."-/./.4,;(W.F.,,ZAIA 1
I
(Netherlands) 557 (3.1) 1 lel./.4rWIA I
Czech Republic 557 (3.1) 1 1,,,://./M/7"./17 I
England*^ 551 (3.3) 1 WIZ/7 / / /x /11../11 I1
wI .m.erwzirt ICanada 549 (3.0) I
Singapore 547 (5.0) 1 fr."..,,-...,Auvz-v..4
(Slovenia) 546 (3.3) 1 17,/./ZMNIZ/VA i
Ireland 539 (3.3) 1 v .........mrzivz 1
Scotland^ 536 (4.2) ....-.
Hong Kong 533 (3.7) 164....... yv,....,A1
1I
(Hungary) 532 (3.4) II Ill /r' JZZA 1
New Zealand 531 (4.9) 1 1/ /Z 1../M=4"."/"./V I
Norway 530 (3.6) 1 11.4,,,,W/4/7/1 1
I
(Latvia {LSS}) 512 (4.9) $ 1.0W+MMVZ1
(Israel) 505 (3.6) 1 1/7/./W.MeZeMA II I
Iceland 505 (3.3) I IIVZIZZAIWWWZA 1
I
Greece 497 (4.1) I L.WZ/WeV.,/
Portugal 480 (4.0)
11
v/ /l1..,mew.....4
Cyprus 475 (3.3) 1", Z1M/.44.44,11
(Thailand) 473 (4.9) LW AMP ii`A
Iran, Islamic Republic 416 (3.9)I
1/Z.W.AMW ./ 1
(Kuwait) 401 (3.1) 14z.......1.-Arm...4.A
Percentiles ofPerformance
5th 25th 75th 95th
1
Mean and Confidence Interval (+/-2SE)
200 250 300 350 400 450 500 550 600 650 700 750 800 850
SOURCE: Student Achievement Data, NCREL; tables 1.1, C.1,. and C.3 in Martin, Michael 0., et aL (1997). Science Achievement in the PrimarySchool Years: lEA's Third Internastional Mathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and
Educational Policy, Boston College.
* Nations in which more than 10 percent of the population was excluded from testing. Latvia is designated LSS because only Latvian-speaking
schools were tested, which represents less than 65 percent of the population.^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
BEST COPY AVAILABLE
B-3
Exh
ibit
B-3
: Fou
rth-
Gra
de M
athe
mat
ics
Perf
orm
ance
, by
Subt
opic
Ave
rage
Per
cent
Cor
rect
Mea
ning
of W
hole
Num
bers
Inte
gers
& W
hole
Num
ber
Ope
ratio
nsC
omm
on F
ract
ions
Dec
imal
Fra
ctio
nsE
stim
atin
g Q
uant
ity&
Siz
e
Rou
ndin
g &
Est
imat
ing
Com
puta
tions
Mea
sure
men
t Uni
ts
Hig
her
than
FiW
Sam
e as
FiW
Sam
e as
FiW
Hig
her
than
FiW
Hig
her
than
FiW
Sam
e as
FiW
Hig
her
than
FiW
Kor
ea88
.1
Sam
e as
FiW
Kor
ea82
.5
Japa
n80
.2
Sin
gapo
re79
.2
Firs
t in
the
Wor
ld78
.6
Hon
g K
ong
76.9
Cze
ch R
epub
lic74
.4
Low
er th
an F
iW
Sin
gapo
re71
.8
Firs
t in
the
Wor
ld65
.7
Hon
g K
ong
65.3
Japa
n62
.5
Kor
ea61
.5
Hun
gary
61.2
Net
herla
nds
60.2
Low
er th
an F
iW
Sin
gapo
re83
.2
Kor
ea75
.0
Hon
g K
ong
72.6
Japa
n71
.3
Sam
e as
FiW
Japa
n78
.0
Hon
g K
ong
70.5
Cze
ch R
epub
lic67
.9
Hun
gary
67.3
Kor
ea67
.0
Net
herla
nds
65.9
Sin
gapo
re65
.3
Aus
tral
ia61
.2
Nor
way
60.6
Sam
e as
FiW
Kor
ea76
.8
Japa
n74
.8
Hon
g K
ong
74.8
Sin
gapo
re74
.3
Firs
t in
the
Wor
ld73
.8
Net
herla
nds
68.7
Hun
gary
68.6
Cze
ch R
epub
lic68
.1
Uni
ted
Sta
tes
67.2
Low
er th
an F
iW
Japa
n76
.5
Kor
ea72
.1
Net
herla
nds
72.0
Cze
ch R
epub
lic68
.2
Hon
g K
ong
67.6
Hun
gary
66.3
Sam
e as
FiW
Firs
t in
the
Wor
ld81
.7
Sin
gapo
re81
.6
Japa
n79
.9
Low
er th
an F
iWF
irst i
n th
e W
orld
59.7
Low
er th
an F
iWH
ong
Kon
g76
.4
Hun
gary
74.5
Net
herla
nds
73.3
Cze
ch R
epub
lic72
.9
Isra
el69
.8
Uni
ted
Sta
tes
69.8
Aus
tral
ia68
.9
Can
ada
68.7
INT
ER
NA
TIO
NA
L 65
.9N
orw
ay61
.4
New
Zea
land
59.1
Eng
land
59.0
Tha
iland
57.3
Net
herla
nds
73.2
Hun
gary
71.5
Uni
ted
Sta
tes
67.6
Isra
el66
.3
INT
ER
NA
TIO
NA
L 64
.1C
anad
a61
.5
Aus
tral
ia61
.5
Nor
way
57.6
Eng
land
53.1
Tha
iland
52.7
New
Zea
land
50.8
Sin
gapo
re61
.3
Aus
tral
ia60
.2
Nor
way
60.0
Firs
t in
the
Wor
ld58
.0
INT
ER
NA
TIO
NA
L56
.8
Low
er th
an F
iW
Uni
ted
Sta
tes
53.7
Isra
el53
.3
Aus
tral
ia52
.0
Can
ada
50.4
Cze
ch R
epub
lic49
.7
INT
ER
NA
TIO
NA
L 50
.0E
ngla
nd49
.1
Tha
iland
46.4
New
Zea
land
45.3
Nor
way
'39
.2
Net
herla
nds
47.1
Uni
ted
Sta
tes
46.8
INT
ER
NA
TIO
NA
L 44
.6A
ustr
alia
43.7
Tha
iland
43.6
Can
ada
41.5
Cze
ch R
epub
lic40
.1
Hun
gary
36.4
Isra
el34
.2
Eng
land
32.0
Nor
way
28.9
New
Zea
land
26.3
INT
ER
NA
TIO
NA
L 57
.2F
irst i
n th
e W
orld
52.9
Eng
land
52.2
New
Zea
land
51.9
Can
ada
50.7
Isra
el50
.5
Tha
iland
49.8
Uni
ted
Sta
tes
47.1
.
Can
ada
61.6
Isra
el61
.2
Aus
tral
ia59
.4
INT
ER
NA
TIO
NA
L 55
.2N
ew Z
eala
nd47
.9E
ngla
nd47
.7
Nor
way
46.4
Tha
iland
39.1
Eng
land
52.3
Can
ada
51.7
Isra
el51
.0
New
Zea
land
49.7
Uni
ted
Sta
tes
48.0
Tha
iland
43.2
Per
imet
er, A
rea,
&V
olum
eG
eom
etry
: Pos
ition
&S
hape
s
Sym
met
ry,
Con
grue
nce,
&S
imila
rity
Pro
port
iona
lity
Pat
tern
s, R
elat
ions
, &F
unct
ions
Equ
atio
ns &
For
mul
asD
ata
& S
tatis
tics
Hig
her
than
FiW
Sam
e as
FiW
Sam
e as
FiW
Sam
e as
FiW
Sam
e as
FiW
Hig
her
than
FiW
Sam
e as
FiW
Sin
gapo
re75
.9
Kor
ea73
.5
Sam
e as
FiW
Firs
t in
the
Wor
ld77
Aus
tral
ia71
.9
Low
er th
an F
iW
Sin
gapo
re88
.9
Kor
ea88
.8
Hon
g K
ong
85.4
Firs
t in
the
Wor
ld85
.2
Eng
land
80.9
Low
er th
an F
iW
Sin
gapo
re63
.7
Net
herla
nds
62.7
Firs
t in
the
Wor
ld59
.4K
orea
58.1
Japa
n57
.7
Cze
ch R
epub
lic57
.3
Hun
gary
55.5
Hon
g K
ong
53.3
Low
er th
an F
iW
Kor
ea85
.0
Firs
t in
the
Wor
ld78
.0
Hon
g K
ong
77.8
Sin
gapo
re77
.5
Japa
n77
.2
Low
er th
an F
iW
Kor
ea85
.3
Sam
e as
FiW
Firs
t in
the
Wor
ld86
.4
Low
er th
an F
iWJa
pan
84.3
Sin
gapo
re82
.6
Firs
t in
the
Wor
ld78
.5
Hon
g K
ong
76.9
Hun
gary
72.4
Net
herla
nds
72.2
Low
er th
an F
iW
Sin
gapo
re80
.8
Kor
ea79
.8
Japa
n78
.3
Hon
g K
ong
75.8
Net
herla
nds
75.4
Uni
ted
Sta
tes
73.0
Can
ada
68.0
Cze
ch R
epub
lic67
.1
Aus
tral
ia-
67.1
Eng
land
65.8
Isra
el63
.4
New
Zea
land
61.5
INT
ER
NA
TIO
NA
L61
.8
Hun
gary
60.3
Nor
way
59.5
Tha
iland
55.5
Net
herla
nds
71.0
Hon
g K
ong
70.8
Cze
ch R
epub
lic68
.2
Japa
n67
.5
Firs
t in
the
Wor
ld65
.9
Hun
gary
62.8
Aus
tral
ia62
.2
Low
er th
an F
iW
Eng
land
70.6
Net
herla
nds
70.3
Hon
g K
ong
70.1
Can
ada
69.4
Uni
ted
Sta
tes
67.7
Cze
ch R
epub
lic67
.4
Japa
n66
.7S
inga
pore
65.3
New
Zea
land
65.3
Kor
ea65
.1
Hun
gary
62.3
INT
ER
NA
TIO
NA
L 61
.7Is
rael
59.1
Nor
way
58.3
Tha
iland
47.6
Aus
tral
ia79
.9
Japa
n79
.5
Uni
ted
Sta
tes
78.5
Can
ada
78.5
Cze
ch R
epub
lic77
.8
Net
herla
nds
74.I
Hun
gary
73.3
INT
ER
NA
TIO
NA
L 70
.5N
ew Z
eala
nd69
.1
Isra
el69
.0
Tha
iland
66.5
Nor
way
56.1
Hun
gary
71.4
Net
herla
nds
70.4
Cze
ch R
epub
lic69
.8
Aus
tral
ia69
.6
Uni
ted
Sta
tes
69.1
Can
ada
67.1
INT
ER
NA
TIO
NA
L 63
.2Is
rael
.62
.6
Eng
land
62.0
New
Zea
land
57.5
Nor
way
54.7
Tha
iland
52.4
Aus
tral
ia51
.3
Eng
land
46.6
Isra
el46
.0
INT
ER
NA
TIO
NA
L 45
.6U
nite
d S
tate
s45
.5
Can
ada
43.8
New
Zea
land
43.7
Nor
way
40.8
Tha
iland
34.0
Cze
ch R
epub
lic71
.0Is
rael
.66
.8
Uni
ted
Sta
tes
65.6
INT
ER
NA
TIO
NA
L 62
.0C
anad
a61
.4
Aus
tral
ia60
.0T
haila
nd54
.0
New
Zea
land
50.5
Nor
way
49.5
Eng
land
48.2
Can
ada
59.3
INT
ER
NA
TIO
NA
L 58
.8U
nite
d S
tate
s58
.3
Isra
el57
.1
Eng
land
55.2
Nor
way
54.5
New
Zea
land
49.1
Tha
iland
48.2
SO
UR
CE
: U.S
. Nat
iona
l TIM
SS
Cen
ter,
Mic
higa
n S
tate
Uni
vers
ity.
BE
ST
CO
PY
AV
AIL
AB
LE
Exhibit B-4: Distributions of Mathematics Achievement in the Eighth Grade
Country Meant Mathematics Achievement Scale Score
- Higher than FiWi
I
Singapore 643 (4.9) ..Same FiW I I I- as
I I I I
Korea 607 (2.4) I"W./ /./.1" /IA
i 1
Japan 605 (1.9) 1v..... ...... I
I
Hong Kong 588 (6.5) .0.0,05/A ' 1
First in the World 587 (11.8) "././../% ifalreAl
Belgium-Flemish^ 565 (5.7) r "I"A11"././.I =MI
Czech Republic 564 (4.9) WI' AVA".1 ../.0.1.7.1%I
Slovak Republic 547 (3.3) . .I- Lower than FiWI
Switzerland^ 545 (2.8) I1 1
(Netherlands) 541 (6.7) I sr.,,rzz J
(Slovenia) 541 (3.1) I
i774777=6.7.).7.11 1
(Bulgaria) 540 (6.3) I v.. .."1
(Austria) 539 (3.0) Iv. ./././M111.1., A I
i 1 1
France 538 (2.9) t II I' t"."/"Mr.."./.4A /
Hungary 537 (3.2) 1...1.....=4/....1 1
1
Russian Federation 535 (5.3) I 11./.17.11WAIA I1 i
(Australia) 530 (4.0) I vw,T,-,.elfzzir.ri 1
Ireland 527 (5.1) I lezzzy.m..,....-A 1
ICanada 527 (2.4) IWZ,1WA,ZZA II I I I
(Belgium-French) 526 (3.4) 1v......m......-A 1
I I I 1
Sweden 519 (3.0) 1 v....T.m-ff.A t
(Thailand) 522 (5.7) I 141., ""o7/1".1 /A II I I
(Israel)* 522 (6.2) 1v.....././....
1
)...-...-/m/.. ..' A I(Germany)" 509 (4.5)1 1
New Zealand 508 (4.5) St,"/".././.1 ' .
England" 506 (2.6) !".01%./../ "4"/W.1"/
Norway 503 (2.2)1
(Denmark) 502 (2.8) -rm.% -i I
United States^ 500 (4.6)(Scotland) 498 (5.5) '01/#.0.0 !AellealreI
MENLatvia {LSS}^ 493 (3.1) Wage/ WAVAA
Spain 487 (2.0) " ../ WA,.111111111Iceland 487 (4.5) SA/AA VW!"
(Greece) 484 (3.1) -..w.... - ,I
(Romania) 482 (4.0) 'Ad' . / d".1. "1 I
Lithuania* 477 (3.5) 1v......m.r."1
Cyprus 474 (1.9) i v..i..m...r..." 1
;
Portugal 454 (2.5) I1/./..."/W/ZA
1
I1
Iran, Islamic Republic 428 (2.2) ".01/./5 rd'a05,1 1
(Kuwait) 392 (2.5) 11/1",..=4,11
i 1
(Columbia) 385 (3.4) I"viW All I
1
(South Africa) 354 (4.4) I v.z.:/./.
Percentiles of
Performance 15th 25th 75th 95thC=ZEZMEMZEM=
Mean and Confidence Interval (+/-2SE)
200 250 300 350 400 450 500 550 600 650 700 750 800 850
SOURCE: Student Achievement Data, NCREL; tables 1.1, E. I, and E.3 in Beaton, Albert E., et al. (1996). Mathematics Achievement in the Middle
School Years: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and
Educational Policy, Boston College.
Nations in which more than 10 percent of the population was excluded from testing. Latvia is designated LSS because only Latvian-speaking
schools were tested, which represents less than 65 percent of the population.
^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
B-5
06
Exhibit B-5: Distributions of Science Achievement in the Eighth Grade
Country Mean+ Science Achievement Scale Score
Same FiWasSingapore 607 (5.5) IrAWIMMIW.W1
First in the World 584 (8.7) I V.,.1.11,/, II i
Czech Republic 574 (4.3) t v....-z,-.."-mAiJapan 571 (1.6) I MN"! ....W1
1 VIII.W/BIZZ/rnelI
Korea 565 (1.9)I vym,-..m....-..-..w..,(Bulgaria) 565 (5.3)
I
(Netherlands) 560 (5.0) zi i
1 vm.,--.=,-zzmi a(Slovenia) 560 (2.5)r vm,-2...-..-rmi(Austria) 558 (3.7)
- Lower than FiWHungary 554 (2.8) I vm.,,m.-zzmi i
i
England" 552 (3.3) I vz..-m.m.-..z..wri a
I VAIZeWZAIBelgium-Flemish^ 550 (4.2)I mz.,..,...wzmzzi(Australia) 545 (3.9)
Slovak Republic 544 (3.2) I, v-..cm=,-.".....1
Russian Federation 538 (4.0) I1
Ireland 538 (4.5) I .iiii's i;
Sweden 535 (3.0) , I ge....m.m...-..-zmi 1
, I
United States^ 534 (4.7) a I
t(Germany)" 531 (4.8) I trz.ww..ACanada 531 (2.6) I vz.w...w..-Awl I
I ilwrzmi INorway 527 (1.9)1
New Zealand 525 (4.4) I 111,ZMAMWMA I!
(Thailand) 525 (3.7) 1.,MWZZIA tl
I 1/"..WZ.W.We'll(Israel)* 524 (5.7)I
I 1,./WIWZIZZZAIHong Kong 522 (4.7)
z.- .Switzerland^ 522 (2.5)I V.W.,;(?Wr.f."1
I(Scotland) 517 (5.1)
VA/WiWZ.r.I
IlSpain 517 (1.7)1
France 498 (2.5).
I vz.w.....zzmi I(Greece) 497 (2.2)
Iceland 494 (4.0) t I.W.1./WZ.WA
(Romania) 486 (4.7) Imz...-,m...zA
Latvia {LSS}^ 485 (2.7)
II
t vz...-..-..m.,..-mi a
I vz.r.-...is,Portugal 480 (2.3)(Denmark) 478 (3.1) I IIVWFM.WZ/Z1 I
Lithuania* 476 (3.4) I 1///,M/ZWAII I
I(Belgium-French) 471 (2.8)t 1....-..m.m.,-1 aIran, Islamic Republic 470 (2.4) tvefcrawcreerJICyprus 463 (1.9)
,
(Kuwait) 430 (3.7) 1 II
, ; 1
(Columbia) 411 (4.1) I vw.f.,...-zmi I
(South Africa) 326 (6.6) I v.....-..z..ezzA1
Percentiles of
Performance 75th 25th 75th 95th
Mean and Confidence Interval (+/-25E)
150 200 250 300 350 400 450 500 550 600 650 700 750 800
SOURCE: Student Achievement Data, NCREL; tables 1.1, E.1, and E.3 in Beaton, Albert E., et al. (1996). Science Achievement in the Middle
School Years: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the Study of Testing, Evaluation, and
Educational Policy, Boston College.
Nations in which more than 10 percent of the population was excluded from testing. Latvia is designated LSS because only Latvian-speaking
schools were tested, which represents less than 65 percent of the population.
^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
BEST COPY AVAILABLE 87 B-6
Exh
ibit
B-6
: Eig
hth-
Gra
de M
athe
mat
ics
Perf
orm
ance
, by
Subt
opic
Ave
rage
Per
cent
Cor
rect
Who
le N
umbe
rsC
omm
on F
ract
ions
Dec
imal
Fra
ctio
ns &
Per
cent
ages
Rel
atio
nshi
ps o
fF
ract
ions
Est
imat
ions
of
Qua
ntity
& S
ize
Rou
ndin
gE
stim
atin
gC
ompu
tatio
nsM
easu
rem
ent
Uni
tsP
erim
eter
, Are
a,V
olum
eM
easu
rem
ent
Est
imat
ions
& E
rror
sH
ighe
r th
an F
iWH
ighe
r th
an F
iWS
ame
as F
iWS
ame
as F
iWS
ame
as F
iWS
ame
as F
iWS
ame
as F
iWH
ighe
r th
an F
iWH
ighe
r th
an F
iWH
ighe
r th
an F
iWS
inga
pore
80.4
Sam
e as
FiW
Sin
gapo
re82
.4
Sam
e as
FiW
Sin
gapo
re77
.9
Firs
t in
the
Wor
ld73
.8
Cze
ch R
epub
lic72
.5
Hon
g K
ong
71.1
Japa
n69
.8
Hun
gary
68.7
Kor
ea68
.4
Low
er th
an F
iW
Sin
gapo
re85
.1
Firs
t in
the
Wor
ld78
.3
Kor
ea77
.6
Japa
n75
.3
Hon
g K
ong
74.1
Bel
gium
(F
I)71
.3
Low
er th
an F
iW
Japa
n77
.9
Sin
gapo
re77
.3
Firs
t in
the
Wor
ld76
.2
Eng
land
72.2
Hon
g K
ong
71.8
Sw
itzer
land
71.8
Cze
ch R
epub
lic71
.7
Net
herla
nds
71.2
Bel
gium
(F
I)70
.7
Kor
ea70
.7
Aus
tral
ia69
.3
Hun
gary
69.3
Sw
eden
69.3
Fra
nce
69.0
Ger
man
y69
.0
Low
er th
an F
iW
Firs
t in
the
Wor
ld89
.1
Cze
ch R
epub
lic86
.9
Sin
gapo
re86
.6
Can
ada
84.7
Kor
ea83
.8
Low
er th
an F
iW
Sin
gapo
re75
.8
Firs
t in
the
Wor
ld74
.5
Cze
ch R
epub
lic69
.8
Japa
n69
.0
Kor
ea66
.7
Low
er th
an F
iW
Sin
gapo
re77
.9
Sam
e as
FiW
Sin
gapo
re84
.5
Japa
n72
.6
Sam
e as
FiW
Cze
ch R
epub
lic82
.9
Sin
gapo
re81
.0
Sw
eden
80.8
Sam
e as
FiW
Japa
n72
.0
Sw
itzer
land
70.2
Bel
gium
(F
1)68
.6K
orea
67.6
Cze
ch R
epub
lic67
.4
Firs
t in
the
Wor
ld67
.0
Hon
g K
ong
65.4
Fra
nce
64.5
Rus
sian
Fed
erat
ion
62.7
Spa
in60
.8
Hun
gary
60.6
Sw
eden
58.9
Isra
el59
.0C
anad
a57
.9
Low
er th
an F
iW
Japa
n72
.2K
orea
69.9
Hon
g K
ong
69.4
Bel
gium
(F
1)68
.7
Firs
t in
the
Wor
ld65
.8
Sw
itzer
land
62.4
Cze
ch R
epub
lic59
.6N
ethe
rland
s58
.9F
ranc
e57
.8
Hun
gary
57.7
Isra
el57
.4
Low
er th
an F
iW
Japa
n74
.8
Cze
ch R
epub
lic73
.2
Bel
gium
(F
I)71
.8
Sw
itzer
land
71.6
Sw
eden
69.5
Fra
nce
69.0
Kor
ea67
.9
Hun
gary
67.5
Net
herla
nds
66.8
Firs
t in
the
Wor
ld66
.4
Nor
way
66.4
Rus
sian
Fed
erat
ion
65.9
Aus
tral
ia65
.0
Hon
g K
ong
65.2
Ger
man
y63
.5C
anad
a63
.0
Isra
el63
.0
Spa
in62
.4
Tha
iland
62.3
INT
ER
NA
TIO
NA
L61
.4
New
Zea
land
61.3
Eng
land
60.5
-
Low
er th
an F
iW
Hon
g K
ong
63.9
Kor
ea61
.8
Firs
t in
the
Wor
ld49
.9
Cze
ch R
epub
lic48
.9
Sw
itzer
land
46.6
Bel
gium
(F
I)45
.6F
ranc
e.
44.1
Hun
gary
42.9
Rus
sian
Fed
erat
ion
44.6
Net
herla
nds
44.0
Aus
tral
ia42
.1
Can
ada
40.4
Tha
iland
40.0
Sw
eden
39.3
INT
ER
NA
TIO
NA
L39
.0
Eng
land
37.1
Isra
el36
.9
Low
er th
an F
iW'
Kor
ea79
.5
Bel
gium
(F
I)78
.9
Hun
gary
78.2
Sw
itzer
land
78.2
Net
herla
nds
76.1
Fra
nce
75.2
Firs
t in
the
Wor
ld75
.0
Aus
tral
ia74
.5
Nor
way
74.5
Ger
man
y72
.9
Hon
g K
ong
72.7
Japa
n72
.5
Eng
land
71.8
New
Zea
land
70.6
Can
ada
69.7
Rus
sian
Fed
erat
ion
696
Low
er th
an F
iW
Bel
gium
(F
1)81
.0
Hun
gary
79.8
Japa
n79
.6U
nite
d S
tate
s79
.2
Net
herla
nds
78.5
Sw
eden
78.5
Aus
tral
ia77
.5
Sw
itzer
land
77.4
Tha
iland
75.4
Fra
nce
74.7
New
Zea
land
74.4
Hon
g K
ong
74.2
Eng
land
74.2
Rus
sian
Fed
erat
ion
73.5
Nor
way
72.3
INT
ER
NA
TIO
NA
L70
.7
Ger
man
y69
.9Is
rael
68.8
Spa
in62
.2
J
Bel
gium
(F
I)64
.1
Sw
itzer
land
62.5
Can
ada
61.3
Sw
eden
60.0
Aus
tral
ia59
.1
Net
herla
nds
58.7
Hon
g K
ong
57.8
New
Zea
land
57.6
Hun
gary
57.3
Nor
way
57.0
Uni
ted
Sta
tes
55.7
Eng
land
54.4
Ger
man
y52
.0Is
rael
51.9
Fra
nce
50.8
INT
ER
NA
TIO
NA
L50
.6
Rus
sian
Fed
erat
ion
50.0
Tha
iland
49.5
Spa
in44
.1
Sw
itzer
land
70.0
Fra
nce
67.2
Can
ada
65.1
Cze
ch R
epub
lic64
.9
Ger
man
y62
.6S
wed
en62
.2
Aus
tral
ia61
.0
Hun
gary
60.7
Nor
way
59.6
Net
herla
nds
62.5
Tha
iland
59.6
Rus
sian
Fed
erat
ion
59.2
Uni
ted
Sta
tes
58.5
Isra
el57
.6
INT
ER
NA
TIO
NA
L56
.9
New
Zea
land
56.8
Eng
land
53.8
Spa
in48
.2
Rus
sian
Fed
erat
ion
65.9
Fra
nce
64.9
Bel
gium
(E
l)64
.6
Sw
itzer
land
61.9
Can
ada
61.0
Sw
eden
59.2
Tha
iland
58.0
Net
herla
nds
57.8
Ger
man
y57
.2Is
rael
56.3
Uni
ted
Sta
tes
55.8
INT
ER
NA
TIO
NA
L55
.4
Nor
way
54.1
Aus
tral
ia53
.5
Spa
in50
.6N
ew Z
eala
nd49
.3
Eng
land
45.9
.
Can
ada
56.2
Aus
tral
ia55
.4R
ussi
an F
eder
atio
n55
.8
Tha
iland
54.2
Sw
eden
53.1
INT
ER
NA
TIO
NA
L52
.1
Ger
man
y51
.7
New
Zea
land
51.5
Nor
way
51.2
Uni
ted
Sta
tes
49.3
Eng
land
49.2
Spa
in49
.2
Can
ada
67.8
Nor
way
66.6
New
Zea
land
65.3
Rus
sian
Fed
erat
ion
63.5
Uni
ted
Sta
tes
62.7
INT
ER
NA
TIO
NA
L61
,0Is
rael
58.4
Tha
iland
57.1
Spa
in54
.8
e
Tha
iland
56.0
INT
ER
NA
TIO
NA
L55
.5
Ger
man
y55
.4
Aus
tral
ia55
.0
Net
herla
nds
53.5
Nor
way
53.4
Uni
ted
Sta
tes
52.3
New
Zea
land
48.6
Eng
land
46.8
Nor
way
36.4
New
Zea
land
35.0
Ger
man
y32
.9
Spa
in29
.9
Uni
ted
Sta
tes
27.8
Tha
iland
66.1
INT
ER
NA
TIO
NA
L64
.7
Isra
el63
.3
Uni
ted
Sta
tes
62.5
Spa
in55
.3U
nite
d S
tate
s53
.0..
2D G
eom
etry
Bas
ics
..
Pol
ygon
s &
Circ
les
3D G
eom
etry
&T
rans
form
atio
nsC
ongr
uenc
e &
Sim
ilarit
yP
ropo
rtio
nalit
yC
once
pts
Pro
port
iona
lity
Pro
blem
sP
atte
rns,
Rel
atio
ns,
& F
unct
ions
Equ
atio
ns &
For
mul
asD
ata
Rep
rese
nt.
& A
naly
sis
Sta
tistic
s &
Pro
babi
lity
Hig
her
than
FiW
Hig
her
than
FiW
Sam
e as
FiW
Hig
her
than
FiW
Hig
her
than
FiW
Hig
her
than
FiW
Sam
e as
FiW
Hig
her
than
FiW
Sam
e as
FiW
Sam
e as
FiW
Japa
n78
.9
Sam
e as
FiW
Kor
ea73
.0
Japa
n70
.8
Sin
gapo
re70
.6
Sam
e as
FiW
s Ja
pan
84.3
Sin
gapo
re78
.6
Firs
t in
the
Wor
ld78
.0
Cze
ch R
epub
lic77
.7
Fra
nce
77.7
Hon
g K
ong
77.6
Sw
itzer
land
75.9
Bel
gium
(F
I)73
.7
Kor
ea73
.3
Hun
gary
71.0
Net
herla
nds
70.6
Rus
sian
Fed
erat
ion
70.2
Can
ada
70.1
Low
er th
an F
iW
Japa
n79
.7K
orea
76.9
Sin
gapo
re76
.7
Sam
e as
FiW
Sin
gapo
re70
.1
Sam
e as
FiW
Sin
gapo
re76
.7
Sam
e as
FiW
Japa
n70
.0
Kor
ea68
.4
Sin
gapo
re67
.2
Firs
t in
the
Wor
ld63
.2
Hon
g K
ong
60.8
Cze
ch R
epub
lic60
.2
Eng
land
60.1
Sw
itzer
land
59.5
Bel
gium
(F
I)59
.2
Hun
gary
59.1
Can
ada
58.2
Aus
tral
ia57
.6
Net
herla
nds
56.9
New
Zea
land
55.7
Fra
nce
55.0
Low
er th
an F
iW
Sin
gapo
re79
.8
Sam
e as
FiW
Japa
n84
.3
Firs
t in
the
Wor
ld83
.2
Kor
ea81
.9
Fra
nce
81.4
Sw
itzer
land
81.3
Sin
gapo
re81
.0
Net
herla
nds
80.6
Cze
ch R
epub
lic79
.6
Sw
eden
79.5
Bel
gium
(F
I)79
.5
Hon
g K
ong
'76
.4
Ger
man
y76
.0
Low
er th
an F
iW
Firs
t in
the
Wor
ld81
.3
Sin
gapo
re78
.5
Kor
ea77
.8
Bel
gium
(F
1)77
.7
Japa
n75
.0
Hon
g K
ong
73.0
Net
herla
nds
72.4
Low
er th
an F
iW
Sin
gapo
re78
.3.
Kor
ea75
.5
Hon
g K
ong
73.5
Firs
t in
the
Wor
ld65
.0
Rus
sian
Fed
erat
ion
63.4
Net
herla
nds
62.9
Tha
iland
62.4
Cze
ch R
epub
lic62
.1
Hun
gary
61.6
Aus
tral
ia58
.8
Sw
itzer
land
57.9
Bel
gium
(F
I)58
.8
Fra
nce
56.1
New
Zea
land
55.8
Eng
land
54.9
Can
ada
54.7
INT
ER
NA
TIO
NA
L53
.8
Isra
el52
.5
Nor
way
52.1
Ger
man
y49
.3
Low
er th
an F
iW
Kor
ea57
.0
Japa
n56
.9F
irst i
n th
e W
orld
50.6
Hon
g K
ong
50.1
Net
herla
nds
41.9
Low
er th
an F
iW
Japa
n67
.8
Kor
ea65
.1
Hon
g K
ong'
64.0
Cze
ch R
epub
lic61
.3
Firs
t in
the
Wor
ld61
.1
Bel
gium
(F
I)59
.4
Sw
itzer
land
57.3
Rus
sian
Fed
erat
ion
56.3
Tha
iland
56.0
Hun
gary
55.9
Net
herla
nds
55.9
Fra
nce
55.7
Low
er th
an F
iW
Japa
n72
.3
Hon
g K
ong
70.5
Kor
ea70
.0
Firs
t in
the
Wor
ld67
.8
Cze
ch R
epub
lic64
.3
Bel
gium
(F
I)62
.1
Rus
sian
Fed
erat
ion
61.4
Hun
gary
60.9
Isra
el58
.3
'Lo
wer
than
FiW
Hon
g K
ong
65.4
Bel
gium
(E
1)64
.1
Fra
nce
61.9
Isra
el59
.7
Cze
ch R
epub
lic57
.4
Rus
sian
Fed
erat
ion
57.2
Firs
t in
the
Wor
ld56
.1
Tha
iland
53.5
Hun
gary
51.6
Can
ada
50.5
Aus
tral
ia50
.4IN
TE
RN
AT
ION
AL
48.9
Sw
itzer
land
48.6
Eng
land
47.0
Net
herla
nds
44.0
Low
er th
an F
iW
Hon
g K
ong
71.0
Fra
nce
69.7
Cze
ch R
epub
lic66
.1
Tha
iland
62.7
Rus
sian
Fed
erat
ion
62.3
Firs
t in
the
Wor
ld62
.2
Bel
gium
(F
I)57
.5
Isra
el54
.8
Hun
gary
53.4
Can
ada
52.8
Net
herla
nds
51.0
Low
er th
an F
iW
Sw
itzer
land
72.4
Can
ada
70.0
Sw
eden
68.7
Aus
tral
ia67
.5
Fra
nce
66.5
Hun
gary
66.2
New
Zea
land
65.9
Eng
land
65.3
Nor
way
64.7
Cze
ch R
epub
lic63
.4
Uni
ted
Sta
tes
63.1
Ger
man
y62
.4
Isra
el62
.3
INT
ER
NA
TIO
NA
L59
.6
Spa
in58
.0R
ussi
an F
eder
atio
n56
.9
Tha
iland
56.5
Can
ada
40.3
Sw
eden
39.2
Bel
gium
(E
l)39
.0
Aus
tral
ia37
.8
Fra
nce
37.3
Sw
itzer
land
36.8
Cze
ch R
epub
lic37
.3
Tha
iland
36.5
Eng
land
34.8
Hun
gary
34.6
Rus
sian
Fed
erat
ion
33.5
INT
ER
NA
TIO
NA
L33
.4
New
Zea
land
33.4
Uni
ted
Sta
tes
31.5
Isra
el31
.4
Nor
way
30.9
Spa
in27
.4
Ger
man
y26
.5
Tha
iland
54.0
Fra
nce
53.6
Aus
tral
ia52
.8
Can
ada
52.4
Spa
in52
.2
Sw
itzer
land
51.9
Net
herla
nds
51.7
INT
ER
NA
TIO
NA
L50
.6
Uni
ted
Sta
tes
49.4
Ger
man
y46
.4N
ew Z
eala
nd46
.6
Nor
way
44.9
Eng
land
44.1
Sw
eden
43.7
Nor
way
75.5
Aus
tral
ia75
.2C
anad
a75
.0
Eng
land
74.8
Tha
iland
74.5
New
Zea
land
74.1
Hun
gary
73.7
Isra
el73
.2
Uni
ted
Sta
tes
72.6
Rus
sian
Fed
erat
ion
72.0
INT
ER
NA
TIO
NA
L70
.9
Spa
in70
.4
New
Zea
land
67.3
Eng
land
66.3
Aus
tral
ia65
.5Is
rael
64.8
Tha
iland
64.5
Ger
man
y63
.7IN
TE
RN
AT
ION
AL
63.6
Nor
way
58.6
Uni
ted
Sta
tes
58.4
Spa
in58
.2
Sw
eden
55.9
..
Aus
tral
ia52
.2 T
haila
nd53
.4C
anad
a51
.6 R
ussi
an F
eder
atio
n53
.2
INT
ER
NA
TIO
NA
L50
.1Is
rael
53.2
Sw
eden
49.2
INT
ER
NA
TIO
NA
L51
.4
Ger
man
y49
.0 U
nite
d S
tate
s51
.1
Eng
land
47.2
Spa
in50
.3
New
Zea
land
47.1
Nor
way
48.7
Isra
el46
.5 G
erm
any
47.6
Nor
way
47.1
Sw
eden
47.2
Uni
ted
Sta
tes
45.2
Spa
in43
.6_
INT
ER
NA
TIO
NA
L52
.6
Sw
itzer
land
51.2
Aus
tral
ia50
.9
Nor
way
46.9
New
Zea
land
45.9
Sw
eden
45.9
Eng
land
45.6
Ger
man
y45
.1
Uni
ted
Sta
tes
44.1
Spa
in43
.6
Ger
man
y44
.3
New
Zea
land
43.2
Sw
eden
42.8
Nor
way
42.4
Spa
in39
.3
Uni
ted
Sta
tes
39.1
II
Spa
in49
.5
Uni
ted
Sta
tes
47.9
Sw
eden
46.1
wM
¢htg
an S
tate
Ivy.
BE
ST
CO
PY
AV
AIL
AB
LE
Exhibit B-7: Distributions of General MathematicsAchievement in the Twelfth Grade
Country Meant Mathematics Achievement Scale Score
Same as FiW 1
(Netherlands)*Sweden(Denmark)First in the WorldSwitzerland(Iceland)(Norway)(Australia)
Lower FiW
560552
547545
540534528522
(4.7)(4.3)(3.3)(3.5)(5.8)(2.0)(4.1)(9.3)
I I I I
I vm............e..,...-A i
1 1 1
1 ViViWAMMV.46.i9VA I
1 1 1
I vziem.,wm....e.el II I
I VZ46,./....W.W.,,YA I
1 i 1
o1/7....w.,mm.,.w..1 I
1 1
I 11r 4,VAAMPWAIM I
I 1.-zzzr,":"....wi II
II 11:44174,MMEMA/AIA I
I Ithan(France)New Zealand^(Canada)(AUstria)*
(Slovenia)(Germany)"Hungary(Italy)Russian Federation*LithuaniaCzech Republic(United States)Cyprus*(South Africa)
523522519518512495483476471
469466461
446356
(5.1)(4.5)(2.8)(5.3)(8.3)(5.9)(3.2)(5:5)(6.2)(6.1)(12.3)(3.2)(2.5)(8.3) I__rtecmra-7--,,--1
I
I V.,,,,AMMV/WVA I
1 1
I V".44.10:4=0741/1 $
I
1
I WYW/JMNAZAA44.4 I
I
1 1 1
I V/MAMMWMA1 1 1
I 11./AriliZZAi 1 1
we...4,Ammvvzi.....?1
wz.),./..a..ezewn II1
I Vor..WYMMR.:05W.01 I
1 1
$ViVAI.77MMWMA
wr.4)..m.m.....,eyzn
I
III 1
1 1 I
I 1,4".411 ,,AW-40WZA I
1 1 1
I I/M.,,WIZZAIL $
1 i 1
Percentiles ofPerformance
5th 25th 75th 95th
200 250 300 350 400 450 500 550 600 650 700 750 800 850
Mean and Confidence Interval (+/-2SE)
SOURCE: Student Achievement Data, NCREL; tables 2.1, E.2, and E.7 and figure B.4 in Mullis, I.V.S., et al. (1998). Mathematics and Science
Achievement in the Final Year of Secondary School: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the
Study of Testing, Evaluation, and Educational Policy, Boston College.
* Nations in which more than 10 percent of the population was excluded from testing.^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
B-8
Exhibit B-8: Distributions of General Science Achievement in the Twelfth Grade
Country Meant Science Achievement Scale Score
Same as FiW
Sweden .
(Netherlands)*(Iceland)First in the World(Norway)(Canada)New Zealand^(Australia)
Lower FiW
559558549547544532
.529527
(4.4)(5.3)(1.5)(4.8)(4.1)(2.6)(5.2)(9.8)
c
I I I I
/ 1,7M./.1%/././1 I
1 I I I
I 1/.444/....1.1././..1I I
I
1
I lietWeiri./Z,ZA II I I
I 1,7/7":". .1"/./././J I1 1 1
I 14,-...,......,:m...../.41 1
t vz,,,...............,,,,,...-A
I 11/./1 M:M=411.1,ZZAI I 1
I i
thanSwitzerland(Austria)*(Slovenia)(Denmark)(Germany)"Czech Republic(France)Russian Federation*(United States)(Italy)HungaryLithuaniaCyprus*(South Africa)
523
520517509497487487481
480475471
461448349
(5.3)(5.6)(8.2)(3.6)(5.1)(8.8)(5.1)(5.7)(3.3)(5.3)(3.0)(5.7)(3.0)
(10.5)
,I I I
.
1 1/Z/Zei":".447.1 II I 1
1IVIZZ.F.rWZMA I
I I I 1
1 VIZ".../ZZA II I I 1
I trm....v.7....zzzzA 1
WW.F.AMM/M.1131 i 1
WZ.1.1.0iWZAI I 1
I
I LIZZZAMMeZMA II I 1
/ IlZer.Z./././.11 tI I I
I lerWe.1.4M1././MA II 1
I LeZZ/11.,W.e/W1 II I
I 14.1:4460./W..AMI I1 1
ILWAIJf./././VA
I I
1VeZZIAMMVZ/ZA
I 1
I wiz/m.1Percentiles of
. Performance 15th 25th 75th 95thrMean and Confidence Interval (+/-2SE)
200 250 300 .350 400 450 500 550 600 650 700 750 800 850
SOURCE: Student Achievement Data, NCREL; tables 2.2, E.3, and E.8 and figure B.4 in Mullis, I.V.S., et al. (1998). Mathematics and ScienceAchievement in the Final Year of Secondary School: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the
Study of Testing, Evaluation, and Educational Policy, Boston College.
Nations in which more than 10 percent of the population was excluded from testing.^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
BEST COPY AVAILABLE
91B-9
Exhibit B-9: Distributions of Advanced Mathematics Achievementin the Twelfth Grade'
Country Meant Mathematics Achievement Scale Score
Higher than FiW 1
FranceRussian Federation*Switzerland(Denmark)Cyprus*LithuaniaSweden
Same FiW
557542
533
522518516512
(3.9)(9.2)(5.0)(3.4)(4.3)(2.6)(4.4)
I ! I
I IM./..././ZA Ii1 1
1 1/.41,W. W.7.4/./1 i1 i i 1
I LIZZliZMWW.F.FA I
I 1 1
1 VZ/7./WZIZZA I1 1 1
1 LIZ(/./.7.4/ZilI
i v......)...,,I,A1 1 1
I Erzz.e... .e....../1
Ias
(Australia)Greece^CanadaFirst in the World(Slovenia)(Italy)Czech Republic
Lower than FiW
525513509490475474
469
(11.6)(6.0)(4.3)(4.9)(9.2)(9.6)
(11.2)
I I
1 11:44,7.7.177:11 11 1
1 w...w.m...-...7-,-,,,,A i1 1 1
W.T.44f1W.r./Z/JI1 1 I
I 1
I V/fiJiMMEMW,,,,VA i1 1
1 viii_____ W17 II 1 1v...," ZIA
I I
Germany^(United States)(Austria)*
465442436
(5.6)(5.9)(7.2)
I I I I
I
I1 11%.".77. Z.IZZ A j1 1 1
1 vA,.//ammor.tae;
Per6entiles ofPerformance 1
5th 25th 75th 95th
1Mean and Confidence Interval (+/-2SE)
200 250 300 350 400 450 500 550 600 650 700 750 800 850
SOURCE: Student Achievement Data, NCREL; tables 5.1, E.4, and E.9 and figure B.5 in Mullis, I.V.S., et al. (1998). Mathematics and ScienceAchievement in the Final Year of Secondary School: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the
Study of Testing, Evaluation, and Educational Policy, Boston College.
Approximately 19 percent of the international school leaving age cohort, 14 percent of the U.S. cohort, and 65 percent of the FiW cohort were cover
by the TIMSS sample for this assessment.
* Nations in which more than 10 percent of the population was excluded from testing.^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
B-10
Exhibit B-10: Distributions of Advanced Science (Physics) Achievementin the Twelfth Grade'
Country Meant Science Achievement Scale Score
Higher than FiWNorway^SwedenRussian Federation*(Denmark)(Slovenia)Germany^(Australia)Cyprus*SwitzerlandGreece^CanadaFrance
Same FiW
581
573
545
534523
522518494488486485466
(6.5)(3.9)
(11.6)(4.2)(15.5)(11.9)(6.2)(5.8)(3.5)(5.6)(3.3)(3.8)
I I I
I ww-,-....,==ezzmA II i
1 1/77.46/7.4/./A II
1FAIZTI,Y4eVer/A II
I
I II/f/ //i0// /A I
I I I
I v........../.....vA1
Ii 1
1 V/ZZIWZJI II I
I vm-,-...v. AA 1
1 V1 /,,,M=e: /MA ITI vz...6,-..,./Aw.......0-A I
1 1
1 VIZZAMIWAeZetel I
I I
v...,,,,ww......,-...-.41 1
I VZIAMEWIA I
I Ias
Latvia (LSS)Czech RepublicFirst in the World(Austria)*
Lower than FiW
488451
446435
(21.5)(6.2)(4.1)(6.4)
Solff 11
I I
I V777.,i417.4 I
II 1.),....w....:ez.,.....A1I 1 1
I 1
1 1
(United States) 423 (3.3)I
I vix....:/zzra
Percentiles ofPerformance
5th 25th 75th 95th
200 250 300 350 400 450 500 550 600 650 700 750 800 850
Mean and Confidence Interval (+/-2SE)
SOURCE: Student Achievement Data, NCREL; tables 8.1, E.5, and E.10 and figure B.6 in Mullis, I.V.S., et al. (1998). Mathematics and Science
Achievement in the Final Year of Secondary School: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the
Study of Testing, Evaluation, and Educational Policy, Boston College.
I Approximately.13 percent of the international school leaving age cohort, 15 percent of the U.S. cohort, and 67 percent of the FiW cohort were cover
by the TIMSS sample for this assessment.* Nations in which more than 10 percent of the population was excluded from testing. Latvia is designated LSS because only Latvian-speaking
schools were tested, which represents less than 65 percent of the population.^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
B-11
Exhibit B-11: Distributions of Advanced Mathematics Achievement forTwelfth-Grade FiW AP Students'
Country . Meant Mathematics Achievement Scale Score
Same FiW APasFirst in the World AP- Lower than FiW AP
587 (4.3) I VZI.M./.4i4IZZ/71 I
I 1
FranceRussian Federation *.Switzerland(Australia) .
(Denmark)Cyprus*LithuaniaGreece^SwedenCanada(Slovenia)(Italy)Czech RepublicGermany^(United States)(Austria)*
557 (3.9)542 (9.2)533 (5.0)525 (11.6)522 (3.4)518 (4.3)516 (2.6)513 (6.0)512 (4.4)509 (4.3)475 (9.2)474 (9.6)
469 (11.2)4655.6)442 (5.9)436 (7.2)
. -
.
F, ..--.4.,1
I
I wir./Aw ...,1
1,-....w ........w...,I
I 145% lffriV.VA
zwz...wmez ..,-...wA1
I WIZ, Ze ,..vi
I ......wz....../. ,......11 1
l v ..,/./.444,11 1,/, ./.17.4 I
I
I ifz," ,-./1
I VW,/ ArZZA
1144,1:4,14444,I I/..../7.44/1
Percentiles ofPerformance
5th 25th 75th 95thUMW/ .01 r
Mean and Confidence Interval (+/-2SE)
200 250 300 350 400 450 500 550 600 650 700 750 800 850
SOURCE: Student Achievement Data, NCREL; tables 5.1, E.4, and E.9 and, figure B.5 in Mullis, I.V.S., et al. (1998). Mathematics and ScienceAchievement in the Final Year of Secondary School: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the
Study of Testing, Evaluation, and Educational Policy, Boston College.
Approximately 19 percent of the international school leaving age cohort, 14 percent of the U.S. cohort, and 28 percent of the FiW cohort were coverby the TIMSS sample for this assessment.
* Nations in which more than 10 percent of the population was excluded from testing.
is Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
04B-12
Exhibit B-12: Distributions of Advanced Science (Physics) Achievementfor Twelfth-Grade FiW AP Students'
Country Meant Science Achievement Scale Score
Same FiW APas
Norway^SwedenFirst in the World APRussian Federation*(Slovenia)Germany^
Lower FiW AP
581
573
564
545
523
522
(6.5)(3.9)(5.9)
(11.6)(15.5)(11.9)
I FIZZ///,417.0WA
1VZ/Z/ZinneMZZA
I
l 4%./.07,4,17/11 11
11
1/Z47.44.6,717/A I
1 wz.I z....":44,.-zi 1
1- than(Denmark)(Australia)Cyprus*Latvia (LSS)SwitzerlandGreece"CanadaFranceCzech Republic(Austria)*(United States)
534
518494
488488486485466451
435423
(4.2)(6.2)(5.8)(21.5)(3.5)(5.6)(3.3)(3.8)(6.2)(6.4)(3.3)
ii
1 VZ,./ZIMMIWZrA 1
I
I vz..../..,:mm/m/1il
I 1/VM,7=/...WA
/ 16,7.41./AWZMA I.i t
1 1/M./.,?7./.4071
1 iv,w.,,,:mmi 1
1
1 v...w-Amezz..A1 I
I IIZZ/V1=4/A II
I ILIZZ.4.0%/A I
_..
Percentiles ofE Performance
5th 25th 75th 95th
200 250 300 350 400 450 500 550 600 650 700 750 800 850
Mean and Confidence Interval (+/-2SE)
SOURCE: Student Achievement Data NCREL; tables 8.1, E.5, and E.10 and figure B.6 in Mullis, I.V.S.,etal. (1998). Mathematics and Science
Achievement in the Final Year of Secondary School: lEA's Third International Mathematics and Science Study. Chestnut Hill, MA: Center for the
Study of Testing, Evaluation, and Educational Policy, Boston College.
Approximately 13 percent of the international school leaving age cohort, 15 percent of the U.S. cohort, and 7 percent of the FiW cohort were covere
by the TIMSS sample for this assessment.* Nations in which more than 10 percent of the population was excluded from testing. Latvia is designated LSS because only Latvian-speaking
schools were tested, which represents less than 65 percent of the population.^ Nations in which a participation rate of 75 percent of the schools and students combined was achieved only after replacements for refusals
were substituted.
t Standard errors appear in parentheses. Because results are rounded to the nearest whole number, some totals may appear inconsistent.
NOTE: Nations not meeting international guidelines are shown in parentheses.
95.B-13
United StatesDepartment of Education
Washington, DC 20208-5573
Official BusinessPenalty for Private Use, $300
SA: 1999-3011
647
Postage and Fees PadU.S. Department of Education
Permit No. G-17
Standard Mail (A)
U.S. Department of EducationOffice of Educational Research and Improvement (OERI)
National Library of Education (NLE)Educational Resources Information Center (ERIC)
NOTICE
REPRODUCTION BASIS
IC
This document is covered by a signed "Reproduction Release(Blanket) form (on file within the ERIC system), encompassing allor classes of documents from its source organization and, therefore,does not require a "Specific Document" Release form.
u7 This document is Federally-funded, or carries its own permission toreproduce, or is otherwise in the public domain and, therefore, maybe reproduced by ERIC without a signed Reproduction Release form(either "Specific Document" or "Blanket").
EFF-089 (9/97)