DOCUMENT RESUME

ED 433 117 PS 027 813

AUTHOR Graves-Desai, Kelly, Ed.; Eaton, Susan, Ed.; Walser, Nancy,Ed.

TITLE The Harvard Education Letter, 1998.INSTITUTION Harvard Univ., Cambridge, MA. Graduate School of Education.ISSN ISSN-8755-3716PUB DATE 1998-00-00NOTE 49p.; Published six times a year. For 1995-1997 issues, see

PS 027 810-812.AVAILABLE FROM Harvard Education Letter, P.O. Box 850953, Braintree, MA

02185; Tel: 617-495-3432 in Massachusetts; Tel: 800-513-0763(Toll Free outside Massachusetts) ($32 for individuals; $39for institutions; $40 for Canada/Mexico; $42 other foreign;single copies, $5).

PUB TYPE Collected Works Serials (022)JOURNAL CIT Harvard Education Letter; v14 n1-6 Jan-Dec 1998EDRS PRICE MF01/PCO2 Plus Postage.DESCRIPTORS Academic Achievement; Adolescents; Attention Deficit

Disorders; *Bilingual Education; Change Agents;Disadvantaged; Educational Change; Elementary SecondaryEducation; Hispanic Americans; Integrated Services; *MixedAge Grouping; Newsletters; Nongraded Instructional Grouping;Professional Development; *Reading Difficulties; Scheduling;School Safety; School Security; School Size; *ScienceEducation; *Student Employment; Student Evaluation; *TeacherCollaboration; Teacher Shortage; Teacher Supply and Demand;Teachers

IDENTIFIERS Emotional Intelligence; Phonemic Awareness

ABSTRACTThis document is comprised of volume 14 of the Harvard

Education Letter, published bimonthly and addressing current issues inelementary and secondary education. Articles in the six issues of this volumeinclude the following: (1) January-February--"Multi-Age Classrooms: AnAge-Old Grouping Method Is Still Evolving" (Walser), "Teachers Wanted:Schools Look for Creative Solutions to Upcoming Teacher Shortage" (Walters),"Schools Should be Safe, But Are They?" (Posner); (2) March-April--"Teachersin the Driver's Seat" (Lewis), "Time and Learning" (Sadowski), "Small SchoolsWork Best for Disadvantaged Students" (Farber); (3) May-June--"The BilingualEducation Debate" (Walters), "Bilingual Education and California's 'Englishfor the Children' Initiative," "Paying Attention to ADHD" (Lynn); (4)

July-August--"Working Teenagers: Do After-School Jobs Hurt?" (Kelly),"Full-Service Schools Respond to Families' Needs" (Farber), "DiscussingStudent Work Gives Teachers New Perspective"; (5) September-October--"FromSputnik to TIMSS: Reforms in Science Education Make Headway Despite Setbacks"(Freundlich), "Latino Achievement Reexamined" (Walters); and (6)November-December--"Learning To Listen May Help Children Learn To Read"(Walser), "Programs Fostering 'Emotional Intelligence' Show Promise"(Sadowski), "Building Collaborative Relationships: Educational Research inSchools" (Rubinstein-Avila and Suarez-Orozco), "A Parent's Influence isPeerless" (Kagan). Regular features include editorial statements andsummaries of recent educational research. (KB)

see

Volume XIV, Number 1

1

U.S. DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement

EDUCATIONAL RESOURCES INFORMATIONCENTER (ERIC)

Sft This document has been reproduced asreceived from the person or organizationoriginating it

Minor changes have been made toimprove reproduction quality

Points of view or opinions stated in thisdocument do not necessarily representofficial OERI position or policy

Published by the Harvard Graduate School of Education January/February 1998

Multi-age Classrooms: An Age-Old GroupingMethod Is Still EvolvingDespite mounting challenges, the multi-age classroom continues to bean attractive option for educators

BY NANCY WALSER

Seven years ago, when ConnieChene took over as principal ofthe Puesta Del Sol ElementarySchool in Rio Rancho, NM, she

issued this challenge to her teachers: Ifthey had any ideas about how to dothings differently to benefit kids, allthey had to do was talk to her.

Located just outside Albuquerque inone of the state's fastest growing cities,Puesta Del Sol had a not-so-progressiveclassroom arrangement. All specialeducation students were taught out-side the school in portable buildings;all regular education students were

INSIDE

Classroom GroupingTerms / 2

Teacher ShortagesAhead / 4

Are Schools SafeEnough? / 6

School SafetyResources / 8

taught inside the main building. Chenewas immediately besieged by proposalsfrom regular and special educationteachers who wanted to combine theirstudents. Teachers knocked down wallsbetween rooms, more proposals camein, and Chene now presides over asmorgasbord of classrooms: bothsingle grades andmixed ages in bothregular and inclu-sion classrooms, in-cluding one inclu-sion class that spanskindergartenthrough the 3rdgrade.

"Teachers began

PERMISSION TO REPRODUCE ANDDISSEMINATE THIS MATERIAL HAS

BEEN GRANTED BY

Ketklkx-cx.2s-DesctiTO THE EDUCATIONAL RESOURCES

INFORMATION CENTER (ERIC)

1

only Haitian-Creole transitional bilin-gual education program, and with bi-lingual Haitians now making up morethan 30 percent of the school popula-tion, principal Len Solo is reassessingtheir system of combining two gradesper classrooms in 1st through 8thgrades. Especially in the higher grades,

he contends, thereare simply too many

"Society is multi-age,families are multi-age,and we wanted theclassrooms to reflectreal life."

to see the power ofkids with different abilities and differ-ent points of view working together inthe classroom," says Chene. "They be-gan to buy into the idea that society ismulti-age, families are multi-age, andwe wanted the classrooms to reflectreal life."

On the other side of the country, atthe Graham and Parks AlternativeSchool in Cambridge, MA, studentshave been combined in multi-age class-rooms for 25 years. However, Grahamand Parks is also home to the district's

ability levels in eachroom for teachers tohandle.

"You can have aneight- or nine-yearspread in terms oflevels," Solo says."I've got a staff thatkills themselves to

meet kids' needs, and it's beginning toget really difficult to do that."

A Long HistorySince the days of the one-room

schoolhouse, multi-age grouping inthis country has been buffeted bychanging times and shifting priorities.largely abandoned for single grades be-ginning in the mid-19th century, thepractice was revived in the 1960s and1970s with the growing interest in de-velopmentally based education. Today

EDITORIAL DIRECTOR: Kelly Graves-Desai. CONTRIBUTING EDITORS: Susan Eaton and Nancy Walser. PRODUCTION EDITOR: Dody Riggs. EDITORIAL BOARD,HARVARD GRADUATE SCHOOL OF EDUCATION: Mildred Blackman, Director, The Principals' Center; Sally Dias, Superintendent, Watertown Public Schools, Watertown,MA; Jay R Heubert, Assistant Professor; Harold Howe II, Senior Lecturer Emeritus; Susan Moore Johnson, Professor and Academic Dean; Robert Kegan, Senior Lecturer;Jerome T. Murphy, Professor and Dean; Gary A. Orfield, Professor; Robert S. Peterkin, Senior Lecturer; John Ritchie, Superintendent/Principal, LincolnSudburyRegional High School, Sudbury, MA; Judith D. Singer, Professor; Jay Sugarman, Teacher, Runkle School, Brookline, MA; Dennie Palmer Wolf, Lecturer on Education.NATIONAL ADVISORY BOARD: John Brademas, President Emeritus, New YorkUniversity; Constance E. Clayton, former Superintendent, School District of Philadelphia;Alonzo A. Crim, Professor of Education, Spelman College; Linda DarlingHammond, Professor, Teachers College, Columbia University; Andrew Heiskelt ChairmanEmeritus, New York Public Library; Marya Levenson, Superintendent, North Colonie Central Schools, NY; Deborah Meier, Founder, Central Park East Secondary School,NY; John Merrow, President, The Merrow Report; Arthur J. Rosenthal, Publishing Consultant. --1

BEST COPY AVALABLE

Classroom Grouping TermsLooping: Students stay with a teacher(s) for more than oneyear. A multi-age classroom is considered one form oflooping.

Multi-age (or mixed-age): Students in two or three gradelevels are mixed in one classroom based on the philosophythat this form of grouping improves learning and attitudestoward school. Multi-age grouping is practiced more often inelemenary schools than in secondary schools.

Multi-grade (also combination grades or split grades):Students from one or more grade levels are grouped due tolow enrollments or uneven class sizes. Students in these dif-ferent grade levels are usually taught separately although theyare in the same classroom.

Multi-year: An umbrella term for any classroom where stu-dents stay with the same teacher(s) for more than one year.Includes multi-age classrooms and looping.

Non-graded (or ungraded): Students of different ages aremixed based on the mixed-age philosophy that emphasizescontinuous progress by individuals rather than grade-levelexpectations. Since state education departments require re-porting by grade level, however, this form of mixed-age class-room is more rare.

Single-grade: Students of roughly the same age are assignedto the same grade for one year. This is the most commonform of grouping.

most multi-age classrooms mix two, ormore rarely three, grade levels contain-ing a minimum of four chronologicalages, according to Jim Grant, executivedirector of the Society for Developmen-tal Education of Peterborough, NH.

Supporters say the practice of mixingdifferent ages in the same classroomstill holds much promise, arguing thatit improves learning by emphasizingproject-based curricula, continuousprogress (as opposed to an annualpass-fail system), cooperation, and thesharing of knowledge. In theory, teach-ers in multi-age classrooms focus on in-dividuals rather than on grade-level ex-pectations. They also have more timeto address individual needs becausechildren spend more than one year intheir class.

Detractors focus on the difficulties ofmanaging multi-age classrooms andsome practitioners report a new threatto the viability of multi-age education:the emergence of tests tied to grade-specific curriculum frameworks, whichadds to the difficulty of teaching morethan one grade level at a time.

Types of GroupingMulti-age classrooms have been used

in a variety of ways for a variety of rea-sons. One type, multi-grade class-rooms, are forced upon teachers as away to manage low enrollments andshrinking budgets (see box). Becauseof this variety, simply counting thenumber of districts with mixed-age

classrooms has proved illusive andmade it difficult to judge their effective-ness.

When comparing achievement andother kinds of success in multi-age andsingle-grade classrooms, research is de-cidedly mixed. In a study published in1995, Simon Veenman of the Universityof Nijmegen of Norway examined 56studies, including 33 in the U.S., whichcompared standardized test results forsingle-grade classrooms with multi-

The emergence of teststied to grade-specificcurricula adds to thedffficulty of teachingmulti-age classes.

grade and multi-age classrooms. Whilehe concluded that multi-age class-rooms "appear to be generally equiva-lent" to single-graded in terms ofachievement, he found that tests meas-uring self-concept and attitudes towardschool registered "a small positive ef-fect for students in multi-age classes."Veenman excluded studies of non-graded classroomsa more deliberateand rarer form of multi-age classroomthat rejects all things labeled by gradesince non-grading is "a philosophythat permeates the entire school or-ganization and program." In an attempt

to look at only the effects of differentgrouping methods, he excluded stud-ies in which teachers had received train-ing in multi-grades or multi-ages.

In a 1992 study comparing non-graded with single-grade classrooms,however, Temple University professorBarbara Nelson Pavan reviewed 64studies in the U.S. and Canada andfound that the majority (58 percent) ofnon-graded classrooms performed bet-ter on achievement tests, as well as ontests for mental health and school atti-tude (52 percent). All seven studies thatcompared students who spent their en-tire elementary years in a non-gradedschool with single-graded counterpartsfound "superior performance by non-graded students." According to Pavan'sdefinition, a non-graded school "doesnot use grade-level designations forstudents or classes" and progress is re-ported "in terms of tasks completedand the manner of learning, not bygrades or rating systems."

Given the range of types of multi-ageclassrooms that exist, according tothese studies, duplicating these posi-tive results is a tricky business. In Ken-tucky, confusion over how to set upnon-graded primary classrooms re-sulted in changes to the state's six-year-old education reform act mandating"multi-age/multi-ability" classrooms.Now school-based councils decide howprimary-grade classrooms will be struc-tured within general goals outlined bylaw. At least some of Kentucky's ele-

THE HARVARD EDUCATION LETTER (ISSN 8755-3716) is published bimonthly by the Harvard Graduate School of Education, Appian Way, Cambridge, MA 02138-3752. Second-class postage paidat Boston, MA, and additional mailing office. POSTMASTER: Send address change(s) to The Harvard Education Letter, 6 Appian Way, Cambridge, MA 021383752.

Signed articles in THE HARVARD EDUCATION LETTER represent the views of the authors. Address editorial correspondence to Editors, The Harvard Education Letter, Gutman Library, 6AppianWay, Cambridge, MA 02138; phone 617-495-3432; fax 617-4963384; email address: edletter@hugsel.harvard.edu; web site: http://hugsel.harvard.edu/hepg/hel.html

© 1998 by the President and Fellows of Harvard College. Published as a nonprofit service. All rights reserved. Special written permission is required to reproduce in any manner, in whole or inpart, the material herein contained. Call 617-495-3432 for reprint permission information.

How to subscribe: Send $32 for individuals, $39 for institutions ($40 for Canada/Mexico, $42 other foreign, in U.S. funds only) to The Harvard Education Letter, 6 Appian Way, Cambridge MA02138; or call us at 617-495-3432 in Massachusetts or 800-513-0763 outside Massachusetts between 8:30 a.m. and 5:00 p.m. EST Monday-Friday. Subscription prices subject to changewithoutnotice. Single copies, $5.00. Back issues and bulk subscriptions available at special reduced rates; call 800 - 513-0763.

3 BEST COPY AVAiLABLi',

mentary schools now use multi-agegrouping for only a portion of the day,with basic skills and drills often re-served for single-age grouping, accord-ing to Pam Williams, a consultant forthe Kentucky Department of Educa-tion.

Emerging ConsensusAfter more than 30 years of dedicated

study, some areas of consensus haveemerged about mixing ages withinclassrooms. No one argues the fact thatmulti-age classrooms are harder toteach and require more preparationand training. And multi-grade class-roomsthe kind forced on teachers forbudgetary purposesare particularlybeset by "common problems and con-cerns," according to Veenman. Theseinclude "lack of time for teaching therequired content, a greater work load,lack of time for individual attention andremediation, lack of adequate class-room management skills, lack of ade-quate preparation during teacher train-ing, inadequate materials, and parentalconcerns about the academic achieve-ment of their children."

Conversely, schools that use multi-age classrooms successfully are markedby several common traits, according toBruce Miller, a researcher with theNorthwest Regional Educational Labo-ratory in Portland, OR, who has writtenextensively about multi-age educationin rural America. In a 1996 review of fiveschools in the Northwest that have usedmulti-age classrooms for more thanfour years, Miller found these common-alties: dedicated teachers, supportiveprincipals and parents, and solidarityand teamwork among the staff. He alsoidentified five requirements for imple-menting multi-age classrooms: reviewthe research before beginning, don'tsettle on a single model, avoid "bottomup or top-down" mandates, recognizethat a major conceptual change is re-quired in terms of attitudes towardteaching and children, and get pre-pared for "evolving long-term change"through "strategic, incremental" steps."Too many educators are implement-ing multi-age classrooms and schoolswith insufficient forethought, plan-ning, and participation of key stake-holders," he concludes.

Threat of TestsBut even supporters say the best

multi-age classrooms are threatened bythe trend toward grade-specific cur-

riculum requirements and tests. "Multi-age is a philosophy that is truly wonder-ful for learning, and in a perfect worldwhere you do not have curricular bar-riers, it is doable," says Char Forsten,who leads seminars around the countryfor the Society for Developmental Edu-cation. Forsten, who taught mixed agesin New Hampshire for 18 years, says shelearned these difficulties first-hand."The minute you were asked to teachthem as 4th and 5th graders instead ofthe blend, you were going against themulti-age philosophy to address themas one group of learners."

Other teachers, however, have got-ten around these barriers by outlining

No one disputesthe fact that multi-ageclassrooms are harderto teach and requiremore preparation andtraining.

district expectations for all the gradelevels in their classroom and concen-trating on the common areas; by teach-ing one curriculum one year and an-other the next; or by using sufficientlybroad themes to cover all the bases.Loopingwhich groups a single ormixed-age class with the same teacherfor more than one yearcan also helpbring teachers up to speed on combin-ing curriculums, according to Forsten."If I'd been a 3rd-grade teacher for 10years before I took on a 3/4 class, I'dtake the 3rd graders up to 4th grade soI would know what the 4th-grade cur-riculum is like before I taught bothgrades together in the same class-room," she says

It's exactly this kind of flexibility thatpractitioners point to as one of themost useful aspects of multi-aging: as atool that can be used by teachers whowant to use it and by schools that wantto offer it to interested parents.

In Rio Rancho, principal ConnieChene has not formally compared testresults of children before and after theintroduction of inclusion and multi-ageclassrooms. But teachers have noticeda dramatic increase in writing skills andimprovement on tests in areas theyhave identified as important. In a 1997survey of all 200 parents whose chil-

dren were in multi-age classrooms, allbut two said they wanted their childrento stay in the classrooms with mixedages. Results like these prove the con-cept is working, says Chene. "There aremany parents who still don't like it, andI make sure I have enough single-gradeclassrooms to accommodate them,"she says. But even these classes havebanded together to do mixed-age activi-ties once a weekan arrangement thatalso gives one teacher some extra plan-ning time each week, she says. "There'sno one way of doing things aroundhere," she emphasizes.

At Graham and Parks, principal LenSolo says he will be asking his schoolcommunity to look into looping by sin-gle grades as a substitute for some or allmulti-age classrooms to reap the bene-fits of having children stay with oneteacher for more than one year, whileat the same time reducing teacher workload and the need for tutoring afterschool and at home. "The measurablebenefit [of multi-aging] comes fromhaving a child for two years," he says."The kids might have a good first year,but they have a much better secondyear; the achievement goes up by morethan a year."

Despite the changes he is contem-plating at his school, Solo still considershimself a staunch advocate of multi-ageclassrooms. "I just see so many bene-fits," he says. "They really contribute tothe sense of community that we try tobuild here and we'll use them as longas we can."

For Further InformationM. A. Bozzone. "Straight Talk From Multi-age Class-rooms: Why Teachers Favor Nongraded Classes andHow They Make Them Work." Instructor 104, no. 6(March 1995): 64-70.P Chase and J. Doan. Full Circle: A New Look atMultiage Education. Portsmouth, NH: Heinemann,1994.B. A. Miller. "What Works in Multiage Instruction,"Education Digest 61, no. 9 (May 1996): 4-8.B. N. Pavan. "The Benefits of Nongraded Schools."Educational Leadership 50, no. 2 (October 1992):22-25.

Society for Developmental Education. The MultiageResource Book, 1993. PO Box 577, Peterborough,NH 03458; 800-462-1478.S. Veenman. "Cognitive and Noncognitive Effects ofMultigrade and Multi-Age Classes: A Best EvidenceSynthesis." Review of Educational Research 65, no.4 (Winter 1995): 319-381.

Nancy Walser is a freelance Journalist and theauthor of Parent's Guide to Cambridge Schools. Sbelives in Cambridge, MA.

The Harvard Education Letter, January/February 1998 3

Teachers Wanted: Schools Look for CreativeSolutions to Upcoming Teacher ShortageBY LAUREL SHAPER WALTERS

perennial predictions of im-pending teacher shortageshave made researchers looklike the boy who cried wolf.

Despite projections that widespreadteacher shortages would hit in the1990s, most school districts face tightmarkets in only certain subject areas.Widespread shortages during this dec-ade have so far been limited to special-ized fields such as science, special edu-cation, math, and bilingual education,according to a 1995-96 University ofMichigan survey of education adminis-trators in 294 public school districts.Nationwide shortages, however, ap-pear to have only been delayed: inevi-table retirements and rising enroll-ments from children of baby-boomersare forcing districts to come up with avariety of solutions to make hiring eas-ier in the next decade.

While some urban and rural schooladministrators are struggling to fill a va-riety of positions, rising teacher salariesand the economic anxieties caused bya recession have kept many retirement-age teachers in the classroom longerthan expected, which has delayedacross-the-board teacher shortages.

Recent political and demographicchanges are already creating acuteshortages for many types of teachers instates including California, Texas, Ne-vada, and Florida. Due to an influx ofimmigrants and booming birth rates,districts in these states are finding itmore and more difficult to fill positionswith qualified candidates, according tostudies by the American Association forEmployment in Education in Evanston,IL. In California, a 1996 state law man-dating class-size reductions will requirethe state to hire 15,000 to 17,000 teach-ers annually for the next several years,for a total of 260,000 new hires in thenext decade, according to the Califor-nia Commission on Teacher Creden-tialing. If more states take California'slead and pass laws reducing class size,it will make the national teacher short-age even more serious, says DavidHaselkorn, president of RecruitingNew Teachers, a nonprofit group inBelmont, MA. "That's just one of the

many unprojectable issues of teachersupply and demand," he adds.

Serious Shortagesto Come

As the demographic double wham-my of accelerating retirements and ex-panding enrollments hits the nation inthe next decade, demographers sayschool administrators better bracethemselves for serious nationwideteacher shortages.

For the past two years, public-schoolenrollment has exceeded the baby-boom records. Demographers predictthe peak will come by 2006 when theyexpect U.S. schools to enroll more than54 million children, a dramatic increasesince the mid-eighties when enroll-ments were around 38 million. To keepup, we'll need two million new elemen-tary and secondary teachers in the nextdecade, according to last fall's back-to-school report on demographic trendsfrom the U.S. Department of Educa-tion. Or, as Patricia A. Ashton, a profes-sor of educational psychology at theUniversity of Florida, wrote in 1996:"Virtually the entire teaching force willretire within about 15 years."

What we're talkingabout is a shortage ofpeople who are well-educated and have theskills to teach.

Early signs of a tightening teacher la-bor market are showing up in the largenumber of educators who are teachingoutside their fields of study. More thanone-third of teachers in English, foreignlanguages, mathematics, or social stud-ies neither majored nor minored intheir main assignment field as under-graduates, according to a 1993-94 studyof public-school staff conducted by theU.S. Department of Education.

While the impending shortage pre-sents huge logistical difficulties, espe-cially for larger urban districts that have

the hardest time attracting qualifiedcandidates, some experts view the situ-ation as a golden opportunity to diver-sify and upgrade the quality of the na-tion's teaching force. "You have to becareful when you use the word short-age," says Richard Murnane, a Harvardeconomist and author of Who WillTeach, a 1991 book examining the labormarket for teachers. "It's not a questionof bodies; it's a question of quality.What we're talking about is a shortageof people who are well-educated andhave the skills to teach."

In fact, wealthy suburban districtsmay never feel the crunch. They alwayshave a surplus of applicants for mostpositions. Schools serving higher con-centrations of poor students report thegreatest difficulty hiring teachers infields where there is a shortage. Sinceschool administrators cannot simplyleave classrooms unattended, theyoften make concessions in filling vacan-cies. In some cases, administrators hireunderqualified teachers who thenwork toward full teaching credentials.In other cases, educators trained in onefield are assigned to teach another sub-ject. This poses the most troublesomequestions for policymakers, since ashortage of qualified teachers meansthe nation's most needy students areput at even more risk of sub-standardteaching.

Responding to theChallenge

Such problems are gaining more at-tention than ever, however. This pastJuly, President Clinton proposed a $350million program to train about 35,000new teachers who agree to work for atleast three years in poor urban and ru-ral schools. The Clinton administrationand three members of Congress haveproposed teacher-training and recruit-ment bills. Several states, includingNorth Carolina and Arkansas, recentlypassed major legislation revampingand increasing support for teacher-education programs.

"I've been talking about the issue ofimproving teaching for 20 years, andthis is really the first time I've seen this

4 The Harvard Education Letter, January/February 1998

kind of activity," says Linda Darling-Hammond, a professor at Teachers Col-lege, Columbia University, and a lead-ing expert on teacher preparation.

Foundation funding for teacher re-cruitment and retention programs isalso at an all-time high. Since 1989, theDeWitt Wallace-Reader's Digest Fundhas invested $40 million in model train-ing programs to increase and diversifythe supply of qualified educators.Through the foundation's Pathway toTeaching Careers Program, 42 collegesand universities are using grant moneyfor scholarships that help returnedPeace Corps volunteers, career chang-ers, and teachers' aides to become cer-tified teachers. The Ford Foundationhas spent another $25 million provid-ing support services like child care andtutoring for Native American, AfricanAmerican, and Latino teacher trainees.

A Question of QualityQuality over quantity of teachers was

the focus of the most recent inde-pendent examination into the state ofthe nation's teaching force by the Na-tional Commission on Teaching andAmerica's Future. In its influential 1996report, "What Matters Most: Teachingfor America's Future," the bipartisangroup of 26 public officials, educators,and business leaders argued for theneed to create a high-quality system ofteacher development and account-ability while challenging districts tostaff every classroom with a fully quali-fied teacher by 2006the same yearthat projected enrollments reach theirpeak nationwide.

"What we have are major problemsof distribution and quality," says Dar-ling-Hammond, who serves as the com-mission's executive director. Too manypeople are preparing to be elementaryand English teachers while not enoughare training in such areas as math, sci-ence, and special education, she ar-gues. And while the largest numbers ofteachers are being trained in the Mid-west at the big land-grant universities,more teaching jobs are on the coasts, inCalifornia, Texas, New York, and Flor-ida. Moving states toward a commonlicensing system would more evenlydistribute teachers, says Darling-Ham-mond. As it stands today, teachers whoare licensed in one state often cannotteach in another.

Creative SolutionsAs the labor market for teachers

heats up, personnel directors are think-

ing more creatively about how to buildtheir future teaching ranks. Alternativecertification programs, initially consid-ered a panacea for predicted shortages,are receiving more scrutiny at the sametime interest in other kinds of trainingprograms is growing.

Since New Jersey created the first al-ternative certification program forteachers in 1983, 41 more states haveset up alternative or emergency licens-ing routes. An estimated 50,000 peoplehave been certified through these pro-grams since 1990, according to the Na-tional Center for Education Informa-tion, a private research group inWashington, DC. Ranging from sixweeks to two years or more, alternativecertification programs are designed tosupplement the existing pool of teach-ers by bringing in older, more experi-enced candidates who have at least abachelor's degree and may be profes-sionals switching from another field.

Too many people arepreparing to be Englishteachers, while notenough are training inmath, science, andspecial education.

Yet the early evidence suggests thatfew alternatively certified teachers arebeing attracted from other professions.More than 75 percent of people enter-ing teaching by an alternative pathcame either straight from college orfrom another job in the education field,says Jianpeng Shen, an assistant profes-sor at Western Michigan University inKalamazoo, who surveyed 15,000teachers for a 1996 study on alternativecertification. Research also suggeststhat teachers trained through alterna-tive certification programs may bemuch more likely to leave the profes-sion than those trained in traditionalprograms. Some studies also suggestthat student achievement suffers underalternatively certified teachers. Re-search in Dallas, where many teacherswere hired with alternative certifica-tion, shows these teachers' perform-ance to be much more uneven than thatof traditionally trained teachers. The ef-fects of this showed up most stronglyon students' achievement in the lan-guage arts, where student performancegains were significantly lower.

Murnane, however, argues that tradi-tional education programs are "dead-ening" and that alternative certificationis necessary to attract college graduatesto a field that does not tend to attractthe best and brightest. "The challengeis to attract talented college graduates,give them in-depth pre-service educa-tion and significant support duringtheir first years on the job," he says. Healso favors performance-based licens-ing where teachers would have to dem-onstrate their competence.

Rather than limiting themselves to al-ternative certification programs, manyschool administrators are creating part-nerships with local colleges and estab-lishing mid-career programs within tra-ditional schools of education. Theseprograms target professionals whowant to become teachers. A new surveyby Recruiting New Teachers of Bel-mont, MA, found more than 300 mid-career transition programs across thecountry; 46 percent were created since1990.

Programs that help paraprofession-alsor teaching aidesbecome fullylicensed teachers also are gaining mo-mentum. More than 9,000 people areenrolled in 150 such programs nation-wide. With a 77 percent minority en-rollment, these programs are helpingbring more African American and Lat-ino teachers into U.S. classrooms. Earlyresearch suggests that teachers' aideswho become fully certified throughthese programs are performing wellabove the average, says Beatriz Clewell,a researcher at the Urban Institute whois conducting a study for the DeWittWallace-Reader's Digest Fund. Thefund-supported programs also retain90 percent of their students while tra-ditional teacher-education programslose almost a third by graduation.

Other districts are looking to recruittalent from within their own communi-ties. In Rochester, NY nearly 100 highschool students participate in a two-

COMING SOON

Why Is Small Better WhenIt Comes to School Size?

Techniques for AssessingStudent Work

Does More Time MeanMore Learning?

The Harvard Education Letter, January/February 1998 5

year program that exposes them toteaching, counsels students on how tomove into a career in education, andencourages teaching careers at everyopportunity. The school district alsoworks with local colleges to help stu-dents receive scholarships. Minnea-polis schools, in addition to trainingparaprofessionals, have a "residencyprogram" that puts several studentteachers from local colleges into highschools for a year at 80 percent of a fullteacher's salary.

But for personnel directors alreadyfeeling the hiring crunch, aggressive re-cruiting goes on even as they pursuethese long-term options. In Los Ange-les, the district expects an influx of50,000 new students in the next fiveyears and already hires 4,500 teachersa year. Long-term needs are being ad-dressed through programs that helpparaprofessionals work toward teachercertification and in high schools that fo-cus on early teacher-preparation op-portunities. Short-term approaches in-clude hiring college graduates to workin shortage areas for one year throughthe Teach for America program. An al-ternative certification program pro-

vides a two-year pipeline for candidatesinterested in working in shortage fieldssuch as special education. The district'smost controversial strategy is paying abonus of up to $5,000 a year for bilin-gual teachers, who are sorely neededand difficult to find.

"You have to take a multifaceted ap-proach to recruitment," says MichaelAcosta, the administrator of employ-ment in the Los Angeles Unified SchoolDistrict. "It's like playing golf. You haveto use a variety of clubs to get to thepoint where you need to be. We've util-ized every club in the bag."

For Further InformationMichael Acosta, Administrator of Certificated Em-ployment Operation. Los Angeles Unified SchoolDistrict, 415 North Grand Ave., Los Angeles, CA90051; 213-625-6334.P T. Ashton. "Improving the Preparation of Teach-ers." Educational Researcher 25, no. 9 (1996): 21-22, 35.J. Shen. "Has the Alternative Certification Policy Ma-terialized Its Promise? A Comparison Between Tra-ditionally and Alternatively Certified Teachers inPublic Schools." Educational Evaluation and Pol-icy Analysis 19, no. 3 (Fall 1997): 276-283.B. C. Clewell. Recruiting, Preparing, and RetainingTeachers for America's Schools. Progress Report:Pathways to Teaching Careers (August 1997). Dewitt

Wallace-Reader's Digest Fund, Two Park Avenue,23rd Floor, New York, NY 10016; 212-251-9700.

L. Darling-Hammond. "Teaching and Knowledge:Policy Issues Posed by Alternate Certification forTeachers." Peabody Journal of Education 67, no. 3(1990): 123-154.S. M. Johnson. Teachers at Work: Achieving Successin Our Schools. New York: Basic Books, 1990.Augustin Melendez, Supervising Director of HumanResources, Rochester Public Schools, 131 WestBroad St., Rochester, NY 14614; 716-262-8522.R. J. Murnane et al. Who Will Teach? Policies ThatMatter. Cambridge, MA: Harvard University Press,1991.

National Commission on Teaching and America'sFuture. What Matters Most: Teaching for America'sFuture, September 1996. PO Box 5239, Wood-bridge, VA 22191-5239; 212-678-3015.

Recruiting New Teachers. Take ThlsJob and Love It:The Mid-Career Guide for Becoming a Teacher 385Concord Ave., Suite 103, Belmont, MA 02178; 617-489 -6000.

Katrina Reed, Director of Labor Relations, Minnea-polis Public Schools, 807 NE Broadway, Minneapo-lis, MN 55413; 612-627-2011.U.S. Department of Education, National Center forEducation Statistics. America's Teachers: Profile ofa Profession, 1993-94 (NCES 97-460). Washington,DC, 1997; 202-219-1744.

Laurel Shaper Walters is an education writer livingin St. Louis, MO.

Schools Should Be Safe, But Are They?Research tells us a lot about common school injuries, but schools lacksystems for preventing them

BY MARC POSNER

0 n June 18, 1993, CharlesRichey was coaching hisson's Little League team inRoss, a small town near

Pittsburgh. His attention was divertedby a toppling slide at the adjacent ele-mentary school. Running to the play-ground, Richey was horrified to findthat the slide had landed on his 9-year-old daughter, Jillian. She died of inter-nal bleeding an hour-and-a-half later.

Investigation revealed that the slide'sleg supports had rusted through. Ateacher had warned the principal and aschool board member about a brokensupport on the slide two months beforeit fell, but no repairs were made.

Jillian Richey's death was not an iso-lated incident. Every year, children are

hospitalised, disabled, and even killedby injuries occurring at school or dur-ing school-sponsored activities. Twomillion children are treated by schoolnurses, hospitals, or the family physi-cian every year for such injuries; morethan 700,000 children require emer-gency room treatment. Research hastaught us much about the most com-mon school injuries. However, mostschool districts have not taken advan-tage of this research in ways that couldsignificantly reduce the number of in-juries in schools.

School DangersGovernment researchers and aca-

demics studying injury prevention haveidentified four areas in which students

are most likely to be seriously injured,as well as some concrete steps that canhelp prevent such injuries. Elementaryschool children, for example, are atgreatest danger on the playground.Falls from slides and climbing equip-ment account for over 75 percent ofserious and fatal playground injuries.Children colliding with equipment, oreach other, also cause large numbers ofinjuries. Collisions with moving swingsare especially dangerous. Children arealso in danger of being strangled if theirheads become lodged between the barsor stairs of climbers or the openings of"adventure" equipment, or whensweatshirt drawstrings become tangledin slides or merry-go-rounds.

Elementary school children also face

6 The Harvard Education Letter, January/February 1998

dangers on the way to and from school.The contemporary school bus protectsits passengers so well that neither theNational Transportation Safety Boardor the National Research Council rec-ommends seat belts in school buses. Infact, the most dangerous part of aschool bus journey is the few minutesbefore and after the ride, when stu-dents are at risk of being struck by amotor vehicle. Sixty percent of the stu-dents killed in such incidents are struckby the bus. Most of these children arebetween the ages of five and seven;their small size makes it difficult fordrivers to see them.

For older students, sports and physi-cal education classes pose the greatestinjury risks. Over 2.3 million junior andsenior high school athletes are injuredannually, including 35,000 who arehospitalized and 600,000 who aretreated by a doctor. The most danger-ous athletic activities are gymnasticsand contact sports, but no sport is com-pletely risk-free. Students have beenpermanently disabled after sliding dur-ing baseball games. Cheer leading hasbecome a competitive sport in whichparticipants have been paralyzed afterlanding on hardwood gymnasiumfloors during stunts. Students are alsoinjured during athletic practice, espe-cially when using areas not designedfor athletics, such as the common prac-tice of running laps in hallways duringinclement weather. Many serious ath-letic injuries are actually reinjuries oc-curring when students resume physicalactivity before an injury has completelyhealed. The most dramatic of such re-injuries are cases of "second impactsyndrome," in which students with un-healed concussions suffer brain dam-age or die as a result of a blow to thehead.

Off the playing fields and sports are-nas, older students are also at risk inscience labs and vocational shops. Fin-gers and hands have been amputatedby power saws and wood jointers. Stu-dents have been burned by explosionsin chemistry laboratories and poisoned

by toxic materials in biology and artclass.

Prevention MethodsThanks to researchers and the efforts

of a number of national organizations,information on simple methods forpreventing injuries is now widely avail-able (see box, p. 8). Resurfacing a play-ground with the proper amount ofwood mulch, rather than dirt, for exam-ple, can make the difference between abump on the head and a concussion.Placing equipment far enough fromother equipment or paths can preventcollisions. Inspecting playgrounds sev-eral times a year for wear and tear can

Injuries are just as muchan indicator of health asvaccination status, casesof the flu, or outbreaks offoodborne illness.

also lessen chances of injury. Technicalinnovations for buses like larger rear-view mirrors, crossing arms that swingout from the front bumper, and elec-tronic motion detectors can help busdrivers cope with blind spots that makeit hard to see children and other pedes-trians. A bus monitor can act as a sec-ond pair of eyes for the driver and en-force discipline when the bus is inmotion. For safer loading, school busloading zones can be separated fromother traffic, pedestrian, and bicycle en-trances to the campus. Painting an out-line of a bus at bus stops combined witha short safety lesson can mean saferboarding behavior by elementaryschool children.

Advice is available on preventingsports injuries. Pre-participation physi-cals, pre-activity warm-up and stretch-ing, use of appropriate protective gear,strict enforcement of rules, and rigor-ous adherence of return-to-partici-pation guidelines are recommended

procedures. Safety in science and voca-tional programs includes instruction insafe practices at the beginning of eachsemester or whenever a new activity ortechnology is introduced. Other rec-ommendations include periodic safetyaudits of labs, art rooms, and voc-edworkshops, with special attention paidto protective guards on power equip-ment and the presence and mainte-nance of protective and emergencyequipment, first aid kits, and fire extin-guishers.

A System for SafetyMost of what we know about school

injuries comes from research and fromliability suits. Some districts have begunassessing their own safety records tobecome more proactive in preventinginjuries and reducing their own legalrisks. In Washington state, the SchoolInjury Surveillance Project has revealedthat 85,000, or a full 17 percent of stu-dents in kindergarten through 5thgrade, are injured during school hourseach year. But those numbers could behigher, since the most diligent schooldistricts often reported much higher in-jury rates, according to Richard Ellis,who directs the project for the Washing-ton State Department of Health. Ellisbelieves local districts should be doinga better job tracking school injuries na-tionwide. "Injuries are just as much anindicator of health as vaccinationstatus, cases of the flu, or outbreaks offoodborne illness," he says. 'All of theseexcept injuriesare tracked by mostschools." In addition to Ellis' project,school injury data programs have be-gun in Arizona, Utah, and Virginia.

There is resistance, however, totracking school injuries more thor-oughly. The two most common objec-tions, according to researchers, are bur-dening staff with more paperwork andcreating injury records that could beused against the schools in lawsuits.

At the same time, some local and na-tional groups are emphasizing local re-cord-keeping as a first and necessarystep toward a comprehensive approach

STATEMENT OF OWNERSHIP, MANAGEMENT, AND CIRCULATION

(Required by 39 U.S.C. 3685) 1. Title of publication: The Harvard Education Letter. 2. Publication no. 8755-3716. 3. Date of filing: December 16, 1997.4. Frequency of issue: bimonthly. 5. No. ofissues published annually: 6. 6. Annual subscription price: $32.00. 7-8. Mailing address of known office of publication/publisher: Gutman Library, Room 349, 6 Appian Way, Cambridge, MA02138. 3752.9. Harvard Graduate School of Education, Longfellow Hall, 13 Appian Way, Cambridge, MA 02138-3752; Editor: Kelly Graves-Desai, Harvard Graduate School of Education, Gutmanlibrary, Room 349, 6 Appian Way, Cambridge, MA02138-3752. 10. Owner: President and Fellows of Harvard College, Neil L Rudenstine, President, Massachusetts Hall, Cambridge, MA02138-1423.11. Known bondholders, mortagagees, or other security holders owning or holding 1 percent or more of the total amount of bonds, mortgages, or other securities: none. 12. The purpose,function, and nonprofit status of this organization and the exempt status for federal income tax purposes have not changed during the preceding 12 months. 13.Publication name: The HarvardEducation Letter. 14. Issue date for circulation data below: Sept./Oct. Issue, Sept. 9, 1997. 15. Extent and nature of circulation (average no. of copies each issue during preceding 12 months/actualno. of copies of single issue published nearest to filing date): a. Total no. copies (17,658/13,000). B. Paid and/or requested circulation: 1. Sales through dealersand carriers, street vendors, andcounter sales (271/183). 2. Paid or requested mail subscriptions (10,791/8,470). c. Total paid and/or requested circulation (11,062/8,653). d. Free distribution by mail (521/521). e. Freedistribution outside the mail (350/348). E lbtal free distribution (871/869) g. Total distribution (11,933/9,522). h. Copies not distributed: 1. office use, leftovers, spoiled (5,725/3,478). 2. Returnfrom news agents (0/0). i. Total (17,658/13,000). I certify that the statements made by me above are correct and complete. /s/ Kelly Graves-Desai, Editorial Director.

BEST COPY AVAILABLE

Resources on School SafetyGeneralInjuries in the School Environment: A Resource Guide. Children'sSafety Network, Education Development Center, Inc., 55 ChapelStreet, Newton, MA 02158-1060; 617-969-7101, ext. 2207; Website:www.edc.org/HHD/csn/sc.hooinj/schoopack.hunl.

Getting to and from School SafelyWalk-Ride-Walk: Getting to School Safely. National Safety Council,1121 Spring Lake Drive, Itasca, IL 60141-3201; 800-621-7619.

Wary Walker Pedestrian Safety Curriculum. Harborview Injury Pre-vention and Research Center, Box 359960, 325 Ninth Avenue, Se-attle, WA 98104-2499; 206-521-1520.

Playground SafetyHandbook for Public Playground Safety. U.S. Consumer ProductSafety Commission, Washington, DC 20207; 800-638-2772;Website: wwwcpsc.gov /.

ASTM/CPSC Playground Audit Guide. Playworld Systems, 315Cherry Street, PO Box 505, New Berlin, PA 17855; 800-233-8404.

Sports InjuriesSports Medicine: Health Care for Young Athletes (2nd ed). Ameri-can Academy of Pediatrics Publications Department, PO Box 927,Elk Grove, IL 60009-0927; 800-433-9016.

Minimizing the Risk of Injury in High School Athletics. NationalAthletic Trainers' Association, 2952 Stemmons Freeway, Dallas TX75247; 214-637-6282.

Science LaboratoriesSafety in the Academic Chemistry Laboratories. American ChemicalSociety, PO Box 57136, Washington, DC 20037-0136; 800-ACS-9919.

Safety in the Elementary Science Classroom. National ScienceTeachers Association, 1840 Wilson Blvd., Arlington, VA 22201-3000;800-722-NSTA.

similar to that used by public health of-ficials to successfully reduce occupa-tional and traffic injuries in the past."Preventing injuries in schools is a mat-ter of looking at the patterns of injuriesthat exist in the school and then at thestrategies that have been used in otherplaces to prevent these types of inju-ries," says Elaine Frank, director of theNew Hampshire Safe Playground Pro-ject at Dartmouth College Injury Pre-vention Center. All they need to do isto apply methods already shown to pre-vent injuries in the school environ-ment."

In 1996, a School Injury Work Groupat the California Conference on Child-hood Injuries recommended that thestate's education code be amended torequire school district safety boards(including parents) to develop, imple-ment, and evaluate school safety pro-grams and violence prevention pro-grams. The American School HealthAssociation (ASHA) National Taskforceon Injury Prevention has also recom-mended that "a uniform process for re-porting injuries and health problems inthe school environment should be inplace and analyzed for the purpose ofmonitoring risk factors, trends, and pat-terns, and suggesting possible preven-tative measures."

According to Jan Ozias, a member ofthe ASHA task force, school injury pre-vention is best overseen by a school-based committee composed of staffhaving responsibility for areas whereserious injuries are most likely to occur,as well as those responsible for studenthealth and safety. Committee members

would include an administrator, theschool nurse, the district's attorney, therisk manager, and representatives fromthe science, vocational education,physical education, and sports pro-grams, as well as staff from the buildingand grounds, security, transportation,and counseling departments. Parentsand student members can also bring afresh perspective to the committee,says Ozias. She adds that another im-portant duty of the committee is to cre-ate and periodically review an emer-gency preparedness plan.

New Plan for EmergenciesSince no school is completely "injury

proof," one of the most serious mis-takes a school can make is not to havean established procedure for respond-ing to emergencies. Students have diedwhile waiting for a principal to be calledto the scene to allow a call to 911 to bemade. Emergency medical technicianshave watched students with brokenlimbs suffer while waiting for parents tobe called for permission to transportthe child to an emergency room. "Plan-ning is critical to effectively respondingto an emergency," says Elaine Brainerd,a consultant who developed the SchoolNurse Emergency Medical Services forChildren Program at the University ofConnecticut Health Center. "We'vedone fire drills for years in schools," shesays. "We also need to prepare for othertypes of emergencies."

Experts recommend that emergencyplans be expanded beyond proceduresdealing with suicide attempts or vio-lencea traditional focusto include

guidelines detailing how to proceed inthe event of any serious injury. Otherrecommendations include encourag-ing parents to sign release forms allow-ing the school to take children to theemergency room when necessary,equipping school buildings with basicfirst aid equipment and high schoolswith portable defibrillators, and train-ing teachers, coaches, and other per-sonnel in basic first aid, CPR, and theHeimlich maneuver. A list of trainedpersonnel could be available in everyroom so they could be summonedwhile waiting for professional help toarrive.

Time to ActSince school districts and parents are

accepting the notion that injuries re-sulting from violence are preventable,Jan Ozias believes it is not a great leapto apply this same preventive model tounintentional injuries. Richard Ellisconcurs. "Injury prevention should bemade part of a school culture," he says.Every time a student is seriously injuredor killed while in a school's care, theirparents, the press, and administratorsand teachers ask if there was some waythe tragedy could have been avoided.The implementation of a school injuryprevention program can help avoidhaving to face this question.

Marc Posner is a Senior Research Associate at Edu-cation Development Center, Inc., in Newton, MA. Heis currently writing a book about preventing schoolinjuries.

8 The Harvard Education Letter, January/February 1998

Volume XIV, Number 2 Published by the Harvard Graduate School of Education March/April 1998

Teachers in the Driver's SeatCollaborative assessment proves a positive way to reform schools and improve teaching

BY ANNE C. LEWIS

Wshen teachers in the Mon-aca school district nearPittsburgh were first pre-ented with new content

standards in 1995-96, "they did whatevery good teacher would doputthem in a drawer," says Kathy Dabrow-ski, who is principal of three smallneighborhood elementary schools inthe 880-student district. But thenDabrowski countered with her own re-quirement: teachers had to turn in stu-dent work along with weekly lessonplans based on the new standards.Dabrowski followed up by discussingthese plans and student work with theteachers, asking such questions as,"What are you doing here to get thiswork to a higher level?"

Monaca teachers from each gradelevel now spend half a day every month

INSIDE

Protocols for AssessingStudent Work / 2 & 3

Does More Time MeanMore Learning? / 4

Small Schools Do Make aDifference / 6

looking at student work together. Staffdevelopment experts from the Institutefor Learning at the University of Pitts-burgh's Learning Research and Devel-opment Center (LRDC) help teacherswith these discussions. Also, teacherrepresentatives from each of the gradelevels in the district meet with parentsweekly for a few hours after school tolook at how student work is measuringup to standards.

"What has happened here is tremen-dous," says Dabrowski. "If you go intoany group of teachers and ask, 'couldyou be doing yourjob better?' you willlose most of them im-mediately. But if youask them to look atstudent work andtalk about how itcould be better, thenteachers become re-ally student focused."

proof that students are learning tohigher standards; reform efforts thatnow target schools as well as districtsand that encourage teachers to shareresponsibility for student success; and,finally, the emergence of a researchbase that is giving teachers better cluesas to how to move to higher levels oflearning.

Beyond the PortfolioTeachers have always examined stu-

dent work, but almost always alone andfor the purpose of grading an individ-

ual's work. The ideaof sharing student

If you ask teachers tolook at student workand talk about it, thenthey become reallystudent focused.

Persuading teach-ers to discuss theirown students' work and compare itwith that of their colleagues' studentsis a delicate process that would havebeen a rarity only a few years ago. Yettoday there are thousands of teachersusing student work for a variety of pur-poses that go far beyond the practice ofassembling portfolios. Three thingshave only recently made this possible:a political and policy climate that wants

work, however, hasbecome more com-mon with the grow-ing use of portfoliosas an alternative orsupplement to stan-dardized and otherformal tests. A decadeago, the practice of fo-cusing on student

work was begun by a small number ofacademics as a way of getting teachersin touch with how students learn. Thearchival project of student work col-lected by the Prospect School in Ver-mont and Project Zero, which was de-veloped by Howard Gardner andothers at the Harvard Graduate Schoolof Education, were primary influences.

In 1994, the federal Title I program

EDITORIAL DIRECTOR: Kelly Graves-Desai. CONTRIBUTING EDITOR: Nancy Walser. PRODUCTION EDITOR: Dody Riggs. EDITORIAL BOARD, HARVARD GRADUATESCHOOL OF EDUCATION: Mildred Blackman, Director, The Principals' Center; Sally Dias, Superintendent, Watertown Public Schools, Watertown, MA; Jay P. Heubert,Assistant Professor; Harold Howe II, Senior Lecturer Emeritus; Susan MooreJohnson, Professor and Academic Dean; Robert Regan, Senior Lecturer; Jerome T. Murphy,Professor and Dean; Gary A. Orfield, Professor; Robert S. Peterkin, Senior Lecturer; John Ritchie, Superintendent/Principal, Lincoln-Sudbury Regional High School,Sudbury, MA; Judith D. Singer, Professor; Jay Sugarman, Teacher, Runkle School, Brookline, MA; Dennie Palmer Wolf, Lecturer on Education. NATIONAL ADVISORYBOARD: John Brademas, President Emeritus, New York University; Constance E. Clayton, former Superintendent, School District of Philadelphia; Alonzo A. Crim,Professor of Education, Spelman College; Linda Darling-Hammond, Professor, Teachers College, Columbia University; Andrew Heiskell, Chairman Emeritus, New YorkPublic Library; Marya Levenson, Superintendent, North Colonic Central Schools, NY; Deborah Meier, Founder, Central Park East Secondary School, NY; John Merrow,President, The Merrow Report; Arthur J. Rosenthal, Publishing Consultant.

0 BEST COPY AVAILAP,F.

Collaborative Assessment:Looking at Learning Through Careful Examination of Student Work

The following protocol was developed by Steve Seidel andcolleagues at Harvard Project Zero. This "Collaborative As-sessment Conference" brings teachers together to talk aboutstudent work. The protocol is based on the notion that stu-dents are often working on problems or exploring interestsbeyond the parameters of a given assignment. Project Zerois an education research organization based at HarvardGraduate School of Education.

(This process should take approximately 45 to 60 minutes.)

The purpose of this practice is to provide opportunities forteachers to examine and discuss pieces of student work in anonjudgmental, structured conversation. It is designed to fa-cilitate discussion in small groups. Through these conversa-tions, participants can learn from the various perspectivesand expertise of their colleagues, raise questions about thestudent and his/her work, see that student work is a reflectionof the child and the learning environment, and use thisopportunity to reflect on ways to improve learning en-vironments for groups of children as well as the individualchild.

1. Getting started. The group chooses a facilitator to guideparticipants through each phase of the conference. The pre-senting teacher shares copies of the selected work, withoutmaking comments about the work or the assignment.

2. Describing the work. The group describes any aspect of

the work they notice. They do not make judgments aboutthe quality of the work or their personal preferences.

3. Raising questions. The group asks questions about thechild, the assignment, the curriculum, or any other area. Thepresenting teacher takes notes but does not respond.

4. Speculating about what the student is working on. Thegroup "guesses" about what the child was working on whenhe/she created the piece. This could include ways the studentwas trying to fulfill the assignment, skills the child was tryingto master, questions the child was trying to answer, or ideashe/she was trying to express.

5. The "presenting teacher" speaks. The presentingteacher now adds her perspective on each of the previousphases of the conference. She provides her own perspectiveon the students' work and responds to any questions or is-sues raised by the group.

6. Implications for teaching and learning. Everyone is in-vited to share any thoughts that have been stimulated by theexamination of the student work. These could includethoughts about their own teaching, student learning, or waysto support a particular child in reaching his/her goals.

7. Final reflection. At this time, participants have an oppor-tunity to reflect on the process of their own thinking duringthe conference.

Adapted with permission from Harvard Project Zero.

began requiring that disadvantaged stu-dents receiving Title I services be heldto the same standards as all other stu-dents. Since then, every state has devel-oped new standards setting forth whatstudents should accomplish. In addi-tion, many districts are developingstandards and assessment systems ontheir own. No matter what type ofstandards are adopted, however, stu-dent work has become a tool for mak-ing sure those expectations get trans-lated into the classroom. Since fewteachers are actually involved in devel-oping standards, the rest "are not goingto learn this stuff by reading books ofstandards," says Katherine Nolan, aconsultant who helped write the NewStandards, a comprehensive set of per-formance standards for five subject ar-eas. These standards have so far been

adopted by 25 states and 50 districts,including Monaca. Only when teacherscome together to discuss standards andwhat high-quality work looks like canthat knowledge get out of teachers'heads and into discussions with col-leagues, Nolan says.

While the standards movement hasfueled interest in looking at studentwork, the process continues to be usedfor a variety of purposes: for scoringand holding schools responsible forstudent performance; for writing stand-ards and helping teachers understandthem; or simply for helping teachersthink about their teaching and learnmore about their students. For exam-ple, student work was used as a tool bythe LDRC and the National Center onEducation and the Economy in Wash-ington, DC, to formulate proficiency

exams based on the New Standards. InKentucky, professional developmentactivities centered on student work arehelping teachers understand the differ-ences among levels on the state's as-sessment tests (novice, apprentice,proficient, and advanced). Other re-form organizations are using studentwork in schools they consult with toupgrade teaching, including the Coali-tion of Essential Schools and the Edu-cation Trust's K-16 Compact. ProjectZero holds open, informal discussionsof student work using the "collabora-tive assessment conference" everymonth. The meetings draw 30-40teachers, administrators, and other in-terested educators from New Englandwho come to engage in thoughtful con-versations about teaching and learningthrough looking at student work.

THE HARVARD EDUCATION LETTER (ISSN 8755-3716) is published bimonthly by the Harvard Graduate School of Education, Appian Way, Cambridge, MA 02138-3752. Second-class postage paidat Boston, MA, and additional mailing office. POSTMASTER: Send address change(s) to The Harvard Education Letter, 6 Appian Way, Cambridge, MA 02138-3752.

Signed articles in THE HARVARD EDUCATION LETTER represent the views of the authors. Address editorial correspondence to Editors, The Harvard Education Letter, Gutman Library, 6AppianWay, Cambridge, MA 02138; phone 617-4953432; fax 617-496-3584; email address: edletter@hugsel.harvard.edu; web site: http://hugsel.harvard.edu/hepg/hel.html

© 1998 by the President and Fellows of Harvard College. Published as a nonprofit service. All rights reserved. Special written permission is required to reproduce in any manner, in whole or inpart, the material herein contained. Call 617-495-3432 for reprint permission information.

How to subscribe: Send $32 for individuals, $39 for institutions ($40 for Canada/Mexico, $42 other foreign, in U.S. funds only) to The Harvard Education Letter, 6 Appian Way, Cambridge MA02138; or call us at 617-495-3432 in Massachusetts or 800-513-0763 outside Massachusetts between 8:30 a.m. and 5:00 p.m. EST Monday-Friday. Subscription prices subject to change withoutnotice. Single copies, $5.00. Back issues and bulk subscriptions available at special reduced rates; call 800-513-0763.

BEST COPY AVAILABLE

Early FindingsWhile consistent links to higher

achievement are only anecdotal at thisstage, collaborative assessment of stu-dent work is showing some positive ef-fects, especially in the area of profes-sional development. Researchers whostudied the reactions of 250 teacherswho participated in New York State'seffort to develop performance assess-ments through scoring student workdecided the exercise did provide a vari-ety of learning experiences for teach-ers. "Looking closely at student work incollaboration with colleagues helpedteachers learn about standards, theirdisciplines, their students, and teach-ing," reported Beverly Falk andSuzanna Ott of Teachers College at Co-lumbia University.

Lauren Resnick, the director of Pitts-burgh's Learning Research and Devel-opment Center, told those attendingthe 1997 assessment conference of theCenter for Research on Evaluation,Standards and Student Testing thatwhen professional development cen-tered on discussions of standards andstudent work, teachers of low-perform-ing students in districts using the NewStandards were able to "move thesekids off the bottom."

How to Do ItThough the phenomenon of analyz-

ing student work to promote change inclassroom practice is spreading quickly,

it's still an elusive idea. No commercialentity has yet boxed the process into aneasy, over-the-counter set of directions.No one proponent or strategy yetdominates either, although two ap-proaches have emerged as the mostcommon.

The first approachillustrated by"Standards-Based Professional Devel-opment: Getting Standards into theClassroom" developed by the Educa-tion Trust (see box below)assumesthat standards are in place, and the taskis to make sure student work reflectsthem. The other approachexempli-fied by the "Collaborative AssessmentConference" developed by Project Zero(see box p. 2)seeks to emphasizewhat teachers can learn by examiningstudent work before it's judged. Ac-cording to Steve Seidel, a research as-sociate at Project Zero, such discus-sions can lead teachers to look beyondtheir assignments and toward recogniz-ing students' creativity whether or notit is directly related to an assignment.

Despite the philosophical differencebetween these two approaches, theyshare some commonalties when itcomes to the actual process of lookingat student work. Often a trained facili-tator leads a group of teachers througha step-by-step format, usually called a"protocol." Discussions may involvescoring work according to a specificallydefined set of rubrics (descriptions thatdefine different levels of quality fromworst to best). But discussions can also

focus on the quality of the assignmentsthemselves. Much of the professionaldevelopment now given by the Educa-tion Trust, for example, is on designinggood assignments, often beginningwith teachers doing their assignmentsthemselves. Essentially, contends RuthMitchell, an assessment expert whoworks with the Education Trust, exam-ining student work "is a strategy to lookat teachers' work." Students can do nobetter than the assignments they aregiven, she contends, and if a teacher'sassignment is divorced from standards,it becomes a "mystical experience" forstudents.

Crucial IngredientsFor many educators and schools, the

hardest part may be getting started.Most experts with experience in col-laborative assessment say it takes a highlevel of trust among staff. In the "TuningProtocol" developed by the Coalition ofEssential Schools, for example, discus-sions are carefully designed to keep theconversations about student work fo-cused and away from personal criti-cisms. Its structure gives equal atten-tion to feedback that is "warm"(supportive) and "cool" (more dis-tanced). To break the ice, Katherine No-lan brings her own personal collectionof student work to discussions. "If youuse work too close to teachers at first,they will shut down. You have to beneutral," she says. Once teachers beginto share ideas, they can move to bring-

Standards-Based Professional Development:Getting Standards into the Classroom

This tool was developed by Education Trust staff, with fund-ing from the Office of Educational Research and Improve-ment. It is meant to be used by teams of teachers who worktogether to revise their assignments to meet standards. Itsguidelines stress the importance of keeping standards inmind when choosing or adapting an assignment. The Edu-cation Trust is a nonprofit organization based in Washing-ton, DC, that promotes educational success for all studentsfrom kindergarten through college.

(This process should take approximately 90 to 120 minutes.)

The following model is for teachers who are in a process ofaligning classroom work with standards. Although it includesdesigning a scoring guide, the model is not primarily con-cerned with giving numbers to student work. It uses scoringas a tool to focus -attention on the quality of classroom as-

signments and their direct connection with standards. If care-fully followed through these steps, the model will result inrigorous assignments and scoring guides that will enable stu-dents to recognize and reach for high standards.

1. We all complete the assignment.2. We identify the standards that apply to this assignment.3. We generate a rough scoring guide from the standards and

the assignment.4. We score the student work, using the guide.5. We ask: Will this work meet the standards? If not, what are

we going to do about it?6. What action can we plan at the classroom, school, district,

and state levels so that all students meet the standards onassignments like this?

Reprinted with permission from the Education Trust.

12 The Harvard Education Letter, March/April 1998 3

ing work from their own classroom.She pushes them to ask, "Can our kidsdo any better than this?" and helpsthem build exemplars of student workfrom across their district.

As Nolan's example shows, anothercrucial ingredient to all models of col-laborative assessment is the presence ofan objective third party. Most schoolsthat are trying to focus on student workrely on outside experts and/or outsidefunding to help them develop teachers'capacity to analyze the standards in stu-dent work and change their assign-ments and practices. Seidel's effort todevelop conversations based on stu-dent work in Massachusetts began withconsultants who spent 15-20 days ayear in a schoola model he says is tooexpensive. The project is now workingwith "district coaches," usually staffmembers who serve as resources forteachers.

Even if the expertise, time, and trustneeded to stimulate group conversa-tions around student work exist, therestill is the problem of whether teachersare able to understand what the proc-ess can teach them. Researchers whostudied the three-year effort to useteachers to write new standards for theNorth Philadelphia effort of the Educa-tion Trust found that the process "was

frustrating to those without a strongbackground of content knowledge anda wealth of teaching techniques."

Similarly, when the Philadelphia dis-trict sought to help teachers developunits of study based on standards,school officials learned how muchteachers lacked theoretical knowledgeabout teaching and learning. So for1997-98, teachers were given curricu-lum frameworks based on the district'sstandards, and summer and school-year workshops are being held to helpteachers integrate standards in theirdisciplines, one by one, using studentwork as the guide. It will take four yearsto cover all the standards, according toNolan, who is working with the district.

Such barriers will have to be over-come if standards-based reform is totake hold, according to Lauren Resnick.Raising standards will become the per-sonal goal of teachers and students, shesays, "only if a concerted effort is madeto engage teachers and students in amassive and continuing conversationabout what students should learn, whatkinds of work they should do, and howwell they should be expected to do it."

For Further InformationD. Allen. Tuning Protocol: A Process for Reflection.Providence, RI: Coalition of Essential Schools, Feb-ruary 1995.

T. Blythe et al. A Guide to Looking Collaborativelyat Student Work. New York: Teachers College Press,in press.K. Cushman. "Looking Collaboratively at StudentWork: An Essential Tool Kit." Horace 13, no. 2(1996). Available from Coalition of EssentialSchools, Box 1969, Brown University, Providence, RI02912; 401-863-3384.Education Trust. For information about this organi-zation, contact Ruth Mitchell, Education Trust, 1725K Street, NW, Washington, DC 20006; 202-293-1217.

B. Falk and S. Ort. Sitting Down to Score: TeacherLearning Through Assessment. Paper presented atthe annual meeting of the Center for Research onEvaluation, Standards and Teaching, 1997. Availablefrom National Center for Restructuring Education,Schools and Teaching at Teachers College, Box 110,Columbia University, New York, NY 10027; 212 -678-3432.Harvard Project Zero. For more information aboutthe Collaborative Assessment Conference, contactSara Hendren, Project Zero, 323 Longfellow Hall, 13Appian Way, Cambridge, MA 02138; 617-495-4342.R. Lopez et al. Get Rolling! Washington, DC: Educa-tion Trust, 1998.R. Mitchell. Front End Alignment. Washington, DC:Education Trust, 1996.S. Seidel et al. Portfolio Practices: Thinking Throughthe Assessment of Children's Work. Cambridge, MA:Harvard Project Zero (NEA School Restructuring Se-ries), 1997. Available from Karen Chalfen, ProjectZero, 323 Longfellow Hall, 13 Appian Way, Cam-bridge, MA 02138; 617-495-4342.

Anne C. Lewis is an education policy writer in theWashington, DC, area. She has been a national col-umnist for Phi Delta Kappan for 15 years.

Time and LearningAs schools and states tinker with time, other factors determine if it makes any difference

BY MICHAEL SADOWSKI

more "structured learning."Longer school days. Blockscheduling. Year-roundschools. The list of well-in-

tentioned school reforms and reformproposals targeting the use of time goeson and on. Four years ago this April, theU.S. Department of Education pub-lished Prisoners of Time, a report thatcalled for radical rethinking of thetime/schooling formula. Striking asomewhat alarmist tone, its authorswarned that "American students musthave more time for learning. The six-hour, 180-day school year should berelegated to museums, an exhibit fromour education past." U.S. studentsspend far less time on core subjects

than students in France, Germany, andJapan, argued the authors, warning thatthe use of time in U.S. schools is "a rec-ipe for a kind of slow-motion social sui-cide."

But does giving students more timeto learn necessarily mean that they willlearn more? Does changing the schoolday or year change learning in anymeaningful way? Research and experi-ence point to a complex set of answersaffected by a host of other variables.

Different Kinds of TimeIn a review of more than 130 studies

published in 1997 by the Brookings In-stitution, Herbert 3. Walberg, ResearchProfessor of Education and Sociology at

the University of Illinois-Chicago, findswhat he calls a "highly consistent" rela-tionship between increased schooltime and better achievement. In 97 per-cent of these studies, there was a posi-tive relationship between time andlearning. "This is one of the most con-sistent of all relationships in educationresearch, comparable to that of socio-economic status and academic achieve-ment," Walberg says.

While Walberg calls more schooltime for students "an obvious and long-standing implication" of his findings,however, he makes some strong quali-fications. First, he admits that the de-gree of the relationship between timeand learning in many of the studies he

4 The Harvard Education Letter, March/April 1998

reviewed is relatively small. Second, hedistinguishes between three differentkinds of time students can spend inschool and notes that only one is likelyto make a significant difference by itself,suggesting that merely adding time tothe school schedule may not result inmore learning.

For time to be productive, Walbergargues, it must be spent in ways that aresuitable to each individual learner. Hegoes on to differentiate between sev-eral types of time: "allocated time," likethat spent in class; "engaged time,"spent in class activities such as perform-ing a laboratory project or taking noteson a lecture; and "productive time" inwhich a student learns new things. En-gaged and allocatedtime can be wastedtime from the individ-ual learner's perspec-tive, Walberg pointsout, if no account istaken of his or her pre-vious knowledge base,learning needs, andother variables: "It's

has been found to have a direct corre-lation with achievement."

"Structured Time" inMassachusetts

Despite its limitations, time restruc-turing remains a linchpin of many edu-cation reform initiatives. In Massachu-setts, for example, new statewide "timeand learning" requirements werephased in over the past three years.Massachusetts elementary schools havelong been required to provide 900hours a year of instruction (990 for sec-ondary schools), and while the num-bers haven't been changed, they mustnow consist of "structured time" in a

defined range of coresubject areas: mathe-

Time changes thatdon't addressstudents' needscan reinforceinequalities.

wasteful educationalpolicy to teach some kids things theyalready know and other kids thingsthey aren't yet capable of learning."

A third qualification Walberg makesconcerns other influences such ashome environment, television viewing,peer groups, and other things that canhave an impact on learning. These"educational productivity factors" mustbe taken into account along with theamount of learning time built into aschedule, he says.

For these reasons, Walberg says,across-the-board time changes thatdon't also address.students' individualneeds and circumstances can reinforceinequalities: those who are already ad-vantaged educationally tend to benefitthe most from any kind of school re-form that does not account for individ-ual differences.

Other research reviews underscorethe limitations of defining learning timetoo broadly. Synthesizing recent studies(including Walberg's) for the Educa-tional Resource and Information Clear-inghouse (ERIC), Thomas I. Ellis writesin a 1997 ERIC Digest,". . . the relation-ship between time and learning is com-plex and problematic. First, a distinc-tion must be made between timeallocated for instruction, time engagedin instructional activities, and timespent successfully completing instruc-tional activities. Only the last of these

matics, science/tech-nology, history and so-cial science, English,world languages, andthe arts. Study halls,lunch, passing time, ex-tracurricular activitymeetings, and assem-blies (unless they are

directly related to the curriculum) mayno longer be counted. "One of the mostshocking things we discovered [on thecommission] were wide variances inthe academic diet for students," ex-plains S. Paul Reville, who served aschairman of the state's Commission onTime and Learning. "Some studentswere taking three study halls a day andothers were taking six hours of high-level academic work. Now we're sayingall students have a right to that full dayof instruction."

The effects have varied from districtto district. Joseph Rappa, another com-mission member and superintendentof schools in Leominster, MA, says hehas had very little trouble implement-ing the new hourly requirements forcore subjects because his relativelywell-funded schools already hadlonger-than-required hours. He con-cedes, however, that the change is cre-ating a time squeeze in other districtswhere money is spread thin, localteachers' unions are demanding moremoney for the longer instructionalhours, and taxpayers are loath to ap-prove a tax increase for additionalschool funding. Many districts in thestate, for example, have moved to blockscheduling not to meet curriculum ob-jectives, but simply to eliminate 15 orso minutes of student passing time thathad previously been built into the day.

The authors of the Massachusetts re-port, which was patterned in its contentand form after Prisoners of Time, ac-knowledge that adding instructionaltime in itself will do little to boostachievement and that the real need isfor greater flexibility. Reville notes thata more flexible approach to time, whichimplies longer school hours for staffand some students, would be moreequitable because all students wouldhave an opportunity to learn what theyneed to know, not just those who can"get it" in the time allotted. "We [on thecommission] saw it as an issue of eq-uity," he explains. "Under the usual sys-tem, time is the constant and learningvariesstudents either get it or theydon't, then they move on to the nextlesson. We need to have learning be theconstant, and if it takes you x or y or zamount of time, the system should beflexible enough to accommodate that."

Do Parents Favor Change?The state of Virginia provides an-

other case in point. In the early 1990s,the Virginia Department of Educationconducted a research review to deter-mine whether the state should increasethe number of days and hours studentsare required to spend in school. Afterits review, the research team recom-mended in 1992 that the state mandateno changes in the length of the schoolday or year (except to allow for in-creased time for "students at risk").Borrowing heavily from Walberg's 1984study, the team's report emphasizedthe differences between allocated, en-gaged, and productive instructionaltime and noted that only the last wasstrongly linked to improved achieve-ment. Still, the Virginia General Assem-bly requested that the department ofeducation calculate the cost of length-ening the school year. Lissa Power-de-Fur, the policy analyst who led the re-search team, says that once thiscost-impact study was completed, in-terest in lengthening the school yearquickly waned. "The legislators foundout that each additional day would costan additional $13.6 million in state andlocal funds for teaching staff and trans-portation alone," Power-deFur says."Once you run those kind of numbers,you aren't left with too many people onboard."

Power-deFur believes another factorthat influenced her team against rec-ommending an extension of schoolhours was the opinion of Virginia resi-dents. More than two-thirds of those

The Harvard Education Letter, March/April 1998 5

polled in a 1991 statewide survey saidthey thought the length of the schoolyear should remain the same. For peo-ple with children in the public schools,this number was even higher: 77 per-cent favored no change.

Other state polls, as well as nationalstatistics, are in line with the Virginiafindings. In the latest Phi Delta Kap-pan/Gallup Poll of the Public's Attitudesof the Public Schools, only 25 percentof those surveyed believe a longerschool day would improve academicachievement appreciably in their dis-trict.

When the Georgia Partnership forExcellence in Education, a private or-ganization that has been promoting alonger school year in that state, lookedat residents' attitudes, its findings weresimilar. Only 6 percent of Georgia resi-dents polled regarded lengthening theschool day and year as "one of the mostimportant ways to improve the qualityof education in Georgia schools."

Year-Round SchoolsAnother reform movement to alter

radically the way school time is struc-tured seems to be growing, particularlyon the West Coast: year-round schools.According to the San Diego-based Na-tional Association for Year-Round Edu-cation (NAYRE), there are now 2,460year-round schools in 41 states. Morethan 1.7 million public school studentsin the U.S. now attend year-roundschoolsabout five times the numberthat did in 1985and more than a mil-lion of these are California children.Some year-round schools have stu-dents on multiple scheduling tracksadopted mainly to ease overcrowding

by staggering vacations and spreadingattendance out over the calendar year.According to NAYRE, more than three-quarters of year-round schools haveadopted this schedule for the academicbenefits.

There are indications that year-round schooling could make a differ-ence for some students by simply elimi-nating the long summer breaks, and theneed for transitions before and afterthem. In a recent review of 39 studiesof the effects of summer vacation onstudent learning, Harris Cooper, pro-fessor of psychology at the University ofMissouri, found that the summer breakresulted in an average of one month oflearning loss (based on a grade-levelequivalent). The declines were mostmarked, Cooper says, in math compu-tation and spelling skills. Perhaps mostsignificantly, Cooper notes that the ef-fects of summer break seem to be mostdetrimental for students of low socio-economic status (those eligible for freeor reduced-price lunch) and that somestudents from middle- and upper-in-come families actually show gains insome areas.

"The summer break has a differentimpact on students, depending ontheir economic resources," concludesCooper. "Children from high socioeco-nomic status homes are more likely toengage in activities that provide prac-tice of academic material."

The Real QuestionRevile, Walberg, and others who call

for change in the way schools use timeadmit that it is merely a structure, butone that is critical to the kinds of reformthat can really make a difference in stu-

dent learning. "Time is not a panacea,"says Walberg. "But now kids are defi-nitely handicapped by a school yearthat's too short."

Further, as the Prisoners of Timeauthors noted four years ago, the realquestion may not be whether changesin school hours make any difference,but whether there is sufficient publicwill to commit the resources that arenecessary for real change. "None of[our recommendations] will makemuch difference unless there is a trans-formation in attitudes about education.The transformation we seek requires awidespread conviction in our societythat learning matters," the authors say."Until our nation embraces the impor-tance of education as an investment inour economic future...nothing will re-ally change."

For Further InformationC. Ballinger. "Prisoners No More." EducationalLeadership 53, no. 3 (November 1995): 28-31.H. Cooper. "The Effects of Summer Vacation onAchievement Test Scores: A Meta-Analytic Review"Review of Educational Research 66, no. 3 (Fall1996): 227.Massachusetts Department of Education. Unlockingthe Power of Time: The Massachusetts Commissionon Time and Learning Final Report. Malden, MA:MA Department of Education, November 1995; 781-388 -3300, ext. 306; www.doe.mass.edu.U.S. Department of Education. Prisoners of lime:Report of the National Education Commission onTime and Learning. Washington, DC: U.S. Depart-ment of Education, April 1994.H. J. Walberg. "Uncompetitive American Schools:Causes and Cures." In Brookings Papers on Educa-tion Policy. Washington, DC: The Brookings Institu-tion, 1997.

Michael Sadowski teaches English and drama atDennis-Yarmouth Regional High School in Massa-chusetts.

Small Schools Work Best forDisadvantaged StudentsNew research looks at who benefits most from small schools

BY PEGGY FARBER

In October 1997, the New YorkTimes published a draft proposalby the New York City school chan-cellor's office recommending that

the city establish minimum school en-rollments of 400 students for elemen-

tary schools, 600 for middle schools,and 800 for high schools in its districtof more than 1 million students.

Just five years earlier, the city an-nounced a program to create dozens ofsmall schools with enrollments of un-

der 400 students. Accordingly, commu-nity groups, school reformers, and lo-cal districts got together and startedabout 150 small public schools, somewith enrollments of under 100 stu-dents.

6 The Harvard Education Letter, March/April 1998

Although Chancellor Rudy Crewbacked away from the recommenda-tions and assured reformers that he re-mains absolutely committed to smallschools, in February 1998 he orderedabout the half the city's alternative highschools to start increasing enrollmentsimmediately, causing alarm amongsmall school principals.

The New York controversy under-scores the continuing tension betweenresearchers who say small schools arebetter and policymakers who make de-cisions about how big schools will be.Is there an optimal school size? Dosmall schoolsor large schoolsbenefit some children more than oth-ers? What does the research tell usabout the impact of school size on aca-demic achievement?

Research Favors SmallSchools

Although the definition of what asmall or large school is varies fromstudy to study, there is, in fact, a largebody of research showing that whenlooking at a group of schools of a varietyof levels and sizes, there is a powerfulrelationship between school size andstudent success most of which favorssmaller schools. Among the findings:

The largest schools, those with 2,000students or more, are ineffective foralmost all students and drasticallyimpede the progress of impover-ished children.The well-documented positive rela-tionship between socioeconomicstatus and achievement is disruptedin small schools. In schools with en-rollments under 1,000, economicstatus plays a significantly smallerrole in determining who succeedsand who fails.The impact of school size on aca-demic achievement is identical in ru-ral and urban United States.Four hundred students are enoughfor a high school to offer an adequaterange of courses. However, one re-cent study indicates that high schoolssmaller than that may be too small.

Two recently published reviews ofthe research literature on school size,"School Size, School Climate and Stu-dent Performance" by Kathleen Cotton,and "The Small School Movement: AReview of the Literature" by RobertGladden, reach identical conclusionsabout what the research says: Smallschools are safer, have significantly

lower dropout rates, better attendancerecords, and better participation in awide range of school activities.

Ironically, given the ongoing debatein New York, some of the sturdiest evi-dence that small schools work best forstudents from impoverished familiescomes from New York City, where a vi-brant network of small public schoolswith enrollments of under 300 stu-dents at the elementary level and 500at the secondary level was started 25years ago by Deborah Meier. Meier saysher daily encounters as a young teacherwith the anonymity and humiliationpoor kids endure in large urbanschools appalled her.

"Education comes through relation-ships," Meier says, summing up thecentral concept demonstrated in re-search and practice in small schools."You cannot introduce young people tothe idea of what being an educated per-son is all about in the absence of rela-tionships with adults who representthat."

Trend Toward BignessDespite research findings, the trend

in school enrollments throughout the20th centuryand continuing todayhas been toward consolidation andlarger schools. Before World War II, theaverage U.S. school served 130 stu-dents. That average has increased to650 students. The average U.S. highschool has about 720 students.

Cities with sprawling ghettos areoften assumed to be the districts thathave schools with soaring enrollments.Indeed, of the 49 high schools in LosAngeles, 17 have more than 3,000 stu-dents, and nine have close to or morethan 4,000 students. Most of the 74high schools in Chicago are over 1,000students strong, 11 have close to ormore than 3,000 students, and the larg-est has over 4,000 students. In NewYork City, 35 out of 200 high schoolshave more than 3,000 students, and adozen of those have more than 4,000.

However, rural children have alsobeen subject to the effects of schoolconsolidations and ever increasing en-rollments. Since 1990, West Virginia hasclosed one-fifth of its schools, for exam-ple. While only about 25 percent of allAmerican children attend high schoolswith more than 1,500 students, thosepercentages are much higher in Nevada(68%), Utah (62%), and Arizona (65%).

The goals driving consolidationeconomies of scale and a variety ofcourse offeringsfirst articulated by

James Conant in his influential 1959book The American High School Today,have been largely debunked in recentyears by economists and researchers.Several studies cited by both Cottonand Gladden show per pupil costs ris-ing as enrollments pass the level of amedium size school because of theadded administrative expenditures thatbecome necessary. Marty Strange, pol-icy manager and research director ofthe Annenberg Foundation's RuralChallenge, says there's more myth thantheory driving the nation's continuingromance with large schools. "Bigness isan article of faith in our society,"Strange says. "We look for reasons tobelieve that big schools are better, bigbusiness is better, big farms are better."

Disadvantaged StudentsSeveral studies have concluded that

small schools are better, especiallywhen the students attending are disad-vantaged. One of the most widely citedstudies, conducted by Noah Friedkinand Juan Necochea in California in1988, was replicated in West Virginia in1995 by Craig Howley. Researcherslooked at achievement of children in3rd, 6th, 9th, and 11th grades as meas-ured by standardized tests in both largeand small schools. Howley concluded,as Friedkin and Necochea did, that chil-dren from affluent families do slightlybetter in large schools than they do insmall schools, but poor children per-form at a much higher level in smallschools than they do in large ones.

For his study, Howley focused onboth elementary and secondaryschools of various sizes. Howley recom-mends that maximum school sizes beconsidered according to the socioeco-nomic status of the students served. Ina 1997 article summarizing his re-search, he wrote that an elementaryschool for an affluent neighborhoodcould accommodate 500 students,while another in an impoverishedneighborhood should be no larger than100. Similarly, middle schools and high

COMING SOON

The Bilingual EducationDebateADD in the ClassroomReading RecoveryContinues to Grow

The Harvard Education Letter, March /April 1998 7

schools should serve no more than 600and 1,000 advantaged youngsters, re-spectively, with much lower enroll-ments in poor neighborhoods.

"If a school district is serving a largeproportion of poor youngsters,"Howley wrote, "it would be well ad-vised to do everything possible to keepsmall schools open."

Strange, who is working withHowley on a series of new studies forthe Annenberg Foundation, com-ments, "We all agree that schools aresupposed to equalize things for kids, tostrip away the weight of class and privi-lege. Big schoolsdon't do that. Theycreate environmentsthat widen gaps."

These findings arereiterated in a 1997study by Valerie E. Leeand Julia B. Smith,"High School Size:Which Works Best and

103 studies for her 1996 article,"School Size, School Climate, and Stu-dent Performance."

Cotton points out that researchersstudying schools within schools oftenleave out such crucial information aswhether the smaller units are truly self-contained or whether they have anycontrol over budgeting, staffing, andcurriculum. She notes that most practi-tioners regard these characteristics tobe essential ingredients of a successfulsmall school.

There is no consensus on optimalschool size, although Lee and Smith see

enrollments in the600-900 range as

Small schools aresafer and have lowerdropout rates andbetter attendancerecords.

for Whom?" For theirstudy, Lee and Smithdivided 800 high schools into eightcategories, with the smallest having 300students or less and the largest withmore than 2,100 students. After track-ing test scores of students over a six-year period, Lee and Smith found thatthe largest inequities occur in the larg-est schools. In schools with over 2,000students, disadvantaged children learnvery little. For the most part, wealthierchildren achieve at a fairly constantlevel.

Two additional findings emergedfrom Lee and Smith's analysis. First, inhigh schools with more than 2,000 stu-dents, all students, whether from privi-leged or impoverished communities,gain almost nothing in reading. Sec-ond, high schools with fewer than 300students may be too small. Gains inreading and math are best for all stu-dents, regardless of wealth, who attendmoderately small high schools with 600to 900 students. However, smallness isnot nearly as injurious as bigness. Stu-dents may learn less in tiny schools thanthey do in moderately small schools,but they still learn considerably morethan they do in the largest schools.

Schools Within SchoolsThe practice of dividing large high

schools into smaller houses has re-ceived far less study, and results are sofar inconclusive, according to Cotton, aresearcher at the Northwest RegionalEducation Laboratory who examined

ideal. Still, a numberof researchers sug-gest roughly 100 to125 students pergrade as a good size.Several researchers,notably David Monkand ChristopherRoellke, have shownthat high schools en-

rolling roughly 400 students are able tooffer a curriculum that compares favor-ably with the range of courses in muchlarger schools.

Power of RelationshipsNumbers can tell how but not why

small schools are better. But re-searchers and small school advocatesecho Meier's emphasis on the power ofrelationships to transform kids. Theyoften use words like connections, trust,engagement, and democracy to de-scribe their school climates.

Jacqueline Ancess, an associate di-rector of the National Center for Re-structuring Education, Schools, andTeaching (NCREST) at Teachers Col-lege, conducted hundreds of inter-views with students in small, inner-cityschools. She has 23 years' experienceas a teacher and an administrator in theNew York City public school system,some of it in alternative schools that shehelped launch. According to Ancess,small schools work because they "allowclose personal relationships betweenkids and teachers, kids and kids, andteachers and teachers. For many kids,the personal relationships, the sense ofcommunity, the power of community,becomes the conduit for learning."

Ancess sees teachers in small schoolsemboldened by the power of these re-lationships. "Adults in these school arerelentless in their nagging. You have aculture of teacher perseverance. There

are all these stories teachers tell one an-other in these schools about differentkids they dragged kicking and scream-ing through resistance to produce qual-ity work. The student resistance nevergoes away, but neither does the perse-veranceit's built into the structure ofthe school." But Ancess warns thatsmallness is only one of the necessaryelements: "Relationships have to beused as levers for high academicachievement. If you're not going to usethem for that, it's a waste of time."

Cotton finds the discrepancies be-tween what the research says andpopular practice very frustrating. "Ifsomeone were setting out deliberatelyto stack the deck against poor or minor-ity students, they couldn't do a muchbetter job," she said. "It's one of thoseareas where the research points in onedirection and the world is running asfast as it can in the opposite direction."

For Further InformationJ. Ancess. "Urban Dreamcatchers: Planning andLaunching New Small Schools." In Small Schools,Big Imaginations: A Creative Look at Urban PublicSchools. Chicago: Cross City Campaign for UrbanSchool Reform, in press; 312-322-4880.

K. Cotton. "School Size, School Climate, and Stu-dent Performance." Close-Up Number 20 in theSchool Improvement Research Series, produced bythe Northwest Regional Education Laboratory 1996;503-275-9615; www.nwrel.org/scpd/sirs/10/c020.html.

N. Friedkin and J. Necochea. "School System Sizeand Performance: A Contingency Perspective." Edu-cational Evaluation and Policy Analysis 10, no. 3(1988): 237-249.

R. Gladden. "The Small School Movement: A Reviewof the Literature." In Small Schools, Big Imagina-tions: A Creative Look at Urban Public Schools. Chi-cago: Cross City Campaign for Urban SchoolReform, in press; 312-322-4880.

C. Howley "Compounding Disadvantage: The Ef-fects of School and District Size on Student Achieve-ment in West Virginia. " Journal of Research in RuralEducation 12, no. 1 (Spring 1996): 25-32.

C. Howley. "Dumbing Down By Sizing Up." SchoolAdministrator 54, no. 9 (October 1997): 24-30.

V Lee and J. Smith. "High School Size: Which SizeWorks Best and for Whom?" Educational Evalu-ation and Policy Analysis 19, no. 3 (Fall 1997): 205-227.

D. Meier. The Power of Their Ideas: Lessons forAmerica from a Small School in Harlem. Boston,MA: Beacon Press, 1995.

D. Monk "Secondary School Enrollment and Cur-ricular Comprehensiveness." Economics of Educa-tion Review 6, no. 2 (1987): 137-150.C. Roellke. "Curriculum Adequacy and Quality inHigh Schools Enrolling Fewer than 400 Pupils."EricDigest (EDO 401 090). Available from Eric Clearing-house on Rural Education and Small Schools, De-cember 1996; 800-624-9120.

Peggy Farber is a freelance education reporterbased in New York City.

8 The Harvard Education Letter, March/April 1998

Volume XR; Number 3 Published by the Harvard Graduate School of Education May/June 1998

The Bilingual Education DebateA long-term view may be necessary to recognize benefits of bilingual programs

BY LAUREL SHAPER WALTERS

This year marks the 30th anni-versary of the federal BilingualEducation Act, the original leg-islation that created special

programs for students who are learningto speak English. Yet instead of celebra-tions, bilingual education is facing at-tacks and reexamination from all sides.

In California, the state with the larg-est percentage of limited-English-profi-cient (LEP) students, voters will decideon June 2 whether bilingual classroomsshould be eliminated and replaced withone-year, sheltered English-immersionclasses (see "Models of Language In-struction," p. 3). Officially named Prop-osition 227, the California ballot ques-tion comes in the wake of complaintsby immigrant parents and some bilin-

INSIDE

Bilingual EducationModels/ 3

The Bilingual EducationDebate in California / 4

ADHD in the Classroom / 6

Strategies for TeachingKids with ADHD / 8

gual educators that students are lan-guishing in bilingual classes withoutlearning enough English.

Throughout the country, the Cali-fornia initiative has sparked a reneweddebate about the rela-tive effectiveness of themany different ap-proaches to educatingthe growing popula-tion of LEP students.And it is also provokingsome soul-searchingamong bilingual advo-

are taught in their native languagewhile learning English. A few recentstudies, however, have turned up someprovocative new evidence of benefitsfor students who get a strong foun-

dation in more thanone language. "Late-exit" transitional bilin-gual programs and two-way programs that areintended to developliteracy in both the na-tive language and Eng-lish appear to help LEP

"A lot of bprogramsas goodought to b

ilingualaren't

as theye."

cates who are arguingfor the reform of bilingual educationrather than its elimination.

After three decades, research on bi-lingual education is extensive, withhundreds of studies available. Yet muchof it has been called into question byconflicting interpretations or has evenbeen dismissed as methodologicallyunsound. In a 1996 review of the litera-ture, for example, Boston University'sChristine Rossell found only 25 percentof 300 program evaluations to bemethodologically acceptable. The oth-ers often failed to use control groups ordid not include a statistical control forsocioeconomic or other differences.

Most research findings so far havefailed to demonstrate the superiority oftransitional bilingual programs, themost common type, in which students

students perform bet-ter throughout high school.

The Great DivideIn 1974, the U.S. Supreme Court

ruled in Lau v. Nichols that schoolsmust take "affirmative steps" to helpstudents who do not speak English.The Court did not specify the type ofprogram required, however. Today, thedifferent approaches to teaching mi-nority-language students comprise adizzying list. Methods break down intodifferent models, including those thatincorporate the native language such astransitional bilingual education, as wellas English-language programs, whichinclude structured immersion and Eng-lish as a Second Language.

Of the nearly 3 million LEP studentsnationwide, 74 percent are Spanish

EDITORIAL DIRECTOR: Kelly Graves-Desai. EDITOR: Nancy Walser. PRODUCTION EDITOR: Dody Riggs. EDITORIAL BOARD, HARVARD GRADUATE SCHOOL OFEDUCATION: Mildred Blackman, Director, The Principals' Center; Sally Dias, Superintendent, Watertown Public Schools, Watertown, MA; Jay P. Heubert, AssistantProfessor; Harold Howe II, Senior Lecturer Emeritus; Susan MooreJohnson, Professor and Academic Dean; Robert Kegan, Senior Lecturer; Jerome T. Murphy, Professorand Dean; Gary A. Orfield, Professor; Robert S. Peterkin, Senior Lecturer; John Ritchie, Superintendent/Principal, Lincoln-Sudbury Regional High School, Sudbury, MA;Judith D. Singer, Professor; Jay Sugarman, Teacher, Runkle School, Brookline, MA; Dennie Palmer Wolf, Lecturer on Education. NATIONAL ADVISORY BOARD: JohnBrademas, President Emeritus, New York University; Constance E. Clayton, former Superintendent, School District of Philadelphia; Alonzo A. Crim, Professor ofEducation, Spelman College; Linda Darling-Hammond, Professor, Teachers College, Columbia University; Andrew Heiskell, Chairman Emeritus, New York Public Library;Marys Levenson, Superintendent, North Colonie Central Schools, NY; Deborah Meier, Founder, Central Park East Secondary School, NY; John Merrow, President, TheMerrow Report; ArthurJ. Rosenthal, Publishing Consultant.

1 (c) BEST COPY AVAILABLE

speakers, according to the NationalCenter for Education Statistics. Viet-namese is the next largest languagegroup, at 3.9 percent. Arguments abouthow best to teach LEP students usuallyfocus on how soon they should betaught in English-only classrooms. Thearguments are often motivated by po-litical agendas, with strong views onboth sides.

Part of the political debate stemsfrom disparate goals. Many supportersof bilingual educa-tion view Englishlearners as an op-portunity to createtruly bilingual andbiliterate graduates.Opponents arguethat tax dollarsshould not be spenton maintaining achild's native lan-guage and culture.The goal, from this perspective, is tohelp students learn English as quicklyas possible and move into mainstreamclassrooms.

At the same time, both sides want tobe sure students keep up with theirpeers in academic coursework. Bilin-gual supporters say this is best accom-plished by teaching in the native lan-guage and gradually introducing moreEnglish as students gain literacy in theirfirst language. Critics say this transi-tional approach isolates students fromtheir peers for too long and delays theacquisition of English.

But in schools whose students speakmany different languages, models thatinclude instruction in the native lan-guageparticularly the two-way modelare not always practical. Classroomswith students from a variety of back-grounds often rely on English as a Sec-ond Language programs, which offerintensive English instruction from aspecially trained teacher for part of theday.

Shortcomings of BilingualEducation

Studies comparing different modelshave failed to demonstrate the supe-riority of transitional bilingual pro-grams, despite widespread federal

support and implementation of this ap-proach. One of the most comprehen-sive studies in the field is a 1991 feder-ally financed project that followedmore than 2,000 elementary childrenover four years. It looked at three pro-grams: structured English immersion(special instruction in English only),early-exit transitional bilingual educa-tion (students mainstreamed intoEnglish-only classrooms by the end offirst or second grade), and late-exit

transitional bilin-gual education (stu-

Students who receivedEnglish-only instructionfailed to sustain theiracademic progress inlater grades.

gual educationagainst it. Students in all three pro-grams learned English and made pro-gress. Therefore, the study concluded,providing substantial instruction in astudent's primary language does notdelay acquisition of English. Yet it didnot demonstrate that this approachboosted achievement or was superiorto the immersion program.

Studies have also shown that manystudents stay in bilingual classrooms

dents not main-streamed until theend of sixth grade).

This evaluation,known as the Ra-mirez study after itsprincipal author,has been used asboth support fortransitional b ilin-and ammunition

60

50

40

30

20

101

FlaalNC61

52MEI MN

longer than the three years usually pre-scribed by the transitional bilingualmodel, and that when they are main-streamed into regular classrooms theyoften lag behind their English-speakingpeers.

Even the staunchest bilingual propo-nents readily admit that the current sys-tem of transitional bilingual educationis failing many students. "A lot of bilin-gual programs aren't as good as theyought to be," says Catherine Snow, aresearcher and professor at HarvardUniversity's Graduate School of Educa-tion. "It's not surprising that parents ofkids in those programs end up beingdissatisfied." Snow chaired a NationalResearch Council committee that an-nounced in March that "initial readinginstruction for children who do notspeak English is best carried out in thechild's home language."

A Case for the Long ViewRecent research, however, suggests

that a long-term view may be necessaryin evaluating the effectiveness of lan-guage programs. In several studiesin-cluding the Ramirez study and studiesby Virginia P Collier and Wayne PThomas of George Mason Universitychildren schooled in English-languageprograms made faster gains in Englishthan comparable children in bilingualprograms. But around the third orfourth year, the children who were

Programs1 Two-way developmental Bilingual Education

2 One-way Developmental BE + Content ESL

3 Transitional BE + Content ESL

4 Transitional BE + ESL both taught traditionally34 5 ESL taught through academic content

24 6 ESL Pullout taught traditionally

En ,m am Average performance of native-Englishspeakers making one year's progress ineach consecutive grade3 5 7 9

GRADE11

Patterns of K-12 English learners' long-term achievement in Normal Curve Equivalents onstandardized tests in English reading compared across six program models. Data are basedon 42,000 student records collected from five districts between 1991 and 1996Chart reprinted with permission from National Clearinghouse for Bilingual Education.

THE HARVARD EDUCATION LETTER (ISSN 8755:3716) is published bimonthly by the Harvard Graduate School of Education, Appian Way, Cambridge, MA 02138-3752. Second-class postage paidat Boston, MA, and additional mailing office. POSTMASTER: Send address change(s) to The Harvard Education Letter, 6 Appian Way, Cambridge, MA 02138-3752.

Signed articles in THE HARVARD EDUCATION LETTER represent the views of the authors. Address editorial correspondence to Editors, The Harvard Education Letter, Gutman Library, 6AppianWay, Cambridge, MA 02138; phone 617-495-3432; fax 617-496-3584; email address: edletter@hugsel.harvard.edu; web site: http://hugsel.harvard.edu/hepg/hel.html

© 1998 by the President and Fellows of Harvard College. Published as a nonprofit service. All rights reserved. Special written permission is required to reproduce in any manner, in whole or inpart, the material herein contained. Call 617- 495.3432 for reprint permission information.

How to subscribe: Send $32 for individuals, $39 for institutions ($40 for Canada/Mexico, $42 other foreign, in U.S. funds only) to The Harvard Education Letter,6 Appian Way, Cambridge MA02138; or call us at 617.495 -3432 in Massachusetts or 800-513-0763 outside Massachusetts between 8:30 a.m. and 5:00 p.m. EST MonidatFriday. Subscription prices subject tochange withoutnotice. Single copies, $5.00. Back issues and bulk subscriptions available at special reduced rates; call 800-513-0763. d

CO?Y AVAILABLE

Models of Language InstructionBilingual Immersion Students are taught in a separate class-room with most instruction in English. Part of the day, however,is set aside to develop concepts, literacy, and critical-thinkingskills in the native language. The goal is for students to improvetheir English so they can join regular classrooms in two to fouryears.

Early-Exit 21.ansitional Bilingual The most common modelin use. The student's native language is used as a bridge, begin-ning with entry into school until they learn enough English tofunction in a regular classroom, theoretically two to three years.Programs vary in how rapidly English is introduced and whichsubjects are taught in the native language. The goal is to developEnglish skills without delaying academic content.

English as a Second Language (ESL) Students get specialinstruction in English from a trained specialist, either in the regu-lar classroom or in a pull-out program varying from 30 minutesto half the day. The goal is to assist students who are learningsubjects primarily in English.

Late Exit Tkansitional Bilingual Education (Maintenanceor Developmental Bilingual Education) Students receivesome instruction in their native language until 6th grade, withthe increasing use of English. The goal is to foster literacy in thenative language as a foundation for later literacy in English, aswell as to instill a positive attitude toward the native culture.

Sheltered English Immersion (Structured Immersion)Academic content is taught in simple English using context andgesture to explain meaning. Some programs allow teachers touse students' home language for clarification or have native-lan-guage aides available when needed. The goal is to mainstreamstudents quickly.

Tivo-Way Bilingual Education Language-majority and lan-guage-minority students are taught together in the same bilin-gual class. Most programs are English/Spanish, but some areavailable in other languages. The goal is for all students to de-velop fluency in both languages.

taught at least partially in their nativelanguage began to catch up in English.While these students began to reducethe gap between their achievement andthat of their native English-speakingpeers, students who received English-only instruction failed to sustain theiracademic progress in late elementaryand secondary school. Collier andThomas looked at scores from nation-ally normed tests in all subjects, not justEnglish, to judge the effectiveness of sixdifferent bilingual models.

Collier blames short-term studies forproducing inconclusive research re-sults on bilingual education programs."In our current research," she says, "wehave found data patterns similar tothose often reported in other short-term studies focused on Grades K-3little difference between programs."Significant differences show up only af-ter students continue their schooling inmainstream courses, says Collier,whose ongoing research follows stu-dents through 11th grade. The prelimi-nary findings show that only studentswho receive both English and native-language instruction through grade 5or 6 are continuing to do well throughhigh school (see chart).

In fact, the question of how long ittakes students to become proficient ina new language is at the heart of thebilingual education controversy. Col-lier's research suggests that it takes stu-dents who receive English-only instruc-tion longer (7 to 10 years) to reachaverage achievement, compared to stu-dents provided with strong native-lan-guage support (4 to 7 years).

But some critics say the Collier andThomas study has not been subjectedto peer review "We only have theirstated findings," says Rosalie PedalinoPorter, director of the READ Institute inAmherst, MA, an organization advocat-ing English-language programs for LEPstudents. "I just cannot believe anychild sitting in an American classroomcan need 10 years to learn English."Collier and Thomas recently posted a96-page research summary on theWorld Wide Web for review by other re-searchers and practitioners.

How Long Does It Take?There appears to be no one answer

to the question of how long it takes fora non-English speaker to become fluentenough to keep up in the classroom.What often gets lost in the discussion isthe highly variable nature of languageacquisition, says Snow. Students of nor-mal intelligence can take vastly differ-ent amounts of time to master English.Socioeconomic background also playsa role.

Another important variable is age."What you need to do for 13-year-oldsand 5-year-olds is quite different,"Snow says. Considerable research sup-ports the theory that students wholearn the "mechanics of literacy" intheir native language can translate thatknowledge into learning another lan-guage more easily. Therefore, saysSnow, "the group that's really highly atrisk are the students who arrive atschool not knowing how to read.They're the ones that we really need toworry about."

The widespread myth that youngerchildren learn languages more easilythan older ones confuses the issue fur-ther. In fact, research shows that olderstudents who are literate in their nativelanguage are faster second-languagelearners.

Mary Cazabon, director of bilingualprograms in the Cambridge, MA, publicschools, sees this in her district. "If stu-dents come in at the high school levelliterate and fully on grade level in theirnative language, they make a smoothtransition into English and don't need7 to 9 years," she says. "The youngerchildren need to develop bilingually inorder to maximize their learning poten-tial," she believes.

In Cambridge, most Spanish-speak-ing students are taught in two-way bi-lingual programs with native Englishspeakers. Research shows strong aca-demic gains for both native-English andLEP students in two-way programs. YetCazabon recognizes that such a modelis impractical for all languages and allages. "You can't say that there is oneexact model that is going to work forevery child," she says.

Evaluation of BilingualClassrooms

Some research, in fact, concludesthat the specific models and languagesof instruction may be less importantthan the quality of teaching that lan-guage-minority students receive. Re-searchers note the lack of interactive in-struction in many programs for LEPstudents. When students have little op-portunity to speak in class, they don't

The Harvard Education Letter, May/June 1998 3

learn as quickly. There is also a concernabout the lack of books available to stu-dents in bilingual programs.

"Many concerned educators realizethat in order to improve the educationof bilingual students, we need to go be-yond the debate on language choiceand support for particular models.... Is-sues of pedagogy need to be ad-dressed," writes Boston University pro-fessor Maria Estela Brisk in her 1998book, Bilingual Education: From Com-pensatory to Quality Schooling.

Even in schools using the samemodel, there can be drastic differencesin instructional approaches. In theRamirez report, for example, studentswho attended one of the late-exitschools performed significantly betterthan students at the other late-exitschools. The researchers noted that the

higher scoring students attended a kin-dergarten that emphasized critical-thinking skills in the native language.Another study noted dramatic differ-ences in instructional practices be-tween the lower and upper grades ofthe models compared, resulting inmore differences between grade levelsthan between models.

"The focus needs to switch from lan-guages to schools," concludes Brisk."Schools, not languages, educate stu-dents."

For Further InformationD. August and K. Hakuta. Improving Schooling forLanguage-Minority Children. Washington, DC: Na-tional Academy Press, 1997.M. E. Brisk. Bilingual Education: From Compensa-tory to Quality Schooling. Mahwah, NJ: LawrenceErlbaum Associates, 1998.V P Collier and W P Thomas. Language Minority

Student Achievement and Program Effectiveness:Research Summary of Ongoing Study. George Ma-son University, 1998; http://www.ncbe.gwu.edu/ncbepubs/resource/effectiveness.National Research Council. Preventing Reading Dif-

ficulties in Young Children,1998; Available from Na-tional Academy Press, 202-334-3313; http://www.nap.edu.R. P Porter. Forked Tongue: The Politics of BilingualEducation. New Brunswick, NJ: Transaction Pub-lishers, 1996.J. Ramirez et al. Final Report: Longitudinal Study ofStructured English Immersion Strategy, Early-Exitand Late-Exit Transitional Bilingual EducationPrograms for Language Minority Children (Vols. 1and 2 prepared for the U.S. Department of Educa-tion). San Mateo, CA: Aguirre International, 1991.C. H. Rossell and K. Baker. "The Educational Effec-tiveness of Bilingual Education." Research in theTeaching of English 30, no. 1 (February 1996): 7-74.

Laurel Shaper Walters is an education writer livingin St. Louis, MO.

Bilingual Education and California's"English for the Children" InitiativeResearchers, practitioners, and policymakers debate the issues

0 n June 2, California voters will decide whether toeliminate bilingual education in their state. Propo-sition 227, also known as the "English for the Chil-dren" initiative or the Unz initiative after the name

of its chief sponsor, proposes replacing current bilingual pro-grams with a maximum of one year of "sheltered English im-mersion" (see "Models of Language Instruction," p. 3). In thiscase, parents would have to explicitly ask for their children tobe placed in native-language programs.

If' roposition 227 passes, it could lead to a seismic shift ineducating the nation's largest population of limited-English-proficient (LEP) students. With 1.4 million LEP students in thestate, California is home to nearly half of the nation's popula-tion of young English learners. In March 1998, the state boardof education gave school districts local control of decisionsabout educating LEP students. However, if Proposition 227passes, it would override any other provisions.

In an effort to advance the dialogue about how best toeducate students who do not speak English, the Harvard Edu-cation Letter asked some key supporters and opponents ofthe California ballot initiative to share their perspectives.

RON K. UNZSilicon Valley entrepreneur, Republican candidate in California's1994 gubernatorial primary, and chief sponsor of Proposition 227.

"The current system of bilingual education just doesn'twork. It may or may not work in theory, but it simply doesn'twork in practice. As far as I can tell, it's never worked anywhereon a large scale in the United States in the last 30 years. Thestatistics are dreadful. Right now in the state of California, a

quarter of all the children in public school don't know English.Of the children who start a given school year not knowingEnglish, by the end of that school year approximately 95 per-cent still don't know English. Assuming that the purpose ofbilingual education is to make sure that children can read andwrite English, it doesn't seem to do a very good job. Of all theimmigrant groups in California, Latino children have the high-est dropout rate, the lowest test scores, and the lowest rate ofadmission to college. So if bilingual education is a good thing,why are the people who are given the most of it doing theworst?"

MIKE HONDADemocratic assemblyman in the California legislature, former highschool teacher and principal.

"This proposition is instructionally unsound and unten-able. It's an untested experiment. Each local school districtshould have the freedom to decide how to teach their limited-English-proficient (LEP) children. The Unz initiative says:"Thou shalt teach only one way." The school districts will haveno authority and the teachers will be mandated to one meth-odology. Unz says that most LEP youngsters are failing. Butwhen you look at the 1.4 million students who are limited-English-speaking, only 30 percent are receiving some sort ofnative-language instruction. Seventy percent are receivingEnglish-only instruction. If you say the LEP population is fail-ing, then its important to realize that the majority of LEP stu-dents are in English-only programs. This is what Proposition227 wantsfailure. Trying to solve a complex issue throughthe simplistic initiative process leaves things up to a lot of

4 The Harvard Education Letter, May/June 199821-

rhetoric, promises, and intent. The real picture is that we don'thave enough qualified teachers to teach youngsters who havelanguage needs. And we don't put enough money behindteacher training."

ROSALIE PEDALINO PORTERDirector of the READ Institute in Amherst, MA, a national organi-zation that helps school districts implement English-language pro-grams. She coordinated the bilingual education programs inNewton, MA, for 10 years.

"This initiative is an attempt to correct a situation in Califor-nia that has been counterproductive for too many years. Thescale is tipped too far in the direction of native-language in-struction that goes on for five to seven years for a lot of chil-dren. To finally have a preference for English-language pro-grams would turn things around in a very important way.However, I don't believe that the one-year limit is a good ele-ment of the initiative. I think two years would have been muchbetter. For 11 years, the California legislature has not been ableto pass a bill clarifying what should be done for language-mi-nority children. Finally, the Unz initiative comes along andcitizens are going to be able to vote on what kind of educationthey think will serve these children best. It needn't have cometo this, but one of the reasons it has is the insistence by thestate's Department of Education on enforcing native-languageprograms, even after the bilingual education law expired. Cer-tainly policy should be made by law and edict of educationdepartment officials. But when these people abdicate theirresponsibility, then somebody like Ron Unz comes along andsays, 'Let's get the people to vote on this.' Bilingual programsdon't produce better results. Is it worth the effort and themoney to put kids into separate programs for five to sevenyears to accomplish the same results as when kids are main-streamed within three to four years?"

MARY T. CAZABONResearcher and director of bilingual education programs for Cam-bridge, MA, public schools.

"From my personal experience of over 20 years in bilingualeducation, I would never make the recommendation that theonly way to teach English-language learners is to give themonly one year of English immersion. I believe that educationfor limited-English-proficient students is not a one-size-fits-allapproach. Efforts should focus on the best ways of servingstudents rather than recommending an overly simplistic for-mula that dictates one year of specialized instruction. We needfirst to ensure that all our students receive exemplary teaching.Then, we need to define what type of program fits their needsand what their parents want for them. That could even includea one-year immersion experience in English. But the key is tohave a variety of viable options open to parents and students."

CHRISTINE ROSSELLProfessor of political science at Boston University, and author ofBilingual Education in Massachusetts: The Emperor Has No Clothes.

"When I look at the empirical research on the subject, itshows that bilingual education is a little bit worse than theregular classroom and a little bit worse than a structured-im-mersion classroom, which is what the Unz initiative proposes.I've gone into classrooms where, after three years of bilingualinstruction, the teacher is still teaching in Spanishthe kidsnever really get to English. If you limit them to one year, then

the danger of this happening is minimal. Research shows thatkids get stuck in structured-immersion programs, too. Theharm of being in a self-contained classroom consisting just ofLEP kids would seem to be less if they're being taught in Eng-lish than if they are being taught in their native tongue. Exceptthat you can imagine how in U.S. schools, where newcomersarrive every day of the year, including the last, that these kidscould get slowed down by the newcomers if they are there formore than a year. Although there's no research suggesting thatone year is sufficient, it's common sense. Remember, mostimmigrant children go immediately into a regular classroomand they do better than those in bilingual education. The limitin the Unz initiative is not on special helpthey can receivethat their whole school careerthe limit is on being in aself-contained, segregated classroom of only LEP children. Ofcourse there needs to be more good research. But if you waitfor research before you do anything, it could be the 22ndcentury. I think you have to go with a certain amount of com-mon sense. American teachers will help a kid who doesn'tknow English. They're not going to give him Fs."

KENJI HAKUTAProfessor of education at Stanford University, and author of Mirrorof Language: The Debate on Bilingualism.

"The Unz initiative goes against what we know about bilin-gual education programs. If you compare bilingual programswith English-only programs, kids in bilingual programs showbetter outcomes. Granted, it's not a huge difference, but it'smeasurable by even fairly crude research methods. What both-ers me the most about the Unz initiative is the very recklessuse of information, as well as open abuse of it. The figure ofa 95 percent failure rate in bilingual education is based on veryloose use of information. For example, those numbers don'ttake into account such things as this being a highly mobilepopulation. There's really no research that says that the ma-jority of kids will learn English in one year. Also, the Unz in-itiative is not written in a way to address academic content; it'sall geared to English acquisition. This initiative certainly wouldput kids on hold for a year. Although many bilingual programsare undoubtedly poorly implemented, there are also manywell-implemented ones that have had tremendous success inhelping LEP students both learn English and advance academi-cally. Despite these successes, bilingual education is not apanacea that will fix all the inequities that language-minoritychildren come into school withsocial inequities that have todo with immigration, poverty and so forth. Language of in-struction is only one of the many factors involved in goodschooling. Other factors include well-trained teachers, a bal-anced curriculum, systematic student assessment, parent in-volvement, and a supportive school climate, to name a few"

ROBERT L. FRENCHSuperintendent of the Orange County Unified School District, whicheliminated all bilingual classrooms for 1,400 students in September1997 after obtaining a waiver from the California State Board ofEducation.

'We sought a waiver because we were getting more andmore students in bilingual education and fewer and fewerwere successfully leaving. In addition, 4th graders going intothe total English program were not doing very well. Somebilingual education programs work if you have outstandingteachers, quality materials, and an educationally sound pro-gram. The question is: "Is there enough time spent on English

22The Harvard Education Letter, May/June 1998 5

skills?" We designed a program that we think is better. We bringin bilingual aides who rotate from classroom to classroom tomake sure the students understand what is being said. Teach-ers in kindergarten and first grade are saying that students whocannot speak a word of English in September are speakingsome English after three weeks. Our program doesn't meetthe Unz initiative. I believe it's a poor piece of legislation. Youhave to do it the Unz way. I'm not for it or against it, but wethink our program is better."

WALDEMAR ROJASSuperintendent of the San Francisco Unified School District, where6% of the 19,000 LEP students are enrolled in bilingual educationclassrooms. He is the first Latino president of the Council of GreatCity Schools.

"Proposition 227 is a one-size-fits-all approach, and one sizedoesn't fit all. We don't need an initiative or a legislativeagenda to tell us how to teach our children. This initiative is

devoid of research, of statistics, and of logic. Because of theshortage of bilingual teachers, many students are already inEnglish-only programsonly 30 percent of California LEP stu-dents are in bilingual programs. If there is any place wherethere is failure, it is much more in the English-only model thanthe bilingual model. This initiative would reinstate the leasteffective model. It's just not doable by most kids. Some kidswill learn some English, of course. They'll know playgroundEnglish, but let them go in and try to read a biology textbookwith that and see how well they perform. This initiative setsus back 30 years. We think it's absolutely in violation of Lau vNichols (the 1974 U.S. Supreme Court decision) and it makesa mockery of public education. We believe we can get thisdefeated because it doesn't make any sense, but otherwisewe'll go in showing the research and ask for a restrainingorder."

Paying Attention to ADHDSuccessful classroom management means training teachers to better servestudents with ADHD

BY LEON LYNN

0 ver the past decade, thenumber of U.S. schoolchil-dren diagnosed with Atten-tion Deficit / Hyperactivity

Disorder (ADHD) and its related form,Attention Deficit Disorder (ADD), hasincreased dramatically. At the sametime, recent changes in federal lawsmandate that schools provide extrasupport to ADHD students in bothregular and special education classes.

Much debate has focused onwhether society has been too quick tooverdiagnose and medicate children,most commonly with a stimulant calledRitalin. Many educators "are not reallyaware of the condition and how to di-agnose it," says Edward Hallowell, aninstructor in psychiatry at the HarvardMedical School. "There are placeswhere half the class has been diagnosedwith ADD, or school systems where it'slooked upon as the plague and no oneis diagnosed with it."

Some educators worry that the sharpupswing in diagnoses is due toovereagerness by health-care profes-sionals to apply the ADHD label to chil-dren deemed unruly or disruptive. Forexample, Thomas Armstrong, an edu-cation consultant and former special-

education teacher, fears that schoolsand parents sometimes put children ondrugs instead of examining whetherchanges in classroom management andcurriculum might improve their behav-ior and performance. "Are so-calledADD kids abnormal because they havetrouble concentrating in a linear, mo-notonous, stimulus-poor, assembly -line classroom?" Armstrong asks.

Behavior modificationtechniques should beused as part of acomprehensive programfor ADHD students.

Others believe the upswing in diag-noses is due instead to greater aware-ness of ADHD among parents and edu-cators. "Ten years ago, probably 80percent of the children with ADHDweren't getting diagnosed," saysRussell Barkley, director of psychologyat the University of Massachusetts Medi-cal Center. "There are lots of anecdotestold about children being diagnosed

too quickly, but there is no evidencethat we have a national problem."

While some educators still expressconcern about overdiagnosis of ADHD,clearly there is increased awarenessabout the effects of ADHD on students'learning. What's more, changes in fed-eral laws have prompted many schooldistricts to focus on how to best serveADHD students in the classroom.

Focus on the ClassroomChildren with ADHD are easily dis-

tracted, have trouble controlling theirimpulses, and are often forgetful. Fre-quently, they have trouble staying stillfor even short periods of time. Theyoften disrupt the classroom by callingout, by getting out of their seats at in-appropriate times, or even by physicallyaccosting their classmates and teachers.Typically, these students also have diffi-culty focusing on the teacher, remem-bering instructions, and finishing theirassigned work.

While many kids may demonstrateADHD-like behaviors, "the differencefor ADHD students is that these behav-iors are debilitating," says George Du-Paul, an associate professor in theSchool Psychology Program at Lehigh

6 The Harvard Education Letter, May/June 1998

About ADHDBetween 3 and 5 percent of U.S. schoolchildren (1.5 million to 2.5 millionstudents) have been diagnosed with Attention Deficit/Hyperactivity Disorder(ADHD). Another 5 percent of students are not hyperactive, but still suffer fromAttention Deficit Disorder (ADD)now considered a form of ADHD.The number of U.S. school-age children designated "other health impaired"acategory that includes ADHDhas grown by 89 percent since 1990. In 1991,the U.S. Department of Education decreed that students with ADHD wereentitled to special education services.About 10 percent of ADHD students are taught in self-contained special edu-cation classrooms, while the rest spend at least part of the day in regularclassrooms.Most children with ADHD appear to have inherited the disorder. About 20percent of those diagnosed have suffered brain damage due to causes such asinjuries, lack of oxygen during birth, and prenatal exposure to drugs or alcohol.Boys are at least four times as likely as girls to be diagnosed with ADHD; someresearchers think this means that girls simply aren't diagnosed with ADHD asoften as they should be.More than 20 percent of ADHD students have been held back in school at leastone grade. Between 15 and 25 percent ofADHD students have been suspendedor expelled from school due to behavior problems. Up to 35 percent of ADHDstudents may never complete high school.

Sources: U.S. Department of Education, Russell Barkley, Linda Pfiffner.

University "If children are working in aclassroom and something interesting ishappening outside the window, mostkids will glance at it and go back towork," he says. "But a student withADHD won't be able to delay respond-ing to the distraction. They can't delayresponse to stimulus. That's what getsthem into trouble in school."

Simply medicating students whohave ADHD isn't enough, experts say.Only half of the students who take Ri-talin show any improvement in aca-demic performance, according to Du-Paul. Linda Pfiffner, director of theHyperactivity, Attention, and LearningProblems Clinic at the University of Chi-cago agrees: "That's one reason not touse medications in isolation. The stu-dent needs a complete program of in-tervention."

Behavior modification techniquesare increasingly being used as part of amore comprehensive program for stu-dents with ADHD. Recent research anda growing body of classroom experi-ence have identified techniques thatteachers and other staff can use to helpADHD students stay focused and con-nected to classroom activities (see"Classroom Strategies, p. 8"). Some ofthese techniques require only minimalnew effort by teachers. While some

have been proven to work, more re-search is needed. "There isn't an ade-quate number of outcome studies,"says Pfiffner. "The problem is trying toseparate out the influences of one partof an overall treatment program."

One District's ProgramIn Kenosha, WI, Kathy Hubbard

oversees a teacher-education programthat many experts and researcherspoint to as a model. About 5 percent ofthe district's 20,000 students have beenidentified as having ADHD; 70 percentof them are taught in regular class-rooms. The program includes an exten-sive teacher-training program, staff inevery school building to work withteachers and parents, and "attentiondeficit disorder education plans" forevery student identified with ADHD.Hubbard, a clinical social worker, be-gan the program in part because herson has ADHD. School leaders, some ofwhom also have children with ADHD,have given her strong support.

At the center of the program is a 16-hour teacher-training course. Thecourse covers such topics as how tospot children who have ADHD and howto approach parents and other profes-sionals about having a child evaluatedfor services. Teachers learn behavior

management theory and techniquesthat do not detract from instruction forother students, as well as how to workwith parents to reinforce these tech-niques at home. They also learn aboutcurriculum adjustments that help kidswith ADHD succeed in the classroom,and about the potential side effects ofcommon ADHD drugs and the limits ofdrug therapies. At present, 900 of thedistrict's 1,250 teachers have com-pleted this course, Hubbard says.

A big part of Hubbard's job is raisingteacher awareness about ADHD. Manyteachers, for example, are unaware thatthe condition often persists in childrenthrough high school. ADHD kids areoften harder to spot at that age becausethey aren't as hyperactive, Hubbardsays, "and so teachers can think that thekids are just lazy or undisciplined."Sometimes teachers, especially in oldergrades, "just don't want to go the extramile for these students because theydon't have to do it for the others," shesays. "But the longer we do this pro-gram, the more teachers are comingaround. They see that there are ways towork with ADHD students that are lessstressful for everyone. Then we winthem over."

Teachers in Kenosha say the traininghas helped them in their jobs. "If theteacher acts appropriately, things canbe a lot better," says Jolene Schneider,a 7th-grade English teacher. "And if youhave the parents' support, you can bepretty effective."

Beth Luckhardt, a special educationteacher, found that putting a containerfilled with rubber balls and joiningblocks on the teacher's desk has helpedseveral ADHD children sit still in a regu-lar classroom. "Now those studentsknow they can play appropriately withtheir hands when they need to, insteadof doing other things that disrupt theclass," she says.

Sharon O'Brien, a social worker inthe Kenosha schools, has also seen first-hand how simple techniques can payoff. She cites the case of 6-year-oldJamie (not his real name), who fre-

COMING SOON

Full-Service Schools

How After-School JobsAffect Student LearningHispanic Achievement

4- '4The Harvard Education Letter, May/June 1998 7

Classroom Strategies for Working with ADHD StudentsHere are some techniques recommended by experts and teach-ers who work with students who have ADHD. While all studentscan benefit from these techniques, those with ADHD often can'tsucceed academically without them, experts say.

Don't Thy to Handle Things Yourself If you suspect a studenthas ADHD, seek help from school personnel with specific train-ing in the disorder, such as counselors, social workers, psycholo-gists, and special education teachers. Working with parents isalso critical to developing a consistent, effective plan.

Minimize Distractions Seat ADHD students near your deskor in the front row. Have all students put away unnecessary itemsduring work times. Designate special places for tools, materials,and books, so everyone knows where everything belongs.

Create a Clear Structure Students with ADHD do best in astructured classroom, where expectations, rules, and routinesare well-defined and communicated. Spell out directions clear-ly, and remember that ADHD students might not rememberthem all.

Consider Adjusting Your Curriculum Break down largetasks or assignments into smaller, less complex units, and giveADHD students positive reinforcement for finishing each part.

Pay Special Attention to Thansitions Stop and discuss thespecific behaviors you want children to think about as theymake their way to the next activity (lunch, recess). Spell out aclear incentive for cooperating, as well as the consequences for

breaking the rules or being disruptive. Try to intercede beforetrouble begins.

Start a Daily Progress Report A daily report that awardspoints for good behavior for a few concrete goals can be helpfulin motivating ADHD students and in communicating with theirparents.

Create a "Tbken Economy" Pick a specific goal for an ADHDstudent, such as turning in all homework or behaving well inclass for a set period of time, and provide a concrete reward formeeting that goal. Reward or penalize the child quickly, so theymake a dear connection between their behavior and the results.

Provide a Specified Throe -Out Area This should not be seenas a punishment, but as a place where the student can go to calmdown if necessary. Older students can be taught to sense whenthey are getting out of control and go to the time-out area ontheir own.

Balance Discipline with Chances for Success Teachers canestablish some form of signal (touching their ear or the student'sdesk) that reminds the ADHD student that they need to get backon task without embarrassing them. Also, look for ways the childcan experience success or make use of special talents.

Be Prepared to Change Strategies Be ready to try a newtechnique if an old one isn't working anymore. Sometimes olderstudents have good insight into something they tried in the pastthat might work again.

quently disrupts his first-grade class,sometimes by hitting and biting hisclassmates. Currently Jamie is respond-ing well to a "remembering bracelet," aMickey Mouse watch he's invited towear as a reminder to be kind to others."If he acts out, he has to take it off, be-cause it's not working," O'Brien says."But he can put it back on as soon as hethinks it might work again."

Kenosha's program costs relativelylittle to operate, says Hubbard. The dis-trict pays her salary as program consult-ant and spends perhaps another $2,000per year on her continued training.Having a full-time person on staffmeans the district avoids spending

hiring outside consultants toconduct trainings, she says. What'smore, "we're saving money becausewe're evaluating kids early, we can treatthem appropriately, and prevent un-necessary placements" to separate spe-cial education classrooms. Although noformal research study has been done todetermine whether the students inKenosha are better off since the pro-

Need for Teacher TrainingWhile Kenosha's program is consid-

ered extraordinary, it is not unique. In1997, the University of Kentucky's Fed-eral Resource Center conducted a na-tional search for school-based practicesholding promise for ADHD studentsand selected 25 examples, includingKenosha. From San Diego, CA to Tow-,

son, MD, teachers, staff, and parents areparticipating in training and other pro-

. grams to better identify and serve stu-dents with ADHD.

These programs are part of a neces-sary trend, since most kids with ADHDspend their time in the regular class-room, say ADHD experts. "I fully appre-ciate that teachers have a lot of de-meads placed on them," says Barkley."But at the same time, public schoolsare growing increasingly inclusive. Wecan't have it both ways. We can't saywe're going to teach these children inthe regular classrooms and then nottrain the teachers to do it."

For Further Information

B. G. Burcham and L. B. Carlson. "Attention DeficitDisorder: School-Based Practice." Federal ResourceCenter, University of Kentucky 1997.Children and Adults With Attention Deficit Disorder(cnAnn), 499 Northwest 70th Avenue, Suite 101,

chadd.org.G. DuPaul, K. McGoey, and T Eckert. "Interventionsfor Students with Attention-Deficit/HyperactivityDisorder: One Size Does Not Fit All." School Psychol-

o.' Review 26, no. 3 (1997): 369-381.E. Hallowell and J. Raley. Driven to Distraction: Rec-°grazing and Coping with Attention Deficit Disor-der from Childhood Through Adulthood. New York:Pantheon, 1994.E. K. McEwan. The Principal's Guide to AttentionDeficit Hyperactivity Disorder. Thousand Oaks, CA:Corwin Press, 1998.K. Hubbard, ADHD consultant, Kenosha School Dis-trict, 3600 52nd St., Kenosha, WI 53144; 414.6536332.

The Merrow Report. 'Attention Deficit Disorder: ADubious Diagnosis." PBS Documentary 1995.National Attention Deficit Disorder Association(ADDA), PO Box 972, Mentor, OH 44061-0972; 800 -487.2282; http: / /wwwadd.org.L. Pfiffner. All About ADHD: The Complete PracticalGuide for Classroom Teachers. New York: Scholas-tic, 1996.

S. Rid. How to Reach and Teach ADD/ADHD Chitdren. West Nyack, NY: Center for Applied Researchin Education, 1993.

gram began, Hubbard feels the district T. Armstrong. The Myth of the ADD Child. New York: Leon Lynn is a freelance writer living in Milwaukee,is in a better position "to nip the emo- Dutton, 1995. WI. He can be reached at LeonLynn@com-tional and learning problems of atten-tion deficit disorder in the bud."

R. Barkley. Taking Charge of ADHD: The Complete,Authoritative Guide for Parents. New York: Guil-ford, 1995.

puserve.com.

1/

8 The Harvard Education Letter, May/June 1998

bs

Volume XIV, Number 4 Published by the Harvard Graduate School of Education July/August 1998

Working Teenagers: Do After-School Jobs Hurt?High schoolers who work more than 20 hours a week may be at higher risk for failure

BY KAREN KELLY

Two decades ago, politiciansand educators extolled the vir-tues of part-time jobs for highschool students as a way to fos-

ter independence, responsibility, andgood work habits. Government panelslike the 1980 National Commission onYouth also praised part-time work, sug-gesting it was "the single most impor-tant factor" in the transition from youthto adulthood.

Only a year before the Commission'sreport was published, however, the firstof several warnings appeared, challeng-ing the prevailing view that high schoolstudents were benefiting from theirafter-school jobs. That warning camefrom Temple University researcherLaurence Steinberg, who, in a 1979 re-port, concluded that after-school workhad a detrimental effect on schoolachievement.

INSIDE

Do Test Scores SufferBecause of Work? / 3

Full-Service Schools / 4

Discussing StudentWork / 6

While Steinberg's workamong themost frequently cited in the fieldhasgenerally withstood the test of time, heand other researchers are beginning tofocus on the 30 percent or more of stu-dents who work more than 20 hoursduring the schoolweek. Growing evi-dence suggests thatteenagers who workmore than theirpeers after schoolmay be more likely tohave been disen-gaged from schoolbefore they took ajob. Since all jobs arenot equal, however,the search is on to de-fine the most optimalwork experience forteens.

These inquiriesare not merely aca-demic. Researcherswho have surveyedhigh schoolers onhow much they workestimate that one-third are working onany given day andthat more than 80percent hold a pay-ing job at some pointduring their high

600

500

400

school career. Data released this year bythe Third International Math and Sci-ence Study shows that U.S. 12th-grad-ers work at a far higher rate than theircounterparts in the 20 other countriesthat participated in this test (see chart).

Sweden Norway U.S.Netherlands Iceland Switzerland

I. Math scores

Science scores

Percentage of students working3 or more hours per day

100

80

60

40

20

0

National average scores on the Third International Mathe-matics and Science Study (TIMSS) of students in their finalyear of secondary school, and the percentage of students ineach country who reported working three or more hours aday. Chart compares scores of students in the top five scoringcountries with those in the United States.

EDITORIAL DIRECTOR: Kelly Graves-Desai. EDITOR: Nancy Walser. PRODUCTION EDITOR: Dody Riggs. EDITORIAL BOARD, HARVARD GRADUATE SCHOOL OFEDUCATION: Mildred Blackman, Director, The Principals' Center; Sally Dias, Superintendent, Watertown Public Schools, Watertown, MA; Jay P Heubert, AssistantProfessor; Harold Howe II, Senior Lecturer Emeritus; Susan MooreJohnson, Professor and Academic Dean; Robert Kegan, SeniorLecturer; Jerome T. Murphy, Professorand Dean; Gary A. Orfield, Professor; Robert S. Peterkin, Senior Lecturer; John Ritchie, Superintendent/Principal, LincolnSudbury Regional High School, Sudbury, MA;Judith D. Singer, Professor; Jay Sugarman, Teacher, Runkle School, Brookline, MA; Dennie Palmer Wolf, Lecturer on Education. NATIONAL ADVISORY BOARD: JohnBrademas, President Emeritus, New York University; Constance E. Clayton, former Superintendent, School District of Philadelphia; Alonzo A. Crim, Professor ofEducation, Spelman College; Linda Darling-Hammond, Professor, Teachers College, Columbia University; Andrew Heiskell, Chairman Emeritus, NewYork Public Library;Marya Levenson, Superintendent, North Colonie Central Schools, NY; Deborah Meier, Founder, Central Park East Secondary School, NY; John Merrow, President, TheMerrow Report; ArthurJ. Rosenthal, Publishing Consultant.

f3 BEST COPY AVAILABLE

A :RA

While many parents and educatorshope a part-time job experience willteach students how to manage theirmoney, research findings suggest other-wise. Steinberg found that only 11 per-cent of students reported saving mostof their money for college, and only 3percent contributed earnings towardtheir family's living expenses.

Unlike young immigrants who hadto help pay the family's rent in the earlypart of the century Steinberg says teensworking in jobs today are mainly mid-dle class, suburban, and whitea find-ing confirmed by other studies. Thebulk of their income goes to clothing,cars, entertainment, and, in somecases, drugs and alcohol.

Negative Effects of WorkAs one of the first to investigate the

effect of part-time jobs on academicsuccess and aspirations, Steinberg hasexamined indicators such as academicachievement, class attendance, timespent on homework, and attitudestoward school among both workingand non-working students. In mostcases, he says, the working student is ata disadvantage, with negative effects in-creasing with the number of hoursworked.

"Students who worklonger hours report di-minished engagementin schooling, loweredschool performance, in-creased psychologicaldistress, higher drug andalcohol use, higher ratesof delinquency, andgreater autonomy from parental con-trol," says Steinberg.

For example, in a 1991 survey of4,000 high schoolers in California andWisconsin, the more hours worked, thesteeper the drop in grade point aver-ages and amount of time spent onhomework. At the same time, studentsworking more than 20 hours reportedusing drugs and alcohol 33 percentmore often than their non-employedclassmates. They also experiencedgreater psychological symptoms suchas anxiety, depression, and fatigue. Theemerging consensus among re-

searchers, says Steinberg, is that thenegative effects of employment arelinked to how much, not whether, a stu-dent works.

This view has been confirmed mostrecently in a 1997 study by David Stern,director of the National Research Cen-ter for Vocational Education at UCBerkeley. Stern looked at the body ofresearch conducted on working teensover the past 20 years, separating thesurveys into two groups: one that in-cluded students who worked fewerthan 15 hours a week, and one includ-ing those who worked more.

In this latter group, he found 10studies that reported students working15 hours or more had lower grades, didless homework, had higher dropoutrates, and were less likely to go to col-lege. Only three found no negative ef-fects. "The preponderance of evidenceindicates that students who work morethan 15 to 20 hours a week while inhigh school perform less well academi-cally," he concluded.

Is Work a Symptom?At first glance, Stern says, it may ap-

pear that working long hours causesstudents to earn lower grades, but he

stops short of makingthat claim: "Is it a ques-tion of cause and effect,or just some kind of spu-rious correlation?" heasks. "It could well bethat this is just a selectionprocess that's going on,where some kids areheading off to work and

putting school behind. It may be thatthose who are more interested in work,work more."

That's a suspicion shared by re-searcher Jerald Bachman, an investiga-tor with the University of Michigan'sMonitoring the Future Project, whichsurveyed 70,000 high school seniors.Bachman wondered if the students get-ting into trouble after working longhours were the same students whowould run into problems at school any-way. He looked at their grades beforeentering the work force, their plans forcollege, and whether they had ever

Kids need timeto talk abouttheir on-the-jobexperiences.

been held back a grade. He found stu-dents with low GPAs, no college plans,and a record of retention were morelikely to choose a job with longerhours.

"I would argue that most of the prob-lems that correlate with working longhours are more fundamentally caused,"says Bachman. "That may contribute tothe spiral, but I think that the spiral iswell underway at the time they elect towork the long hours."

In his earlier studies, Steinberg didnot consider the prior academic recordof students. But for a 1993 study, heasked the students to recall their GPAs,their educational plans, the amount oftime they spent on homework, andtheir interest in school before they tookpart-time jobs. He discovered that theadolescents who worked longer hourswere less academically inclined to be-gin with, and concluded that their dropin GPA was not necessarily the result oftheir work schedule. In addition, one-third of the working adolescents he sur-veyed admitted to taking easier classesin order to protect their grade point av-erage.

Not All Work Is BadBachman and other researchers are

not willing to write off part-time jobsaltogether. They say the most successfuljob experiences are those that areclosely linked to an official school pro-gram. Often called School-to-Work orWork-Based Learning (see HEL, March/April 1997), these partnerships be-tween school and industry provide stu-dents with learning experiences in areal-world atmosphere that can, inturn, strengthen their academic under-standing.

The majority of teenagers are notworking in this type of job, says Stern,but there is evidence that those who arehave a more positive attitude aboutwork and fewer academic problemsthan those whose jobs are not linked toa school program. A successful school-to-work program "must be carefullyplanned and monitored by people whounderstand both the work setting andwhat is to be learned there," says Stern.

In fact, some researchers have found

THE HARVARD EDUCATION LETTER (ISSN 875513716) is published bimonthly by the Harvard Graduate School of Education, Appian Way, Cambridge, MA 02138.3752. Second-class postage paidat Boston, MA, and additional mailing office. POSTMASTER: Send address change(s) to The Harvard Education Letter, 6 Appian Way, Cambridge, MA 02138-3752.

Signed articles in THE HARVARD EDUCATION LETTER represent the views of the authors. Address editorial correspondence to Editors, The Harvard Education Letter, Gutman Library, 6AppianWay, Cambridge, MA 02138; phone 617-495-3432; fax 617-496-3584; email address: edletter@hugsel.harvard.edu; web site: http://hugsel.harvard.edu/hepg/hel.html

1998 by the President and Fellows of Harvard College. Published as a nonprofit service. All rights reserved. Special written permission is required to reproduce inany manner, in whole or inpart, the material herein contained. Call 617-495-3432 for reprint permission information.

How to subscribe: Send $32 for individuals, $39 for institutions ($40 for Canada/Mexico, $42 other foreign, in U.S. funds only) to The Harvard Education Letter, 6 Appian Way, Cambridge MA02138; or call us at 617.495 -3432 in Massachusetts or 800-513-0763 outside Massachusetts between 8:30 a.m. and 5:00 p.m. EST Monday-Friday. Subscription prices subject tochange withoutnotice. Single copies, $5.00. Back issues and bulk subscriptions available at special reduced rates; call 800-513-0763. r ti

BEST COPY kiiiiLALI.L

Do U.S. Math and Science Scores Suffer Because of Work?The test scores of U.S. high school students in math and

science are lagging behind those of other countries, but so far,researchers have not been able to lay the blame on any onefactor. Class size, age, gender, curriculumnone of thesealone can adequately explain the difference in scores betweenhigh- and low-performing countries.

But Albert Beaton, a professor at Boston College and theinternational study director of the Third International Mathand Science Study (TIMSS), has a working theory. At the Ameri-can Educational Research Association conference held inApril, he pointed to after-school jobs as a possible culprit: U.S.12th-graders work at a far higher rate than their global peerswho took the TIMSS test.

"Not only do our kids work their way through college, butour kids seem to be working their way through high school,"says Beaton. Pointing to statistics that show one-half of U.S.students working three or more hours a day outside of normalschool hours, he said, "That is very different from other placesin the world, and I think we're going to have to look at that."

In fact, the findings are dramatic. Fifty-five percent of U.S.12th-graders work more than three hours a day at a paid job,compared to an international average of only 18 percent. Incontrast, three-quarters of the international students reported

working less than one hour a day.Temple University professor Laurence Steinberg devoted a

chapter to this phenomenon in his 1996 book, Beyond theClassroom: Why School Reform Has Failed and What ParentsNeed to Do, noting that student employment is so rare in othercountries that it's actually difficult for researchers to get anaccurate measurement.

"Most European and Asian high school students would findit utterly astonishing that their American counterparts havefour or five hours of free time each day to devote to an after-school job," wrote Steinberg. 'And most would find it incred-ible that the average American teenager spends only about onehour per day on homework."

Steinberg adds that the United States "is the only countryin the world that actively encourages large numbers of college-bound high school students to take on jobs during the aca-demic year."

But the correlation between after-school jobs and low U.S.TIMSS scores is still a hypothesis. The authors of the TIMSSreport warn, "It is too early in the process of data analysis toprovide strong evidence to suggest factors that may be relatedto the patterns of performance."

KAREN KELLY

students in work-based learning pro-grams perform better academicallythan students in the traditional highschool program. Larry Rosenstock, for-mer director of the New Urban HighSchool Project, studied work-basedlearning programs in 23 cities. Hefound a big spike in college entrancerates among the students involved inthese programs.

"In Durham, NC, the mayor took meto visit the health careers program at alocal high school," explains Rosen-stock. "There, we discovered 90 per-cent of the kids in work-based learningwere going to college, compared to 70percent in the high school at large."

Most schools don't have the re-sources to implement such a program,however, and if they do, they're oftennot sure where to start, according toCarol Clymer, a senior program officerat Public/Private Ventures in Philadel-phia, a nonprofit organization thathelps high schools start work-basedprograms. One of the key componentsin work-based programs, says Clymer,is the opportunity for working teens totalk about their on-the-job experiences.Kids often can't articulate what they'relearning at work, so they need time tothink about it and talk about it in orderto identify the positive components,she says.

While her organization hasn't con-ducted rigorous research on the sub-ject, Clymer says her observations sug-gest that many working teens arelearning about responsibility, team-work, and communication on the job.

But in the absence of true school-work partnerships, researchers agreethat there's a need for parents to restrictthe number of hours a student works,especially if their academic achieve-ment is suffering.

Federal and state governments settheir own guidelines: under federal law,students under 16 cannot work morethan three hours on a school day and18 hours in an entire week. The govern-ment has not set guidelines for 16- and17-year-olds, although several stateshave placed their own restrictions onolder teens.

Stern would like to see governmentsinstitute a permit process, whereby stu-dents would have to have a certain GPAbefore they could get a job. Bachmanagrees that more supervision could behelpful, but he contends that re-searchers, educators, and parents mustbegin to identify the characteristics ofthe ideal workplace experience. Hesays this is crucial, given the likelihoodthat school-year employment will re-main an important part of the life ofmany, if not most, adolescents.

For Further InformationE. Bedenbaugh. "Competing for Time: School andTeenage Employment." NASSP 76, no. 549 (January1993): 74-81.J. Bachman and J. Schulenberg. "How Part-TimeWork Intensity Relates to Drug Use, Problem Behav-ior, Time Use, and Satisfaction Among High SchoolSeniors: Are These Consequences or Merely Corre-lates?" Developmental Psychology 29, no. 2 (March1993): 220-235.R. McNeal. 'Are Students Being Pulled Out of HighSchool? The Effect of Adolescent Employment onDropping Out." Sociology of Education 70, no. 3(1996): 206-221.G. Ruscoe. "Students Who Work." Adolescence 31,no. 123 (Fall 1996): 625-633.L. Steinberg. Beyond the Classroom:Why School Re-

form Has Failed and What Parents Need to Do. NewYork: Simon & Schuster, 1996.L. Steinberg. "Negative Impact of Part-Time Work onAdolescent Adjustment: Evidence from a Longitudi-nal Study" Developmental Psychology 29, no. 2(March 1993): 171-180.L. Steinberg and S. Dombusch. "Negative Correlatesof Part-Time Employment during Adolescence: Rep-lication and Elaboration." Developmental Psychol-ogy 27, no. 2 (March 1991): 304-313.D. Stem. "What Difference Does it Make if Schooland Work are Connected? Evidence on CooperativeEducation in the U.S." Economics of Education Re-view 16, no. 3 (June 1997): 213-229.D. Stem. "The Continuing Promise of Work-BasedLearning." Centerfocus, no. 18 (November 1997).

Karen Kelly is a freelance education reporter basedin Albany, NY

BEST COPY AVAILABLEThe Harvard Education Letter, July/August 1998 3

Full-Service Schools Respond to Families' NeedsQuality of collaborations is a major factor in their successBY PEGGY FARBER

To New York City principalGinny Connelly, the rise of full-service schools means there isa safe place for her students to

go after school to extend what they'velearned during the day.

To Ann Aimes, a Kentucky school of-ficial, it means parents have some-where to turn for emergency helpwhen food runs short.

To Larry Bilick, a parent in Berkeley,CA, it means his school can begin tooffer the basic health and recreationservices his district used to provide be-fore the passage of Proposition 13wiped out extracurricular "frills" in hisstate in 1978.

Communities across the country areestablishing school-based human serv-ices programs in record numbers tohelp poor and middle-class familieshandle an array of contemporary socialproblems caused by everything fromwelfare cutbacks to the rise of singleparenting and dual-career families.

The schools go by a variety of names,including wrap-around schools, com-munity schools, and extended schools.Still, whether in California, New Jersey,Kentucky, or New York City, full-serviceschools have several features in com-mon:

They stay open and active late, onweekends, and during vacations.

They provide an unparalleledrange of human serv-ices. In various combi-nations, the followingexist in full-serviceschools across thecountry: psychologi-cal counseling, pri-mary health care, den-tal care, after-schoolrecreation, one-stopcase management,homework help, job-training, adult literacy classes, parent-ing seminars, clothing and food banks.

They bring into schools agenciesthat until now have not worked directlywith the public education system, aswell as professionals whose access tochildren has been hampered by schoolcalendars and locations. The UnitedWay, the Children's Aid Society, the New

Jersey Department of Human Services,and the Missouri Department of MentalHealth are all examples of agencies thathave recently established school-basedprograms.

They force unprecedentedandsometimes difficultcollaborationsbetween school professionals and lead-en of outside organizations.

Only a few schools, notably the fourChildren's Aid Society/New York CityBoard of Education schools in NewYork, offer a truly complete package ofhealth and human services to studentsfrom sunup to sundown. But the term"full-service school," popularized by re-searcher and author Joy Dryfoos, isused by many people to designate arange of schools engaged in service de-livery and after-school care.

Historical RootsThe concept of full-service schools

can be traced back to the 1930s, whenCharles Stewart Mott, a director of Gen-eral Motors and a wealthy philanthro-pist, decided to do something about re-ports of juvenile delinquency in Flint,MI. With visions of transforming schoolbuildings into neighborhood centers ofrecreation, civic activity, and lifelonglearning, Mott gave money to local pub-lic schools to stay open late. His experi-ment drew national praise and, in1936, Eleanor Roosevelt's personal en-

dorsement.Today's full-service

schools are a responseto a wider range ofchallenges that chil-dren face, and thistime around, federalofficials, state legisla-tors, and social serviceproviders are at the ta-ble alongside the phi-lanthropists. The Clin-

ton administration has put the force offederal funding behind the concept,giving grants this year to 300school/community partnerships thatprovide after-school learning, recrea-tion, and family-support programs.Clinton has asked Congress to increasefunding for these 21st Century Com-munity Learning Centers from $40 mil-

Today's full-serviceschools are aresponse to a widerange of challengesthat children face.

lion this year to $200 million in each ofthe next five years.

Contemporary full-service schools,like the community schools of the1930s, have emerged from local initia-tives, even though they may receive fed-eral funds. "One of the wonderfulthings about this development is thatthere is no single model," says MargotWelch, director of Harvard University'sCollaborative for Integrated SchoolServices, a resource center for counsel-ors and others seeking to bring servicesinto schools. "What's happening is thatthey are growing in communities inways that really allow the people in aneighborhood to choose what is mostuseful to them."

Three Communities, ThreeDifferent Programs

A look at three communities gives anidea of the ways that full-service schoolsare evolving. Junior High School 22 onNew York City's Lower East Side, forexample, is open every weekday from3:00 until 10:00 p.m., enabling about200 neighborhood children to stay anddo their homework, explore the arts,and engage in community actionallwhile establishing strong connectionswith a caring team of adults. The pro-gram is run separately from the schoolby the Grand Street Settlement House.

The relationship between JuniorHigh School 22 and Grand Streetevolved because the school is a Beaconschool, one of 40 in New York City.Funded by the New York City Depart-ment of Youth and Community Devel-opment, the Beacon Initiative spon-sored after-school programs in 10schools, beginning in 1991, during theworst years of urban violence, whensafety was a top priority, according toGeoffrey Canada, a founder of the Bea-con Initiative. As incidents of childrenbeing shot by stray bullets shockedNew Yorkers, Canada and other com-munity activists lobbied the state legis-lature to keep inner-city schools openas lighted buildingsbeaconsoffer-ing children safe passage through ado-lesence.

Ginny Connelly, principal of JuniorHigh School 22, is profoundly gratefulthat her school is a Beacon school.

4 The Harvard Education Letter, July/August 1998

"Even if we have the most whiz-bangwonderful school in the universe be-tween the hours of 8:40 and 3:00,"Connelly declares, "most kids get intotheir first scrapes with the law betweenthe hours of 3 p.m and 6 p.m." If there'sno one to connect with her students inthose hours, she says, "then we're slam-ming our heads up against a brick wall."

At the youth center inside MezeekMiddle School in Louisville, K' staffersembrace a "whatever-it-takes" attitudeto make learning possible for studentswho live in the poorest area in the state.The center's services run the gamutfrom a mental healthclinic, which providesintensive, long-termcounseling to about100 students and theirfamilies, to a supply ofemergency cash, food,and clothing for fami-lies in urgent need.

Mezeek's youth cen-ter is one of hundredsestablished by the Ken-tucky Education Reform Act (KERA) of1990 in schools where at least 20 per-cent of the student population lives be-low the poverty line. Ann Aimes, ahome-school liaison, says the Mezeekcenter has given staffers like herself away to respond to the pressing prob-lems of poverty. "I would talk to kids allday long, but they were just drowning,"remembers Aimes, who was a securityguard at the school before the centeropened. "Now we've got some place torefer kids, some place to give themsome hope."

At Jefferson Elementary School inBerkeley, CA, parents are organizing towin a Healthy Start grant in order toestablish a one-stop social service cen-ter for families who routinely have totravel to separate agencies aroundtown to get help. They also plan an af-ter-school arts and enrichment pro-gram at the school to replace programscut as a result of Proposition 13. UnderCalifornia's Healthy Start initiative, en-acted in 1991, the state awards three-year seed grants to schools to link upwith neighborhood service providers.

Although Larry Mick and other Jef-ferson parents are working hard to wina grant, they are also critical of the statefor using Healthy Start as a way to re-store some services to schools, whilestill relying on local initiatives and fund-ing. Twenty years ago, Berkely schoolshad nurses and free after-school recrea-

tion programs, Bilick points out. Now,in order to qualify for a Healthy Startgrant, "you have to get a ton of in-kindservices," he says. "You get a mentalhealth agency in town to help, a copycenter to do some xeroxing, the localuniversity to provide interns. There arefour other Healthy Starts in Berkeley.These merchants, the YMCAtheycan't do it for everybody and they'restarting to say, 'We've done enough.' "

Obstacles AriseWhile there is almost no published

research yet showing the effect full-service schools have onstudent academic out-

Successfulcollaborationsrequire dynamicschool and agencyleadership.

comes, there is emerg-ing evidence that thequality of the collabora-tions between schoolsand social serviceproviders will be a ma-jor factor in whetherthe full-service conceptcan contribute to learn-ing. A study of Ken-

tucky schools by John Kalafat andRobert Illback has suggested that stu-dents in schools with successfully im-plemented service centers out-performstudents in schools with poorly runcenters on a wide range of measures,including Kentucky achievement tests,teacher evaluations of classroom be-havior, and dropout rates. Kalafat andIllback defined successful full-serviceschools as those that, among otherthings, had functioning advisory coun-cils composed of a wide range of com-munity members, consistently evalu-ated themselves, and explicitly focusedattention on children's educational im-provement.

The Kentucky study underscores atrouble spot that is showing up in otherpreliminary reports. Researchers arefinding that in many schools there is awideand undermininggulf be-tween school people and resource cen-ter staff Principals do not always wel-come the service providers, nor do theyalways grasp what social services can dofor their students.

"Here we use the phrase 'HealthyStart Is Out Back' because so manyschools put the family resources centerin portables out back in the playgroundor on the side where the street is, sopeople could come to the center andNOT come into the school," says MaryWagner, a researcher at SRI Interna-tional, who did the first statewide evalu-

ation of service delivery in the Califor-nia program. "One of things that's keyto which Healthy Starts survive andwhich don't is the extent to which theyare really integrated into the school,"she adds.

Keys to SuccessDryfoos, who recently summarized

the state of the movement in a paperentitled 'A Look at Community Schoolsin 1998," says that successful collabora-tions are extremely "people-depend-ent," requiring a lot of time, a lot ofpatience, and dynamic school andagency leadership.

James Connell, president of the In-stitute for Research and Reform in Edu-cation, a nonprofit education consult-ing firm, calls the gulf between schoolpeople and family service people "atragedy." Connell has looked at dozensof school-based service programs andis an advisor to San Francisco's effort toimplement a New York-style Beaconprogram. He advises bridging the gapby putting social service programs intothe broader context of education re-form efforts. When the call for socialservices comes out of a school's reformefforts, he says, "it's fully integrated intothe school. It's not being plopped inbecause of some grant or somebody'sbrilliant idea that you ought to put serv-ices in schools."

He points to what happened at ClaraBarton Elementary School in Roches-ter, NY, in the early 1990s, when theteachers and principal, working withConnell, instituted sweeping educa-tion reformsrefining academic stand-ards, keeping children and teachers to-gether for several years, and loweringthe teacher-pupil ratio. 'And then theysaid, okay, we need kids to come toschool having been fed, we need to re-spond to the needs of families," Con-nell recalls. At that point, the principalreached out to community serviceproviders to put in place a wellness cen-ter that is treasured by everyoneteachers as well as families, he says.

COMING SOON

Teaching SciencePeer Harassment in MiddleSchoolsLatino Achievement

The Harvard Education Letter, July/August 1998 5

What's AheadThe trend of placing comprehensive

services for families and children inschools is growing vigorously. TheDeWitt WallaceReader's Digest Fundand the Charles Stewart Mott Founda-tion have just committed tens of mil-lions of dollars to a variety of initiativesthat expand the movement to new cit-ies and rural districts. This spring, bil-lionaire businessman and philanthro-pist George Soros announced that hisfirst foray into giving in the UnitedStates will be a $25 million grant to NewYork City for school/community agencycollaborations.

Dryfoos reports that full-serviceschools advocates are beginning towork with school reform advocates in

order to bring the best school/commu-nity models to the national agenda. "It'stime to turn a good idea into an ongo-ing sustained effort," she says, "with na-tional reach and implications."

For Further InformationJ. Connell, M. Gambone, and T Smith. "Youth De-velopment in Community Settings: Challenges forOur Field and Our Approach." Philadelphia: Insti-tute for Research and Reform in Education, 1998;732-288-0066; E-mail: jnirre@aol.com.

J. Dryfoos. Safe Passage: Making It Through Adoles-cence in a Risky Society. New York: Oxford Univer-sity Press, 1998.

J. Dryfoos. 'A Look at Community Schools in 1998."Occasional Paper No. 2, National Center for Schoolsand Communities, Fordham University, 113 West60th St., Room 704, New York, NY 10023; 212 -636-6699.G.Maeroff. Altered Destinies: Making Life Better forSchoolchildren in Need. New York: St. Martin'sPress, 1998.

The Charles Stewart Mott Foundation, 21st CenturyCommunity Learning Centers Office, 1200 MottFoundation Building, Flint, MI 48502-1851; 810-238- 5651; www.mott.org.

The Children's Aid Society, 105 East 22nd St., NewYork, NY 10010; 212-949-4800; wwwchildrensaid-society.org.

Emerging Coalition for Community Schools, Insti-tute for Educational Leadership, 1001 ConnecticutAve., NW, Washington, DC 20036; 202-822-8405;wwwieLorg.National Center for Community Education, 1017Avon St., Flint, MI 48503; 810-238-0463;www.nccenet.org.

21st Century Community Learning Centers, U.S. De-partment of Education, Office of Educational Re-search and Improvement, 555 NewJersey Ave., NWWashington, DC. 20208; 202-219-2187;www.ed.gov/offices/OERI/21stCCLC.

Peggy Farber is a freelance education reporterbased in New York City.

Discussing Student Work Gives TeachersNew PerspectiveProject Zero's "Rounds" adapts a medical tradition for teachers

0 nce a month from Octoberthrough May, some 35teachers, principals, stu-dents, and other visitors

from around the U.S. gather in a con-ference room at the Harvard GraduateSchool of Education to discuss a pieceof student work in a process called"Rounds." Named after the medical tra-dition of "Grand Rounds" in which doc-tors come together to consider chal-lenging cases, these sessions aresponsored by Project Zero, a 30-year-old research group at HGSE.

The Rounds at HGSE is one of sev-eral major efforts to get teachers to takea more in-depth look at their craft, in-cluding the "Critical Friends Groups"sponsored by the Annenberg Institutefor School Reform and the "Tuning Pro-tocol" used by the Coalition for Essen-tial Schools (see HEL, March/April1998). While some models use collabo-rative assessment to reach consensusabout a rating or grade for a piece ofwork, Project Zero uses these sessionsto "get closer to a child's purpose, andrecognize the challenges and accom-plishments in the work," says SteveSeidel, a Project Zero research associatewho leads the monthly sessions. Teach-

ers who participate by bringing sam-ples of their own students' work oftenfind the collaboration helpful, becauseit either confirms their view of the workor introduces new perspectives thatthey can use in working with studentsin the future, says Seidel. Participantssometimes start similar conversationsin their own schools because they valuethe chance to give a single piece of stu-dent work the attention they feel it de-serves, and because it enables them toreflect on their experience as educa-tors, he says.

An Insider's ViewTo get an inside view of this process,

Harvard Education Letter editorNancy Walser sat in on the AprilRounds, which focused on a story writ-ten by a 6th-grader. The story "A Regu-lar Legalist Day" (see p. 7), was writtenfor Ann Jaquith, a California teacherwho assigned her students a role fromthe Han Dynasty in Chinese history andasked them to write about it as part ofan interdisciplinary course on historyand English, called "Making Connec-tions." Participants (identified by firstnames only) are shown only the writtenpiece as it was turned in to Jaquith.

Through the use of five principles thatlay out ground rules for the discussion(see box, p. 8), participants begin bydescribing the work, then raise ques-tions about it, and, finally, speculate onwhat the student was trying to accom-plish. The presenting teacher then an-swers these questions about the writerand the assignment before bringing upsome dilemmas of her own.

In this edited excerpt of a one-hourconversation, participants have alreadyfigured out that the story is historicalfiction set in China. They note some in-herent contradictions between the so-phisticated imagery and glaring gram-matical problems. Some participantsalso note that major conflicts in thestory seem to have been left unfinished.Jaquith tells the group that the studenthas trouble with the mechanics of writ-ing and often lacks motivation. Al-though she's read the piece numeroustimes, this is the first time she's awareof the parallels between the stories shehas told the class about her travels inChina and her student's story.

Arm: I told the students a story aboutbeing in a restaurant in China andeating duck and duck's brain, and, lo

6 The Harvard Education Letter, July/August 1998

A Regular Legalist Day

I opened my eyes and looked out the window. The sun was so bright I had to squint. I saw the same thing Isaw every day, a maple tree and four beautiful lilies surrounding it. I stretched my arms and stepped outof bed.

I went and fetched three buckets of water from the small well outside my house. I put them on the stove andheat them up.since it took a while to heat up, I took off my robes in advance. !saw if the water is hot enough(which it was) and poured them into my bath tub.The water was soothing so I took longer than usual. I felt ithad been long enough.My skin was even beginning to wrinkle. I dried off and put my red shirt on and my pants.

I went in to my kitchen for breakfast.' had my usual; a bowl of rice with duck on top. My favorite part is thehead because the brain is so tender. Just when I'm finished up I heard a knock at the door. I look to see who itwas and it was none other than my brother.

"Oh hello brother ,what a surprise to see you here."I said"I'm having a tea party ,may I borrow twelve plates from you ."he asked."Sure, but am I invited?""Yes of course my brother,"he said"I'll be right out with the plates."As I was taking the plates out of the cupboard a huge crash that almost made me drop the plates I was holding.I ran out to see what happened with plates in hand. My bother was standing and an innocent look on his face

with my precious vase at his feet broken into a million pieces.) let out a yelp and jumped back in horror."I was just looking at it when all of a sudden it floated out into mid air and dropped." he said lying through

his teeth."It's one thing that you broke it but to lie on top of it is way to extreme.Get out of my house. I'll have to deal

with you later!!!."Now I was late for work. I ran out my door ,into my courtyard, down my unusually quiet street and in to the

busy city of Changan,where I caught my usual horse drawn carriage."To the palace," I requested.The driver slapped the reigns on the horse and we were off.In about ten minutes we arrived at the palace."That will be four yuan sir."I paid him and stepped down. I saw the usual thing ; two beautiful hand carved jade gates with very detailed

dragons on them. Standing to the right of them I saw Letu, the guard."Hello Letu," I said"Hey Leje, The emperor is waiting for you!"As I walked in I saw many beautiful pieces of art by famous crafts people. I walked up the stairs and saw the

emperor sitting on his throne. I bowed to him and he offered me some jasmine tea."Hello Leje, I have a very important issue to discuss with you""And what is that your highness?" I asked"Well a scrawny peasant stole a ruby from my personal treasury.What should I do?""I think you should cut of his hands because it will teach him a very good lesson and others will learn his

mistakes.""That's a very wise idea," he saidWe discussed a few more issues and before I knew it, it was lunch time. We ate fish, pork, duck, rice, vegetables

and tea. While we were eating we heard yelling out side. We instantly got up and walked to the balcony to seewhat was happening. When we looked down we saw an angry mob of peasants yelling about not having enoughfood and money to support their families.

"Guards, go shoot a crossbow into the crowd to break up the riot," He commanded. They did as they weretold and sure enough every one scattered in different directions.

"Well now that that's over with, I give you permission to leave."I quickly ran down the stairs and out the door saving just enough time to say goodbye to Letu. Since the

carriage didn't come in the after noon I had to walk home. After Forty five minutes of walking I finely reachedmy front gate. I went inside and saw my broken vase right were I had left it .1 didn't want to deal with it then soI went into my kitchen to eat. You might think I shouldn't be hungry now after that lunch but after walking sofar I was Ready to eat. I made myself a bowl of rice and fish.) ate it and after a long day of work I went straightto bed.

TYCO -E E ND

The Harvard Education Letter, July/August 1998 7

and behold, it's here. And the heat-ing of the water. That's another storythat I told about China. You can'tdrink the water, so you have to boilit. So I'm noticing now many of thedetails that are in this piece are notfrom things that she read, but arefrom stories that I told in class. Aboutthe conflicts, I also wondered aboutthe rubyhow someone could getaccess to it in a guarded palace? If Iwere writing this piece, that's whereI would have put my energy, not withthe conflict with the brother.Barbara P: How did the student feelabout the piece?Ann: The way she feels about manythings that she doesthat she wasglad to be done.Chris: As in the big, bold font sheuses for "THE END"?Ann: Yeah. This was supposed to bea third draft, which is interesting tome. What do you do with a final piecethat comes in and has so many er-rors? I think that there were thingsabout it that she felt quite goodabout. I think that she was workingon trying to use sensory images anddetails to give a sense of place. Sheloves art. One of the things that thestudents did in the process of writingwas to draw a sketch of the housethat their character would live in. Wealso looked at pictures of places likethe Forbidden City to get a sense ofwhat a palace might have looked like.Chris: I think that with this kind ofpiece there's a real dilemma or atrade-off, because it's about historyand it's also about writing. And whenkids are doing historical fiction, oftenthey're not going to do their bestwriting because they're trying tobring in this other thing that isn't somuch what they own. On the otherhand, I think it's a great way to dohistory. Kids usually like it, and say,"Oh, this is so much fun." Also, Ithink this is a finished thing. Iwouldn't even ask her to do anotherthing on it.Ann: I struggle with how hard topush her and when to just let go.Other students, when asked a coupleof those questions, say, "Can I do an-other draft ?" And her question ismuch more likely to be, 'Am I goingto have to do anything else?"Beth: I wondered if she had donethe drawing [of the house] beforeshe did the writing. I wonder abouthaving her visualize the other partsof the story, because it might helpher key in to those parts that aren't

Project Zero'sFive Principles of Collaborative Assessment

1. Start by looking at the work knowing littleor nothing about the context inwhich the work was produced.

2. Describe the piece, keeping in mind that no observation is too obvious.Withhold judgments as much as possible.

3. Most students approach their work with purpose and intent even if it is notimmediately apparent.

4. When speculating about what the student was trying to accomplish, beready to cite examples from the work. Keep coming back to the work.

5. Don't worry if you come away from Rounds with more questions thananswers.

as finished.Barbara S.: It has to be our job tohelp her do a better piece. You know,I can't see this as the end state. Be-cause her strength is the artistic, thevisual, that's where she was really de-scriptive, and that's where you gotthat wonderful feeling. So I was won-dering for her, first of all, if she ex-pressed it artisticallythe conflicts,the inhumanityand then if shecame back to the writing, how muchmore powerful it might be.Kevin: This might be tough becauseit's part of a particular history unit,but I wonder if this wouldn't besomething you could come back toafter a while. Not right away, but inthree weeks, in two months, say tothis girl, let's look at the ending nowand see if there are any ways we canpush it.

Putting "Rounds" to WorkJaquith did go back to school and use

this conversation in two different con-texts. In a study group with eight otherteachers at her school, Jaquith contin-ued the debate about when a piece ofstudent work should be considered"done." "We didn't reach a consensus,but we decided it's really important forstudents to get honest feedback," shereports. She also had a conversationwith the story's author. Though it hadbeen three months since the studentwrote the story, she remembered theassignment well and was anxious toknow what had been said about it dur-ing the April Rounds. "We talked aboutthe verb tenses and I think I raised herawareness about the errors. We alsotalked about the unresolved conflictwith the brother, and I was struck byhow much thought she put into that

8 The Harvard Education Letter, July/August 1998

characterthat for him, punishing thepopulace was probably much easierthan executing justice at home." In fact,the next two pieces the student wrotewere stronger, Jaquith reports. "I hadthe sense that she had a renewed pridein what she was doing, that the one-on-one attention she got on that piece hadsome impact, because she showedmore care and consideration in hernext assignments."

Jaquith says the Rounds process is of"tremendous value" to her as a teacherbecause "others notice things you don'tand raise questions that keep youthinking about your practice," she says.She also feels it is valuable for helpingteachers to keep an open mind whenlooking at a piece of work before jump-ing to conclusions about what a studentcan or cannot dosomething that canbe key to helping students achieve aparticular goal, she says. 'A teacher whocontinues to think about her practice isalways asking, why am I doing what I'mdoing and how can I be more effective?"says Jaquith. "These questions are veryvaluable for modeling learning andrisk-taking to students, because that'swhat they need to do to succeed. Andin the process of doing that, I becomea better teacher. "

For Further InformationHarvard Project Zero, 323 Longfellow Hall, 13 Ap-pian Way, Cambridge, MA 02138; 617-495-4342;www.LEARN@PZ. Send e-mail to info@pz.har-vard.edu.D. Allen, ed. Assessing Student Learning: FromGrading to Understanding. New York: Teachers Col-lege Press, in press.N. Lyons, ed. With Portfolio in Hand: Validating theNew Teacher Professionalism. New York: TeachersCollege Press, 1998.

411

4

Volume XEV, Number 5 Published by the Harvard Graduate School of Education September/October 1998

From Sputnik to TIMSS: Reforms in ScienceEducation Make Headway Despite SetbacksMore time is needed for widespread classroom changes

BY NAOMI FREUNDLICH

hen final science scoresfrom the Third Interna-tional Mathematics andScience Study (TIMSS)

tests came out late last year, U.S. stu-dents proved to be behind half of thosein other developed countries by theeighth grade, and dead last by the finalyear of secondary school. The lowscores, though shocking, were not un-expected: The U.S. has been trying tooverhaul science education since theRussians launched Sputnik in the1950s, but progress has been slow.

While many new science courseswere developed in the 1960s in re-sponse to Sputnik, that period of re-form halted shortly after U.S. astro-nauts landed on the moon, according

INSIDE

Latino Achievement / 4

The TIMSS, the Times,and the Times / 7by Howard GardnerBystanders Can HelpBattle Harassment / 8

to James Rutherford, former director ofProject 2061, the American Associationfor Advancement of Science's programfor revamping K-12 science education.With the crisis over, science reformstopped short of producing a "criticalmass" of newly trained teachers. "Itwasn't rooted in the schools yet, so wewent back to the traditional ways of do-ing things," says Rutherford.

Worries over whether the U.S. cankeep up in the global economy haveushered in the currentperiod of science edu-cation reform, saysRutherford, whose1990 book, Science forAll Americans, is con-sidered "the bible" forthese efforts. Unlikethe 1960s, though,there is general con-sensus on what con-cepts students shouldbe learning from kin-dergarten through 12th grade. Buttranslating those expectations into theclassroom remains the central task. In-depth studies of TIMSS results, com-bined with case studies of reform ef-forts, are shedding even more light onthe many changes that are still neces-sary. Among them are wider dissemina-

tion of newer inquiry-based sciencematerials and intensive training ofteachers in how to use them. "Trying toput together the pieces is the challengenow," Rutherford says.

Beyond StandardsAlthough the TIMSS scores don't re-

flect it, there have been major nationaland statewide efforts to improve sci-ence literacy in the schools. Some 46states have adopted science standards

or curriculum frame-works. Many of theseframeworks are basedon standards set by theAmerican Associationfor the Advancement ofSciences (AAAS) andthe National Academyof Sciences. In addi-tion, about half of allstates have institutedstate-wide tests in sci-ence, just like those in

reading and math.But standards have proved to be only

a starting pointteachers also neednew materials and training to interpretand follow them. In a 1998 analysis ofstate standards, the Thomas B. Ford-ham Foundation gave only six statestop marks for having standards that

The TIMSS scoresdon't reflect themajor effort nowin place to improvescience literacyin schools.

EDITORIAL DIRECTOR: Kelly Graves-Desai. EDITOR: Nancy Walser. PRODUCTION EDITOR: Dody Riggs. EDITORIAL BOARD, HARVARD GRADUATE SCHOOL OFEDUCATION: Mildred Blackman, Director, The Principals' Center; Sally Dias, Superintendent, Watertown Public Schools, Watertown, MA; Harold Howe II, SeniorLecturer Emeritus; Susan Moore Johnson, Professor and Academic Dean; Robert Kegan, Senior Lecturer; Jerome T. Murphy, Professor and Dean; Gary A. Orfield,Professor; John Ritchie, Superintendent/Principal, Lincoln-Sudbury Regional High School, Sudbury, MA; Catherine Snow, Professor; Jay Sugarman, Teacher, Runkle

School, Brookline, MA; Ariadne Valsaunis, Director of Public Information; Dennie Palmer Wolf, Lecturer on Education. NATIONAL ADVISORY BOARD: John Brademas,President Emeritus, New York University; Constance E. Clayton, former Superintendent, School District of Philadelphia; Alonzo A. Crirn, Professor of Education, SpelmanCollege; Linda Darling-Hammond, Professor, Teachers College, Columbia University; Andrew Heiskell, Chairman Emeritus, New York Public Library; Marra Levenson,Superintendent, North Colonie Central Schools, NY; Deborah Meier, Founder, Central Park East Secondary School, NY; John Merrow, President, The Merrow Report;ArthurJ. Rosenthal, Publishing Consultant.

3 BEST COPY AVAILABLE

were sufficiently specific when it cameto content. Since releasing its own"Benchmarks for Science Literacy" in1993, the AAAS's Project 2061 (namedfor the year that Haley's Comet will re-turn) has recently embarked on a seriesof workshops aimed at introducingstandards to teachers in individualschool districts, and is working on a da-tabase to help middle school teachers,in particular, find curriculum materialsto meet them. "That's the big obstacle,"says Mary Koppal of Project 2061. "Peo-ple want to use standards and bench-marks, but they want to know whatworks."

Less Is MoreSince the 1980s, evidence has been

mounting that "hands-on, inquiry-based" materials recommended by na-tional science standards impart factsjust as well as the traditional lectureapproach, with the added bonus of im-proving student attitudes toward sci-ence, (HEL, May/June 1990). Assess-ments like those devised by Ohioresearchers (see box, p. 3) are also con-firming that new inquiry-based ap-proaches can raise scores, even amonggirls and minorities, groups that tradi-tionally have not performed well in sci-ence.

Unfortunately, the textbooks thatmost schools rely on, especially in mid-dle and high schools, have not been re-vamped to emphasize these new meth-ods. "Despite a recognition that Project2061 and other national reform docu-ments are calling for reduced contentand different approaches to teachingthan in the past, publishers have madeonly incremental changes, layered ontop of existing textbook formats andcontent," say the authors of an SRI In-ternational report on Project 2061.

The problem with U.S. science textsis that they take the "mile-wide, inch-deep" approach, says William H.Schmidt, research coordinator at Michi-gan State University's U.S. TIMSS Na-tional Research Center. Schmidt and hiscolleagues analyzed 491 curriculumguides and 628 textbooks from aroundthe world to get a picture of how theU.S. taught math and science as corn-

pared to other countries. Based ontheir study "A Splintered Vision,"Schmidt argues that the U.S. did poorlyon TIMSS primarily because its sciencecurriculum attempts too much and isrepetitive from year to year. Countriesthat did better on TIMSS have more fo-cused curriculum, he says. "If you lookat a [U.S.] 8th-grade science book, it has65 topics in it," he says. "The Germanand Japanese textbooks have five topicsin them and the international averageis 20. Science has big theorems, but theway it's conveyed to U.S. students is likea large laundry list."

Minnesota's experience with theTIMSS tests confirms Schmidt's thesis,says Bill Linder-Scholer, executive di-rector of SciMathMN, a nonprofit re-form organization that sponsored Min-nesota's participationin the test. Minnesotastudents tied with Sin-gapore's for the bestscores on the TIMSS8th-grade earth sciencetest, primarily because95 percent of its 8th-graders take earth sci-ence (as opposed to amix of science subjects)there was "common agreement"among teachers to focus on only fourtopics within that subject, says Linder-Scholer. "We have no magic bullet," hesays. "The fundamental lesson is aboutalignment."

Some new curriculum is being devel-oped to correct this problem. In starkcontrast to the 65 topics found in the650-page earth science text most com-monly used in middle schools, the Na-tional Science Resources Center, for ex-ample, is developing an earth sciencecourse that focuses on only three topicstaught entirely with activities and thin"student guides."

Centralized Efforts WorkDespite these pockets of progress,

reformers at the National Science Foun-dation (NSF) have discovered just howdifficult it is to bring about widespreadimprovement. Between 1991 and1993, the NSF signed cooperativeagreements with 25 states to carry out

"standards-based systemic reformthroughout their jurisdictions." TheseStatewide Systemic Initiative (SSI)grants averaged about $10 million, andthere was great latitude in how the stateprograms could be executed. Statestypically contracted to universities orother outside groups to set up teacherdevelopment programs, introduce newcurriculum, and try to measure success.

A study of the SSI program releasedin March by SRI International con-cluded that it "had limited, moderate,and uneven impacts on classroom prac-tice." The programs that failed oftensuffered from what program managerPatrick Shields calls the "let a thousandflowers bloom approach"givingmoney to many local initiatives to dowhat they want with it. However, cen-

trally managed pro-grams like Ohio's (seebox, p. 3) that featureintensive teacher train-ing did far better. "Pro-grams that were able toshow a real impact onstudents and teacherswere those that startedearly, were able to pro-

vide intensive high-quality professionaldevelopment, and used pre-made cur-riculum materials," said Shields.

Learning by DoingTeacher training plays an especially

important role in science education ef-forts for two major reasons: science asa field is constantly buffeted by new dis-coveries, and the learning curve issteeper because teachers generallyhave had little specialized training inscience. The average elementary schoolteacher has taken only one collegecourse in science.

Reform efforts focused only on newcurriculum without adequate teachertraining appear to be "doomed," sayJohn Cannon and David Crowther ofthe University of Nevada. After spend-ing $1 million on a new hands-on, ac-tivity-based curriculum called "SciencePlace," one large Nevada school districtsaw its reforms fail because teachers atits 56 elementary schools didn't knowhow to use it, according to a 1997 case

"The way scienceis conveyed to U.S.students is a likelarge laundry list."

and because

THE HARVARD EDUCATION LETTER (ISSN 8755 -3716) is published bimonthly by the Harvard Graduate School of Education, Appian Way, Cambridge, MA 021383752. Second-class postage paidat Boston, MA, and additional mailing office. POSTMASTER: Send address change(s) to The Harvard Education Letter, 6 Appian Way, Cambridge, MA 02138-3752.

Signed articles in THE HARVARD EDUCATION LETTER represent the views of the authors. Address editorial correspondence to Editors, The Harvard Education Letter, Gutman Library, 6AppianWay, Cambridge, MA 02138; phone 617-4953432; fax 617 - 496.3584; email address: edletter@hugsel.harvard.edu; web site: http:// hugsel.harvard.edu / hepg/hel.html

1998 by the President and Fellows of Harvard College. Published as a nonprofit service. All rights reserved. Special written permission is required to reproduce inany manner, in whole or inpart, the material herein contained. Call 617.495 -3432 for reprint permission information.

How to subscribe: Send $32 for individuals, $39 for institutions ($40 for Canada/Mexico, $42 other foreign, in U.S. funds only) to The Harvard Education Letter, 6 Appian Way, Cambridge MA02138; or call us at 617-495-3432 in Massachusetts or 800-513-0763 outside Massachusetts between 8:30 a.m. and 5:00 p.m. EST Monday-Friday. Subscription prices subject to changewithoutnotice. Single copies, $5.00. Back issues and bulk subscriptions available at special reduced rates; call 800-513-0763.

't BEST COPY AVAILABLE

Ohio Reform Shows ResultsWhen the National Science Foundation announced its in-

tention to award millions of dollars in grants for statewidescience education reforms, Jane Butler Kahle, a professor atOhio's Miami University, teamed up with physics Nobel laure-ate Kenneth G. Wilson to come up with a plan. Their idea wasto train middle school teachers in inquiry-based physicsa"gatekeeper" subject that students need to take in order totake higher-level science courses. The goals were to create amore knowledgable teaching force and to engage more femaleand minority students in the subject, since these students aremore likely to opt out of science after middle school.

What they came up with "was successful beyond anyone'sexpectations," wrote Kahle in a 1997 article for Science Edu-cator. Dubbed the "Discovery Institute," the six-week summerprogram trained teachers in math, life science, and physics.Taught by master teachers, mathematicians, and scientists ateight sites throughout Ohio, the institutes stressed teachingteachers "the way we wanted them to teach," says Kahle.Teachers received stipends of up to $60 a day and could earncollege credits by attending. Unlike the typical lecture-lab sci-ence course, those taking Discovery's Physics By Inquiry insti-tute learned by doing. For example, teachers were told to puttogether light bulbs, batteries, and strips of aluminum to makethe bulbs light as part of learning about electricity.

To reach even more teachers, a two-week summer work-shop for individual districts was developed to help teacherswork through actual curricula in a way consistent with na-tional science standards. "Teachers who go through the two-week institutes are very likely to go the next summer to thesix-week institute because we whet their appetite," said Kahle.Whereas many programs peak at reaching the 10 percent ofteachers who usually volunteer for reforms, the Discovery Pro-gram reached almost 40 percent of middle school teachers inparticipating districts. The initial cost of both the two- and

six-week training was about $25 per hour per participant, butthis figure decreased as the program grew.

To measure the program's success, Kahle and her col-leagues surveyed teachers, principals, and parents, and visitedschools to gauge changes in teacher practices. They also de-veloped their own test to see how well students of teacherswho did and did not participate in Discovery training wereable to think conceptually and interpret data. Results showclearly that students in "reform" classrooms with teachers whohave attended Discovery Institutes are more likely to writeabout how they solve problems, to work in small groups, andto use hands-on materials than those in "non-reform" class-rooms. Results also show that teachers in reform classroomsuse more inquiry-based teaching approaches, which requirestudents to support their claims, encourage questions, andinspire students to discuss subjects. These practices remainedevident three years after teachers were trained.

Test scores were not only higher in reform classrooms, butgirls actually outscored boys within their racial groups. AfricanAmericans in reform classrooms scored as well as white stu-dents in non-reform classrooms. Studies like these are doneevery year, and results are published and given to state legis-lators to boost support for the program. Ohio has alsochanged teacher certification criteria to require "intense expe-riences with inquiry," says Kahle.

"We have changed attitudes about what constitutes goodprofessional development in the state," she says. "That may bethe long-lasting effect of what we've done."

For Further InformationJ.B. Kahle. "Systemic Reform: Challenges and Changes." Science Educator 6, no.1 (Spring 1997): 1.6.

Ohio's Systemic Initiative: Discovery, 420 McGuffey Hall, Miami University, Oxford,OH 45056-1693; 513-529-1686 (evaluation) or 937-775-2726 (K-12 programs);http://www.discovery.k12.0h.us.

study by the two researchers. Although ers will have to pass to graduate begin- said Loucks - Horsley. "Inquiry-basedthe publisher provided training for 10 ning in 2003. In New Jersey, a partner- learning is not just posing questions tolocal teachers, who were then paid to ship with the Merck Institute for Sci- teachers and kids; it's exploring physi-provide workshops for other teachers,only 59 percent of the district's teachersattended even a single workshop.

ence Education and local districtsstresses fundamental science concepts,as well as classroom management skills

cal and natural phenomenon in a waythat they learn serious scientific princi-ples."

Teachers complained that there was not for keeping teaching materials stocked In the end, the goal is not simply forenough time during the year to com-plete the units or to work out prob-

and organized.Some of the most effective training is

U.S. students to score better on TIMSS,reformers say, but to construct the class-

lems. Without continued training forteachers using "Science Place" and in-centives for teachers to attend work-

being done far away from schools ingovernment labs, science museums,and even on beaches, says Susan

room of the future. "The longer thing,which we think will take 25 to 50 yearsto accomplish," says Rutherford, "is to

shops, the reform was "doomed from Loucks-Horsley, a director of profes- really put in place the next system, thethe beginning to trip and fall," theauthors concluded.

sional development at the National Re-search Council. Programs in these set-

next curriculum that will turn out,across the board, students who are re-

The best professional developmentin science is collaborative, stresses con-

tings can show teachers exactly what ismeant by inquiry-based learning so

ally comfortable with math, science,and technology, and know how to use

tent learning, and is done over a period they can model and guide discussions it and control it and who can participateof weeks with opportunities for follow- in the classroom, she says. "They can in a more interesting and safer world."up discussions, experts say. There is no take a field trip to the beach to look atone model. In Massachusetts, for exam- erosion, develop questions, and do an For Further Informationple, state education officials are using investigation. All of a sudden they have "A Splintered Vision: An Investigation of U.S. Science

and Mathematics Education," 1997. Available fromvideotapes to show teachers how to questions about what they are seeing the U.S. TIMSS National Research Center, Michiganteach to statewide tests that 10th-grad- and realize they need to know more," State University, College of Education, 455 Erickson

The Harvard Education Letter, September/October 1998 3

Hall, E. Lansing, MI 48824-1034; 517-353-7755.Available on the web at http://ustimss.msu.edu."Benchmarks for Science Literacy" 1993. Availablefrom the American Association for the Advancementof Sciences/Project 2061, 1333 H. St. NW Washing-ton, DC 20005; 202-326-6666; http://project2061.aaas.org. For more information on Project2061 Professional Development Programs, call ScottMay at 1-888-PDP-2061.

J.R. Cannon and D.T. Crowther. 'An Autopsy of anElementary Science Program Implementation,"1997. Available from the National Association for Re-search in Science Teaching, Ohio State University1929 Kenny Rd., Room 200E, Columbus, OH 43210;614-292-3339. Available on the web at http://

science.coe.uwfedu/narst/conference/cannon.pdf."Evaluation of the National Science Foundation'sStatewide Systemic Initiatives (SSI) Program," 1998.Available from SRI International, Center for Educa-tion and Human Services, 333 Ravenswood Ave.,Menlo Park, CA 94025; 650-326-6200. Reports areavailable on the web at: http://www.sri.com/policy/cehs/edpolicy/html.S. Loucks-Horsley et al. Designing ProfessionalDevelopment for Teachers of Science and Mathe-matics. Thousand Oaks, CA: Corwin Press, 1998."Minnesota TIMSS Report," 1997. Available from Sci-MathMN, 638 Capitol Square, 550 Cedar St., St. Paul,MN 55101; 612-296-4058; http://www.informns.k12. mn. us/scimathmn.

F.J. Rutherford and A. Ahlgren. ScienceforAllAmeri-cans. New York: Oxford University Press, 1990."Science and Technology Concepts for MiddleSchools," in press. Available from the National Sci-ence Resources Center, Arts & Industries Building,Room 1201, MRC 403, Smithsonian Institution,Washington, DC 20560, 202-287-2063; http://www.si.edu/nsrc.

Naomi Freundlich is a freelance writer in New YorkCity and a former science reporter for BusinessWeek.

Latino Achievement ReexaminedResearchers seek new ways to help this population succeed in schoolBY LAUREL SHAPER WALTERS

Latino children are the largestminority group in U.S. schools,now outnumbering African-American children by 35,000.

They are also the most academicallytroubled racial or ethnic group, withpersistently high dropout rates and lowtest scores. While the nation's overallschool completion rate rose steadilyduring the past two decades, the Latinodropout rate remained at 30 to 35 per-cent, more than dou-ble the rate for AfricanAmericans and 3.5times that for whites."According to everyeducational indicator,Hispanic Americansare making progress atalarmingly low rates,"concluded PresidentClinton's Advisory

range of subgroups all from Spanish-speaking countries, each with its ownhistory and culture. Much of the re-search on Latino educational achieve-ment focuses on Mexican Americans,since that subgroup accounts for two-thirds of all Latinos in the United States.

Given the persistent and uniqueproblems Latinos as a group have inschool, some explanations have cen-tered on the culture itself. One widely

held theoryincreas-ingly viewed with suspi-

"The moreAmericanizedLatino immigrantstudents become,the worse theirattitude in school."

Commission on Edu-cational Excellence forHispanic Americans in a 1996 report.

Researchers looking at why thisachievement gap exists have come upwith a long list of possible influences,including poverty, discrimination, seg-regation, the stresses of immigration,frequent language limitations, andpoorly educated parents. Yet evenwhen economic backgrounds, immi-grant status, and language differencesare taken into account, Latino studentsstill have higher dropout rates thanother ethnic groups.

The umbrella term "Latino"oftenused interchangeably with the term"Hispanic"encompasses a wide

cion by researchersisthat Latinos value familyloyalty and work overgrades and advance-ment to college, espe-cially for girls. Recentresearch is chippingaway at that thesis.Some of the most valu-able research has mixedquantitative data with

an ethnographic approach of inter-viewing students, parents, and schooladministrators to help identify thekinds of struggles students face. The re-sults of this approach are debunkinglong-held myths and providing betterinsight into how schools can help thisgrowing segment of the populationsucceed in school.

Questioning the DeficitModel

One study in particular has madewaves for refuting the idea that Latinoscome to school lacking the capabilitiesand attitudes to succeed. This vein of

thought is considered to be part of abroader "deficit theory," which, from itsbeginnings in the 1920s, sought to ex-plain the academic failings of minoritiesby blaming the students and their fami-lies, according to Richard Valencia, as-sociate professor of educational psy-chology at the University of Texas andauthor of The Evolution of DeficitThinking, a 1998 book oh the theory.

Recent work by Harvard culturalanthropologist Marcelo M. Suarez-Orozco and his wife, research psychol-ogist Carola Suarez-Orozco, has shownpersuasively that it may be the processof assimilating into American culturethat has more influence over Latinoachievement than any factor related totheir native culture. Beginning in 1990,the couple conducted a three-year re-search project comparing the schoolexperience of immigrant students fromMexico with children still living in Mex-ico, second-generation Mexican Ameri-cans, and non-Latino whites. They in-terviewed 189 students ages 13 -18, aswell as teachers and parents, to studythe impact of school dynamics, peerand family life, and attitudes towardachievement on school performance.They found that immigrants are per-forming better than second-generationor U.S.-born Latinos. As one teachernoted, "The more Americanized [Lat-ino immigrant students] become, theworse their attitude in school."

Marcelo Suarez-Orozco believesthese findings refute the deficit theory."Children do not arrive with a culturaldeficiency or baggage," he says. "New

4 The Harvard Education Letter, September/October 1998

arrivals come with tremendously high Some have theorized that the immi- riential knowledge should not conflict

aspirations. If you control for every- grants' optimistic attitudes may give with teachers' responsibility for provid-

thing else, immigrants get better way to frustration in the face of real-life ing students with particular academic

grades, they study more, they spend obstacles, which may affect family or content knowledge and learning skills."

less time watching '11; they spend less student attitudes toward school. John While few researchers would resur-

time hanging out with their friends. U. Ogbu, an anthropologist at the Uni- rect arguments along the lines of the

The problem is that over time, for many versity of California, Berkeley, argues "deficit model," there also is a shift away

reasons, those things aren't sustained." that some minority groups give up on from simply blaming school practices

Pinpointing the causes of declining education after seeing that it does not such as tracking, poor counseling, and

motivation and achievement among offer them the same rewards it brings retention, and a renewed recognitionLatino immigrants is proving difficult. the majority population. Over time, mi- that sociocultural factors might, in fact,

"It's not going to be a one-dimensional nority youth learn not to invest in edu- play a role. For example, it's still impor-

answer," says Gary Orfield, a professor cation because they do not see a long- tant to examine student and family atti-

at Harvard Graduate School of Educa- term benefit in the job market, Ogbu tudes towards school among ethniction, who is editing a new book on Lat- argues. Census data from 1990 support groups because such factors do affect

ino civil rights. "There are a lot of dif- this argument: the mean annual in- educational outcomes, argues Russell

ferent things going on, and trying to come for white male high school gradu- W Rumberger, an education professor

weigh the relative importance of them ates was $22,521; for Latinos it was at the University of California, Santa

is not easy" $14,644. Barbara. "If you believe in cultural dif-

To begin with, the self-selection proc- ferences, that doesn't necessarily imply

ess for immigrants may inflate the pic- A Closer Look at Schools a deficit," he says.

Lure of their initial success, researchers Other researchers are looking more But other researchers warn againstsay, because immigrant families arrive closely at what schools may be doing to drawing any conclusions about Latino

with optimism about the future and a hinder Latino achievement. Scholars culture as a whole. Harriet Romo and

desire to better themselves. A 1996 such as Enrique T Trueba of the Univer- Toni Falbo studied 100 "at-risk" Mexi-

RAND report found that immigrant sity of Texas argue that many schools can-origin students in Austin, TX, over

children and parents of all ethnic fail to provide a culturally appropriate four years. "We found no single culture

groups have higher educational aspira- education for minority students. of the home, nor was there a single cul-

tions than their U.S.-born counterparts. "Latino experiences and cultural Lure of the school," they wrote in Lat-

Inexplicably, this gap in aspirations is capital need to be counted as strengths," ino High School Graduation: Defying

three times larger for Latinos than for Trueba writes. "The incorporation of the Odds, a 1996 book based on their

other groups. students' language, culture, and expe- study. "Furthermore, our statistical

How Schools Can Help Latino Students SucceedMuch of what Latino students need to succeed academically

applies to all students, regardless of ethnicity. But beyond ba-sic school reform, there are some strategies schools can useto help both Latino students and their parents notonly survivethe school years, but also thrive. Here is a synopsis of resear-chers' recommendations for improving Latino achievement:

Emphasize prevention of problems. Most students showwarning signs when struggling academically or consideringleaving school. If, for example, a student has two unexcusedabsences in a row, schools should contact the parents directly.Personalize programs and services for Latino students.Connecting with adults and establishing mentoring relation-ships can make a significant difference, particularly for highschoolers. Some strategies for achieving this include reducingclass sizes, creating separate houses or academies within alarge high school, teachers "adopting" a group of students,and older students mentoring younger ones.Make schools accessible to parents. This may require pro-viding transportation to meetings, going to homes to meetwith parents who do not have phones, providing translators,and sending home correspondence written in Spanish.Provide parents with frequent feedback about their chil-dren's academic progress. Many Latino parents are unawareof the schools' expectations for their children. Standards mustbe communicated clearly and repeatedly

At the elementary level, provide after-school tutoringand enrichment related to in-class assignments. Trainedvolunteers in the classroom offer another opportunity foradult-student connections.Set high academic and behavioral standards for students.Latino students should be actively recruited into the highestlevel classes and provided the extra support needed.Provide intensive academic counseling for students thatincludes parents. Encouraging college attendance requiresearly and regular attention to courses taken and preparationfor standardized testing, as well as parental support.Incorporate Latino culture and traditions into the class-room when possible, particularly at the elementary level.Students should feel that their cultural background is wel-comed and respected by the school.Replicate programs that have proven effective. In schooldistricts across the country, targeted efforts are improving Lat-ino achievement and lowering dropout rates. School do notneed to start from scratch. The Hispanic Dropout Project's1998 report, for example, lists nearly 50 successful programsnationwide (See For Further Information, page 6).

Laurel Shaper Walters

Sources: Romo and Falbo, Rossi, and Secada.

The Harvard Education Letter, September/October 1998 5

analyses indicate that characteristics ofthe home culture did not predictwhether the student graduated." Forexample, whether students spoke Eng-lish or Spanish at home had no statisti-cal link to whether the student gradu-ated from high school.

Romo and Falbo, both professors atthe University of Texas, argue that "cul-tural differences between homes andschools do not inevitably become edu-cational barriers unless school policiesmake them so." They found that mostschools overestimatedthe resources of theparents in their study.Many could not reador write and had littleknowledge of the fun-damentals of the edu-cation system, makingit difficult for them tosupport their chil-dren's education. For

In spite of stereotypes suggestingthat Latino families don't value educa-tion, this study and many others showthat nearly all Latino students and theirparents want a college education."They just don't know what to do to getthere," explains Eugene Garcia, dean ofthe graduate school of education at theUniversity of California, Berkeley, andchairman of the task force. "These par-ents didn't even go to high school, sothey have no idea what kind of hoopstheir children have to jump through to

go to college."This lack of knowl-

Nearly all Latinostudents want acollege education.They just don'tknow what to doto get there.

example, many par-ents could not read correspondencebecause it was sent home in Englishonly, and they had no way of getting toteacher conferences because theydidn't have a car.

Lowering Barriers to CollegeOther studies point to a lack of "edu-

cational capital"knowledge of theeducational systemas an explanationfor why a very small percentage of Lat-ino graduates are eligible for highereducation. In California, Latinos are 30percent of the K-12 population, yet lessthan 4 percent of Latino high schoolgraduates in the state meet the require-ments for admission to the Universityof California. The University of Califor-nia's Latino Eligibility Task Force istracking a group of students from the10 California high schools graduatingthe most Latino students, weighted bygeographical distribution for rural-ur-ban representation. The median aver-age school completion rate for parentsof these students is less than six years.

COMING SOON

Building Social Skills in theClassroomValue-Added: The NewAccountability ToolResearchers and TeachersCan Help Each Other

edge about educationalinstitutions is not beingcompensated for by aca-demic counselors in thehigh schools, the Cali-fornia study found.Many Latino studentswere not on a college-prep academic track,and 50 percent of those

who were taking all the right coursesnever took the SAT In an effort to boostLatino college enrollment, the Eligibil-ity Task Force has recommended thatthe University of California eliminatethe SAT requirement for admission.This would increase the proportion ofeligible Latino high school graduates by59 percent, according to the report.

The task force also notes that nearly60 percent of Latino applicants to theUniversity of California from 1993 to1997 were female. "There's a culturalnotion that Latinos will not send girlsto college," Garcia says. "But when weinterviewed girls they said there's noproblem going to college as long as it'snear home."

Future DirectionsTo best understand the Latino expe-

rience in U.S. schools there needs to becomparative data from other immi-grant groups, researchers say. MarceloOrozco-Suarez is in the first year of afive-year comparative interdisciplinarystudy of Afro-Caribbean children fromHaiti and the Dominican Republic,Asian children from China and Taiwan,and Latinos from Mexico and CentralAmerica. The study includes 35 re-search sites on both U.S. coasts. "Thecomparative data allows you to look atmore than one level," says Suarez-Orozco, "and find the processes thatmediate the adaptation of immigrantchildren to U.S. schools."

Yet much of what needs to be done

to help Latino students does not re-quire further study, researchers argue."These kids need schools that are safe,that are integrated, where the materialsare up to date, and the teachers aretrained properly," says Suarez-Orozco."Those kinds of things don't requirerocket science."

"Attention to Hispanic school com-pletion must become a salient part ofthe national agenda on education,"write the authors of "No More Ex-cuses," the final report of the HispanicDropout Project, a group commis-sioned by Education Secretary RichardRiley to study the problem of Hispanicdropout rates. "That this crisis has re-mained largely invisible results in inac-tion and allows the many excuses fordoing nothing to go unchallenged," theauthors wrote.

In the past several decades, a na-tional effort has helped lower dropoutrates and close the achievement gap be-tween whites and African-Americans,says Walter Secada, director of the His-panic Dropout Project and a professorat the University of Wisconsin in Madi-son. "We can do the same things forHispanics."

For Further InformationJ.U. Ogbu. "Understanding Cultural Diversity andLearning." Educational Researcher 21, no. 8(1992): 5-14.

"Our Nation on the Fault Line: Hispanic AmericanEducation. Final Report from the White House In-itiative on Educational Excellence for HispanicAmericans." Available from the U.S. Department ofEducation, 202-401-1411; http: / /www.ed.gov/offices/OlIA/Hispanic.

H.D. Romo and T Falbo. Latino High School Gradu-ation: Defying the Odds. Austin: University of TexasPress, 1996.

R.W. Rumberger and K.A. Larson. "Toward Explain-ing Differences in Educational Achievement amongMexican-American Language-Minority Students."Sociology of Education 7, no. 1 (1998): 68-93.W.G. Secada et al. "No More Excuses: The Final Re-port of the Hispanic Dropout Project." Washington,DC: U.S. Department of Education, February 1998.Available on the web at www.ed.gov/offices/OBEMLA/hdprepo.pdf, or by calling 1-800-USA-LEARN.

M. Suarez-Orozco and C. Suarez-Orozco. Transfor-mations: Migration, Family Life, and AchievementMotivation Among Latino Adolescents. Stanford,CA: Stanford University Press, 1995.

University of California Latino Eligibility Task Force."Latino Student Eligibility and Participation in theUniversity of California YA BASTA!" (Report No. 5),July 1997. Chicano/Latino Policy Project, Institutefor Social Change, University of California atBerkeley; 510-642-6000.

Laurel Shaper Walters is a freelance educationwriter based in St. Louis, MO.

6 The Harvard Education Letter, September/October 1998

The TIMSS, the Times, and the TimesBY HOWARD GARDNER

Like most Americans, I was surprised and disappointedby the performance of 12th-graders on the recentThird International Mathematics and Science Study(TIMSS). After a promising performance at the 4th-

grade level, American youngsters fell sharply, vying with Cy-prus and South Africa for the lowest scores on these lengthy,primarily short-answer tests. Even though nearly every aspectof the Study has been justifiably criticized, and it is thereforedifficult to know exactly how to interpret the results, no onecan take comfort in this poor performance.

I was invited to comment by the New York Times. Insteadof critiqing the TIMSS, I proposed to discuss what should nowbe done in our country. A few days later the Times ran my pieceunder the misleading title, "Low Scores Are No Disgrace." Alas,neither the Times nor the annoyedreaders who wrote in understood myargument. I welcome the opportunityto state my position more clearly.

Fifteen years ago, when the U.S.economy was in the doldrums, manycommentators blamed the schools. "If only we could improveour students' learning," so the argument went, "we wouldonce again be a competitive country" Today the United Statesstands in the midst of the greatest economic revival in memory.Still, by most accounts, our schools have changed little. Evi-dently there is not a strong tie between the performances ofthe schools, however measured, and the country's economicsuccess.

This state of affairs, I argue, gives us a special opportunity.We do not have to accept other countries' definitions of whatcounts for an educated student. Instead, we have the luxuryof asking, "What, really, do we want our students to know andhow can we be convinced that they know it?"

At present this question is approached historically andretrospectively. We take our curricula and assessments largelyfrom what has been done in the past; we let graduate schooldictate what is needed in college and college dictate what isneeded in high school. This scenario makes sense if (like Sin-gapore and Brazil) we seek a nation of engineers. It proveseccentric if we seek an informed and engaged citizenry.

Let me put my cards on the table. I do not care that muchif a student has mastered algebra, geometry, and precalculusif she knows all the basic facts of geology, astronomy, physics,and biology. Indeed, if we are honest with ourselves, we mustadmit most of us never use this information. Unless we be-come scientists, engineers, or mathematicians, the major usewe make of this knowledge is to help our children prepare forexaminations twenty years later!

What is important for citizens? In mathematics, broadly con-strued, adults need to be able to manage their own personalfinancesbalance a checkbook, arrange the proper mort-gages, invest shrewdly for old age; to make prudent decisionsin the light of the laws of probability; to understand and evalu-ate the actions of legislators in collecting and dispersing largesums of money. In sciences, individuals should understand thenature of the scientific enterprise: what is a fact, a theory, ahypothesis, an experiment. Individuals should be able to un-derstand a newspaper report about a new medical treatment,evaluate the evidence on which it is based, and decide whether

to believe it. Put bluntly, in a country where half the peoplefollow astrology rather than astronomy, and where many re-ject the theory of evolution, some appreciation of science writlarge is far more important than committing to memory factsthat will soon be forgotten.

If these sentiments were taken seriously, we would have torevisit vast parts of the curriculum. Following this, we wouldtackle far fewer subjects and probe them in much greaterdepth, we would give students a sense of what it means tounderstand a topic in detail, and, one hopes, whet their appe-tites for continuing to learn and understand deeply for the restof their lives.

I am under no illusion that American youngsters, any morethan others, would do well initially on measures of such un-

derstandings. Indeed, most of our as-sessments the world over do not evenprobe such understandings, thoughthere is no reason why they could not.But at this unique moment in scholas-tic and economic history, we have a

rare opportunity to revisit what we are doing, to make majorchanges, and to put our assessments in order.

How to get started? We need a few dramatic counterexam-ples to the standard view of science and mathematics educa-tion epitomized by the TIMSS. In the spirit of the Coalition ofEssential Schools, such counterexamples must foreground anunderstanding of how one thinks scientifically and the capac-ity to carry out the mathematics needed to negotiate the dailylife of an engaged citizen. Many curricula can yield these out-comes. The crucial complement is a set of performance-basedexaminations in which students have the opportunity to dem-onstrate these pivotal understandings. Students who do wellon such measures will not only be better prepared for theworld, but will also prove more attractive to colleges and uni-versities. And should such effectiveness be demonstrated, Ihave little doubt that other countries will follow suit, just asthey have aped American practices of IQ testing, SAT exami-nations, and other short-answer instruments.

Back to my title. The TIMSS is not a bad instrument, theTimes was kind enough to print my article. Yet neither of theseinstitutions appreciates the crucial role of "time" in the edu-cation that I am advocating. The TIMSS surveys whether onehas encountered and gained a glancing familiarity with hun-dreds of facts, concepts, and equations; it does not probe anyunderstanding in depth. It takes a great deal of time to developthe understandings that are needed to be an informed citizenin the world. Still, that investment of time is worthwhile be-cause once it has been made, it can be drawn upon for therest of one's life and passed on to succeeeding generations. Atone time it was called a liberal education.

INSI GHTS

Howard Gardner is theJohn H. and Elisabeth A. Hobbs Professor of Cognitionand Education at the Harvard Graduate School of Education. He is alsoauthor of Frames of Mind: The Theory of Multiple Intelligences and many otherpublications.

Copyright 0 Howard Gardner, 1998.

4 jThe Harvard Education Letter, September/October 1998 7

Bystanders Can Play a Role in Battling HarassmentDay after day, researchers from Hofstra Universitywatched as Joey harassed one girl after another inhis 7th-grade class. One day, Joey picked one of hismore unattractive classmates for a long-running

monologue about the various sexual favors she could do forhim. Adding insult to her injury, other classmates egged himon by saying, "Joey would never go for slag like that." Finallythe teacher spoke up, saying only, "Oh, Joey, calm down."Never, in three years of observation, did Joey suffer any realconsequences for his behavior.

For both teachers and students, middle school is seen as aplace "where people make fun of you, where you have to bemore carefula place where youlearn to take it," says Charol Shake-shaft, professor of administrationand policy at Hofstra who is writinga book about the three-year study ofnine New York area middle schools.Ironically, while both students andteachers seemed to accept the environment as a given, bothgroups told Hofstra researchers that they would like "some-thing to be done."

But what if Joey's classmates had behaved differently? Whatif they had told him to stop the harassment, or had gatheredaround the girl to defend her. In fact, enlisting bystanders inthe fight against peer harassment may be the hidden weaponneeded to combat the pervasive problem of bullying and sex-ual harassment (gender-related bullying) in schools.

"The notion of the bystander appears to be really importantto changing the culture," says Eleanor Linn, senior associatedirector at University of Michigan's Programs for EducationalOpportunity. Linn completed a review of interventions thisyear, which was commissioned by the Hamilton Fish NationalInstitute on School and Community Violence for the U.S. Jus-tice Department. As it turns out, no intervention or programto combat peer harassment has yet been evaluated. However,a look at what works when it comes to preventing sexualabuse, date rape, and other violence has turned up some cluesabout just how effective bystanders can be in stopping harass-ment, she says.

When it comes to designing an effective program, researchshows that kids tend to retain information if adults tell themthey can use it "to help a friend," says Linn. This means stu-dents can be motivated to stand up to harassers if they see itas part of protecting their schoolmates. Bystanders are also in

a better position to stand up to harassers, says Linn. Unliketargets, bystanders are not as traumatized, so confronting aharasser can be easier.

Not many educational materials address the role of the by-stander, says Linn. Some that do are published by the WellesleyCenter for Research on Women at Wellesley College. Ratherthan tell teachers what they should tell students to do, how-ever, these materials contain activities for helping studentscome up with their own ideas about how they can respond toharassing acts.

In Flirting or Hurting? (for grades 6-12), for instance, stu-dents discuss how they would have acted if they'd witnessed

several incidents that became well-known court cases. In Bu//y Proof (forgrades 4-5), students discuss the "de-grees" of courage it takes to confronta bully depending on whether thebully is a friend or an unpopular stu-dent. In Quit It! a new curriculum for

grades K-3 published this year by the Wellesley Center, stu-dents are asked how they can "listen to their strong side" tohelp someone being teased on the playground.

The language adults use when encouraging children to talkabout peer harassment is very important, says Nan D. Stein, aWellesley researcher who helped develop and write all threebooks while documenting the high rates of sexual harassmentin schools over the past two decades. Although sexual harass-ment is serious and illegal, children can more readily under-stand the concept when it is put in the context of "gender-re-lated" bullying, she says.

Despite a recent U.S. Supreme Court decision shieldingschools from damages, they can still be held liable for incidentsof sexual harassment under Title IX. And even though familieshave increasingly resorted to lawsuits, incident rates of sexualharassment in schools have not changed much since the early1990s, according to Stein, who is also a coauthor with Linn ofthe new Justice Department report. Surveys show that girlshave an 80 to 97 percent chance of being sexually harassed inschool, while boys have a 60 to 80 percent chance.

Stein, Linn, and other experts say it's time for schools to usea whole arsenal of strategies to battle the problem. Awarenesscampaigns, posters, schoolwide surveys, policies, incident re-porting procedures, and counseling are also part of the solu-tion. So, too, are students.

Nancy Walser

For Discussion

HEL on the WebThis issue of the Harvard Education Letter can be ac-cessed at our Web site. You can also see what's comingup in future issues, read about our reprint publica-tions, browse our subject and chronological indexes,and send us an e-mail message. Our address is:http://hugse1.harvard.edu/hepg/hel.html

For Further InformationE. Linn. "The Effectiveness of Existing Programs and Approaches for ReducingSexual Harassment and Sexual Violence in Schools." Available from the HamiltonFish National Institute on School and Community Violence, 1925 North Lynn St.,Suite 305, Rosslyn, VA 22209; 703-527-4217.

N. Stein. "Incidents of Sexual Harassment and Sexual Violence in K-12 Schools."Available from the Hamilton Fish National Institute on School and CommunityViolence.

Wellesley College Center for Research on Women. Wellesley College, 106 CentralStreet, Wellesley, MA 02181; 617-283-2507.

8 The Harvard Education Letter, September/October 1998

The HarvardEducation Letter

Volume XIV, Number 6 Published by the Harvard Graduate School of Education November/December 1998

Learning to Listen May Help ChildrenLearn to ReadResearchers examine whether lessons in 'Phonemic awareness" can preventreading problemsBY NANCY WALSER

In the early 1990s, language artscoordinator Marguerite Held wasput in charge of working with stu-dents who, for some reason, were

not reading at all by January of their1st-grade year. The school was in a sub-urban district outside Houston, TX.The students were largely middle-class,with parents who had read to them athome. Their teachers had introducedthem to phonics as part of the district'sliterature-based program, yet some stu-dents were still unable to read.

Held, a teacher for 17 years, had ahunch what was wrong. "They all had asimilar problem, and that was that theycould identify the individual sounds [ofthe alphabet], but they couldn't blend

INSIDE

Social and EmotionalLearning Programs / 4

Building CollaborativeRelationships / 7

A Parent's Influenceby Jerome Kagan 1 8

them. They couldn't take the word matand make it into cat, bat, or rat withoutreally thinking about it." In otherwords, these students didn't under-stand that the "m" of "mat" was separa-ble from the "at," that the letter "m"represented that particular sound, andthat putting the sounds (representedby other letters) into the same slot inthe word would generate differentwords. This understanding constitutes"phonemic awareness." In short, theirprognosis for becoming"real readers," Held real-ized, was poor.

After reading a numberof research articles, Helddeveloped her own reme-dial program. She beganby teaching children howto listen to sounds withinwordshow to count thenumber of sounds andmanipulate the sequenceof sounds to make different words. "Itseems like it would be so easy, but thereare so many children who can't do it,"she says. "But still you can teach themand you can see some success." In hersix years of tutoring, all but one of thechildren she worked with became read-ers, she reports.

One in five children is estimated tohave difficulty learning to read, accord-ing to a report entitled PreventingReadingDifficulties in Young Children,recently released by the National Re-search Council (NRC). Reading difficul-ties are more likely to emerge amongchildren who are poor, are minorities,attend urban schools, or arrive atschool not speaking English. But manychildren like those in Held's school alsoface difficulties learning to read. Most

reading problems, the re-port claims, can be pre-vented by appropriate at-tention to language de-velopment, developmentof phonological aware-ness, and development ofliteracy and print skillsduring preschool andkindergarten.

In the search for pre-ventive strategies, some

researchers have emphasized the im-portance of phonemic awareness, askill they believe pre-readers should de-velop to ensure that reading progressessmoothly. Phonemic awareness is theknowledge that words can be seg-mented into smaller units of sound(called phonemes). It is reflected in

adings can be

ed byHaten.

Most reproblempreventappropattentio

EDITORIAL DIRECTOR: Kelly Graves-Desal. EDITOR: Nancy Walser. PRODUCTION EDITOR: Dody Riggs. EDITORIAL BOARD, HARVARD GRADUATE SCHOOL OFEDUCATION: Mildred Blackman, Director, The Principals' Center; Sally Dias, Superintendent, Watertown Public Schools, Watertown, MA; Harold Howe II, SeniorLecturer Emeritus; Susan Moore Johnson, Professor and Academic Dean; Robert Kegan, Senior Lecturer; Jerome T. Murphy, Professor and Dean; John Ritchie,Superintendent/Principal, Lincoln-Sudbury Regional High School, Sudbury, MA; Catherine Snow, Professor; Jay Sugarman, Teacher, Runkle School, Brookline, MA;Ariadne Valsamis, Director of Public Information. NATIONAL ADVISORY BOARD: Constance E. Clayton, former Superintendent, School District of Philadelphia; AlonzoA. Crim, Professor of Education, Spelman College; Linda Darling-Hammond, Professor, Teachers College, Columbia University; Andrew Heiskell, Chairman Emeritus,New York Public Library; Marya Levenson, Superintendent, North Colonie Central Schools, NY; Deborah Meier, Founder, Central Park East Secondary School, NY; JohnMerrow, President, The Merrow Report; ArthurJ. Rosenthal, Publishing Consultant.

42BEST COPY AVAILABLE

children's abilities to produce rhymes,to engage in games involving pho-neme-additions, -deletions, and -sub-stitutions (such as in "eeny, meeny,miney, mo"), and to represent soundswith letters when engaged in inventedspelling.

Most children make this connectionthrough the normal course of languagedevelopment (see box) combined withearly exposure to reading at home or inpreschool. However, children whodon't "get it" will have difficulty sound-ing out and blending new words, re-membering them, and learning to spell,according to the NRC report.

A Strong Research BaseThe research base linking phonemic

awareness to success in reading isstrong and well documented. Analyz-ing the research on this link, authors ofthe NRC report concluded that "dozensof...studies have confirmed that there isa close relationship between phonemicawareness and reading ability, not justin the early grades but throughout theschool years.... Even prior to formalreading instruction, the performanceof kindergartners on tests of phono-logical awareness is a strong predictorof their future reading achievement."

Individual researchers go even fur-ther. "In the last two decades, a scien-tific body of evidence has accumulatedthat points to poor phonological proc-essing deficit as a core cause of poorreading," writes Barbara Foorman, aprofessor of pediatrics at the Universityof Texas-Houston who served on theNRC committee. "Burgeoning evidenceexists that deficits in this area can beameliorated through appropriate train-ing, particularly with children in kin-dergarten through grade 2," she adds.

Among the many recommendationsmade by the comprehensive report isthat children who need it get "explicitinstruction" in phonemic awareness.Such instruction should be "matchedto children's needs," and those whohave mastered it "should move on tomore advanced learning opportunities."

"What the committee is recommend-ing is that each kid get a chance to learn[phonemic awareness], whatever it

takes," says Catherine Snow, a profes-sor at the Harvard Graduate School ofEducation and chair of the NRC com-mittee.

In response to the overwhelming re-search indicating its importance, atleast two states have begun screeningkindergartners for phonemic aware-ness. In Virginia, a voluntary pilot pro-gram attracted 90 percent of the state'sschool districts to a testing programthat also provides some funding to helpchildren who fail the screening. InTexas, a mandatory screening is beingimplemented for the first time this fallfor all kindergarten, 1st-, and 2nd-gradestudents in the state.

Still, researchers caution that in thebig picture of reading development,phonemic awareness is only one pre-dictor of success, along with letter

Researchers cautionthat phonemicawareness is only onepredictor of success.

knowledge, vocabulary, and "readi-ness" skills. For example, vocabularysize also appears to be a critical factorin beginning reading, as is the learningof letter names.

"Although there is a high correlationbetween phonemic awareness and suc-cess in beginning reading,...we knowthat children learn these things withina rich literacy experience that includesreading, writing, listening to stories,and talking about them," says DorothyStrickland, a professor of reading atRutgers University who was also anNRC committee member.

"This [research] should not implythat simply working on these things inisolation will automatically lead to suc-cess in reading," she adds.

The NRC report echoes the convic-tion of most reading educators in em-phasizing that, while developing pho-nemic awareness is an important stepin the prevention of reading difficulties,attention to children's language devel-opment, to their development of print

knowledge, and to maintenance ofpositive affect and motivation to readare equally important, says Snow.

How Explicit ShouldInstruction Be?

Just how explicit "explicit instruc-tion" of phonemic awareness should beis the most contentious issue amongthose specializing in this area of literacy.In this area the research on effective-ness is also less clear.

Because phonemic awareness ap-pears to be so critical to the process ofreading, and because not all childrenespecially those in impoverished innercitiespick it up at home or at school,some argue that schools need to sys-tematically teach it through a sequen-tial program beginning with kinder-garten.

Part of the urgency to teach phone-mic awareness skills no doubt comesfrom the widely stated national goal ofhaving all students reading at gradelevel by the 3rd grade. But others pointto evidence that problemsif not cor-rected by the end of 1st gradecanhave reverberations that go far beyondthe realm of reading.

"I have seen kids who can memorize,and they kind of slip through until 2ndgrade, when the number of words theyhave to memorize just gets to be toomuch," says Held. "Then you are notonly dealing with reading problems,but emotional issues as well, becausethey know they are behind."

To reading adults who already un-derstand that words are made of differ-ent sounds, the concept of phonemicawareness may seem obvious. But justtry to get a 6-year-old to listen to thesounds in a word before jumping to itsmeaning, says NRC committee memberMarilyn Jager Adams, a reading con-sultant based in Cambridge, MA. Theauthor of the 1990 book, Beginning toRead: Thinking and Learning AboutPrint, Adams has also co-authored twocurricula for teaching phonemic aware-ness for kindergarten and 1st grade.

In her curricula, lessons are struc-tured around interactive sound gamesthat teachers can use with a whole classor small groups. "It begins by diving

THE HARVARD EDUCATION LETTER (ISSN 8755.3716) is published bimonthly by the Harvard Graduate School of Education, Appian Way, Cambridge, MA 02138-3752. Second-class postage paidat Boston, MA, and additional mailing office. POSTMASTER: Send address change(s) to The Harvard Education Letter, 6 Appian Way, Cambridge, MA 02138-3752.

Signed articles in THE HARVARD EDUCATION LETTER represent the views of the authors. Address editorial correspondence to Editors, The Harvard Education Letter, Gutman Library, 6AppianWay, Cambridge, MA 02138; phone 617. 495.3432; fax 617-496-3584; email address: edletter@hugsel.harvard.edu; web site: http://hugsel.harvard.edu/hepgffiel.html

1998 by the President and Fellows of Harvard College. Published as a nonprofit service. All rights reserved. Special written permission is required to reproduce inany manner, in whole or inpart, the material herein contained. Call 617. 495 -3432 for reprint permission information.

How to subscribe: Send $32 for individuals, $39 for institutions ($40 for Canada/Mexico, $42 other foreign, in U.S. funds only) to The Harvard Education Letter, 6 Appian Way, Cambridge MA02138; or call us at 617-495-3432 in Massachusetts or 800 - 513.0763 outside Massachusetts between 8:30 a.m. and 5:00 p.m. EST Monday-Friday. Subscription prices subject to change withoutnotice. Single copies, $5.00. Back issues and bulk subscriptions available at special reduced rates; call 800-513-0763.

4 3 BEST COPY AVAILABLE

How Children LearnThat the Alphabet Is a Symbol System for Sounds

FROM BIRTHBegin to distinguish the sounds of human language

ONE TO TWO YEARS OLDEarly vocalization gives way to enjoyment of rhyming lan-guage, nonsense word play, etc.Word comprehension emerges, followed by a sharp in-crease in vocabularyDevelop syntax, ability to use 3 or 4 words in a sentence

PRESCHOOL YEARS (THREE TO FOUR YEARS OLD)Know that alphabet letters are a special category of visualgraphics that can be individually namedPay attention to separable and repeating sounds in language(e.g., "Peter, Peter, Pumpkin Eater")Phonemic awareness begins as the perception of speechmoves from holistic to segmented units of sound, includingthe appreciation of rhyme and alliteration

Begin to match letters of the alphabet, like the first letter ofone's name, to a sound.Increasing vocabulary allows conversation about complexand abstract ideas while sharing books with adults, fosteringgreater comprehension skills.Develop ability to think about language, pronunciation,word usage, sentence structure, and text form

KINDERGARTEN/FIRST GRADE (FIVE TO SIX YEARS OLD)Learn to identify letters and letter-sound correspondencesUse phonemic awareness and letter knowledge in inventedspelling; begin to decode words by mapping letters to pho-nemes.Begin to read words by sightMake transition to "real reading" of novel words

Source: "Preventing Reading Difficulties in Young Children." National ResearchCouncil, 1998

into language play so kids will under-stand and listen to the sounds ofwords," says Adams. "It also uses textsdesigned to exercise the phonics. Kidsare invited to do invented spelling fromthe moment they can, and there's lotsand lots of reading.

"It's a balanced program," she con-tinues, "but it's not the only way. Theissue is, are you watching the kids tomake sure they are with you? If you'redrilling, that's a sign that something'sgone awry"

A 1998 study by Foorman and otherscompared the progress of 285 disad-vantaged students in classrooms usingAdams's "Open Court" curriculumwhich features 15-20 minutes of se-quenced phonemic awareness acti-vities ("direct code")with thosereceiving two other types of instruc-tion. In one other type, students weregiven sequenced phonics lessons incombination with texts that feature cer-tain spelling patterns ("embeddedcode"); in the third type, students weretaught phonemic awareness skills in-formally within the context of a litera-ture-based program ("implicit code").

According to the authors, students inthe direct-code classrooms "improvedin word reading at a faster rate and hadhigher word-recognition skills." Theimpact was greater among childrenwho had weaker phonological aware-ness coming into the classroom. How-ever, students taught through the im-plicit-code method had more positive

attitudes toward reading, and there wasno measurable difference among thethree groups on spelling achievementor vocabulary growth.

Nevertheless, Foorman and her col-leagues concluded that "results showadvantages for reading instructionalprograms that emphasize explicit in-struction...for at-risk children."

Opposing ViewsThat is not the conclusion drawn by

Gerald Coles, a psychologist and for-mer teacher in Ithaca, NY, who analyzedphonemic awareness research and pro-grams for his 1998 book, Reading Les-sons: The Debate Over Literacy. Pro-grams that focus on teaching phonemicskills have not proven their effective-

"If there is a crisis inreading, it's withminority children notlearning to read."

ness, he says, since they are unable toseparate out other effects like initialskill levels. In some studies, even chil-dren who had high phonemic aware-ness did not show improved readingability, he notes.

"The problem I have with this inor-dinate emphasis on skills is that it pre-sumes that it's the instructional magic

bullet for solving the literacy problem,particularly for poor children," Colessays. He argues that scripted programstaught with basals like "Open Court"limit children's reading choices and de-emphasize motivation by teaching in a"lock-step" fashion. What teachersneed more, he says, are smaller classesso they can individualize instruction tosuit their students' needs.

For her part, Foorman decries thelack of flexible grouping in classrooms,but feels there's a need for basal pro-grams to teach phonemic awarenessbecause of high teacher turnover andlack of adequate preparation to teachreading, especially in inner cities. "Ifthere is a crisis in reading, it's with mi-nority children not learning to read,"she says. Foorman urges administratorsto look at the background and pre-school experience of their students, aswell as their families' level of literacy, inmaking curriculum decisions.

When it comes to individualizingreading instruction, designers of newphonemic awareness screening tests inVirginia and Texas say the tests shouldhelp teachers do that by pinpointingwhich students are having trouble inwhich areas. Virginia's PhonologicalAwareness Literacy Screening Program(PALS) sponsors a web page that allowsteachers to input test results and givesinstant feedback on which students areweak in which phonemic skills. In ad-dition to getting test results, teacherscan also use the page to learn about

The Harvard Education Letter, November/December 1998 3

instructional strategies for teaching theneeded skills, says PALS co-directorMarcia Invernizzi.

With data collected on 57,000 stu-dents so far, PALS researchers intend toidentify which districts can help stu-dents improve the most in terms ofphonemic awareness between the falland spring tests, according to Inver-nizzi. "When we find them," she says,"we'll go find out what they did."

For Further Information"Preventing Reading Difficulties in Young Children,"1998. National Research Council, National AcademyPress, 2101 Constitution Ave. NW Lockbox 285,Washington, DC 20055; 1- 800.624 -6242. This re-port, and "Starting Out Right," a shorter versionwritten for parents and others, is also available onthe web at http: //wwwnap.edu.Phonological Awareness Literacy Screening Program(PALS), 125-A, Ruffner Hall, 405 Emmet St., CurrySchool of Education, University of Virginia, Char-lottesville, VA 22903; 804-243-8685; http://curry.edschool.Virginia.EDU/curry/centers/pals.

G. Coles. Reading Lessons: The Debate Over Liter-acy. New York: Hill & Wang, 1998.

M.J. Adams et al. Phonemic Awareness in YoungChildren: A Classroom Curriculum. Baltimore, MD:Paul H. Brookes, 1998.

B.R. Foorman et al. "The Role of Instruction inLearning to Read: Preventing Reading Failure in At-Risk Children. "Journal of Educational Psychology90, no. 1 (1998): 37-55.

Programs Fostering "Emotional Intelligence"Show PromiseSome practitioners see critical needs being met by social and emotional curriculumBY MICHAEL SADOWSKI

Acting out, fighting, racial andother slurs, bullying, willfuldisruption of learning. Manyteachers are faced with prob-

lem behaviors like these at one time oranother: some face them virtually everyday. And as every teacher knows, theeducational costs of these behaviorscan be tremendous.

Debby Collins, principal of the K-5Plymouth School in Monrovia, CA, saysthat until recently most of her staff be-lieved lack of discipline was a seriousimpediment to learning in their class-rooms. Collins recalls several teachers'comments:

"Kids don't know how to cooperateor act nicely toward each other.""There is far too much acting-out be-havior.""What's killing us, keeping us frombeing the best we can be, is a lack ofdiscipline."

Having tried traditional approachessuch as detention, time-outs, and callsto parents without much success, Col-lins and her staff decided it was time tolook for an alternative. After readingTeaching Children to Care: Manage-ment in the Responsive Classroom byRuth Sidney Charney, the Plymouthstaff became interested in and eventu-ally adopted the Responsive Classroomapproach to social and emotionallearning. They started the programwith the 1997-98 school year, and Col-

lins says things are already beginning tochange.

Developed at the Northeast Founda-tion for Children in Greenfield, MA, theResponsive Classroom approach fo-cuses on building such skills as coop-eration, healthy assertion, responsibil-ity, empathy, and self-control. Programcomponents geared to building thesecompetencies include "morning meet-ing," in which students practice conver-sation and listening skills as they sharestories and concerns;"guided discovery,"whereby teachers de-liberately guide stu-dents' introduction toclassroom materials tofoster curiosity and re-spect for these materi-als; "academic choicetime," which encour-ages students to makepersonal investmentsin their own learning;and classroom organi-

ing so that messages and expectationswould be consistent throughout theschool. While it is too early to gauge theprogram's effect on academics, Collinshas been encouraged by teachers' com-ments and the changes she notices inchildren's attitudes and behaviors."The kids are developing more empa-thy for each other. They're better ableto see another person's point of view,"she observes. "There are fewer disci-pline referrals, and kids seem to be able

to cooperate better. Itreally does create a

The overarchingidea behind SELprograms isdeveloping socialand emotionalskills, includingself-esteem andrespect for others.

zation that includesstudent "interest areas" and facilitatescollaborative work in different-sizegroups.

The Plymouth School completed itsfirst year of the Responsive Classroomprogram last spring, with all 630 stu-dents in 31 classes participating. Col-lins made sure that all adults in thebuildingnot only teachers, but alsosecretaries, custodians, and otherbuilding staffparticipated in the train-

more orderly environ-ment where kids aresafer."

Of course, class-room and playgroundconflicts still occur, butCollins has noted dif-ferences in the waychildren handle them:"In situations wherethere has been a physi-cal fight or somebodytakes something from

somebody else, now kids are morelikely to be honest about it and to takemore responsibility for their actions."

Responsive Classroom is just one of23 programs listed on the website ofthe Consortium for the Advancementof Social and Emotional Learning(CASEL), an organization established in1994 and dedicated to the develop-ment, promotion, and evaluation of so-cial and emotional learning (SEL) pro-

4 The Harvard Education Letter, November/December 1998 ti

grams. While these programs vary inobjectives, approaches, and grade lev-els reached (ranging from preschoolthrough grade 12), the overarchingidea behind SEL programs is develop-ing social and emotional skills, in-cluding self-esteem, respect for others,personal decision-making, avoidinghigh-risk behaviors, conflict resolution,and effective communication. In thisway, SEL educatorshope to preempt prob-lem behaviors andhelp students becomehappier, more empa-thetic, more sociallyskilled adults. "Themost successful SELprograms are thosethat address the needsof the whole child,"says CASEL executivedirector Roger Weiss-berg, professor of psy-chology and education at the Universityof Illinois at Chicago.

Interest in social and emotional skillshas been growing lately, as teacherslook for new strategies to deal withproblem behaviors. In his 1995 best-seller Emotional Intelligence, DanielGoleman argues, based on research inpsychology, neurobiology, and otherfields, that knowing how to manageone's emotions is at least as importantfor success in life as intellect. He alsowrites that SEL programs play an inte-gral role in helping prepare children forall of the challenges they face now andwill face in the future: "Beyond [the]educational advantages, the coursesseem to help children better fulfill theirroles in life, becoming better friends,students, sons and daughtersand inthe future [they] are more likely to bebetter husbands and wives, workersand bosses, parents, and citizens."

New Haven's SEL ProgramThe specific reasons why educators

adopt social and emotional learningprograms differ, but many have two pri-mary goals that they believe are closelylinked: improved social and emotionalskills, and better academic perform-ance. Karol DeFalco, a facilitator for so-cial development in the New Haven(CT) Public Schools, says poor aca-demic achievement, especially amonghigh-risk segments of the school popu-lation, prompted that system's decisionto begin developing an SEL program adecade ago. DeFalco and her col-leagues also were deeply concerned

about the district's high rates of teenparenting, dropouts, and illegal druguse, all of which appeared to be closelylinked to low academic achievement.

About 75 people representing vir-tually all constituencies of the NewHaven school communityparents,teachers, administrators, businessowners, clergy, and othersmet anddiscussed ways to improve the achieve-

ment of underperform-ing students. The group

"These prosocialbehaviors serve asquite powerfulacademic enablersfor many of thestudents we workwith."

concluded, based ontheir observations, thatmany of these studentsshared common socialand emotional charac-teristics, including lackof impulse control,poor problem-solvingskills, low self-esteem,difficulty in acceptingindividual differences,and poor communica-

tion skills. 'We realized we had to dosomething about students' social andemotional development if they weregoing to have any hope of improvingacademically," DeFalco says.

Working with Weissberg, then a psy-chology professor at Yale University,district staff members developed a pro-gram in social and emotional learningthat has grown into the New Haven So-cial Development Project, a wide-rang-ing K-12 curriculum in which all 20,000students in the city's public schoolsparticipate. The objectives of the cur-riculum are to improve students' skillsin three areas: self-monitoring, prob-lem-solving and decisionmaking, andcommunication. Within these areas,the program focuses on such issues asself-awareness and anger managementat the elementary level; peer-pressureresistance in middle school; and transi-tion-making, violence prevention, andthe understanding and acceptance ofdifferences at the high school level.

In an effort to gauge the program'seffectiveness, the project's staff devel-oped an evaluation instrument in 1992;they now survey all of the city's 6th-,8th-, and 10-graders (about 2,500 stu-dents) every two years. Their resultshave been encouraging. While thestudy does not establish definite linksbetween SEL and academic achieve-ment, statistics for several academicallyrelated indicators have shown im-provement. According to their surveys,from 1992 to 1996 (the last year forwhich data are available), the dropoutrate, students' attitudes about school,

feelings of school safety, retention rates,and the percentage of students whoplan to go on to college have all im-proved. Students also have a greatersense that there is respect for diversitywithin the schools. Statistics about al-cohol and drug use have essentially re-mained steady. "Kids are fighting less,they report they are less sexually active,they feel happier and safer at school,and more want to go on to college.There's definitely reason for opti-mism," DeFalco says.

Results Encouraging butLimited

Several recent larger studies of theeffects of SEL programs also showpromising results, but their findings re-quire some qualification. Stephen N.Elliott, professor of educational psy-chology at the University of Wisconsin,conducted a study of 301 ResponsiveClassroom and control group studentsin the Springfield, MA, public schools.Using three social skills measurementinstruments and the Iowa Test of BasicSkills (ITBS) to measure academicachievement, Elliott found that over the1996-97 school year:

34 percent of students in ResponsiveClassroom classes showed reliableimprovement in social skills, whileonly 20 percent of students in thecontrol group did;30 percent of students in ResponsiveClassroom classes showed reliableimprovement in problem behaviors,while only 10 percent of students inthe control group did;ITBS scores rose 22 percent for theResponsive Classroom students and3 percent for the control group stu-dents;there is a correlation (determined byregression analysis) between socialskills improvement and improvedITBS scores.

Elliott, who has also studied Respon-sive Classroom programs in other set-tings, says he sees the changes effected

COMING SOON

Professional DevelopmentThrough Inquiry SeminarsSchool ArchitectureRetention

4bThe Harvard Education Letter, November/December 1998 5

by the program occurring in several"waves": "In the first wave of change,problem behaviors decrease. This cre-ates the opportunity for more prosocialbehaviors to replace some of the prob-lem behaviors. And these prosocial be-haviors serve as quite powerful aca-demic enablers for a good 30-40percent of the students we work with."

In another study of SEL's effective-ness, Metis Associates, a New York re-search firm, is examining the ResolvingConflict Creatively Program (RCCP) inthe Atlanta publicschools. A project ofEducators for SocialResponsibility (basedin Cambridge, MA),RCCP is geared to de-veloping students'social skills and theirunderstandingabout alternatives to

Carolyn Lee Hart, liaison specialist inthe Atlanta public schools' departmentof professional development, who hasworked closely on the RCCP imple-mentation, says, "I think the programhas had a great impact because theschools that we selected [for RCCP par-ticipation] were in a difficult area for us,an area where we've had low testscores. The environment of thoseschools has changed so. It's just an up-lifting environment now Teachers havemade changes in what they do, and stu-

dents respond."One thing SEL ad-

The research does littleto answer the questionof whether the benefitsof learning stay withstudents.

violence. As RCCP di-rector Linda Lantieri explains, the pro-ject focuses on "problem-solving andcommunication skills used in deesca-lating conflict." These skills include ac-tive listening, expression of feelings,perspective-taking, negotiation, andencountering bias.

Data covering two school years (fall1995 to spring 1997) show a 17 percentdecline in the RCCP high school's out-of-school suspension rate, compared toa 6 percent increase citywide, and adropout rate that is 1.7 percentagepoints lower than the city average.Qualitative data from the study showimproved self-esteem, more willing-ness to help others, and greater respon-sibility for handling conflicts among theRCCP students. Metis Associates seniorvice president Stan Schneider says thefindings are especially encouraging,given that the RCCP high school hadpreviously been one of the lowest per-forming in the city. In what is perhapsthe study's most impressive statistic,out-of-school suspensions in the RCCPmiddle school (West Fulton) decreasedby 10 percent, compared to an 86 per-cent increase citywide.

West Fulton Principal Robert Bellcredits RCCP and several other new in-itiatives with improving the suspensionstatistics. The school has also addedpeer mediation and Saturday detentionas deterrents to potentially suspend-able offenses. He adds that RCCP train-ing has also helped teachers manageclassroom behavior better, reducingthe need for discipline referrals.

vocates are quick topoint out is that nei-ther the need for SELnor its benefits arelimited to urbanschools and stu-dents. The Oakland,CA-based Develop-mental Studies Cen-

ter recently released findings of threeevaluation studies, representing urban,suburban, and rural schools participat-ing in its Child Development Project(CDP). Geared to fostering "caringcommunities" in schools, key compo-nents of CDP include the integration ofsocial and ethical content into litera-ture-based reading and language artsprograms; specific lesson formats forcollaborative classroom learning; "de-velopmental discipline" that empha-sizes problem-solving approaches in-stead of punishments and rewards; anda range of family involvement efforts.

Students in CDP schools, repre-senting eight districts in four differentstates, were found to have improvedprosocial attitudes and behaviors and agreater sense of their schools as com-munities. Instead of geography, the keyfactor for success, according to Devel-opmental Studies Center president EricSchaps, was whether most teachers ina school made "meaningful movementtoward implementation" of the pro-gram.

Weakness in the ResearchOne weakness in the overall body of

SEL research is that many evaluations(including those of Responsive Class-room, RCCF: and the Child Develop-ment Project previously mentioned)have been commissioned or conductedby the organizations whose programsare under evaluation. SEL advocatesagree that more independent researchis needed. Second, the research does

little to answer the question of whetherthe benefits of social and emotionallearning stay with students beyondtheir experiences in a classroom orschool with an SEL component. Inother words, do SEL students reallygrow up to be better adjusted, morecaring adults?

Weissberg notes that evaluating thelong-term effects of social and emo-tional learning programs is compli-cated by a variety of factors. "Doing anaccurate longitudinal study with thiswork would be extremely costly andcomplex," he says. "How do you evalu-ate a 12- or 13-year multicomponentsocial and emotional learning pro-gram? Basically, any program that canbe evaluated scientifically is probablyone that is less ambitious in scope."

Instead of checking outcome data,Weissberg suggests that educators whoare considering an SEL curriculum lookat what goes into each program: "Howare the skills taught? How clearly de-signed are the lesson plans? How arepeople trained to implement the pro-grams? Is there follow-up training tosupport skill applications? These arethe kinds of questions to ask."

For Further InformationJ.L. Aber, S.M. Jones, J.L. Brown, N. Chaudry, and F.Samples. "Resolving Conflict Creatively: Evaluatingthe Effects of a School-Based Violence PreventionProgram in Neighborhood and Classroom Con-texts." Development and Psychopathology 10(1998): 187-213.R.S. Charney. Teaching Children to Care: Manage-ment in the Responsive Classroom. Greenfield, MA:Northeast Foundation for Children, 1991.Consortium for the Advancement of Social and Emo-tional Learning (CASEL). The CASEL website listsdescriptions, target grade levels, and contact infor-mation for 23 social and emotional learning pro-grams. Beginning next year, CASEL will also includeits evaluations of some existing SEL programs on thesite: http://www.cfapress.org/caselicasel.html.M.J. Elias, J.E. Zins, R.P. Weissberg, KS. Frey, M.T.Greenberg, N.M. Haynes, R. Kessler, M.E. Schwab-Stone, and T.P. Shriver. Promoting Social and Emo-tional Learning: Guidelines for Educators.Alexandria, VA: Association for Supervision and Cur-riculum Development, 1997.D. Goleman. Emotional Intelligence. New York:Bantam Books, 1995.L Lantieri and J. Patti. Waging Peace in Our Schools.Boston: Beacon Press, 1996.

Michael Sadowski is a freelance education writer.He teaches English and drama at Dennis-YarmouthRegional High School in Massachusetts.

6 The Harvard Education Letter, November/December 19984 `i

Building Collaborative Relationships:Educational Research in SchoolsBY ELL4NE B. RUBINSTEIN-AVILA AND CAROLA SUAREZ-OROZCO

The cultural gap between institutions of higher educa-tion and schools is not a new phenomenon. Educa-tional researcherssometimes armed with muchtheory and little practiceare viewed by teachers as

being out of touch with the challenging realities of teachingstudents in today's complex society.

As researchers, we have often been in the position of askingteachers to participate in studies. Many teachers have beenpleased to have the opportunity to participate in researchprojects. Others have been skeptical about whether, ulti-mately, research findings will justify the disruption of theirdaily routines or the additional work involved. Some teachersask, "Why do we need more research?What we need are more books andmanipulatives, not more research"although it is important to note thatfunding for research does not com-pete with funding for books and class-room materials. Or "What will comeout of all this anyway?" And, finally, "What's in it for us?"

Apprehension is a natural consequence of a lack of infor-mation regarding research. In order to build effective collabo-rative relationships, teachers must ask and researchers mustanswer questions about study logistics and goals. Based onour experiences, we offer the following topics that adminis-trators and teachers should discuss with researchers beforedeciding whether to participate.

The most important issue is whether the proposed researchproject has the potential to provide information that couldlead to improved educational opportunities for students. Par-ticipating teachers should know what researchers are tryingto discover and who may benefit from the findings. On a morepragmatic level, the question of what impact the proposedresearch project will have on the day-to-day quality of life ofteachers and students is a key consideration. It is importantfor the teacher to ask the researcher to be specific about howmuch student and/or teacher time will be required for theproject.

The amount of time participants need to commit to a pro-ject can vary greatly. A teacher may be asked simply to identifystudents who meet the study's inclusion criteria, a task thatmight take as little as half an hour. On the other hand, anintervention project may require a teacher to teach a newcurriculum two mornings a week for two years.

Other key questions to ask researchers are: Will classroomobservations be shared with district administrators? How willthe confidentiality of teachers, students, parents, and schoolsbe protected? How will the results be presented and wherewill they be published? Will the analysis of the data requireteacher input? If so, how?

If these questions are not answered to teachers' satisfaction,they should feel justified in turning down the research project.It on the other hand, concerns are met with thoughtful re-sponses, teachers should feel confident in agreeing to partici-pate in a project. Of course, issues may arise during the courseof the study. Researchers should make themselves accessibleto the participants in case concerns arise. A timeline and pro-

gress reports can also help sustain smooth collaborative rela-tionships.

In some of the best school-university collaborations, re-searchers have worked reciprocally with participating class-rooms, lending their expertise in exchange for data. Theseexchanges evolve naturally out of the relationships that devel-op during the course of a project. Below are some examples:

Some university researchers volunteered to help asteacher's helpers either before, during, or after the data col-lection period. They team-taught, co-planned a unit, helpedreview students' work, and answered journals, among otherthings. Numerous teachers have told us that a researchers'

systematic presence in the classroomgave students extra attention, Thistype of collaboration can enhance thequality of data collected, as well as theauthenticity in the analysis of thatdata.

A doctoral student offered to con-duct a few technology workshops for teachers and aides. Inreturn, the great majority of school personnel agreed to beinterviewed three times during the school year and to meetonce after each interview to discuss their experiences.

A veteran teacher asked a researcher to help formulate amini-study for a self-evaluation that her district was piloting.In exchange for numerous classroom observations, the re-searcher met with the teacher to lend her expertise on datacollection and analysis for the self-evaluation. The researcheralso helped make a video documenting the process the teach-ers used to make the self-evaluation.

A graduate student conducting a small qualitative studyfor a methods course solicited ideas from teachers to come upwith a question that was relevant to them. Two team teachersasked to be "shadowed" in their planning and teaching effortsin order to explore the nature of their collaboration. The resultwas an intimate description of the process, providing theteachers with a systematic perspective on their team-teachingfor a semester. The insights they gleaned from this experience,they later told her, fostered better communication betweenthe teachers, which ultimately enhanced their partnership, jobsatisfaction, and, very likely their students' learning.

While there are many wrinkles to be ironed out, teachersand researchers, armed with more knowledge about the re-search process, can arrange for reciprocal exchanges that willpotentially enrich the learning and teaching experiences forall involved. Together, researchers and practitioners mayachieve the stimulating marriage of theory with practice oftenlacking both in universities and in schools.

Eliane B. Rubinstein Avila is a doctoral student at Harvard Graduate School ofEducation, and a research assistant with the Harvard Immigration Projectalongitudinal study of immigrant students being conducted in eight school dis-tricts in the San Francisco and Boston areas.

Carola Suarez-Orozco is a lecturer at the Harvard Graduate School of Edu-cation. Sbe is also co-director of the Harvard Immigration Project.

BEST COPY AVAILABLE48 The Harvard Education Letter, November/December 1998 7

A Parent's Influence Is PeerlessBY JEROME KAGAN

Editor's Note: A new book by Judith Rich Harris, The Nur-ture Assumption: Why Children Turn Out the Way They Do,argues that peers have more influence than we think andparents have less. For the sake of discussion, we askedJerome Kagan, a psychologist at Harvard University, toshare his views about the importance of parents in their chil-dren's lives.

That parents exert a minor influence and peers a majorinfluence on a child's developmentthe chief claimin The Nurture Assumptionignores some impor-tant facts, ones that are inconsistent with this book's

conclusions. Indeed, there is ample evidence that, for betteror worse, parents do shape their children.

Consider, for example, that the best predictor of a child'sverbal talent is the frequency withwhich the parents talk and read to thechild. A verbally talented child is morelikely to get better grades in school,and therefore is a little more likely toattend a better college. That, in turn,makes it more likely that in adulthood he or she will land abetter job.

Moreover, a parent's education (whether they are, for ex-ample, a high school dropout or a college graduate) can pre-dict how their child will fare in the world. Parental educationalattainment, which is related to their child-rearing practices,predicts, among other things, the probability of aggressivebehavior and the likelihood of psychiatric problems whenthese children become adults.

Although parents clearly do matter, parental influences aresubtle, complex, and, at the moment, very difficult to measureobjectively. One of the most important of these processes hasbeen called identification. Children assume that some of thequalities of their parents belong to them and, as a result, theyexperience the emotions that would follow that belief.

A girl with a competent, well-liked mother feels pride be-cause she is identified with a parent who has desirable char-acteristics, and she will carry that pride through adolescence.By contrast, the child with a parent who is incompetent, im-pulsive, drinks too much, and is not liked by neighbors is likelyto carry shame into adolescence and adulthood. The child-hood shame that Frank McCourt, author of the Pulitzer Prize-winning memoir 'Angela's Ashes," felt when he found his fa-ther drunk in a local bar is one example of the consequencesof identification.

Unfortunately, a great deal of the evidence cited to supportthe idea that parents don't matter much is based on question-naires in which people are asked to describe themselves. Theyrespond to such questions as: "Do you like parties?" "Are youafraid to take risks?" "Do you prefer to be alone or withfriends?" and "Do you worry a lot?" But this source of evidenceis too crude to capture a child's pattern of identification withparents.

Indeed, such questionnaire evidence yields answers thathave a special quality. For example, when parental question-naires are used, the estimated degree of genetic influence onthe child's shyness increases as the child grows older. Butwhen the evidence on shyness comes from actual observationof the child at home or in the laboratory, the estimate ofgenetic influence decreases. The direct observations are prob-

ably more accurate reflections of achild's personality than the parents'descriptions.

Put simply, reliance on question-naires as a strategy to evaluate parentalinfluence is a little like trying to under-

stand the galaxies without the advantages of a telescope. Forthis reason, the claim that parents are relatively unimportantis premature.

It is useful to recall that at the turn of the century a majorityof Americans believed that the poor school performance anduncivil behavior of children of European immigrants wasprobably due to genes. Careful scientific work during the suc-ceeding decades corrected that serious misjudgment. In fact,many of those children became some of America's leadingcitizens during the first half of this century.

So why has The Nurture Assumption received so much pub-licity? One reason may be that many American parents feeltense, worried, and occasionally guilty about how their chil-dren turn out. Any book that reduced some of that concernwould be greeted with joy and might be followed by a tempo-rary muting of the burden of responsibility for the child'sgrowth.

But the evidence of parents' importance remains woventhroughout everyday life. The child who is told that he cannothave dessert because he is too loud or impetuous at dinneris, most of the time, better behaved the next day. The childwho is consistently encouraged by a parent to overcome afear of swimming in the ocean will take more risks the nextsummer.

Telling parents that they have little influence on their chil-dren, in light of the scientific evidence and their daily encoun-ters with their children, is a little like declaring on a foggySeptember morning that all the trees have disappeared be-cause you cannot see them.

INSI GHTS

HAPPY HOLIDAYSfrom the

Harvard Education Letter

Jerome Kagan is Starch Professor of Psychology at Harvard University. His re-search during the last 15 years has been on temperamental biases in children,especially fearful and fearless profiles.

4 [zi

8 The Harvard Education Letter, November/December 1998

la

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.

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)

FS o a7 B13