beyond islands of excellence - learning first · the cost for beyond islands of excellence: what...

BEYOND ISLANDSOF EXCELLENCE:What Districts Can Do to Improve Instructionand Achievement in All Schools

A Project of theLearning First Alliance

Development and production of this report were supported by grants from the Ewing Marion KauffmanFoundation and from the Office of Educational Research & Improvement, U.S. Department of Education.The findings and conclusions presented in this report do not necessarily represent the official positions orpolicies of the Office of Educational Research & Improvement, U.S. Department of Education (grant no.R215U000007-01) or the Ewing Marion Kauffman Foundation.

BEYONDISLANDS OFEXCELLENCE:What Districts Can Do to ImproveInstruction and Achievement in All Schools—A Leadership Brief

A Project of theLearning First Alliance

AuthorsWendy Togneri, Learning First Alliance

Stephen E. Anderson, Associate Professor,International Centre for Educational Change,Ontario Institute for Studies in Education of theUniversity of Toronto

Ordering InformationFor additional copies of Beyond Islands of Excellence: What Districts Can Do to Improve Instructionand Achievement in All Schools (Stock No. 303368) or Beyond Islands of Excellence: What DistrictsCan Do to Improve Instruction and Achievement in All Schools—A Leadership Brief (Stock No.303369), contact the Association for Supervision and Curriculum Development (ASCD) at 1-800-933-ASCD (2723), extension 2; in the Washington, DC, metro area, call 703-578-9600.You may mail your order to ASCD, P.O. Box 79760, Baltimore, MD 21279-0760 or order fromthe ASCD website at www.ascd.org.

The cost for Beyond Islands of Excellence: What Districts Can Do to Improve Instruction andAchievement in All Schools is $10 per copy plus shipping and handling.

The cost for the Beyond Islands of Excellence: What Districts Can Do to Improve Instructionand Achievement in All Schools—A Leadership Brief is $3 per copy plus shipping and han-dling (10 percent discount for orders of 10–49 copies; 15 percent discount for orders of 50or more copies). For member organizations of the Learning First Alliance and their state orlocal affiliates or members, orders of 400 or more copies are $1 per copy plus shipping andhandling.

Additional Learning First Alliance Publications Available from ASCD

Every Child Learning: Safe and Supportive SchoolsThis guide to safe and supportive schools outlines the core elements of safe and supportive learn-ing communities; systematic approaches to supporting safety and positive behavior; involvementof family, students, school staff, and the surrounding community; and standards and measures tosupport continuous improvement based on data. The guide concludes with recommendations forensuring that all students experience their schools as safe and supportive learning communities.(Stock No. 301279)

Every Child Reading: An Action PlanThis paper expresses the consensus view of Learning First Alliance member organizations regard-ing the teaching of reading. The report calls for an end to the “reading wars” and a sensible bal-ance between whole language and phonics approaches. It outlines eight steps necessary to ensurethe reading success of every child as well as a detailed action plan for making these things hap-pen. (Stock No. 300342)

Every Child Reading: A Professional Development GuideThis guide, which follows from Every Child Reading: An Action Plan, describes what teachersneed to know and be able to do to teach and improve the reading skills of students. It includesthe conditions that must be in place for teachers to improve reading instruction; the eight com-ponents of effective, research-based reading instruction in which teachers must have a thoroughgrounding; and examples of specific professional development experiences that enable teachers toacquire the knowledge and skills necessary to teach reading effectively. (Stock No. 300303)

Every Child Mathematically Proficient: An Action PlanThis action plan puts forward research-based strategies “to bring American students to world-class levels in mathematics.” To accomplish the report’s goal of virtually all students masteringthe content now included in algebra I and geometry by the end of ninth grade, the Alliance proposes several action steps to strengthen professional practice. (Stock No. 300343)

These Learning First Alliance publications are available from ASCD for $3 per copy.

Learning First Alliance publications, including this report, the leadership brief, and individualcase studies of the five districts in this report, may be downloaded free of charge at www.learningfirst.org.

Copyright © 2003 by the Learning First Alliance. Permission is granted to copy portions or allof this document. Each copy must include the above notice of copyright.

Printed in the United States of America.

LEARNING F IRST ALLIANCE

BEYOND ISLANDS OF EXCELLENCE: WHAT DISTRICTS CAN DO TO IMPROVE INSTRUCTION AND ACHIEVEMENT IN ALL SCHOOLS

Space unfortunately does not permit the Alliance to thank by name the many staff members ofour study districts or to outline the depth of their contributions. Many staff members in each

district gave generously of their time and expertise. We offer deep appreciation to Wanda Bambergand Nadine Kujawa of Aldine Independent School District; to Jon Baker, Lorraine Costella, andBonnie Ward of the Kent County Public Schools; to Lowell Billings, Dennis Doyle, and Libby Gilof Chula Vista Elementary School District; to Ginny Craig, Carol Johnson, and Rick Spicuzza ofthe Minneapolis Public Schools; and to Melody Johnson, Diana Lam, and Mike Sorum of theProvidence Public Schools.

Many people from the research and policy community generously and voluntarily gave of their timeand expertise. We thank Kermit Buckner, Gordon Cawelti, Tom Corcoran, Betty Hale, Rick Hess,Stephanie Hirsh, Ellen Foley, Susan Fuhrman, Willis Hawley, Susan Moore Johnson, DaleKalkofen, Bruce King, Michael Knapp, Diane Massell, Lauren Resnick, Linda Skrla, SamStringfield, Gary Sykes, and Michael Usdan. We also thank the members of the Learning First State Alliances who contributed their time and suggestions to this study.

This study would not have been possible without the guidance and hard work of the InternationalCentre for Educational Change at the University of Toronto’s Ontario Institute for Studies inEducation (OISE). We are deeply grateful to the OISE research team: Stephen Anderson, whohelped shepherd this study from infancy to completion, and Shawn Moore and Nancy Watson,who provided extensive guidance. We also thank Nina Bascia, Amanda Datnow, and AndyHargreaves of the University of Toronto for their important contributions to this report.

This study represents several years of work by many people affiliated with the Learning FirstAlliance. We thank staff members of Alliance organizations who contributed to the research design,data gathering, and drafting of this report, including Diane Airhart, Michael Allen, AgnesCrawford, Gail Dickson, Richard Flanary, Twanna Hill, Julia Lara, John Mitchell, Joyce Munro,Virginia Roach, Marilyn Schlief, Sylvia Seidel, Joan Snowden, Irina Terehoff, Adria Thomas, andAnn R. Walker. We would also like to thank several board members from Alliance organizations,including Mary Barter, Darrell Rud, and Mossi White. We are especially grateful to Gene Carter,Tony Rollins, Judy Seltz, and Mikki Terry for their guidance and support.

We wish to thank members of the Learning First Alliance staff and the team of editors. We offerour deepest gratitude to Ann Walker, whose perseverance contributed immeasurably to the comple-tion of this study. The careful editing of Anne Edwards, Pat George, and Bob Rothman contributedsignificantly to creating what we hope is a clear and engaging report. We thank Cathy Horn for herthorough review of the achievement data. We also extend appreciation to Abby Emerson, AdrienneGault, Marissa Hopkins, Lisa Lazarus, and Jennifer Oliva for their important contributions to theproject and its completion.

Finally, we are grateful to the U.S. Department of Education’s Office of Educational Research &Improvement and the Ewing Marion Kauffman Foundation.

Acknowledgments

BEYOND ISLANDS OF EXCELLENCE: WHAT DISTRICTS CAN DO TO IMPROVE INSTRUCTION AND ACHIEVEMENT IN ALL SCHOOLS

Acknowledgments

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1The Study and Its Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

The Districts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2

The Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

Challenges That Remain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8

II. Building the Will and Vision for Districtwide Instructional Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 Improving Instruction: A Systemwide Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

Strategic Visions Guide Instructional Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12

A Vision for Instruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

Developing Districtwide Curricula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

Decisionmaking Based on Data, Not Instinct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19

A Systemwide Approach and School Flexibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

III. Adopting New Approaches to Professional Development: Implementing Coherent and District-Organized Strategies to Improve Instruction . . . . . . . . . . . . . . . . . . .23 Principles for Professional Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

Networks of Instructional Experts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

Support Systems for New Teachers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26

Strategic Allocation of Resources to Improve Instruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27

A Fundamental Shift in Practice: New Approaches to Professional Development . . . . . . . . . . . . . . .28

IV. Redefining Leadership Roles: Multiple Stakeholders Drive Instructional Reform . . . . . . . . . . . . . . . . . . . .31 Establishing Sound Stakeholder Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31

Working on Working Together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32

Redefining Instructional Leadership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32

The School Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33

The Central Office . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34

Superintendents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34

The Union . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36

Principals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38

Table of Contents

Teacher Leaders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40

The State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42

Universities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .44

Parents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45

V. Challenges to Districtwide Instructional Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47

VI. Ten Lessons Learned . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49

VII. Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51

Appendix I Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57

Appendix II District Achievement Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59

Research That Informed the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71

LEARNING F IRST ALLIANCE

Table of Contents

Heroic principals who turn around low-per-forming schools, innovative charter schools thatbreak established molds, inspiring teachers whomotivate students to excel—those are the famil-iar prescriptions for improving student achieve-ment in high-poverty schools. While suchefforts may mean brighter educational futuresfor the children involved, they produce isolatedislands of excellence.

Our nation has a moral imperative to close theachievement gap between low-income studentsand their more advantaged peers. The NoChild Left Behind Act makes this a legalrequirement as well. Yet improving learningopportunities for all children will require morethan individual talents or school-by-schoolefforts. It will demand systemwide approachesthat touch every child in every school in everydistrict across the nation.

The Learning First Alliance calls for policymak-ers, practitioners, and the public to accept thechallenge of improving student achievementacross entire school systems. We believe thatsubstantial gains will result only if we recognizethat to increase student achievement, we mustimprove instruction and commit the politicalwill and resources necessary to develop dis-trictwide solutions. As a permanent partnershipof organizations representing parents, teachers,principals, administrators, local and stateboards of education, and colleges of education,the Learning First Alliance recognizes that suchimprovements will require both individual andcollective action. Without efforts to create suc-cess across school systems, far too many stu-dents will continue to languish. We find thatunacceptable.

Moving beyond islands of excellent schools tosystems of success will require that all thoseinvolved in education better understand whatthey must do to help students succeed. Stateleaders need greater knowledge about where totarget resources and how to set policies to sup-port entire school systems. District-level educa-tors—board members, superintendents, unionleaders, principals, and teachers—need guid-ance about policies and practices that willimprove instruction. And community membersand parents need good ideas about how to mosteffectively participate in, and support, high-quality teaching and learning.

The Study and Its PurposeTo address the need for better information, the Learning First Alliance studied five high-poverty districts making strides in improvingstudent achievement. Recognizing that effectiveinstruction is crucial to improving achievement,we were interested in learning more about howdistricts promoted good instruction across theirsystems. Furthermore, in keeping with the mis-sion of the Learning First Alliance, we startedfrom the premise that many actors play impor-tant roles in improving instruction and achieve-ment. Given these premises, we sought toaddress the following questions:

■ How did the districts create the will tobegin instructional reform?

■ What strategies guided their reform efforts?

■ In what ways did districts change theirapproaches to professional development?

■ How did interactions among the stakehold-ers facilitate or hinder instructional reform?

BEYOND ISLANDS OF EXCELLENCE: WHAT DISTRICTS CAN DO TO IMPROVE INSTRUCTION AND ACHIEVEMENT IN ALL SCHOOLS 1

I.Introduction

■ How was leadership distributed acrossstakeholders to facilitate improvement?

The DistrictsTo find answers to these questions, this studyexamined instruction and achievement in fiveschool districts across the country:

■ Aldine Independent School District, Texas

■ Chula Vista Elementary School District,California

■ Kent County Public Schools, Maryland

■ Minneapolis Public Schools, Minnesota

■ Providence Public Schools, Rhode Island

To select the districts, we used both primaryand secondary criteria. In applying our primarycriteria we sought districts that exhibited all ofthe following characteristics:

■ Success in increasing student achievementin math and/or reading over three or moreyears

■ Improvement in student achievement acrossgrade levels, races, and ethnicities

■ A poverty rate of at least 25 percent, asdefined by students eligible for free orreduced lunch

■ A reputation for effective professional devel-opment practices, based on recommenda-tions from education leaders

To identify the districts, we solicited recom-mendations from Learning First Alliance mem-ber organizations, education researchers, andnonprofit leaders. We received over 50 recom-mendations. To narrow the field, we conducteda careful review of district achievement data,using standardized test results from 1998–2000

as the primary data source. On the basis of pri-mary data, 14 districts emerged as potentialstudy sites. We then applied secondary criteriaby reviewing a mix of demographic factors,including size, geographic distribution, urbanic-ity, and union affiliation. We also sought dis-tricts whose current districtwide reforms hadnot already been studied. As a final measure, weinterviewed superintendents and staff develop-ment leaders to learn more about their profes-sional development work.1 After applying allcriteria, we selected the five districts listed here.

The selected districts varied in some importantways (see Table 1). They included a small ruraldistrict of fewer than 3,000 pupils (KentCounty) and large urban districts of more than45,000 students (Aldine and Minneapolis). Theannual expenditures of the districts ranged fromAldine’s $6,822 per pupil to the $10,854 perpupil that Minneapolis spent. The percentageof children in poverty ranged from 38 percentin Kent County to 80 percent in Providence.The districts also shared some important demo-graphic characteristics. All saw a rise in povertyover the past decade, and all experienced a sig-nificant change in the ethnic and racial makeupof their populations. Most shifted from pre-dominantly white communities to communitiesof increasing diversity, and many of the districtssaw a dramatic rise in overall student popula-tion. The districts also saw an increase in thepercentage of students for whom English was a second language.

Taken as a whole, the five districts demonstrat-ed improvement in academic achievement—asmeasured by test scores—across grades, sub-jects, and racial/ethnic groups. In Aldine, theachievement gap between white and black stu-dents who took the Texas Assessment ofAcademic Skills (TAAS) math test closed signif-

2 LEARNING F IRST ALLIANCE

Introduction

1 See Appendix I for a complete explanation of the study methodology.

icantly. In 1994, 65 percent of black studentsmet minimum expectations compared with 84percent of whites. By 2002, the percentage ofblack students meeting minimum expectationshad risen to 94 percent, while the number forwhite students rose to 96 percent. KentCounty, meanwhile, increased the proportionof students scoring “satisfactory” on state testsand was the highest-scoring district in the statein 1999 and 2000. Increases in achievementalso were evident in the other districts.2 Forexample, the percentage of fifth-grade blackstudents in Minneapolis passing the MinnesotaComprehensive Assessment of Reading rosefrom 14 percent in 1998 to 33 percent in2002. Chula Vista experienced similar gains. In1999, the percentage of second-grade Hispanicstudents scoring at or above the 50th percentileon the state-sanctioned Stanford 9 math testwas 37. By 2002, 55 percent of second-gradeHispanic students were rated at or above the50th percentile. In Providence, the percentageof fourth-grade students who met or exceededthe standard on the New Standards EnglishLanguage Arts Reference Exam rose 12 percent

from 1998 to 2002. (See Appendix II fordetailed information about student achieve-ment results for each district.)

In general, test scores in the five districtsreflected both the successes and the challengesof reform. For example, elementary studentsgenerally made steady progress, while studentsin the higher grades demonstrated more volatileperformance. Where districts had chosen a par-ticular subject area (e.g., reading) on which tofocus initially, gains were more readily seen onthe associated test. In addition, while all racialand ethnic groups made progress, it occurred at different rates. The gap between white andminority students closed at varying rates aswell. Finally, districts that had implementedreform efforts over longer periods of timeshowed clearer improvement than did the dis-tricts in which reform was a newer endeavor,such as Providence. Overall, the data revealedthat although the districts had not experiencedcomplete success, they had made districtwidegains—particularly at the elementary level—inimproving student achievement.


2 The trends must be considered with two caveats. First, in some cases, because certain subgroup populations (e.g., Native American students) were sosmall, data were not provided. Second, the data do not follow cohorts of students but rather track trends over time across groups.


Introduction

The FindingsLearning First Alliance leaders and researchersspent several days in each district and conduct-ed more than 200 individual interviews, 15school visits, and 60 focus groups. We foundthat districts implemented a strikingly similar

set of strategies to improve instruction. Sevenfactors emerged as essential to improvement:

1. Districts had the courage to acknowledgepoor performance and the will to seeksolutions.

Table 1. District Statistical Data: 2001–2002 School YearAldine Chula Vista Kent County Minneapolis Providence

Total budget ($) 391,362,709 182,325,535 21,828,714 664,480,530 254,492,680

Per pupil budget ($) 6,822 5,500 8,000 10,854 9,897

Number of schools 61 39 8 128 48

Number of students 52,520 23,132 2,795 47,470 27,192

Student racial/ethnic distribution (%)

White 9 20 70 26 17Black 34 5 27 44 22Hispanic 55 62 3 11 52Asian/Pacific Islander/Filipino 3 12 0 15 3Native American/Alaskan Native 0 1 0 4 1

Free and reduced lunch eligibility (%) 74 44 38 67 80

English as a second language (%) 23 33 1 24 22

Number of teachers (full-time equivalent) 3,496 1,122 179 3,629 2,100

Average salary ($) 43,732 48,644 50,240 49,190 NA

Average years of teaching experience 11 14 17 11 NA

Current and previous superintendents Nadine Kujawa Lowell Billings Bonnie Ward Carol Johnson Melody Johnson(2001–present); (2002–present); (2002–present); (1997–present); (2002–present);

Sonny Donaldson Libby Gil Lorraine Costella Peter Hutchinson Diana Lam(1986–2001) (1993–2002) (1994–2002) (1993–1997) (1999–2002)

School board 7-member 5-member 5-member 7-member 9-memberboard elected board elected board elected board elected appointed

at large at large at large at large board

Notes: NA = not applicable. Percentages may not sum to 100 due to rounding error.

2. Districts put in place a systemwideapproach to improving instruction—onethat articulated curricular content and pro-vided instructional supports.

3. Districts instilled visions that focused on stu-dent learning and guided instructionalimprovement.3

4. Districts made decisions based on data, notinstinct.

5. Districts adopted new approaches to profes-sional development that involved a coherentand district-organized set of strategies toimprove instruction.

6. Districts redefined leadership roles.

7. Districts committed to sustaining reformover the long haul.

Below is a brief overview of each finding. Laterin the report, we discuss these findings in moredetail.

FINDING 1: Districts had the courage toacknowledge poor performance and the willto seek solutions.The emergence of public reporting of testingresults drove many districts to look at studentachievement data in new ways, and they didnot like what they saw: low achievement, par-ticularly for poor and minority children. Ineach district, some combination of leaders—school board members, superintendents, and/orcommunity members—acknowledged poor per-formance, accepted responsibility, and beganseeking solutions.

That courage to acknowledge negative informa-tion was critical to building the will to change.Leaders said that in the past they had assumedthat their systems were effective and that allparticipants were doing the best they could.

Today, the willingness of leaders to questionpractices publicly has spurred stakeholders at alllevels to implement and support new strategiesto improve teaching and learning.

FINDING 2: Districts put in place a sys-temwide approach to improving instruction.To improve student achievement, leaders real-ized they would need to fundamentally changeinstructional practice. Teachers would need tobe more effective in helping every child suc-ceed, and principals, central office staff, andboard members would need to better supportteachers in their classrooms.

Before reforms began, the districts had neitherclear, well-understood goals nor effective meas-ures of progress. Supports to improve instruc-tion were haphazard. Today, much has changed.The most common components of these newsystems are:

■ A vision focused on student learning andinstructional improvement

■ Systemwide curricula that connect to statestandards, are coherent across grade levels,and provide teachers with clear expectationsabout what to teach

■ A multimeasure accountability system andsystemwide use of data to inform practice,to hold schools accountable for results, andto monitor progress

■ A new approach to professional develop-ment—one that involves a coherent anddistrict-organized set of strategies toimprove instruction

■ Instructional leadership distributed acrossstakeholders

■ Strategic allocation of financial and humanresources


3 Items 3–6 are core elements of finding 2, the systemwide approach. We have chosen to highlight these findings separately because of their impor-tance in district efforts to improve instruction.

■ Use of high-quality research to inform deci-sionmaking and practice

FINDING 3: Districts instilled visions thatfocused on student learning and guidedinstructional improvement.Acknowledging poor student performance pro-vided district leaders with the ammunition topush for change. The districts began by devel-oping visions to guide them down this path.The visions, while differing across the districts,shared four common elements:

1. Increasing achievement for all students

2. Improving instruction

3. Creating a safe and supportive environmentfor students

4. Involving parents and the community

What distinguished these districts was not theexistence of a student-focused vision but theextent to which—and the ways in which—thedistricts used their visions to guide instructionalimprovement. Visions were clearly outlined instrategic plans, board meeting agendas, schoolimprovement plans, and newsletters. Further-more, superintendents made it clear that thevision was to drive programmatic and financialdecisions at every level of the system. Most dis-tricts succeeded in embedding the vision intothe actions of stakeholders, particularly at theadministrative level.

FINDING 4: Districts made decisions basedon data, not instinct.Leaders determined that to improve instruc-tion, they would need to put in place systemsto assess district strengths and weaknesses.Therefore, the districts did three things:

1. They systematically gathered data on mul-tiple issues, such as student and school

performance, customer satisfaction, anddemographic indicators.

2. They developed multimeasure accountabili-ty systems to gauge student and schoolprogress.

3. They encouraged teachers and administra-tors to use data to guide decisionmaking.

As a result of these actions, principals, boardmembers, teachers, and central office staff in alldistricts exhibited significant use of data to guidedecisionmaking.

FINDING 5: Districts adopted newapproaches to professional development.The districts made remarkable shifts in theirapproaches to professional development. To vary-ing degrees, they all rejected the traditional, one-time workshop approach to developing theirteachers’ skills. Instead, they implemented coher-ent, district-organized strategies to improveinstruction. The strategies included the following:

■ Principles for professional development.Districts used research-based principles ofprofessional development to guide theirwork. They connected teacher and principalprofessional development to district goalsand student needs; based the content of professional development on needs thatemerged from data; and implemented mul-tiple strategies to foster continuous learning.

■ Networks of instructional experts. Districtssought to augment instructional leadershipby building well-trained cadres of instruc-tional experts among the teacher and prin-cipal corps. Principals were not expected tolead schools by themselves, and teacherswere not expected to work in isolation. Bycreating networks of instructional experts,including instructionally proficient princi-pals and teacher leaders (e.g., content spe-cialists, mentor teachers), districts increased


Introduction

their leadership capacity to assist teachers inimproving practice.

■ Support systems for new teachers. Districtsimplemented multiple strategies to assistbeginning teachers. In most districts, men-toring programs provided the main support,but many also included a series of seminarsand other assistance.

■ Strategic allocation of financial resources.Districts invested financially in their goalsof improving instruction and achievement.Before allocating their dollars, schoolboards, superintendents, and principalslooked carefully at how to stretch and pri-oritize their funds to address instructionalneeds.

■ Encouragement and assistance in using data.Districts provided teachers and principalswith better data—and with more assistancein how to use them to guide instructionalpractice.

FINDING 6: Districts redefined leadershiproles.District leaders determined that no single groupwould be expected to tackle instructionalimprovement alone. Instead, they redistributedleadership roles. Over time, the districts extend-ed leadership from traditional positions—superintendents and principals—to includeother actors: assistant principals, teacher lead-ers, central office staff, union leaders, andschool board members. In addition, most dis-tricts included representatives from universities,state offices, and communities in their leader-ship efforts. Leadership in the districts was notmerely shared; most stakeholder groups soughtto take on the elements of reform that theywere best positioned to lead.

■ School boards shepherded instructional improve-ment efforts. In many districts it was thecourage of the school board that jump-started

reform efforts. Yet the boards did not simplygalvanize change; they followed through bypromulgating policies that supported instruc-tional improvement (e.g., higher salaries forteachers and principals, mentoring programsfor new teachers, and systemwide curricu-lum). In most districts, the boards held thesuperintendent and staff accountable forprogress but did not engage in the dailyadministration of the reform effort.

■ Central offices drove systemwide change.Superintendents used central office policies,structures, and human resources to guideinstructional improvement. In most of thedistricts, central offices assumed roles thatthey were uniquely situated to fulfill—responsibilities that, if not taken up by dis-tricts, would have been left unperformed orhighly fragmented. Examples of such prac-tices included establishing strong principaltraining and support systems, coordinatingthe development of districtwide curriculum,establishing and implementing multimea-sure accountability systems, and creatingsystemwide supports for new teachers.

■ Principals and teacher leaders were crucial tothe districts’ systems of instructional leadership.Nowhere was the district commitment tobuilding instructional expertise more evi-dent than in the development of principalsand teacher leaders. Districts expected prin-cipals to act as the primary instructionalleaders at the school sites and provided sig-nificant support to help reach this ideal.Some districts required training in observingclassrooms, providing instructional feed-back, and using data. All districts regularlyconvened principals to share challenges,exchange strategies, and learn about emerg-ing issues. Today principals regularly engagein classroom observation, use data to analyzestudent performance and teaching strategies,and seek to build structures that encouragecollaboration.


To expand instructional leadership, districtsalso relied on teacher leaders. In each dis-trict, these individuals provided additionalinstructional support to teachers by model-ing lessons in the classroom, assisting strug-gling teachers, and providing materials andideas. Furthermore, they relieved principalsof administrative duties related to instruc-tion, such as professional developmentplanning and overseeing test administra-tion. In Aldine, for example, most schoolsused teacher leaders to analyze data and toexplain the analyses to teachers. As a result,teachers had a better sense of student per-formance and more information on whichto base instructional adjustments.

Teacher leader networks also deepened thecoherence of instructional practice amongschools within the districts. Districts fre-quently convened teacher leaders for profes-sional development and information shar-ing. As a result, those individuals were inti-mately engaged in districtwide planningand were able to incorporate new strategiesinto classrooms.

FINDING 7: Districts committed to sustain-ing reform over the long haul.The districts in our study understood that making a difference takes time. They set theircourses and stayed with them for years. Theyalso experienced remarkable stability in theirleadership. In three of the five districts, thesuperintendents who sparked change servedtheir districts for at least eight years.4 In addi-

tion, in most districts many board membersserved for 10 or more years. That continuityallowed superintendents and boards to growtogether in their approaches to change and tobetter understand each other’s work.

Even when superintendents left the districts,the boards chose to hire from within in order tomaintain continuity. In four of the five districts,the superintendent changed during the courseof this study and was replaced by a deputysuperintendent, usually one who had beenresponsible for curriculum and instruction.5

The original superintendents had served toshake up district practice. After their departure,the boards sought to sustain the reformsthrough continued stability in leadership.

Challenges That RemainWhile the story is largely one of forward momen-tum, a few caveats are in order. First, althoughthe districts demonstrated improvements, not allwere high achieving. Two districts, Kent Countyand Aldine, registered scores in the top tier oftheir states; the remaining districts experiencedimprovement at the elementary grades in mathand/or reading. Second, while not all districtsexhibited each characteristic we describe, at leastthree districts implemented each of the strategieswe outline in this report. Finally, while our studyconcentrated on district efforts to improveinstruction, the districts employed additionalstrategies (e.g., family support systems) that mayhave contributed to academic success but werebeyond the scope of this study.


Introduction

4 In Kent County, Lorraine Costella served as superintendent for eight years before retiring in 2002. Libby Gil spearheaded Chula Vista for nineyears before stepping down in 2002. Before retiring in 2001, Sonny Donaldson had been Aldine’s superintendent for 15 years. His deputy, NadineKujawa, who has worked in Aldine for more than 30 years, became superintendent after his departure. 5 Providence’s superintendent, Diana Lam, left her post in September 2002 to become deputy chancellor for teaching and learning for the New YorkCity Department of Education. She was replaced by her deputy, Melody Johnson. In Kent County, Lorraine Costella was replaced by her assistantsuperintendent for instruction, Bonnie Ward. In Chula Vista, Lowell Billings, a former Chula Vista assistant superintendent for both curriculum andfinance, replaced Libby Gil. As noted above, in Aldine Nadine Kujawa replaced Sonny Donaldson.


Although the districts in the study have madesignificant strides toward their goals, they stillface considerable challenges. We will addressthree challenges in a final section of this report.

* * *We do not presume that the study districtshave all the answers. Those we interviewed

were candid about the challenges they face andabout problems that impede their ability tomake even greater progress. Nonetheless, webelieve that this report will contribute toreform efforts in other districts by highlight-ing policies and practices to improve teachingand learning across entire systems.

Why were these districts willing to tackle reform?What distinguished them from other districtsthat operated under similar policy contexts andexperienced similar challenges? The answer is thepresence of key leaders who were willing toaccept ownership of difficult challenges and seeksolutions without placing blame. The leaders var-ied by district—school board members, superin-tendents, community leaders. But in each dis-trict, one or more leaders pushed their colleaguesto do something about the poor performancethat state and local test data revealed.

The emergence of public reporting of testingresults drove many districts to look at studentachievement data in new ways, and they didnot like what they saw: low achievement, par-ticularly for poor and minority children. Thecourage to accept this negative information wascritical in building the will to change. In thepast, district leaders had questioned neithertheir own practices nor the achievement of their students. Today, the study districts do not ignore disappointing data; instead, they usepublic accountability to spur stakeholders tosupport new approaches to improving practice.

The districts created the will for reform in dif-ferent ways. In Kent County and Aldine, themotivating factor was pressure to improve thestate test scores of poor and minority children.District leaders were dissatisfied with low stu-dent achievement and demanded change. Infact, these leaders demonstrated precisely whatadvocates of standardized testing had hoped:the data focused attention on student perfor-mance and encouraged educators to addresstheir responsibilities to improve achievement.The story in Aldine illustrates how the conflu-ence of leadership and state tests can galvanize

change. Prior to the implementation of theTexas Assessment of Academic Skills (TAAS),district administrators had not looked at stu-dent achievement in an organized way. Theysimply thought that students were performingwell. Yet the TAAS results of the early 1990sbelied that belief: Aldine, in fact, ranked nearthe bottom of the state. Data revealed substan-tial gaps in achievement: although white stu-dents scored relatively well, black and Hispanicstudents scored much lower. Concerned aboutlow levels of achievement, the superintendentand school board determined that a dramatical-ly new approach to teaching and learning wasneeded. Noted previous superintendent SonnyDonaldson in 1999:

[Eight to ten years ago] we didn’t have thedata that showed that not everybody was per-forming at the level they’re performing [at]today. We never disaggregated test scores tenyears ago. We had a black valedictorian atAldine High School.…We had Hispanic kidsthat were just outstanding students and wewould look at that and say, well, yeah,Hispanic kids are getting a fair shake in Aldinebecause we’ve got Hispanic kids that are doinggreat. But no they weren’t, because we didn’tlook at the data. (Koschoreck, n.d.)

In Minneapolis, by contrast, reform effortsstemmed from a steady accumulation of trou-bling data, coupled with strong community pres-sure. In the mid-1990s, data revealed that theachievement gap between white and minoritystudents was widening and that the high schoolgraduation rate was scarcely above 40 percent.Community leaders expressed deep concernabout student achievement and the equality ofopportunity for all students. In response, the


II. Building the Will and Vision forDistrictwide Instructional Reform

municipal government, the business community,the school board, and community advocacygroups demanded that the district take action toimprove performance. In perhaps the most dra-matic action, the Minneapolis Chamber ofCommerce refused to support the school board’srequest for a tax levy, contending that the per-formance data indicated that the district’s spend-ing was not producing adequate results. Inresponse, the school board and the communityhired a superintendent with a strong instruction-al background. Together the superintendent andother district and community leaders developeda set of performance indicators to rank andmonitor school progress. The greater level ofaccountability reinstated trust among internaland external partners and contributed to localsupport, district innovations, and, eventually,new levy proposals.

In all districts, regardless of the catalyst, leaderscame to the same conclusion: To improve stu-dent achievement, they needed to emphasize akey factor within their control—improvinginstruction.

Improving Instruction: A Systemwide ApproachLeaders realized that, in order to improveachievement, they would need to fundamentallychange both instructional support and instruc-tional practice. Teachers would need to be moreeffective in helping every child succeed, andprincipals, central office staff, and board mem-bers would need to support classroom effortsmore effectively. To address these needs, districtssought the necessary infrastructure to supportinstructional improvement. This step represent-ed a sharp break from past practice. Before cur-rent reform efforts, the districts lacked a univer-sal understanding of expected outcomes. Someschools had common texts, but no districts hadsystemwide curricula. Boards did not make

instruction and achievement central to theirwork. Principals were more likely to focus onthe operations of the school than on the activi-ties in the classroom. Teachers alone determinedtheir curriculum and instructional methods. Ifthey interacted with colleagues about instruc-tion, they did so primarily in hallways andlunchrooms, not in regularly scheduled meet-ings. Without a common base on which towork, teachers and principals received frag-mented guidance about instruction.

Today, much has changed. In general, districtsare engaged in building systems in which theparts coalesce to collectively support instruc-tion. While the components are not yet fullyimplemented, districts are making progress.The components of these new systems are:

■ A vision focused on student learning andinstructional improvement

■ Systemwide curricula that connect to statestandards, are coherent across grade levels,and provide teachers with clear expectationsabout what to teach

■ A multimeasure accountability system andsystemwide use of data to inform practice,to hold schools accountable for results, andto monitor progress

■ A new approach to professional develop-ment—one that involves a coherent anddistrict-organized set of strategies toimprove instruction

■ Instructional leadership distributed acrossstakeholders

■ Strategic allocation of financial and humanresources

■ Use of high-quality research to inform deci-sionmaking and practice

These new systems have resulted in changes inpractice. The vision of improving instruction


drives the work. Structures such as systemwidecurricula and a multimeasure accountabilitysystem provide a path for improvement and sig-nal expected outcomes. And a new approach toprofessional development has pushed teachersand administrators to use data and to assess andadjust instructional practice. Moreover, ourdata revealed that teachers and administratorsare acting in new ways. More principals areinstructional leaders, and more districts employteacher leaders to assist principals and teacherswith instructional needs. In the sections thatfollow, we examine these strategies in detail.

Strategic Visions GuideInstructional ReformFour of the five districts began their reformefforts by reassessing and revising their visions.Over time, those visions became the guidingforces of all strategic planning. The existence ofvisions in these districts was not particularlyunusual. What was notable, however, was theextent to which and the ways in which thesedistricts used their visions to guide instructionalimprovement. We will explore the role of thevision below.

Vision Statements and Strategic PlansIn crafting their vision statements, districts gen-erally developed both a one-line vision and abroader and written strategic plan. Each dis-trict’s one-line vision connected to the goal ofimproving student achievement. Throughoutthe interviews, teachers, principals, board mem-bers, and central office staff wove the one-linevision about student achievement into the dis-cussions: “all our students will achieve on gradelevel,” “we must increase the achievement of allchildren in our district,” and the like.Stakeholders had internalized the vision andspoke about it in their own words.

In addition to the one-line vision statements,

each of the five districts had a more detailed setof goals and strategies that comprised thestrategic plan. Four main goals emerged acrossthe districts:

1. Increasing achievement for all students

2. Improving instruction

3. Creating a safe and supportive environmentfor students

4. Involving parents and the community

Although the districts’ strategic plans wereshaped in part according to those four goals,their specific strategies varied. For example, thestrategic plans in Kent County, Minneapolis,and Providence contained both goals and theaction steps necessary to achieve the goals. KentCounty and Minneapolis added “indicators ofsuccess” so that they could gauge progress.


Building the Will and Vision for Districtwide Instructional Reform

Box 1:Excerpt from Kent County Strategic Plan

Kent County Strategic Plan—Goal 1: Kent CountyPublic School students will demonstrate knowl-edge of basic skills and higher-order thinkingskills to solve problems and communicate results. ■ Objective 2: Kent County students will score

at or above standard on state assessments. ■ Measure of progress: Maryland School

Performance Assessment Program (MSPAP,grades 3, 5, 8)

■ Indicator: By July 2001, 70 percent of stu-dents will achieve satisfactory standards onMSPAP at grades 3, 5, and 8.

■ Recommended steps to achieve the objectives:❑ Hold an assessment symposium in

January to analyze student achievementon MSPAP assessments.

❑ Use data to determine opportunities forimprovement.

❑ Schedule an evening informational meet-ing for parents to explain MSPAP testresults and show prototype test items.

❑ Monitor teacher use of newly developedcurriculum frameworks for grades 4/5.


Chula Vista, on the other hand, presented fivestrategic goals in conjunction with a set ofshared values under which all members of thedistrict were expected to operate. Furthermore,some plans evolved over time. For example,when Minneapolis determined that its initialplan was insufficient to guide its efforts to closethe achievement gap, district leaders developeda supplemental strategic plan to focus specifical-ly on minority students’ achievement. Examplesof the Kent County, Minneapolis, and ChulaVista approaches are outlined in Boxes 1–3.

Using the Vision to Guide ReformMost districts had a written vision, but to bringthat vision to life across multiple stakeholders

was a rigorous task. To accomplish deep under-standing and use of this vision, the districtsused four primary strategies.

Building the Vision by ConsensusTo craft their visions, districts sought the inputof educators and community members. Theyconvened community meetings, held focusgroups, and issued surveys to learn more aboutstakeholder goals for their children. They thencrafted vision statements that reflected theinput (see Box 4). In Chula Vista, for example,the superintendent and her cabinet of assistantsuperintendents spent a full year conducting18 focus groups with a myriad of stakeholders,including parents who lived in the district butsent their children to private schools. Seekingstakeholder input in a structured way was asubstantial strategy for building broad owner-ship of the vision.

Making the Vision VisibleIn these districts, the power of the vision wasin its use. The vision was written into high-profile documents that were widely disseminat-ed, such as strategic plans, board meeting agen-das, school improvement plans, and parentnewsletters. Perhaps because it was seen andheard in so many formats, the vision guidedthe thoughts and actions of most stakeholdersand became a part of daily decisionmaking inthe districts.

District leaders deeply integrated the vision andits strategic elements into their work. Leadersmade it clear that decisions, whether at theschool or central office level, needed to reflectthe strategic vision. When district leaders madeprogrammatic, personnel, or finance decisions,they continually referred to the strategic visionto guide their actions. Moreover, superintend-ents and assistant superintendents frequentlyvisited schools and discussed strategies to meetthe vision for improved instruction andachievement. As another example, staff develop-

Box 2:Minneapolis’s Twelve-Point Plan

for Improving theAcademic Performance and Graduation

Rates of Students of Color

The schools will:1. Use student data to direct action steps2. Ensure quality teaching and focused profes-

sional development3. Create a more diverse workforce4. Target resources to needy schools5. Restructure the secondary experience to

increase graduation rates and [support] thetransition to postsecondary [education]

6. Reduce over-referral to special education

Families and students will:7. Improve student attendance8. Strengthen family-school partnerships and

foster positive peer influence9. Leverage community partnerships

Together we will:10. Invest in school readiness11. Give students more time and greater

opportunities12. Increase support for students with behavior-

related issues

ers used the vision to ground the focus of theirprofessional development efforts.

Using Principals to Transmit the VisionIn all districts, principals were important liaisonsfor building systemwide understanding of thedistrict vision and goals. Central office leadersworked to ensure that principals understood thevision and could communicate it to teachers andparents. Yet principals did not simply implementthe district vision. Rather, they molded it to fittheir schools’ own contexts. Aldine, Chula Vista,and Kent County experienced particular successin infusing their visions and strategic plansthrough the principal level. This success was in

part attributable to the structured way in whichdistricts convened their principals to addressinstructional issues. In Chula Vista and Aldine,principals gathered weekly to share in discussionabout district goals and school-specific chal-lenges. As a result, those principals regularlyreceived common messages and continually dis-cussed improvement strategies with their peers.

Using Structures and Policies to Reinforce the VisionMany districts also employed a variety of toolsto deepen the use of their visions. Each districtembedded its vision into its policies. For exam-ple, district and school leaders were required to



Box 3:Chula Vista Elementary School District

Strategic Goals and Shared Values

Strategic Goals1. Literacy. All students will exit elementary

school as multiliterate lifelong learners with amastery of essential skills.

2. Equity. All students will have access to aca-demic programs and resources that will enablethem to achieve their full potential.

3. Collaboration. With the school as the center,the entire community will become full part-ners in education, responsible for each child’ssuccess.

4. Technology. All participants in the educationalprocess will have the resources and knowledgeto successfully participate in the information-based society of the twenty-first century.

5. Safe and supportive environment. All membersof the school community will enjoy a safe,caring, and stimulating environment.

Shared Values1. Equality. We believe each child is an individ-

ual of great worth entitled to develop to hisor her full potential. All children can and willlearn, and deserve equal access to a qualityeducation.

2. Equity. We believe there is no significant dif-ference in educational outcomes based onrace, gender, or economic status. Solutions,resources, programs, services, and support are

applied in a manner that develops the fullpotential of each child.

3. Accountability. We value and recognize indi-viduals who assume responsibility for anddemonstrate commitment and dedication toserving the interests of all children.

4. Ethical responsibility. We value each individualwho practices, teaches, and serves as a rolemodel of dignity, respect, honesty, integrity,and trust.

5. Diversity. We seek, encourage, and respecteach individual’s contributions and value amulticultural perspective.

6. Teamwork. We believe that families are theprimary role models for our children. We arecommitted to teamwork and collaboration toprovide maximum services for students, staff,and community members. This partnershipamong families, community members, andschool staff is the foundation of our children’seducational success.

7. Innovation. We are committed to challengingthe status quo and embracing a technologicalworld.

8. Excellence. We are committed to high stan-dards of performance throughout the districtand continuously seek and utilize new knowl-edge and skills.

use the vision to justify budgetary decisions, todevelop school improvement plans, and the like.

Depth of Vision SaturationIn general, leaders in each district harbored adeep understanding of the district vision. Yetwhile the board, central office staff, and princi-pals strongly embraced the vision, districtsexperienced a greater challenge in spreading thestrategic vision to the teacher corps. Aldine andKent County, the two most centrally drivendistricts, appeared to experience the greatestsuccess in building teacher understanding ofthe vision and strategic plan. This may in partbe attributed to their significant use of princi-pal and teacher leader networks, which madethe vision visible at the classroom level.

A Vision for InstructionAs the districts refined their overarching vision,they also sought to develop a more specificvision for good instruction. In general, instruc-tional visions were not a series of practices—forinstance, cooperative learning or direct instruc-tion—but rather a philosophy of practice. Morespecifically, district leaders sought to infuse areflective and evidence-based approach to teach-ing practice. This meant that they expectedteachers to actively engage students in rigorouscontent, assess the impact of instructionalmethods, reflect on their practice, work withcolleagues to research and share effective prac-tice, and make appropriate adjustments to helpstudents learn effectively.


Box 4:Building a Vision-Guided System—Kent County, Maryland

In the summer of 1994, Dr. Lorraine Costella arrived in Kent County as the new superintendentwith a mandate from the school board to increase the achievement level of the county’s children.Costella knew that the board and community were concerned about the low test scores that hadcome to light through the state’s accountability testing—the Maryland School PerformanceAssessment Program (MSPAP). However, she was less clear about the vision and goals of the stake-holders in her new community. She recognized that her first imperative would be to bring focus to adistrict that was operating under a 35-goal strategic plan. After meeting individually with principals,board members, and other stakeholders, she felt that individual meetings were not sufficient to movethe district collectively to a new focus. As a next step, district leaders convened a full-day meeting ofeducation stakeholders. Costella told the group of teachers, parents, board members, business lead-ers, and others that their charge was to think fully about what they wanted children in Kent toachieve. The meeting resulted in a series of outcomes and processes that were drafted into a new setof five strategic goals. This process has become part of the district lore of inclusiveness.

Like leaders from other districts in our study, Kent County’s leaders did not engage stakeholders inthe development of a vision and then let it sit in a closed binder on a shelf. Rather, the leadershipmade sure that staff members engaged with the new vision. The district formed strategic work teamscharged with setting performance indicators and strategies to achieve the five elements in the strate-gic vision. The board, superintendent, and central office staff used the vision as a guide to determinebudgeting decisions, to monitor progress, and to make hiring decisions. In addition, Kent Countyrequired that all of its schools use the districtwide structure of school improvement planning as avehicle to connect school goals and strategies with district goals.

The Kent County story is rooted firmly in the ability of district leaders to focus staff members onimproving instruction and achievement. Kent County began its efforts to create a shared focus bybringing stakeholders together in a strategic planning process. What emerged was a living strategicplan that prioritized increasing student achievement. With the strategic plan in place, Kent Countyleaders set about to determine ways in which they could bring this vision to life within the schools.



Two districts, Minneapolis and Providence, for-mally adopted written principles of practice. InMinneapolis, stakeholders throughout the dis-trict designed six Standards of EffectiveInstruction, which were ultimately included,with strong leadership from the MinneapolisFederation of Teachers, into the union contract:

1. Teachers are committed to students andtheir learning.

2. Teachers have a depth of knowledge of thesubjects they teach and how to teach thosesubjects to students.

3. Teachers manage and monitor studentlearning for continuousimprovement/progress.

4. Teachers reflect systematically about theirpractice and learn from experience.

5. Teachers participate as members of learningcommunities.

6. Teachers commit to professional develop-ment consistent with Minnesota Basic andHigh Standards, National Standards, andalignment of standards and goals.

Each standard included 5 to 15 indicators thatoutlined expectations for teaching. For instance,an indicator under standard 2 required thatteachers help students gain mastery of basic andhigher-order skills by “generating multiplepaths to knowledge through experiential activi-ties, discussions, study of text, interactive groupwork, inquiry, technology.” An indicator understandard 3 asserted that teachers must “createmotivating environments for student learningby using multiple instructional strategies.”

In Providence, district leaders adopted aninstructional philosophy developed by theUniversity of Pittsburgh’s Institute for Learning(IFL). The Principles of Learning (see Box 5) area research-based framework of the characteristicsof an effective learning environment. The districtused the principles as a core set of beliefs about

both pedagogy and the conditions of effectivelearning. IFL leaders explained that the princi-ples “[were] designed to help educators analyze[both] the quality of instruction and opportuni-ties for learning that they offer to students.” Likethe Minneapolis Standards, the principles wereaccompanied by indicators of how each principlecould be achieved in practice. For example, prin-ciple 5, academic rigor, encouraged teachers toreadily engage students in “high thinking” andrequire “active use of knowledge” from their stu-dents. Principle 9, self-management of learning,asserted that teachers must help students “man-age their own learning by evaluating the feed-back they get from others; bringing their knowl-edge to bear on new learning; anticipating learn-ing difficulties and apportioning their timeaccordingly; and judging their progress toward alearning goal.” While Aldine, Kent County, andChula Vista did not adopt written standards,their leaders expressed similar goals of using mul-tiple instructional strategies to build students’basic and higher-order thinking skills and ofusing reflective and evidence-based approaches to teaching practice.

In addition to their philosophies of practice, thedistricts also promoted a variety of research-based strategies to improve instruction. In KentCounty, the district promoted pedagogical meth-

Box 5:Principles of Learning

1. Organizing for Effort2. Clear Expectations3. Recognition of Accomplishment4. Fair and Credible Evaluations5. Academic Rigor in a Thinking Curriculum6. Accountable Talk7. Socializing Intelligence8. Learning as Apprenticeship9. Self-management of Learning

Source: Institute for Learning (© 2001 University of Pittsburgh).

ods associated with Baldrige in the Classroom—a classroom-level offshoot of Baldrige inEducation—that emphasized using data, settinglearning goals, and taking responsibility forlearning. In Minneapolis, among other strategies,the district emphasized activity-based learning inmath and science and was involved in a sys-temwide effort to promote arts integration as ameans of making the learning process moremotivating for students and more sensitive tovariations in student learning styles. In additionto the Principles of Learning, Providence leaderspromoted professional development activitiesthat introduced teachers to specific instructionalpractices associated with “balanced literacy.” InAldine, the central office provided professionaldevelopment to all teachers in the use of multi-ple research-based instructional strategies associ-ated with effective student learning. WhileChula Vista leaders did not promote a standardset of instructional strategies, leaders stronglyencouraged schools to adopt research-basedreform strategies (e.g., Accelerated Schools,Micro-Society, California Governors ReadingInitiative), most of which advocated specific andresearch-based pedagogical methods. In additionto the above examples, all districts sought toimprove teacher capacity to address the diverselearning styles of their students.

In these districts, the vision for instruction wasnot simply about what to teach and how toteach it, but also about how to assess the effectsof teaching on student learning. Central officeleaders pushed principals and teachers to try var-ious instructional methods, reflect on the efficacyof those methods, and adjust practice when datarevealed that change was needed. District leaderssought to address the reality that no singleinstructional method would yield results for allchildren. Therefore, they pushed teachers todevelop expertise in a range of proven instruc-

tional approaches and to differentiate their prac-tices as needed based on review of data.

Developing Districtwide CurriculaKent County, Minneapolis, and Aldine begantheir instructional reform with curricular over-haul. This reconstruction was a response in partto state standards movements and in part toresearch that revealed that teachers soughtgreater curriculum guidance. Leaving curricu-lum decisions up to individual schools had cre-ated difficulty for both teachers and students inthese districts. Teachers were uncomfortablewith the lack of guidance in how to reach stateand district standards. And students, because ofhigh rates of mobility, often encountered a cur-ricular maze as they moved to new schools.

To provide greater clarity on what to teach andgreater cohesion from school to school, the dis-tricts developed their own curricula, aligned tostate standards and district goals. In the mid- tolate 1990s, in the early stages of reform, thedistricts convened teams of teachers and admin-istrators to write district-specific curricula basedon state standards and local needs.6 These dis-tricts did not simply seek to build their newcurricula around the state standards. Teachersand administrators throughout the systemsengaged in lengthy dialogue about what theywanted to ensure that students learned withinand across each grade level (see Box 6).

As the process evolved in each district, teachersdeveloped lesson plans and sample strategies toprovide instructional guidance in the new cur-riculum. In addition, once the curriculum wasdeveloped, most districts did not simply passout documents to teachers. Rather, theytrained teacher leaders in implementation ofthe new curriculum and expected these leaders


6 The authors acknowledge that the existence of a districtwide curriculum alone does not guarantee rigorous curriculum or instruction. While stake-holders in these districts spoke of efforts to create strong curricula using both districtwide goals and statewide standards, full investigation of curricu-lar rigor was beyond the scope of this study.

to provide school site guidance. They also con-ducted districtwide training in the implemen-tation of the new curriculum in the classroom.

It was clear that administrators and teachers inthese three districts agreed that the new curric-ula provided coherent instructional guidanceconsistent with the districts’ instructionalvisions and that such guidance did not existprior to the standardization. Yet interestingly,teachers in the three districts did not reportthat the systemwide curricula constrained theirwork. Rather, many teachers asserted that theyhad significant pedagogical freedom within theframeworks. They explained that district lead-ers encouraged teachers to use their profes-sional judgment and skills to teach in waysthey deemed effective for their students. Thecurricula also helped them connect better

within their schools and increase the rigor oftheir instructional work. As one Minneapolisteacher put it:

[Because of the curriculum] we have moretypes of conversations going on about what weare doing. I might ask another teacher, “whenare you going to do that standard” so we cancoordinate. [Our] department meetings havetransformed from what they used to be, whichwas preoccupation with topics like “what bookare we going to buy.” That was all we caredabout, and then we would go back into ourrooms and close the door.

Similarly, another Minneapolis teacherexplained:

Today, we take a department meeting and dis-



Box 6:Knowing What to Teach: Creating the “Gospel” of Benchmarks in Aldine

In 1996, Aldine began to align its curriculum to the state standards framework, the Texas EssentialKnowledge and Skills (TEKS), and created what became known as the benchmarks. That intensiveundertaking both created a curriculum and helped staff develop as professionals. At the outset of theeffort, central office leaders asked principals to bring language arts representatives to a districtwidemeeting. Nadine Kujawa, then deputy superintendent, charged the school teams with going back totheir schools to work with teachers at every grade level. Teachers were to provide answers to twoquestions: What do I need to teach this year? and What do students need to know to be successfulnext year? Teachers met in grade-level groups and then across grade levels to answer those questions.In the summer of 1997, 40 language arts teachers reconvened at the district table, bringing withthem the teachers’ responses to those questions. During an intensive month-long summer effort, thisteam produced a specific K–12 language arts scope and sequence aligned to the TEKS. Respondingto teacher requests, the benchmarks evolved into six-week curricular sequences. The district startedwith language arts and math in the first year and then moved to science and social studies in thenext year. Now benchmarks exist in all core subjects as well as art and foreign language.

As teachers became more comfortable with the benchmark curriculum, both administrators andteachers determined that interim aligned assessment measures would be helpful for teachers. As aresult, teachers reconvened to write benchmark assessments—aligned tests in math and reading thatcould be administered every three weeks to check student progress. While the teachers and schoolswere required to use the benchmark curriculum, the benchmark assessments were optional.

In general, administrators viewed the benchmarks as a key to Aldine’s success. According to oneadministrator, “We really push the benchmarks, and we have not backed off. The benchmarks driveeverything we do.” The remarks from an elementary teacher spoke for many: “Benchmarks tell youexactly what to teach. And the units allow you creativity. They are more fun for teachers and kids.You can be creative with benchmarks, but they give you exactly what students should learn.”


patch of business and then move right intograde level teams. And so my [partner]English teacher and I are spending time oncurriculum, which didn’t use to happen. SoI’m in her business and she’s in mine.…Andwe like it.…I think the standards have helpedus to pry that door open.

While leaders in Kent County, Aldine, andMinneapolis saw systemwide curricula as a keystrategy to create cohesion in their districts,Chula Vista and Providence did not take thatapproach. Rather, they focused on building thecapacity of their leaders to drive instructionalimprovement. Providence leaders expressed adesire to move in the direction of a districtwidecurriculum, but because they were only threeyears into their reform efforts at the time of thestudy, the district had not yet moved forwardwith curricular reform.

Decisionmaking Based on Data,Not InstinctIn addition to refocusing their visions andaddressing curricular cohesion, districts soughtto dramatically increase their use of data todrive decisionmaking and improve instruction.Leaders determined that to improve instruc-tion, they would need to more readily assessstrengths and weaknesses in performance andinstruction in their districts. As a result, the districts sought to do three things:

1. Systematically gather relevant data

2. Build multimeasure accountability systemsto assess student and school progress

3. Use data to guide decisionmaking

Revamping the data and accountability systemsmeant identifying and acquiring multiple meas-ures of performance instead of simply relyingon end-of-year standardized test results.

Although state tests jump-started reform in sev-eral districts, the tests provided incompleteinformation. Districts sought to augment statetest data in two ways:

1. By providing and encouraging the use ofmultiple types of data, including:

■ An array of student performance data(e.g., grades, student work, end-of-unittest scores, suspension information,mobility rates, attendance, diagnosticdata)

■ Information on school and communityclimate and customer satisfaction (e.g.,external evaluations, parent and studentsurveys, community focus groups andsurveys)

2. By using formative data to provide anongoing picture of performance

The districts filled gaps in state testing systems.Providence and Kent County, for example,sought annual measures of performance andfilled grade-level gaps in their state assessmentswith the SAT-9 and California Test of BasicSkills (CTBS), respectively. Aldine was not fullysatisfied with the rigor of the TAAS and aug-mented it with the more rigorous Iowa BasicSkills Tests. Minneapolis developed its own test,the Northwest Achievement Levels Test, totrack how much a student had learned fromgrade to grade.

In addition to standardized test results, districtssought measures that would provide instruction-ally relevant information about student perform-ance throughout the year. Some districts institut-ed interim and diagnostic testing mechanisms tomonitor progress at different points in the year.Other districts used diagnostic tests in literacy tomonitor ongoing growth in achievement. Aldineimplemented a set of teacher-developed diagnos-tic tests connected to the districtwide curricu-lum. The optional tests were administered every

three weeks to assess student progress in mathe-matics, writing, reading, and other core contentareas. For all of the study districts, however,formative assessment meant more than testing.Teachers and administrators increased theirreview of student work as a vehicle to assess stu-dent learning. Other data, such as principalobservation feedback, retention statistics, andsatisfaction surveys, increasingly guided decisionsat the school and central office levels.

All districts used a combination of data both toincrease accountability and to guide instruc-tional improvement. Perhaps the most sophisti-cated accountability system among the districtswas that of Minneapolis, which used more than15 indicators to assess school progress. In addi-tion to a wide array of testing measures, theMinneapolis system included such indicators asattendance rates, suspension rates, and studentand staff perceptions of school safety. Schoolswere ranked according to their aggregateprogress on all indicators. Minneapolis leadersbelieved that their system provided them with amore accurate picture of school success thandid the state ranking, which relied solely on testscores. Some schools deemed to be failingunder the state system showed improvementunder the Minneapolis system, in part becausethe multiple indicators accounted for year-to-year improvement by individual students.

Using DataOur study revealed that these districts did notjust talk about data; they used them to guideimportant decisions about teaching and learn-ing, particularly at the central office and princi-pal levels. Noted a Providence administrator:

Our decisions are made based on data, quali-tative and quantitative. We look at studentachievement data on an ongoing basis. Weaddress it at principals’ meetings. OnThursday of last week we had a half-day dataanalysis session with all the [teacher leaders].

…We had voluntary professional developmentsessions on data analysis and interpretation.We use data all the time. The schools have todevelop a school improvement plan and allo-cate their budgets based on data.

In all five districts, staff used data to guide deci-sions related to instruction, such as budget allo-cation, staff hiring, and teaching and learninggap identification. At the school level, the degreeof data use varied, but there were promisingexamples in all districts. Principals and teachersanalyzed data to monitor progress, to determinethe effectiveness of their instructional approach-es, and to figure out where to make adjustments.Teachers looked to data to determine specificlearning patterns—for example, whether certainstudents exhibited difficulties in identifyingwords by sight, or whether they were still strug-gling with sounding words out. A teacher inKent County explained how data guided work:

We looked at our CTBS scores and ourMSPAP scores, and our reading scores wereflat. We needed a way to raise them. So themajority of everything that we are going tofocus on this year is reading. We looked intoresearch. Just last week, we visited a school inDelaware. They had a reading incentive pro-gram that was very successful for them. Wewent over and took a look at their practicesand decided we should spend money on thisapproach.

Why did the study districts seem to make bet-ter use of data than others? They made con-scious efforts to make data more usable and tomake users more comfortable with them. Tovarying degrees, districts implemented severalstrategies to foster data use:

■ Making the data safe

■ Making the data usable

■ Making use of the data



Making the Data SafeDistricts used data in part because leaderswere willing to accept the information thatdata revealed—whether positive or negative.In fact, leaders embraced data as tools to helpthem improve. As noted earlier, when test dataoriginally revealed that the districts were per-forming poorly, board members and superin-tendents took ownership of the data. Theyacknowledged the challenges highlighted inthe data and used the information to spurchange.

Districts not only accepted difficult data butreadily put themselves forward for study. Eachdistrict voluntarily commissioned surveys andagreed to participate in external research on itswork. Importantly, when challenging findingsemerged, the leaders modeled a willingness toshare the information and seek solutions. Forexample, in 2000, after a year of implementingthe Baldrige strategic planning process, KentCounty submitted to a comprehensive volun-tary review by an external team of educatorsand business experts. The findings revealedmany positive aspects of Kent County’s workbut also addressed areas that needed improve-ment. The report noted that the district lackeda comprehensive strategy for engaging teachersin decisionmaking around professional develop-ment. District leaders, rather than putting thereport on a shelf, sought solutions to the prob-lem. Within a year, the district had formed aprofessional development council made up ofteachers, principals, union leaders, and centraloffice staff. Furthermore, district leadersengaged teachers in the implementation of districtwide professional development.

Making the Data UsableThroughout our interviews, stakeholders spokeof the heightened pressures the new accountabil-ity system brought to their work. Teachers, inparticular, often noted that the demands relatedto data were daunting and overwhelming.

Finding time to disaggregate and digest the largereams of data was not possible for most teachersin most schools. Recognizing that, districtssought to provide data and data analysis toolsthat were easy to access and understand. Somedistricts supplied teachers and principals withinterpreted data reports, some assisted schoolsby funding intermediaries to help interpretschool-specific data, and others provided toolsto facilitate in-school disaggregation of data. InMaryland, the state also aided schools by estab-lishing a website that allowed school and districtstaff to interpret school-specific data.

Making Use of the DataWhile much of district-level professional devel-opment for teachers focused on content andpedagogical strategies, several districts promot-ed teacher professional development on datause. Kent County was particularly aggressive,dedicating several district professional develop-ment days, over the course of a few years, to aseries of assessment workshops. In that series,teachers and leaders convened to dissect dataand to talk about the implications of the find-ings. Thanks to assessment training, a KentCounty teacher noted, “[you] feel that you canlook at the assessments and control the resultsin your room. You are not at the mercy of amysterious force.” In addition to workshops,the assistant superintendent and superintendentregularly visited with principals and schoolleadership teams to discuss the implications ofstudent performance data. Kent County hadsome success with its data efforts. Teachersthroughout the system were clearly conversantin the potential of data use. They understood,at a basic level at least, how to interpret data. Insome schools and among some teachers, multi-ple sources of data were being used in sophisti-cated ways to assess learning needs.

* * *While district leaders heavily assessed studentprogress, they struggled to assess instructional


practice. We found little evidence that districtsformally assessed the impact of professionaldevelopment, for instance. Although princi-pals engaged in classroom observation andexpressed a greater confidence in their knowl-edge of the teaching practices in their build-ings, districts lacked a formal way of measur-ing the effect of professional development onteaching practice.

A Systemwide Approach andSchool FlexibilityAlthough the creation of an infrastructure forinstructional improvement might suggest thatthe districts imposed top-down reforms at theexpense of school-level flexibility, that does notappear to have been the case. Over time, dis-trict leaders determined that to improveinstruction, schools needed to have the flexibili-ty to hire teachers, to use funds, and to struc-ture their staffs and time as they saw fit.

This was particularly interesting given that thedistricts varied in the degree of their centraliza-tion. Aldine, the most centralized of the dis-tricts, provided teachers with curriculum frame-works that were divided into six-week units.Teachers and principals were expected to usethe curriculum in their classrooms, monitorprogress carefully by using data, and makeadjustments to instruction based on regularreview of data. Many schools used district-developed benchmark tests that were adminis-tered every three weeks and measured studentprogress toward standards.

By contrast, Chula Vista was purposefully decen-tralized. The district philosophy suggested that amajority of power be vested at the school level.In Chula Vista, schools had neither curriculumunder which they were required to operate norbenchmark testing. The district depended heavi-ly on school leaders, who were centrally support-ed and held accountable for reaching predeter-mined targets of student performance. WhileKent County, Providence, and Minneapolis wereneither as centralized as Aldine nor as decentral-ized as Chula Vista, they were oriented toward acentralized approach that provided varying levelsof curricular guidance, centralized professionaldevelopment, and the like.

Despite varying degrees of centralization, prin-cipals and teachers from all districts felt thatthey had a high degree of flexibility in theirwork. Noted one principal from Aldine:

The district leaves us open to be innovative.We can be as innovative as we are willing tobe.…I believe that it has been instrumental toour success.…Our school board and oursuperintendent have worked with us to ensurethat we have that flexibility. If a principalthinks a certain strategy is going to work…[t]he administration will let them go for it.

Fostering a balance between district-level sup-port and school-level flexibility to innovatewas a philosophy echoed by leaders through-out the districts. Leaders expressed the under-standing that, because challenges varied fromschool to school, school leaders would needflexibility to address challenges specific totheir environments.



BEYOND ISLANDS OF EXCELLENCE: HOW DISTRICTS CAN IMPROVE INSTRUCTION AND ACHIEVEMENT IN ALL SCHOOLS 23

As districts sought to improve instruction, newapproaches to professional development becameessential. A clear vision, curricular coherence,and improved data and accountability systemswere unlikely to have much impact unlessteachers in the classroom learned how to usethese supports to improve instruction for indi-vidual students. In the next pages, we describehow the districts made remarkable shifts intheir approaches to professional development.

When we began our study, we imagined thatthe term professional development would be toonarrow to capture what we would see—that thecommon understanding of professional devel-opment would not depict the broad range ofpolicies, practices, and supports the districtswere using to improve instruction. Our inter-views with hundreds of stakeholders confirmedthat belief. To varying degrees, all districts inthe study moved beyond the traditional, one-time workshop approach to professional devel-opment and put in place coherent, district-organized strategies to improve instruction.

In the early 1990s, prior to current reforms,districts engaged in traditional, fragmentedprofessional development practices. For exam-ple, central offices used their district profes-sional development days to offer a long menuof training opportunities. The opportunitiestended to be short term—a day, a couple ofhours—and offered little follow-up support.School-based professional development effortshad little connection to district-level profes-sional development. Teachers did not com-monly visit other classrooms or view practices

in other districts, and principals were not regu-lar figures in the instructional discussion.Decisions at both the district and school levelswere made without serious analysis of studentand teacher needs.

Today, the picture looks quite different. Itincludes deliberate strategies to use research-based principles of professional development,widespread use of data in decisionmaking, andclear connections between district goals andschool-level practices. This is in large part theresult of coherent strategies that districts putinto place to support and improve instruction.These strategies included the following:

■ Principles for professional development.Districts used research-based principles ofprofessional development to guide theirwork. They connected teacher and principalprofessional development to district goalsand student needs, based the content of professional development on needs thatemerged from data, and implemented mul-tiple strategies to foster continuous learning.

■ Networks of instructional experts. Districtssought to increase instructional leadershipby building well-trained cadres of instruc-tional experts among the teacher and prin-cipal corps. Principals were not expected tolead alone, and teachers were not expectedto work in isolation. By fostering networksof instructionally proficient principals andteacher leaders (e.g., content specialists,mentor teachers), districts increased theircapacity to improve instructional practice.

III. Adopting New Approaches toProfessional Development:Implementing Coherent and District-Organized Strategiesto Improve Instruction


Adopting New Approaches to Professional Development

■ Support systems for new teachers. Districtsimplemented multiple strategies, particular-ly mentoring programs, to assist noviceteachers.

■ Strategic allocation of financial resources.Districts invested financially in their goalsof improving instruction and achievement.Before allocating their dollars, schoolboards, superintendents, and principalslooked carefully at how to stretch and pri-oritize their funds to address instructionalneeds.

■ Encouragement and assistance in using data.Districts provided teachers and principalswith better data—and with more assistanceon how to use data to guide instructionalpractice.7

Principles for Professional DevelopmentAs central office leaders searched for strategiesto improve instruction, they quickly came tounderstand that they lacked principles to guidetheir professional development work. Districtleaders firmly believed in the research on effec-tive professional development, which, amongother things, called for professional develop-ment practices to be data driven, to providegreater opportunity for collaboration amongcolleagues, to push for greater reflection onpractice, to provide opportunities for continuallearning, and to use learning strategies appro-priate to participants. District leaders recog-nized that to change professional developmentpractices, they needed to rethink the criteria onwhich professional development was based.Over time, central office leaders began to pro-mote a set of research-based principles to guidethe way in which teachers and administratorsshould learn new practices.

Districts took different approaches to establish-ing principles. In Minneapolis, the union ledthe effort to put principles of professional de-velopment in place. Today, professional devel-opment principles are codified in the teachercontract and used as a tool by the central officeand the union leadership to promote more rig-orous, ongoing training for teachers. Whileother districts had not formally adopted profes-sional development criteria, leaders in these dis-tricts promoted the use of such principles. Mostdistricts outlined principles in written docu-ments, including grant proposals, strategicplanning documents, and school improvementplans. Using the same networks that helped tospread the vision, the districts created largecohorts of stakeholders who had internalizedthe principles. In the following sections, we dis-cuss further how the principles of professionaldevelopment framed the districts’ approaches toimproving instruction.

Networks of Instructional ExpertsAll five districts invested heavily in buildingnetworks of instructional experts. Doing thisrequired a casting off of traditional role expec-tations. Prior to district-level reform, instruc-tion was left largely to teachers in their class-rooms. In the old system, the instructional roleof the principal was idiosyncratic or nonexist-ent. Principals were largely building managers,and teachers had limited leadership roles.

Today, the districts are calling on central officestaff, principals, and teachers to change the waythey work and are building networks of instruc-tional experts from the central office to theclassroom. To understand the change, it is use-ful to examine two layers in the instructionalsupport network: principals and teacher leaders.

7 Because this strategy was also part of the districts’ systemwide framework for improvement, it was discussed in Chapter II. Therefore, we will notreaddress data use at length in this section.

Districts recast the role of the principals in theirsystems. Most of the districts actively sought toestablish clear expectations of the principal as aninstructional leader. In Providence and ChulaVista, central office leaders established writtenexpectations of principal work. In Providence,the district adopted the nationally developedPrinciples of Learning, created by the Institutefor Learning. By contrast, in Chula Vista districtleaders and principals worked together over sev-eral years to develop written standards for prin-cipals. The standards included outcome indica-tors outlining the tasks principals were expectedto accomplish, such as increasing studentachievement, creating supportive learning envi-ronments, and building the leadership capacityand professional skills of teachers. Whether thedistricts used written or orally communicatedexpectations, the goals for principal work werequite similar across all districts. Principals wereexpected to create environments conducive toreflective and rigorous teaching. District leadersintended that principals would create structuresthat allowed teachers to use data to assessinstructional practice and would work collabora-tively to share ideas. District leaders also intend-ed principals to become active instructionalguides, to observe teachers daily, and to provideregular feedback on their work.

To help principals meet these high expectations,central office leaders and staff provided princi-pals with significant support and training. In all districts, principals met regularly to discusschallenges and share ideas. Most districts estab-lished intensive training programs for princi-pals—by sending principals to respected train-ing academies and by bringing consultants intothe district to train principals in becoming moreeffective classroom observers and questioners.

While it would be incorrect to suggest that allprincipals met all expectations, significantchange in principal practice occurred across thedistricts. In interview after interview, principals

spoke about their work in terms of improvinginstruction, not simply managing the building.Many principals expressed greater comfort inregularly observing teachers and providing feed-back. They noted that they were increasingtheir use of data and encouraging more system-atic use of data among teachers. Explained oneMinneapolis principal:

It is my job to create a system of accountabilityso that teachers are more accountable. One ofthe things we do is conduct a reading sampleon every child in the building to assess thelevel [at which] each child is reading. Once thediagnostic teacher does that, she gives me a list.I review it until 9 at night. When I see chil-dren that are not reading at grade level, I makea note to the teacher. I will then sit down andask lots of questions of the teacher. What kindof instruction are you providing? Is this stu-dent [an English Language Learner]? What doI need to do to help you with this student?

Teacher leaders also represented a fundamentalcomponent in district efforts to improveinstruction. These leaders sat at both the districtand school levels and provided assistance as sub-ject matter specialists, mentors, professionaldevelopers, data specialists, and other experts ininstruction. School-based teacher leaders were aparticularly crucial support for principals andteachers. These teacher leaders worked closelywith other teachers to provide individual coach-ing, model lessons, and spread good instruction-al practice within a school. School-based teacherleaders also extended a principal’s capacity toobserve teachers, provide instructional guidance,and coordinate school-based professional devel-opment. Teacher leaders often assisted principalsin a variety of administrative tasks, such asscheduling, test administration, and data analy-sis. Whether at the district or school levels, thekey function of teacher leaders was to providean extra layer of instructional support to class-room teachers and to administrators.


To build stronger connections among the recastgroup of instructional leaders, each districtinstituted a variety of tools to facilitate commu-nication and learning. Districts brought togeth-er principals and central office staff regularly toshare ideas and discuss districtwide and school-based concerns (see Box 7). In most districts,principals met regularly (from once a month toonce a week). Such meeting structures assisteddistricts in making principals the channel

through which information traveled betweenthe central office and the schools. Bringingprincipals together to learn about district goalsand to share ideas increased the coherence ofgoals and strategies across the district.

In addition to structures to bring together prin-cipals, many districts nurtured networks ofteacher leaders. Districts brought togetherschool-based teacher leaders to take part in com-mon training exercises. School-based and dis-trict-based teacher leaders also met to plan andcarry out districtwide professional development.By developing and using teacher leader net-

works, districts were able to increase the level oftraining across the system on key areas of need,such as data use or increasing teachers’ capacityto address diverse learning needs. In general,these district-coordinated tools increased communication within schools, enhanced thecapacity of the central office to transfer ideas toschools, and deepened the engagement of schoolstaff in district-level policy and practice.

Support Systems for NewTeachersBecause of growth, retirement, and turnover,the study districts faced a continual stream ofnew teachers. Some districts in the study hiredbetween 200 and 500 new teachers each year. A central office administrator in Kent Countyexplained: “Half of our teachers will be gonewithin five years. And so every young teacherthat we can get, that we can nurture, we haveto do what we can.”

Four of the five districts implemented multiplestrategies—mentor programs, seminars for newteacher cohorts, peer observation opportunities,and the like—to provide support for this grow-ing cadre of first-year practitioners. Mentorprograms, however, were the primary strategyused. Although all mentor programs matchednew teachers with veteran teachers, the pro-grams varied widely in scope and intensity. Insome districts, mentors met weekly, while inothers they met less frequently. In some dis-tricts, a veteran teacher was paired with onenew teacher. Other districts employed full-timementor teachers to work with a caseload of 15to 30 new teachers. In general, mentorsobserved teaching practices, modeled lessons,and provided resources and general support. Inseveral districts, the programs were developedin partnership with teacher unions.

Although the districts understood the value ofsupporting new teachers, they struggled to



Box 7:Facilitating Communication in Aldine

Central office leaders nurtured the Aldinereform effort by building a structure thedistrict called vertical alignment. In 1994, inresponse to quickly growing enrollment andresearch on vertical K–12 alignment, thedistrict restructured itself into four quad-rants. The leadership created vertical teamsdefined by the high schools and their feederschools. The vertical structure was used tobring together principals and central officestaff to address district goals and to creategreater alignment in practices across gradelevels. Aldine supplemented the verticalstructure with a horizontal structure, bring-ing together leaders in grade-similar orschool-similar groupings. Principals andcentral office staff met in either vertical orhorizontal groupings weekly to share ideasand discuss district needs.

maintain these efforts in the face of budget cutsand diminished grant funding.

Strategic Allocation of Resourcesto Improve InstructionIn addition to investing in human resources,districts invested financially in their goals ofimproving instruction and achievement. Beforeallocating their dollars, school boards, superin-tendents, and principals looked carefully athow to stretch and prioritize their dollars toaddress instructional needs. District leadersalso allocated their time strategically and usedexternal resources carefully to maximize theirinvestments.

Using Meeting Time to Focus onInstructional IssuesTeachers, principals, and central office staff inthese districts came together fairly regularly.They met as grade-level teams, as principalcohorts, and as faculty. What distinguishedthese districts and many schools within themwas the way in which they used this meetingtime. When educators came together to work,they focused significantly on improving instruc-tion and student achievement. Many principalsand teachers noted that grade-level or facultymeetings were not gripe sessions; the meetingsfocused on students and how to help themachieve at higher levels. The use of data andclear goals helped to keep such meetings cen-tered on issues relevant to improving teachingand learning.

Allocating Financial Resources to ImproveInstructionIn addition to being prudent about the use ofmeeting time, the districts paid careful atten-tion to how they used their money. As theyallocated resources, they asked themselves questions about the resources they needed to

improve instruction. Because the districts hadclear goals and understood their challenges,they were able to channel financial resources toneeds. The districts worked hard to ensure thatfunds were available for principal training,teacher leader training, and other priorities tobuild instruction.

Using External Resources to Fund NewApproaches to Professional DevelopmentThe districts exhibited a strikingly pervasive useof external resources—both money and expert-ise—to undergird their efforts to improveinstruction. Much of the work described abovewould not have occurred without the assistanceof such resources, including federal, state, andprivate funds. Districts used external funds tobuild mentor programs, to augment principalsupport efforts, and to increase their teacherleader ranks. Such funds also paid to bringexternal expertise into the schools and increasedrelease time for teachers to work together.There was significant agreement among inter-viewees that professional development wouldnot have occurred at a significant scale withoutthose external resources. Explained one ChulaVista teacher:

We have been meeting before school and afterschool and on Saturdays. We did curriculumdevelopment over the summer. All of whichwould have been impossible without [external]funding, because without it you are asking theteachers to come in on their own time, with-out providing…any compensation.

The story of using external resources extendedbeyond funding. Districts made considerableuse of external expertise. Leaders in all districtswere diligent readers of research and seekers ofexternal partners to help them fulfill theirvisions. Superintendents pushed central officestaff and principals to seek input from research,and principals provided teachers with research.Noted a principal in Chula Vista: “We receive a


lot of support from the district. In my first yearin the district the superintendent probably pro-vided me with 20 helpful articles and books toread.” In addition to research, central officestaff, principals, and teachers sought out thebest practices of other districts. For instance,when Minneapolis leaders were seeking guid-ance on their high school reform initiative, theyvisited practitioners from like districts through-out the country, including Boston, Chicago,and Seattle.

Whether seeking external funds or looking forguidance, these districts were strategic in the useof outside resources. Central office leaders madeit clear that they did not accept funds or go afterexternal expertise simply because the resourceswere available. Rather, they looked at their goals,determined the programming and resources theyneeded, surveyed existing programs and struc-tures, and then sought assistance. Noted anadministrator in Aldine: “What we do is look forpeople who can take our goals to the next step.Unless they are willing to look at all our [stan-dards] and do it our way then we are really notinterested in having them in the district.”

While external resources provided districts witha powerful boost to their capacity-buildingefforts, these resources presented a double-edged sword. On the one hand, without suchresources the districts would have been unableto provide many of the types of professionalgrowth opportunities that drove their reformefforts. On the other hand, the heavy relianceon such funds presented challenges. Obtainingsuch resources created a drain on human laborin some of the districts, as it took considerablehuman investment to write and monitor grants.In addition, the districts’ heavy reliance onshort-term grants to fuel professional develop-ment created difficulty in sustaining someefforts. Throughout the districts, there wereexamples of programs that struggled to remainafloat after the funding had ceased.

A Fundamental Shift in Practice:New Approaches to ProfessionalDevelopmentNew support strategies across the districts led tonew approaches to professional development.Gone was the traditional single workshop. In itsplace, districts implemented a series of practicesorganized at the central office level to improveinstruction.

Although the professional development shiftwas not complete, districts were making goodprogress. Teachers and principals more readilyshared ideas. The content of professional devel-opment was heavily rooted in the weaknessesrevealed by student achievement data.Moreover, teachers worked more deliberately toassess student needs and adjust instructionalpractice. (See Table 2 for an analysis of theshift.) The thoughts of a Kent County teacherreflected those of many interviewed: “We arebeginning to work smarter. We are doing indi-vidual assessments and are identifying students’needs and tailoring instruction.”

More specifically, districts were no longer sim-ply offering workshops and sending teachers toconferences. They were shifting their practice tomeet the newly adopted principles of profes-sional development and were making use ofnew networks of instructional leaders. As anexample, districts changed the way in whichthey used traditional resources, such as district-level professional development days. Under thenew systems, districts focused professionaldevelopment days on a few key topics ratherthan on a large menu of disconnected issues.Central office and school staffs determined thetopics by looking at needs that emergedthrough the data. Furthermore, as a way tooffer additional opportunities for job-embed-ded professional development, the districtsshifted a majority of their release days back to




schools. In most districts, schools were asked touse the time to extend the discussion on dis-trict-level training but to do it in a school-basedcontext. Teachers and principals highly praisedthe shift.

Changes in professional development were notinitiated at only the district level. School-levelstaff followed the district example and adaptedthe new principles of professional develop-ment to their schools. In many schools, teach-ers and principals felt empowered to tacklechallenges together. They also expressed a pro-fessional responsibility to seek new ways toimprove instruction. Each district had manyschools that had created staffing and schedul-

ing structures that enabled teachers to worktogether effectively to address instructionalchallenges. These shifts were particularly evident in elementary schools. Furthermore,some schools used resources and policy oppor-tunities in ways that freed teachers for grade-level meetings, observations of colleagues, andthe like. Many schools used teacher leaders to provide coverage for peer observation orsmall group collaboration. As an example, anelementary school in Minneapolis used statecompensatory funds to hire a regular substituteteacher who worked weekly in the school andprovided coverage so that teachers couldobserve colleagues.


Table 2. New Strategies for Improving Instruction: Professional Development Characteristics Before and After Instructional Reform

Before Instructional Reform After Instructional Reform

Districts did not provide Districts implemented a framework to support instructionala systemwide framework improvement. Common elements included:to support good instruction. ■ A vision focused on student learning and instructional improvement

■ Systemwide curricula that connected to state standards and were coherent across grade levels

■ A multimeasure accountability system and systemwide use of data■ A new approach to professional development—one that involved

a coherent set of strategies to improve instruction■ Instructional leadership distributed across stakeholders■ Strategic allocation of financial and human resources■ Use of high-quality research to inform decisionmaking and practice

Districts lacked a set of research- Districts and schools used research-based principles to guide professional based principles to guide development implementation. professional development efforts.

Little connection existed between Significant connections existed between district visions and school strategies todistrict goals and school-based improve instruction.professional development.

Instructional leadership was diffuse, Districts created networks of instructional leaders that provided significantand leaders were not trained in support to teachers.a coordinated way.■ There was limited support ■ Districts expected principals to be instructional leaders and provided

for principals to become significant support.instructional leaders. ■ Districts formed networks of teacher leaders who provided instructional

■ Teacher leaders existed assistance to teachers, principals, and central office administrators.but were not used in a coordinated, explicit manner.

Data were not widely used to inform Professional development decisions at the school and district levelsinstructional and professional were based on needs that emerged from data.development decisions.

Professional development at the district Districts modeled research-based professional development.level was ad hoc. Districts incorporated the criteria of goal-focused, ongoing,

data-driven professional development by:■ Ensuring that districtwide professional development days had a

year-long focus and were connected to district goals■ Shifting districtwide professional development days to the school level■ Promoting action research among stakeholders

Resources were not used in a targeted Districts strategically used internal and external resources to improve instruction.way to leverage instructional Districts used data and their visions strategically to target the allocation of improvement. resources to improve instruction.


When we began our study we were interestednot simply in learning about how districts builtteaching capacity, but also in understanding theroles and interactions of the multiple stakehold-ers in this effort. What we found was that dis-trict leaders determined that no single stake-holder could tackle instructional improvementalone. The expansion of instructional leadershipdid not occur overnight. But during the courseof the reforms, the districts extended the leader-ship from a few traditional positions—thesuperintendent and principal—to include otheractors: assistant principals, teacher leaders, cen-tral office staff, union leaders, and school boardmembers. In addition, in most districts, externalactors—representatives from state offices, uni-versities, and communities—worked in a coor-dinated manner with district staff. In these dis-tricts leadership was not simply shared; moststakeholder groups took on the elements ofreform they were best positioned to lead. In thenext sections, we discuss how those qualifica-tions fostered mutual leadership.

Establishing Sound StakeholderRelationsFor stakeholders in the study districts, simply get-ting along was not the goal. Leaders in these dis-tricts determined that strong relationships heldlittle value if they did not create positive changefor children. As a result, a key goal of most dis-trict leaders was to learn to work together toimprove teaching and student learning.

Across the districts we saw evidence of success inreaching that goal. In most districts, boards andsuperintendents shared mutual respect and a

strong ability to work with each other. Super-intendents and central office staff repeatedlypraised boards for setting policies that providedthem with the flexibility to innovate. Principalsalso cited significant support from central officeleaders. In addition, across the districts, teacherleaders worked closely with district staff and prin-cipals to increase the level of instructional sup-port for teachers.

Three districts—Aldine, Kent County, andMinneapolis—developed particularly effectiveworking relationships across major stakeholdergroups. In Kent County and Minneapolis, theboard, union, and central office leaders workedtogether closely, communicating regularly aboutinnovations and challenges. In Aldine, the boardand central office shared a similar relationship.8

In those highly collaborative districts, leaders andpractitioners were able to introduce innovationmore readily and with less suspicion. Less timewas consumed attending to disagreements anddiscord, and more time and resources wereinvested in support structures for principals andteachers. Positive interactions in those districtsdid not mean that the stakeholders always gotalong or that they ignored challenges. In fact, anability to productively discuss differences wasmost evident in districts where the greatest har-mony existed among stakeholders.

Positive interactions between the central officeand union were less consistent across the dis-tricts than among other stakeholders in the system. Yet as we discuss later, where a strongrelationship existed between union and centraloffice leaders, it provided a powerful force forchange in a district.

IV. Redefining Leadership Roles: MultipleStakeholders Drive Instructional Reform

8 Aldine did not have an active union (fewer than 10 percent of teachers belonged to the union); our findings related to unions do not reflect datafrom that district.

Working on Working TogetherCollaboration and trust did not simply hap-pen in the districts; rather, they were the resultof deliberate and involved processes. Led bytheir boards and superintendents, the mostcollaborative districts in the study worked onworking together. They engaged in ongoingdialogue, created cross-role leadership struc-tures to facilitate communication amongstakeholders, and intentionally sought tools to facilitate collaboration.

In Kent County, for instance, the Baldrige inEducation process provided a framework forthe central office, the board, principals, andteachers to work together. The process helpeddistrict stakeholders at all levels set a collectivevision, develop strategies to reach the vision,use data to monitor progress toward reachinggoals, and act on challenges that surfaced.Central office staff worked closely with union,principal, and teacher leaders to set systemwidegoals. Simultaneously, teachers used the processto guide them in grade-level meetings. Theframework provided staff at all levels with anonthreatening vehicle to assess progress,acknowledge weaknesses, and tackle challenges.Noted one principal:

In the last two years since [Baldrige imple-mentation]…we’ve really been looking at con-tinuous improvement.…And we are allallowed as leaders of schools to make riskydecisions. Sometimes they work. Sometimesthey don’t.…We try something. We check itout. If it doesn’t work, we change it.

As another example, in Minneapolis, when theschool board sought to improve its work, itengaged in a training process called the Carvermethod. The Carver approach emphasized theboard’s role in establishing goals, setting indica-tors, aligning resources to the goals, monitoringprogress, holding the system accountable through

the superintendent, and communicating with thepublic. The process pushed the Minneapolisboard to speak with a common voice once deci-sions were made. It also provided the board withset norms that assisted its members in focusingon policy issues rather than engaging in the dailyadministration of district affairs.

In addition to using external tools to facilitatecoalition building, the most highly collabora-tive districts strove to be inclusive. Districtleaders seeking to implement new programs orpolicies included key stakeholders right fromthe beginning. For instance, when Kent Countybegan its work with the Baldrige process, lead-ers established a committee of teachers, princi-pals, union leaders, university colleagues, boardmembers, parent leaders, and other interestedparties. Stakeholders in the most collaborativedistricts were not simply informed about newefforts but involved in their development andimplementation.

Redefining Instructional LeadershipThough working well together was important,it was equally critical to be doing the rightwork. Over the course of their reform efforts,district leaders determined that improvinginstruction would require redefining leadershipand finding ways for all stakeholders to drivethat improvement. No single group would beexpected to tackle instructional reform alone.Instead, leadership would be shared, and mem-bers of each stakeholder group would take onroles they were best suited to lead. Elmore(2000) called that the theory of comparativeadvantage. Leaders, he asserted, should tacklearenas they are best positioned, or have a com-parative advantage, to lead. In his model, poli-cymakers should set performance targets andhold leaders accountable for reaching them.Central office staff should design systems andprovide support networks to assist school-basededucators. And principals and teachers should


Redefining Leadership Roles


have the freedom to engage in practices bestsuited to their talents and to the needs of chil-dren in their schools.

By and large, we saw districts working toward adistributed leadership of comparative advantage.Boards in the study districts focused primarilyon developing policies that supported instruc-tional reform but did not get involved in thedaily administration of district work. Super-intendents and central office staff supportedschools by providing more responsive trainingand by building and financing networks ofteacher leaders. And many principals used thecombination of district support and freedom to guide their staffs to assess challenges andrethink practice (see Table 3). In the followingsection, we explore this forward momentum aswell as some of the challenges in strengtheningand clarifying the roles of stakeholders ininstructional reform.

The School BoardSchool boards in most of the study districtswere policy and accountability driven. Boardsheld the superintendent and his or her col-leagues accountable for progress but did notengage in the daily administration of schools.Explained one board member: “I am not anadministrator; that is not my job. I am not aprofessional educator.…[The superintendentand her staff ] are the professionals, and we sayto them, ‘These are the results we want to see;you are in charge of how to do it.’ ”

In addition, the boards took their policy rolesseriously and promulgated policies to supportinstructional reform. For example, in ChulaVista, when the district’s top administratorsdetermined that they needed to create anexceptional cadre of principals, they asked theboard for help. As a first measure, the boardpassed a policy endorsing higher salaries forprincipals, which gave the superintendent con-

siderable leverage in attracting strong princi-pals to the district. In addition, the board pro-vided the superintendent with the flexibility toadminister raises and bonuses, and supportedthe superintendent in dismissing principalswho were not meeting performance standards.

As important as the issues on which these dis-tricts focused was the manner in which theyworked. Above all, the boards were driven by thegoal to improve student achievement. That notionwas not simply chimerical; the boards main-tained a focus on teaching and student learningneeds in their decisionmaking. They adoptedvisions and strategic plans that placed children’slearning needs at the center, and they attemptedto implement policies to support the strategicplan. An Aldine board member explained,“Everything we do is based on what’s best forthe children, period. Whether you are dealingwith an administrative issue or a student issue,we ask, ‘What’s best for the children?’”

Most boards also placed importance on speakingpublicly with one voice. The boards strove forconsensus and collegiality, and members actedrespectfully toward one another in public dis-course. Although philosophical differencessometimes existed, the norm of respect amongcolleagues guided their efforts to work throughthose differences. That cohesion was aided inpart by the composition of the boards; in fourof the five districts, board members were electedat large rather than as representatives of regionswithin the district. This appeared to help sus-tain collegiality and avoid jurisdictional divi-sions. In addition, the boards promoted a solu-tion-seeking orientation. Boards encouragedtheir staffs to tackle difficult issues and seekinnovative solutions. The collegial manner andsolution-seeking orientation of these boards set atone that permeated their districts. An adminis-trator in Kent County summed up the board’swork as follows: “The board recognizes its roleas a policymaker. [Board members] are very



professional. They never humiliate each other.They have no hidden agendas. The goal is whatis best for the children.”

In Aldine, Chula Vista, Kent County, andMinneapolis, the school boards were particular-ly effective, using their authority to shift districtagendas toward improving teaching and learn-ing. The boards were not only catalysts toreform efforts, but also significant forces in setting policy to improve district instructionalpractice. Yet this high degree of focus fromboards is not assured. Further study is neededto determine if new, instructionally focusedboard candidates are being cultivated and ifcommunity members fully understand theirimportance in sustaining instructionally focusedschool boards.

The Central OfficeIn all districts, the central offices were powerfulguiding forces for improving instruction. Indeveloping reform strategies, district leadersbegan to rethink and revise the core elements ofcentral office work. In many cases, leaders deter-mined that there were certain roles that centraloffices were uniquely positioned to play. In fact,leaders reasoned, if the central office did not takethe lead, the role would not be performed. Suchroles included creating a districtwide curriculum,building a high-quality principal corps, anddevising systemwide supports for new teachers.

As central offices undertook these new respon-sibilities, they paid considerable attention tocollaboration with other stakeholders. Theyengaged stakeholders at every level of the sys-tem in the design and implementation of inno-vations. As an example, when Aldine began itsefforts to align its curriculum to state standards,the central office worked closely with principalsand teachers. As a first step, district leaders con-vened principals to discuss the development ofthe districtwide curriculum and then askedprincipals to return to their schools to engage

teachers in discussions about expectations forstudent learning.

While superintendents sought to restructure thecentral office role, they paid close attention tothe interaction with schools. In general, superin-tendents used the central office to provide aframework of supports within which schoolswere encouraged to innovate and address chal-lenges specific to their buildings.

SuperintendentsA striking number of similarities emerged inthe actions of the superintendents in the fivestudy districts. They were visionary and visi-ble, instructionally focused, data oriented,and solution seeking. We address these char-acteristics below.

Superintendents in the districts were visionaryleaders. Although the urgency for change oftenemerged from boards and other external forces,the superintendents brought substance to theurgency. They pushed for greater focus onteaching and learning and introduced newstructures and policies to improve instruction.In most districts, the school boards were savvyabout understanding the need for instructionalreform and hired superintendents explicitly fortheir skills in the instructional arena.

Superintendents in the study districts spent thevast majority of their time focusing on improv-ing instruction and instructional supports. Theextent of their efforts was visible in the districtgoals and strategies for improvement. In mostdistricts, the superintendent led the charge tobolster the rigor of the curricula and to makeprofessional development more effective.Moreover, superintendents pushed stakeholdersto allocate resources to increase instructionalsupport. Superintendents also provided directsupport for instructional leadership, visitingprincipals regularly in their schools to celebrateimprovements and address challenges.


Stakeholders Observed RolesBoard members ■ Made decisions based on both students’ needs and data

■ Designed policies that supported instruction■ Held the chief administrator accountable for results but did not get involved

in day-to-day management decisions■ Engaged in productive, collegial discourse■ Confronted difficult issues and sought solutions

Superintendents ■ Viewed instructional reform as an important way to improve student achievement and central office ■ Designed systems to support school-level educators in improving instructionstaff ■ Built multimeasure accountability systems

■ Used data to guide decisionmaking■ Spearheaded efforts to align curriculum to standards and to create curricular

cohesion across the district■ Built a cadre of principals as instructional leaders■ Fostered cadres of teacher leaders to extend instructional assistance to teachers■ Implemented a system to support new teachers■ Reallocated resources to support instruction■ Tackled challenges by seeking solutions

Union leaders ■ Worked actively with district leaders to increase instructional supports for teachers■ Communicated with district leaders about teacher professional development needs■ Built trust across stakeholders by communicating visibly and regularly with

district leaders■ Introduced and supported research-promoted approaches to professional development

Principals ■ Provided instructional leadership to the school■ Transmitted and operationalized the district vision into the school building■ Used data to drive decisionmaking and to address instructional

strengths and weaknesses■ Engaged in collaboration with peers across the district on a regular basis■ Built a framework for productive professional development■ Championed induction by actively supporting efforts to recruit and retain new teachers■ Reallocated resources to support instruction

Teacher leaders ■ Extended the level of instructional support provided to teachers■ Served as a bridge between the administration and the classroom■ Assisted principals by overseeing administrative roles related to instruction

(e.g., data analysis, professional development planning)

State education ■ Provided seed funding for district-level professional development efforts for teacher leaders leaders, mentoring, principal training, and so forth

■ Engaged district leaders in developing statewide instructional supports (e.g., curriculum frameworks, literacy initiatives, mentoring programs)

Universities ■ Began to shift practice from working with individual schools to working withentire districts

■ Began to target assistance based on the vision and needs that emerged from district data

■ Began to partner strategically with districts

Parents ■ Engaged minimally in instructional improvement efforts

Table 3. Distributing Instructional Leadership: The Roles of Multiple Stakeholders in Improving Instruction

Note: The actions of board members, superintendents, principals, teacher leaders, and parents were observed in at least four districts. The actions of unionleaders, state education leaders, and universities were observed in two or more districts.

The power of many of the superintendentsemerged not only from their well-respectedideas but also from their willingness to take onchallenges. Across all districts, the superintend-ents were solution seekers. They looked to allresources—human, financial, and external—tohelp them attack a problem. Furthermore, theircommitment to seeking solutions, and notdwelling on problems, helped to create a solu-tion-oriented ethos in the districts.

One characteristic that stood out amongsuperintendents was their openness toaccountability and data as tools for improvinginstruction and achievement. With supportfrom their boards, superintendents did notrun from negative data and poor performance,but instead viewed such data as a wakeup call.As an example, Minneapolis’s superintendent,Carol Johnson, used data as a powerful tool toaddress low performance. In 1999, frustratedby poor results in high schools, Johnson askedher research department for a statistical reporton student failures by subject area. The datarevealed that the highest percentage of failureswas in mathematics and social studies. Shethen asked for data comparing student failureswith attendance and found that many studentsfailing courses at the high school level were inschool 95 percent or more of the time. Thatfinding ran counter to a previous publicpromise of the superintendent and board guar-anteeing academic success for students withhigh attendance rates. Instead of shelving thedata, Johnson took the information to the dis-trict’s high schools and presented it to teach-ers—not to accuse them of failure but to cre-ate an awareness of the situation and involvethem in seeking solutions. She pointed outthat the fact that student failure was focusedin particular subject areas meant that teaching,not just student backgrounds, played a role inthe low rates of achievement. By presentingthe data the way she did, Johnson showed awillingness to hold herself and school faculties

collectively responsible for understanding andsolving problems.

The superintendents were also skillful in build-ing and maintaining connections among stake-holders. For example, during her tenure inKent County, Superintendent Lorraine Costellaengaged stakeholders in regular and ongoingdialogue. She created the Baldrige leadershipteam, which brought together leaders fromacross the district to address needs and strate-gies for improvement. She met at least monthlywith the union president and board chair. And,she pushed all stakeholders to talk about diffi-cult issues. By working closely with the leadersthroughout the district, Costella raised the barof trust among those entities—a trust that facil-itated her work in reforming the district.

The UnionTeacher unions can play a variety of roles withinschool districts. Traditionally, teacher unions havefocused on bargaining for improved workingconditions—salaries, benefits, and length ofworkday. Another emerging union role is toadvocate for teacher professional growth needs. In the study districts, the unions’ focus varied. Insome districts, the union focused primarily onworking conditions; in others, the union washeavily involved in building better professionalsupports for teachers. Regardless, the study dis-tricts provided instructive examples of how unionand central office leaders can forge productiveagreements to improve instructional capacity.

One role union leaders played was to commu-nicate the voice of teachers to the board andcentral office staff. This was particularly criticalgiven the ambitious reforms established in thestudy districts. In the course of their changeefforts, the districts increased expectations anddemands on teachers. While the districts wentsome distance to provide additional time, train-ing, and instructional supports, there remaineda variance between the expectations on teacher




performance and the structural supports thatwere available to meet those expectations. As aresult, many teachers felt high levels of anxietyand were concerned that they did not have allof the tools needed to accomplish district aims.

In such cases, unions played a role in articulat-ing teachers’ concerns to district leaders. And inthese districts, the central offices and boardsgenerally heeded the union’s input and adjustedtheir strategies accordingly. Providence and KentCounty provide examples of unions voicingteacher concerns and acting as intermediaries tocreate solutions. In Providence, two key ele-ments of the district improvement strategy wereprincipal “walkthroughs” (structured, nonevalu-ative, instructionally oriented teacher observa-tions) and school-based coaches (teacher leaderswho modeled lessons, observed teachers, andprovided other supports). Some teachers, accus-tomed to privacy in their classrooms, objectedto principal walkthroughs and refused to extendinvitations to coaches. When those teacherscomplained to their union, union leaders carriedteacher concerns to the central office leadership.The central office responded by inviting unionleaders to participate in walkthroughs andobservational training and to accompany princi-pals and coaches into classrooms. Convincedthat the observations were both nonevaluativeand a useful instructional support for teachers,union leaders got the word out to members thatthe union supported the strategies, and the lead-ers encouraged teachers to welcome them.

A similar collaboration between Kent County’sunion and central office leadership helpedincrease productive supports to teachers. In theearly years of Kent County’s reforms, the centraloffice planned and implemented centralized,mandatory professional development. Teachers,frustrated by the quality of the professionaldevelopment, used sick and personal days toavoid the training. Aware of the frustrations, theunion president sat down with the superinten-

dent to voice teachers’ concerns. Workingtogether, the union and central office leadersdevised mutually beneficial solutions. The cen-tral office formed a professional developmentcouncil—made up of teachers, principals, andcentral office staff—to guide decisionmaking. In addition, over time the central office turnedover the design and implementation of profes-sional development to teachers. Seeing thoseshifts and realizing that the new professionaldevelopment was a valuable support for teach-ers, the union actively encouraged teachers toattend.

Unions played another important role in thestudy districts by working cooperatively with dis-trict leaders to increase instructional support forteachers. In Minneapolis, the union drove effortsto increase supports to new teachers, to providebetter guidelines for instructional expectations,and to create career ladders for teachers tobecome leaders. While establishing cadres ofteacher leaders was essential to improvinginstruction in all districts, in some districts thisrequired working through a number of issueswith union leaders: Would applicants for thesejobs be chosen based on seniority or based onperformance criteria? Would the leaders andcoaches have a role in teacher evaluation? Inwhat job category would they be classified? Inseveral districts, the union worked with thesuperintendent’s staff to answer these questionsin a way that was within the bounds of existingcontracts and met the need for capacity building.

Two districts—Kent County and Minneapolis—exhibited particularly robust partnershipsbetween the union and central office. In bothdistricts, the union leaders and superintendentssought to find common ground as well as tobridge differences. The union president andsuperintendent met monthly to discuss com-mon issues. While the fact that these leadersmet regularly was important, their ability tocommunicate openly and honestly about ideas

and challenges provided the foundation for pro-ductive engagement. Moreover, the ability tospeak honestly helped to ensure that concernswere voiced and solutions sought before theyescalated to grievances or created rifts at the bar-gaining table. Explained one union leader:

Probably one of the most meaningful thingsthat…developed between the administrationand the teachers’ association is the open dia-logue. [The superintendent] and I meet once amonth.…And we have been able to head offmajor grievances and major confrontationsbecause of our ability to work behind thescenes and within the framework to try toaccommodate people and to come up with thebest solution for all sides. And that’s been very,very effective.

A second key element of the partnership was thatin these districts the central office and the unionacted as partners in the planning and develop-ment of major initiatives. In Kent County, thecentral office engaged union leaders early in theplanning of several instructional reform initia-tives, and in Minneapolis, both the central officeand the union included each other in the designand implementation of such initiatives.

A third element of the productive partnershipin these districts was the role union leaders tookin advocating and supporting efforts to improveprofessional support for teachers. In Minne-apolis, the union was in the vanguard of devel-oping and implementing structures and initia-tives to build teacher capacity. The union tookthe lead in building a tenure process thatrequired high levels of professional develop-ment for teachers. Union leaders worked closelywith administrators on numerous grants andinitiatives to introduce greater opportunities forprofessional development into the schools,including a year-round induction program fornew teachers and the establishment of profes-

sional development centers in low-performingschools to provide greater resources to teachers.The union also initiated and supported the cre-ation of numerous teacher leader positions thatprovided teachers with a continuum of oppor-tunities as they progressed in their careers.

A fourth strategy, used primarily inMinneapolis, was the inclusion of key strategiesfor teacher professional development in theteacher contract. For instance, the Minneapolisteachers’ contract included the district’s princi-ples of professional development; it outlined amultitude of teacher leader positions, providedguidance on a complex peer appraisal system,and codified several joint committees that man-aged district professional development activi-ties. The contract language provided leveragefor both union and district leaders to push forstrong professional development.

In both Kent County and Minneapolis, thestrong relationship between the union presidentand the superintendent helped to model trustthroughout the district. Furthermore, the abilityof union and central office leaders to forge trust-ing and productive relationships signaled thatstakeholders needed to work together to addressproblems and create solutions.

PrincipalsDistrict leaders viewed principals as the primaryleaders of instructional improvement at theschool level. So, while not surprising, it wasencouraging that most principals interviewedfor the study described their roles in terms ofsupporting the instructional work of theirteachers. Throughout the districts, principals:

■ Provided instructional leadership

■ Used data to guide their decisionmaking andfostered the use of data among their staffs

■ Observed classroom instruction and provid-



ed teachers with nonevaluative feedback

■ Created structures and time for teacher collaboration

■ Partnered with cadres of teacher leaders tostrengthen instructional supports

■ Transmitted and operationalized the districtvision into the school

■ Refocused professional development tomeet district principles

While not all principals engaged in all the actionsabove, both principals and administrators spokeabout significant shifts in principal practice toprovide more instructional guidance to teachers.Observed one Kent County administrator: “Ourprincipals are among the best instructional leadersin the state. They are eager learners. They are notresistant to change, and they are thoughtful aboutchange. They learn from each other.”

Overall, principals sought to increase theinstructional assistance and feedback to theirteachers. Many principals, particularly at theelementary level, spent time each week observ-ing classroom practice. Some principals recon-figured their administrative teams to accom-plish this goal. In some schools assistant princi-pals shared curriculum and instruction dutieswith their principals, acting as supervisors forcertain subject matter. Principals also turned toteacher leaders—subject matter specialists,department heads, and other teachers withrelease time—to provide teachers with feedbackon instruction. These strategies increased thedegree of observation, lesson modeling, andfeedback to teachers. Noted one principal:

We spend a lot of time in the classroom. Weuse a lot of strategies to provide feedback. Iencourage…teaming with veteran teachers andnew teachers. The veteran teachers have a lotto offer, but the new teacher comes with newinnovations. So I encourage even my veteranteachers to take a look at new teachers and for

them to share ideas and strategies.…I give [my new teachers] a list of teachers that I wantthem to see.

Principals also worked closely with their staffs toreorient their approaches to professional develop-ment. Like the districts, schools were movingaway from one-shot workshops, and principalsand teachers sought new ways of engaging teach-ers in professional learning. As an example, prin-cipals spoke at length about the use of data as animportant starting point in assessing instruction-al practice. They worked readily with teacherleaders and assistant principals to disaggregatetest data and to review other forms of data onstudent achievement. Elementary principals inparticular spoke of how they used data in theirschools. Two Kent County principals remarked:

PRINCIPAL 1: Our best discussions take placewhen we look at student scores because youcan look and see why [we missed] thesekids. These are bright kids; why didn’t theyachieve at higher levels? We start to havethat discussion.

PRINCIPAL 2: When we looked at the data wefound out…that reading for literary experi-ence was our weakest area. So we had tothen become informed as to what we need-ed to do strategically in the classroom toimprove in this area, which meant lookingat things like current research and booksthat were provided to us in the area of read-ing. One that jumps out in my mind isLiteracy Constructing Meaning, and we onthe school improvement team purchasedthose books under the recommendation ofsomeone in the [Maryland] State Depart-ment [of Education] who was in charge of reading.

In addition, principals worked diligently tomotivate and acknowledge the good practice oftheir teachers. They did this in part by providingsupport for training and instructional resources.


Many teachers noted that principals worked hardto find the resources—money, release time, andsubstitute teachers—to allow teachers to attendprofessional development workshops relevant toschool goals and instructional needs. Principalsregularly provided teachers with written materi-als, manipulatives, and other instructional aids.Explained one Aldine teacher: “Our principalinvites us into her office. You go in and it’s likeChristmas; there are just piles [of resources].She’ll say… ‘just come in and choose what youwant.’” Some principals relied on regular, writ-ten communication or verbal praise. An Aldineelementary principal noted:

We’re really trying to encourage the imple-mentation of higher-order thinking skills andmore advanced problem solving. So when Isee that happening, I discuss it with peopleand let them know. I give a lot of praise andencourage them to share their good ideas.

The pressure on principals to produce resultswas significant in these five districts. If schoolswere not improving, principals were replaced.

Yet principals were not left to their owndevices. All districts provided supports to helpprincipals become astute instructional leaders.Several districts brought in outside experts totrain principals in observational techniquesand in classroom walkthroughs. These effortsprovided principals with specific tools toassess teaching practice and to provideinstructive feedback. Principals noted that thetraining increased their comfort with observ-ing and reflecting on teaching practice (seeBox 8). Districts also worked closely withprincipals to aid them in tying school-levelstrategies to the district agenda and in becom-ing effective analyzers of data. In addition,they worked with principals to increase theirknowledge and understanding of state anddistrict standards.

Districts also created formal structures to increaseprincipal dialogue across the district. In two ofthe districts, principals met weekly to share chal-lenges, exchange strategies, and learn aboutemerging issues. In other districts, principals metmonthly in full groups and in small peer groups.As a result of the formal gatherings, principalsthroughout the districts developed numerousinformal collegial networks. Principals repeatedlynoted that they relied heavily on a small group ofcolleagues with whom they connected by e-mailand phone weekly, and sometimes daily.

In addition to direct supports, the central officeassisted principals by cultivating teacher leadersand assistant principals to provide instructionaland administrative assistance in schools.Leaders in most districts stressed the instruc-tional importance of assistant principals in theirsystems. Assistants were not viewed as overseersof discipline and administration. Rather, inmany schools, they managed core areas of cur-riculum and instruction. Noted one ChulaVista administrator: “It used to be that associate[principals] just pushed pencil and paper. Todaythey are expected to be curriculum leaders.”9

In addition, teacher leaders offered similarinstructional and administrative assistance toprincipals in targeted areas.

Finally, most districts—concerned about a lead-ership shortage—sought to cultivate assistantprincipals and teacher leaders as the next gener-ation of leadership at the school and centraloffice levels.

Teacher LeadersAs districts attempted to improve instructionalpractice, they relied heavily on an emergingcadre of leaders—teacher leaders. As men-tioned earlier, most districts envisioned thisrole as a bridge between the administrationand the classroom, under the theory that



9 In Chula Vista, assistant principals are called associate principals to express a different philosophy of the associate principal role.

exemplary teachers who had recently left theclassroom would provide more meaningfulguidance to teachers.

Most districts employed teacher leaders at theschool and district levels. The roles of district-level teacher leaders varied. Some provided sup-port in a particular curriculum area, such as mathor reading. Others designed systemwide profes-sional development. Still others worked primarilyto provide assistance to struggling schools.

The benefits of the school-based teacher leaderswere many (see Box 9). Teacher leaders provided

teachers with additional instructional guidance.Among other tasks, they modeled lessons in theclassroom, assisted struggling teachers, createdlessons, and provided materials. In addition,they provided a layer of assistance to principals.By engaging in classrooms with teachers, teacherleaders deepened the principals’ capacity to offerteachers instructional support. In many districts,school-level teacher leaders also relieved princi-pals of administrative duties that related toinstruction, including professional developmentplanning, oversight of test administration, anddata analysis. In Aldine, for example, schoolsused teacher leaders to analyze data and explain


Box 8:The Principal’s Role in Chula Vista

Today, principals in Chula Vista are both building leaders and district liaisons, but that was not alwaysthe case. Not long after Dr. Libby Gil came to Chula Vista as superintendent in 1993, it became clearthat life for principals was going to change. She met with all principals and asked them for feedback ontheir jobs. She began to talk about her vision and expectations for principals. Under this new vision,principals would act as executive leaders of their buildings—instructional leaders, business managers,motivators, innovators, and developers of learning communities. Yet as she interviewed the principals,she found that a majority of them were working under a strict building manager model. Thus, the firstimperative was to shift the understanding of a principal’s role.

To help make this shift, Chula Vista leaders—central office staff and principals—engaged in a uniqueprocess to define the expectations for principals. Over the course of five years, these leaders came togeth-er to develop and revise a peer evaluation process that would define the new principal role—what itmeant to be an instructional and managerial leader in Chula Vista. The peer evaluation process wouldbecome the hallmark of Chula Vista’s effort to support principals and hold them accountable for success.Today the evaluation consists of multiple performance standards expressed through a four-level rubric.The principal is held accountable for ensuring student achievement, building staff capacity, ensuring cus-tomer satisfaction, acting with integrity and fairness, and creating a safe and nurturing environment.

To transition to the new way of working outlined in the peer evaluation, principals needed significantsupport from the board and central office. First, the board raised principals’ salaries and gave the super-intendent flexibility to administer monetary incentives to high-performing principals. In addition, cen-tral office leaders devised a multilayered support structure. Principals received training to assist inbuilding their skills as classroom observers, collaboration cultivators, data users, and the like. The dis-trict also facilitated a series of ongoing supports, including weekly meetings with small peer groups andweekly information packets containing relevant data and research.

The pressure on principals to produce results is significant in Chula Vista. Yet many principals reportedthat they came to Chula Vista expressly to work in what they described as an innovative, supportive,high-stakes system. Explained one principal, “You are given a lot of freedom, but there is a lot of liabilityand responsibility along with that.” This principal also asserted that he felt supported by the centraloffice: “The district is very upfront if you are having difficulties. It’s not about chopping you up; it’sabout assisting you to improve. And there are many opportunities to improve.”

the analyses to teachers. As a result, teachers hada richer sense of student performance and wereable to make instructional adjustments based onneeds unearthed from the data.

In addition to the individual benefits to theteacher, networks of teacher leaders expandedthe coherence of instructional practice acrossschools within these districts. Both district- andschool-based teacher leaders were trained at thedistrict level and often were involved in district-level decision meetings. As a result, theybecame intimately engaged in districtwidestrategies and were able to transmit these strate-gies to the classroom. In addition, teacher lead-ers increased the input of teachers in the designof districtwide instructional practices.

Districts used a variety of policy and fundinglevers to build their teacher leader ranks. In sev-eral districts, central office leaders mandatedthe use of teacher leaders in schools, particular-ly in low-performing schools. Some centraloffices shared the cost of these required posi-tions. In Minneapolis, for example, schoolsidentified as struggling were required to hirehalf-time reading and math resource teachers,with the costs shared between the school andthe district. In Chula Vista, principals wereevaluated on their ability to build teacher lead-ership capacity. In Aldine, the district expectedall schools to support at least one teacher leaderand encouraged schools to assign two teachersto the role. To help schools create those posi-tions, the district funded the cost of one teacherleader in each building.

While the growth of teacher leader positionsbrought a significant amount of expertise andresources to the professional development workacross the districts, teacher leaders experienceda few challenges. In many districts, the premisebehind teacher leaders was to increase theintensity of follow-up for district- and school-based training initiatives. While teacher leaders

appeared to increase instructional support, theystruggled to find time within the school day toprovide the desired level of follow-up. Thatwas, in part, because teacher leaders carriedhigh administrative and instructional supportloads. Principals and central office leadersplaced high expectations on teacher leaders toassist classroom teachers through demonstrationteaching, peer coaching, and observation.District leaders and principals also gave teacherleaders significant administrative work. As aresult of the administrative load, teacher leadershad less time for classroom work than theydesired. Despite such struggles, teacher leaderswere a crucial element of the instructionalreform efforts of all the districts.

The StateThe role of the state varied in the study dis-tricts. Several of the districts gave the statehigh marks for policy and administrative sup-port, while other districts felt that state poli-cies and practices distracted them from theirreforms. Yet, interestingly, all five districtstook advantage of state policy initiatives thatthey believed enhanced their reform efforts.

In three districts, Aldine, Kent County, andProvidence, district leaders described the state as playing a catalyzing role in their reforms. InAldine and Kent County, state standards andaccountability policies galvanized local leaders totake action in the face of poor test results. Statepolicies not only jump-started their reforms butalso provided varying degrees of guidance alongtheir journey to improve instruction. In largemeasure, leaders in these districts viewed thestate as partners.

In Aldine, for example, the district used a statelaw mandating mentoring programs for newteachers as a push to implement a needed pro-gram to support its large influx of novice teach-ers. Furthermore, when this policy came about,the district did not seek to minimally satisfy it



but rather worked to create the most productiveprogram possible. Likewise, when the state man-dated that primary teachers be trained in teach-ing reading, the district viewed it as an opportu-nity to further district goals around literacy.

The districts were not merely influenced bystate policies and practices; rather, they took aproactive approach to shaping policies at the

legislative and administrative levels. The dis-tricts charged highly placed staff with learningabout and influencing state legislation and stateboard of education policies. They also madesure that central office leaders sat on importantstate task forces concerning curriculum andinstruction. At the same time, the involvementwas reciprocal: the state departments of educa-tion actively sought district input. Across the


Box 9:Teachers as Instructional Leaders: Improving Literacy in Providence

In 1999, Providence teachers and leaders began to look at their literacy data, and they were notcomfortable with what they saw. Test scores had been flat, and leaders determined that a newapproach to instructional reform was needed to improve student achievement. As part of its literacy-focused reform strategy, Providence created a multitiered system of instructional support. Whileprincipals were to be the overall leaders, the district invested considerable resources in a network ofteacher leaders, which included both “instructional literacy coaches” and “lead team teachers.”

Employing a literacy coach in each elementary school was a key element of the district strategy toimprove student achievement. Yet appointing literacy coaches at each school was a massive undertak-ing that required deep collaboration between district and union leadership. In establishing the litera-cy coach position, central office and union leaders made three strategic decisions. The first was toseek candidates with both strong backgrounds in teaching reading and writing and good interper-sonal skills suitable to the “coaching” role. The second decision was to post jobs as district officepositions so that coaches could be selected based on their qualifications rather than by seniority pro-visions in the teachers’ contract for in-school teacher leader appointments. The third was to set up acommittee of principals and district officials to screen applications and interview candidates, toensure a rigorous selection process. Literacy coaches were appointed in the spring of 2000.

District leaders determined that the literacy coaches would fail without extensive training. To facili-tate training, the district hired the Institute for Learning (IFL). During the summer, the literacycoaches participated in an intensive training program with the IFL. The coaches were trained inboth change process skills (e.g., Learning Walks) and instructional content.

Lead team teachers were another crucial component of the district’s instructional support network.In the spring of 2000, elementary principals selected one lead team teacher per grade level to assistin the effort. Lead team teachers, trained by literacy coaches, were hired to help spread strong litera-cy practices throughout their schools. While the lead teachers received neither stipends nor workloadadjustments, they did get access to additional in-service training assistance.

Restructuring the district’s instructional capacity required changing district expenditures. By focus-ing existing funds on the literacy goals and seeking external funding from foundations, the superin-tendent acquired the financial support needed to implement the new plan. After a state waiverallowed class size reduction funds to pay elementary literacy coaches’ salaries, the positions werefunded by Providence’s regular operating revenues. However, a lack of funding prevented middle and high school coaches from being selected until the second year of reform implementation.

While it was too early in the reform effort to know the full impact of literacy coaches, Providence ele-mentary school teachers and principals praised the work of the literacy coaches and their potential toimprove instruction.

states and districts, there was a high level ofcommunication between state and district lead-ers. In Rhode Island and Maryland, for exam-ple, the commissioners of education metmonthly with superintendents from across theirstates. Those meetings were used as a forum toexchange ideas and address problems.

The role of the state in the Chula Vista andMinneapolis districts was mixed. While thetwo districts leveraged a significant amount ofstate resources to enhance their capacity-build-ing efforts, almost unanimously leaders inthese districts agreed that state policymakersdid not operate as full partners. In both dis-tricts, leaders outlined counterproductive poli-cies and practices from the legislature and thestate department of education. While leadersin Kent County, Aldine, and Providenceworked readily with state department leadersin shaping policies, Chula Vista and Minne-apolis expressed a low level of communicationand collaboration.

Despite those difficulties, both Chula Vistaand Minneapolis used multiple state policiesand programs to their advantage. For instance,in Minnesota, the state provided importantcompensatory funds to high-poverty schools.In addition, state policy mandated that dis-tricts set aside 2 percent of their budgets forprofessional development. In turn, Minne-apolis required that 2 percent of school bud-gets be set aside for professional development.Interestingly, we observed that leaders of sever-al turnaround schools in Minneapolis usedcompensatory funds to increase opportunitiesfor teacher collaboration and training. Thefunds paid for teacher leaders, professionaldevelopment materials, expert trainers, substi-tutes, and the like.

The story of Chula Vista helps to furtherexplain the complicated role of the state inthese districts. While Chula Vista leveraged a

significant amount of state resources to enhanceits professional development efforts, its leadersasserted that state policymakers did not focuson what they were best positioned to do.Explained one Chula Vista leader, “I think thestate department spends too much time dealingwith all of the elements of education whentheir main concern should be about results.” Inaddition, leaders suggested that state policy andpractice were at times counterproductive totheir efforts. An administrator referred to thestate accountability system as an example.“While the state system has introduced a levelof accountability, they have selected an off-the-shelf, norm-referenced test that isn’t alignedwith standards. Only about 40 percent of thestandards are reflected on the test.”

Despite the district’s ambivalence about thestate’s role, many Chula Vista schools reliedheavily on state programs and state funds toenhance their professional development efforts.State influence was seen most heavily in thearea of reading. Many of Chula Vista’s high-poverty schools used a state-developed and -funded program, the Governor’s ReadingInitiative, commonly referred to as Results, as their primary reading focus. A majority ofteachers in schools using the Results programtook part in both summer and year-roundtraining. The state funded both the trainingand stipends for teachers. In addition, someschools deemed “low performing” received statefunds. These funds were used in large part forprofessional development needs. State grantsalso provided a significant resource for district-level professional development efforts, includ-ing resources to enhance the district’s mentorprogram and to fully fund district specialists inmath and reading.

UniversitiesOver the past two decades the term “school-uni-versity partnership” has become part of the lan-guage of school reform. Traditionally, universities




in these partnerships have worked with individualschools or a cluster of schools to provide profes-sional development. Yet what we saw emerging inthe study districts was a more strategic set of dis-trict-university partnerships. Rather than workingwith individual schools, universities collaboratedat the system level to provide programmaticresources across the districts.

Because the study districts were savvy con-sumers of data, they had a deep sense of theirweaknesses and needs. As a result, districtsapproached universities with a much keenersense of the scope of aid they needed and wereable to leverage resources to support the col-laboration. Conversely, when universitiesapproached districts with a partnership idea,the central office was able to determinewhether and how the university offeringmatched district needs.

The districts and universities worked togetherprimarily in two areas—licensure and profes-sional development. With licensure, universitiesworked at the system level to help districts fillgaps in their licensed teacher corps. As an exam-ple, two of the districts sought to increase thenumber of licensed minority teachers in theirsystems and worked collaboratively with localuniversities to develop intensive alternative certi-fication programs for minority candidates. Asanother example, when the state of Minnesotaincreased licensure requirements for middleschool teachers, the Minneapolis Public Schoolsdetermined that it would have a shortage ofteachers certified to teach the middle grades. To mitigate the problem, the district and theUniversity of Minnesota entered into a partner-ship to provide a specialized certification pro-gram. The district and the university shared thecosts of instruction and credits and offered theprogram at no cost to current Minneapolis mid-dle school teachers. The classes were arranged tofit into teacher schedules. Furthermore, curricu-lum was tailored to the district’s needs.

Districts also worked closely with universities toaddress content and pedagogical needs. Somedistricts turned to universities for assistance inaddressing weaknesses in literacy, math, science,and other disciplines. For example, when datarevealed weaknesses in math, Aldine adminis-trators supported an effort by Rice Universityto assist with training teachers throughout thesystem in math content and pedagogy.

While institutions of higher educationincreased their strategic work with districts,they were not yet engaged as systemic partnersin instructional reform. Universities were gener-ally not involved in the strategic planningefforts of districts. Neither were the districtsreadily engaging with universities regardingteacher preparation. This is particularly impor-tant, since stakeholders in all of the districtsexpressed concern that many new teachers werenot well prepared in such areas as using data,integrating standards into instruction, and col-laborating with peers on instructional matters.Without systematic engagement with highereducation, these needs may remain unmet.

ParentsDistricts adopted a range of practices toengage parents in instructional reform efforts.Many of the districts sought to involve parentsat the outset of their reforms. In addition tobringing parents to the table in the visiondevelopment process, districts commissionedsurveys to learn more about parent concernsand ideas. Furthermore, they sought to informparents about reform measures throughnewsletters and forums.

In large measure, however, districts left parentengagement primarily to school-level staff.Schools included parents on leadership com-mittees, distributed newsletters to parents, andconducted homework nights that includedparents, children, and teachers. However, theseefforts were generally not systematic, and dis-



tricts had not developed significant policiesand practices related to parent involvement.

Indeed, despite the strong rhetorical commit-ment to parent involvement, most districtsadvanced their instructional reform effortswithout the robust engagement of parents.District leaders acknowledged that engagingparents around day-to-day schooling issuesremained a constant struggle, one that the dis-tricts had not adequately addressed. Some dis-trict leaders acknowledged that, early in theirreform processes, they had concentrated theirresources and energy primarily on improvinginstruction and had spent fewer resourcesengaging parents in the general reform effort.


Although the districts in the study have madesignificant strides toward their goals, they stillface considerable challenges. While we do notattempt to address all of those challenges, wehighlight a few significant ones that emergedacross all districts:

■ Old system structures do not easily supportnew approaches to professional development.

■ High schools struggle to improve achieve-ment.

■ Finding funding to support new approachesto instructional improvement remains difficult.

Old system structures do not easily support newapproaches to professional development. Ourinterviews suggested that leaders in the studydistricts expected school staff to take on moreresponsibility than ever before—to analyze dataand diagnose student needs, to determine theefficacy of their own instruction, to researchnew practices, and to collaborate frequentlywith colleagues. Yet district leaders had not cre-ated the full complement of supports neededfor teachers to meet these new expectations.

We saw clear attempts on the part of manyteachers and schools to live up to these expecta-tions. Many schools increased the level of collab-orative opportunity—carving out an hour or twoa week for shared work, for example. But it wasnot enough. Through our interviews and site vis-its, it appeared that only a limited number ofschools had significantly overhauled the schoolday to provide time for teachers and principals tofit the demands of their jobs into their workdays.In many schools, much collaboration occurredduring informal gatherings—in the hallways, at

lunch breaks, or in between classes. Noted threeteachers from Kent County in a dialogue aboutfinding time for collaboration:

TEACHER 1: People come early at our schooland people stay late to do it whenever youcan.

TEACHER 2: You have to meet on the sly almost,just whenever you get a chance.

TEACHER 3: You grab what time you can.

TEACHER 2: You drop a note in the mailbox orsend an e-mail to collaborate about lessons.The music teacher might ask me, “Whatdecades or what years are you studying nowso that we can…match up [social studies]with music?” Or the art teacher sends an e-mail to me, “What are you studying?” sothat she can coordinate it. But it’s done thatway, drop a note in the mailbox, do itthrough e-mail.

In many schools, the culture of working togeth-er took root more quickly than did the struc-tures to support it. Perhaps not surprisingly, thevariance between the expectations placed onteachers and the structural supports to meetthese expectations created great anxiety formany teachers in these systems. District leadersand teachers noted that both educators and pol-icymakers must develop new models of supportthat allow teachers to engage in the type ofinstruction, data analysis, and collaborationdemanded of them.

High schools struggle to improve achievement. Byand large, the success of the districts in thestudy was confined to elementary schools.Throughout the six- to ten-year trajectories oftheir reforms, the districts focused heavily on

V. Challenges to Districtwide Instructional Reform

Challenges to Districtwide Instructional Reform

instructional improvement at the elementarygrades, almost to the exclusion of the highschools. This appears to have been a deliberateeffort in most districts to tackle a subset of theirchallenges. Thus it was not surprising to seestudent achievement gains peak in the elemen-tary grades. By focusing on the earlier grades,districts created greater support for improve-ment at the elementary and middle school lev-els. As an example, in most districts, curricu-lum alignment was more complete in gradesK–6 than in higher grades. This may be in partbecause the districts were more comfortableconfronting challenges in the elementarygrades, and in part because the state standardsframeworks were more developed at the ele-mentary level. In addition, it appeared thatmore resources were invested at the elementarylevel—particularly for math and literacy.

While all districts started with elementaryreform, several districts were exploring newefforts at the high school level. Districts wereapproaching their high school reform in muchthe same way that they started their elementaryefforts. They were reviewing data, identifyingcore challenges, assessing the research, visitingother districts to learn about exemplary prac-tice, and making informed decisions. As highschool reforms in most study districts began in2000 or 2001, it is too soon to determinewhether they will take hold.

Finding funding to support new approaches toinstructional improvement remains difficult. Aswe have outlined throughout this report, thedistricts in this study engaged in a variety of

innovative practices to improve instruction andinstructional leadership. To bring such practicesto life, the districts relied significantly on exter-nal state, federal, and private grant sources.Providence, for example, received multi-mil-lion-dollar grants from private foundations tohelp train a cadre of instructional coaches andteacher leaders to deliver its literacy reform. InChula Vista, a three-year grant from the stateallowed the district to increase the central officecapacity and to hire a math instructional spe-cialist to train teachers in math content andpedagogy. Districts in the study were savvyabout finding additional funds to implementkey programs. And it is in part this know-howthat helped study districts to engage in profes-sional development activities not seen in otherdistricts.

However, while external resources provided dis-tricts with a powerful boost to their instruc-tional improvement efforts, these resources presented a double-edged sword. On the onehand, without such resources the districtswould have been unable to provide many of theprofessional growth opportunities that currentlydrive their reform efforts. On the other hand,the heavy reliance on such funds presented con-siderable constraints. Obtaining such resourcescreated a significant drain on human labor insome of the districts. In addition, such moniesat times came with strings that shifted the focusof carefully considered reform efforts. The dis-tricts’ heavy reliance on short-term grants tofuel professional development also challengedthe sustainability of these efforts.



At a time when districts nationwide face enor-mous pressure to raise achievement for all stu-dents, particularly those who have traditionallylagged behind their peers, educators and policy-makers are eager for ideas. The work of the fivedistricts in this study offers 10 important les-sons for those seeking to improve instructionand student achievement.

LESSON 1: Districts can make a difference. If asa nation we are serious about improvingachievement for all students, we cannotexpect the staffs of each of the nation’sapproximately 95,000 public schools to fig-ure out how to do this work on their own.As these five districts demonstrate, schooldistricts play an essential role in providing acoherent instructional framework to helpschools, particularly low-performingschools, succeed.

LESSON 2: Let truth be heard. These districtscreated a climate for change where it wassafe to acknowledge poor performance andsafe to seek solutions. They reviewed theirdata and publicly acknowledged the needfor improvement. Leaders neither madeexcuses for poor achievement nor wastedtime placing blame. Rather, they acceptedthe challenge of educating high-povertychildren and made sure that superintend-ents, principals, and other leaders sharedthis goal.

LESSON 3: Focus on instruction to improve stu-dent achievement. It is basic: students learnwhat they are taught; students will learnmore if they are taught well. Yet so oftenreform efforts look at everything excepthow to help teachers help their studentslearn. In these districts, reforms focused on

improving instruction, and this approach ispaying off.

LESSON 4: Improving instruction requires acoherent, systemwide approach. To help allschools improve achievement, district lead-ers created a framework of supports withseveral critical parts. They established aclear vision that focused teachers andadministrators on improving instruction.Districts then specified the outcomesexpected for students and schools; they cre-ated districtwide curricula to help teachersknow what to teach; they used data at everylevel to inform their work, and they createda coherent set of strategies to support andimprove instruction.

LESSON 5: Make decisions based on good data.These districts used multiple—not single—measures of student and school perform-ance to gauge progress and inform instruc-tion. Moreover, they helped teachers andadministrators learn to use the data effec-tively.

LESSON 6: Rethink professional development.These districts abolished traditional andineffective approaches to teacher trainingand replaced them with research-basedstrategies to improve teacher and principalskills. They used student performance datato guide what teachers needed to learn andcreated cadres of principal and teacherleaders to provide quality instructionalguidance.

LESSON 7: Everyone has a role to play inimproving instruction. No single stakeholderwas expected to lead instructional reform.Leadership was shared across the system,and stakeholders generally took on the lead-

VI. Ten Lessons Learned


Ten Lessons Learned

ership roles for which they were best suited.School boards focused primarily on devel-oping policies to support instructionalreform. Central office staff worked to sup-port schools; they provided significant sup-ports to principals to become instructionalleaders, they built and financed networks ofteacher leaders, and they facilitated struc-tures that encouraged collaboration.Principals guided their staffs to assess chal-lenges and improve practice, and teacherleaders provided instructional support andcoherence.

LESSON 8: Working together takes work. Theexpansion of leadership required significantcollaboration among stakeholders. Simplygetting along was not the goal; leadersdetermined that amity held little value if itdid not create positive change for children.Led by the efforts of their boards andsuperintendents, the most collaborative districts in the study worked on workingtogether. Districts deliberately sought andimplemented tools to guide collaboration.To be sure, not all of these districts involvedall of the stakeholders to the same degree,but the record so far suggests that the col-laboration of important stakeholders is vitalto school improvement.

LESSON 9: There are no quick fixes. Leaders inthese districts recognized that success wouldtake time and that they would have to stickwith their efforts for the long haul. Districtleaders encouraged practitioners to try newideas and did not expect immediate results.

Board leaders supported superintendentsover many years and many initiatives.Leaders assessed the impact of their effortsand made adjustments along the route.

LESSON 10: Current structures and fundinglimit success. Current district and schoolstructures do not fully provide the time andsupports necessary for systemwide instruc-tional improvement. Moreover, these dis-tricts’ heavy reliance on external and short-term funding to support their efforts putstheir continued success in jeopardy—andraises questions about how many other dis-tricts can follow in their path.

Although some may seem commonsensical,these lessons are important because they are notbeing applied systemically in our nation. Asthese districts illustrate, when these lessons areapplied, improvement in high-poverty schoolsystems is possible. These districts earned theirgood results. While the districts have not fig-ured out all the answers, they show that whendistricts support schools and plan carefully andcollaboratively, they can translate their visionsinto improvement—for their communities,their leaders, their teachers, their parents, and,most important, their students.

For the Learning First Alliance, these lessons arenot academic. They lead to an action agendafor the future. On the basis of these lessons, theLearning First Alliance has adopted a set of rec-ommendations directed to all those involved inimproving our nation’s public schools.


1. Mobilize political will to improve instruc-tion across the district; engage everyonefor the long haul.A. Use student achievement data to galvanize

political will.

B. Recognize that improving instruction isessential; create top-level support for instruc-tion among board members, superintend-ents, and community and parent leaders.

C. Allow for innovation that may not showimmediate results.

2. Implement a systemwide approach toimproving instruction that specifies the out-comes to be expected, the content to betaught, the data to inform the work, andthe supports to be provided.A. Develop a clear and concrete vision for

improving instruction districtwide, and use itto guide decisionmaking at all levels of thesystem.

B. Provide curricular guidance to help teachersknow what to teach.

C. Use data to assess needs, guide decision-making, and measure improvement.■ Create multimeasure accountability sys-

tems that specify desired student andschool outcomes.

■ Provide usable data to stakeholders.■ Train stakeholders to use data effectively.

D. Make professional development relevantand useful.

E. Align human, financial, and other resourceswith instructional priorities.

F. Be a savvy and active consumer of the bestavailable research and expertise.

3. Make professional development relevantand useful.A. Agree on and use research-based principles

to guide professional development.

B. Eliminate inefficient single-workshopapproaches to professional development.

C. Create a robust corps of teachers and prin-cipals who are instructional leaders.

D. Use data and research to guide professionaldevelopment content.

E. Create support systems for new teachers.

4. Redefine school and district leadership roles.A. Work together to ensure that stakeholders—

boards, central offices, unions, principals,teachers and teacher leaders, universities,and parent and community leaders—areengaging in the roles that they are bestpositioned to lead.

B. Build a network of instructional expertise,including a strong corps of principals andteachers as instructional leaders.

C. Focus the central office on developing a sys-temwide framework to support instruction.

D. Within a clearly defined district framework,allow schools the flexibility to make deci-sions based on data and to allocateresources as needed to address goals andchallenges.

5. Explore ways to restructure the traditionalschool day and year. Provide adequate time and supports for teachersand principals to carry out the new vision fortheir work and instructional improvement.

6. Attend to funding. Make funding for new approaches to profes-sional development central to district budgets,and call for dependable state and federal fund-ing for this essential work.

VII. Recommendations

We have outlined recommendations to help all partners in education address instructional reform.


Recommendations

Recommendations for IndividualStakeholdersThese six recommendations have important implica-tions for everyone with a stake in improving instruc-tion and achievement. Doing the hard work of dis-trictwide improvement requires all stakeholders tostep forward and lead where they are best posi-tioned to lead. As a beginning step, the Allianceurges stakeholders to consider the following:

School Boards1. Maintain the district focus on improving instruc-

tion and achievement.2. Work collaboratively with the central office,

union, and other leaders (1) to frame andimplement a district vision focused on instructionand achievement and (2) to adopt and useresearch-based principles regarding effectiveteaching and effective professional development.

3. Use data to regularly monitor the efficacy of theschool system. Hold yourselves and the centraloffice responsible for results. When results aredisappointing, seek solutions rather than assign-ing blame.

4. Hire top-level leaders—a superintendent anddeputy superintendent—who will lead instruc-tional improvement and will make decisionsbased on instructional and academic needs.

5. Set clear, coherent policies that support betterinstruction. Avoid involvement in day-to-daydecisionmaking that constrains the operation ofthe district.

6. Recognize that improving instruction and studentachievement is an ongoing process. Allow forinnovation that may not show immediate results.

Superintendents/Central Office1. Work collaboratively with the board, union, and

other leaders (1) to frame and implement a dis-trict vision focused on instruction and achieve-ment and (2) to adopt and use research-basedprinciples regarding effective teaching andeffective professional development.

2. Help to ensure adequate resources for districtneeds.

3. Make improving instruction and achievement theguide for decisionmaking and budgeting.

4. Inspire and encourage leadership at all levels ofthe system. Collaborate with leaders across thedistrict. Meet regularly with union leaders toaddress concerns and instructional issues.Create structures that bring together principalsfrom across the district to collaborate regularlyon improving instruction.

5. Take a systems approach to improving instruc-tion and achievement, and align core systemcomponents to support one another. ■ Provide clear curricular guidance to help

teachers know what to teach.■ Expect principals to be instructional leaders,

and provide significant training and supportto help them reach that ideal.

■ Foster networks of teacher leaders at thedistrict and school levels who provideinstructional assistance to other teachersand leaders.

■ Use research-based principles to guide pro-fessional development.

6. Assess the needs of teachers in the district usingteacher survey data, attrition rates, achievementdata, and other information. Propose and col-laborate on strategies that address these needs,such as induction programs, provision of differ-entiated professional development for veteranteachers, and development of teacher leaders.

Union Leaders1. Work collaboratively with the central office,

board, and other leaders (1) to frame andimplement a district vision focused on instruc-tion and achievement and (2) to adopt and useresearch-based principles regarding effectiveteaching and effective professional develop-ment.

2. Advocate for a system of teacher leaders thatcan provide needed supports to classroomteachers.

3. Assess the needs of teachers in the district usingteacher survey data, attrition rates, achievementdata, and other information. Propose and col-laborate on strategies that address these needs,such as induction programs, provision of differ-entiated professional development for veteranteachers, and development of teacher leaders.

4. Negotiate for contracts that support high-quality professional development, such as building career ladders for teacher leaders and creating strong induction programs.

Principals

1. Continually improve your skills in using data,observing instructional practice, providinginstructional feedback, motivating teachers, andso forth. Work with colleagues to advocate forgreater district-level supports and training.

2. Foster professional learning communities sothat teachers work and learn together as partof their regular practice. Encourage teachersto engage in research-based professionaldevelopment.

3. Use your resources to create teacher leaderpositions and employ teacher leaders to extendinstructional support in the school. Advocate forcentral office support for teacher leadersthrough district funds and contracts.

4. Make improving instruction and achievement theguide for decisionmaking and budgeting.

5. Support new teachers and act as a champion at the school and district levels for effective induc-tion practices.

Parent Leaders1. Demand data regarding student performance,

curriculum quality, teacher qualifications, thequality of instruction, fund allocation, and strate-gies to improve achievement.

2. Build parent and community support for instruc-tional reform. Help parents understand reform inthe district, the importance of instruction, andthe relationship between instructional improve-ment and student achievement.

3. Learn about why teachers need ongoing on-the-job professional development to improve studentachievement, and work with parents to supportit. Support policies such as early-release time oradditional funds to build the instructional skillsof teachers and leaders.

4. Actively support school board candidates whowill sustain the district focus on improvingachievement and instruction.

Universities1. Aggressively pursue opportunities to be an

effective long-term partner in the design, devel-opment, implementation, and evaluation of thedistrict’s professional development plan.

2. Ensure that teacher candidates are prepared toteach in high-poverty school districts.Candidates should receive the training theyneed to use data effectively, to work successfullywith parents and peers, to meet the learningneeds of diverse students, and to teach in stan-dards-based and assessment-driven schools.

3. Ensure that aspiring administrators are pre-pared to design and implement research-based,systemwide approaches to improve instruction.

4. Revise promotion and tenure policies to supporteffective district partnerships.

State Departments and State Boards of Education1. Recognize and support the pivotal role that dis-

tricts play in creating success across all schools.Ensure that state policies do not circumvent thedistrict. Engage district leaders in developingand implementing state policies and programs.Deliver technical training on a districtwide scaleinstead of focusing on individual schools.

2. Support district efforts to align standards,assessment, curriculum, and professionaldevelopment.

3. Support instructional leadership and high-quality professional development across schoolsystems. Allocate the time, policy supports, andfunding to support research-based professionaldevelopment.

4. Encourage districts to use multiple data meas-ures to gauge student success. Model the use of multiple measures for decisionmaking at thestate level.

5. Provide student performance and other data to dis-tricts in a manner that allows districts to use themeffectively for instructional improvement. Workclosely with districts to determine how they usedata and how better to respond to their needs.

6. Help build district capacity and expand leaders’opportunities to learn from each other. Highlightimproving districts and the lessons that can belearned from them. Bring district leaders togeth-er to share ideas.


Next Steps for the Learning First AllianceTo ensure that the implications of this reportand our recommendations are understood andimplemented, the Learning First Alliance and itsmembers will undertake the following actions:

1. Disseminate the findings and recommenda-tions of this report broadly to educators,policymakers, parents, and the public.

2. Consider the implications of the report foreach organization’s work and policies.

3. Encourage school districts and states to usethe report for learning, reflection, andaction. To accomplish this, the Learning FirstAlliance will share the findings and recom-mendations with superintendents, schoolboard members, principals, union leaders,parent leaders, university deans of education,chief state school officers, governors, andstate school board members. We will developtools to assist stakeholders in considering theimplications of this report. In addition, wewill partner with interested Learning FirstAlliance partner states to convene state poli-cy roundtables for this purpose.

4. Build greater understanding of newapproaches to professional developmentand address the ways that stakeholders willhave to work differently to improveinstruction. As part of this effort, we willidentify the implications of this report forspecific stakeholders, such as principals,board members, colleges of education, andstate policymakers.

5. Address key challenges identified in thereport:

■ Advocate at the federal, state, and locallevels for sustainable funding to createcoherent systems of instructional sup-ports such as those identified in thisstudy.

■ Examine in greater depth the challengescreated by attempting to carry out newprofessional development practices with-in the current school structures.Acknowledge and address the fact thatcurrent practice does not provide ade-quate opportunity for teachers and prin-cipals to carry out the new demands oftheir work—to analyze data and diag-nose student needs, to determine theefficacy of their own practice, to aligntheir instruction to new curriculumstandards, and to collaborate regularlywith peers.

6. Synthesize existing research on districtwidereform to make such information accessibleto practitioners and policymakers.

7. Call for high-quality research to answerimportant questions that practitioners andpolicymakers wrestle with as they seek dis-trictwide success:

■ Time and structures for instructionalimprovement. What are the underlyingreasons why schools struggle to buildstructures and find time to implementnew principles of professional develop-ment? How are schools and districtsusing time and resources differently tocreate structures that allow for the newkind of professional development out-lined in this report?

■ Continuity of leadership. What strategiesare districts using to create and sustaineffective instructional leadership? Whatare the most significant barriers to sus-taining instructionally focused leaders atthe board and central office levels? Howare community and parent leadersinvolved in sustaining board and centraloffice leaders?

■ Funding. Given the highly complex androbust set of instructional supports need-ed to improve instruction systemwide,


Recommendations


how much does it really cost to enactdistrictwide instructional reform? Whatare the most cost-effective ways of train-ing leaders and providing ongoingteacher training? Are some componentsof instructional reform more essentialand cost-effective than others? To whatextent does current state and local fund-ing cover the cost of research-based pro-fessional development?

■ Measuring changes in instructional practice.How can districts hold principals, teach-ers, and administrators accountable forhigh-quality instructional practice? Towhat extent are teachers changing instruc-tional practices as a result of new ap-proaches to professional development?Which tools could help districts measurethe impact of professional developmenton instructional practice and instructionalleadership?

■ Curriculum rigor and alignment to statestandards. Given the variability of therigor and quality of state standards andassessments, what are effective strategiesfor districts in states with low-level or

narrow standards and/or assessments toensure high-level curricular content?

■ High versus low capacity. Strategies forinstructional reform are not universallysuited to all schools and districts; somehave a greater capacity to carry outcomplex reform than others. To whatextent are districts assessing the capacityof their schools and themselves beforeimplementing instructional reform? Aresome approaches to large-scale instruc-tional improvement better suited tohigh-capacity districts than to low-capacity districts, and vice versa?

■ Middle and high school success. Whatneeds to happen for district reforms toprogress not only at the elementaryschool level but also at the middle andhigh school levels?

■ Addressing the needs of the whole child. Asdistricts engage in systemwide instruc-tional improvement, how can they effec-tively integrate approaches to helpingchildren grow socially, emotionally, andethically as well as academically?


District SelectionThe investigation discussed in this reportinvolved a comparative case study of five schooldistricts. Districts were selected according to pri-mary and secondary criteria. In applying our pri-mary criteria, we sought districts that exhibited:

■ Success in increasing student achievementin math or reading over three or more years

■ Improvement in student achievement acrossgrade levels and ethnicities

■ A poverty rate of at least 25 percent, asdefined by students eligible for free orreduced lunch

■ A reputation for effective professional devel-opment practices, based on recommenda-tions from education leaders

In addition to the primary criteria, districts wereselected according to a mix of demographic fac-tors, including size, geographic distribution, statepolicy frameworks, union affiliation, and theextent to which they had already been studied.

We solicited district recommendations fromLearning First Alliance member organizations,education researchers, and nonprofit leaders.We received over 50 recommendations of dis-tricts and conducted a careful review of districtachievement data. Standardized test data werethe primary sources of achievement data used.On the basis of data, 14 districts emerged aspotential study sites. Using the secondary crite-ria, we chose five of those districts for study.

Data Gathering and AnalysisWe gathered data during two visits to each dis-trict (except Providence) over the course of theacademic year 2001–2002. The first site visits(three days), which were conducted fromNovember 2001 through January 2002 by ateam of four researchers from the University ofToronto and the Learning First Alliance, servedas the primary data-gathering exercise. In thespring of 2002, a larger team of senior staff andboard members from Learning First Alliancemember organizations conducted a secondround of site visits (two days). Providence wasnot visited in the second round because ofscheduling difficulties. Learning First Allianceresearch staff attended all nine site visits. Twomembers of the University of Toronto facultyparticipated in each first-round site visit. Onemember participated in the second round ofvisits. Six kinds of data were gathered prior toand during these visits.

1. Interview data. During the first site visit,the research team conducted approximately35 individual semistructured interviews ineach district (176 total) with educationstakeholders, including teachers, unionleaders, central office staff, board members,teacher leaders, principals, university part-ners, community leaders, and parents. TheLearning First Alliance senior staff andresearch team members conducted 31 fol-low-up interviews with a similar set ofstakeholders during the second site visit.

In addition, the research team conductedinterviews with state-level education stake-holders, including the commissioners of

Appendix I Methodology

education from four states, state legislators,state board of education members, statedepartment of education staff, and execu-tive directors of state alliance organizations.

2. Focus groups. Focus groups were conductedduring both site visits. A total of 63 focusgroups were convened (10–15 per district).Focus groups were used to gain feedbackfrom stakeholder groups with larger num-bers of staff, such as teachers, principals,and community members. Focus groupswith teacher mentors, instructional coach-es/teacher leaders, principals from turn-around schools, and the like were used togain deeper information about specificinstructional reform efforts or professionaldevelopment innovations.

3. Observations. Nonparticipant observationswere conducted during the first site visit.The research team attended professionaldevelopment workshops, principal train-ings, grade-level meetings, board meetings,and other meetings related to professionaldevelopment and the implementation ofinnovations.

4. School site visits. During both phases of thestudy, the research teams visited threeschools per district, ensuring a mix of ele-mentary, middle, and secondary schoolswhere feasible. The school site visit designincluded an interview with the principal,two interviews with teachers, and a schoolwalkthrough.

5. Documentation. From each district we gath-ered a series of documents to bolster ourinterview and observational data. We gath-ered both district- and school-based docu-ments, such as strategic plans, union con-tracts, budgets, curricular frameworks, pro-fessional development guidelines and activi-ties, achievement data, accountabilityreports, and professional development tools.

6. Field notes. Each member of the site visitteam from the first and second site visitcompleted a formal observational memothat outlined information about stakeholderinteractions, stakeholder roles in instruction-al reform, and other general observations.

All interviews and focus groups were taped.Given the volume of interviews and focusgroups and limited resources, researchers used acombination of verbatim transcription andsummarization to create a written record of theinterviews. All focus groups were transcribed.On the basis of the study questions and anoriginal conceptual framework, a codingscheme was designed for the data and a themat-ic outline was developed to guide case studywriting. Interview data were coded and enteredinto nud*st. On the basis of a combination ofinterview transcript review, nud*st thematicreports, field notes, and documentation, casestudies were written for each district. Multiplemembers of the research team participated inthe analysis and drafting of each case study.Once written, the case studies became dataused to inform the analysis for the cross-caseanalysis report. Drafts of the case studies wereprovided to the districts to cross-check theaccuracy of the factual data.


Methodology

Table A.1. Aldine Independent School District . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Table A.2. Aldine Independent School District . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Table A.3. Chula Vista Elementary School District . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .64

Table A.4. Chula Vista Elementary School District . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65

Table A.5. Kent County Public Schools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66

Table A.6. Kent County Public Schools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66

Table A.7. Minneapolis Public Schools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68



Table A.10. Providence Public Schools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69

Table A.11. Providence Public Schools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70


Appendix II District Achievement Data

Table A.1 Aldine Independent School DistrictPercentage of Students Meeting Minimum Expectations on the

Texas Assessment of Academic Skills (TAAS)—Reading—1994–2002

1994 1995 1996 1997

READING, Grade 3Asian NA 95 90 95African American 67 77 66 79Hispanic 78 89 88 91White 85 87 93 91








District Achievement Data


Total 1998 1999 2000 2001 2002 % Change

97 89 91 90 93 –283 83 83 81 86 1986 89 87 86 89 1192 92 92 91 94 9

98 93 91 96 98 888 86 86 89 92 2495 89 89 91 95 1796 92 93 95 96 10

98 92 92 95 91 088 77 80 88 86 2191 81 83 90 91 1793 92 95 93 96 12

92 93 95 93 99 1389 85 87 84 86 2785 84 88 87 89 2195 91 94 93 95 14

92 96 89 95 99 983 81 79 86 91 2785 84 81 90 93 2594 92 92 95 98 12

90 95 96 98 99 1687 88 90 92 94 2684 90 91 93 96 2795 95 96 93 96 8

80 87 94 90 93 2185 85 89 89 95 3080 83 90 87 92 2895 95 95 96 96 8

Notes: NA = Not applicable. In order to meet minimum expectations, a student must answer at least 70 percent of the testquestions correctly.

Source: These data were provided by the Aldine Independent School District.

Table A.2 Aldine Independent School DistrictPercentage of Students Meeting Minimum Expectations on the

Texas Assessment of Academic Skills (TAAS)—Mathematics—1994–2002

1994 1995 1996 1997

MATHEMATICS, Grade 3Asian NA 95 96 99African American 56 73 68 84Hispanic 71 84 86 94White 77 84 92 94










Total 1998 1999 2000 2001 2002 % Change

98 96 97 95 97 279 77 73 80 86 3086 87 85 89 93 2294 89 87 90 95 18

98 99 96 99 100 483 86 87 89 95 4694 92 91 95 97 2696 95 93 96 98 21

99 99 100 99 97 1083 82 85 90 91 3293 90 91 94 96 2894 92 94 96 98 17

98 98 98 100 99 1687 84 85 89 93 5391 89 91 94 96 4396 92 92 96 96 26

93 98 96 99 99 1975 81 84 88 91 5183 89 92 94 95 4893 92 93 94 98 29

96 99 100 100 100 2284 82 88 90 90 5185 89 94 95 96 4995 95 96 95 95 23

87 92 97 98 100 2569 70 82 87 90 5776 82 90 91 96 5192 92 94 99 95 28

Notes: NA = Not applicable. In order to meet minimum expectations, a student must answer at least 70 percent of the testquestions correctly.

Source: These data were provided by the Aldine Independent School District.

Table A.3 Chula Vista Elementary School DistrictPercentage of Students Scoring at or above the 50th Percentile on the Stanford 9—Reading—1999–2002

Total 1999 2000 2001 2002 % Change

READING, Grade 2Asian 67 66 70 71 4Filipino 71 64 75 81 10White 59 62 67 67 8Hispanic 28 33 41 39 11African American 39 39 43 51 12




READING, Grade 6Asian 61 64 63 59 –2Filipino 67 66 73 74 7White 65 64 67 69 4Hispanic 34 35 39 35 1African American 45 36 45 52 7

Notes: Chula Vista was originally selected based on disaggregated data from 1998–2000. Due to changes in the state’s methods fordisaggregating data, data prior to 1999 are no longer used.

Source: These data were provided by the Chula Vista Elementary School District.



Table A.4 Chula Vista Elementary School DistrictPercentage of Students Scoring at or above the 50th Percentile on the Stanford 9—Mathematics—1999–2002

Total 1999 2000 2001 2002 % Change

MATHEMATICS, Grade 2Asian 77 85 90 89 12Filipino 73 71 84 85 12White 64 74 75 77 13Hispanic 37 51 55 55 18African American 45 52 51 59 14





Notes: Chula Vista was originally selected based on disaggregated data from 1998–2000. Due to changes in the state’s methods fordisaggregating data, data prior to 1999 are no longer used.

Source: These data were provided by the Chula Vista Elementary School District.


Table A.5 Kent County Public SchoolsPercentage of Students Scoring Satisfactory or Higher on the

Maryland School Performance Assessment Program (MSPAP)—Reading—1994–2002

1994 1995 1996 1997

READING, Grade 3African American 22 21 26 25White/non-Hispanic 52 46 55 54



Table A.6 Kent County Public SchoolsPercentage of Students Scoring Satisfactory or Higher on the

Maryland School Performance Assessment Program (MSPAP)—Mathematics—1994–2002

1994 1995 1996 1997

MATHEMATICS, Grade 3African American 36 32 36 40White/non-Hispanic 58 66 65 80






1998 1999 2000 2001 2002 Total % Change

49 62 60 55 39 1769 75 72 57 54 2

27 30 29 28 41 2751 57 58 63 51 11

10 17 25 29 28 1338 38 43 50 39 4

Notes: Due to reasons of privacy (n < 5), data are not represented for Hispanic and Asian students. Kent County’s scoresin 2001 and 2002 show a decline that is reflective of a broader statewide decline in MSPAP scores.

Source: These data were provided by the Kent County Public Schools.

1998 1999 2000 2001 2002 Total % Change

76 71 70 46 44 881 95 90 77 67 9

21 25 24 16 19 458 69 59 51 49 4

29 30 42 29 27 669 75 75 73 63 1

Notes: Due to reasons of privacy (n < 5), data are not represented for Hispanic and Asian students. Kent County’s scoresin 2001 and 2002 show a decline that is reflective of a broader statewide decline in MSPAP scores.

Source: These data were provided by the Kent County Public Schools.

Table A.8 Minneapolis Public SchoolsPercentage of Students Scoring at or above Level IIb on the

Minnesota Comprehensive Assessment—Mathematics—1998–2002

Total1998 1999 2000 2001 2002 % Change

MATHEMATICS, Grade 3American Indian 22 25 39 35 38 16Asian/Pacific Islander 19 28 34 41 44 25Hispanic 17 30 30 25 31 14Black 13 17 24 28 28 15White 58 67 72 72 73 15


Notes: Most students in Level IIb are working successfully on grade-level material and are on track to achieve satisfactory work in thestate’s content standards. Minneapolis was selected based on achievement data from grades K–8, the grades on which district-levelreform had focused thus far. The district began high school reform efforts in 2001–2002.

Source: These data were provided by the Minneapolis Public Schools.

Table A.7 Minneapolis Public SchoolsPercentage of Students Scoring at or above Level IIb on the Minnesota Comprehensive Assessment—Reading—1998–2002

Total1998 1999 2000 2001 2002 % Change

READING, Grade 3American Indian 21 23 29 34 41 20Asian/Pacific Islander 13 17 15 30 29 16Hispanic 22 26 24 24 24 2Black 16 18 21 30 28 12White 60 63 68 73 74 14


Notes: Most students in Level IIb are working successfully on grade-level material and are on track to achieve satisfactory work in thestate’s content standards. Minneapolis was selected based on achievement data from grades K–8, the grades on which district-levelreform had focused thus far. The district began high school reform efforts in 2001–2002.





Table A.9 Minneapolis Public SchoolsPercentage of Students Passing the Basic Skills Test—Reading and Mathematics—1998–2002

Total1998 1999 2000 2001 2002 % Change



Notes: Minneapolis was selected based on achievement data from grades K–8, the grades on which district-level reform had focusedthus far. The district began high school reform efforts in 2001–2002.


Table A.10 Providence Public SchoolsPercentage of Students Who Met or Exceeded the Standard on the

New Standards English Language Arts Reference Exam—1998–2002

Total 1998 1999 2000 2001 2002 % Change

Grade 4All students 35 53 57 44 47 12

Grade 8All students 24 22 25 27 23 –1

Notes: As Providence had focused its reform efforts on the elementary grades, the district was selected based on data that revealedimprovement in elementary school students’ reading scores from 1998 to 2000 (e.g., according to the originally provided data, the percentage of fourth-grade black students who met or exceeded the standard on the English Language Arts Reference Examincreased from 28% to 42%, and Hispanic student achievement increased from 21% to 37%). These disaggregated data were provid-ed by the Rhode Island State Department of Education. However, due to changes in the state’s methods for disaggregating data, theoriginal data we received have been withdrawn from public access. Therefore, we print here the aggregate data provided by theProvidence Public Schools. The district began to disaggregate its own data again in 2002.

Source: These data were provided by the Providence Public Schools.



Table A.11 Providence Public SchoolsPercentage of Students Who Met or Exceeded the Standard on the New Standards Mathematics Reference Exam—1998–2002

Total 1998 1999 2000 2001 2002 % Change

Grade 4All students 24 26 34 58 35 11

Grade 8All students 23 27 26 13 18 –5

Notes: As Providence had focused its reform efforts on the elementary grades, the district was selected based on data that revealedimprovement in elementary school students’ reading scores from 1998 to 2000 (e.g., according to the originally provided data, the percentage of fourth-grade black students who met or exceeded the standard on the English Language Arts Reference Examincreased from 28% to 42%, and Hispanic student achievement increased from 21% to 37%). These disaggregated data were provid-ed by the Rhode Island State Department of Education. However, due to changes in the state’s methods for disaggregating data, theoriginal data we received have been withdrawn from public access. Therefore, we print here the aggregate data provided by theProvidence Public Schools. The district began to disaggregate its own data again in 2002.

Source: These data were provided by the Providence Public Schools.


American Federation of Teachers. (2001).Beginning teacher induction: The essentialbridge. Educational Issues Policy Brief.Washington, DC: American Federation ofTeachers.

Bodilly, S. J. (1998). Lessons from new Americanschools’ scale-up phase: Prospects for bringingdesigns to multiple schools. Santa Monica,CA: Rand.

Bryk, A., Sebring, P., Kerbow, D., Rollow, S.,& Easton, J. (September 1994). The stateof Chicago school reform. Phi DeltaKappan, 66(1), 74–78.

Cohen, D., & Hill, H. C. (1998). State policyand classroom performance: Mathematicsreform in California. Philadelphia, PA:Consortium for Policy Research inEducation.

Corcoran, T. (1995). Helping teachers teachwell: Transforming professional development.Philadelphia, PA: Consortium for PolicyResearch in Education.

Corcoran, T., & Fuhrman, S. (2002). Going toscale: Building effective infrastructure.Philadelphia, PA: Consortium for PolicyResearch in Education.

Council of Chief State School Officers and TheCharles A. Dana Center at the University ofTexas at Austin. (2002). Expecting success:An analysis of education policies in Texas.Washington, DC: Council of Chief StateSchool Officers.

Darling-Hammond, L., & Ball, D. L. (June1997). Teaching for high standards: Whatpolicymakers need to know and be able to do.Philadelphia, PA: Consortium for PolicyResearch in Education.

Education Commission of the States. (2001). Acloser look: State policy trends in three keyareas of the Bush education plan—Testing,accountability and school choice. Denver,CO: Education Commission of the States.

Elmore, R. F. (2000). Building a new structurefor school leadership. Washington, DC: TheAlbert Shanker Institute.

Elmore, R. F. (1996). Getting to scale withgood educational practice. HarvardEducational Review, 66(1), 1–26.

Elmore, R. F., & Burney, D. (2000). Leadershipand learning: Principal recruitment, induc-tion and instructional leadership in commu-nity school district #2, New York City.Pittsburgh, PA: High Performance LearningCommunities Project at the LearningResearch and Development Center,University of Pittsburgh.

Elmore, R. F., & Burney, D. (1997). Schoolvariation and systemic instructional improve-ment in community school district #2, NewYork City. Pittsburgh, PA: High Perfor-mance Learning Communities Project atthe Learning Research and DevelopmentCenter, University of Pittsburgh.

Research That Informed the Study



Foley, E. (2001). Contradictions and control insystemic reform: The ascendancy of the centraloffice in Philadelphia schools. Philadelphia,PA: Consortium for Policy Research inEducation.

Fuhrman, S. H. (1993). The politics of coher-ence. Designing coherent education policy:Improving the system. San Francisco: JosseyBass.

Fullan, M., & Hargreaves, A. (1996). What’sworth fighting for in your school. New York:Teachers College Press.

Fullan, M., & Stieglbauer, S. (1991). The newmeaning of educational change. 2nd ed. NewYork: Teachers College Press.

Greenwald, R., Hedges, L. V., & Laine, R. (Fall1996). The effect of school resources onstudent achievement. Review of EducationalResearch, 66(3), 361–396.

Grossman, P., Thompson, C., & Valencia, S.(2001). District policy and beginning teach-ers: Where the twain shall meet. Seattle, WA:Center for the Study of Teaching and Policyat the University of Washington and Centeron English Learning & Achievement at theUniversity of Albany.

Hargreaves, A. (1994). Changing teachers,changing times: Teachers’ work and culture inthe postmodern age. Toronto: OISE Press.

Harwell, M., D’Amico, L., Stein, M. K., &Gatti, G. (February 15, 2000). Professionaldevelopment and the achievement gap in com-munity school district #2. Pittsburgh, PA:High Performance Learning CommunitiesProject at the Learning Research and Deve-lopment Center, University of Pittsburgh.

Haslam, M. B., & Seremet, C. P. (2001).Strategies for improving professional develop-ment: A guide for school districts. Arlington,VA: New American Schools.

Herman, R., & Stringfield, S. (1997). Tenpromising programs for educating all children:Evidence of impact. Washington, DC:Education Research Service.

Hess, F. M. (1999). Spinning wheels: The politicsof urban school reform. Washington, DC:The Brookings Institution.

Hightower, A. M. (2002). San Diego’s big boom:District bureaucracy supports culture of learn-ing. Seattle, WA: Center for Study ofTeaching and Policy at University ofWashington.

Hirsch, E., Koppich, J. E., & Knapp, M. S.(2001). Revisiting what states are doing toimprove the quality of teaching: A workingpaper prepared in collaboration with theNational Conference of State Legislatures.Seattle, WA: Center for the Study of Teachingand Policy at the University of Washington.

Kauffman, D., Johnson, S. M., Kardos, S. M.,Liu, E., & Peske, H. G. (2002). “Lost atsea”: New teachers’ experiences with cur-riculum and assessment. Teachers CollegeRecord, 104(2), 273–300.

King, M. B., & Newmann, F. M. (April 2000).Will teacher learning advance school goals?Phi Delta Kappan, 81(8), 576–580.

Koschoreck, J. W. (n.d.). Accountability andeducational equity in the transformation ofan urban district. Cincinnati, OH:University of Cincinnati.

Lewis, S., Jepson, J., & Casserly, M. (1999).Closing the achievement gaps in urbanschools: A survey of academic progress andpromising practices in the great city schools.Washington, DC: The Council of the GreatCity Schools.


Lieberman, A. (1995). Restructuring schools:The dynamics of changing practices, struc-ture, and culture. In: A. Lieberman, ed. Thework of restructuring schools: Building fromthe ground up (pp. 1–17). New York:Teachers College Press.

Lippitt, G., & Lippitt R. (1978). The consultingprocess in action. San Francisco: Jossey Bass.

Maloy, K. (1998). Building a learning commu-nity: The story of New York City CommunitySchool District #2. Pittsburgh, PA: HighPerformance Learning Communities Projectat the Learning Research and DevelopmentCenter, University of Pittsburgh.

Manning, J. B. (2001). Emerging models of gov-erning school districts. Review 10(2).Philadelphia, PA: National Center onEducation in the Inner Cities.

Marsh, J. A. (2000). Connecting districts to thepolicy dialogue: A review of literature on therelationship of districts with states, schools,and communities. Seattle, WA: Center forthe Study of Teaching and Policy at theUniversity of Washington.

Massell, D. (2000). The district role in buildingcapacity: Four strategies. Philadelphia, PA:Consortium for Policy Research inEducation.

Murphy, J., & Hallinger, P. (1988).Characteristics of instructionally effectiveschool districts. Journal of EducationResearch, 8(31), 175–181.

National Clearinghouse for ComprehensiveSchool Reform. (June 2001). District sup-port for teaching and learning. Washington,DC: National Clearinghouse forComprehensive School Reform.

Newmann, F. M., & Wehlage, G. G. (1995).Successful school restructuring: A report to thepublic and educators by the Center onOrganization and Restructuring of Schools.Madison, WI: Board of Regents of theUniversity of Wisconsin System.

Prince, C. D. (January 2002). The challenge ofattracting good teachers and principals instruggling schools. Arlington, VA: AmericanAssociation of School Administrators.

Ragland, M. A., Asera, R., & Johnson, J. F.(n.d.) Urgency, responsibility, efficacy:Primary finding of a study of high-performingTexas school districts. Austin: Charles A.Dana Center, University of Texas at Austin.

Richardson, J. (October 2000). Central officeguidance strengthens the whole district.Oxford, OH: National Staff DevelopmentCouncil.

Skrla, L., Scheurich, J. J., & Johnson, J. F.(September 2000). Equity-driven achieve-ment-focused school districts: A report on sys-temic school success in four Texas school dis-tricts serving diverse student populations.Austin: Charles A. Dana Center, Universityof Texas at Austin.

Spillane, J. P. (1996). School districts matter:Local educational authorities and stateinstructional policy. Education Policy, 10(1),63–87.

Spillane, J. P., Reiser, B., & Reimer, T. (April2002). Policy implementation and cognition:Reframing and refocusing implementation.Evanston, IL: Northwestern University.

Stein, M. K. (1998). High PerformanceLearning Communities district 2 report onyear one implementation of school learningcommunities. Pittsburgh, PA: HighPerformance Learning Communities Projectat the Learning Research and DevelopmentCenter, University of Pittsburgh.


Stringfield, S., & Datnow, A. (May 1998).Scaling up school restructuring designs inurban schools. Education and Urban Society,30(3), 269–276.

Swan, W. W. (1998). Local school system imple-mentation of state policy actions for educa-tional reform. Athens, GA: Department ofEducational Leadership, The University ofGeorgia.

Ward, J., St. John, E. P., & Laine, S. W. M.(1999). State programs for funding teacherdevelopment. Oak Brook, IL: North CentralRegional Educational Laboratory.

Watson, N., Fullan, M., & Kilcher, A. (2000).The role of the district: Professional learningand district reform. The Ontario Institutefor Studies in Education of the Universityof Toronto.


About the Learning First AllianceThe Learning First Alliance, founded in 1997, is a permanent partnership of 12 leading educationalassociations that have come together to improve student learning in America’s public elementaryand secondary schools. Members of the Alliance represent more than 10 million Americans engagedin providing, governing, and improving public education.

Alliance members are committed to developing and delivering common messages to all parts of theeducation system, sharing and disseminating success stories, encouraging collaboration at everylevel, and working toward the continual and long-term improvement of public education based onsolid research.

The three major goals of the Learning First Alliance are to:

1. Ensure that high academic expectations are held for all students

2. Ensure a safe and supportive place of learning for all students

3. Engage parents and other community members in helping students achieve high academicexpectations

1001 Connecticut Avenue, NWSuite 335

Washington, DC 20036202-296-5220

www.learningfirst.org

The Learning First Alliance is composed of the following organizations:

American Association of Colleges for Teacher EducationAmerican Association of School Administrators

American Federation of TeachersAssociation for Supervision and Curriculum Development

Council of Chief State School OfficersEducation Commission of the States

National Association of Elementary School PrincipalsNational Association of Secondary School PrincipalsNational Association of State Boards of Education

National Education AssociationNational PTA

National School Boards Association

beyond islands of excellence - learning first · the cost for beyond islands of excellence: what...

Documents