Dollars & Sense II - Architects of Achievement

&dollars

SENSE IILessons from Good, Cost-Effective Small Schools

By Barbara Kent Lawrence, Ed.D

Paul Abramson

Victoria Bergsagel

Steven Bingler

Barbara Diamond, J.D.

Thomas J. Greene,

Bobbie Hill

Craig Howley, Ed.D

David Stephen

Elliot Washor, Ed.D

Copyright © 2005 by KnowledgeWorks Foundation. All rights reserved.

Printed in the United States of America.

Except as permitted under the United States Copyright Act of 1976, no part of this pub-lication may be reproduced or distributed in any form or by any means, or stored in adatabase or retrieval system without prior written permission of the publisher.

To order a copy of this publication free of charge, please contact KnowledgeWorksFoundation, Concordia, or Architects of Achievement.

KnowledgeWorks FoundationOne West Fourth Street, Suite 200Cincinnati, Ohio 45202-3611Telephone: (513) 929-4777Fax: (513) 929-1122Website: www.kwfdn.org

Architects of Achievement7900 E Green Lake Dr. N, Suite 212Seattle, Washington 98103Telephone: (206) 812-3170¶Fax: (206) 812-3190Website: www.archachieve.org

Concordia, LLC201 Saint Charles Avenue, Suite 4314New Orleans, Louisiana 70170 Telephone: (504) 569-1818 Fax: (504) 569-1820Website: www.concordia.com

Acknowledgement: With thanks to Dollars & Sense, a magazine that covers economicissues from a progressive point of view, for permission to use the title Dollars & Sense for this publication. For more information about the magazine, please visithttp://www.dollarsandsense.org.

Funding for this publication was provided by Bill & Melinda Gates Foundation andKnowledgeWorks Foundation.

KnowledgeWorks Foundation is committed to furthering universal access to high-qualityeducational opportunities for individuals to achieve success and for the betterment of society.The Foundation provides funding and leadership for education initiatives throughout Ohio. Whatmakes the Foundation unique is its multi-faceted approach to education philanthropy: we fund,facilitate, and do. The Foundation believes the approach of investing, supporting, and activelyparticipating maximizes the results of improving education. The Foundation is committed tosharing knowledge gained and lessons learned with others in Ohio and across the U.S.

Concordia is a national leader in the planning and design of innovative environments forlearning. Over the past 22 years, research alliances have included the MIT Media Lab, HarvardUniversity’s Project Zero, the University of New Mexico, the National Aeronautics and SpaceAdministration, the Thornburg Institute, the Appalachian Education Lab, and the West Ed Lab.Domiciled in New Orleans, Louisiana, the firm’s work emphasizes collaboration and integrationas tools for addressing the needs of diverse constituencies.

Architects of Achievement is focused on building bridges between educational practice andarchitecture – especially as it relates to meeting student needs. They provide high-levelconsulting expertise related to new construction, renovation or simple reconfiguration to supportpowerful small schools and educational complexes. Bringing research regarding how people learn,they help school districts, charter organizations and foundations think creatively about effectivedesign solutions that foster synergy between program and building. Working with groupsthroughout the nation committed to developing better learning environments, they believe thatALL students can achieve if provided the opportunity, tools and support.

Sponsors

D o l l a r s & S e n s e I I : L e s s o n s f r o m G o o d , C o s t - E f f e c t i v e S m a l l S c h o o l s

iv

Index of Select Small Schools viii

Overview x

About This Report x

Introduction 1

Operating Costs of Good Small Schools 2

Selecting Good Small Schools 4

Small Schools, Big Ideas 6

Introduction 6

Staffing 6

Juggling Act 6

Using Rules Strategically 6

Double Duty 7

Sharing Administrative Duties 7

Sharing Staff 8

Importing Teachers 8

Accepting No Substitutes 8

Using Mentors and Tutors 8

Adjunct Faculty 8

Using Volunteers 8

Multiple Credentials 9

Hiring Young, First-Time Teachers 9

Hiring Retirees and Career Changers 9

Project- and Internship-Based Curriculum 9

Other Incentives 10

Educational Program 10

Focused Mission 10

Attracting Students 10

Course Cycles 10

Ordering Supplies 11

Services 11

Transportation 11

Food Service 12

Athletics 12

Health Care 13

Table of Contents


v

Sources of Funding 13

Renting Space 13

Business Located in Schools 13

Trading Spaces 14

Building Shared Space 14

Selling Memberships 14

Renting Computers 14

Grants 14

Donations 14

Financial Benefits of Good Relationships with the Community 15

Sharing Space 15

Recruiting Students 15

Federal Funding 16

Title 1 16

Impact Aid 16

Work Study 16

Emergency Repairs and Renovations 16

Qualified Zone Academy Bonds 16

Carl D. Perkins Vocational and Technical Education Grants 16

New Market Tax Credit 16

State Programs 16

Facilities Design and Construction 17

Site Selection 17

Smaller School Sites 17

Donated Sites 18

Design and Construction 18

Free Design 19

Construction Bidding 19

Renovation 19

Additions 20

Leasing 20

Adapting Existing Space Within a School 20

Using Students and Volunteers 20

Flexibility/Adaptability/Multiple Uses 21

Sharing Academic Work Space with Partners 21

Reducing Square Footage per Student 22

Sharing Space for Athletics and Extracurricular Activities 22

Partnerships in Construction 22

Green Construction 23


vi

Furnishings, Fixtures, and Equipment (FF&E) 23

Furnishings and Furniture Donations 23

Build it Yourself 24

Sharing Internally 24

Flexibility/Mobility 24

IT Built or Rebuilt by Students 24

IT Serviced by Students 24

Maintenance 24

Investing in Preventive Maintenance 24

Reducing Trash and Graffiti 25

Maintaining the Building 25

Maintaining the Grounds 25

Using Computers Effectively for Maintenance and Operations 25

Facilities Operations 25

No Janitorial Staff 25

Reduced Security Staff and Equipment 26

Utilities 26

Natural Lighting 26

Dual Fuel 27

Automatic Light Sensors 27

Solar Power 27

Keeping Cool 27

Automatic Water Shut-off 27

Wind Power 27

Computerized Control Systems 27

Retrofitting 27

School Profiles 28

Suggestions for Further Work 60

Conclusion 60

The Dollars & Sense Team 62

Resources 65


vii

Illustrations

Boxes Definition of Terms 5

Benefits of Networks 7

PowerSchool 8

Saving on Transportation 11

Creativity Pays 15

Site Sizes – New Guidelines 18

Tips for Designing Spaces for Autonomous Small Schools 22

Savings from Solar Power 23

ChartsAnalysis of School Budgets 3

Appendix 1: 68

Information by SchoolContact Information

School Type and Location, Grades Served, Number of Students, Demographics

Budget

Facilities

Appendix 2: 104

School Facility Construction Analysis Summary

Appendix 3: 110

Test Scores for Schools in This Report

Appendix 4: 113

Strategies Grid

Appendix 5: 115

Criteria for Selecting and Evaluating Schools

Appendix 6: 122

Selected References


viii

Index of Select Small Schools

Avalon Charter High School St. Paul, MNUrban charter high school 125 students (2004-2005)Grades 9-12 • Project-based programFacility: renovation

This small, urban, teacher-owned charter high school offers a rich,project-based curriculum. It is located in the Warehouse District in acost-effectively renovated building that it shares with two othercharter schools.

See page: 28

C.C. Blaney Elementary School Hollywood, SCRural elementary school190 studentsGrades PK-5 • Arts infusion programFacility: additions and renovations

This older rural school was recently modernized to reflect the district’snew commitment to provide a first-class facility and program to alargely African-American population. More than 98% of students areeligible for free or reduced-cost lunches, and Title 1 money supportsa curriculum infused with the arts.

See page: 30

Bronx: Small Schools at Morris High Bronx, NYUrban high schoolGrades 9-12 • Four schools within a school

Bronx International: 300 studentsBronx Leadership Academy: 340 students School of Excellence: 330 studentsViolin and Dance: 170 students

Facility: renovation

The four small schools carved out of a failing large urban high schoolsuggest ways schools within a school can raise attendance andgraduation rates. These schools, with separate identities andprograms, also have raised the aspirations of students and teachers.

See page: 32

Camino Nuevo Charter Academy Los Angeles, CAUrban charter elementary school257 students Burlington Street CampusUrban charter middle school294 studentsGrades: PK-8Facilities: renovation

Camino Nuevo elementary and middle schools occupy two renovatedbuildings in the most densely populated section of Los Angeles. Theschools have turned a dangerous area into an oasis for its Hispanicpopulation and have created models for school-communitypartnerships.

See page: 34

Flagstaff Arts and Leadership Academy (FALA) Flagstaff, AZUrban charter high school160 Students (2004-2005)Grades: 9-12 • Arts-based programFacility: modular units

This charter high school attracts students from a wide geographicarea to its rigorous arts-based academic program. Located on thegrounds of the Museum of Northern Arizona, the school is ideallysituated to share activities with the museum and its neighbor, ThePeaks senior living center.

See page: 36

High Tech High San Diego, CAUrban charter high school368 students (2002-2003)454 students (2004-2005)Grades 9-12 • Internship and project-based curriculumFacility: renovation

Putting a high school in a former storage building on adecommissioned naval base might not seem sensible, but the artfulrenovation of this facility has created a building ideally suited to thisproject-based school.

See page: 38

Interdistrict Downtown School (IDDS) Minneapolis, MNUrban multi-district school500 students (2004-2005)Grades K-12Facility: new construction

This new school in downtown Minneapolis draws its students from 10 districts in an effort to create a diverse student body. The locationoffers many benefits, including a wide variety of private and publicresources. The facility offers lessons in a cost-effective andenvironmentally sound design.

See page: 40

Laurel-Concord Public School Laurel, NERural school388 students (2004-2005)Grades K-12Facility: renovations and additions

Although set in the cornfields of Nebraska, Laurel-Concord PublicSchool is a hotbed of entrepreneurial activity. A resourceful schoolboard and superintendent have found a number of innovative waysto manage the school and maximize its resources.

See page: 42

Media and Technology Charter Boston, MAUrban charter high school185 students (2004-2005)High School • Grades 9-12Facility: renovation

Media and Technology Charter (MATCH) has been remarkablyeffective in helping its students, most of whom are Hispanic orAfrican-American, succeed academically. This small school, located ina former car dealership near public transportation, provides firm andcaring support, including 10 hours of one-on-one tutoring eachweek.

See page: 44


ix

Metropolitan Regional and Technical Center Providence, RIUrban charter high school438 students (2003-2004)705 students (2006-2007) (projected capacity)Grades 9-12 • Internship-based programFacility: new construction

The Big Picture Company schools in Providence, known collectively asThe Met, include four small schools on the main campus at PublicStreet and two others located nearby. The Met’s advisory andinternship programs are based on close relationships betweenstudents and adult members of the community.

See page: 46

Oak Valley School Lincoln, NERural elementary - middle school30 students (2004-2005)Grades K-8Facility: new construction

This tiny rural elementary school has found creative ways to be costeffective while offering a rich, academically sound program thatemphasizes the fun of learning. Students say they love going toschool and feel like members of an extended family.

See page: 48

R.D. Schroder Middle School Hollywood, SCRural middle school380 students (2004-2005)Grades 6-8Facility: addition and renovation

R.D. Schroder Middle School and C.C. Blaney Elementary Schoolshare a 31-acre site. The schools donated the area between theirrespective buildings to the parks department, which created anathletic facility that serves the schools and the community.

See page: 30

The STAR School Flagstaff, AZRural elementary – middle school65 students (2004-2005)Grades K-8Facilities: adapted modulars

STAR (Service to All Relations) operates a campus of modularbuildings that are fully solar-powered and adapted for the aestheticof the local culture and architecture. The curriculum combines themodern skills of critical thinking with the Navajo traditions ofmaintaining harmony with the environment and all who share it.

See page: 50

Tacoma School of the Arts Tacoma, WAUrban high school348 students (2004-2005)Grades 10-12 • Arts-based curriculumFacilities: renovated, leased in multiple locations

Students at Tacoma School of the Arts (TSOTA) attend classes in ninedifferent buildings. The school leases or trades space in facilities thatoffer first-class accommodations for its arts-based curriculum, andprovides a wide range of resources for students built on partnershipswith local cultural organizations.

See page: 52

Todd Beamer High School Federal Way, WAUrban high school with three academies1,350 students (2003-2004)1,550 students (2004-2005)Grades 9-12 • Schools within a schoolFacility: new construction

This new facility – which can be configured as one school or as manyas eight separate schools – provides the flexibility and foresight for itsschools-within-a-school concept. Its planners knew the school’seducational program might change and they wanted the building tolead reform, not block it.

See page: 54

H.S. Truman High School Federal Way, WATwo urban high schools197 students (2004-2005) (total for two schools)Grades 9-12 • Internship-based programFacility: new construction

Because many students lived in cramped apartments, the architectsfound cost-effective ways to create the airy and light facility for thisalternative high school. The principal developed an internship- andadvisory-based program that encourages students to stay in school.

See page: 56

World of Opportunity Birmingham, ALUrban alternative school35 to 65 students per day • High school and adult Facility: leased, small addition

World of Opportunity (WOO) is a remarkable education and job-readiness program that serves students who, in many cases, wereforced out of the local public schools. WOO understands thechallenges its students face, and it meets their needs in effective andaffordable ways.

See page: 58


x

Overview

This report deepens the argument for small schools in three important ways. First, analysis of morethan three thousand construction projects shows that smaller schools are no more expensive to buildthan much larger schools. Second, analysis of the budgets of 25 good small schools throughout theUnited States demonstrates that on average they spend less per student on educational program,maintenance and operations than the per-pupil expenditure in their districts, yet they achieve resultsthat are equal to or better than schools in the same area. Third, these schools offer innovative andeffective educational programs, facilities, and strategies for cost effectiveness that can serve as modelsand inspiration to people interested in cost-effective good small schools.

A complete Appendix containing contact information, budgets, test scores, and references followsthe text. The website www.goodsmallschools.org supplements the written report, and contains manydocuments from the schools and links to additional resources. This website will be updatedperiodically, and will collect and disseminate strategies for good, cost-effective small schools beyondthose contained in this report.

About This Report

Dollars & Sense II was written for parents, students, school board members, school officials, teachers,policy-makers – and all other members of the community who have an interest in establishing andoperating good small schools. It presents findings and offers cost-saving suggestions from 25 sampleschools, information that can be used as a jumping-off point for discussions and planning for othersin similar circumstances.

This second report assumes that readers are familiar with the first Dollars & Sense report, whichsummarizes the benefits of small schools for students, teachers, parents, and communities. This reportis designed to be used as a working reference, allowing readers to move around in the text accordingto their interests.

Most cost-saving strategies appear at least twice in the report – in the section about that type ofstrategy, and in the profile of the school(s) that use the strategy. Most readers are likely to find ithelpful to move back and forth between the strategies and school profiles, as the profiles provide thecontext for the strategies, and show how they are being used in support of student learning. Toidentify which school employs specific strategies, consult the “Strategies Grid” in Appendix 4.

The appendices provide detailed information on each school, including contact information, keystatistics, maintenance and operations budgets, construction and renovation costs, test scores, and asummary of the research on school construction projects. Finally, the appendices include the schoolselection criteria and the site-visit guidelines used with each of the 25 schools featured in this report.

This report and additional information, including photographs, tables, charts, and sample documents,are available on the companion website at www.goodsmallschools.org.


1

I N T R O D U C T I O N

Good small schools make sense. Good small schools

create supportive communities where students succeed, and both students and teach-

ers thrive. Years of research and experience – not to mention common sense –

support this notion. Yet legislators and other decision-makers continue to close small

schools and build large ones because they think large schools are more cost effective

to build, maintain, and operate. The first report, Dollars & Sense: The Cost

Effectiveness of Small Schools, outlines the economic and social arguments in support

of smaller schools and demonstrates that the true costs of large schools are enormous

and the benefits dubious (D&SI, pp. 8 - 19).1

Dollars & Sense II: Lessons from Good, Cost-Effective Small Schools strengthens the

case for the cost effectiveness of small schools with new research and examples of

what’s working at 25 such schools2 across the nation. It answers the question – “Can

small schools actually be built and run at a cost per pupil that is comparable to that

of large schools?” – with a resounding “yes.”

First, the report offers an analysis of the budgets of 25 good small schools from across

the nation with various styles of education and diverse student populations. On

average, these schools spend less per student than do other schools in their districts.

These 25 schools prove that good small schools are possible to build, maintain, and

operate cost-effectively (D&SI, p. 20).

Lessons from Good, Cost-EffectiveSmall Schools

1 Throughout the text of Dollars & Sense II, the reference (D&SI) and a page number indicate that further infor-mation on a specific topic is available in Dollars & Sense: The Cost-Effectiveness of Small Schools (available in theResource Library at www.kwfdn.org and www.goodsmallschools.org).2 The 17 schools listed at the beginning of this document add up to 25 small schools, because there are four smallschools at Morris High School, two at Camino Nuevo, two at H.S. Truman, and four at The Met.


2

The central section of Dollars & Sense II offers specific,creative, cost-saving ideas that any school can use – fromfinding space in multiple buildings in a revitalizeddowntown to working math problems on desktops usingerasable markers to save paper.

These real-life ideas assembled from 25 small schoolsshould inspire school board members, parents, teachers,students, school administrators, policy-makers, community members, and taxpayersto come up with their own techniques forcreating and running good, cost-effectivesmall schools. The report concludes withprofiles of each of the small schools or smallschool complexes. The profiles offer briefinsights into why these schools are “good,”and set the context for the strategies theyemploy to run cost-effectively.

Appendix 2 provides an analysis of recentschool construction projects. Like similarwork highlighted in Dollars & Sense 1(D&SI, pp.18-20), this research shows thatsmall schools can be built as affordably aslarger schools. Extremely large schoolsappear to be cheaper only because theyprovide less space per student. As demonstrated in theoriginal Dollars & Sense report, the educational and socialcosts of such “mega-schools” are exorbitant (D&SI, pp.11-17). Appendices 1 and 3 offer detail on school budgetsand test scores. Appendix 4 summarizes the strategies usedby the schools on a single grid. The criteria for selectingstudy schools and the site visit guidelines are detailed inAppendix 5, and selected references are presented inAppendix 6.

While this report provides remarkably convincingevidence that small schools can operate cost-effectively,there are undeniably some inherent difficulties in doing so.Administrators, staffs, and communities supporting smallschools must be creative and inventive to achieve theseresults. This “out of the box” thinking is hard work, but itcan be done – these schools have done it.

Small schools also operate within systems and structuresthat are organized to support large schools, and sometimesmust seek solutions outside normal bureaucraticprocedures. The hard work of seeking creative solutions isworth it, though, and not just because of the benefits tothe students. The opportunities for participation andcommunity connection also make the effort worthwhile.

A key finding of this report is that smallness – somethingthat research and experience prove is good for students –in itself opens doors on ways to stretch dollars strategicallyin support of high quality education. Small schools can beflexible and nimble, just because they are small. It is easierfor them to use volunteers and form deep, lastingpartnerships with other agencies because they are workingat a smaller scale. In large schools, more bureaucratic and

systematic solutions are necessary justbecause of the size of the enterprise. Sowhile some of the cost-saving strategieshighlighted in the report could certainlybe employed in a large school, a mainpoint of the report is to demonstrate thatsmall schools have unique strategies forcost effectiveness.

Operating Costs of Good Small Schools

Another key finding of this report, inaddition to the conclusions on theaffordability of small school construction,concerns the actual per-student costs ofoperating small schools. Here, too, thefacts defy the common wisdom. Twenty ofthe 25 schools in this study spent less per

student than the average expenditure in their districts.3

The four schools comprising The Met spent more thanthe district’s per-pupil allocation, but this may be due inpart to the fact that the Met has chosen to grow slowly, sothat it does not yet serve its potential enrollment (eventhough it has a waiting list). At the time of the analysis,the school had 438 students, but by 2004-2005 it had sixhundred.4 Spending by the MATCH school exceeds theper-pupil expenditure for Boston, but the amount thedistrict allocates per student has been steadily decreasingbecause of budget constraints. MATCH considers itstutoring program essential and found outside sources tofund it.

“A key finding of this

report is that

smallness –

something that

research and

experience prove is

good for students – in

itself opens doors on

ways to stretch

dollars strategically in

support of high

quality education.”

3 Note that the four small schools at Morris High School in theBronx, the three academies at Todd Beamer High School, andtwo schools at H.S. Truman High School are each reported asone school for purposes of the budget summary. 4 The Met actually has six schools, and for the purposes of thebudget all six are included. However, the discussion and profileinclude only the four schools at Public Street because the teamdid not visit the other campus.


3

Avalon $10,176.36 $11,876.00 $1,699.64 14.31% St. Paul School District Dist office Ramsey 2002-03

CC. Blaney** $8,366.00 $11,533.90 $3,167.90 27.47% Charleston School Distr/State Report Dist office Charleston 2002-03

Bronx Schools** $11,138.00 $11,640.00 $502.00 4.31% New York City Public Schools Dist office Kings 2002-03

Camino Nuevo ES** $7,487.00 $8,408.00 $921.00 10.95% Los Angeles Unified School Dist. NCES* Los Angeles 2003-04

Camino Nuevo MS** $6,848.00 $8,408.00 $1,560.00 18.55% Los Angeles Unified School Dist. NCES Los Angeles 2003-04

FALA $6,102.00 $11,783.00 $5,681.00 48.21% Flagstaff Unified District Dist office Coconino 2003-04

High Tech High** $7,614.00 $9,092.00 $1,478.00 16.26% San Diego Unified School Dist. NCES San Diego 2004-05

IDDS*** $11,543.00 $12,616.34 $1,073.34 8.51% Minneapolis School District #1 Dist office Hennepin 2002-03

Laurel-Concord**** $7,927.00 $8,089.93 $162.93 2.01% Cedar County Public Schools NDE Cedar 2001-02

MATCH** $12,580.00 $11,943.00 ($637.00) -5.33% Boston Public Schools BPS Suffolk 2003-04

MET** $13,953.00 $11,426.00 ($2,527.00) -22.12% Providence School District RIDOE Rhode Island 2002-03

Oak Valley**** $4,965.00 $6,698.60 $1,733.60 25.88% Lancaster County Public Schools NDE Lancaster 2003-04

RD Schroder** $9,440.00 $11,533.90 $2,093.90 18.15% Charleston School Distr/State Report Dist office Charleston 2002-03

STAR $8,661.00 $11,783.00 $3,122.00 26.50% Flagstaff Unified District Dist office Navaho 2003-04

TSOTA** $5,922.00 $8,738.00 $2,816.00 32.23% Tacoma School District 10 Dist office Pierce 2003-04

Todd Beamer $6,747.00 $7,377.00 $630.00 8.54% Federal Way School District Dist office King 2003-04

Truman $6,893.00 $7,377.00 $484.00 6.56% Federal Way School District Dist office King 2003-04

WOO***** $2,557.26 $8,778.00 $6,220.74 70.87% Birmingham City School District NCES Jefferson 2003

Average $8,273.31 $9,950.09 $1,676.78 16.85%

*In those few cases in which the districts were either unable or unwilling to provide comparative data, the latest available National Center for Education Statistics (NCES) financial datawas used and this was for school year 2001-2002

**The financial data obtained from many of the schools did not include expenditures for food service as well as a number of other expenditures known to have been incurred. Sincethe districts did include all expenditures in their averages, in order to make the figures comparable the NCES data for each of those items was added to the financial data obtainedfrom each of these schools. The two Camino Neuvo schools did in fact know and capture these expenditures but as separate cost centers. Those allocations were added back to the twoschools’ original numbers as well.

***A weighted average of per pupil spending for the 10 schools which send students to IDDS was used as a basis of comparison and while only 8 of the 10 schools’ financials wereavailable, the 8 reporting schools comprise 96.1% of the total IDDS enrollment. The IDDS average annual cost per pupil also had to be adjusted to include other expenditures that wereincluded in the Minneapolis School District numbers.

****Because the two Nebraska schools are, uniquely, also their own districts, meaningful pupil expenditure comparisons could only be made to the countywide pupil expenditures ofthe county in which each school resides. The county data was obtained from the Nebraska Department of Education website.

*****WOO reports its financials on a calendar basis, i.e., January thru December. The average annual cost per pupil is based upon an average daily attendance of 45.

Average AnnualCost per pupil

Percent DistrictSchool School District Difference Difference District Data Source County Year

Because the way in which schools and districts determinebudgets varies enormously, it is extremely difficult tocompare costs. The team has made every effort to makeaccurate comparisons, which sometimes required addingor subtracting items that were or were not included inindividual school and district budgets. These adjustmentsare noted in the budget summary and individual schoolbudgets in the Appendices.

According to the estimates produced by this analysis, theschools in this study spent, on average, almost 17% less perstudent than the per-pupil expenditure for their districts.This finding shows that small schools in very differentlocations, serving different populations and grades, areoperating efficiently. These schools prove that communitiescan have small schools that offer an excellent education andalso are cost effective to maintain and operate.

Analysis of School Budgets


4

Selecting Good Small Schools for Analysis:What Makes a Good Small School?

The schools studied in this report were not selectedrandomly. The Dollars & Sense team identified more thanone hundred schools throughout the country throughdiscussion with other researchers and educators andthrough its own knowledge. The team then applied aseries of screens to determine each school’s suitability for asite visit and inclusion in the study. Schools had to meetsize limitations recommended in Dollars &Sense I – four hundred or fewer students in aschool at the time of selection, or no morethan one hundred students per grade in highschool, 75 per grade in a middle school, and50 per grade in an elementary school. Tomaximize diversity in a small sample, theteam looked at schools in urban, suburban,and rural communities in different regions ofthe nation (D&SI, pp. 7-8) and schools thatserved diverse populations. Again, as agroup, the schools needed to be governeddifferently, use different approaches to curriculum, andinclude different configurations of grades. The finalrequirement was that within the previous five years, eachschool had completed a facilities project such as newconstruction, addition, or renovation.

In addition, the team used specific criteria5 to assess thequality of schools selected:

• autonomy, • good leadership,• a defined mission, • making choices to ensure coherence and support the

distinctive nature of the school, • developing good working relationships within the

school and between the school and community, • equity, • respect, • a supportive culture, and • a commitment to academic achievement.

These, in the team’s estimation, are the most importantelements of a good small school. The schools also had toagree to share information and provide the Dollars andSense team the opportunity to talk with parents, students,teachers, and staff. The team also studied budgets todetermine if a school was spending within the sameallocation per pupil as other schools in its district.

Site teams visited the schools that appeared to meet thecriteria. The purpose of the site visits was to confirm that

the schools met the criteria and to gatherinformation on the strategies they used tooperate and to house themselves cost-effectively. Each site visit team includedan architect, a budget analyst, acommunity engagement specialist, and theproject director. Most of the teammembers were also experienced educators.During the site visits, the team observedclasses; interviewed teachers,administrators, architects, custodians, andother staff members; and held focus groups

with students and teachers. Each member of the site visitteam submitted a report from which the project directorwrote a group report. (Summaries are available on thewebsite at www.goodsmallschools.org.) Finally, the entireteam reviewed its findings and selected 25 schools as thefocus of this report.

No single school is strong in all areas, and what constitutesa good small school is linked to local history, culture, andneeds. These schools respect the needs and concerns ofstudents, parents, and communities. These schools areagile and willing to adapt to changing and unexpectedcircumstances. They are entrepreneurial; they use resourceseffectively and thoughtfully and they are able to thinkcreatively. Also, they have an identifiably goodeducational program6 supported by the facility. All schoolshave strengths – and weaknesses. The goal of this report isto highlight strengths that can be shared and emulated.

“All schools have

strengths – and

weaknesses. The goal

of this report is to

highlight strengths

that can be shared

and emulated.”

5 A more in-depth discussion of the criteria is available inAppendix 5.

6 The term “program” for the purpose of selecting schools appliesto those activities, including governance and administration,most directly connected to teaching and learning, as distinctfrom those related to the facility and its operations. Programcriteria reflect both Linda Darling-Hammond’s “10 Features ofGood Small Schools” as set forth in Redesigning Schools: WhatMatters and What Works, as well as those defined on page 7 ofDollars & Sense: The Cost Effectiveness of Small Schools. See alsothe Bill & Melinda Gates Foundation’s list of “What Makes aGreat High School” at www.gatesfoundation.org. A fullstatement of all criteria is set forth in Appendix 5.


5

Small schools typically demonstrate greater agility andflexibility than large schools in meeting changingcircumstances. Still, small schools can be fragile.Leadership, demographics, finances, facilities, curriculum– all of these may change, sometimes very quickly, andany of these changes can affect a small schoolsignificantly. It is important to point out, therefore, thatthe descriptions of the schools in this report are justsnapshots of a particular moment.

Most of these schools have operated successfully forseveral years. Some have existed for decades, but manyface continuing pressures from state legislatures, schoolboards, and school administrators who want to close small schools.

The selected schools offer a diverse sample of innovativesmall schools, but this is by no means an exhaustive list.Many good small schools across the country are operatingcost-effectively and have wonderful ideas to offer. Theschools in this report are simply ones that met the criteriafor this report, both individually and as a group. As aresult, these schools provide a wide array of strategies forbuilding and running good small schools cost-effectively.Dollars & Sense II aims to start conversations that willcontinue on the website or in future publications. Thatsaid, each of these schools has an important story to telland each is serving its students and community in waysthat are worth observing closely.

DEFINITION OF TERMS:

Advisory

An advisory is a small group of students, usuallyranging in number from 14 to 18, who work togetherand with their advisor for the entire time they attendthe school. Much more than a traditional homeroom,the advisory functions as “home-base” for students,and the advisor plays a major role in their lives. Thesmall size of the group and the time the studentsspend together allow close relationships to developand for advisors to know not only their students buttheir students’ families as well. In some schools, theadvisor is also the primary teacher who guidesstudents in developing projects and internships andmaking connections with their academic subjects.

For more information see: www.essentialschools.org

Internships

In an internship, a student works with an experiencedadult outside the school on a specific project in afield that interests the student. The internshipmentor supervises and offers opportunities to learnabout the field and work that expand the student’sskills in much the same way as skilled craftspeoplecoach apprentices.

The student’s teacher or advisor monitors that thework is meaningful and that the student is learningskills that connect to and supplement his or heracademic work. Because students help create anacademic program based on their own interests, theyare already invested in making it successful. Whatthey learn in the internship becomes the basis forlearning academic skills.

For more information see:The Big Picture Company: Learning ThroughInternships: www.metcenter.org

Mentor

A mentor is an older and more experienced personwho develops a relationship with a student and servesas a role model. Mentors are not usually teachers, butreceive training and support from the school. Thementor gives the student help in making decisions,encouragement, assistance in setting and fulfilling

(continued next page)


6

goals, and support in daily situations. The mentormay also help the student with academic work, butthe scope of a student’s relationship with a mentor isusually broader than that with a tutor.

Program

The educational program includes the curriculum,but also the way in which teachers teach and studentslearn. The educational program is designed to helpthe school achieve its mission and encompasses theway in which material in the curriculum is taught.For example, an internship-based program will differfrom a project-based program.

Project-based

In project-based education, students and teachers maywork together on projects such as writing, producing,and presenting a play, or students may workindividually on a project with the guidance of theirteacher. By being involved in a project, students havea chance to apply not only skills of analysis andproblem-solving, but also to gain academic knowledgein a way that is both rigorous and engaging.

For more information see:www.bie.org/pbl/pblhandbook/intro.php

Teacher-owned

In schools working under the EdVisions model,“teachers work not as employees, but as part of aprofessional partnership which contracts with schoolboards to operate schools...the teachers haveauthority over all aspects of running the school, frombudget to personnel decisions and salaries, tocurriculum and pedagogical methods."www.edvisions.com

Volunteer Tutor

A tutor gives private instruction that is oftenremedial. In the case of most schools in this report,all tutors volunteer their time to work with students,but MATCH funds its extensive tutorial programthrough the federal Work Study program.

Small Schools, Big Ideas

Introduction

The 25 schools highlighted in this report may be small insize, but their innovation and determination are immense.This section organizes the cost-saving ideas from theseschools into five broad categories: staffing, educationalprogram, services, sources of funding, and facilities(including construction, maintenance, and operations). Ineach category, the report describes cost-saving strategiesand offers specific examples from schools that have usedthem. Readers can refer to the “Strategies Grid” inAppendix 4 for a complete list of schools employing aspecific strategy.

Staffing

Salaries are the largest expense for any school. However,there are ways to attract and retain good teachers andadministrators without raising costs, including offeringattractive working conditions and providing incentivesbeyond compensation. The schools in this study useseveral strategies to reduce the costs of staffing and toimprove working conditions for their teachers.

Juggling Act

Savvy schools take advantage of opportunities in districtfunding formulas for staffing. They look carefully at theirown needs and juggle district allocations. For example,H.S. Truman High School in Federal Way, WA, has nolibrarian; it uses the money allocated for that position toemploy a tech specialist because that role is critical to itsalternative internship program. The school also redirectedthe money earmarked for one guidance counselor into ateacher’s salary. The counselor is not needed, because theteachers serve as advisors.

Using Rules Strategically

Schools can use rules to their advantage. For example, theTacoma School of the Arts (TSOTA) in Tacoma, WA,hires most of its staff in April and May. This timing givesschool officials the maximum flexibility in selection,while still operating within guidelines of the unioncontract. Precise and clear job descriptions help theschool find the best teachers for the job and also deflectprotests from the teachers union if the school chooses notto hire a teacher who meets the contract’s requirements,but not the school’s.


7

Double Duty

Small schools can conserve funds by havingadministrators who also teach. Administrators withteaching credentials can keep in touch with students on adaily basis and can better understand students’ concernsand teachers’ problems. Combining teaching andadministration also reduces the staff required in a smallschool. Avalon Charter High School and Oak ValleySchool have administrators who teach and teachers whoadministrate.

Sharing Administrative Duties

Small schools have found a way to obtain theadministrative leadership they need, without incurringdisproportionate costs. At Avalon Charter High School inSt. Paul, MN, teachers have a proprietary interest in theschool and share administrative duties. Each is paid asmall stipend for these duties in addition to his or hersalary for teaching, but the total expense is far less than itwould be to hire a full-time principal or bookkeeper.Avalon Charter High School, by having its teachersabsorb many of its administrative functions, savesapproximately $40,000 annually (the average staff salaryplus benefits).

Working within a network such as EdVisions (see Benefitsof Networks box) in order to have access to outside adviceand expertise, as well as funding, can also be costeffective. Schools such as Avalon Charter High Schooland Camino Nuevo Charter Academy in Los Angelesfound such affiliations invaluable, not only in the processof applying for the charter but in finding sites andestablishing the schools.

Laurel-Concord Public School in Laurel, NE, is a K-12school whose strong superintendent works closely with aforward-looking board.7 Together, they have put theschool on such a steady course that the superintendenthas taken on leadership of a second district, therebyrelieving Laurel-Concord Public School of half his salary,saving the district approximately $53,000 annually. Usingthe computer program PowerSchool (see box, page 8)simplifies many administrative functions and helpsadministrators, teachers, and students save time. Theschool has recently been able to share principals – as wellas its superintendent – with a smaller district.

Tiny Oak Valley School in rural Nebraska shares its part-time principal (a former superintendent and a practicinglawyer) with several other small schools, making his timeand expertise affordable. The shared principal costs theschool $4,599, including benefits, or three percent of thebudget in 2001-2002. Schools in the area more typicallyspend 10% of their budget for the principal, whichsuggests that Oak Valley School is saving between$15,000 and $50,000 annually.

BENEFITS OF NETWORKS

Several schools are allied with networks of schoolswith a similar vision, or affiliated with organizationsthat provide them with particular expertise. Forexample, Avalon Charter High School is allied withEdVisions and follows that group’s model of teacher-owned and -operated schools. The Burlington Streetcampuses in Los Angeles are part of Camino NuevoCharter Academy, a network of schools under theguidance of Pueblo Nuevo Development, a nonprofitcommunity outreach effort.

In schools that are part of a consortium, the cost ofexpertise and support services for accounting, payroll,curriculum development, facilities purchase orleasing, renovation, and maintenance is sharedwithin the network. For example, C.C. BlaneyElementary and R.D. Schroder Middle schools buytesting services, but not supervision, from the for-profit school management company Edison Schools,Inc. Although this costs rather than saves money atthe outset, over time it will reduce outlay as people inthe district assume responsibility for testing using thetraining they have received from Edison. Experts atExcellent Education Development helped CaminoNuevo Charter Academy find a site for its middleschool that was affordable and also helped negotiate acontract for its renovation. For more information oneducation networks, visit the websites listed below.

The Big Picture Company: www.bigpicture.org

Edison Schools, Inc.: www.edisonschools.com

EdVisions: www.edvisions.com

Excellent Education, ExEd: www.exed.net

High Tech High Learning: www.hightechhigh.org

New Visions: www.newvisions.org

Pueblo Nuevo: www.pueblonuevo.org

7 Dan Hoesing is superintendent of the district, which runs a K-12 program in the same building; however, there is a principalfor each of the three separate sections of the school: elementary,middle, and high school.


8

PowerSchool

PowerSchool, a division of Apple, provides web-basedstudent information systems (SIS) for small, medium,and large school districts. PowerSchool’s products areflexible and comprehensive systems that allowdistricts to use demographic data and studentperformance factors to inform instruction, increaseparental involvement, drive individual studentachievement, and produce local, state, and federalreports. Through PowerSchool, administrators build acommunity of interaction and accountability,teachers gain timesaving administrative tools, parentsgain immediate access to their children’s grades, andstudents can track their own progress and takeownership in their education. PowerSchool has builta record of success in a range of districts, from publicto private, rural to urban, and small to large. Fourthousand schools and more than two million studentsuse PowerSchool daily.

www.apple.com/education/powerschool

Sharing Staff

Some schools share staff other than administrators. Forexample, Camino Nuevo, a nonprofit organization in LosAngeles that runs a group of schools, uses two highlyqualified and experienced staff members to test andevaluate scores for all its schools and to serve as faculty.C.C. Blaney Elementary and R.D. Schroder Middleschools share several staff members, including an assistantsheriff who is in charge of security. Many people at TheMet in Providence also work for its sponsoringorganization, The Big Picture Company. Laurel-ConcordPublic High School plans to increase the number ofEnglish teachers it shares with a smaller district.

Importing Teachers

Some schools in areas where it is hard to attract teachersrecruit internationally. C.C. Blaney Elementary and R.D.Schroder Middle schools, for example, sent thesuperintendent to Jamaica to interview and hire teachers.These schools have also hired a teacher from NewZealand through a teacher-exchange program.

Accepting No Substitutes

A few schools – H.S. Truman High School and HighTech High in San Diego, CA, for example – choose not

to hire substitutes to cover for absent teachers. Insteadteachers work in teams and cover for each other.

Using Mentors and Tutors

MATCH in Boston hires tutors from local colleges towork individually with its ninth- and tenth-gradestudents. Students are required to spend 10 hours perweek, in addition to the regular school day, working withtutors. Since tutors are paid with federal work-studymoney, the school provides only $1.75 of their $15 hourlywage. Not only is this arrangement good for MATCHstudents, but the colleges appreciate having meaningfulwork-study jobs for their students. Colleges also welcomethis opportunity because they are now required to allocatea percentage of their federal work-study money toprograms that benefit the local community. It is easier forsmall schools to find mentors and tutors for all or at leastthe majority of their students simply because fewer areneeded.

Adjunct Faculty

TSOTA hires full-time teachers for core subjects, but usesadjunct faculty, much as colleges do, for classes in the arts.Hiring adjuncts allows TSOTA to attract practicingartists, dancers, and actors who prefer to teach part timeand who are willing to do so for a per diem rate withoutbenefits. The Visiting Artists program would not bepossible without adjunct faculty who help TSOTAincrease its offerings cost-effectively.

Using Volunteers

Interdistrict Downtown School in Minneapolis has almostas many volunteers as students. Volunteer help reducesthe load on teachers and makes the school a moreattractive place to work. Successful programs such asthose in place at IDDS, The Met, Avalon Charter HighSchool, The STAR School, Laurel-Concord PublicSchool, Camino Nuevo Charter Academy, and WOO relyon community and parent volunteers and some invest in apaid volunteer coordinator. This liaison at Camino NuevoCharter Academy in Los Angeles coordinates parentvolunteers, each of whom signs an agreement stating thathis or her family will give 15 hours of work per year tothe school.

Parent volunteers at Camino Nuevo Charter Academyassist in a variety of tasks, including helping to superviseplaygrounds and streets around the school, answeringphones, translating, maintaining an information line, andgreeting visitors.


9

The coordinator at IDDS helped bring more than 500volunteers into the school, many of them on a daily orweekly basis. In addition, she organized an “e-mentor”program in which students and mentors communicateregularly by e-mail. Some of the gains, such asimprovement in reading scores, are measurable, whileothers are immeasurable. Students see that people in thecommunity care about them and their futures, andteachers have many extra hands and eyes to help students learn.8

Laurel-Concord Public School initiated a small programthat matches people from the communitywith students in particular need of an olderbuddy. This informal advisory programbecame so successful that the schoolimplemented an advisory program based onthe TEAMMATE concept developed byUniversity of Nebraska football coach TomOsborne in which athletes tutor in localgrammar schools.9

Multiple Credentials

Many small schools hire people who have multiplecredentials, which is less expensive than hiring separatepeople to cover the same areas. For example, the leadteacher for the seventh and eighth grades at The STARSchool is also an experienced special needs teacher andwrestling coach. Teachers at H.S. Truman High School,by contrast, have a state waiver to cover a wider range ofsubjects without multiple credentials. Teachers of coresubjects at High Tech High are credentialed for eithermath and science or humanities (English/language artsand social studies). This lets a team of two teachers workclosely with as many as 50 students.

Hiring Young, First-Time Teachers

Some schools hire young teachers, not only because theycan be trained in the principles and philosophy of theschool but also because their salaries are lower. CaminoNuevo Charter Academy, for example, hires staff fromTeach for America, which gives them significant supportthrough summer camps and mentoring. While it was not astrategy Camino Nuevo implemented to save money,Dollars & Sense estimates the school may have savedabout $100,000 by hiring first-time teachers. (Estimatedby comparing salaries for first-time teachers at Camino

Nuevo with salaries for tenured teachers inthe LA Unified School District.)

Hiring Retirees and Career Changers

Some schools hire people who have retiredor who want to change careers. Thesepeople may not, at least initially, haveteaching credentials, but some states offera grace period before requiring theappropriate licensing. Retirees and peoplewho had careers in other fields bringexperience and expertise they can share

with students. They add richness and diversity to a schoolcurriculum and can also fill part-time and unexpectedvacancies. Sometimes they are willing to work asvolunteers. World of Opportunity (WOO) inBirmingham, AL, for example, relies on a teacher withsuperb credentials who used to work for the county. HighTech High has created its own accredited teachereducation program that allows young or mid-careerteachers to earn their California teaching credential forfree within a two-year period through an after-schoolprogram.

Project- and Internship-based Curriculum

Schools that have a project- or internship-based programaugment their teaching staff (and not their facilities) bygiving students the opportunity to work outside theschool. The regional director of outreach at High TechHigh develops connections with the business community.Through these connections, businesses serve not only assponsors, but also as sites for student internships. Otherschools, including Avalon Charter High School, H.S.Truman School, and The Met, also carefully match theirstudents with workplaces. Advisors play an ongoing rolein monitoring students’ progress and helping themintegrate and reflect on their work experiences in anacademic context. One outcome of these innovative

“The coordinator at

IDDS helped bring

more than 500

volunteers into the

school, many of

them on a daily or

weekly basis.”

8 Because the school thinks this position is so important, in 2004IDDS created three new part-time positions as LearningPartnership Coordinator (LPC). LPC teachers are relieved fromsome teaching duties in order to coordinate partnerships betweenthe school and the community.9 To find out more about this mentoring program, see: www.teammates.org.


10

programs is that students have a chance to work with avariety of people in interesting fields. Sending studentsout to internships and for work on projects is not an easyway to avoid responsibility for educating them. Well-designed programs can truly benefit students, but creatingand maintaining such programs take dedication, hardwork, and a lot of time.

This strategy may work most easily for a small schoolbecause larger schools may find it difficult to locateenough placements near the school. Internships locatedtoo far away from the school might requirethat students and teachers make long trips,which are time-consuming and expensive.

Other Incentives

Schools can allocate resources to retaingood teachers, a strategy that can prove lessexpensive than hiring and training newpeople. Laurel-Concord Public School givesteachers bonuses for extra work, and a localbusinessman set up a fund to rewardteachers for special effort. Laurel-Concord Public Schoolpays its teachers well, which may account for its lowteacher-turnover rate. In nine years, only seven teachershave been hired, four to replace teachers who retired.Laurel-Concord Public School also allows teachers toexchange health insurance benefits for payouts if theyelect to take coverage through a spouse’s health plan.Teachers at Laurel-Concord Public School can get moneyfor taking classes and a bonus if they agree to mentorother teachers. After studying the feasibility, MATCHcreated a dormitory-like space on its vacant third floor tohouse recent graduates from college. The schooladvertised for full-time tutors and received more than twohundred applications for 45 positions. Tutors receive $600per month plus room and board in exchange for workingevery day with the same MATCH student.

Educational Program

Every school makes decisions, in addition to staffingchoices, about how to teach and support learning. Thesechoices have an impact on how much it costs to run theschool. Small schools, precisely because of their size, arewell positioned to use strategic, cost-saving approaches.

Focused Mission

No school can be all things to all people. Although somemay see it as a weakness, a focused mission is really asource of strength. Small schools can offer depth in fields

of study and in methods of teaching (D&SI, pp. 8-10, 17).A focused mission means not doing some things, andavoiding the expenditures that these activities wouldrequire. Many of the schools in this study, for example,have chosen not to provide athletics, food service, ortransportation for students and allocated money saved tostrengthening the academic program.

Attracting Students

A school’s mission can attract students, along with therevenue that accompanies them. C.C. Blaney Elementary

School, for example, is creating a programbased in the arts and hopes to evolve intoa magnet school for students who livewithin its district but outside theimmediate area. Because Charleston allowsstudents to attend any school in thedistrict, the school may increase itsfunding by adding the per-pupil allocationthat follows students who attend fromoutside the immediate area. Conversely,

schools that focus on children in their ownneighborhoods reduce the time and money spenttransporting them and give their relatives and members ofthe community easier access to the school.

Course Cycles

Limiting the grade range of a school and structuring thecourse cycles accordingly can reduce expenses. TSOTAdecided not to serve ninth-grade students because itsclassrooms are scattered throughout downtown Tacoma.Because TSOTA was concerned that ninth-grade studentsmight not handle well the responsibility of attending classin widely scattered locations, the school decided to enrollonly students in grades 10-12. The school looks forstudents who are relatively mature and have developed apassion, although not necessarily a remarkable talent, forthe arts. Because the school does not serve ninth-gradestudents, administrators and teachers offer classes in athree-year cycle. This means that students from the threehigh school grades take classes together. A junior couldtake the same courses as a sophomore or senior. Studentssay that they like this arrangement because they learn fromthe older students, and older students learn from helpingyounger ones. This system reduces preparation time forteachers and simplifies scheduling. It also reduces the needto have teachers from as many different disciplines. Thesocial studies teacher, for example, teaches courses thatrepeat every three years as well as electives.

“Small schools,

precisely because of

their size, are well

positioned to use

strategic, cost-saving

approaches.”


11

Ordering Supplies

Supplies are expensive, particularly if they are wasted. Insmall autonomous schools, teachers can order only whatthey plan to use in their classrooms. At larger schoolsusing a central ordering system, some supplies are neverused because they don’t suit the purposes or preference ofindividual teachers. In contrast, at Oak Valley School thethree teachers participate in deciding what to purchase,and the school buys supplies a year ahead. Students helpsave money on paper by using their desktops and erasablemarkers for solving math problems. TSOTA is frugal aboutsupplies and saves money by having students sharewhenever possible. WOO provides materials such as paper,pens, and pencils to students who can’t afford them, andstudents understand they must use all supplies carefully.

Services

Services such as transportation, food, athletics, and healthcare are essential to students, and are expensive. Bothsmall and large schools can increase cost efficiency byoffering these services through partnerships in thecommunity. The schools in this study demonstrate greatcreativity in providing and, in some cases, expandingservices, while at the same time containing costs.

Transportation

Busing can be an enormous financial burden on a schoolor school system. The cost of buses, gas, oil, maintenance,garages, drivers, and parking has made some small schoolsdecide against providing transportation (D&SI, pp. 13-14). At Oak Valley School, parents not only drive theirchildren to and from school – often in carpools – but alsoprovide transportation for field trips. Flagstaff Arts andLeadership Academy (FALA) in Arizona has taken thesame approach, and parents or older students drive toschool. Several schools give parents a transportationstipend for driving their children to school, so the schoolsspend less than they would offering a traditionaltransportation program.

The location of the school can minimize transportationissues. Truman is located by a new transportation center(the nexus for a variety of transit routes in Federal Way)which helps facilitate the transportation of students toand from their learning internships. MATCH has easyaccess to Boston’s public transportation system. The cityoffers a reduced-cost pass to students, which MATCHsupplements when necessary. TSOTA relies on passes forstudents so they can use public transportation, and theycan earn physical education credit by walking to and fromclasses located in facilities throughout the downtown area.

Saving on Transportation

As noted in Dollars & Sense (pps. 14-15), the cost oftransporting public school students doubled between1970-1971 and 1995-1996 to nearly $10.4 billion.The National Center for Education Statistics reportsthat in 1999-2000 the average per-pupil cost fortransportation was $521. Many of the schools in thisreport spent well below that figure in 2003-2004.

www.ncela.gwu.edu/pubs/seareports/95-96/2enroll.htm

Bronx Small Schools spent on average between $5.45 and $18.64 per year per student.

IDDS spent $35,000 or $72.10 per student.

H.S. Truman High School spent $21,734 or $109 per student.

TSOTA spent $94,500 or $270 per student, based onmonthly passes for 10 months at $27 per month per student.

Laurel-Concord Public School spent $134,886 or$314 per student.

The STAR School spent $27,125 or $502 perstudent, because it is a rural school and most studentstravel more than an hour to and from school.

The Met spent a total of $100,000 for transportation,including the cost of taking students to internships.The average cost is $228.31 for 438 students.


12

Food Service

Some schools provide food options, not food servicebecause full food service, with cafeterias and kitchens, isexpensive. Schools that eliminate food service avoid thecosts of kitchen space and equipment, cooks, kitchenaides, food, storage, utensils and plates, and a separatecafeteria. Many of the schools in this study found creativeoptions for feeding their students.

• Avalon Charter High School has a small kitchen wherestudents can heat water and use a microwave. Studentsat all three schools in the building can buy lunch andsnacks at the food court located in the basement.

• At FALA and High Tech High, students buy mealsfrom private vendors who pull up atlunchtime. Camino Nuevo CharterAcademy contracts food service to acatering company that brings meals tothe schools.

• High Tech High decided not to haveon-site food because it didn’t want togive up any of the 40,000 square feet inits building to non-academic purposes.Instead, it has contracted with two localfood sellers who set up booths on site. A separatelyoperated mobile food seller (the “roach coach”) alsocomes to the school.

• The Met uses a slightly different approach. There aresmall warming kitchens in each of the four separateschools, but food is prepared in a centralized kitchenand distributed, with the help of students, to thesmall schools. The Met spends $330,000 a year onfood service.

• Schools in warm-weather cities such as San Diego andLos Angeles can afford not to have an inside cafeteria.High Tech High, for instance, relies on tables underumbrellas in a courtyard and benches in the open air.FALA has no inside cafeteria, but students eat in theopen space between the classrooms, or inside theclassrooms on inclement days.

Elementary schools, particularly those that serve a needypopulation, believe that offering students at least onegood meal a day is an essential part of their mission.Camino Nuevo Charter Academy serves meals preparedby a catering service in the cafeteria in the middle schooland a covered cafeteria in the elementary school. The

STAR School has a spacious and handsome kitchen thatopens into a dining area with large windows, revealing aview of the mountains surrounding the school. Twoparents work full time running the kitchen. The STARSchool spent about $538 per pupil on food for 54 studentsin 2003-2004. Other schools have opened their cafeteriasto senior citizens and use their kitchens to train studentsin food service.

Athletics

Providing athletics is particularly challenging for smallschools. Communities committed to competitive sportsoften don’t want to reduce school size because it limits thepool of athletes for competition. Sometimes, state anddistrict policies prevent a small school from offering

athletics because requirements are tooonerous. Some small schools choose not tooffer sports because they want to focus onacademics. Many schools omit athletics toavoid the expense of coaches, fields,equipment, gyms, uniforms, andtransportation to and from competitions,as well as liability in case of accidents.Whatever the reason, the result is thatstudents don’t have easy access to exerciseor the ability to develop physical skills.

There are ways to pare down the expenses of athletics.Some small schools offer less expensive alternatives toteam sports, such as aerobics, dance, kick-boxing, andyoga. Schools may also choose to eliminate some of thespaces for athletics that, after appropriate study, seemunnecessary. For instance, Todd Beamer High Schoolchose not to build separate varsity locker rooms. Ratherthan spend money on an amenity serving a few varsityathletes, the district decided to allocate funds to theacademic program.

Other schools have forged partnerships with organizationsthat provide athletic programs.

• IDDS has formed a particularly good relationshipwith the YMCA, and staff from the Y lead a varietyof sports activities and an after-school program forstudents of all ages.

• H.S. Truman High School does not offer an athleticprogram, but students use the facilities at a nearbypark and the Boys and Girls Clubs.

“Some small schools

offer less expensive

alternatives to team

sports, such as

aerobics, dance, kick-

boxing, and yoga.”


13

• Students at C.C. Blaney Elementary and R.D.Schroder Middle schools use the playgrounds andfields located between the two schools that were builtby the Charleston County Parks and RecreationDepartment on land given by the school district. TheParks and Recreation Department and the schoolsshare the cost of maintenance and the salary for theathletic director.

The STAR School takes advantage of its land andclimate. The school built an outside half court forvolleyball and basketball, constructed a straw balebuilding for wrestling, and laid out a cross-country trail.The STAR School offers competitive teams in wrestling,basketball, volleyball, and cross-country running. Allchildren are involved in one or more sports,and the school’s teams have competed withgreat success in their league.

One school, Laurel-Concord Public School,has discovered a way to make athletic programshelp pay for themselves. (See discussion of“selling memberships” on page 14.)

Health Care

Providing in-house health care can beexpensive, and small schools find it difficult to have afull-time nurse on staff. Some schools, such as C.C.Blaney Elementary and R.D. Schroder Middle schools, arelocated near enough to each other to share a nurse. Otherschools, such as MATCH, have decided that if a studentfeels ill, he or she, with parental permission, will go homein a cab. In an emergency, a teacher will take a student toa nearby hospital or call an ambulance. By not having anurse on staff, MATCH saves between $25,000 and$50,000 a year. Instead of health courses previouslyoffered by a school nurse, MATCH instituted a weeklyseminar on health care and illness prevention taught bymedical interns and nonprofit health agencies. The Metand Camino Nuevo Charter Academy each have a clinicwithin their facility that is open to the school community.The clinic in Camino Nuevo serves about four thousandpeople annually. Though this clinic does not save theschool money, it does help keep students healthy, whichincreases the school’s average daily attendance figures –important for reimbursement and fulfilling state andfederal requirements.

Sources of Funding

Many of the schools in the study raise funds from sourcesother than state and local fund allocations. As costsincrease, this strategy becomes even more important.10

Renting Space

Some researchers estimate that a school facility is usedless than 30% of the time.11 Late afternoons and evenings,weekends, and vacations are all times when the facilitycould be open to appropriate users. MATCH rents outspace to a church group for $17,000 per year. MATCHalso rents space on top of its three-story building for a T-Mobile cell telephone tower – a deal that brings in$24,000 annually.

At Laurel-Concord Public School, whenthe local education foundation built a newfootball field and track for the school,board members suggested including fourelevated boxes that could be rented forparties. Each of the boxes, which holds 20to 24 people, now rents for between $250and $450 an event, which includes studentcatering of pizzas and other lightrefreshments. This activity can bring in up

to $900 per night, and yearly revenue is about $20,000.

Business Located in Schools

For two years, MATCH housed a student-run cell phonefranchise in its first-floor storefront. The school earned$20,000 for a fund that was usually used to help itsstudents with college tuition. The store has since closed because of the glut in the cell phone market, but the school has solicited students’ ideas for a new retail enterprise.

Club SOTA at the Tacoma School of the Arts serves non-alcoholic beverages and music to students andmembers of the community, but it does so without chargeto increase its involvement with the community and givestudents a venue for performing. Nevertheless, it would befeasible for other small schools to adopt this strategy as amoney-making proposition.

“Many of the schools

in the study raise

funds from sources

other than state

and local fund

allocations.”

10 For more examples of partnerships, see Nathan, J. & Febey, K.,2001, and Lawrence, B.K., 2004.11 Lyons, J.B., personal communication, June 2002.


14

Trading Spaces

From 2001 until 2003, TSOTA paid a total of $27,000 touse space at the University of Washington. Now, though,the university trades space rather than renting it,exchanging it for use of TSOTA’s sculpture studio. FALAswaps space with the Museum of Northern Arizona. Inthe summer, the Museum uses FALA’s photography lab,four classrooms for its art and dance programs, and twoother classrooms for tribal art. In exchange, the schoolcan use the museum’s meeting hall, the gardens, and thedarkroom without charge. IDDS trades space at theYMCA and YWCA for space in the school in which theY’s hold computer classes.

Building Shared Space

With help from the city, Laurel-Concord Public Schoolbuilt its library and fitness and computer centers with theunderstanding that city residents could use the facility.C.C. Blaney Elementary and R.D. SchroderMiddle schools built the playground andathletic fields with help from the parksdepartment, and students and communitymembers use the facilities.

Selling Memberships

During the day, community residents canuse the fitness center at Laurel-ConcordPublic School for free. The school sellsmemberships for after-hours and weekenduse. At these times, members enter the well-appointed club using their personal creditcard, which is keyed with an individualelectronic code. In 2004, 160 families paid $10 per monthto use the club, and fitness center memberships typicallyraise $15,000 to $20,000 per year.

Renting Computers

Laurel-Concord Public School uses PowerSchool, a web-based student information system (see box on page 8).Parents and students with access to a computer can checkimportant school-related information – everything fromlunch menus to a student’s exam score. Many families inthis rural area did not have computers so the board andsuperintendent turned that problem into an opportunity.Like many schools, Laurel-Concord Public Schoolreceives donated computers, but these older machines arehard to maintain and upgrade, and operating the samesoftware programs on so many different machines isalmost impossible. Laurel-Concord students recondition

the donated computers, and the school rents them at $15per month to parents and students. This computer rentalsystem raised approximately $100,000 between 1992 and2004, income sufficient to purchase all the software theschool needs.

Grants

Many schools search for grants to supplement tax dollarsand other sources of income. WOO’s finance committeehas recruited and trained board members, staff, andstudents to help research and write grants, conduct sitetours of the facility, and request donations from potentialdonors. This effort has produced more than $80,000 indirect funding and grants. In 2004, The STAR Schoolreceived a grant of $220,000 for a well so that it will nolonger have to buy water. That year it also received an$85,000 grant to purchase solar panels, which will doublethe school’s capacity to produce electricity. Both were

emergency grants from the federal ImpactAid Discretionary Fund. IDDS received agrant of $50,000 from the State ofMinnesota Department of Energy for ademonstration project in solar energy, andMATCH received a Green Schools grantof $400,000 from the MassachusettsTechnology Cooperative that made itpossible to install passive solar panels.

Donations

As private schools already know well,alumni are a good source of donations.The four small schools within a school at

Morris High School in the Bronx have received help fromColin Powell, the most famous graduate of Morris HighSchool, and Caroline Kennedy Schlossberg has adoptedthe new small schools and worked on their behalf – bothmaking a donation and helping the schools raise funds.Corporations, philanthropies, and businesses that eithersupport the academic focus or are located in theneighborhood are also potential donors. The Miller WireWorks, Inc. in Birmingham leases space to WOO but alsosupports it with a monthly donation. Hewlett-Packardcontributed 25 laptops to MATCH, and Atamai inCambridge, MA, gave MATCH a donation of $100,000to support the math program, to be used for professionaland curriculum development. Laurel-Concord PublicSchool has a “Giving Tree” sculpture in the front hallhonoring donors to the Community Learning Center, andHigh Tech High displays a “donor tree” to honor residentsand businesses in the community that have contributed to

“With help from the

city, Laurel-Concord

Public School built its

library and fitness

and computer centers

with the

understanding that

city residents could

use the facility.”


15

the school. Several schools, including tiny Oak ValleySchool and WOO, raise money through their own non-profit 501(c)(3) corporations. WOO receivescontributions annually from approximately 116community donors totaling more than $80,000.

Financial Benefits of Good Relationships with the Community

Because the local constituent board for C.C. BlaneyElementary and R.D. Schroder Middle schools hasdeveloped close relationships with local businesses andgovernment agencies, it has been able to developstrategies that save taxpayers money and improve thequality of life for students, teachers, and members of thecommunity. It has helped forge partnerships between theschools and the Charleston County Parks and RecreationDepartment to build, operate, and staff athletic facilities.Similarly, it has fostered a partnership between theschools and the town to build sidewalks and a sewersystem to serve the school. Much of this work would havebeen impossible without good working relationshipsbetween local board members and the larger district.

Many schools have discovered the benefits of involving thecommunity and allowing local residents to use the schoolfacility regularly (D&SI, pp. 15-17). Laurel-Concord, forexample, found that the community supported a taxoverride in a year when other school districts were not aswell funded by their communities. The school credits thecommunity’s support to two things: first, the work of anintermediary hired to work with the community after aprevious attempt failed; second, the goodwill built throughresidents’ regular use of the school library, computer center,and fitness center. This strategy works because residents –even those without children – become attached to theschool and want to support it financially.

Sharing Space

Some schools rent space in their facilities, while othersoffer space as a community service. Laurel-Concord PublicSchool is open seven days a week so the community canuse the school. Half of one staff person’s time goes toscheduling community events in the school, and thecommunity pays half of her salary. Every day, five or sixgroups meet at the school well into the evening. The Metis also generous in welcoming community use of its facilitybecause it believes this fosters strong relationships thatserve the school and its students in a variety of ways,some of which save money.

Creativity Pays

Laurel-Concord Public School has employed a varietyof creative financing strategies to raise money. After arestaurant across the street from the school closed itsdoors, the site was vacant for many years. The schoolapproached the bank holding the mortgage and askedwhether, if the assessed value were high enough, thebank would donate the building to the school for atax deduction. The bank agreed, so people from theschool cleaned up the building and found a buyer – abusiness person who agreed to pay $55,000 over 30years so he could convert the building into ahardware store. This arrangement provides income tothe school and manageable cash flow for thestoreowner, and at the same time has contributed tothe sustainability of the community.

To fund its library and learning center, Laurel-Concord attracted donations from a variety ofsources. Over five years, the city paid $100,000, aneducation foundation contributed $100,000, and theschool paid $175,000. In addition, the schoolreceived a distance-learning grant for fiber optics anda conference room. The person in charge of thedistance-learning program works half time for theschool and half time for Wayne State University.

Recruiting Students

Schools in states and districts that allow parents to selecta public school have found ways to attract students andthe per-pupil allocation that comes with them. Nebraskaallows parents to send their children to any school in thestate, and Oak Valley School now draws 22 of its 30students from outside its immediate area. FALA spent$2,000 on a brochure to tell parents and prospectivestudents about its programs. This effort helped attract 36new students, who brought $225,000 in additionalincome. C.C. Blaney Elementary School in lowlandSouth Carolina is working to draw students from its largedistrict of Charleston by becoming an arts magnet school,using the mission itself to attract students. AvalonCharter High School sends representatives to localchurch meetings, the Saint Paul School Fair, and somelocal middle schools to recruit. The STAR Schoolencourages its families to recruit new students. TSOTAhas teams that include students, teachers, parents, andschool administrators to interview prospective studentsand be sure that the school is a good match for them,saving money by preventing attrition.


16

Federal Funding

Although the federal government provides only a smallpercentage of the money for public schools, this fundingcan be enormously beneficial to schools that receive it.The paperwork may be daunting, but the rewards aregreat.

Title I

C.C. Blaney Elementary and R.D. Schroder Middleschools receive $575,330 in Title I funds per year. Morethan 98% of families with children at C.C. BlaneyElementary School are eligible because theyreceived free or reduced-cost lunch. Ninety-five percent of families with children atR.D. Schroder Middle School, which drawsfrom a larger area, are eligible. Theelementary school uses its Title I money fora program based in the arts that has changedhow the school looks and works, the wayteachers teach and kids learn, and the wayits students measure themselves and theirpossibilities.

Impact Aid

Schools that serve children from military bases or NativeAmerican reservations are eligible for Impact Aid. TheSTAR School received $196,360 in 2003-2004 because 85% of its students are Navajo. This funding, in additionto the per-pupil allocation from the state and grants theschool has won, help make the school viable.

Work Study

MATCH can afford to have tutors who spend 10 hours aweek working with each of its ninth- and tenth-gradersbecause the federal Work Study program pays most oftheir salaries, a contribution worth $405,000 a year.Work-study students at Boston’s 66 colleges are delightedto earn money by helping younger students rather thandoing clerical work. Their colleges and universities arepleased because federal law now requires that receivinginstitutions spend a percentage of their work-study moneyon outreach to their neighborhoods. Students at MATCHappreciate the tutoring they receive because it is vital totheir success in the rigorous courses at MATCH.

Emergency Repairs and Renovations

This federal program didn’t last long, but it was a realopportunity for schools on the alert for facilities funding.FALA received a $94,000 grant that allowed the school

to build a wide deck that connects all its modularbuildings. The deck gave the modular units a sense ofpermanence and presence, and also provided space forpeople to study, eat, and socialize.

Qualified Zone Academy Bonds

The Qualified Zone Academy Bonds (QZAB) programmade $400 million available annually in credits tolending institutions for offering no-interest bonds toschools for facilities projects. MATCH used a zero-interestQZAB of $2 million to help fund the purchase of its

building. The program was re-authorizedfor 2004-2005. MATCH’s QZAB fundingdemonstrates that it is wise to be alert fornew federal programs. The applicationprocess may be onerous and the chances ofsuccess may seem small, but schools thatreceive funds from federal programs suchas QZAB can benefit substantially. (See:www.qzab.org)

Carl D. Perkins Vocational andTechnical Education Grants

These grants support vocational education and training.Funds may be used for equipment, curriculum materials,staff development, and other purposes relating tovocational education and training. For more information,see: www.ed.gov/offices/OVAE/CTE/perkins.html. TheMet received a Perkins grant of $189,000.

New Market Tax Credit

The New Market Tax Credit Program gives taxpayers acredit against their federal income tax for investing inqualified Community Development Entities (CDEs). TheCDE must invest the equity in projects that benefit low-income communities. Camino Nuevo Charter Academysecured a New Market Tax Credit of $900,000 through itsparent organization, Pueblo Nuevo, to help finance therenovation of its facilities. After one year, Pueblo Nuevorefinanced the loan and now estimates that it pays $750per student annually in debt service. For moreinformation on this program, see:www.cdfifund.gov/programs andwww.nationaltrust.org/community_partners.

State Programs

Some states offer programs that can help enterprisingschools. For example, Arizona offers an innovative taxcredit of up to $250 for married couples, which allowstaxpayers to deduct that amount from their personal taxes

“Students at MATCH

appreciate the

tutoring they receive

because it is vital to

their success in the

rigorous courses at

MATCH.”


17

for contributions made to public schools. The money goesdirectly to the schools, and the credit is available to anystate resident, regardless of whether he or she has a childin the public schools. Schools often use the money tosupport extracurricular activities and field trips that mightnot otherwise be possible. Taxpayers simply write a checkto the school and can even specify the program or activitythey want to support. FALA in Flagstaff and The STARSchool located in a rural area outside the city encourageparents to donate $250 to the school for field trips andextracurricular activities. Because the amount of thedonation is deducted from the amounttaxpayers owe, it directly benefits both themand the schools. In 2003, these donationsresulted in a contribution of $26,000 toFALA and about $2,000 to The STARSchool. For more about this program see:www.flagstaff.k12.az.us

Facilities Design and Construction

School facilities require enormous investment on the partof districts and taxpayers. The perception that largeschools are necessary for cost effectiveness is one of themost deeply held misconceptions about small schools.Together these two realities pressure communities to buildlarge school facilities (D&SI, pp. 3-4). Providing facilitiesis a particularly daunting challenge for people who wantto start new small schools in states that give them nofunding for facilities and where policy promotes largeschools. Nevertheless, the successful small schools in thisstudy have found a variety of ways to work within theseconstraints. These schools offer innovative strategies forsecuring sites and then renovating, building, andmaintaining their facilities.

Site Selection

The site is usually one of the most expensive items in aschool’s facilities budget, and obviously one of the mostimportant. Access to roads, public transportation, andutilities is crucial, and building connections to thoseservices is expensive. Proximity to local communityresources, both public and private, also is important. Toooften, large schools are built in isolation outside thecommunities they serve. This isolation can make largeschools more expensive to build, operate, and maintainthan they need to be and make it very difficult for theschool to serve students or the community effectively.12

Smaller School Sites

Besides the enormous cost of large sites, there are manygood reasons to locate schools on small, in-town lots. Untilrecently, however, most districts assumed that they needed alarge site to accommodate a new school. State policypromoted large sites and often followed guidelines about sizeoffered by the Council of Educational Facilities PlannersInternational (CEFPI). In 2004, these guidelines wererevised substantially to be more flexible. (See box, page 18.)

Schools are powerful agents of sprawl, and several statesare reconsidering the implications of large sites for schools: the cost ofinfrastructure to serve such sites, thedecimation of open and natural land, andthe costs of transporting students longdistances. Such states have begun tochange their requirements about schoolsite size (D&SI p. 6).13

Several of the schools in this study standon relatively small to very small lots. Todd Beamer HighSchool, with a capacity of more than 1,500 students, islocated on 38 acres; C.C. Blaney Elementary and R.D.Schroder Middle schools occupy 31 acres, including thecommon recreation facility the schools share with thecommunity. The Met site includes four separate schoolson approximately eight acres, each housing no more than120 students. The site also houses amenities such as thefitness center, performance center, TV studio, recordingstudio, health/medical/dental center, and commercialkitchen, that also serve the two other Met schools locatedon a separate campus nearby. Bronx Small Schools atMorris High occupy 1.38 acres. High Tech High has 1.35acres; Avalon Charter High School shares its site of oneacre with two other charter schools; and IDDS has oneacre. MATCH has no land at all, and the building is10,000 square feet on each of three levels.

“Too often, large

schools are built in

isolation outside

the communities

they serve.”

12 For more information on the issue of sprawl and schools, see“Schools for Successful Communities: An Element of SmartGrowth,” Council of Educational Facility Planners Internationaland the U.S. Environmental Protection Agency, September2004. See also: Beaumont, C. & Pianca, E., HistoricNeighborhoods in the Age of Sprawl: Why Johnny Can’t Walk toSchool, November 2000. 13 CEPFI & US EPA, September 2004, pp. 27-29.


18

SITE SIZES – New guidelines

The Council of Educational Facility PlannersInternational (CEPFI) recommended specific sitesizes in its 1953 guidelines that remained in placeuntil the organization revised them in 2004.

The old guidelines suggested a minimum of 10 acresfor elementary schools, 20 acres for middle schools,and 30 acres for high schools, plus one additional acreper 100 students. Many state policies still reflect theserecommendations. For a review of state policy onacreage requirements for schools facilities see:www.cefpi.org/pdf/state_guidelines.pdf.

In its new guide, CEFPI suggests a more open andsite-specific approach: “The site must be large enoughto support current and future educational programs.… [The traditional] rule of thumb does not take intoconsideration variations in educational program orthe difficulties of obtaining sizeable tracts of land indensely populated areas.”14

How can schools function on such small lots? The schoolsin this study use a variety of strategies, from formingpartnerships with diverse organizations, to sharingfacilities and functions. They also require that everyelement of the facility serve multiple functions:

• Todd Beamer High School – the athletic field islocated right next to the school, not at a distance asis the more common practice.

• IDDS and MATCH – use space in partnerinstitutions.

• H.S. Truman High School – uses space in Boys andGirls Clubs and an adjacent park.

The four small schools in the Bronx occupy most of theformer Morris High School, which was built on a smallurban site in the early 1900’s. The schools will eventuallyshare the entire space when the large high school isphased out. Students in all of the schools share the gyms,fields, two cafeterias, and a theater. Though in some waysthey are forced to function as one large school, the smallschools seem remarkably adept at sharing the spaceamicably and retaining their individual identities.

On the Met’s eight-acre campus, the four buildings, onefor each of the four schools, are separated, but the schoolsshare a kitchen, an athletic field and a well-equippedathletic facility that includes a climbing wall andbasketball court. Each of the schools has a warmingkitchen and a large space that serves as a cafeteria as wellas the place for meetings, presentations, studying, andsocializing.

Donated Sites

Donated land can save money, but school boards mustcarefully consider gifts of land. Often the expense ofbringing infrastructure to the site erases any savings from“free” land. This is particularly true when a developerdonates land near his or her new residential communityand hopes to benefit from services brought to the site bytaxpayers.

Donations can take a variety of forms. The founders ofThe STAR School, for example, literally mortgaged theirranch to buy 40 acres for the school and to develop thesite. The school leases the site and buildings from thefounders. Oak Valley School sits on land donated for itsuse when the school was first built. If the school evercloses, this land will revert to the owners of thesurrounding land. This same strategy would work for othergroups hoping to start a school.

Schools may be able to buy land below market rate withthe excess value credited to the seller as a tax deduction,as Laurel-Concord did when it received donated propertyfrom the local bank that had foreclosed on the previousowner.

Design and Construction

Some states and school districts promote the use of a pre-approved architectural plan sometimes referred to as a“cookie cutter” as a way to lower design and constructioncosts. This approach may not actually save money,however, and certainly won’t result in a building thatmaximizes its site or reflects the history and culture of itsneighborhood. Using the same design again and againprobably doesn’t save money for several reasons.Adjusting the land to the design is usually more expensivethan designing a building that fits the contours of the siteand maximizes natural light and cooling breezes. Acreative and attentive architect can design a school that isless expensive to build, operate, and maintain than aschool patterned on a generic design. A good architectwill listen to the people who will be using the building so

14 CEFPI and US EPA, September 2004, pp. 7-19,10.


19

the facility also reflects their needs, preferences, localculture, and history. An architect who understands theschool and the community will be able to suggest costefficiencies that do not compromise good design.

The architects of H.S. Truman High School held focusgroups with students and learned that many of them livedin dark, cramped apartments and cravedspace and light. The architects incorporatedthe students’ desires for space and light intotheir plan. By choosing unfinished plywoodas the final ceiling cover and omittingexpensive window trims, they saved moneythat they then put into soaring ceilings andlarge, light-filled rooms. The floors areconcrete, which is easy to install andmaintain; a barn door announces passagefrom each school’s common area into itsmore intimate learning spaces; garage doors open thebuilding to its surroundings. Light streams throughclerestory windows, and exposed heating and ventilationsystems work effectively and inexpensively. No schoolfacility based on a template could serve the needs of theschool and students so well, and few could have beenbuilt so cost-effectively.

Other schools and their architects have used similarinventiveness to save money and create workable spacesin aesthetically pleasing schools:

• The architect used cost-effective materials and madespaces flexible at Avalon Charter High School.

• The architects of Todd Beamer High School designeda facility that can serve multiple purposes andscenarios.

• The parents and teachers at STAR School coveredthe exterior walls of its modular units with a stucco-like substance to make them look as if they were builtfor the site.

• The architect for MATCH used as much as possibleof the existing structure and detail of the former cardealership that was renovated to house the school.

Free Design

Some schools have found architects who are willing todonate their time and expertise. TSOTA found anarchitect to donate time, and FALA plans to use a retiredarchitect who lives in the senior community with whichthe school shares the site.

Construction Bidding

The bidding process may be more competitive and resultin cost savings for smaller schools, because relatively fewcontractors are bonded for or willing to take on bigprojects. Sometimes policies and regulations limit theability of small contractors to compete in local school

construction projects, but morecontractors are able to compete for smallerbuilding projects.

In addition to saving money, the use ofsmaller contractors offers opportunities tolocal, minority, or female companies. TheMet, for example, is proud that 42% of itscontractors were minority or female.Constructing a school is an opportunity toinvest in a community, but large projects

usually must use not only labor but also materials fromoutside the area, often at greater cost.15

Renovation

Some state policies make it more difficult to renovate abuilding than to build a new one.16 (D&SI, pp. 5-6) Infact, though, building infrastructure to serve a newlocation is expensive, as is hauling the debris from ademolished school to a landfill. Using existing buildings,on the other hand, saves materials and preserves theculture and history of the area. Often older schools,constructed at the turn of the 20th century, were builtbetter than those that were built later to serve what wasexpected to be a bubble in school population. Architectsfamiliar with renovation can evaluate a building todetermine whether it is feasible to renovate it to serve21st century needs.

Buildings that never served as schools can also beadapted, and structural elements can be incorporated intothe new school design. The entrance hall, metal staircase,and Corinthian columns at MATCH, for example, arearchitectural elements that no one would include nowbecause to do so would be prohibitively expensive. Thesefeatures give the space in this 1917 building that was oncea dealership for luxury cars an enviable elegance and asense of the past.

“In addition to saving

money, the use of

smaller contractors

offers opportunities

to local, minority, or

female companies.”

15 For more information about the ways in which a school con-struction project can benefit its community see: Weiss, J. 2004.16 Beaumont, C. & Pianca, E., 2000.


20

The buildings that serve the Burlington Street sites atCamino Nuevo Charter Academy were once a mini-malland a warehouse. It was more cost effective to remodelthem than to build a new school. Existing spaces, such asthe open courtyard at the elementary school, provideddesign opportunities that would have been unlikely in anew space. The middle school facility, the formerwarehouse, features an interior atrium with walls that leanslightly in, creating the feeling of a small village. Thisenclosed courtyard, like a village plaza, serves as a hallwayand a gathering place for students, parents, and teachers.

The High Tech High facility seems ideal for the school,and it is hard to believe that it was actually designed to bea naval training and storage area. Theschool wanted a facility that looked like theoffices of a successful business yet couldserve the school’s needs. The architect keptas much of the original structure as possible,saving ribbon windows that run high alongthe walls and brighten an expansiveentrance hall leading to the “great room.”

Laurel-Concord Public School maximizedspace while minimizing expenses. Theaddition for the library and computer center was fundedthrough partnerships with the town and a local educationfoundation. The old library became a second mediacenter, and the space between two wings of the schoolwas enclosed to create a fitness center. Although thefacility reflects a number of styles from different periodsand can be confusing to people unfamiliar with its layout,it serves its constituency well and has been adapted tomeet changing needs. In 1978, shortly after construction,the open plan in the elementary wing was modifiedslightly to reduce noise. It may, however, be one of thefew school facilities that still functions as an open schoolin the way its designers intended.

TSOTA renovated two buildings, one for visual arts, theother for performances. The first was built in 1906, thesecond in 1908. Both offered a historic context that couldbe enhanced to support the TSOTA program.

Additions

Building an addition onto an existing facility can be acost-effective way to add space. Both C.C. BlaneyElementary and R.D. Schroder Middle schools needed alarger facility. The site permitted construction of anaddition that was much cheaper than relocating theschools. Laurel-Concord Public School has expanded by

constructing walls that join wings. At Oak Valley School,parents constructed two additions at minimal cost. Thefirst addition provided bathrooms and locker space, andthe second added another classroom and computer room.Parents also built a small playhouse for students. Studentsand teachers at WOO built a roof that joins the leasedbuilding to a trailer, which houses additional classroomspace and the school’s library.

Leasing

Leasing space can be a cost-effective alternative to buyingor building. WOO leases space inexpensively from MillerWire Works, Inc., its supportive neighbor and landlordthat assists the school and welcomes its impact in

diminishing crime in the area. AvalonCharter High School leases space and keptrenovations to $8.47 per square foot, inpart because it does not own theimprovements. FALA leases some of itsbuildings and has kept improvements to aminimum.

TSOTA leases spaces in two facilities withfive-year options to buy, which helps toprotect its investment in the buildings and

gives time to raise funds. MATCH leased space before itwas able to buy a building and used the time to define itsprogram and needs before investing in a facility. Many ofthese schools would not have been able to find a facility ifthey hadn’t initially leased space.

Adapting Existing Space Within a School

Another way to maximize space cost-effectively is toadapt existing space. This strategy can work if donejudiciously in conjunction with remodeling or building anaddition. Unfortunately, too many schools are forced todo this on a nearly continuous basis. To find space forincreasing numbers of students and services, they findthemselves converting closets into offices or basementsinto classrooms – a practice that is not recommended.Turning the unfinished third floor of MATCH school intodormitory space for tutors is a positive example ofadapting previously unused space to benefit students.

Using Students and Volunteers

The school “family” often includes people with wide-ranging skills. Inventorying these talents and intereststhrough a survey and compiling the results electronicallymay provide a useful resource to students, teachers, andeven members of the community as long as there is a way

“Many of these

schools would not

have been able to

find a facility if

they hadn’t initially

leased space.”


21

to assure privacy. For example, the Met and its parent, theBig Picture Company, inventory the auxiliary skills andinterests of everyone associated with its schools. Theinventory of parents, faculty, and staff should not onlyinclude current talents, but also identify skills people wantto learn. Not only can schools find volunteers with skillsin the building trades, but they may also uncoverexpertise that can help with individual student interests.For example, adults who play the dulcimer, speak fluentSwahili, or collect artifacts from the Civil War can bematched with students who want to learn more aboutmusic, language, or history. And not all volunteers needto bring a skill. They can learn on the job, provided theyhave expert supervision so people can do the tasks safelyand successfully. Schools that employ volunteers of anysort should, of course, be sure that their insurance policiescover their liability for these helpers.

Many of the schools in this study have joined studentswith local community volunteers to help build skills inthe immediate community. For example, at The STARSchool, parents and teachers covered the modularbuildings with stucco and built a straw bale structure forscience classes and wrestling. Parents, students, and staffat FALA created cinder paths and built wide decking toconnect the school’s modular buildings.Laurel-Concord Public School students didall the wiring for the new computer lab.They worked under the supervision of theirinstructor, and a licensed electricianchecked all their work. They not only savedthe school money, but also learned animportant skill. WOO students helped builda carpentry shop and the roof that connectstwo buildings.

Flexibility/Adaptability/Multiple Uses

One of the best ways to save money is todesign and construct spaces for multiple purposes.Creating successful multi-purpose space requires carefulplanning. Todd Beamer High School can house one largeschool or as many as eight small schools, with only minoradjustments, because planners designed the building to beflexible. Burying waste lines deeper than normal was moreexpensive initially, but allows for easy and inexpensiveplumbing changes, including adding sinks if necessary.Non-load-bearing walls can be removed easily to createnew spaces. Offices can be centralized or dispersed toseparate schools. Thanks to thoughtful planning, Todd

Beamer High School can relatively easily and cost-effectively adapt to new styles of teaching and learning.

Schools can also save money by sharing space. Studentsneed access to a variety of spaces, but it is usually lessexpensive to build or renovate and maintain a cafeteria,gym, or auditorium that serves multiple small schools thanto construct separate spaces for each school. The Met andthe Bronx Small Schools at Morris High School use thisstrategy effectively. Admittedly, however, sharing spacecan be challenging. Some schools have decided that it isnecessary to have some redundancy in facilities evenwhen they share the building. The small schools at MorrisHigh School, for example, received a grant in 2005 tocreate separate science laboratories. Even superbscheduling and gracious working relationships may notalways surmount conflicts over use of the space, and, inany case, it is important to give students and theirteachers their own space so they can build relationshipswithin their small school.

Sharing Academic Work Space with Partners

Using space in the community is one of the most effectiveways to reduce the amount of space a school needs toconstruct. As the quantitative research in this report

demonstrates, schools of all sizes reduceconstruction costs by reducing the numberof square feet per student in the facility.Sharing facilities expands the amount ofspace available to students and teacherswithout increasing construction costs. Forexample, IDDS uses local theaters fordrama classes and the facilities andpersonnel from the nearby YMCA for gymclasses. In return, IDDS lets the YMCAuse space in its facility at the end of theschool day for computer classes. IDDS hasalso traded space with the MacPhail

Center of the Arts to reduce fees for its Saturday Suzukiprogram and other music lessons. FALA uses a lab forphotography classes that is owned but not used by its host,the Museum of Northern Arizona. Laurel-Concord PublicSchool shares its library and computer room withmembers of the community because the town and a localeducation foundation contributed to funding thesefacilities. TSOTA uses space in several downtownlocations in exchange for space in its facilities. Thesespace-sharing agreements help schools and communitiesavoid duplicating functions and share not only space, butthe resulting savings.

“Not only can schools

find volunteers with

skills in the building

trades, but they may

also uncover

expertise that can

help with individual

student interests.”


22

TIPS FOR DESIGNING AUTONOMOUSSMALL SCHOOLS

That Share a Site

Plan enough distance between the small schools todiscourage students and teachers from spending muchtime out of their own school.

Create a welcoming common area in each school inwhich students, teachers, visitors, and parentsassociated with that school can socialize, study, meet,attend presentations, eat, and relax.

Design small comfortable nooks in which individualsand very small groups can study, relax, and socialize,as well as space for larger groups to meet, worktogether, and socialize. (From Elliot Washor of TheBig Picture Company)

That Share a Building

The school’s educational program, vision, and valuesshould drive facility design.

Each school must have its own space and physicalidentity.

Each school should have a separate entrance, lobby,graphics, color scheme, and furniture style.

Flexibility is a particularly important principle forsmall schools that share a building. Staggeredschedules and pairing schools with students ofdifferent ages can help make it easier to sharefacilities such as the cafeteria, gym, auditorium,library, and athletic fields.

Schools must be truly autonomous to be truly small.The physical structure must support the socialstructure to promote small schools.

Adapted from Facilities Design Considerations forSchools that Share a Building, KnowledgeWorksFoundation, 2003. www.kwfdn.org

Reducing Square Footage per Student

Students at schools such as Avalon Charter High School,The Met, High Tech High, and H.S. Truman HighSchool require less space per student because studentsspend part of many days outside the school. AvalonCharter High School, for example, allocates 118 squarefeet per student, but the space does not seem cramped orcrowded, because many students are in seminars orworking on projects at any given time. By contrast, large,public suburban high schools, with full athletic andperformance facilities, generally build 160 to 200 squarefeet per student. Similarly, MATCH built classrooms thataverage 500 square feet rather than the more usual 700 to800 square feet, because it intends never to have morethan 20 students in a class. The number of square feetthat is appropriate in a small school reflects the needs ofthe academic program and the number of students theschool anticipates will be in a class.

Sharing Space for Athletics and ExtracurricularActivities

Athletics can require significant and customized space.Many schools take advantage of the athletic fields, gyms,and pools in their community, saving space and moneywhile providing important outlets for students. H.S.Truman High School uses the adjacent park and Boys andGirls Club for athletics and other activities, while C.C.Blaney Elementary and R.D. Schroder Middle schoolsshare their common playground and athletic fields withthe community. IDDS provides space for its 475 studentsthrough partnerships with local institutions, including theYMCA and YWCA, MacPhail Center for the Arts,Orchestra Hall, the Guthrie and Orpheum theaters, andMinnesota Center for Book Arts. IDDS runs all of itsathletic and extracurricular programs through the Y’s.Laurel-Concord Public School runs its swimming programin a pool adjacent to the school that is owned by theRotary Club. MATCH students use the nearby Y forathletics and other programs. The school hopes todevelop its relationships with Boston University sostudents can use more of the university’s facilities.

Partnerships in Construction

Schools can reduce construction costs by finding partnersto share their initial expense, as well as the cost ofstaffing. C.C. Blaney Elementary and R.D. SchroderMiddle schools formed a partnership with the CharlestonCounty Parks and Recreation Department to constructthe playground, athletic fields, and courts, and they now


23

share in the cost of maintenance and staffing. The districtpersuaded the town of Hollywood, SC, to install a newsewer line, and a sidewalk over it, in an area whereneither of these basic amenities had been available.

Green Construction

Although it is difficult to eliminate utility costs, threeschools have reduced their dependence on electricity byusing a solar alternative. The STAR School purchases noelectricity; it is solar-powered with a back-up generator tosupply additional power on cloudy days. IDDS inMinnesota got a grant to install a trombe wall and heat-air exchange system that warms air before it gets to theboiler, reducing the school’s heating cost. MATCH alsoreceived funding from the Massachusetts TechnologyCollaborative Green Buildings Program for solar panelson its flat roof, which it uses to supplement electricityfrom the local utility company.

SAVINGS FROM SOLAR POWER

Alternative energy allows a school to beenvironmentally sound and cost-conscious.

Using supplementary solar power saves MATCH anestimated $12,000 to $15,000 annually.

The STAR School, which is entirely “off the grid,”estimates it saves $2,400 to $3,000 per year. In orderfor The STAR School to be on the electrical grid, itwould have had to pay the utility company to run thepower to the school at $20,000 per mile (for 1 to 2miles). The cost per month for using the power linewas estimated at the time to be about $200. The solarpanel and later upgrades and additions cost about$30,000. The investment, therefore, was recouped inno more than one year.

In 2004 STAR received a second grant to double thenumber of solar panels, which means that the schoolwill almost never need to use the back-up generator.

Architects can save money on electricity by orienting abuilding to receive maximum sunlight through amplewindows and skylights. Good siting, design, andlandscaping can also utilize the sun’s heat as well asprevailing breezes. The STAR School faces south so thatthe sun helps heat the buildings in the winter, whileoverhangs protect the interior from over-heating insummer when the angle of the sun’s rays are more direct.Deciduous trees provide shade in summer, but allow thesun to help heat buildings in winter.

Furnishings, Fixtures, and Equipment

Schools have found many ways to reduce the high cost offurnishings, fixtures, and equipment (FF&E) byencouraging donations and recycling, and using the helpof volunteers and students.

Furnishings and Furniture Donations

Just like small startup companies, new small schoolsshould take advantage of used furniture and discardeditems, so long as the items do not adversely affect theeducational program. For example, schools can encouragearea businesses to donate used furniture and equipment.In many cases, these donations are of a higher qualitythan the school can afford to purchase. There is apotential downside to this strategy, however. Thearchitect for the renovation of Avalon Charter HighSchool, for example, designed work stations that fit thespace and needs of the project ideally. When a majorcorporation was able to donate work stations, however,the school couldn’t afford to turn it down. The donatedfurniture suits neither the physical layout nor thestudents’ needs quite as well as a customized system. Still,it works and the savings were substantial.

In many cases, corporate FF&E donations exceed aschool’s expectations. MATCH opened its doors just assome area dot-coms were closing and a major Bostoncorporation was getting rid of a lot of quality officefurniture. MATCH’s conference room rivals that of manybig-business facilities thanks to corporate donations

Another strategy to get usable equipment, furniture, andfurnishings at little or no cost could be called“scavenging.” One of the co-directors of TSOTA hadexcellent contacts in the Tacoma school system so whenexisting public schools were refurbishing they were able toobtain free furniture and equipment.


24

Build It Yourself

Using simple but effective designs, students, teachers, andparents built much of the furniture at TSOTA, includingtables, work stations, and storage units. Similarly, parentsbuilt bookcases, wall dividers, and desks for the computerstations at Oak Valley School, and at The STAR Schoolparents, teachers, and students built furniture, storagesystems, and playground equipment.

Sharing Internally

To make the most of available FF&E,particularly equipment, the answer is simple:share. At TSOTA, people expect to share.Because the community is small and hasestablished a deep sense of trust, studentsleave their work out, and the school doesn’tneed lockers. Equipment, desks, workstations, and other furnishings are rotatedamong many students. The same ethos existsat WOO, where the only items students donot share are their personal notebooks, which containtheir work, assessment scores, and journal entries.

Flexibility/Mobility

The same principle of flexibility that maximizes thepotential and minimizes the costs of a facility can work forFF&E. Tables and some chairs at Todd Beamer HighSchool are on casters, so they can be moved easily tocreate different configurations. Teachers have storagespace in movable metal cabinets that allow them to lockup items, but also to move them easily when necessary. Adumbwaiter makes it easy for science teachers to movedelicate and expensive equipment between floors. At TheMet, food prepared in a central kitchen is transportedquickly and easily on carts to serving kitchens in the fourschools. These techniques save money because equipmentcan be used almost constantly throughout the school dayand often into the evening for after-school and adulteducation programs.

IT Built or Rebuilt by Students

Ideally, schools should have the best computers andprograms available to give students the training they needto succeed in today’s workplace, yet often IT equipment isdated and works poorly. With a little training, studentscan provide the necessary knowledge and skills tomaximize computer resources. Several schools in thisstudy use students to service computers and other ITequipment. Laurel-Concord Public School students

rebuild donated computers that the school then leases tolocal families who need them. Students gain experienceand training in computer repair and servicing, while theschool gains income and families get computers at home.

IT Serviced by Students

Students at The Met staff a call center, answering requestsfrom anyone on the campus for technical help andcomputer repairs. Students gain work experience, and the

school reduces the number of computersthat are out of service. This system alsoallows the head of campus IT to devote hisattention to more complex issues.

Maintenance

Although the school facility is a valuableasset, too often the pressures of fundingsalaries, new textbooks, and othernecessities take precedence over buildingmaintenance. Schools that put off

investing in their facilities undermine the structuralintegrity of their buildings and increase the amount ofmoney they will have to put into future repairs. In somecases, deferring maintenance can even force the closure ofa building and the loss of a school.17

Investing in Preventive Maintenance

One of the most cost-effective ways to maintain a schoolfacility is to establish a proactive maintenance plan.Emergency repairs are always more expensive than routinemaintenance. The Laurel-Concord Public School boardhas a sensible and cost-effective approach to maintenance;it reviews needs with the head custodian on a regular basisand has a 10-year plan for renovating or refurbishing everysection of the school. The head custodian can ordermaterials ahead of time, thus saving money, and plan workfor when it is cheapest. If maintenance problems arise thathave not been scheduled, the plan can, of course, change.This disciplined approach has saved the schoolconsiderable money.18

“One of the most

cost-effective ways to

maintain a school

facility is to establish

a proactive

maintenance plan.”

17 Lawrence, B.K., June 2003.18 NCES, February 2003.


25

Reducing Trash and Graffiti

All the schools in this report have remarkably fewproblems with trash or graffiti, despite the fact that manyof them are surrounded by neighborhoods where both arecommon. One of the benefits of being a small school isthat people identify with the school and take pride in thecampus. As a result, small schools save money on upkeep.

Maintaining the Building

Every morning a parent stops by Oak ValleySchool and asks the head teacher ifanything needs fixing. If it does, the parentfixes it. One FALA parent is a plumber, andfixes any of the school’s plumbing problems.Parents at The STAR School alsocontribute to maintaining the buildings andgrounds, and students, teachers, andvolunteers clean and maintain the buildingsat WOO. Parents at Camino Nuevosupplement the work of regular staff in thecafeteria and assist with some maintenancework. Students at Laurel-Concord areresponsible for some of the cleaning andmaintenance at their school. Students havean opportunity to learn skills while gainingpride in their school. As a result, there is no graffiti in thebuilding and maintenance costs are lower. Relying onvolunteers for essential services, however, requirescontinuous investment in recruiting new people andmaintaining good relationships with those who donatetheir services to the school.

Maintaining the Grounds

Schools under financial pressure may not think they caninvest in landscaping, but landscaping is important.Landscaping anchors the building visually to itssurroundings and can reduce energy costs. Designing,planting, and caring for the landscape can increase theinvolvement of students, teachers, and members of thecommunity, as well as teach them about the localenvironment. At FALA, students working with a graduatestudent renovated a greenhouse and built gardens near theschool, and the music teacher tends to the gardens aroundthe school because he enjoys it. Students at The STARSchool build their own gardens using the techniques ofthe people who lived in the same area about 500 A.D.They also built a straw bale greenhouse to show thecommunity that it could be done inexpensively, yet yieldvegetables and greens almost year-round. At Laurel-

Concord Public School, the agriculture classes areresponsible for planning and maintaining the landscaping.

Using Computers Effectively for Maintenance and Operations

IDDS uses a computerized system to monitor temperature,lighting, and other mechanical functions throughout thefacility. Any of the four engineers on staff can instantlysee how the systems are functioning and make necessary

corrections.19 This kind of sophisticatedsystem works well when people in thebuilding know why it exists and do notundermine it by overriding the automaticlight and heat sensors. Indeed, continuoustraining may be necessary to make facultyand administration aware of itsadvantages. The furnace, cooling system,and electricity are used only when needed,so the school saves money.

At Laurel-Concord Public School, thePowerSchool software helps theadministration manage many of theaspects of building operation, includingtracking cleaning supplies, utilities, and

breakdowns. By managing these functions locally, ratherthan through a central office, the software avoidsinefficiencies and saves money.

Facilities Operations

Because small schools have less space, fewer people, andclose relationships and easy interaction between people,they are able to use several strategies to reap additionalsavings that are not available to large schools.

No Janitorial Staff

TSOTA’s janitorial staff works only at night. During theday, teachers and students are expected to care for thebuilding. Until the fall of 2004, FALA had no janitors;students and teachers were entirely responsible forcleaning classrooms. Until 2004, the head of the board ofOak Valley School picked up the trash. The STARSchool leases a trailer on site to a family whose membersserve as the school’s caretakers.

“Designing, planting,

and caring for the

landscape can

increase the

involvement of

students, teachers,

and members of the

community, as well as

teach them about the

local environment. ”

19 The head engineer helps the director in planning, purchasing,and many other responsibilities. The other three engineers, atitle assigned by the union, perform work more usually thought ofas custodial. The K-12 school always has two engineers on duty,but there is little overlap in the shifts so all four are at the schoolat the same time for less than two hours a day. The directorthinks that a K-12 school requires more such staff than oneserving fewer grades.


26

Camino Nuevo Charter Academy parents help staff withjanitorial duties, as do volunteers, students, and teachersat WOO, Oak Valley, and The STAR School. Students atAvalon Charter High School, FALA, and High TechHigh do much of the janitorial work. Not only is thiswork free to the school, but it builds camaraderie,contributes to a sense of ownership, and reduces litter.

Reduced Security Staff and Equipment

One of the many strengths of small schools is that thebest security results from personal relationships amongstudents, parents, teachers, and staff. Manysmall schools spend very little on security. Ingeneral, rural schools are usually lessconcerned about security than urbanschools. The small urban schools in thisstudy are able to spend relatively little onsecurity because people in the schools knoweach other, an intruder would stand outquickly, and people in the neighborhoodhelp protect the school (D&SI, pp. 8 -10).Small schools in this study demonstrateways to save money while maintaining a safe school.

• C.C. Blaney Elementary and R.D. Schroder Middleschools share the services of a deputy sheriff paid bythe county. The deputy sheriff is on campus at alltimes, working primarily as a community liaison androle model – counseling students much more oftenthan correcting them.

• Although students enter Morris High School in theSouth Bronx through a metal detector, the securitystaff has been reduced since the introduction of smallschools. The assistant principal who oversees securitysays there are far fewer problems than in the previouslarge school. Students in several of the small urbanschools seem to feel much safer in the school thanthey do on the street. In fact, safety is one of theelements of a small school that students and parentsat all of the schools in this report value most highly.

• Children in Camino Nuevo Charter Academy,located in one of the highest crime rate areas of LosAngeles, say they love their school because there isno fighting. Parents appreciate the school because ithas brought a sense of peace and safety to theneighborhood.

• The Met has no security fence but has securitypersonnel on campus who are family members ofstudents and members of the community. Goodrelationships with the community are a criticalcomponent of the Met’s security system.

• When WOO suffered a series of break-ins thatdamaged windows and doors and resulted in the lossof two small televisions, the school appointed a localhomeless man as the volunteer building securityofficer, posting his picture with the new title. The jobhelped raise his self-esteem, and, in exchange for

food, clothing, and a daily newspaper, thisman watched over the building. Therehave been no problems with vandalism ortheft since he took over this duty. WOOhas also helped him find housing in thepublic apartments four blocks away andhires him occasionally to help with clean-up projects.

Utilities

Utilities are a necessary and constantexpense, but many schools have found innovative ways tosave money. The schools mix complex systems withcommon sense to cut utility costs.

Natural Lighting

Natural daylight, cooling breezes, the heat of the sun – allcan be used to reduce utility costs. Windows in many of theschools in this study are larger than designers of schoolfacilities would have recommended in the late 1970s and1980s, when the cost of fuel soared and its availabilityplummeted. Natural light – which is free – creates ahealthier work environment than fluorescent orincandescent lighting.20 At High Tech High, skylights andwindows provide enough ambient light so that manyclassrooms and common spaces almost never need electriclight. Furthermore, schools without adequate ventilationbecome dangerous because they foster toxic molds; somethat were built or remodeled at the end of the 1970s havehad to be abandoned because of the health risks resultingfrom the lack of airflow. High Tech High, Todd BeamerHigh School, H.S. Truman High School, MATCH, TheMet, IDDS, and C.C. Blaney Elementary and R.D.Schroder Middle schools are all notable exceptions, withample windows that provide light and allow air to circulate.

“Natural light –

which is free –

creates a healthier

work environment

than fluorescent

or incandescent

lighting.”

20 Lyons, J.B. November 2001, p. 3-4; Schneider, M. 2002


27

Dual Fuel

Oak Valley School has two furnaces, one that uses oil, theother gas. Not only does this provide the school with aback-up for emergencies, but it also allows the school toswitch to whichever fuel is more cost-effective at thetime. The same is true for the Morris High School facilityin the Bronx, which is heated by four dual-fuel gas and oilboilers installed in 1997.

Automatic Light Sensors

Several schools use automatic sensors to turn off lightsafter a room has been unoccupied for a set amount oftime. Such sensors are in classrooms at MATCH, thebathrooms at The STAR School, and throughout The Met.

Solar Power

As mentioned in the previous section on greenconstruction, IDDS, MATCH, and The STAR School alluse solar power to augment, or in the case of The STARSchool, to provide electricity. Each is a good example ofhow solar power can lower a school’s utility costs. TheSTAR School is located in a rural area in the cool,mountainous area of northern Arizona, MATCH inBoston, and IDDS in Minneapolis. None of these placesseems the ideal location for generating solar power, yeteach is doing so successfully.

Keeping Cool

Keeping school facilities cool in hot weather is expensive.Air conditioners came with The STAR School’s portableunits, but they would have drained the solar-poweredelectrical system. The STAR School replaced the unitswith “swamp coolers” that cool the air by diffusing a finemist into the air. They are much less expensive to operate,can also be used to heat the buildings, and are much moreefficient than traditional heating, ventilation, and air-conditioning (HVAC) systems.21

Automatic Water Shut-off

In arid places such as Arizona, using water carefully isobviously important, so it is no surprise that The STARSchool has invested in automatic water shut-off for itsbathrooms. It is more of a surprise to find a similar systemat MATCH in Boston, but the devices conserve water,reducing water and sewer bills at both schools.

Wind Power

Wind is a potential source of energy not fully exploited byany of these schools. The STAR School is, however,exploring the use of a wind-powered generator tosupplement its solar panels.

Computerized Control Systems

Using a computer to monitor and adjust temperature andlighting makes a lot of sense because it saves dollars.IDDS saves on utility bills because the computerizedsystem gives engineers and custodians information 24hours a day, seven days a week.

Retrofitting

During the energy crisis of the 1970s and ’80s, manyschools blocked expansive windows, either partially orcompletely. Now windows that insulate more effectivelycould replace those installed in the 1970s and ’80s,restoring natural daylight and breezes, reducingdependence on energy, and saving money. Architectsrecommended that all the windows at Morris High bereplaced with ones that conserve energy. Many local gasand electric companies offer home owners the chance tohave an inspector visit their house and makerecommendations for conserving energy. The same serviceis available for schools.

21 Schools could also use storage systems that draw power during off-peak times to heat bricks and make ice. Air forced overheated bricks distributes heat, and, conversely, air forced throughice cools the facility.

desks, and their advisees use individual work stationsarrayed behind half-walls that separate the advisories

from thecentralcourtyard.

The schooluses a varietyof means tostay within itsbudget. ThePillsburyCompanydonated workstations, which– though notas ideal for thespace as thoseoriginallydesigned bythe architect –had theenormousadvantage ofbeing free.Students use

public transportation to reach the school, and mostbuy lunch in the basement from a private food servicethat serves all three schools in the building.


28

Avalon Charter High School shares a 1916 brickbuilding with two other charter schools; it occupiesthe third floor, the previous home of a charter thatfailed. Like its sister school, New Country inHenderson, MN, Avalon Charter High School is affil-iated with EdVisions Cooperative and is owned andrun by its teachers.22 Teachers share administrativeduties, as well as the responsibility for teaching andplanning, which gives the school a strong sense of col-legiality. Of course, these multiple roles can at times

burden educatorswho are alreadytrying to fit aproject-based cur-riculum into stateacademic require-ments.

The school cost-effectivelyrenovated its spaceto include areasalong the perimeter

for each teacher and the students he or she workswith throughout one or more years in a group calledan “advisory.” These high-ceilinged spaces open ontoa more intimate interior courtyard known as TheCafé, with tables and chairs and a kitchenette forsimple food preparation. Advisories are paired in thisarrangement. Two teachers work at corresponding

Profile Avalon Charter High School: An urban high schoolSt. Paul, MN • Grades 9-12 • 125 Students

“We, the People of Avalon, in order toprovide for a safe and productivelearning environment, promote theobtaining and usage of knowledge forthe benefit of those around us, andensure general happiness, do ordain andestablish this constitution of AvalonHigh School.”

— From the Constitution for Avalon School writtenby students and their advisor.

“Teachers share

administrative duties,

as well as the

responsibility for

teaching and

planning, which gives

the school a strong

sense of collegiality.”

Avalon Charter High School floor plan

Students work at their own stations inconfigurations they and their advisorarrange.

A student studies in the entrance hall tothe school, which also serves as a galleryfor student projects.


29

The curriculum is based on projects that studentsdefine and execute with the guidance of theiradvisors. Avalon Charter High School has notraditional classes and teaches through a combinationof independent projects and seminars open to allstudents that fulfill graduation requirements, as wellas seminars to cover material that will be on statetests. As often as possible, it uses the community as aclassroom, and students visit nearby sites including

places such as RootRiver Trail inMinnesota, the sitefor a recentseminar inalternativetransportation thatthey reached onbikes.

Avalon CharterHigh Schoolcombines a collegepreparatory

program with project-based learning and creates, asits website notes, “a context-rich curriculum with aglobal emphasis stressing depth over breadth and

drawing upon the passions and interests of itsstudents.” It succeeds remarkably well in this.

What is most striking about this school is thematurity of its students, the depth of their self-knowledge, and their concern for each other andtheir teachers. The student congress, open to allstudents, is remarkable for its level of discussion andthe fact that the views of the students are takenseriously by teachers. Students seem to feel that whatthey think and want to have happen at AvalonCharter High School is important, and they take theopportunity to be a part of forming their school.

“The student

congress, open to

all students, is

remarkable for its

level of discussion

and the fact that the

views of the students

are taken seriously

by teachers.”

22 For more information about EdVisions Cooperative, see:www.edvisions.com.

Low walls separate advisories, which are grouped in pairs,from the central circulation areas, the large interior café, andclassrooms. High ceilings and large windows give a feeling ofspace and let in ample daylight.

The café offers a comfortable place to study, have lunch, and socialize.


30

Profile C.C. Blaney Elementary School and R.D. Schroder Middle School: Ruralelementary and middle schools Hollywood, SC • C.C. Blaney: Grades PK-5 • 190 Students

R.D. Schroder: Grades 6-8 • 380 Students

A first-grade girl at C.C. BlaneyElementary School pops up from her seat andwalks confidently towards the visitors. She extends

her hand and giveseach one a firmhandshake and asmile as she looksdirectly at themand says, “My nameis Taisha and I amour ClassAmbassador. I wantto welcome our

visitors to our classroom. Today we are studying math,and I invite you to share in our learning.”

Ninety-eight percent of the children at C.C. BlaneyElementary School are eligible for free or reduced-cost lunch. The percentage at R.D. Schroder MiddleSchool, which draws from a larger geographic area, isonly slightly smaller. Yet, these schools are brightwith color, laughter, activity, warmth, and a sense oforder and purpose.

The schools serve a rural African-Americanpopulation in the very large and diverse district ofCharleston County. Until recently, both were

severely under-funded. As a result, they lackedsupplies, good teachers, and sufficient facilities. Asthe gifted head of the local constituent board, Leroy

“These schools are

bright with color,

laughter, activity,

warmth, and a

sense of order

and purpose.”

The successful renovation and new construction project as shown in this rendering of the C.C. Blaney Elementary School hastransformed the facility so that it now supports the arts-based program and has become a center of its community.

The Kindergarten room is a welcoming place for students andteachers alike.

Elementary students and their teacher work together on mathproblems.

“I invite you to share in our learning.”

— Student


31

Seabrook, puts it, “There was a time we weren’t seenas family.”

The middleschoolbuilding hadfour windows,and manypeoplereferred to itas “thedungeon.”After afacilitiesevaluation

revealed unacceptably poor conditions, the districtallocated funds to renovate the elementary andmiddle schools. Costs to remedy the district’s facilityproblems were estimated at $600 million, a large sumeven for this relatively affluent school district. Yetofficials committed to upgrading district schools,starting with the worst facilities. C.C. Blaney

Elementary andR.D. SchroderMiddle schoolswere two of thefirst to benefit fromthis decision toright old wrongs.23

This renovationhas brought parents back to the school and givenpeople pride in their community. It has forgedpartnerships between local government and thecommunity. The Charleston County Department ofParks and Recreation bought land between the twoschools from the Department of Education andturned it into a recreation facility for students, as wellas for the community. The parks department pays partof the salaries of maintenance personnel and a full-time coach, and the school pays the balance. Duringthe school day, children use the playground, fields,and courts almost constantly; later in the day, adultsin the community play league games with teams fromsurrounding communities. The Department of PublicWorks built a sidewalk over the new sewer systemthat serves the school, which makes it easier for

“This renovation has

brought parents back

to the school and

given people pride in

their community.”

students, family members, and people from thecommunity to get to and from the school. Results?For the first time, there was standing room only at the

elementary andmiddle schoolgraduations. Peoplelook to the schoolas the center oftheir community,and test scores arerising.

The school iseligible for Title I money because it serves a very poorpopulation. It has used that money to hire art and dramateachers to create a curriculum centered in the arts. Theschool hopes to become a magnet in its district for thearts, but the benefits of investing funds this way arealready evident. Improving the schools is beginning tobring white families back to public education – one ofthe school’s goals. Student art enriches the walls ofhallways and classrooms, and students are growing morearticulate and confident, and learning to enjoy school.This is why, without prompting from the teacher, Taishabounces up out of her seat and welcomes four unknownadults into her world.

“For the first

time, there was

standing room only

at the elementary

and middle school

graduations.”

23 These two schools serve a slightly different population.C.C. Blaney draws elementary students from the immediatearea while R.D. Schroder serves middle school students fromvillages in a wider geographic area.

An exchange teacher from Jamaica enjoysworking with a group of students.

Elementary students are eager to answer the teacher’s question.

Profile Bronx Small Schools: Four new small urban high schools within aschool: International, Leadership, Excellence, Violin and DanceMorris High School • The Bronx, NY • Grades 9-12 • Students: International: 300 •Leadership: 340 • Excellence: 330 • Violin and Dance: 170

“Usually a principal’s job is lonely, but itis good to share and learn and here weprincipals all talk. We have to trusteach other more than anything and notlobby for ourselves in the building withthe other principals.”

— Shael Polokow-Suransky, Principal, BronxInternational High School


32

The Bronx Small Schools’ building,constructed between July 1901 and January 1904 andrestored by the city in 1997, is both beautiful andintimidating.24 It sits on a hill in the South Bronxoverlooking the poorest congressional district in thecontinental United States. In recent years, to anyonein the South Bronx, the words “Morris High School”

evoked images ofcrime, drugs,violence, apathy,racial division,over-workedteachers, and angrystudents.

In the early 1900s,however, MorrisHigh School wasinnovative – thefirst high school inthe developingresidential areaknown as theBronx and the only

co-ed high school in New York City. Built at a cost of$469,383, the 2,679,500-cubic-foot facility hadelectricity, telephones, and an excellent ventilationsystem, then still a novelty. Classrooms with largewindows maximized daylight.25

C.B.J. Snyder, superintendent of school buildings atthe time, designed the facility to house 2,500students, but when it opened three years later it hadto accommodate 2,630 because the population of the

Bronx had expanded so quickly. With the approval ofhis board of education, Snyder, a Bronx resident,created the school as a testament to the new boroughand its pride in education.26

Much has changed in the intervening century. TheSouth Bronx and Morris High have been throughdifficult times, as is demonstrated by the school’selaborate security system. Even so, students in thesmall schools are comfortable enough to smile andgreet visitors. Teachers are laughing and talking withtheir students, and kids are lounging comfortably ondesks and tables in the halls as they wait for classes tobegin. And when given a chance, they say, “I lovethis school.” “We are family.” “Here we are safe.”

What is different now? The answer is that New YorkCity has made a real commitment to small schools. Inthe fall of 2002, Morris High began phasing out thelarge high school and replacing it with four smallschools: International, Leadership, Violin and Dance,and Excellence. A new principal assumed leadership of

“Though the new

schools occupy space

next to each other

with little physical

differentiation

between them, the

extent to which

people have already

imprinted an identity

on their school

is remarkable.”

The auditorium was restored by the city of New York to itsoriginal elegance and can be available to the community whenused by students in the small schools that now share theMorris High School facility.


33

Morris High to promote the transition and watch overthe demise of the old Morris High School. Many studentsand teachers joined the new academies, but otherswanted to graduate from Morris High School. Oncethese students graduate, Morris High will cease to exist.

Though the new schools occupy space next to eachother with little physical differentiation between them,the extent to which people have already imprinted anidentity on their school is remarkable. Lots of studentwork is exhibited on the walls. Students are enthusiasticabout their teachers and the opportunities they seeahead of them if they work hard. Other changes areoccurring as well. Since the small schools opened, thecustodian reports a 20% decrease in vandalism withinthe building, and the security staff was reduced from 36to 12. The community is beginning to see the school asa safe place and a resource. Between 3 and 6 P.M., 50%of the building is used for tutoring, after-schoolprograms, and clubs, including the Police AthleticLeague and the Explorers Club. A volunteer liaison tothe community organizes events and works to bring thecommunity into the schools.

Still, significant challenges exist – one simply theresult of success. There is a waiting list of students whowant to enroll in the four small schools. Forgenerations, the Department of Education has countedon – and budgeted for – serving less than a fullcomplement of eligible students because of lowattendance and high drop-out rates. Now attendance ishigh at the small schools, and fewer students drop out– so the city wonders how it will fund seats for studentswho come to school more often and longer than

students who attend large schools. Teachers say theyhave too little money for supplies, and some complainthat they are exhausted because they work late andare expected to be intouch with students’families in theevening and onweekends. A studenttold us, “You canreach the teachersanytime. They giveyou their e-mail andphone numbers.That tells you they care.” Students appreciate thiseffort – but it does take a toll on their teachers.

Here, teachers and administrators find a realopportunity to work together for better results fortheir students. That is why principals enjoy theirwork, why students feel they matter, and why teachersnow like working in a place that for too many yearsfelt hopeless.

“Students are

enthusiastic about

their teachers and the

opportunities they

see ahead of them if

they work hard.”

24 The cost was more than $50 million, but the vast majorityof this money was spent renovating the historic auditoriumand making repairs on the exterior stonework and theheating system. Less than $2 million went to updating thebuilding to accommodate the small schools.25 Hermalyn, G., 1995, p. 9026 Hermalyn, G, 1995, pp. 81, 84.

An interior hallway of the renovated school is bright and airy with high ceilings, handsome woodwork, and period lighting.

The exquisite detail and grace of the façade reflects the skill ofits designer C.B.J. Snyder, who was also superintendent ofschool buildings for New York City. Snyder and members ofthe school board wanted Morris High School to be a focus ofthe developing community’s pride in education.

© Th

e Bro

nx C

ou

nty H

istorical So

ciety, New

York C

ity

MacArthur Park has the highestpopulation density in Los Angeles. Fourfamilies may share an apartment intended to serveone. In the shadow of expensive, high-rise buildings,the edges of most streets are littered with detritus,men vivisect cars for parts, and drugs change hands.The corner of Burlington and West Seventh streetswas a center for drugs and crime until Camino NuevoCharter School reclaimed it for the community. Theschool purchased and renovated a failed mini-mall,turning it into an elementary school, and thenrenovated a nearby warehouse into the middle school.

Now an older woman who lives in the small housebetween the two schools has planted a rose garden inher front lawn. Children, supervised by parents, andteachers walk safely from the schools along thesidewalk to the playground. After school, ice creamvendors wait on the street, and people congregatearound the carts to talk and enjoy a treat.

Students use the protected exterior corridor to move betweenclasses, which saved the expense and space of internal hallwaysand is possible in the temperate climate of Los Angeles.


34

Profile Camino Nuevo Charter School: An urban elementary and middle schoolLos Angeles, CA • Burlington Street Elementary • Grades PK-3 • 257 Students

Burlington Street Middle School • Grades 4-8 • 294 Students

“In other schools, when kids don’t listenthey throw spitballs.”

— Camino Nuevo Elementary Students

In 1999, former Episcopal priest Philip Lance startedPueblo Nuevo Development (PND), an outreach

organization. PND’smission is to serveits neighborhood bycreating self-reliant,community-basedorganizations. PNDbegan by creatingjobs for people inthe community, butbranched intoeducation as itrealized thecommunity’schildren wereunder-served inlocal public schools,

and that education and proficiency in English wereessential to getting better jobs. Pueblo Nuevo createdCamino Nuevo Charter Schools, a PK-8 schooldivided into elementary and middle schools, andstarting in 2004, a high school.

Camino Nuevo Charter School traded diversity forserving its immediate neighborhood, but the studentsare not as homogeneous as it first appears. Studentscome from many countries in Mexico, Central andSouth America. This cultural unity serves the schooland students well. Parents, teachers, andadministrators share expectations for behavior and

“Parents sign an

agreement with the

schools that they will

contribute 15 hours

of work a year per

family, meet three

times a year with

teachers, and ensure

their child has a clean

uniform and does his

or her homework.”

Students studying Egyptian history are attentive andknowledgeable, working together purposefully yet withexcitement.


35

care for each other’s children. Many of the school’schildren are related to one another. The school isattentive to local culture; children meet in same-sexadvisories to discuss more easily matters traditionallydivided along gender lines.

The children wear uniforms. They laugh and talkquietly, together and with their teachers, as they lineup in the hall waiting to go to their homerooms, butthey are not subdued. Almost all are Hispanic; somespeak indigenous dialects, so English often is theirthird language, after Spanish. Many come to theschool speaking almost no English, but a carefullyorchestrated bilingual program, with a lot ofsupportive tutoring, assures that almost all studentswill be fluent in English and Spanish by fifth grade.Two trained people monitor student progress and helpteachers understand results so they can help childrenindividually as well as work more effectively withentire classes. As a result, test scores are improving.

Throughout the day, the open courtyard is full ofchildren enjoying recess or eating lunch, and thenlater demonstrating dances they have learned to anenthusiastic audience of teachers and parents. Theprincipal says, “All the moms help out,” adding thateach day about 30 parents come to the school to help.Some parents come every day. Parents sign anagreement with the schools that they will contribute

15 hours of work ayear per family,meet three times ayear with teachers,and ensure theirchild has a cleanuniform and doeshis or herhomework. Parentsand the school takethis agreement

seriously. Parents get a report card on the way inwhich they are fulfilling their obligations under theagreement. Parents who fail to fulfill these obligationsreceive counseling and support from the outreachcoordinator, and, while this has not yet happened, ifparents do not help educate their child, the schoolcould say the school might not be the right fit for thefamily.

Students and parents deeply appreciate the schoolbecause they know children are learning. Manystudents at Camino Nuevo Charter Schoolcommented that they like being in school, they likelearning, and they appreciate the good behaviorexpected by “the Camino Nuevo way.” Theirgratitude is touching. As one boy said, talking aboutthe rectangle of dirt with a few climbing bars and asandbox the students use for recess, “Our playgroundis perfect, and our school is too.”

“Students and

parents deeply

appreciate the

school because

they know children

are learning.”

The middle school at Camino Nuevo Charter Academyoccupies a former warehouse and has helped the communityof MacArthur Park reclaim its neighborhood.

Students gather on the steps in the courtyard of the elementaryschool to celebrate Black History.

Profile

“We’d lose the FALA magic with more kids.”

— Kirk Quitter, FALA Director


36

Flagstaff Arts and Leadership Academy(FALA) shares 400 acres with its host, theUniversity of Northern Arizona, and a neighboringsenior-living center, The Peaks. FALA was born in1996 out of a gifted art teacher’s frustration with thedecline of the arts programs in Arizona schools.FALA has weathered some difficult times, and it isnow poised to realize the opportunities of its locationand the potential for partnerships with its neighbors.

The school buildings are ordinary portable unitsleased from a bank. But students, parents, and staffmembers have transformed them into vibrantshowcases for a thriving academic culture rooted in

the arts. Studentshave paintedclassroom doorswith designs thatannounce thesubjects beingcovered inside, andstudent art enrichesthe walls ofclassrooms andoffices. Minor

graffiti in the bathrooms led the student council tosuggest that students cover the stalls with intricateand colorful laminated collages, which are easy toclean and fun to look at. Using federal EmergencyRepair and Renovation Grant funds, the schoolbought a new modular classroom, and, as part of their30 hours a year of community service, studentsassisted by parents and staff built a wide deck thatconnects all of the units and provides a place forsocializing, eating, and studying. Working with agraduate student, FALA students renovated adonated greenhouse and built gardens and an

amphitheater of straw bales that is remarkably simpleand cost-effective, but which offers a welcoming spacefor small audiences and performers.

Not “just” an art school, FALA prepares students forcollege with rigorous courses, many of them project-based and multi-age. In its Young Playwright Series,ninth graders write a play; Theater 1 students, most ofwhom are in tenth grade, perform it; and Theater 2students, who are older, direct and produce it. Thisyear’s play, Romeo and Juliet, was set in a half-eatensandwich and featured a forbidden romance betweenbacteria and fungi. The play was a smash success.Another success: FALA students’ scores on state andnational tests, which are consistently high.

A “do it yourself” spirit helps make this schoolpossible. There is limited busing, no sports or cafeteria,and, until the fall of 2004, the school had no custodialservice. The music teacher still waters the gardens,many students and parents drive, the food vendor pullsup at lunchtime, and students clean classrooms and doservice projects such as covering pathways withcrushed volcanic cinder. A parent provides freeplumbing expertise, but as the director says, “Whentoilets clog, I’m out there with a plunger.”

“FALA was born in

1996 out of a gifted

art teacher’s

frustration with

the decline of the

arts programs in

Arizona schools.”

FLAGSTAFF ARTS and LEADERSHIP ACADEMY: A charter high schoolFlagstaff, AZ • Grades 9-12 • 160 Students

When there was a minor problem with graffiti in the bath-rooms, students decided to create laminated collages on thedoors of stalls. The result is delightful – colorful, funny, interesting, easy to clean – and it deters graffiti.


37

“I love this place.” The student council is animpressive group and has real power to affect policyand find solutions to everyday problems. The studentrepresentative to the board of directors says, “Theboard asks me to talk first. They really listen. Theyare proud of us.” And they should be.

Though still developing, the school’s partnershipswith the museumand senior-livingcenter are importantto students and tothe future of all threeinstitutions. Themuseum allows theschool to use itsauditorium, grounds,darkroom, andclassrooms. Studentshave interned at themuseum and workedwith its professionals.In the fall of 2004,the social studiesteacher introduced aclass about theculture, history, andecology of theColorado Plateau.The teacher’s wife,who heads themuseum’seducational outreachprogram, helpedprepare the class.With a grant fromthe Arizona HeritageFoundation, studentsvisited a Hopi community to repair bread ovens andwill build one at the museum. Students also havepainted sets for a museum fundraising event. Studentsvolunteer regularly at The Peaks senior residence,giving concerts, talking with residents, paintingresidents’ portraits, and socializing. As one studentreported, “Their faces brighten up when they see us.”Everyone seems to hope that these partnerships willdeepen, and the school director and the new directorat the museum are working to make this happen.

SCHOOL CODE:

(Defined by students,and displayed in eachclassroom on posterspainted by students.)

Respect individuality

Be responsible foryourself

Listen to everyone’svoice

Respect others’personal property

Be mindful of gossip

Avoid classism

Have an open mind tolearning

Support each other’spassion

Have respect forteachers

There is a culture of ease and acceptance at FALA; inone classroom, a student lay sprawled on the floorbetween the tables. “Man down,” another joked asvisitors walked in, but no one minded, because all ofthe students are engaged in exciting intellectualwork. A dog comes to class every day with her master,the algebra teacher, because, as the instructor pointsout: “Algebra is everything a teenager isn’t: logical,sequential, and with one solution. The dog helpshumanize it.” The algebra classroom almost pulseswith energy, and the graphics on the walls pullstudents into a subject that is usually less thanenticing to teens focused on the arts.

Though outward indicators might suggest a lack ofstructure – no dress code, students lounging in thecourtyard between classes – the school’s success stemsfrom a clear set of expectations about academicpurpose and a code of behavior that was developed bythe student council. Four students were expelled inthe middle of the year for smoking marijuana because,

as the director explains, “We just can’t tolerate thatkind of behavior.” Most issues, though, are discussedinformally or at student council and resolved in waysthat are possible only in small schools whereeveryone knows everyone else.

There is a warm feeling of mutual respect andpartnership between students and the adults atFALA. This is evident in the way they talkcomfortably with each other, and when students say,

Bales of straw arranged in a semi-circle create a smallamphitheater for readings, rehearsals, and even classes andinformal performances. With the help of a graduate student,students at FALA built a greenhouse nearby that theypainted and gardens that welcome visitors from The Peaks.

Profile High Tech High: An urban high school San Diego, CA • Grades 9-12 • 368 Students (2002-2003) • 454 Students (2004-2005)

“Our accountability is to the students.We owe them that. They aren’t going tolet us get away with anything.”

— High Tech High Teacher


38

High Tech High occupies what was onceBuilding 49 on a site used by the navy. For severalyears, the school was the sole occupant of a 480-acre,mixed-use development project that was designed toinclude an education core, retail space, culturalcenter, office buildings, two hotels, and 470 units of

predominantlyhigh-end housing.Now that familiesand businesses arefinally moving in,the school is wellsituated to attractstudents. Becausethe school’sfounders had theforesight to assumeits lease early inthe project’sdevelopment, theywere able to findspace cheaply andcould afford toremodel to suit the

school’s needs. Eventually, they were able to establishtwo new schools (High Tech Middle and High TechHigh International) in facilities adjacent to Building49. From this base, the project has grown into anetwork of schools around the nation and in othercountries – an impressive accomplishment.

High Tech High offers a project-based curriculumthat emphasizes technology, although not to theexclusion of other fields. Ninth-grade studentsgrouped into clusters and teams have an advisor andtake their classes together. They occupy four

classrooms, two on each side of a corridor called the“fishbowl” because of the glass cubicles that serve asoffices for the advisors. These spaces give the adults aprivate place to talk and plan, while making it easyfor them to supervise their students. Older studentsattend core classes together, but they can takeelectives offered to 10th- through 12th-graders andwork on individual and group projects with mentorsin businesses and other private organizations.Students appreciate the independence andresponsibility; they thrive by working on projects thatinterest them with knowledgeable professionalswilling to share their expertise. Finding, monitoring,and maintaining relationships with such people, many

“Older students

attend core classes

together, but they

can take electives

offered to 10th-

through 12th-graders

and work on

individual and group

projects with mentors

in businesses and

other private

organizations.”

Extraordinary art by students and working artists hangsthroughout the school, turning walls into a gallery. BecauseHigh Tech High hallways contain no lockers, there is plenty ofwall space available for exhibitions.


39

of them from the business community, is a full-timejob for a coordinator with lots of energy andexperience in such work.

But it is the facility that is most remarkable. Therenovation left the basic structure intact and tookadvantage of such details as ribbon windows that lacethe upper reaches of the walls and saw-tooth skylightsthat run much of the length of the building,providing light and a sense of space. High Tech Highwanted the school to feel like a professional workspace, and it does, with the added feature ofexceptional art by students that enlivens the space.The reception area opens to a “great room” wherestudents display their work and make presentations.There are some traditional classrooms, but also“project rooms” where different sorts of work can takeplace. In one room a team is building a submarine,

which the students will later test. All walls above 53inches are made of glass, so the school seems very

open, but noise isnot the factor itcan be in a totallyopen school plan.The thoughtfullyrenovated facilityprovides an idealsetting for thisproject-based smallschool. It offersspaces that supportthe program, andembodies themission and goals

in its design, creating a professional but open andexciting place in which to teach and learn.

“High Tech High

wanted the school to

feel like a

professional work

space, and it does,

with the added

feature of exceptional

art by students that

enlivens the space.”

Students work in small groups on projects – this one involvingmodel cars.

Biology students check out an experiment in the High TechHigh Biotechnology lab.

The Great Room with soaring ceilings provides a dramatic space for presentation and displaying completed projects. Classrooms linethe perimeter and advisories use the spaces between the central area and the classrooms.


40

Profile INTERDISTRICT DOWNTOWN SCHOOL: An urban K-12 schoolMinneapolis, MN • Grades K-12 • 500 Students

“How can you put a price tag onknowing a kid for three to four years?”

— IDDS Teacher

It doesn’t make sense to locate a schooldowntown because no one lives there, right? Wrong.As Interdistrict Downtown School (IDDS)demonstrates, it makes very good sense. This K-12school, which opened in September 1999, drawsstudents from 10 districts around Minneapolis. Theschool was originally conceived to bring togetherchildren from different races and cultures. Foundersalso realized that downtown Minneapolis offersresources unavailable in outlying areas: libraries,theaters, a university, public transportation, and a

wealth ofemployees at localbusinesses andorganizations.

IDDS was designedto fit into itsneighborhood, thetheater district, andmake maximum useof its surroundings.

The five-story, red-brick, 102,500-square-foot school,located on one acre, reflects the tone and scale ofnearby buildings. Architects searched for cost-savingopportunities – including the idea of setting theschool on the foundation of a city-owned parkinggarage, thus eliminating the need for a newfoundation, and using space in neighboring facilities.The MacPhail Center for the Arts, the Orpheum andState theaters, Orchestra Hall, the MinneapolisPublic Library, the University of St. Thomas Schoolof Education, the YMCA and the YWCA all provideservices and space to the school, and the school offersits own space in return when possible. The Y’s hold

athletic and after-school programs because IDDS hasno gym, and the Y’s use the school in the lateafternoons and evenings for computer training.

The air is remarkably clean at IDDS, thanks to thecareful selection of building materials and the use ofefficient heating, ventilating, and air-conditioningsystems. For example, recycled glass tile, sealedconcrete floors, and linoleum are easy to cleanwithout toxic chemicals. Using paint with little or novolatile organic compounds, avoiding solvent-basedadhesives, and selecting emission-free particleboardresult in a much lower level of toxicity than iscommon in many schools.

Using concrete instead of carpeting for floors reducesthe expense of cleaning and maintenance, although itdoes increase noise. The school has had to employsome simple sound-abatement procedures and hascarpeted high-traffic areas to reduce noise. The

“IDDS was designed

to fit into its

neighborhood, the

theater district, and

make maximum use

of its surroundings. ”

The roof offers a place to play and enjoy the view of downtown Minneapolis.

Volunteers at IDDS

Volunteers gave more than 5,450 hours to IDDS in2003-2004. Valuing their time at the amountMATCH pays its work-study students means IDDSvolunteers contributed at least $73,575 to the schoolthat year. But IDDS students benefit from thesevolunteer relationships far more than dollars candemonstrate. Below is a sample of how localorganizations support IDDS with volunteers.

Target provides approximately 160 K-3 readingbuddies for 24 half-hour one-on-one sessionsthroughout the year.

Wells Fargo supplies approximately 20 math tutors for12 one-hour small group sessions with eighth graders.

American Express assists second through fifth graderswith 160 pen pals and with approximately 30 mathtutors for 30-minute one-on-one sessions over a 20-week period.

Various organizations provide approximately 50 “e-mentors,” each of whom spends approximately20 hours per year working with sixth- and ninth-gradestudents.

Public Achievement supplies second and third gradewith approximately 15 coaches for 12 one-hoursessions.

The University of St. Thomas provides 10 tutors whospend 25 hours each with various grade levels and indifferent subject areas.

Twenty-five Big Brothers and Big Sisters serve ascommunity mentors for grades K-6, volunteering onehour for 30 weeks.

facility incorporates such energy-saving technology as solar panels and a perforated metal black trombewall,27 which heats air against the building before it is drawn in through a heat-air exchange system.Computerized monitoring shows that these techniquessave money over the long term, though initialinvestment may be higher than traditional systems.

IDDS’s location allows the school to developpartnerships with an astounding number of localbusinesses and organizations. The walkwayconnection to the University of St. Thomas School ofEducation allowed the architect to create a teachers’lounge shared by teachers at IDDS and the university.The same system of walkways connects to localcorporations, which encourages employees tovolunteer at the schools. IDDS Director Lee Fertig, apassionate and articulate spokesman for multiculturaleducation and the energetic coordinator of the

volunteer program, helped recruit more than fivehundred volunteers for the school. Wells Fargo andTarget employees, for example, walk to the schoolthrough the enclosed walkway to tutor students. Somevolunteers work with teachers and classes, including aretired accountant who assists the math classes for anhour each day. Other mentors correspond regularlywith their students through e-mail.


41

27 For more information about trombe walls see:www.allanstime.com/SolarHome/Trombe_Wall/.

For more information about sustainable design see:www.develop.csbr.umn.edu/msdg2/

Students work on an art project that also teaches skills inmanual arts.


42

Profile LAUREL-CONCORD PUBLIC SCHOOL: A rural elementary through high schoolLaurel, NE • Grades PK-12 • 193 Students – elementary • 195 Students – high school

“The school and main street connectionare vital. If a rural school andcommunity are going to survive, it willbe by surviving together.”

— Superintendent Dan Hoesing

computer resource handle much of the administrativework. For example, teachers use PowerSchool to entergrades. After that, students, parents, and schooladvisors need only a password and an Internetconnection to gain access to the information.

The school takes academic achievement seriously. If astudent is doing poorly, he or she cannot participatein sports or other extracurricular activities. Parents,advisors, and coaches can use PowerSchool to checkgrades and determine whether a student has turned inassignments on time. PowerSchool not only allowsadvisors and coaches to pull students off a team ifthey are behind in their academic work, it also allowsthem to reinstate those students as soon as theirteachers have entered passing grades. This immediacyof response is a strong incentive for students to gettheir work done. Parents like PowerSchool becausethey can check their child’s progress, and advisorsappreciate that parents come to conferences wellinformed about their children’s lives at school.

PowerSchool requires that parents, teachers, andstudents know how to use and have access to acomputer and the Internet. The school does anexcellent job of training teachers and students, andadvisors work with parents who are unfamiliar withcomputers. As noted earlier, the school leases donated,

Cornfields surround Laurel-ConcordPublic School in rural Nebraska. Many of its 388students in grades K-12 have been at the school sincekindergarten, and many have relatives who attendedthe school as well as others who are current students.The facility, anchored by a 19th century traditionalbrick building, is a stylistic hodgepodge reflecting the

eras in which it wasremodeled.Although theexterior looksdated, the school isanything but old-fashioned. Theschool meldstechnology andtraditional values

with foresight, enthusiasm, and realism.

The board and its superintendent, Dan Hoesing, areguided by several goals:

1) Infuse the curriculum with the most advancedtechnology available.

2) Use technology to simplify the day-to-day tasks ofrunning the school.

3) Improve and maintain the physical plant.

4) Retain extracurricular activities that other schoolsare eliminating.

5) Share the costs and benefits with the community.

One of the first decisions the board made was to havethe superintendent learn PowerSchool, a web-basedstudent information system, and then let this

The new Community Learning Center was made possible bysharing costs between the school city, and community.Residents use the library, computer center, and fitness centerduring and after school, which has helped make the school acenter of its community, and perhaps not coincidentally hasmade residents more supportive of school funding initiatives.

“The school melds

technology and

traditional values

with foresight,

enthusiasm, and

realism.”


43

student-reconditioned computers to parents who lackcomputer access. Because the school provides space tothe local telephone company, that utility offers freeservice and Internet hook-up to the school, students,and their parents. PowerSchool and the school’scapabilities for “distance learning”28 give students achance to study a wide variety of subjects. The schoolsets aside a room, and teachers are available to helpstudents in distance-learning courses.

Laurel-Concord is part of a 26-school consortium thatshares online resources and training. The school is awireless facility, so students and residents easily useboth desktop and laptop computers. PowerSchool hashelped Hoesing run his school so efficiently that hehas accepted the superintendency of a neighboringdistrict, which means that two districts now share hissalary – another cost-effective strategy.

As originally intended, the school and the communityshare costs and the school facility. The communityand the school shared the cost of the most recentfacilities projects: a library, computer center, andfitness center. Some “extras,” such as a microfilmprinter, were affordable only because people in boththe school and community would use them.

This reciprocal arrangement is not without itschallenges. Because the community and school sharethe library, some people were concerned about studentshaving access to inappropriate reading material. Thefamily reading center is located next to the entrancefor easy access by members of the community. The area

includes play equipment for toddlers, and families useit regularly. The library also offers a computer centerthat is open to students and people from thecommunity. The reading areas and books for studentsare located in the large open room furthest from thefamily section. During the school day, students maynot use the adult family area, but after school they cango anywhere in the library. Everyone seems to acceptand respect this arrangement.

The fitness center, built by erecting a wall to connectexisting wings, is equipped with several types ofexercise machines. It is open to students andmembers of the community throughout the day, butfor $10 per month per family local residents can use itwhen the school is closed. In 2004, 160 families hadmemberships, and each family gained access to thefitness center by using a coded personal credit card.

The board and Hoesing have developed a 10-yearcycle for building maintenance. Planning helpsreduce project costs, and it also helps preventemergency repairs, which are always the mostexpensive way to deal with facilities problems.

Students and their parents appreciate that theirschool is well run, that students achieve academically,and that they can participate in a wide variety ofextracurricular activities that many financiallystrapped schools have had to cancel. The communityappreciates having access to a fine library, computercenter, and superb fitness center – assets that oftenare unavailable in rural areas.

28 Distance learning, which is more common on the collegelevel, is a method secondary schools can use to increasecourse offerings. Several terms describe essentially the samething: “distance learning,” “distance education,” “distributedlearning,” “remote education,” “online learning,” and “openlearning.” Three common defining elements of distancelearning are learning that takes place with:

1. The instructor and student geographically remote fromeach other.

2. The use of educational media including television, radio,CDs, or Internet technologies.

3.The provision of two-way communication betweenteacher and learner. Instructional interaction can rely onusing the mail, telephone, desktop video and computerconferencing technologies, or the Internet.

www.distancelearn.about.com/cs/glossary/l/aa120202a.htm

Partnerships between the school and community made itpossible to enclose the space between two wings of the schooland create this fully equipped fitness center, which is open tostudents and residents.

Profile MEDIA AND TECHNOLOGY CHARTER HIGH SCHOOL: An urban high schoolBoston, MA • Grades 9-12 • 185 Students

“Whisper my name, and I’ll find you.”

— Alan Safran, Executive Director


44

of the students have repeated a year since coming toMATCH. Their willingness to stay in school and repeata year, rather than drop out or go to a less demandingschool, demonstrates their determination to learn andtheir conviction that at MATCH, they do learn.

MATCH relies on tutors to give students the extra timeand support they need. In 2000-2001, MATCH loggedfour thousand volunteer hours, including tutors’ andmentors’ time. In the next year, that number rose to sixthousand hours, and in 2002-2003 the school benefitedfrom more than ten thousand volunteer hours. Intensivetutoring provides many benefits beyond its impact onstudent learning. Students and teachers have time toknow each other, and adults are accessible to studentsand their parents. Teachers frequently call parents athome to tell them how their children are doing.

Joking that at one time they thought they shouldrename the school “Mentoring and Tutoring CharterHigh,” the founders can now look on proudly asMATCH students rank first in the state on theMassachusetts Comprehensive Assessment System(MCAS) among schools with a predominantlyminority population. In addition, MATCH studentsoutperformed students from the other 22 schools inBoston that don’t require an examination for entrance,and 89% of the class of 2005 passed both sections ofthe test on their first try. As the school website states

The Media and Technology Charter HighSchool (MATCH) grew out of school founderMichael Goldstein’s master’s thesis. Goldsteinassembled a team that included educators, businesspeople, community leaders, and media professionals.Their goal? To create an inner-city school with a smallenvironment where student projects would be closelytied into “the basics.” Goldstein recruited codirectors,including Charlie Sposato, an experienced principal,and Executive Director Alan Safran, who had served inthe Massachusetts Department of Education. Together

they created a schoolthat serves apredominantlyAfrican-Americanand Hispanicpopulation fromsome of the poorestneighborhoods inBoston. Ninety-five

percent of MATCH’s students qualify for free andreduced-cost lunch.

The team was determined to help all MATCH studentsachieve at high levels. When MATCH’s first ninth-grade students had sixth-grade reading levels and fifth-grade math scores, the school took advantage of itsagility and autonomy as a small school to adjust itsprogram. To achieve the team’s high standards,Goldstein, Safran, and Sposato added one-on-onetutors, additional study time, and set clear expectationsand boundaries.

MATCH provides structure to students whose lives areoften chaotic outside of school. Students must wearuniforms, follow firm rules, and be prompt. If a studentis more than two minutes late to school, he or she mustserve detention, even if the cause is a delayed subwaytrain. MATCH expects students to behave well, toparticipate in class, and to work hard. Almost a quarter

“MATCH provides

structure to students

whose lives are

often chaotic outside

of school. ”

The renovated car dealership located on Boston’s public transportation service, “The T,” gives easy access to studentsand tutors from area colleges.


45

“…despite the fact that just 4% of our students hadscored in the top two MCAS levels (proficient andadvanced) in math before coming to MATCH, after 2years here, 73% have scored proficient or advanced.”29

MATCH does nothing without careful analysis – andits research showed that the school should be locatedalong easily traveled public transportation to giveeasy access to students from throughout the area, aswell as volunteers and paid tutors. Using funds fromthe federal Qualified Zone Academy Bond (QZAB)program,30 MATCH was able to buy a 90-year-old,three-story brick building with 10,000 square feet perfloor located directly on a subway line. The site is onthe fringe of Boston University and close to BostonCollege, where students are just a short train rideaway from the Boston Museum of Fine Arts, theBoston Symphony, the Boston Public Library, andother cultural, historic, and medical assets. They alsoare within easy access to businesses and privateorganizations. In short, the school’s location offers a

wealth of opportunities for internships, mentoring,field trips, and projects.

Careful and frugal renovation created a two-storyschool that includes a spacious front hall with ahandsome grand staircase, a reception room, limitedand shared office space, and a large room (once aluxury car showroom) that serves as a study hall,assembly room, dining hall, and a gathering place forstudents. A conference room directly off the largeroom (to save space, there is no hall separating the

two) offers a handsomely furnished area for meetings.Because dot-coms were folding (and liquidating theirassets) while MATCH was under construction, theschool’s classrooms and meeting rooms were well andcost-effectively furnished. MATCH recentlyrenovated space on the third floor as dorms forAmeriCorps volunteers.

The school’s small size provides flexibility andattention to students. Being small allows the schoolto function with less formal structure than isnecessary in a large school. And, as founderGoldstein points out, “If you have good people, don’timpose structure.” The good people will help developa structure that is responsive.

The MATCH students are doing well academically, asdocumented by their test scores, but perhaps the bestindicator of MATCH’s success was observed by one ofthe study’s researchers while she was going toMATCH on the subway. Instead of a conversationthat might have been sprinkled with expletives andreferences to boyfriends and pop stars, five teenagegirls talked about how they did on the last test, howthey could help each other understand an assignment,and where they were applying to college. When theyall got off the train and walked in to MATCH, theresearcher knew she was visiting a remarkable school.

29 See www.matchschool.org. 30 For more information on Qualified Zone Academy Bond(QZAB), see: www.qzab.org.

Victo

ria Berg

sagel fo

r Arch

itects of A

chievem

ent

Tutors work one-on-one with students for ten hours a weekeither after school or on Saturdays. This intensive support hashelped MATCH students pass the MassachusettsComprehensive Assessment in record numbers.

Elegant columns grace the large hall, once the showroom for the auto dealership, and now a multi-purpose room wherestudents study, work with their tutors, eat lunch, and gatherfor presentations.


46

Profile The Metropolitan Regional and Technical Center: An urban high schoolProvidence, RI • Grades 9-12 • 438 Students

“One student at a time.”

— School Motto

The four buildings of The Met’s PublicStreet campus were designed to blend into theneighborhood. Each of the four schools (Liberty,Unity, Justice, and Equality – named by theneighboring community) occupies a corner of therectangular site and echoes the architecture of thesurrounding two- and three-story apartment houses.Paved streets crossing the campus are visualextensions of the city streets that border the school,but automobile access to the school is limited. Only alow wooden rail, easily jumped by a lanky teenager,separates the schools from the neighborhood.

How do you create small schools within a largerorganization and give them the autonomy to developtheir own character? Elliot Washor, cofounder of TheMet and cohead of its umbrella organization, The BigPicture Company, believes such schools must be setapart from each other – by as much as two hundredyards. This, he suggests, is close enough to shareamenities yet far enough away to encourage eachschool to develop its own identity. Students also shouldhave nooks in their own facility in which individuals

and very small groups can gather to talk or study.Without these physical spaces, the relationships that arethe bedrock on which a good small school is foundedcannot develop.

The Met and The Big Picture Company’s philosophy ofproject-basedlearning is groundedin the founders’experience invocationaleducation. Theybelieve that hands-on work chosen bythe studentsprovides the best

way for them to assume responsibility for their owneducation. The Big Picture Company believes thatstudents quickly become alienated from a course ofstudy in which they have no interest. They’reconvinced that building on students’ interests is away to involve them in the wide range of study theyneed to be effective students, pass state tests, and getinto college.

The Met serves a diverse student population fromProvidence and the nearby communities. Although theschool is located in one of the poorest neighborhoodsof the city, its students make the journey because theyare alienated and fed up with the traditional system. As

“Students work in

groups of 14 with an

advisor who stays

with them

throughout their

time at The Met.”

Exterior shot of one of the four small schools at the PublicStreet campus.

The principal of a small school at the Public Street campus ofThe Met addresses her students and their advisors at themorning “Pick-Me-Up,” a school-wide meeting.


47

one student put it, “When I went to other schools, Ihardly ever went. Like I missed 80 days. Last year, atThe Met, I missed five days.”

What’s different for this student? The four schools onthe Public Street campus and the two others thattogether form The Met are limited to 120 studentseach. Students work in groups of 14 with an advisorwho stays with them throughout their time at TheMet. This relationship allows time for real friendshipsand trust to flourish. This is critical for teenagers,many of whom have been alienated from adults. Asone student explains, “I think if you don’t haveteachers who actually care about you – and actuallycare about what you are doing, then why should youcare? My advisor really cares. I was on the phone withhim for an hour and a half yesterday, and we weretalking about different projects and things like that.And he cares, and that makes me want to care aboutmy work because I have somebody who cares andwho’s willing to help me. And I can trust him. I cantell anything to my advisor. He’s the person I run towhen I have a problem. That’s why I love myadvisor.” She says this with such feeling that tearscome to her eyes.

Students take some classes with their advisors andpeers, as well as with specialists at The Met.Students design projects that focus on internshipsbased in their areas of interest. One girl works at anorganic farm, while a boy studies electronics with anelectrician who has mentored many Met studentsover the past several years. With the guidance oftheir advisors and mentors, students work their way

through levels ofresponsibility andaccomplishment.Every quarterstudents developand carry outunique projects,each of whichculminates in apresentation orexhibition. Somemight wonder how

it is possible to have students select their ownprojects and still fulfill academic requirements, but

students at The Met report that they are workingharder than they ever did in previous schools and areaccomplishing much more. They add that theirfriends are astounded when they see the presentationsand portfolios Met students are required to makebefore passing to the next level of work.

At The Met, students pursue their interests with thesupport of advisors and are led back to core subjectssuch as history, math, reading, and writing in theirattempts to better understand their discipline. Astudent studying electronics was eager to learn thehistory of the field and the science of how electricityworks. Eventually this extended to the history of the

industrial revolution, as well as advanced math. Ofcourse, this method takes a well-developed system ofsupport to succeed: students, advisors, a head ofadvising, a weekly newsletter, and continuous trainingand meetings run by The Big Picture Company. Butthe results are exciting. Students who previously haddropped out, avoided school, or been passive abouttheir lives are now engaged in learning. A very smallpercentage fail to handle the freedom andresponsibility The Met offers, but for the 97% whosucceed and graduate, life will never be the same. Nowthey see themselves as lawyers, surgeons, teachers ofspecial-needs students, electricians, and professors.

“Students who

previously had

dropped out,

avoided school, or

been passive about

their lives are now

engaged in learning.”

A student works at her own station within the advisory.Advisors and advisees decide how to use the space, and maychange it throughout the year. Many of the walls betweenadvisories and other rooms can also be moved.

Profile OAK VALLEY SCHOOL: A rural elementary schoolOak Valley, NE • Grades K-8 • 30 Students (2004-2005)

“Here we really care about our school.Our parents care. At public school it’sembarrassing to have your parentsaround. Here it’s great. Parents really help. We’re all equal. We’re afamily here.”

— Oak Valley Students


48

Oak Valley School, founded in 1868, hasoccupied a simple building tucked between cornfieldsat the end of a long dirt driveway since the originalbuilding burned down in 1920. In the 1970s, parentsadded 300 square feet for bathrooms and lockers, and,in the summer of 2001 “the dads” added a 700-

square-foot spacefor a secondclassroom and acomputer room at acost of $50 persquare foot. Theschool groundsinclude a memorialgarden given by theparents’ associationin memory of astudent, aplayground with ametal merry-go-round that hasbeen in use sincethe 1950s and ajungle gym. When

Mrs. Jean Nabity, the lead teacher, rings her bell andchildren line up in front of the door, the scene iscomplete – a typical rural school.

What is atypical about this school, however, is that itattracts two-thirds of its students from outside thedistrict. Parents who live in the university city ofLincoln bring their children here, as do farmers whohave home-schooled their children. Althoughhomogeneous racially, the student body is diverse

“Students also have

the chance to learn

from and with their

peers in multi-aged

classes and to work

with three full-time

teachers who are

resourceful,

engaging, and

determined to offer

their students a fine

education.”

The small white school house is located at the end of a longdirt road in the middle of cornfields.

economically. Why do they bring their children to thissmall rural school? Because this school offers individualattention and a family atmosphere. Students also havethe chance to learn from and with their peers in multi-aged classes and to work with three full-time teacherswho are resourceful, engaging, and determined to offertheir students a fine education.

“Every day when we are planning we ask ourselves,‘What can we do to make this fun?’” says Jean Nabity.Her students’ smiles and the excitement buzzingaround the rooms suggest, “a lot.” Younger childrenlaugh as their teacher squirts strawberry-scentedshaving cream on their desks as they prepare for aspelling bee. The students write the words in the

Younger and older students read to each other each day overseveral weeks. By the end of the year each child will have readwith everyone in the other age group. Teachers monitor eachstudent’s progress and note that reading to a younger or olderstudent is very helpful to both.


49

cream, wiping the cream with their hands to preparea clean “slate” after the teacher has checked eachword. No one misbehaves, and everyone has a goodtime. Meanwhile, older students in the next roomwork in pairs, sticking the names of bones on theappropriate body part of their partner. Later, olderand younger students gather, some sitting at desks,others on couches or easy chairs, to read to eachother. And after reading, older students gather formath, working complex problems on their desks witherasable markers, and then work in the new computerroom. The spirit of play infuses everything they do.

The principal is a practicing lawyer and part-timeschool administrator with lots of experience. He andthe three-person school board have created anonprofit organization to supplement basic publicfunds for the school, but the school has neversolicited or received any money from outside

organizations and does not see itself as experimentalor even particularly innovative.

The school provides a solid and imaginativecurriculum at a cost that is about $600 below theaverage per-pupil cost in the area. To keep costsdown, students clean the rooms, parents do most ofthe maintenance, and the school provides notransportation.Instead, parents drivetheir children to andfrom school and onfield trips – as theydid on a recent visitto the home ofauthor Willa Cather,after the olderchildren had readsome of her books.Teachers order onlythe supplies theyneed, avoiding acentralized orderingsystem because they consider it wasteful. The schoolhas a furnace that can be fired either by oil or by gas,so it can take advantage of whichever fuel is cheaper.And they employ such small innovations as puttingtennis balls on the feet of all chairs so they don’tscrape the floor and make it harder to clean.

Oak Valley School proves that it is possible to have avery good and very small school that offers studentsaccess to an excellent education grounded in familyand community. Teachers have autonomy over whatthey teach and how they teach it, and they offerparents and members of the community ways toparticipate in the education of their children – all at a cost below that of educating children in muchlarger schools.

“Oak Valley School

proves that it is

possible to have a

very good and very

small school that

offers students access

to an excellent

education grounded

in family and

community.”

Before school starts a parent volunteers to lead a popular chessclub in which students of all ages participate.

Spelling is fun when students write words in shaving creamwith their fingers and erase them with their hands after theteacher checks their work.

Profile THE STAR SCHOOL: A solar-powered rural elementary schoolFlagstaff, AZ • Grades K-8 • 65 Students

“Service to all relations.”

— STAR motto


50

The school nestles against a low hill atthe end of a dirt road off a rural highwayoutside the city of Flagstaff. Arrayed behind invitingplaygrounds, adobe buildings form a rough circle. Atthe center of that circle, a bright blue wing of canvasarcs over a wooden deck called “the shady.” Thecloudless sky stretches behind the dormant volcanicrange, and the greens and grays of desert vegetationsoften the dusty land. Purple iris unfurl under rainspouts, and small, rock-encircled gardens dot the landaround the buildings. There is such harmony andsuch a sense of permanence here that it is surprisingthe school opened in August 2001, the buildings are

modular,everythingruns onsolar power,and the areais goingthrough itsworstdrought in1,200 years.

The STARSchool isthe productof a vision

shared by Dr. Mark Sorensen and his wife Kate,developed from working with the Navajo Nation andusing only solar power for more than 12 years at theirranch. To start STAR School, the Sorensensmortgaged their ranch, and bought 40 acrespreviously used for dumping cars and debris. The firstcontribution the school made to its community wasrestoring this land and making it a safe place.

The name STAR, an acronym for “service to allrelations,” suggests the sun and solar power, as well asthe goal of being exemplary. The school’s visioncenters on the Navajo ideal of finding and sustainingharmony with all people and with the land –everything that lives on, beneath, or above it. Some

of the academicprograms at theschool reflectNavajo tradition,such as its strategyfor conflictresolution, thenative sciencecurriculum, andclasses in theNavajo language.Students also learn

reading, writing, and quantitative and critical skills.

The STAR School opened with 23 students in grades1-6. In the fall of 2004, the school served 65 childrenin grades K-8, about 85% of them Navajo, most ofwhom live on the nearby reservation. It has added agrade each year since it opened but plans never toexceed one hundred students. Sorensen points outthat while most charters want to increase enrollment,The STAR School intends to stay small to retain theclose connections that nurture children and families.

The main buildings are modules covered withan adobe-like stucco that make them fit wellwith the landscape and culture of the area.

Modeled after traditional Navajo structures, the Shady is araised platform with picnic tables protected from the sun by ablue canvas canopy. It was built over water storage tanks.

“The first

contribution the

school made to its

community was

restoring this land

and making it a safe

place.”


51

Children living on the reservation have to travelmore than 45 minutes to reach The STAR School.Nevertheless, students choose The STAR Schoolbecause they feel safe and know that they arecherished. The curriculum is rooted in Navajo

culture, but theschool valuesdiversity. TheSorensens believethat children learnto appreciate eachother by playingand workingtogether, so theyhope to welcomemore children fromvariousbackgrounds.

While many larger schools cut athletic programsbecause they are expensive, The STAR School offerswrestling for boys and girls, volleyball, soccer, cross-country, and basketball. Of the 55 children in the K-7 program in 2002-2003, 20 ran cross-country, 20 played volleyball, and more than half wrestled.

The STAR School came in second in its league ofnine schools in soccer and second out of nearly 60schools in cross-country. The climate makesbasketball and volleyball possible on an inexpensiveoutdoor court. The area is ideal for cross-country, andwrestling requires only a small space and little

equipment. A gifted coach, who is a highly respectedmember of the Navajo Nation, makes The STARSchool’s athletic program possible and serves as theliaison between the school and the community.

Many of The STAR School’s challenges are specificto its location and culture. Absenteeism is highbecause the Navajo expect children to be homeduring times of celebration or crisis. Sometimeschildren are out of school for weeks, but The STARSchool is working with students and their families tokeep them connected to the school through theInternet and with home visits by thecoach/community liaison.

The extent to which everyone and everything workstogether at The STAR School is extraordinary.Teachers and parents transformed the modularbuildings with adobe-colored stucco that makes itseem as if they grew out of the desert. Children, usingthe agricultural techniques of their ancestors, aregardening in the desert with rocks to retain water. Inthe summer of 2004, parents helped build a straw balehouse that serves as a science building and mini-gym.Parents work part time at the school as bus drivers,cooks, and assistant teachers. The coach, outreachcoordinator, and facilities director are all parents.Together they have transformed a dumpsite into acenter of community.

“The school’s vision

centers on the Navajo

ideal of finding and

sustaining harmony

with all people and

with the land –

everything that lives

on, beneath, or

above it.”

Recycled tires were used to create a raised area in the playground that can also be used as an informal amphitheater.

Three boys study together in a quiet corner of the classroomfor older students.

Profile TACOMA SCHOOL OF THE ARTS: An urban high schoolTacoma, WA • Grades 10-12 • 348 Students

“We share everything.”

— Jon Ketler, TSOTA Director


52

What do Tacoma Art Museum, the Museumof Glass, the Washington State History Museum, theRialto Theater, the Broadway Center for thePerforming Arts, the School of the Arts PerformingArts Center, the buildings at 1950 Pacific Avenue,and the building at 1818 Tacoma Avenue in Tacoma,have in common? They all offer classrooms forstudents attending the School of the Arts in Tacoma(TSOTA).

When Tacoma high school teacher Jon Ketlerthought about the recent renovation of publicbuildings in his city, he saw an opportunity for a newschool, perhaps even a new type of school. He saw aschool dedicated to teaching students who werepassionately focused on the arts. He saw a school thatallowed them to explore that passion with qualifiedteachers and working artists throughout thedowntown area. He envisioned students using the

newly renovated public buildings dedicated to thearts: the museum, library, university, symphony hall,and theaters. Using his knowledge of local and schoolpolitics and real estate, Ketler was able to craft aschool that uses nine different buildings within a one-mile radius. A combination of hard work, ingenuity,luck, and good will allowed him to make his bold ideafor a good small school a reality.

Students use public transportation as well as theirown feet to get to and from classes. They can earn ahalf credit in physical education for walking to andfrom classes for the three years they attend and areallowed 20 minutes between classes. The school offersa three-year program, and accepts students only assophomores, because they have to be mature enoughto take responsibility for getting themselves to classesin different locations in downtown Tacoma.

The school uses a variety of strategies to be cost-effective. Teachers and students built much of thefurniture, including shelves, work tables, and desks.Instead of installing plumbing lines to serve sciencework stations, they use large buckets that they emptyinto a central sink. Ketler, a consummate scavenger,got a lot of the equipment from other public schools

Constructed between 1906 and 1908 the New York & TedBrown building previously housed a department store, musicstore, and dance studio. Now as the Performing Arts Centerit again hosts recitals and classes. The jazz cabaret, ClubSOTA and a 105-seat theater enrich the cultural life of thedowntown for students and residents alike.

Overhead doors are a cost-effective way to open and separatespaces for different classes.


53

that were getting rid of it. For example, he got largepieces of slate previously used for blackboards, whichthe science teacher and students turned into science

tables. TSOTA hascreated a culture ofsharing that fullymaximizes resources.Students andteachers trust eachother; they leavetheir work out in the

open and share books, equipment, and supplies.

TSOTA spends most of its money for faculty on coreteachers, but supplements core faculty with adjuncts whoteach their specialty, including jazz and print-making.These adjunct teachers are delighted to share theirexpertise, yet do not require benefits. Without them, theVisiting Artists program that offers a wide variety ofclasses in studio and performing arts would not bepossible. Ketler and his codirector and faculty continueto build creative learning experiences for TSOTAstudents through partnerships with the community.

TSOTA shares its space. For example, a universitysculpture professor uses a room in the arts building toteach a college-level sculpture class, which is alsoattended by some TSOTA students. TSOTA’sperformance space is available to the community,and, after school, the chorus practice room becomesClub SOTA, where students and the community canattend performances by local artists free of charge.

TSOTA invests ina full-timecommunity-outreachcoordinator who isin charge of findingpeople who mightoffer internshipsand specialprojects, or teach asadjuncts. Shecoordinates appealsfor donations,oversees recruiting,

and directs interactions with the media. Because ofher energy and dedication, and because she has agreat school to talk about, TSOTA is getting a lot ofpositive attention.

“TSOTA has created a

culture of sharing

that fully maximizes

resources. ”

“Using his knowledge

of local and school

politics and real

estate, Ketler was

able to craft a school

that uses nine

different buildings

within a one-mile

radius. ”

Kevin

Veatch

for A

rchitects o

f Ach

ievemen

t

Students at TSOTA use professional quality facilitiesincluding a large and well-equipped theater with a box officewhere they learn the “behind-the-scenes” arts of stagingperformances.

Students practice together before auditioning for a concert.The music rooms are constructed with double sheetrock wall,acoustical ceiling clouds and soundproofing glass panels toisolate and control sound. Musicians can practice withoutbeing heard in the adjacent recording studios, practice rooms,and classrooms.

Profile Todd Beamer High School: Three academies in a new school within a schoolFederal Way, WA • Grades 9-12 • 1,350 Students divided into three academies (2003-2004) •1,550 Students divided into three academies (2004-2005)

“On time and on budget. He’s the mostenlightened facilities manager one couldever hope to meet.”

— Architect Lorne McConachie, Bassetti Architects,about Rod Leland, Facilities Director for Federal Way


54

The district of Federal Way on thedoorstep of Seattle wanted to build a newhigh school, in part to realign the arrangement ofgrades in schools within the district. While districtadministrators knew that small schools made sense forstudents, teachers, parents, and the community, itseemed everyone in these groups wanted a large highschool. The educators tried but failed to convinceparents and community members of the benefits of

small schools. Ateam of planners,led by VictoriaBergsagel, thedirector ofeducational design,and Rod Leland,the facilitiesdirector for thedistrict, decided tohonor the wishesof the public. Butthey did so in sucha fashion that thefacility can

accommodate a number of different ways to organizeacademies, with the hope that the community willgrow to understand and accept small schools.

To design a facility that reconciled the conflictingideas of the administrators and community, planningteams met over three years for rigorous discussion ofdifferent scenarios – from one large institution toeight smaller schools housed in the same facility.They discussed the need for different types of spaces –not only classrooms, but also nooks and niches set offfrom the hallways where students could gather to

socialize or study, yet still see and be seen, importantconsiderations for teenagers and the adults whosupervise them.

In the fall of 2003, Todd Beamer High School openedto serve 1,350 students in three academies. Theschool sits at the edge of its football field like a villageon a green. The varied shapes and sizes of the joinedbuildings echo the architecture of the older houses inthis growing suburb. The field and a high bermseparate the school from a busy highway, muffling thenoise of cars and trucks yet providing easy access.

The facility’s furnishings, even the walls, are movableand can be adjusted to fit many configurations. Manyspaces serve multiple purposes. The “great hall” worksalong the lines of a baronial banquet hall – a placewhere music, theater, dancing, and dining often occursimultaneously. Its small kitchen and service areas,located across the hall from the central kitchen toreduce the odor of cooking, are also easily accessibleby students and guests at events. The great hall servesthe entire student body and functions as the gym,cafeteria, and auditorium. It has a bank ofcomfortable theater seats that, at the touch of abutton, unfold to provide seating for 320 or slide backinto a neat stack on the back wall to conserve floorspace. Tables and chairs are on casters so they can bequickly reconfigured, and the stage is movable. At thetouch of another button, an acoustic curtain on a

Comfortable seating in the library invites students to studyand talk quietly. Note the chairs at the library tables thataccommodate rocking.

Using different material and heightsbreaks the mass of the building intosections that make it seem smallerand well-integrated into its setting.

Kevin

Veatch

for A

rchitects o

f Ach

ievemen

t


55

track about the height of the basketball nets circlesaround the wall above to brighten or dim the sound.The facility is a marvel of technology and thoughtfulplanning for multiple contingencies.

To save money in future reconfigurations, gas andwater lines extend the length of the wings, as well asvertically in the two-story building. Labs locatedalong a spine with utilities serve the presentarrangement of three academies, but could easilyserve a larger number of separate schools. Adumbwaiter allows teachers to move fragile scienceequipment and supplies safely between floors. Wastelines lie deeper than normal so sinks can be addedanywhere in the building yet drain well. Mechanicaland ventilation systems were oversized to allow for

future capacity.Small bathroomslocated throughoutthe building canserve smallacademies andclusters of students.

Even the library,the main office,

and counseling spaces can be reconfigured easily withmovable walls into separate, smaller academies. Theschool’s furnishings are remarkably adaptable.Teachers use cabinets on casters to store their ownsupplies and can open walls between rooms to createlarge shared spaces.

The building is handsome, inviting, and functional.Stained concrete floors wind down hallways in an

“The facility is a

marvel of technology

and thoughtful

planning for multiple

contingencies. ”

interesting pattern of blues, rusts, and greens. Naturaldaylight floods the hallways and classrooms throughlarge windows that draw the landscape into thebuilding. Wide rock stairs visible from the interiorbeckon teachers and students to impromptuclassrooms and performance areas outdoors. But theperson who may appreciate the building most fully isits head custodian, who says, “I love the buildingbecause it is easy to clean.”

All this was accomplished for $26,640,735 – less thanthe $27,930,000 maximum allowable constructioncost for the project and less than almost all other highschools built in the region. Square foot costs:

• $36.47 for the primary building system, includingsteel frame, floors, and exterior and bearing walls

• $52.63 for the secondary building system,including mechanical, electrical, and interiorwalls

• $51.86 for the tertiary building system, includingdoors, cabinetry, interior finishes, and specialties.

The risk in building a large facility, as educators wellknow, is that students and parents will identifythemselves as part of a large high school, not thesmaller academies, and that the large high school willnever grow into autonomous and distinct smallschools. Some danger signs are already apparent, butthe school’s leadership team has made it possible forsmall schools to coexist successfully in this remarkablefacility. The possibilities at Todd Beamer High Schoolare almost limitless, thanks to the foresight of ahandful of administrators and planners.

Although the school easily houses over 1,500 students, the facility does not overwhelm the site. Instead, the attached buildingsreflect the style of older houses and fit in well with the neighborhood.

Kevin

Veatch

for A

rchitects o

f Ach

ievemen

t

Profile H.S. Truman High School: A suburban alternative high schoolFederal Way, WA • Grades 9-12 • Students: • Forum School: 99 • South School: 98

“The greatest impediment to change is fear.”

— Tom Murphy, H.S. Truman High Schoolsuperintendent


56

When the two small schools that make upH.S. Truman High School moved to their newfacility in March 2003, they left behind a 50-year-oldstructure and a dropout rate twice that of other high

schools in thedistrict. The newspace offeredsoaring ceilings,natural lightstreaming throughlarge windows,garage-style doorsthat open wide tothe outdoors, andan expansive “greatroom” for eachschool. The new

program is modeled after The Big Picture Company’sphilosophy of learning through internships.

Pam Morris-Stendall, who had been principal for 12years before the move, served as the catalyst for thischange at H.S. Truman High School. She could nolonger tolerate the high dropout rate and wasdetermined to find a better way to help not only her

students, but also those who had dropped out of otherprograms and needed a real alternative. Together, sheand her team created a new facility and a newprogram.

Ten of her 12 teachers left when Morris-Stendall saidthey were going to change the school from a traditionalto an internship-based program. But she found newteachers – with a deep commitment to this type ofteaching and learning. Students participated actively inplanning for the school and making those plans work.

The architects really listened to students. Whenstudents said that they lived in cramped apartmentsand craved space, air, and light, the architects madethese qualities central to the design. High ceilings arenot always cost-effective to build, but the architectsfound ways to create space on a budget, usingplywood to cover ceilings and omitting window trim,for example. The school feels larger than itsdimensions would suggest and students and theiradvisors have personalized their own spaces to reflecttheir interests and accomplishments.

By forming partnerships, H.S. Truman High Schoolexpanded the space available to its students. Studentscan use the land and athletic equipment at the parkadjoining the school, and soon, they will also use theadjacent facilities of the Boys and Girls Club. A Head

“When students said

that they lived in

cramped apartments

and craved space, air,

and light, the

architects made these

qualities central to

the design.”

The simple exterior of H.S. Truman High School is functionaland attractive. Note the clerestory windows that run thelength of the central building and provide natural daylight forthe interior.

Garage doors open easily for access to an outside spacethat students and advisors enjoy together for discussionsand meetings.


57

Start program on the same site provides a place forTruman students to have internships.

What do kids at H.S. Truman High School like abouttheir school? They love the building, and they lovethe principal and their advisors. They like each otherand the feeling of safety they have created at theschool. One student said, “There are no fights here.Well, maybe one or two, but nothing like my oldschool.” They appreciate that the district thoughtenough of them to build a new facility, and that theprincipal created a program based on internships thatallows them to pursue their own interests and interactpositively with people in the community. They wantto be sure that they and their school make a goodimpression on the community, and, throughinternships, and the support of their advisors, they arelearning to do that.

Moving to the new facility has lifted spirits andscores. In 2004, 100% of the graduating class wasaccepted into college. What started as a school for“throw-away” children is becoming a place where

students feel valued, grow to care about each other,their advisors, and themselves, and are learning morethan they or anyone else thought they could.

In fall of 2004,several valuedteacher/advisors leftto assume moreresponsibility in thedistrict and in otherschools, and Morris-Stendall retired. ButH.S. Truman HighSchool found acompassionate andarticulate newleader. She isdetermined tocontinue to offer acaring community

and challenging academic opportunities to childrenwho might otherwise have lost their way.

Kevin

Veatch

for A

rchitects o

f Ach

ievemen

t

Students, many of whom live in cramped dark apartments, told architects they craved light and space. Working with a limitedbudget and unlimited imagination and ability to innovate, architects delivered a design with clerestory windows and soaring ceilings.

Students in an advisory share a lot of time together and with theiradvisor and develop closerelationships.


58

Profile WORLD OF OPPORTUNITY: An urban community-based, civil rights, socialjustice, educational and job readiness program. Birmingham, AL • Ungraded • 35-65 Students per day

“Other people try to limit you by telling youwhat you can’t do. WOO tells you whatyou can do. It feels like family here.”

— WOO Student

In 2000, 522 students in schools throughoutBirmingham, AL, were told to leave school betweenFebruary 15th, when they were counted for statereimbursement, and April 15th, the date foradministering state achievement tests. The commondenominators: they were all African-American, manyhad just turned 16, and all were asked to leavebecause of “lack of interest.” Many of these“disinterested” students turned up on the doorstep ofthe city’s alternative adult education program run by

Steve Orel. Whenhe protested theirdismissal, the citylocked the doors ofthe adult educationprogram and firedhim.

Orel and BrotherCharles Todel, aSalesian monk whoworked with theCatholic diocese,founded World ofOpportunity(WOO) to help

students who had been pushed out of city schoolsearn their General Educational Development (GED)credentials. Though the school board locked thedoor, it didn’t own the building. WOO was able tolease the windowless 25-by-50-foot cinder blockfacility in one of the poorest neighborhoods inBirmingham from the sympathetic owner who runs amanufacturing plant next door. Later, the schooladded a trailer, and students built a roof to connect itto the main building. Crime, drugs, disillusionment,

and hopelessness crippled the community, but WOOstaff, volunteers, students, and members of thecommunity are changing that. WOO doesn’t chargetuition and receives no state or federal assistance.Instead, it is funded by donations made to its501(c)(3) non-profit corporation and supported bythe work of volunteers.

The neighborhood is rundown, the building drab, butwhen the school opened its doors a world ofopportunity opened as well. People ranging in agefrom young teenagers to grandmothers sit at desks infront of computers or at tables in three small rooms.Students work at their own pace, some with tutors,some individually. Often students help each other.The walls are papered with announcements ofstudents’ achievements: passing the GED or the testto be a certified nurse’s aide, or getting a job. Thoughpoor by some measures, the people at WOO have anextraordinary capacity for giving – to each other andto their community. They share their resources;volunteer as tutors at an elementary school; assist thelocal food pantry; collect mittens, hats, and toys foran “Angel Tree”; and without hesitation collectedsupplies after 9/11 to send to each of four schools nearNew York’s World Trade Center towers.

“WOO doesn’t charge

tuition and receives

no state or federal

assistance. Instead, it

is funded by

donations made to its

501(c)(3) non-profit

corporation and

supported by the

work of volunteers.”

The facility includes a trailer, and behind it, a small buildingleased from a sympathetic neighbor, Miller Wire Works, Inc.But World of Opportunity has created an oasis of hope inBirmingham, Alabama for students in the Adult EducationProgram, and residents.


59

WOO recognizes that its students have manyresponsibilities to balance: care of children and sickand aging relatives; problems with transportation;money for food, rent, and clothing. WOO serves thewhole student: getting a donated car for a womanwho had to move out of the area and couldn’t get to

school, finding funds to pay for housing for a womanwho suffered a stroke and was caring for hergranddaughters while earning her GED, providingclothes for job interviews, obtaining food whenpeople are going hungry, and offering babysitting sostudents can attend classes. Students set their pace of

study, and thosewhose studies areinterrupted canreturn wheneverthey want.

WOO is a safehaven for itsstudents and peoplefrom thecommunity.Located deep in an

African-American ghetto, it has created a culturethat attracts people of all backgrounds. One studentsaid, “I’ve had my bad times. But these people hereare my family. I go home, and I can’t wait to get back.I’ve never been with so many cultures, and everyonegets along. There are no fights.”

WOO has never turned any student away, and despiteenormous odds it is making a difference in the lives ofits students and its community. Since it started,WOO has helped more than 50 students earn theirGED and has more than 25 former students incollege. Most important, it has revived the heart anddreams of a community that was dying. Today WOOstudents look ahead to earning college and graduatedegrees and aspire to being artists, doctors, lawyers,scientists, teachers, and social workers.

“Though poor by

some measures, the

people at WOO have

an extraordinary

capacity for giving –

to each other and to

their community.”

The rooms have no windows, and spaces are cramped, butWOO students, tutors, and teachers work together as a com-munity, sharing their talents and resources.

A teacher, students, and volunteers work together in thecentral room at World of Opportunity. The leased buildinghas no windows, but that doesn’t stop students from studyingindependently and with tutors to pass exams including theGED and CNA.

The director of WOO and two students enjoy a momenttogether as they stand in front of the binders in which studentsrecord their work and journal entries.


60

Suggestions for Further Work

Inevitably, answers to one set of questions reveal wherefurther study is needed. Dollars & Sense II is no exception.The quantitative and qualitative research in this reporthas raised additional issues that demand examination.Among them:

• Very little peer-reviewed research literature exists onschool construction costs and their relation to schoolsize. This fact is particularly surprising given theamount of money being spent on school construction.Why is there so little credible research on schoolfacilities, and how do we fill this information gap?

• One way large schools “save” money is by allocatingmuch less space per student than small schools do.But how does space affect students, and what types ofspace are increased in small schools and decreased inlarge schools?

Another area of further study is the sustainability ofvolunteerism and the role of volunteerism in makingsmall schools cost effective, as well as in creating vitalcommunities.

The particular schools in this study also warrantadditional research. How will they adapt to newconditions and remain cost effective? In short, there ismuch more to learn, and we hope to continue this work.

Conclusion

This report offers the results of research conducted overtwo years and gathered with three distinct methods: sitevisits to schools that met specific criteria by a team withdiverse experience, analysis of the budgets of thoseschools, and analysis of a large database of schoolconstruction projects. The Dollars & Sense team used bothquantitative and qualitative methods to learn about thecultures, educational programs, and costs necessary toconstruct, maintain, and operate good small schools. Yetthe results of these separate analyses all point to the sametruth: Good small schools are being built, maintained, andoperated throughout the United States so efficiently thatthey spend less per student than the average amount spentper student by their districts.

The small schools in this report are models of costeffectiveness as demonstrated by the fact that as a groupthey are operating at almost 17% less than the per-pupilexpenditure for their districts. The strategies they have

created can inspire people in other schools andcommunities. Furthermore, analysis of more than threethousand school facilities construction projects shows thatsmaller schools are not more expensive to build thanlarger, reasonably sized schools. This is exciting anddramatic information because education research over thepast 30 years shows conclusively that small schools benefitstudents, teachers, parents, and communities. The schoolsthat have the best chance to improve students’ academicachievement – good small schools such as the ones in thisreport – are actually affordable.

The good small schools in this report vary in location,grades, and populations served, and in their form ofgovernance, yet they share fundamental characteristics.Each is grounded in relationships – relationships betweenteachers, students, parents, administrators, and membersof the community. Each has enough autonomy to defineits mission and to determine the best ways in which toaccomplish its goals. Each has created a caringenvironment and programs based on high expectations forachievement and conduct. Furthermore, the facilities thathouse these schools support the educational programrather than block or distort it. Each school also is cost-effective; proof of that lies in the fact that as a groupthese schools spend less per student than the amountspent by their districts.31

By their very existence, these schools prove that goodsmall schools are operating cost-effectively. Thejustification for closing and consolidating schools – thateconomies of scale reduce the expense of public educationto taxpayers – is simply untrue.

Moreover, the schools visited for this study are not servingaffluent students; many work predominantly withpopulations that have long been under-served and allinclude such students. It is to their great credit that theyare teaching poor and marginalized students how to read,write, compute, and think critically, and to trust adultsand enjoy school. Many of these schools are also workingclosely with families and other members of thecommunity. By doing so, they have become communitycenters and centers of community.

31 Note: The four small schools at the Met’s Public Street campusexceed this amount but they are not yet, by their own choice, atfull enrollment. MATCH School also exceeds the per-pupilexpenditure, because the school has received grants and dona-tions, while Boston has had to reduce its allocation because ofbudget cuts.


61

It is important to note that this report does not promotereplication. Rather, schools and communities shouldadapt good ideas to meet the particular needs of theirsituation. Rarely can a school lift an idea or practice andgraft it whole even onto an entirely new school. Thechallenge is more subtle and difficult. To succeed, schoolsmust tailor new ideas and practices to fit their uniquesituation.

It has been more than 30 years since any credibleresearcher in education has recommended large schools.The burdens of large schools are now well documented:higher drop-out rates, lower rates of graduation andpostsecondary enrollment, more violence and vandalism,alienated students, and higher costs for security,administration, and transportation than in smallerschools. The cost to society of students who drop out, getin trouble with the law, or become dependent on welfareis enormous, and it correlates closely with the lack of ahigh school education (D&SI, pp. 8-11).

These costs are literally incalculable. Perhaps even moreimportant is the loss to individuals, families, communities,and our nation of what might have been accomplishedand contributed by students if they had not dropped out

of school. We cannot afford large schools because wecannot afford to lose the contributions of well-educatedhigh school graduates. The high percentage of studentswho are disaffected or drop out in large schools is simplyunacceptable and unnecessary. The power of good smallschools is that they engage their students. They let theirstudents dare to dream, show them the steps they musttake to realize their dreams, and help them becomecontributing members of their communities and theirnation.

The research in this report shows that small schools canbe cost-effective and offer the elements of a goodeducation: close relationships with caring adults; programsthat are deep, diverse, and engaging; and environmentsthat motivate students to come to school, work hard, andgraduate. The good news is that what works best forstudents, particularly those from poor families, and fortheir parents, families, teachers, and communities – goodsmall schools – can be cost effective. As a nation, we mustinvest in small schools. The costs of large schools aresimply too great.


62

The Dollars & Sense Team

Barbara Kent Lawrence is an educational consultant specializing in small schools and schoolfacilities. Her academic background includes a BA in anthropology from Bennington College, an MAin sociology from New York University, and an Ed.D. from Boston University. She has taughtstudents from middle school through college level, and headed the history and guidance departmentsin a high school. For many years she ran a real estate brokerage and construction company in Maine,where she started an environmental education program for children (Grades K-9), and a foundationfor a local school. She was a policy analyst for The Rural School and Community Trust and is theauthor of many articles and two books. Lawrence is an adjunct professor at Lesley University whereshe teaches writing, research methods, and an introductory course in sociology, and is currentlyfinishing a third book.

Paul Abramson is president of Stanton Leggett & Associates (SLA), an educational consulting firmspecializing in analysis of long range needs, program change, facilities development and revision,administrative, faculty and support staffing, and the organization of the educational process. SLAhelps clients explore ideas and opportunities that open up new directions and provide successful andsometimes unique solutions to school management problems with a particular emphasis ondeveloping facilities to support the educational program.

Mr. Abramson has been editing publications serving school administrators and teachers since1957 and worked as a consultant and writer for Educational Facilities Laboratories. Since 1974, hehas provided an annual study of school and college construction throughout the United States. Datafrom that study forms the basis for analyzing the cost of large and small schools as shown in thisstudy. Abramson writes a monthly column on planning issues – A Final Thought – for School Planning& Management magazine for which he also serves as education industry analyst.

Abramson is a member of the Council of Educational Facility Planners International and waspresident of the organization’s Northeast Region. He is a graduate of Reed College, Portland, Oregon.

Victoria Bergsagel is an educator passionate about designing schools where all students achieve.Harvard-educated, Bergsagel has been a teacher, counselor, principal, community relations director,school district administrator, and adjunct professor. As the director of educational design in a large publicschool district, Bergsagel led the educational program planning upon which the construction of newschools was based. She has also served as the director of educational partnerships at Talaris ResearchInstitute. Working collaboratively with teams of researchers and educators to conduct, integrate, andinterpret some of the world’s leading brain research, her team helped connect relevant research findingsto practical applications – combining the science of learning with the practice of learning.

Bergsagel is a creative thinker with an unrelenting advocacy for children and a vast knowledgeof educational research. She founded and directs Architects of Achievement, a group partnering withthe University of Washington Small Schools Project to help educators and architects across thenation integrate the work of facility design into school reform.

Steven Bingler is the founder of Concordia LLC, a community planning and architectural designfirm. Concordia’s award-winning designs have appeared in national publications, includingArchitectural Digest, Progressive Architecture, Architecture, Interiors, the Los Angeles Times, the New YorkTimes, USA Today, and the Wall Street Journal.

Concordia has extensive experience in the planning and design of a wide range of community-based environments for living and learning. The firm’s planning and design process employs theConcordia Model, a comprehensive community engagement and systemic planning tool thatintegrates physical, cultural, social, educational, organizational and economic assets and needs.Research alliances include the MIT Media Lab, Harvard University’s Project Zero, the University of


63

New Mexico, the National Aeronautics and Space Administration, the Thornburg Institute, theAppalachian Education Lab, and the West Ed Lab. Bingler’s papers have been published in a widerange of books and journals in the fields of urban planning, architectural design, education, publichealth, and smart growth.

Mr. Bingler is a frequent speaker at national symposia and conferences related to systemsthinking and innovations in community-based planning and design. He served as a special consultantto the Office of the Secretary of the U.S. Department of Education for policy related to the design ofschools as the centers of the community.

Barbara Diamond is the vice-president responsible for one of KnowledgeWorks Foundation’s majorprogram areas – Communities and School Facilities planning and design – and for its public policywork. The Foundation is Ohio’s largest education philanthropy, with a mission to increase access toeducation throughout Ohio. Its other major program areas are College and Career Access, andSchool Improvement.

Diamond received her bachelor’s and J.D. degrees from Harvard University. She has beenadmitted to the bar in Minnesota, Massachusetts, and Ohio and has extensive experience as anattorney and policy analyst. She has served as staff attorney for the Education and Higher Educationcommittees of the Minnesota House of Representatives, and for the committees on Criminal Justiceand Ways and Means of the Massachusetts Senate. Before moving to Cincinnati in 1999, she servedas Counsel for Policy Development for the Massachusetts Supreme Judicial Court, where she oversawstatewide master planning for courthouse construction.

Thomas J. Greene, whose background is in finance and accounting, is the budget consultant on theDollars & Sense II project team. He is the CEO, CFO and publisher of Greene Bark Press Inc., andsenior consultant for TJGreene Associates, a division of Greene Bark Press Inc. He is the formervice-president of finance and controller of Powerwinch (a Warren Buffet owned company); formerbusiness manager of Fortune Magazine (a Time-Warner/AOL company); former controller of RRBowker (formerly a Xerox-owned company); and former senior consultant with Lucas, Tucker &Company CPA’s. He holds a BBA in accounting and economics and a MS in industrial relations andeconomics. He was an adjunct in accounting at Sacred Heart University, Norwalk CommunityCollege, and York Community College in New York City. He has recently cofacilitated severalbudgeting workshops for small school leaders sponsored by KnowledgeWorks Foundation.

Bobbie Hill is the director of planning for Concordia LLC. Through the Concordia Planning Model,Hill works with communities to help them realize their potential through consensus-building andexploring learning opportunities through collaboration. She is committed to public scholarship byhelping communities become healthy civil societies that are interconnected – not homogeneous, butintegrated. She has directed projects from the rural North Country of New Hampshire to inner cityCincinnati, OH, urban Los Angeles, and Plainfield, NJ.

She has organized many local and statewide networks and associations concerned with issuesrelated to education, the arts, and community advocacy. Hill has served on regional and state taskforces such as the Governor’s Education First Committee, which developed policy and consensus-building for education reform in West Virginia. Her work on this committee and other organizationshas brought about significant change. For example, Hill worked with government leadership to createlegislation that supports community-based planning as a prerequisite for requesting state support forschool construction.


64

Craig Howley has researched rural education and published widely in that field. He has taughtmathematics at the University of Charleston and has evaluated mathematics professional developmentprojects in rural schools. He is coauthor of Small High Schools That Flourish (AEL, 2000) and Out of OurMinds: Anti-intellectualism in American Schooling (TC Press, 1995). Howley codirects research efforts ofthe NSF-funded Appalachian Collaborative Center for Learning, Assessment, and Instruction inMathematics (ACCLAIM) and is an adjunct professor in the Educational Studies Department at OhioUniversity. His work for ACCLAIM entails cultivation of research capacity in rural mathematicseducation. Previously, he directed the ERIC Clearinghouse on Rural Education and Small Schools. Hisempirical work is mostly quantitative; it investigates the mediating influence of school and district sizeon academic achievement. Other studies have examined construction costs of smaller schools, ruralschool busing, and rural principals’ perspectives on planning.

David Stephen’s professional life straddles the worlds of architectural design and education reform. Asa Registered Architect he has more than 18 years of experience in building design and construction. Asa teacher and education reformer, he has worked with high schools nationwide to envision, develop,and implement innovative practices in teaching and learning. Increasingly, Stephen’s focus has turnedto the design of school buildings that explicitly address and support their program’s curricular goals.During the past five years, he has facilitated the architectural design process for a wide variety of newand reforming schools.

At present, Stephen works as program officer and director of facilities design for High Tech HighLearning, the replication branch of the nationally acclaimed High Tech High (HTH) charter school inSan Diego, CA. He led the architectural design effort for the HTH campus’s three school buildings:High Tech High, High Tech Middle, and High Tech International. The HTH facility received a 2001Educational Design Excellence Award from the American School & University Architectural Portfolio,and an Honors Award from the School Construction News and Design Share Awards 2002. Stephen alsoworks with Architects of Achievement, a group partnering with the University of Washington SmallSchools Project to help educators and architects across the nation integrate the work of facility designinto school reform. He received his B.Arch from Rhode Island School of Design, and an M.Ed fromLesley College.

Elliot Washor is the cofounder of The Big Picture Company and The Met. He has been involved inall aspects of school reform for more than 30 years as a teacher, principal, video producer, and developerand director of small school networks. As a practitioner he is particularly interested in real worldlearning and literacy, from kindergarten through college, in urban and rural settings, and across alldisciplines. His work concerns what it means to be “smart” in a given situation, both alone and in agroup, and how learning is amplified when hands and mind are used in tandem. He has served as aconsultant to schools throughout the United States, Europe, and Asia, and has written on many topics,including leadership to learning. His professional development programs won an Innovations in Stateand Local Government Award from the Ford Foundation and the Kennedy School of Government atHarvard. In addition, he has been at the national forefront of thinking about small schools and theirfacilities design within communities, and was recognized by DesignShare (The International Forum forInnovative Schools) with an International Award of Merit for the design of The Met.

The Dollars and Sense team wishes to express its appreciation for the contributions of the indispensableFrances Saad to this project. She assembled the original data base of schools, made travel arrangementsfor all the site visits, organized team meetings, tracked the budget, matched captions to photos, proof-readevery copy, and performed other tasks too numerous to name – all with grace, intelligence and good sense.Without her, this publication would not exist, and the team’s admiration and gratitude are boundless.


65

Resources

Dollars & Sense: The Cost Effectiveness of Small Schools, www.kwfdn.org.The first report in the Dollars & Sense series summarizes the history of the move to larger schools andreviews the research on the benefits of small schools. It makes the case for the value of small schoolsand shows that they have been built cost-effectively. Dollars & Sense: The Cost Effectiveness of SmallSchools has become an essential resource for advocates of small schools.

Education Information

National Center for Education Statistics (NCES)nces.ed.gov1990 K Street, NW Washington, DC 20006202-502-7300

NCES is the primary federal entity for collecting andanalyzing data that are related to education in theUnited States and other countries. It offers specificinformation about all K-12 schools in the United Statesincluding, size, demographics, and fiscal data.

Facilities Information

Architects of Achievementwww.archachieve.org

Architects of Achievement provides facilities-relatedservices to schools, and offers a website rich ininformation about and photographs of successful smallschools and ways in which their facility supports theirprogram. This site includes floor plans of some of theschools in this report.

Council of Education Facilities Planners,International (CEFPI) www.cefpi.org

CEFPI is the trade organization for people involved inthe planning, design, and construction of schoolfacilities. Though some of the site is open to membersonly, a lot of information is available to the public.

National Clearinghouse for EducationalFacilities (NCEF) www.edfacilities.org

The NCEF is the primary source for articles and reportsabout school facilities.

Schools as Centers of Community

21st Century School Fund (21st CSF)www.21csf.org1816 12th Street NWThurgood Marshall CenterWashington, DC [email protected]

21st CSF focuses on urban schools, particularly those inthe Washington, DC, area, but its innovative approachto forming partnerships for facility planning andfinancing has earned national respect.

KnowledgeWorks Foundationwww.kwfdn.orgOne West Fourth Street, Suite 200Cincinnati, Ohio 45202513-929-4777

KnowledgeWorks Foundation empowers communitiesand individuals to improve education. Its website is aresource on small schools, schools as centers ofcommunity, and community engagement.

Schools as Centers of Community: A Citizen’sGuide for Planning and Design Bingler, S., and L. Quinn, Washington, DC: U.S.Department of Education.Available for download:http://www.kwfdn.org/resource_library/_resources/schools_as_centers2.pdf

This important publication introduces key elements inplanning a school as the center of its community, andoutlines a process for stakeholders to use in achievingthat goal.

Smaller, Safer, Saner Successful SchoolsNathan, J. and K. Febey. 2001. Minneapolis,MN: Center for School Change andWashington, DC.: The National Clearinghousefor Educational Facilities.Available through:http://www.hhh.umn.edu/centers/school-change/reform.htm

This publication offers examples of good small schoolsand ways in which they are effective.


66

Small Schools

AELwww.ael.orgP.O. Box 1348 Charleston, WV, 25325-1348 800-624-9120 or 304-347-0400.

AEL is the former host of the ERIC Clearinghouse onRural Education and Small Schools, known asERIC/CRESS. AEL was a major source of informationand research on small schools and continues to makethis work available.

Rural School and Community Trustwww.ruraledu.org1530 Wilson Blvd., Suite 240Arlington, VA 22209703-243-1487703-243-6035 fax

Many of the schools in rural areas are small. The RuralSchool and Community Trust is a source of researchand information about such schools that can also beuseful for people concerned with urban education.

School Redesign Networkwww.schoolredesign.comStanford UniversitySchool of Education520 Galvez MallStanford, CA 94305-3084650-725-0703

School Redesign Network is a resource for peopleworking to improve schools and student achievement,and offers Study Kits on a variety of topics to supportthis effort.

Small Schools Projectwww.smallschoolsproject.org7900 East Greenlake Drive North, Suite 212Seattle, WA 98103206-616-0303206-543-8250 fax

The Small Schools Project based in Washington Stateis a resource for information about all aspects ofcreating a good small school.

Small Schools Workshop www.smallschoolsworkshop.org1608 North Milwaukee Avenue, Suite 912Chicago, IL 60647773-384-1030773-384-1226 fax

The Small Schools Workshop provides guidance andexpertise particularly to large school systems workingto restructure into smaller learning communities.


68

ContactMs. Andrea Martin, Program CoordinatorAvalon Charter High School1745 University Avenue WestSt. Paul, MN 55104(651) 649-5495, ext. 202

Website: www.avalonschool.org

Location: UrbanDistrict: Avalon Secondary SchoolsType of school: CharterGrades served: 9-12Number of students: 125NCES school ID: 2770187

Appendix 1 • Information by School

Note: Floor Plan and Site information, when available, are at www.goodsmallschools.org

Avalon Charter High SchoolSt. Paul, MN

Population served:1

White, non-Hispanic — 67.6%Black, non-Hispanic — 13%Hispanic — 3.7%Asian/Pacific Islander — 12%American Indian/Alaskan Native — 3.7%

Percent of students eligible for freeand reduced-cost lunch: 14%

continued >

All Costs

% Description Amount

INSTRUCTIONAL PROGRAM

SALARIES

30.52% Salaries: instructional $366,446.163.74% Salaries: student support/special ed $44,861.658.40% All staff benefits $100,831.0842.65% Subtotal $512,138.89

NON-SALARY INSTRUCTIONAL EXPENSES

0.37% Professional development $4,385.772.34% Curriculum materials and supplies $28,057.920.95% Special ed consultant fees & outside support $11,442.005.10% Non-instructional supplies & materials $61,211.764.93% Instructional services, lecturers, program services $59,176.1213.68% Subtotal $164,273.5756.33% Total all instructional $676,412.46

EXPENSES: OTHER OPERATIONAL

0.09% School Board $1,048.332.25% Transportation $27,056.364.32% Administrative expenses $51,850.114.81% Office salaries & benefits $57,745.471.34% Supplies & postage $16,045.01

1 Information in this appendix about population served and percent of students who receive free and reduced-price lunches is taken fromthe National Center for Education Statistics website, at www.nces.ed.gov/globallocator/. This data is for the 2002-2003 school year, thelatest available, whereas the school contact information and the number of students is for 2004-2005.


69


0.12% Dues & memberships $1,385.000.19% Telephone $2,270.000.21% Miscellaneous $2,569.6013.32% Subtotal $159,969.8869.65% Subtotal non-facility costs $836,382.34

FACILITY COSTS

5.25% Capital outlays (new and equipment replacement) $63,014.7618.21% Building lease $218,718.074.98% Utilities (propane, disposal, & water) $59,809.790.81% School insurance $9,737.661.09% Repairs, maintenance, & supplies $13,148.5230.35% Total facility costs $364,428.80

100.00% Total all costs $1,200,811.14

Building History

Year constructed: 1916Year of remodeling: 2001Total acreage: 1 acre for total site that serves three schoolsTotal constructed space: 64,000 sq. ft., three floors andbasementFormer uses: The building was originally used by Brown,Blodgett and Sperry Company, a firm that produced stockhouse plans.

ArchitectureRandall Fielding, Architect and Planner(612) 925-6897(612) 922-6631 [email protected]

Jamie Wellik (612) [email protected]

Facility-Related DetailsRenovation, new construction, or addition: RenovationSquare footage: 15,000 sq. ft. (doesn’t include freightelevator + mechanical room)Square footage per student: 118 sq. ft. per studentCost to acquire site: LeasedCost to acquire building: LeasedCost of construction: $127,108

Cost per square foot: $8.47 (The cost was minimized bythe fact that the space had been occupied by anothercharter school.)

Floor plan: www.goodsmallschools.org

Financial NotesAvalon’s total actual expenditures were $1,200,811 with an average annual attendance of 118 students and a staff of 11. Theannual cost per pupil of $10,176.36 is somewhat inflated, compared to non-charter schools, because of its lease and contractcosts of $218,718 per year. If we adjust for the lease expense, the average cost per pupil would be approximately $8,322. Removingthe lease cost from the equation would also improve the percentage of the budget spent on instructional programs from 56.3% to74.4%, which would be somewhat higher than other small schools. Avalon also devoted almost five percent of its budget to technologyand Internet activities ($57,905 or 4.8%). A significant amount of money was set aside in the original budget for out-of-schoolresources such as field trips ($14,000 or 1.1%) in keeping with the project-based and technology vision of the school. The originalbudget also provided for $16,887 for a parent liaison, a position that has proven to be invaluable to small schools in urbanenvironments.


70

Contact Dr. Patricia Cooper, PrincipalR.D. Schroder Middle School7224 Hwy 162Hollywood, SC 29449(843) 889-2391

C.C. Blaney – R.D. SchroderYonges Island, SC

Contact Ms. Cindy Smalls, PrincipalC.C. Blaney Elementary School7184 Highway 162Hollywood, SC 29449(843) 889-3992

Contact Mr. Leroy Seabrook, Chairman St. Paul Constituent District 237865 Toogoodoo RoadYonges Island, SC 29449

Location: RuralDistrict: Charleston CountyType of school: PublicGrades served: PK-5 Number of students: 190 (2004-2005)NCES school ID: 4501440

Population served:White, non-Hispanic – 1.6%Black, non-Hispanic – 94%Hispanic – 4.4%

Percent of students eligible for free andreduced-cost lunch: 92

C.C. Blaney Elementary School

All Costs



SALARIES

31.21% Salaries & benefits: instructional $522,702.009.07% Salaries & benefits: staff support $151,854.2828.39% Salaries & benefits: student support $475,472.3168.67% Subtotal $1,150,028.59


2.63% Professional development (including travel & conferences) $44,028.830.98% Curriculum materials and supplies $16,386.520.23% Printing and binding $3,773.003.31% Media center $55,435.942.34% Instructional services, lecturers $39,250.000.48% Rental/leases $8,085.849.97% Subtotal $166,960.1378.63% Total instructional expenses $1,316,988.72


9.30% Office of the Principal $155,780.360.04% Transportation $634.290.19% Telephone $3,189.400.08% Other miscellaneous services $1,392.32

continued >

Website: www.blaney.ccsdschools.com


71


1.04% Community services $17,336.7110.65% Subtotal $178,333.0889.28% Subtotal non-facility costs $1,495,321.80

FACILITY COSTS

4.24% Property services $71,085.420.07% Repairs & maintenance $1,174.532.14% Custodian costs $35,840.594.26% Utilities (heat, gas, & electricity) $71,407.1010.72% Subtotal facility costs $179,507.64

100.00% Total all costs $1,674,829.44

Financial NotesThe C.C. Blaney Elementary school has approximately 253 students K thru fifth grade (and temporarily sixth grade). It has a totalbudget of $1,674,829. The average annual spending per pupil amounts to $6,620. Nearly 75% of the budget is devoted tosalaries and benefits. Moreover almost 80% of the total budget is devoted to the regular instructional programs. More than eightpercent of the instructional budget is devoted to student impairments, e.g., speech, learning, and emotionally impaired, with the bulkof this spending on the emotionally handicapped (their term) or $63,007 of the total $81,605 special education budget. Contractedservices and utilities account for more than 81% of the total facilities spending and the custodial staff reports to the principal and isincluded within the school budget. Capital spending, transportation, and food service are not included in this budget.

Location: RuralDistrict: Charleston CountyType of school: PublicGrades served: 6 -8Number of students: 390 (2004-2005)

Population served:White, non-Hispanic – 3.8%Black, non-Hispanic – 94.6%Hispanic – 1.6%Percent of students eligible for free andreduced-cost lunch: 91

NCES school ID: 4501440

Website:http://schroder.ccsdschools.com/home.asp

R.D. Schroder Middle School

All Costs



SALARIES

45.43% Salaries & benefits: instructional $1,009,037.0019.58% Salaries & benefits: special needs $434,912.3265.02% Subtotal $1,443,949.32


2.11% Professional/curriculum development $46,890.171.58% Instructional services $35,000.004.34% Curriculum materials $96,310.001.69% Technology & software supplies $37,496.990.93% Printing & binding $20,614.800.05% Media center $1,101.000.12% Miscellaneous other $2,713.2110.81% Subtotal $240,126.1775.83% Total instructional expenses $1,684,075.49

continued >


72



10.71% Salaries & benefits: administration $237,884.070.46% Pupil support services: non salary $10,200.290.72% Student transportation $15,989.640.06% Supplies & postage $1,411.620.38% Telephone $8,518.770.06% Travel & conference $1,296.170.04% Miscellaneous other $841.0712.43% Subtotal $276,141.6388.26% Subtotal non-facility costs $1,960,217.12

FACILITY COSTS

0.92% Custodial salaries & benefits $20,347.293.71% Property services $82,355.416.43% Utilities (heat, gas, & electricity) $142,752.370.69% Building other costs, including purchased services $15,263.6911.74% Total facility costs $260,718.76

100.00% Total all costs $2,220,935.88

Financial NotesThe middle school currently has between 275 and 280 students in 7th and 8th grades only. The 6th grade was transferred two yearsago to allow for renovations to the school. It is slated to return in the fall of 2004 and will add two to three classes or a total ofbetween 40 and 60 students. The current annual cost per pupil is approximately $7,932. Although a relatively high percentageof the budget is devoted to salaries and benefits (almost 77%), nearly 90% of the budget is devoted to instructional programs. Theschool caters to a high percentage of special needs pupils and nearly 15% of the budget is devoted to that population. These programsinclude the physically, mentally, learning, and emotionally challenged. Total facilities cost of $260,718 account for almost 12% ofthe budget, the biggest chunk of this being utilities of $142,752. Capital spending and food service were not part of this budget andwould add substantially to the overall budget since the school has a cafeteria and staff. A high portion of the population qualifies forfederal entitlements.

Year constructed: Schroder: 1920,renovated 1950 and 1970s Year of addition: 2002Year of remodeling: 2002Total acreage: 31Total constructed space: 49,577 sq. ft.(Blaney), 90,267 sq. ft. (Schroder)Former uses: schools

ArchitectureHeery International(843) 566-8168

Bill LewisExecutive Director, FacilitiesImprovement, Charleston County SD(843) [email protected]

Facility-Related DetailsRenovation, new construction, or addition:Renovation and additionSquare Footage:

49,577 (Blaney)90,267 (Schroder)

Cost of construction:Blaney – $3,826,510Schroder – $4,515,585

Cost of FF&E:Blaney – $83,762Schroder – $92,714

Cost of technology:Blaney – $151,647Schroder – $172,172

Cost per square foot:Blaney – $77Schroder – $50

Town installed sewer for about $7 million.

Building History


73

Morris High School (being phased out)

High School for Violin and DanceBronx International High SchoolBronx Leadership Academy II School for Excellence

Contact Norma Morales, PrincipalBronx International High School1110 Boston RoadBronx, NY 10456(718) [email protected]

All Costs


11.37% Leadership $108,770.00 4.10% Parent coordinator $39,200.00 4.22% Office staff $40,354.00 54.60% Instructional staff $522,377.00 1.39% Classroom substitutes $13,304.00 0.00% Social worker $0.00 3.51% Support staff $33,556.00 0.00% Professional development $0.00

STUDENT SUPPORT

4.63% Guidance counselor $44,283.00 0.00% Family workers $0.00 4.39% Teachers after school $41,958.00 0.00% School secretary $0.00

NON-PERSONNEL EXPENSES

0.19% Contract services $1,800.00

Bronx Small SchoolsBronx, NY

Websites:

Bronx International High School: www.nycenet.edu/OurSchools/Region2/X403/default.htm?searchType=school

High School for Excellence: www.nycenet.edu/OurSchools/Region2/X404/default.htm?searchType=school

Bronx Leadership Academy II: www.nycenet.edu/OurSchools/Region2/X527/default.htm?searchType=school

High School for Violin and Dance: www.nycenet.edu/OurSchools/Region2/X543/default.htm?searchType=school

Location: UrbanDistrict: New York CityType of school: PublicGrades served: 9-10Number of students: 340 (2004-2005) NCES school ID: 3620580

Population served:White, non-Hispanic – 2.8%Black, non-Hispanic – 56%Hispanic – 34.7%Asian/Pacific Islander – 5.3%

Percent of students eligible for free andreduced-cost lunch: Unknown

continued >

Bronx Leadership Academy II


74


0.00% Admissions fees $0.00 0.52% Curriculum development $5,000.00 3.29% Educational consultants $31,456.00 0.09% Educational software $876.00 1.47% Equipment $14,075.00 0.00% Food and forage $0.00 0.05% Library books $471.00 1.99% Noncontractual services $19,058.00 0.15% Non-DP equipment repair $1,450.00 1.08% Supplies $10,361.00 0.30% Transportation of pupils $2,908.00 2.19% Textbooks $21,000.00 0.48% Miscellaneous $4,562.00 11.81% Subtotal non-personnel expenses $113,017.00

100.00% Total all costs $956,819.00

Financial NotesThe Bronx Leadership Academy has 156 students in two grades (9th and 10th). Last year at this time, there were only 76 studentsin one grade, the ninth. Based on this enrollment, the average annual cost per student would be $6,133.46. Uncommittedbudgeted dollars (see note below) of approximately $50,000 would reduce the average annual spending per child to $5,816.00. Keepin mind, however, that these costs do not include facilities or food service costs. It would appear that the plan is to add a grade a yearfor the next two years. The instructional staff of 13 includes all the traditional disciplines: four language arts (English), three math,four science, and two social studies. One science and one math position are vacant.

The word Committed in NYC terms is not synonymous with encumbered. Committed seems to allow the school to move moneybetween line items while at the same time positioning the budgeted funds to be spent on a known person, place, object, or event.Facilities are budgeted and controlled at the Central Board of Education.

All Costs


8.32% Leadership $108,770.00 3.00% Parent coordinator $39,200.00 5.10% Office staff $66,771.00 53.13% Instructional staff $694,937.00 0.85% Classroom substitutes $11,136.00 5.08% Social worker $66,403.00 9.44% Support staff $123,494.00 6.79% Professional development $88,763.00

Location: UrbanDistrict: New York CityType of school: PublicGrades served: 9-11Number of students: 300 (2004-2005)NCES school ID: 3650580

Population served:White, non-Hispanic – 2.6%Black, non-Hispanic – 27%Hispanic – 69%Asian/Pacific Islander – 1.4%

Percent of students eligible for free andreduced-cost lunch: Unknown

continued >

Bronx International


75


STUDENT SUPPORT

0.08% School social worker $1,105.00 0.00% Family workers $0.00 4.05% Teachers after school $52,950.00 0.16% School secretary $2,119.00


0.00% Contract services $0.00 0.00% Admissions fees $0.00 0.03% Curriculum development $340.00 0.00% Educational consultants $0.00 0.13% Educational software $1,661.00 0.37% Equipment $4,874.00 0.00% Food and forage $0.00 0.11% Library books $1,492.00 0.14% Noncontractual services $1,808.00 0.00% Non-DP equipment repair $0.00 2.64% Supplies $34,546.00 0.32% Transportation of pupils $4,203.00 0.26% Textbooks $3,432.00 0.00% Miscellaneous $0.00 4.00% Subtotal non-personnel expenses $52,356.00

100.00% Total all costs $1,308,004.00

Financial NotesThe Bronx International HS has 155 students in two grades (9th and 10th) and this amounts to approximately $8,439 averageannual cost per student. The instructional staff positions are all whole classroom positions. Four of these are 11th grade and thebalances are all combination 9th and 10th grade. The school has had 9th and 10th grades both in 2004-2005 and 2003-2004. Thestudent enrollment has only increased by 11 students. Eleventh and 12th grades will be phased in during upcoming school years.

Location: UrbanDistrict: New York CityType of school: Public

Grades served: 9-11Number of students: 330 (2004-2005)Population served: Unknown

Percent of students eligible for free andreduced-cost lunch: UnknownNCES school ID: 3650580

Bronx High School for Excellence

All Costs


16.17% Leadership $199,035.00 3.21% Parent coordinator $39,550.00 3.31% Office staff $40,728.00 55.86% Instructional staff $687,758.00 1.58% Classroom substitutes $19,510.00 0.00% Social worker $0.00 3.58% Support staff $44,061.003.60% Professional development $44,381.00

continued >


76

% Description AmountSTUDENT SUPPORT

0.27% Guidance counselor $3,264.00 1.58% Family workers $19,474.00 2.35% Teachers after school $28,967.00 0.20% School secretary $2,419.00


0.37% Contract services $4,543.00 0.00% Admissions fees $0.00 0.00% Curriculum development $0.00 1.40% Educational consultants $17,247.00 0.55% Educational software $6,718.00 0.55% Equipment $6,819.00 0.20% Food and forage $2,500.00 0.15% Library books $1,843.00 1.37% Noncontractual services $16,832.00 0.00% Non-DP equipment repair $0.00 2.08% Supplies $25,562.00 0.32% Transportation of pupils $4,000.00 0.61% Textbooks $7,523.00 0.69% Miscellaneous $8,452.00 8.29% Subtotal non-personnel expenses $102,039.00

100.00% Total all costs $1,231,186.00

Financial NotesThe HS for Excellence added an 11th grade in 2004-2005 and currently has 220 students. It does not yet have any 12th gradestudents, but a phase-in scheduled for the 12th grade is planned over several semesters. On a budgeted basis this would put its annualcost per student at $5,596, excluding facilities and food service, both of which are controlled as well as budgeted at the CentralBoard of Education. Approximately $97,000 of the budget remains uncommitted and if taken out of the equation would lower theannual cost per student to $5,157 per year.

Location: UrbanDistrict: New York CityType of school: PublicGrades served: 9

Number of students: 170 (2004-2005)Population served: Unknown

Percent of students eligible for free andreduced-cost lunch: UnknownNCES school ID: 3650580

Bronx High School for Violin and Dance

All Costs


10.29% Leadership $113,288.00 3.56% Parent coordinator $39,200.00 3.65% Office staff $40,221.00 46.43% Instructional staff $510,999.00 1.06% Classroom substitutes $11,635.00 5.72% Social worker $62,946.00 3.02% Support staff $33,252.00 5.13% Professional development $56,426.00

continued >


77


STUDENT SUPPORT

0.32% Guidance counselor $3,528.00 0.42% School secretary $4,639.00


2.37% Contract services $26,055.00 0.04% Admissions fees $400.00 4.06% Curriculum development $44,722.00 0.33% Educational consultants $3,600.00 0.09% Educational software $1,038.00 6.23% Equipment $68,588.00 0.09% Food and forage $1,000.00 0.05% Library books $557.00 0.55% Non-contractual services $6,027.00 0.09% Non-DP equipment repair $1,000.00 3.44% Supplies $37,818.00 3.05% Textbooks $33,602.00 20.39% Subtotal non-personnel expenses $224,407.00

100.00% Total all costs $1,100,541.00

Financial NotesThe Bronx HS for Violin and Dance is new and currently only has 90 9th-grade students with a phase-in scheduled for the uppergrades over several semesters. On a budgeted basis this would put its annual cost per student at $12,228.23. However, only atotal of $806,256 of the budget has been committed reducing the actual annual cost per student to $8,958 for the present.

Year constructed: 1899 (cornerstone laidJuly 23, 1901; completed January 1904;dedicated June 10, 1904)Contractor: Louis Wechsler Original construction cost: $469,383Original site cost: $112,637.49 Designed for: 2,500 students, ultimatelybuilt to accommodate 2,630Former uses: Morris High School

ArchitectureOriginal: C.B.J. Snyder, superintendent ofschool buildings (1891-1923), and aresident of the Bronx

Facility-Related DetailsRenovation, new construction, oraddition: Minor renovation of MorrisHigh School. A modest amount ofconstruction was done to adjust a fewrooms (partitions, etc) to fit multiple small

schools into the building. It now holds 4small high schools that are phasing ingrades and one large high school that isphasing out grades.

Square footage: original – 2,679,500 cubicfeet, 40.62% in classrooms, balance inoffices, library, lecture rooms, lockers,gymnasiums, store rooms, and laboratories. Year of remodeling: 1997 (cost of $50million to restore landmark building)Total acreage: 1.896 acres Total constructed space: 182,000 sq. ft.(186,000 including trailers)Cost of construction: $180,000 (+$800,000 planned to convert twoadditional spaces to classrooms, move anexisting health clinic to another location,and provide signage and symbols toidentify locations of individual schoolswithin the campus building.)

Cost of FF&E: Unknown (New schoolorganizations in New York City receive“start-up” funds to cover the costs of newtechnology and furniture. There is norequirement that these funds are restrictedto that use, however. The amount for thispurpose provided to each of the four smallhigh schools in the building is $90,000.)

Cost of technology: Unknown (Newschool organizations in New York Cityreceive “start up” funds to cover the costsof new technology and furniture. There isno requirement that these funds arerestricted to that use, however. Theamount for this purpose provided to eachof the four small high schools in thebuilding is $90,000.)

Cost per square foot: $0.99 (including future planned expansion: $5.38)

Building History


78

Contact Ms. Ana Ponce, Executive DirectorCamino Nuevo Charter Schools635 S. Harvard Blvd.Los Angeles, CA 90005(213) 736-5542 ext. 106

Location: UrbanDistrict: Los AngelesType of school: CharterGrades served: PK-3NCES school ID: Unknown

Camino Nuevo Charter Academy/BurlingtonStreet Campus – Elementary School

Los Angeles, CA

Number of students: 257 (2004-2005)Population served:

UnknownPercent of students eligible for freeand reduced-cost lunch: Unknown

All Costs



SALARIES

45.55% Salaries: instructional $567,600.009.03% Salaries: administration $112,500.000.72% Salaries: staff support $9,000.002.60% Salaries: office staff $32,450.0014.60% All staff benefits $181,967.0072.51% Subtotal $903,517.00


0.56% Professional development (including travel & conferences) $7,000.003.80% Curriculum materials, including textbooks $47,400.000.02% Dues & memberships $300.003.21% Instructional services, lecturers $40,000.007.60% Subtotal $94,700.0080.11% Total instructional expenses $998,217.00


0.14% Transportation $1,800.000.56% Equipment rental, repair, & leases $7,000.000.54% Other miscellaneous services $6,692.001.24% Subtotal $15,492.0081.35% Subtotal non-facility costs $1,013,709.00

continued >

Website: www.caminonuevo.org/partnerships.html


79


FACILITY COSTS

0.08% Capital outlays (new and equipment replacement) $1,000.000.04% Construction cost (depreciated capitalized assets) $500.0011.33% Building lease $141,192.006.13% Utilities (heat, gas, & electricity) $76,440.000.83% Maintenance $10,400.000.23% Building other costs, including trash hauling $2,869.0018.65% Total facility costs $232,401.00

100.00% Total all costs $1,246,110.00

Financial NotesThe Burlington Street elementary school houses 240 students in grades K-3 but will add a Pre-K program next year withapproximately 30 children. The average annual spending per child is $5,192.13 with more than 80% of the budget devoted toinstructional expenditures. The school spends a lot of money on outside programs and consultants including art and dance. A total of$40,000 ($30,000 allocated to dance for an outside company and $10,000 for art programs) of the budget is devoted to theseprograms. In relation to other charter schools, the amount devoted to the facilities is higher than average due to lease expenses, whichrepresent almost 12% of the budget. A state-funded rent assistance program ended this year but allowed the parent company to amasscapital over the three years that the program existed although the schools were never certain of how much or even when until it wasactually paid. The school uses separate outside maintenance/custodial services for the two schools and plans to expense (as opposed tocapitalize) improvements and/or repairs to the building in the amount of approximately $3,000. Special education and food servicesare handled by the parent company and are not part of an allocated cost to the school. Compliance and reporting concerns are alsohandled by the parent company and are not an allocated cost to the school.

Unavailable

ArchitectureDaly Genik ArchitectsKevin Daly(310) 656-3180 (310) 656-3183 [email protected]

Building History

Facility-Related DetailsSquare footage: UnknownSquare footage of renovation: 11,319Cost of acquiring building: $651,190Cost of renovation: $1,250,350Design fees and permitting costs: $81,272Cost per square foot of renovation: $110.46Cost of FF&E: $22,532Cost of technology: $26,389


80

ContactMs. Ana Ponce, Executive DirectorCamino Nuevo Charter Schools635 S. Harvard Blvd.Los Angeles, CA 90005(213) 736-5542 ext. 106

Location: UrbanDistrict: Los AngelesType of school: CharterGrades served: 4-8Number of students: 294NCES school ID: Unknown

Camino Nuevo Charter Academy/BurlingtonStreet Campus – Middle School

Los Angeles, CA

Population served:Unknown

Percent of students eligible for freeand reduced-cost lunch: Unknown

All Costs



SALARIES

42.94% Salaries: instructional $520,050.006.81% Salaries: administration $82,500.000.74% Salaries: student support $9,000.002.68% Salaries: office staff $32,450.0013.64% All staff benefits $165,175.0066.82% Subtotal $809,175.00


0.52% Professional development (including travel & conferences) $6,250.004.20% Curriculum materials, including textbooks $50,910.000.02% Dues & memberships $300.000.41% Instructional services, lecturers $5,000.005.16% Subtotal $62,460.0071.98% Total $871,635.00


0.14% Transportation $1,650.000.29% Equipment rental, repair & leases $3,500.000.61% Other miscellaneous services $7,418.001.04% Subtotal $12,568.0073.01% Subtotal non-facility costs $884,203.00

continued >

Website: www.caminonuevo.org/partnerships.html


81


FACILITY COSTS

2.75% Capital outlays (new and equipment replacement) $33,250.000.00% Construction cost (depreciated capitalized assets) $0.0017.90% Building lease $216,804.000.00% Custodian costs $0.005.22% Utilities (heat, gas, & electricity) $63,180.000.86% Maintenance $10,400.000.00% Insurance $0.000.26% Building other costs, including trash hauling $3,180.000.00% Service contracts: elevator, heating $0.0026.99% Total facility costs $326,814.00

100.00% Total all costs $1,211,017.00

Financial NotesThe Burlington middle school includes grades four through six and seven through eight, and these divisions have 158 and 108 studentsrespectively. During renovations the fourth grade was housed with the elementary school. The 2003-2004 school year was the firstyear back in the middle school building. The annual average cost per student is $4,553 and while more than 70% of the budgetis devoted to the instructional program, facilities costs – as is also true of the Burlington street elementary school – take a more thanaverage bite out of the budget at almost 27%. This is more than twice as much as is generally true for other charter schools. Rentagain takes the largest chunk, 17.9% or approximately $217,000. Included in the facilities costs was more than $33,000 devoted tooutfitting the six new classrooms the school added this year with computers, accessories, and furniture. The two buildings are ownedby the parent company, which sets and is paid those rents. As is true of the elementary school, central administration absorbsinsurance as well as special education, food service, compliance, and publicity issues. The school plans to expense $3,180 on buildingrepairs or improvement and set aside nearly $36,429 for equipment and furniture.

Unavailable

ArchitectureUnavailable

Building History

Facility-Related DetailsSquare footage: UnknownSquare footage of renovation: 19,500 Cost of acquiring building: $1,002,668 Cost of renovation: $4,126,438 Design fees and permitting costs:Cost per square foot of renovation: $211.61Cost of FF&E: $23,073Cost of technology: $16,431


82

ContactKirk Quitter, DirectorFlagstaff Arts and LeadershipAcademy3100 North Fort Valley RoadFlagstaff, AZ 86001(928) 779-7223(928) 779-7041 fax

Website: www.fala.apscc.k12.az.us

Location: UrbanDistrict: Excel Education Centers, Inc.Type of school: CharterGrades served: 9-12 Number of students: 160NCES school ID: 0400091

Flagstaff Arts and Leadership AcademyFlagstaff, AZ

Population served:White, non-Hispanic – 46.9%Black, non-Hispanic - .6%Hispanic – 13.9%Asian/Pacific Islander - .6%American Indian/Alaskan Native – 38%


All Costs



SALARIES

45.32% Salaries & benefits: instructional $425,856.0012.86% Salaries & benefits: administration $120,812.004.14% Salaries & benefits: student support $38,873.0062.31% Subtotal $585,541.00


0.43% Textbooks $4,000.002.43% Instructional supplies $22,800.000.59% Purchase services $5,508.002.04% Miscellaneous $19,200.005.48% Subtotal $51,508.0067.79% Total all instructional $637,049.00


4.07% Professional services $38,285.000.47% Postage (UPS, FedEx, USPS) $4,400.000.53% Supplies $5,000.000.32% Dues & fees $3,000.000.48% Telephone $4,500.005.87% Subtotal $55,185.0073.67% Subtotal non-facility costs $692,234.00

FACILITY COSTS

0.07% Security $700.001.06% Equipment lease $10,000.000.61% Property lease $5,700.00

continued >


83


1.45% Custodian costs $13,600.002.49% Utilities (electricity & water) $23,400.000.21% Maintenance $2,000.001.60% Insurance $15,000.009.46% Trailer lease $88,922.000.11% Disposal/recycling $1,032.000.21% Technology $2,000.000.00% Miscellaneous & supplies $20.009.05% Emergency repair $85,063.0026.33% Total facility costs $247,437.00

100.00% Total all costs $939,671.00

Financial NotesFALA’s total FY 2003-2004 budget is $939,671 or an annual average spending of $6,102 per student for its currentenrollment of 154. Included in the budget, however, is $85,063 from a federal emergency repair grant which was used for thetransportation, installation, and hook-up of a new trailer; the construction of a deck connecting the classrooms; and cinder pathsaround the campus courtyard. Eliminating this onetime expenditure would lower the average annual spending per pupil to $5,550.Total spending devoted to instructional service, including special education, entitlements, and Proposition 301, but excludingadministration wages/benefits, totals $516,237 or 54.9% of the total school budget but 60.4% minus the emergency repairexpenditure. FALA does not offer transportation (no buses, drivers, fuel, repair/maintenance), does not offer meals (no kitchenstaff, cafeteria, kitchen, or foodstuff), and does not offer sports (no gym, sports equipment, etc.).

The school staff of 23 includes 10 full-time and six part-time teachers, one director, one secretary, one administrative assistant, andthree study hall substitutes. FALA spent approximately $2,000 on a promotion and advertising program and drew an additional 20+students, giving the school a cash windfall of $100,000+ for the year. Facilities costs total $166,874 or 17.8% of the total schoolbudget and almost 20% if we exclude the emergency repair grant. More than half (56.7%) of the budgeted facilities expenditures areaccounted for by the lease of the trailers and the property from the museum or $94,622. However, FALA lumps some operationalcosts under facilities, such as PC Insurance ($15,000), technology ($2,000), equipment leasing/supplies ($10,000), dues/misc.supplies ($20), and telephone expenses ($4,500). The total of all these operational expenses is $31,520. If this amount wasexcluded total facilities cost would represent 14.4% of the total budget or $135,354. There are no custodians; the director andstudents do maintenance. Lastly, FALA is able to take advantage of Arizona’s state tax law, which allows a tax credit of up to $250per family for donations given to any school. Last year FALA was able to raise $26,000 through these donations.

Year of remodeling: leased starting in1996, decks constructed 2002Former uses: housing


Facility-Related DetailsRenovation, new construction, and/oraddition: originally five modularbuildings on museum property;

currently seven modular buildings (one is owned)Square footage: six modulars, each 30’X 60’ for a total of 10,800, whichprovides classrooms and space foradministrative officesCost to acquire site: leased space frommuseum for $5,700 per year Cost to acquire building: Leased:$88,922

Cost of construction: originally leasedmodulars from museum; currently ownone modular;$94,000 to buy modular and builddecks connecting the modularclassrooms.

Building History


84

ContactMs. Simi BumiaHigh Tech High2861 Womble RoadSan Diego, CA 92106-6025(619) 243-5000(619) 243-5050 fax

Website: www.hightechhigh.org

Location: UrbanDistrict: UnknownType of school: CharterGrades served: 9-12Number of students: 368 (2002-2003);454 (2003-2004)NCES school ID: N/A

High Tech HighSan Diego, CA

Population served: N/APercent of students eligible for freeand reduced-cost lunch: Unknown

All Costs



SALARIES

56.10% Salaries & benefits: instructional $1,494,324.0013.54% Salaries & benefits: administration $360,738.0069.65% Subtotal $1,855,062.00


1.18% Curriculum materials $31,350.001.18% Subtotal $31,350.0070.82% Total instructional expenses $1,886,412.00


1.00% District payment $26,642.000.28% Student transportation $7,359.000.26% Meetings & hospitality $7,001.000.98% Printing & duplicating $26,045.001.18% Supplies & postage $31,530.001.68% Telephone $44,685.000.09% Dues & memberships $2,285.000.34% Travel & conference $9,139.000.01% Student meals $314.007.00% Charter management fees $186,492.002.50% Contingency reserve $66,604.0015.32% Subtotal $408,096.0086.15% Subtotal non-facility costs $2,294,508.00

continued >


85


FACILITY COSTS

8.92% Rent & leases $237,690.000.56% Housekeeping $14,894.002.78% Utilities (heat, gas, & electricity) $73,964.000.45% Maintenance $12,009.001.14% Insurance $30,474.0013.85% Total facility costs $369,031.00

100.00% Total all costs $2,663,539.00

Financial NotesAlmost all of High Tech High’s revenues (96.3%) come from the per-student allotment of $5,744 (which in FY2004-2005 isanticipated to drop to $5,612). In contrast, the San Diego Unified School District has a per-student allotment of $7,600 (dropping to$7,100 in FY 2004-2005). Proposition 13 limits the amount of money that can be spent on education to one percent of propertytaxes. Thus, with only $99,709 coming from other state and federal sources, the per-pupil spending, for the 457 students of HighTech High, amounts to approximately $5,830 per pupil. While more than 70% of the budget is devoted to instructional expenses,High Tech High pays a management fee to the CMO (Charter Management Organization) of $186,492 or a total of seven percentof the school’s budget or $408 per pupil. The school also pays the district a one percent oversight fee or $26,642. The school devotesalmost 14% of its budget to facility-related expenses or a total of $369,031. More than half of this is rent and lease expenses or$237,690. The school’s budget also includes a contingency allowance of $66,604, which is 2.5% of its overall budget. This is clearlya teacher’s school, i.e., a school run by the teachers with a full time equivalent staff (FTE) of 26 teachers. Reportedly, the teachersplan their own professional development with only $9,139 devoted to travel and conferences. There may be an additional FTE of 3.7teachers for special education and arts through a Regional Occupation Program. What is not included in the numbers is therenovation cost of $3.5 million nor the monies spent on the computers, furniture, common room, bio-lab, portable lockers, stage,curtain, etc., which may be another $500,000. The school has a three-year lease with an option to buy for $2.7 million (before a taxcredit of $1 million).

Year constructed: 1941 Year of remodeling: 2000-2004Completed Sept. 2000: 33,500 sfrenovation of everything but the 9th-grade areaCompleted Sept. 2001: 5,000 sfrenovation of 9th-grade areaCompleted Sept. 2003: renovation ofgreat room to add two additionalproject/seminar roomsCompleted Sept. 2004: renovation ofgreat room to add four additionalproject/seminar roomsTotal acreage: 58,600 sq. ft. or 1.35 acresTotal constructed space: 38,500 sq. ft.Former uses: Building 49, formerly a

US Navy technical training centerthen a storage facility for a communitycollege.

ArchitectureDavid StephenHigh Tech High Learning

Facility-Related DetailsSquare footage of renovation: 38,500Cost and year of acquiring building:$1,700,000 (2003)Cost and year of renovation:

$3,500,000 (2000)$495,000 (2001)$198,000 (2003)

Design and permitting fees: $251,580Cost per square foot of construction:$108.90Cost of FF&E:$530,000.00 (2000-2004)Cost of technology:$1,850,000 (2000-2004)

Building History


86

ContactLee Fertig, PrincipalInterdistrict Downtown School10 South 10th StreetMinneapolis, MN 55403

Website: www.education.state.mn.us/

Location: UrbanDistrict: West Metro EducationProgramType of school: PublicGrades served: K – 12Number of students: 475 (2002)

Interdistrict Downtown SchoolMinneapolis, MN

Population served: (reported by theState of Minnesota, 2002-2003)

White, non-Hispanic – 34.9%Black, non-Hispanic – 54.9%Hispanic – 3.4%Asian/Pacific Islander – 3.6%American Indian/Alaskan

Native – 3.2%Percent of students eligible for freeand reduced-cost lunch: 35NCES school ID: 2700161

All Costs


INSTRUCTIONAL PROGRAM SALARIES

54.64% Salaries & benefits:instructional $1,931,122.005.05% Salaries & benefits: administration $178,404.008.72% Salaries & benefits:staff support $308,024.004.86% Salaries & benefits: student support $171,818.007.26% All benefits $256,485.0080.53% Subtotal $2,845,853.00


0.28% Professional development $10,000.001.27% Curriculum materials $45,000.001.56% Subtotal $55,000.0082.09% Total instructional expenses $2,900,853.00


1.29% Miscellaneous/other $45,559.000.99% Transportation $35,000.002.69% Parking $95,000.000.28% Field trips payments to other agencies $10,000.000.83% Supplies & postage $29,300.000.46% Telephone $16,400.000.36% Local travel, conference travel $12,750.000.66% Interdepartmental charge back’s $23,484.000.29% Consultant fees for services $10,185.000.48% Operating leases and rentals $17,087.000.02% Dues and memberships $750.008.36% Subtotal $295,515.0090.45% Subtotal non-facility costs $3,196,368.00

continued >


87


FACILITY COSTS

0.14% Remodeling and improvements $5,000.000.69% Equipment $24,287.004.74% Utilities (heat, gas, & electricity) $167,500.002.91% Repair and maintenance $102,875.001.07% Insurance $37,944.009.55% Total facility costs $337,606.00

100.00% Total all costs $3,533,974.00

Financial NotesThe expense of the program is influenced greatly by its prime space and downtown location. For example, limited downtown parkingnecessitates that the school rent space in a parking lot for its teachers, which costs $95,000 a year. Utilities account for $167,500($70,000 of which is electrical and heating); property liability insurance accounts for $37,944; and repairs and maintenance accountfor $102,875. There is not a lot of money spent on staff development. The majority of instructors is on loan from the participatingschool districts so their professional development may be scheduled by their respective schools. The school devotes $50,000 to after-school programs and is in partnership with the local YMCA, trading classroom and gym spaces respectively. The average annualspending per pupil amounts to $7,347.

Year constructed: 1999Year of remodeling: opened Sept. 1,1999 (not fully finished)Total acreage: 1 acreTotal constructed space:102,500 sq. ft.Foundation was reused.

ArchitectureCuningham Group Architecture, P.A.John Pfluger, Principal612-379-3400952-457-9457 [email protected]

Facility-Related DetailsRenovation, new construction, and/oraddition: New constructionSquare Footage: 102,500 Cost to acquire site: N/ACost to acquire uilding: N/ACost of construction:

$11.8 million (construction cost)$14.2 million (total project cost)

www.develop.csbr.umn.edu/msdg2/Cost of FF&E: $800,000Cost of technology: $800,000Cost per square foot:

$115(construction cost)$139 (total project cost)

Building History


88

ContactDaniel J. Hoesing, SuperintendentLaurel-Concord School District5th & Wakefield StreetP.O. Box 8Laurel, NE 68745-0008(402) 256-3731(402) 256-9465 [email protected]

Website: www.laurel.esu1.k12.ne.us

Laurel-Concord ElementaryLocation: RuralDistrict: Laurel-Concord PublicType of school: PublicGrades served: K-6Number of students: 193NCES school ID: 3100004Population served:

White, non-Hispanic – 98.6%Black, non-Hispanic – 0.9%Hispanic – 0.5%

Percent of students eligible for freeand reduced-cost lunch: 28

Laurel-Concord Public SchoolLaurel, NE

Laurel-Concord High SchoolLocation: RuralDistrict: Laurel-Concord PublicType of school: PublicGrades served: 7-12Number of Students: 195NCES school ID: 3100004Population served:

White, non-Hispanic – 99.6%Black, non-Hispanic – 0.4%


All Costs Elementary and High School



SALARIES

38.27% Salaries: instructional $1,351,432.004.45% Salaries: special education $157,282.001.72% Salaries: staff support $60,659.001.98% Salaries: student support $70,042.0012.15% All staff benefits $429,147.0058.58% Subtotal $2,068,562.00


0.81% Purchased services $28,739.003.17% Supplies and materials $112,048.002.10% Capital outlays (new and replacement) $74,123.000.26% Other expenses $9,046.002.82% Tuition paid to other districts (spec ed) $99,576.009.16% Subtotal $323,532.0067.74% Total instructional expenses $2,392,094.00


0.73% Board of Education $25,903.003.81% Transportation $134,671.005.09% Office of Superintendent $179,870.005.10% Office of the Principal $180,277.001.34% Purchased business services $47,430.003.22% State categorical programs $113,681.00

continued >


89


3.07% Federal programs $108,533.000.21% Debt service $7,558.000.51% Other miscellaneous services $18,130.0023.11% Subtotal $816,053.0090.85% Subtotal non-facility costs $3,208,147.00

FACILITY COSTS

3.50% Custodian costs $123,772.005.65% Maintenance $199,531.009.15% Total facility costs $323,303.00

100.00% Total all costs $3,531,450.00

Financial NotesThe Laurel-Concord elementary and high school budgets are combined by the Nebraska Department of Education. The schools’ pupilpopulations are almost identical (222.9 and 222.6 respectively) resulting in an annual cost per pupil of $7,927. Unlike almost allof the other schools in the study, however, Nebraska’s unique organizational model includes the cost of a superintendent’s office, andsome board of education and debt service expenses in the mix. These three add a little more than six percent to the cost burden.Nevertheless, nearly 70% of the schools’ budget is devoted to the instructional program.

Year constructed: 1923Year of addition:1960: turned assembly room intoclassrooms to accommodate increase instudents from Concord School, whichwas closed in 1958.

1966: added physical science room,biological science room, twoclassrooms, home economics room,music and band room, two restroomsand a heating plant.

1976: added elementary school wingconnected to gymnasium.

1978: activity area with gym. Oldbuilding demolished except for façade.

1979: built corridor connecting 1965building to new one.

1981: offices and a conference roomconstructed on east side of thecorridor.

1996: city and school joined in aproject spearheaded by the Laurel-Concord Athletic Boosters forconstruction of a school/communityhealth and fitness center, locatedbetween the old gym and band room.Total cost was $54,170.

1998: joint school/communitylibrary/media center. Through apartnership of City of Laurel, Laurel-Concord School District, and theLaurel-Concord EducationFoundation. 10,239 sq. ft., $348,700.Cost about $23.00 per student and$34.00 per square foot.

2000: football field, track, andelevated box seats


Facility-Related DetailsLibrary/technology centeraddition/renovationSquare footage of new construction:15,000 Square footage of renovation: 5,000 Cost of new construction: $348,000Cost per square foot of newconstruction: $23.00

Building History


90

ContactMr. Michael GoldsteinMr. Alan SafronMr. Charlie SposatoMATCH School1001 Commonwealth AvenueBoston, MA 02215(617) 232-0300(617) 232-2838 fax

Website: www.matchschool.org

Location: UrbanDistrict: Media & Tech CharterType of school: CharterGrades served: 9-12Number of students: 185NCES school ID: 2500066

Media and Technology Charter High SchoolBoston, MA

Population served:White, non-Hispanic – 10.9%Black, non-Hispanic – 64.7%Hispanic – 18.5%Asian/Pacific Islander – 5.9%


All Costs



SALARIES

33.13% Salaries & benefits: instructional $785,225.008.28% Salaries & benefits: administration $196,305.002.60% Salaries & benefits: staff support $61,700.008.05% Salaries & benefits: student support $190,770.006.33% Tutoring $150,000.0058.39% Subtotal $1,384,000.00


0.72% Professional development $17,000.001.27% Curriculum materials $30,000.000.89% Materials, printing and recruitment $21,000.000.63% Contracted special education $15,000.000.21% Summer academy materials & costs $5,000.003.71% Subtotal $88,000.0062.10% Total $1,472,000.00


1.22% Professional services $29,000.000.59% Transportation $14,000.000.63% Parking $15,000.0016.88% Adjusted debt service $400,000.001.05% Supplies & postage $25,000.000.32% Furniture & fixed noncapitalized $7,500.000.21% Clubs, sports elective materials/equipment $5,000.001.01% Community culture fund $24,000.001.27% Student meals $30,000.000.78% Book trips $18,500.002.83% Contingency reserve $67,000.0026.79% Subtotal $635,000.0088.89% Subtotal non-facility costs $2,107,000.00

continued >


91


FACILITY COSTS

7.59% Construction cost (depreciated) $180,000.000.26% Construction cost (depreciated capitalized assets) $6,250.000.84% Custodian costs $20,000.001.18% Utilities (heat, gas, & electricity) $28,000.000.11% Maintenance $2,500.000.63% Insurance $15,000.000.38% Building other costs, including trash hauling $9,000.000.11% Service contracts: elevator, heating $2,500.0011.11% Total facility costs $263,250.00

100.00% Total all costs $2,370,250.00

Financial NotesIncluding the cost of facilities, the debt service (both principal and interest), the contingency fund of three percent (approximately$67,000) as well as the cost of tutors, the average annual cost per MATCH student is $13,168. However, without all these extra itemsthe annual cost per student drops to approximately $9,111. The tutoring expense sustains the apparently successful rigorousacademic program. The tutors are paid by their respective colleges through a federally funded work-study program at hourly rates between$10 and $17 per hour. MATCH reimburses the colleges for any difference between the funded rate and the payout rate to the students,which costs MATCH about $5 per hour. For the approximately 30,000 tutor hours, estimated by Michael Goldstein, that MATCHengages each year the school pays out $150,000 but the actual value of those hours is approximately $405,000. The school also has a“books trip” program in which each student is allotted $100 per year to purchase books from the local Barnes & Noble. Once a monthstudents take a trip to the local bookstore to execute this program. The total annual cost of this program is $18,500.

The total cost of the facility is approximately $9,050,000, which includes construction, site acquisition, design, legal, FF&E, andtechnology. The building has a 50-year life beginning FY 2002 and thus the building associated costs average $180,000 per year. TheFF&E and technology were depreciated over an eight-year period, which amounts to approximately $6,250 per year. The amount ofthe budget going towards the instructional programs accounts for over 62% of the overall budget including facilities, but in the absenceof the facilities cost the instructional costs would account for almost 70% of the overall budget. The annual allotment to the school bythe City of Boston for each student is only $9,500, which can be adjusted downward and retroactively depending upon the economichealth of the city at the time. Interestingly, the money is actually paid directly to the school by the state and deducted from whateveroverall funds the state has appropriated for the city. MATCH would appear to have all of the ingredients of a cost-effective smallschool: a rigorous and apparently successful academic program, cost efficiencies, sustainable partnerships, a bright and newlyrenovated facility with built-in design flexibilities, and an entrepreneurial management team of teachers and administrators. Thestaffing consists of 12 full-time and eight part-time teachers, and approximately 10 staff members, including two tutoring directorswho oversee 100 tutors during the academic year and 60 during the summer, a principal, a CEO, and an executive director.

Year constructed: 1917, built forLincoln Motorcar CompanyYear of remodeling: 2001Total acreage: 10,000 sq. ft.Total constructed space: 30,000 sq. ft.Former uses: car dealership

ArchitectureF.A. Norcress

Facility-Related DetailsSquare footage of renovation: 22,500 Cost and year of acquiring building:$4,250,000 (2001)Cost and year of renovation:$4,000,000 (2001)Design and permitting fees: $750,000

Cost per square foot of renovation:$211.11Cost of FF&E: $55,000Cost of technology: $95,000

Building History


92

Metropolitan Regional and Technical Center -Public Street Campus

Providence, RI

ContactMs. Jill Olson-CrowleyMetropolitan Career and TechnicalCenter325 Public StreetProvidence, RI 02905(401) 752-2634(401) 465-0868 [email protected]

Website: www.bigpicture.org

Location: UrbanDistrict: Board of Regents SchoolsType of school: PublicGrades served: 9-12Number of students: in four separatefacilities: 320 (FY 2003); 720 (FY2006 capacity) NCES school ID: 4400070

Population served:White, non-Hispanic – 35%Black, non-Hispanic – 24.4%Hispanic – 33.6%Asian/Pacific Islander – 2.8%American Indian /AlaskanNative – 4.2%


All Costs



SALARIES

19.54% Salaries: instructional $1,222,156.0014.92% Salaries: staff support $933,220.0011.37% All staff benefits $711,274.0045.84% Subtotal $2,866,650.00


2.64% Professional development (including travel & conferences) $165,000.007.36% Curriculum materials including textbooks $460,540.000.83% Program support $52,000.0010.84% Subtotal $677,540.0056.68% Total instructional expenses $3,544,190.00


10.11% Administration salaries & benefits $632,086.001.60% Transportation (including bus tokens) $100,000.0011.78% Operations salaries & benefits $736,820.000.48% Internet services $30,000.0010.23% Business services $640,000.0034.21% Subtotal $2,138,906.0090.88% Subtotal non-facility costs $5,683,096.00

continued >


93


FACILITY COSTS

5.44% Contract maintenance $340,000.003.68% Utilities (electrical/gas/water) $230,000.009.12% Total facility costs $570,000.00

100.00% Total all costs $6,253,096.00

Financial NotesThe average cost per student is $14,276.47. However, the six schools are only at 62% of capacity. Even though there is awaiting list of more than 250 students, the plan is for a controlled build-up to full capacity over the next two school years. The budgetdoes not include nutritional expenses, which are estimated to be in the range of $327,000. Approximately a third of the nutritionalcosts are federally subsidized but the MET hopes to make the operation self-sustaining by offering catering to the community atcompetitive prices. The instructional portion of the budget as a percentage of the total would appear to be low when compared to otherschools in the study. However, the unique instructional model of the school places the leadership staff (six principals and the director)in an instructional or advisory role as well. The transportation costs include the cost of six buses and drivers who not only transportstudents to and from school but to and from their daily internship programs as well. The building maintenance and repairs, as well asgrounds and landscaping, are done by the builders under the terms of a unique contractual arrangement. An evaluation of the costeffectiveness of this $340,000 contract will be undertaken by year end. The MET was constructed from the proceeds of a $17.7million bond issue, a small portion of which is as yet unused. One of the unique aspects of the instructional model is that each teacheror advisor, as they are called, is allowed to spend $1,500 for a trip, activity, event, or materials – whatever he or she feels isappropriate. Few, however, take advantage of this allowance. Finally, although the school is only at 62% of capacity, when fullyoccupied the annual average cost per student will remain high in comparison to larger schools, most likely in the $12,000 to $13,000 range.

Year constructed: 2002 (opened) Total acreage: about 8 acres Total constructed space: 85,000 sq. ft.Former uses: none

ArchitectureConcordia, Inc.New Orleans Studio201 Saint Charles Ave., Suite 4314New Orleans, LA 70170Steven Bingler(504) 569-1818(504) 569-1820 fax

Facility-Related DetailsRenovation, new construction, oraddition: New constructionSquare footage: 85,000Cost to acquire site: $0.9 millionCost of Construction: $13.5 millionCost of FF&E: $2.3 millionCost of technology: $1.3 millionCost per square foot: $159 (includessite development and constructioncosts)

Building History


94

Oak Valley Elementary School Lincoln, NE

ContactDan Alberts, PrincipalOak Valley Elementary School3200 W. McKelvie RoadLincoln, NE 668524

Randy Wiegand (Treasurer) (402) 435-8510

Jean Nabity (Teacher) (402) 470-3151

Location: RuralDistrict: Oak Valley Public SchoolsType of school: PublicGrades served: K-8Number of students: 30NCES school ID: 3116470

Population served:White, non-Hispanic – 100%

Percent of students eligible for freeand reduced-cost lunch: 0

All Costs



SALARIES

53.02% Salaries: instructional $64,528.007.91% Benefits $9,629.0160.93% Subtotal $74,157.01


3.08% Purchased services $3,744.353.26% Supplies and materials $3,964.036.33% Subtotal $7,708.3867.27% Total $81,865.39


3.78% Office of the Principal $4,598.895.94% Board of Education: business expenses $7,230.190.77% Purchased services $943.110.14% Supplies and materials $174.500.25% Mileage to parents $307.2010.89% Subtotal $13,253.8978.16% Subtotal non-facility costs $95,119.28

FACILITY COSTS

17.28% Purchase services $21,034.081.97% Supplies and materials $2,400.492.59% Depreciation $3,150.0021.84% Total facility costs $26,584.57

100.00% Total all costs $121,703.85

continued >

Website: www.oakvalleyschool.com/index.htm


95

Financial NotesOak Valley's total annual budget amounts to approximately $122,000 or $4,965 per student. More than half (55%) of itsfunding comes from local taxes, chiefly property taxes, and another 45% from state funds. The balance comes from othernonrecurring miscellaneous state and city sources such as late payment fees, and a PTA offers occasional support to capital projectsor funds some student activities. Almost 70% (68.3%) of the budget is spent on regular instructional programs. In lieu of regularstudent transportation, i.e., buses, drivers, vehicle maintenance, insurance, gas and oil, etc. Oak Valley reimburses its parents via amileage allowance and thus its transportation expense for all FY 2001/2002 was $307. Unlike most public school systems, OakValley captures depreciation as an expense. Lastly, all other services, including special education needs, staff development, and pupilsupport services are purchased from outside sources and account for less than two percent of the budget.

Year constructed: 1920, 1,150 sq. ft. Year of addition:1970: 300 sq. ft., bathrooms andcubbies

1979: playhouse on playground

2001: 700 sq. ft. Built by parents,classroom and computer room

Total acreage: UnknownTotal constructed space: 2,150 sq. ft.Former uses: none

Facility-Related DetailsSquare footage of renovation: 700Cost and year of renovation: $35,000(2001)Cost per square foot of renovation:$50.00Design fees and permitting costs:$1,725Cost of FF&E: $470Cost of technology: $4,100

Building History


96

ContactDr. Mark Sorenson, DirectorThe STAR Charter School77 Leupp RoadFlagstaff, AZ 86004(928) [email protected]

Website: www.starschool.org

Location: RuralDistrict: Painted DesertDemonstration Projects, Inc.Type of school: CharterGrades served: K-8 (2004-2005)Number of students: 65NCES school ID: 0400213

The STAR Charter SchoolFlagstaff, AZ

Population served:White, non-Hispanic – 23%American Indian /Alaskan Native – 77%


All Costs



SALARIES

30.41% Salaries & benefits: instructional $144,869.0012.89% Salaries & benefits: administration $61,400.005.67% Salaries & benefits: student support $27,000.0048.96% Subtotal $233,269.00


0.36% Professional development $1,726.002.73% Curriculum materials and supplies $13,000.002.32% Materials, classroom supplies $11,051.003.08% Contracted/consultancy classroom related $14,675.350.67% Support services $3,200.001.41% Miscellaneous $6,713.0410.57% Subtotal $50,365.3959.54% Total instructional expenses $283,634.39


0.21% Professional services $1,000.005.69% Transportation $27,125.001.78% Parental reimbursement $8,500.000.04% Supplies & postage $200.005.68% Student meals $27,075.000.10% Telephone $500.001.85% Miscellaneous $8,810.0015.37% Subtotal $73,210.0074.90% Total non-facility costs $356,844.39

FACILITY COSTS

3.57% Facilities manager $17,000.005.54% Rent $26,400.00

continued >


97


0.95% Power system upgrade $4,532.001.55% Custodian costs $7,400.001.72% Utilities (propane, disposal, & water) $8,200.002.46% Maintenance $11,700.003.04% Insurance $14,500.003.22% Building other costs, greenhouse, concrete $15,350.003.04% Consultant services (well project) $14,473.6325.10% Total facility costs $119,555.63

100.00% Total all costs $476,400.02

Financial NotesSTAR’s total per-pupil allocation is $276,450 or an average of $5,026 per pupil for the 55 pupils enrolled in 2003-2004 at the school.However, because the majority of its pupils are from the Navajo reservation (approximately 85%), the school is entitled to apply forImpact Aid from the federal government (reserved for students on military bases and reservations). With some overlap from 2003, thisaid amounted to $143,892 in FY 2003-2004. The total budget for STAR including entitlements from the state and county, the impactaid, and per-pupil allotment amounted to $474,400.02 or $8,661 per pupil. Approximately 60% of the total budget is devoted toinstructional aid. There are, however, a number of unusual expenditures in the budget, e.g., the school is powered by solar power and aportion of the budget, $9,532, was devoted to upgrading the solar system. The school does not have access to sewer and water; it uses aseptic system and has water delivered every two weeks. The monthly water consumption is approximately 10,000 gallons at an annualcost of approximately $3,500. The budget includes a similar amount ($3,223) to hire a consultant to survey for and write a grant for awell. The Impact Aid also allowed the school to erect a canopy over the playground, install insulated shades, and build a facility storageshed, all for $4,750. Other construction projects, including a greenhouse, add another $5,600. There is, therefore, approximately$25,000 of capital spending included in the budget or $455 per student. Total facilities cost of $119,556 accounts for 25.1% of theoverall budget. The biggest piece (5.5%) of the facilities cost is the rent of $26,400 the school pays to lease the land from the Sorensens.In addition to a facilities manager, the school employs one full-time and one part-time custodian, and one part-time maintenance person.The STAR campus sits on 40 acres which provides ample room for expansion. Transportation is an issue because the students live atextreme ends of the reservation. Travel time averages an hour or more for the students who are transported via the school’s one bus andtwo vans. The school spends approximately $36,000 on transportation, including $8,500 reimbursing parents for transporting theirchildren. The school employs three drivers. The school also provides breakfast and lunch (free, reduced-price, and paid) and spendsapproximately $27,000 providing these meals to the students (only half of that is reimbursed). An eighth grade was to be added inSeptember 2004 and bring the total enrollment closer to 70 students, which is on or about breakeven for the school. The per-studentaverage annual spending of $8,661 is understandable given the size of the facility and complexity of the program and campus. Moreover,the addition of the well, the building of a permanent greenhouse, the enhancement of the solar system (adding another 30 panels), andthe addition of the eighth grade should drop the average annual cost to below $7,500 per pupil.

Year constructed: 2001Total acreage: 40Total constructed space: 5,784 sq. ft.Former uses: site was a dump for derelictcars and appliances that became a hangoutfor people drinking and using drugs.


Facility-Related DetailsRenovation, new construction, and/oraddition: Three buildings (one newmodular, one completely renovated,and one under construction)Square footage: 5,784Cost to acquire site: $45,000 Cost of construction: three buildingstotaling $400,000 (not includingsidewalks and playground but includingsolar electric and water storage)

Cost of FF&E: $46,700 (kitchenequipment, $27,000; classroomfurniture, $15,000; office furniture andequipment, $4,700)Cost of technology: $48,000Cost per square foot: $69.16

Building History


ContactMs. Kim BaldwinCommunity and Fund DevelopmentTSOTA Partners601 South 8th StreetTacoma, WA 98405(253) [email protected]

Website:www.tacoma.k12.wa.us/schools/hs/sota

Location: UrbanDistrict: Tacoma School District 10Type of school: PublicGrades served: 10-12Number of students: 348NCES school ID: 5308700

Tacoma School of the Arts (TSOTA)Tacoma, WA

Population served:White, non-Hispanic – 82.6%Black, non-Hispanic – 8.3%Hispanic – 3.9%Asian/Pacific Islander – 3.6%American Indian /Alaskan

Native – 1.6%Percent of students eligible for freeand reduced-cost lunch: 13%

All Costs



SALARIES

47.58% Salaries: instructional $861,863.001.45% Salaries: administration $26,250.001.13% Salaries: student support $20,500.001.93% Salaries: office staff $35,000.0052.10% Subtotal $943,613.00


1.55% Professional development (including travel & conferences) $28,000.004.76% Departmental expenses (including equip) $86,250.003.62% Departmental expenses: vocational (including equip) $65,500.000.55% Instructional services, lecturers $10,000.0010.48% Subtotal $189,750.0062.57% Total instructional expenses $1,133,363.00


5.22% Transportation $94,500.000.83% Parking $15,000.003.42% Equipment rental, repair, & leases $62,000.002.13% Marketing and public relations $38,500.0011.59% Subtotal $210,000.0074.17% Total non-facility costs $1,343,363.00

FACILITY COSTS

25.83% Building lease $467,900.0025.83% Total facility costs $467,900.00

100.00% Total all costs $1,811,263.00

continued >98


99

Building History

Financial NotesThe budget for the Tacoma School of the Arts 2003-2004 is based on a student count of 360 and amounts to $1,811,263. It comesfrom two funding sources: the district, which includes a separate vocational allocation, and the 501(c)(3) partnership (corporation),receiving grants and funds from partners and many other sources. The 501(c)3 corporation accounts for $324,063 or 17.9% of thetotal budget. The average annual spending per pupil amounts to $5,031. However, there are certain costs that are notcaptured or reported against the school’s budget. The cost of utilities, for example, is not part of the school’s budget because the entiredistrict is on a statewide grid and is billed for the total amount by the utility company. Also, textbooks, which are provided to TSOTAby the district, are not included.

The TSOTA program is spread over 10 venues, six of which receive lease payments. The total combined amount of those leases is$467,900 or 25.83% of the total budget. The two main buildings, the visual arts building and the performing arts building, accountfor the bulk of the lease cost (95.1%) and are funded by the district. One of the 10 venues, the “1818” humanities building, is inconstant use by TSOTA and is owned by the district. TSOTA’s budget is not charged for this use. Approximately half ($75,000) ofthe $151,250 devoted to program spending (basic education courses) is funded by the 501(c)(3) corporation. All of these funds areused for supplies, materials, and equipment needs for the courses. All of the $62,000 in the budget for computer leasing and databasesystems is provided by the 501(c)(3) corporation. The 501(c)(3) corporation also funds the pre-school summer camp ($32,000) andall marketing and public relations. The staffing budget includes provisions for 20 full-time teachers, including career and technicaleducation, 10 part-time adjunct artists, one secretary, and one director. The staffing budget also provides for science class assistantsand a visual arts consultant. The two other items in the budget that stand out are the expense of $15,000 for staff parking becausethe venues are downtown and $94,500 for student transportation, both of which are funded by the district. The students use publictransportation and are provided with a monthly pass that costs $27 per student per month for their 10-month school year.

Year constructed: New York Building,1906; Ted Brown Building, 1908; alsorenovations in downtown publicbuildings

Year of remodeling: 2001Total constructed space: remodeled inthese two buildings, 20,000 sq. ft.Former uses: department stores, TedBrown music store, Arthur MurrayDance Studio

ArchitectureMcGanahan Architects2111 Pacific, Suite 100Tacoma, WA [email protected]

Facility-Related DetailsVisual Arts CenterSquare footage of renovation: 20,000Cost of acquiring building: Leased for$1.00 per sq. ft. (with option to buyafter five years)Cost of renovation: $400,000Cost per square foot of renovation:$20.00 (renovated by district withtheir work crew to reduce costs)Cost of FF&E: furniture donated orbuilt

Performing Arts Center: Ted Brown BuildingSquare footage of renovation: 20,000Cost of acquiring building: Leased(with option to buy after five years)Cost and year of renovation: $1.8million (2003)Cost per Square Foot of Renovation:$90.00

TSOTA also uses space in severalbuildings in downtown Tacoma.


100

ContactMs. Jane Rusten, Coordinator ofOffice OperationsTodd Beamer High School35999 16th Avenue SouthFederal Way, WA 98003(253) 945-2570(253) 945-2599 fax

Website: http://schools.fwps.org/tbhs/

Location: UrbanDistrict: Federal WayType of school: High SchoolGrades served: 9-12Number of students: Divided intothree academies: 1350 (2003-2004),1550 (2004-2005)

Todd Beamer High SchoolFederal Way, WA

Population served:Unknown (new schools)

Percent of students eligible for freeand reduced-cost lunch: 32%NCES school ID: Not yet listed.

All Costs


INSTRUCTIONAL PROGRAMS

62.77% Basic education $5,997,517.00 8.29% Special education $792,131.00 9.54% Career & technical education $911,629.00 0.22% Community services $21,263.00 13.13% District-wide education $1,254,231.00 2.83% Other compensatory education $270,000.00 96.78% Total educational programs $9,246,771.00

STUDENT SUPPORT

2.20% Nutritional services $209,893.00 1.02% Pupil transportation $97,791.00 3.22% Total student support $307,684.00

100.00% Total all costs $9,554,455.00

Financial NotesTotal number of students is 1,416 with an average cost per student of $6,747. This is lower than the district’s average cost perpupil of $7,377, for the most part, because the total district spending for compensatory education is $15,654,971, of which Beamerreceives only $270,000.


101

Year constructed: 2000; opened fall 2003Total acreage: 37.67 acresTotal constructed space: 162,000 sq. ft.

ArchitectureBassetti Architects71 Columbia Street, Suite 500Seattle, Washington 98104(206) 340-9500(206) 340-9519 faxCoordinator: [email protected]@bassettiarch.comwww.bassettiarch.com/default.htm

Facility-Related DetailsSquare footage: 162,000Cost to acquire site: $3,300,000(2000)Cost and year of renovation:$27,000,000 (2001-2003)Design and permitting fees:$4,330,000Cost per square foot: $167Cost of FF&E: $3,640,000Cost of technology: Included in theFF&E except an additional $420,000for servers and telecommunicationsroom.Great Hall is 6500 sq. ft.

Building History


102

ContactMs. Judy Kraft, PrincipalH.S. Truman High School31455 28th Avenue SouthFederal Way, WA 98003(253) 945-5800

Website:http://schools.fwps.org/truman/

Location: UrbanDistrict: Federal Way School District 210Type of school: Public AlternativeHigh SchoolGrades served: 9-12Number of students: Forum: 99 (2004-2005); South: 98 (2004-2005)NCES school ID: 5302820

H.S. Truman High SchoolFederal Way, WA

Population served:White, non-Hispanic – 52.3%Black, non-Hispanic – 19.9%Hispanic – 10.9%Asian/Pacific Islander – 10.4%American Indian /AlaskanNative – 6.5%

Percent of students eligible for freeand reduced-cost lunch: 35.3%

All Costs


INSTRUCTIONAL PROGRAMS

61.66% Basic education $853,058.00 11.33% Special education $156,776.00 8.91% Career & technical education $123,331.00 0.22% Community services $3,024.00 12.75% District-wide education $176,396.00 3.21% Other compensatory education $44,448.00 98.08% Total educational programs $1,357,033.00

STUDENT SUPPORT

0.31% Nutritional services $4,284.00 1.61% Pupil transportation $22,275.00 1.92% Total student support $26,559.00

100.00% Total all costs $1,383,592.00

Financial NotesThe total number of students is 201 (in two schools, 2003-2004) with an average cost per student of $6,893. The budget forall schools in the Federal Way School District is established via a staffing formula based on the number of students enrolled. Ingeneral, schools in the district are entitled to a specific allocation of core teachers, secretaries, administrators, etc., or comparableFTE, for every 26 students. Truman is allowed a “sweeter” allocation formula, using a student-to-teacher ratio of 17 to 1. Thisformula compensates for the fact that Truman does not receive $16 million for sports programs and facilities, a cafeteria, ortransportation because it does not offer these services.


103

Building History:Year constructed: 2003Site acquired: 1942 for $10,000Total acreage: 9 acresTotal constructed space: 23,300 sq. ft.Former uses: another school

Site Plan: www.designshare.com/Floor Plan: www.designshare.com/Elevations: www.designshare.com/Clusters: www.designshare.com/Images: www.designshare.com/

ArchitectureMahlum ArchitectsDavid Mount 71 Columbia, Suite 400Seattle, WA 98104(206) 441-4151

Facility-Related DetailsSquare footage of new construction:23,300Cost of acquiring site: N/ACost and year of new construction:$3,983,000 (building 2002-2004)

Cost per square foot of construction:$171 (building and site) $113(building only)

The building and site were quite costeffective. There was no need to buildgyms or fields due to communitypartnerships. The total building cost(2002 dollars) was $1,791,010 or$76.76 per square foot ($1,880,561total or $80.60 per square foot in 2004dollars). The site costs were more,because the school district providedsite prep, utilities, and parking for twocommunity partners.The total site costwas $2,111,890 in 2002 dollars, or$2,217,485 in 2004 dollars.

The DesignShare website providedthe following figures: Cost per square foot: $117 Completion date: 2003Square footage: 22,330 Cost: $2,620,000www.designshare.com/

Building History


104

ContactSteve Orel, DirectorWorld of Opportunity7429 Georgia RoadBirmingham, AL 35212-2921(205) 271-9532(205) 592-3725 fax

Website:www.worldofopportunityWOO.org

Location: UrbanDistrict: UnknownType of school: Private nonprofitGrades served: Alternative schoolserving adolescents, young adults, andadults, as well as one class ofelementary, middle, and high schoolstudents.Number of students: between 35 and65 each day NCES school ID: NA

World of OpportunityBirmingham, AL

Population served:Adolescent and adultPercent of students eligible for freeand reduced-cost lunch: More than99% of students were eligible forfree and reduced-cost lunch at theirprevious schools.

All Costs



SALARIES

55.64% Salaries: instructional $64,031.399.11% All staff benefits $10,479.6464.75% Subtotal $74,511.03


2.10% GED expenditures $2,422.103.29% Curriculum materials, including textbooks $3,785.317.58% Capital outlays $8,725.0012.98% Subtotal $14,932.4177.73% Total instructional $89,443.44


3.48% Professional services $4,000.002.03% Depreciation $2,340.001.45% Direct assistance $1,664.351.50% Employee reimbursable expenses $1,730.271.63% Supplies & postage $1,871.471.53% Other miscellaneous services $1,764.4011.62% Subtotal $13,370.4989.34% Subtotal non-facility costs $102,813.93

FACILITY COSTS

7.04% Building & trailer lease $8,098.820.82% Maintenance $941.072.80% Insurance $3,222.9610.66% Total facility costs $12,262.85

100.00% Total all costs $115,076.78

continued >


105

Financial NotesIncluded in the total annual cost of running the school is in-kind services and donated items. The World of Opportunity (WOO) hadout-of-pocket expenses of $80,149.74 for reporting year September 30, 2002 to September 30, 2003. In-kind volunteer teacherhours were the equivalent of $23,202 in salary dollars; equipment and textbook donations had a market value of $11,725. Theschool incurs an expense category unique to its mission and student population: direct assistance (food, rent, car fare, medical, etc.,dispensed in emergencies to students as needed) which accounted for $1,664.35 of its budget. While the school’s daily studentpopulation currently averages 45, due to the nature of the program it actually serves approximately 150 discrete students and if lookedat in that context, the average annual cost per student amounts to $767. Capacity and repetitive funding sources are its mainhurdles because it receives no public funding and has reached capacity even with the addition of a leased trailer.

Year constructed: 1974 Year of addition: Unknown

Total acreage: UnknownTotal constructed space: 2,235 sq. ft.Former uses: meat market,convenience store, and electricalcontractor’s shop before beingconverted for use by WOO.

ArchitectureNo architect involved.

Facility-Related DetailsRenovation, new construction, oraddition: rented concrete masonrybuilding attached to a trailerSquare Footage: 2,235 (1,850 forcinder block building, remainder fortrailer)Cost to lease building: $730.90 permonth (includes utilities but notbottled water or maintenance; manyin-kind donations) Cost of construction: $3,000 (lessthan $2,000 to renovate first year, plus$1,000 second year; many in-kinddonations here)Cost of FF&E: 90% (or greater) ofthe furniture, fixtures, and equipmenthave been donated. (The copymachine, including maintenancecontract is included with the rent.)Cost of technology: $70 per month forInternet (less than $5,000 this pastyear for hardware and software, lessthan $1,000 per year previous years;many in-kind donations in thiscategory)Cost per square foot: $1.34

Building History


106

The analysis explored two questions:

– Are planned costs of construction equal for

smaller as compared to larger schools?

– How do variables relating to the schools and

their contexts affect construction costs?

To answer these questions the research team worked witha database of 3,471 school construction projects compiledby Paul Abramson for School Planning and Managementfrom information provided by School Construction Alert,a Dun and Bradstreet Company. This database purportedto be new schools begun between 1995 and 2003 andscheduled for completion between 1996 and 2009.2 Asummary of the quantitative work follows, but a fulldiscussion of the methodology and findings is available onthe companion website at www.goodsmallschools.org

Challenges

Researchers faced two serious problems. First, theinformation contained in the database was originallycollected for people who hope to sell products andmaterials to schools. It contains only eight variables andlittle identifying information, making it difficult to link tothe demographics of students, poverty levels of theirfamilies, level of parents’ educational attainment, andother important variables. Without these variables, thislarge database provided only limited information forinvestigation.3 Second, very little peer-reviewed researchliterature exists on school construction costs and theirrelation to school size. This is particularly surprising giventhe amount of money spent on school construction andthe potential impact of school facilities on students,teachers, and communities. This study and future workwill begin to fill this significant information void.

Methodology

Trimming the Data SetTo reduce error, the team removed the top and bottomfive percent of cases in the large database. Extreme valuesin key derived variables such as cost per square foot areprobably results of recording or reporting errors (includinga project other than new construction, like a repair oraddition, for example). Dropping cases with extremevalues – a common research practice – reduces likelyerror, producing a much more believable range of costs persquare foot and costs per student.

Need to Link to More VariablesThe trimmed data set of elementary, middle, and highschools (n: 3,124) still contained only eight variables. Todeepen the analysis possible with the data, the researchteam needed to link it to demographic information. TheCommon Core of Data (CCD) is the most complete,credible, and up-to-date information available aboutschools in the United States.4 Linking information fromthe CCD to the data set on construction figures wouldallow the team to ask more probing questions than waspossible with the larger set of data.

Augmented Data Set To link the construction data to the CCD, the researchteam created a subset drawn from the trimmed data setthat included only “reasonably sized” high schools —schools under the team’s threshold for a “mega-school” atthe high school level (see box on next page). The teamselected high schools because they are less numerouswithin any given district, which improves the odds of anaccurate match to the CCD; additionally, high schoolshave been a focus of contention about school size. Tofurther reduce the possibility of an incorrect match, thisdata set included schools from districts with only one highschool. The team then matched cases in the two data setsand called all schools to determine the accuracy of thematches. This careful work produced a subset of 211 highschools that could be linked with comparative confidence

Appendix 2 • School Facility Construction Analysis Summary

2 Paul Abramson, author of School Planning and Management’s“Annual Construction Report,” and a member of the Dollars &Sense team, made this data set available.3 There are only eight variables attached to the data set: state, zipcode, educational level (elementary, middle, or high school),total cost, planned enrollment, planned square feet, start year,and end year.

4 Common Core of Data produced by the National Center forEducation Statistics. See: www.nces.gov.


107

to the variables in the CCD. The team used this smaller,richer data set to conduct more extensive analyses thanwas possible with the large data set.

Locale Differences Associated with “Reasonable Size” Mega-schools (high schools enrolling more than 1,000students) are much more common in metropolitan areasthan in rural areas. In the United States in 2001-2002,6,795 high schools enrolled one thousand or fewerstudents, 50.8% of them in rural locales and only 8.8% ofthem in large or mid-size cities. These proportions closelymatch the proportions in the augmented data set. Thissimilarity supports generalization of findings from theaugmented data set to the nation as a whole. It alsodemonstrates that the balance of rural cases in theaugmented data set reflects the existing condition, not abias introduced by the methodology for selecting cases.

Assumptions About Size and CostsAs we have stated in this report and in Dollars & Sense:The Cost Effectiveness of Small Schools, empirical researchand the experience of educators, students, and parentssuggest that very large schools negatively affect teachingand learning. Students, teachers, parents, andcommunities suffer when schools are too large. Schoolsize, moreover, dramatically affects the academicachievement of impoverished students. Minority studentsare disproportionately represented in poor communitiesand in schools that are too large, particularly in large

SCHOOL SIZE5

Definitions

Appropriately small: refers to the smaller schools inthe data set of reasonably small schools.

Mega-schools: refers to all schools with a plannedenrollment above the median for schools larger thanreasonable size.

Size Limits

Small Schools (recommended upper limits)6

Elementary Schools: (1–8): 25 students per gradelevel (200 total enrollment)

Elementary Schools: (1–6): 25 students per gradelevel (150 total enrollment)

Middle Schools: (5–8): 50 students per grade level(200 total enrollment)

High Schools: (9–12): 75 students per grade level (300 total enrollment)

Reasonably Sized Schools (upper limits)

Elementary Schools: up to 500 studentsMiddle Schools: up to 750 studentsHigh Schools: up to 1,000 students

Mega-Schools:

Elementary Schools: 700 students and aboveMean in this study is 900 students

Middle Schools: 1,000 students and aboveMean in this study is 1,300 students

High Schools: 1,000 students and aboveMean in this study is 2,150 students

DESCRIPTION OF DATA SETS

Untrimmed Data Set: 3,471 schools.The data set from School Planning and Management isdescribed as containing only new schools begunbetween 1989 and 2003 and scheduled for completionbetween 1996 and 2009. Researchers did not analyzethis data set because very low and high values seemedto have been entered in error.

Trimmed Data Set: 3,124 schools. Researchers created a data set by trimming the upperand lower five percent of cases from the untrimmeddata set to reduce the number of projects included inerror.

Augmented Data Set: 211 schools. Researchers created this small data set containingonly high schools so they could link it to the NationalCenter for Education Statistics Common Core ofData. The Common Core of Data contains the mostup-to-date and complete information on schools inthe United States, and linking the augmented data setto this information gave researchers many additionalvariables to analyze.

5 As defined and used in Dollars & Sense: The CostEffectiveness of Small Schools.6 As defined and used in Dollars & Sense: The CostEffectiveness of Small Schools.


108

Findings of Quantitative Analysis The first question the research addressed is: Are plannedcosts of construction equal for smaller as compared tolarger schools?

The analyses conclude that:

• Cost per square foot is no higher in smaller than inlarger schools. In fact, cost per square foot issignificantly and consistently lower for smaller highschools, whether smaller and larger is defined over thefull data set, the reasonably sized schools, or forappropriately small – the smaller group of “reasonablysized” schools – as compared to “mega-schools.” Mega-schools at the high school level are the most expensiveper square foot of all categories considered.

• Cost per student is always higher in smaller ascompared to larger schools, whatever the grade range,with one exception – smaller middle schools in thereasonably sized group are no more costly per studentthan larger reasonably sized schools. Mega-schools atthe elementary and high school levels allocated lessthan half the square feet per student of appropriatelysmall schools. This difference consistently and clearlyrenders cost per student lower in larger schools ascompared to smaller schools, much lower in the case ofmega-schools.

• Square feet per student is consistently higher amongsmaller schools (i.e., for all levels in both schoolgroups). This allocation of space explains the tendencyof smaller schools to cost more per student. It alsounderscores the importance of a finding from thequalitative research that forming partnerships withinthe community for sharing library, cafeteria, and athleticspace reduces the size and cost of a small school and/orspreads cost of construction more evenly throughout thecommunity.

cities. Yet, the view persists that large schools are cheaperto build, maintain, and operate. This assumption becomesparticularly worrisome when poor communities face thechallenge of building a school because they are oftenforced to build schools that are far too large to serve theirstudents well.

Put differently the question becomes “Are the savingsgained by constructing large schools so great that theyjustify creating an educational facility that can’t support agood educational program?” In light of the knowledgethat large schools produce negative results, includinghigher dropout rates, lower graduation rates, and increasedviolence and vandalism, with consequent high costs tosociety, the answer is clearly NO! Even though we believedistricts should not build large schools, the research teamlooked at the costs of building small and unreasonablylarge schools using the large trimmed data set and thesmaller augmented data set.

Next the research team considered a more importantquestion: Is the cost of building smaller reasonably sizedschools significantly higher than building larger schoolsthat are still reasonably sized? It is within these parametersthat policy-makers, educators, and voters should makedecisions about locating and sizing schools, so thisresearch looks most deeply at the costs of constructingsmaller versus larger reasonably sized schools.7

7 The team used the definition of reasonably sized suggested inDollars & Sense: The Cost Effectiveness of Small Schools:enrollments less than 500 for elementary, 750 for middle, and 1,000 for high schools.


109

Median size for all schools by level and comparison group

MedianLevel Comparison Group N in Group (mean) Enrollment

Elementary All schools (trimmed data set) 1,681 600 (610)Reasonably sized schoolsa 609 400 (381)Mega-schoolsb 520 800 (872)Appropriately small schoolsc 320 300 (295)

Middle All schools (trimmed data set) 720 750 (775)Reasonably sized schoolsa 371 575 (537)Mega-schoolsb 116 1,200 (1292)Appropriately small schoolsc 187 424 (408)

High School All schools (trimmed data set) 723 912 (1063)Reasonably sized schoolsa 407 567 (584)Mega-schoolsb 134 2,000 (2157)Appropriately small schoolsc 204 400 (373)

Notes. The comparison groups used in the study are related to one another statistically and are drawn from the large, trimmed data set(n=3,124).

a Reasonably sized schools are those schools in the data set that are below the upper limits of size endorsed by the Dollars & Sense team: 500,750, and 1,000 for elementary, middle, and high schools, respectively. The study compared costs for these groups divided into two groups:those above the median for this group (larger reasonably sized schools) and those below the median for this group (smaller reasonably sizeschools). The augmented data is composed only of reasonably sized high schools.

b Mega-schools, as defined in the study, are those at each level that are larger than the median size for “unreasonably sized schools” – for allthe schools that are not “reasonably sized” in the trimmed data set. These are very large schools indeed – far too large in the view of theDollars & Sense team. These medians (which define the lower limits of enrollment for identifying mega-schools) are 700, 1,000, and 1,600for elementary, middle, and high schools, respectively. The average mega-school is, of course, substantially larger than these minimumvalues for the groups (e.g., 2,157 for high schools in this group).

c Appropriately, small schools are those below the median size for “reasonably sized schools” at each level. These schools, in the view of theDollars & Sense team, are close to the right size range for impoverished communities; such communities, however, often have “mega-schools,” which is a very common problem in urban areas. The team emphatically does not recommend the average enrollments of schoolsin this group as ideal or optimal.

d Medians in this data set do not precisely divide groups into two equal halves because planned enrollments, as estimates, cluster into round-number values (e.g., by 10s or 50s).


110

Discussion

Costs for Mega-SchoolsComparing small and very large schools in the trimmeddata set shows that the largest schools reduce the squarefootage per student. Because square feet of constructedspace is the most important predictor of cost, limitingspace per student is the most efficient way to reduce cost.The tendency in impoverished communities, andespecially such communities in urban areas, has been toconstruct large schools, enrolling too many students, witha reduced number of square feet per student. This mayreduce the short-term costs of construction, but is likelyto increase long-term costs to society that develop as aresult of students who drop out and do not graduate fromhigh school (Dollars and Sense I pp. 8 - 17).

Future research should analyze how larger schools are ableto allocate less space per student than smaller schools do.Does the reduced space affect areas used by students suchas cafeterias and classrooms or does it affect areas used bysupport services and administration?

Though mega-schools are comparatively rare,8 they aremore likely to exist in urban places and are far morecommon in rural areas than in small towns.9 In the largetrimmed data set 520 very large elementary schools (morethan the number of K-6 mega-schools in existence in2001-2002) were planned for construction, but theprojected number of very large middle and high schools isalso disturbing. In fact, the trends suggest that planningfor high school construction in 1989-2002 was morereasonable with respect to size than for middle schools,and much more reasonable than planning for elementaryschools, which is alarming.

Costs for Reasonably Sized SchoolsThe research showed that with respect to costs, thesmaller schools in the reasonably sized schools set are lessexpensive to build than the larger schools per square foot,and that they are no more expensive per student thanlarger schools. The smaller schools in this data set actuallyallocated 26% more space to each student as compared tothe larger schools. In fact, the data suggests that smaller ascompared to larger reasonably sized 9-12 high schools(those enrolling no more than 1,000 students) can be lesscostly to build, which is demonstrated in one test persquare foot and per student in another.

In analyses based on planned enrollment of 9-12 schools,cost per square foot was significantly less (by about $15

per square foot) in smaller schools.10 In analyses based onactual subsequent enrollment of 9-12 schools, cost perstudent was significantly less in smaller schools (by about$1,300 per student).11 These results are consistent withearlier work by Paul Abramson as reported in Dollars &Sense: The Cost Effectiveness of Small Schools and showthat smaller reasonably sized high schools are no more,and probably less, costly to build than larger reasonablysized high schools.

The second question the research addressed is: How doconstruction costs vary in light of school and contextvariables, including planned enrollment, square footage,district revenues, locale, socioeconomic conditions, andethnicity?

Using a series of regression analyses, the researchersstudied nine predictor variables:

1. total planned square feet,2. planned enrollment,3. average district enrollment for 1997-2001,4. average district revenue per student for

1997–2001,5. proportion of adult population 25 and older who

were high school graduates in 2000,6. percent of district families living with children

above the poverty line in 2000,7. percentage of students in the school who were

African-American,8. planned completion date, and9. location (urban or rural).

Because there is so little empirical research on either theeffects of size on school construction or on constructioncosts for schools, it was difficult to hypothesize

8 In 2001-2002, the nation maintained 9,881 regular-educationpublic K-6 schools, but just 466 of these enrolled more than 900students, most of them (n=332) in large cities and the exurbanand suburban fringe of large cities. Similarly, there were 11,3859-12 high schools, but just 921 enrolled more than 2,150 stu-dents, and 787 of these were in large cities. (All data from the2001-2002 Common Core of Data.) 8 Mega-schools are more common in rural areas than in smalltowns by a factor of about 20.10 Because planned enrollment often changes between the time afacility is designed and opens for the first day of school, actualrather than planned size is a more accurate variable.11 This result underscores the fluctuations in both these measure-ments that come about because enrollment fluctuates, and under-scores the point that a more permanent measurement such asamortized value over lifespan should be used to track costs andvalue of a school facility.


111

relationships in terms of cost per square foot, cost perstudent, or of square foot per student, though researchwith the trimmed data set shows that a large proportion oftotal cost can be predicted by total square feet. Squarefootage is very important, however, because the way largeschools realize cost efficiencies, to the extent that they doso, is by reducing the amount of space allocated perstudent. In fact, as mentioned earlier, larger schoolsallocate about half as much space per student asappropriately small schools do (i.e., in the smaller groupof the “reasonably sized schools” group).

Predicting the effect of variables on total cost is lessproblematic, however. Square feet and enrollment are theprincipal cost factors in any construction project, andlocal revenues, education level in the community, andpercent of minority or low-income population affectdecisions about what kind of schools to build. Plannedcompletion date affects costs because facilities constructedlater will generally cost more than those built earlier,barring recession or other unexpected economicconditions. While all these variables tend to increase cost,the number of planned square feet is the most important.In declining order of influence, the following variablespredicted total cost for 9–12 high schools: footprint (totalsquare feet), inflation (construction end date), localrevenue wealth, rural location, and proportion of familiesin living in poverty. Only location in a rural communitytends to reduce the cost of construction.

Not surprisingly, the study concludes that local affluencedrives up the cost of a school per student and povertydrives it down. Because of the assumption that buildingand operating very large schools is cheaper than buildingand operating much smaller schools, impoverishedcommunities are unlikely to build appropriately smallschools, and yet they are precisely the places that needthem most. The irony is that small schools are costeffective not only because they offer students, families,and communities advantages in program andpersonalization that are not possible in very large schools,but also we now know that they are cost effective to buildin comparison to larger schools. The largest high schoolsare the most expensive to build per square foot and offerstudents, families, and communities the least return oninvestment in relation to the benefits associated withsmaller schools.

GeneralizabilityFirst, the augmented subset of high schools representsschools in 30 states, but the proportions of schools bylocale matches the national distribution. These factssuggest that results can be generalized, though cautiously,to the nation as a whole. Second, costs per square foot inthis data set are consistent with those reported byAbramson for 2003, with a national median cost of $108per square foot. Third, and very importantly, therestrictive selection rules (unique zip code and highschool unique in the district) did not produce sharplydifferent means on cost. These facts, too, support somedegree of national generalizability.

Error can creep into any work no matter how vigilant theresearchers are in creating systems to reduce or eliminateit. The Dollars & Sense team was diligent in implementinga methodology that reduced error. A complete explicationof their work and the results is on the KnowledgeWorksFoundation website at www.goodsmallschools.org.

Conclusion of the Quantitative AnalysisThe quantitative analyses demonstrate that schools thatare far too large to educate students effectively have beenbuilt more cheaply than smaller schools, primarily becausethey include fewer square feet per student. Thesubsequent costs to society of these schools, however, asdescribed in Dollars & Sense: The Cost Effectiveness ofSmall Schools, are enormous.

Most importantly, this research shows that small schoolsare being constructed at a cost similar to or less thanthose that are still within the limits of reasonable size butare significantly larger. This important finding shouldencourage educators, policy-makers, and taxpayers tosupport construction of smaller schools because educationresearch has shown they are the most effective ineducating students, particularly poor and marginalizedstudents. Now, for the first time, solid researchdemonstrates that the facilities for smaller schools are nomore expensive to build than larger reasonably sizedschools.


112

Purpose

The following analysis compares the test scores of studentsin most of the schools in this report against theperformance of students in larger schools in the samedistrict that appear to serve comparable populations. Thepurpose of this analysis is to determine whether studentsin the smaller schools are achieving academic results thatare on par with those of students from larger schools. Thefindings show that in most cases, even though many ofthe smaller schools have only been open for a few years,their students are achieving significantly higher academicsuccess than their peers in larger schools.

Note: There are many reasons that might explaindifferences in scores. This analysis does not attempt toidentify or explain the differences. The goal is simply toshow how academic performance of students in the smallschools in this report compares to students in comparableschools.

Guidelines

Because there is no national test that all students take, itis impossible to compare performance across states. Statesuse different tests and some make more informationavailable about the results than others. For the most part,the review compares English/language arts and mathscores because these measure fundamental skills. Thefollowing guidelines governed the comparison of theperformance of students in the schools in this report.

1. Wherever possible test scores for each of theschools in this report are compared to a largerschool within its zip code and district that serves asimilar population.

2. When a school constitutes the district, as is thecase with Avalon Charter School, the comparisonis to a school in the same zip code or nearby areathat seems to serve a similar population.

3. In a case, such as Laurel-Concord Public Schools,when there is no nearby school, the comparison isagainst district or state scores.

Some schools do not have reportable scores for thefollowing reasons.

1. The only scores available for the New York Citytests are for the year 2002-2003 and do notinclude figures for the Bronx schools in thisreport. As soon as the scores are available, theywill be posted at www.goodsmallschools.org

2. STAR school and Oak Valley Elementary Schoolare too small to report results withoutcompromising individual students, so no scores areavailable.

3. WOO serves students who have dropped out ofhigh school and do not take state tests.

Results

A table for each school is posted on the website with theweb address for the state department of education thatsupplied the information.

Summary

Avalon Charter High SchoolNo students at Avalon scored in the lowest two categorieson the reading test for the Minnesota ComprehensiveAssessments (MCAS), while 27% of students at CentralSenior High School were in those categories. A higherpercentage of Avalon students scored in levels 3 and 4than students in Central High School; however, therewere more students from Central who scored in thehighest level, level 5. Avalon students did not fare as wellon the math test as they did on the reading test,particularly at the highest level; however, their scores inlevels 1-4 are very similar to scores achieved by studentsfrom Central Senior High.

Bronx Small SchoolsScores for the four small schools in the Bronx are not yetavailable on the Department of Education website. Oneindicator of their success, however, is that students arecoming to school more often and fewer are dropping out.The website will add the data as soon as it is available.

Appendix 3 • Test Scores for Schools in This ReportCOMPARISON OF TEST SCORES


113

C.C. Blaney Elementary School and R.D. SchroderMiddle SchoolThese schools are benefiting from a recent commitmentby the district to bring them up to the same level as otherschools through funding for facilities, staff, andequipment. Students, particularly those in the elementarygrades, are making significant gains. For example, in 200375.2% of students tested at or below Basic, but in 2004only 64.5% of students tested in this category. In 200324.8% of students scored in the Proficient or Advancedcategories but in 2004 47.9% of Blaney students scored inthese categories. Blaney’s test results compare well withscores achieved by other African-American students inthe district in 2004.

Students at R.D. Schroder have also improved theirperformance, though not by as much as Blaney students,but they score better in most categories than the averagefor African-American students in the district.

Camino Nuevo Elementary and Middle SchoolsIn almost every category, students from Camino Nuevooutscore their peers in nearby schools serving a similarpopulation. For example, between 17 and 22% of studentsin Grades 2-5 scored in the Advanced or Proficientcategory, while between three and nine percent ofstudents at comparable Gratts Elementary scored in thetop categories. Similarly, between 15 and 18% of studentsfrom Camino Nuevo Middle school in grades 6-8achieved scores in the highest two categories, while onlynine to 11% of students in nearby Berendo School scoredin the Proficient or Advanced categories.

Flagstaff Arts and Leadership AcademyThe AIMS (Arizona Instrument to Measure Standards)scores for 2004 were 86.5 in Reading, while the stateaverage was in the 50s. Math scores were 77.8, while thestate average is in the 30s, and students’ scores on thewriting test were even with the state average at 59.5.FALA students have consistently outscored the stateaverages, including Flagstaff Unified School Districtschools. In 2002 FALA students scored 13 percentagepoints higher than the national average in reading and7.2 higher in math.

High Tech HighStudents at High Tech High far surpass their peers atPoint Loma Senior High. For example, between 69 and77% of students in grades 9-11 at High Tech High scoredin the Proficient or Advanced category in 2004, while

between 36 and 43% of students in grades 9-11 at PointLoma scored in these categories. There is a similardisparity between scores achieved by students at HighTech High and those at Point Loma High School foralmost all other categories.

Interdistrict Downtown SchoolStudents at IDDS scored higher on the MinnesotaComprehensive Assessments (MCAs) than theircounterparts at Sullivan Elementary School andWashburn High School. For example, 55% of thirdgraders and 60% of fifth graders at IDDS scored abovegrade level in reading, while 34% of third-grade studentsand 43% of fifth graders at Sullivan Elementary scoredabove grade level. Students’ scores in math followed thesame pattern. Thirty-seven percent of tenth-gradestudents at IDDS scored in the two highest levels, and16% of the 11th graders were as successful on the MCAs.Eighteen percent of Washburn High School’s tenthgraders scored above grade level in reading and only sevenpercent of its 11th graders.

There was a similar disparity at the lowest end of thescores. Twenty-three percent of third graders at IDDSscored below grade level in reading, and 40% of fifthgraders scored in the two lowest categories, while 40% ofthird graders at Sullivan were below grade level and 47%of fifth graders scored in the lowest two categories. On themath test, 23% of IDDS third graders were in the lowestcategory, while 64% of Sullivan third graders were belowgrade level. Similar patterns hold true in upper grades.

Laurel-Concord Public SchoolsThe Laurel-Concord public schools compare favorablywith national averages. Seventy-five percent of studentsin grades 3-5 score above the national average in reading,and their performance is improving. In 2003-2004 almost75% of students in grades 7-8 were above the nationalaverage, as were 72% of 10-12th-grade students. In mathonly 15% of students in grades 3-5 at Laurel-Concordscored below the national average, and 69% of students ingrades 7-12 were above the average.

Students at Laurel-Concord scored above the state andnational averages for the ACT (American College Tests):Laurel-Concord: 24.2; State of Nebraska: 22.6, National: 21.9.

Media and Technology Charter High SchoolIn 2000, only 18% of incoming ninth-grade students hadpassed the eighth-grade Massachusetts Comprehensive


114

Assessment (MCAs). Yet after just two years at MATCH,the pass rate on the tenth-grade exam by these samestudents was 82%. By December 2002, after the firstretake opportunity, the pass rate was 100%. Similarly, bySeptember 2003, 89% of the class of 2005 passed bothmath and English on their first try. By May 2004, after thesecond retake opportunity, 100% of the class of 2005passed both tests. Moreover, despite the fact that just fourpercent of students had scored in the top two MCA levels(“proficient” and “advanced”) in math before coming toMATCH in 2001, after two years, 73% scored “proficient”or “advanced” in 2003. On MCAs, in 2002 and 2003,MATCH was number 1 of 22 open-admissions highschools in Boston, and the top predominantly African-American high school in the state.

Metropolitan Career and Technical CenterOne of the most important indicators of the success Metstudents are achieving is the graduation rate – 94.2% in2003. This is in stark contrast to the 48.1% graduationrate for Hope High School, which many of the Metstudents would otherwise attend. Scores on the math andreading tests show not only that students at The Met areimproving their academic performance on state tests, butthat they score much higher than students at Hope HighSchool. For example, on the English language arts tests,22.2% of Met students achieved the standard or above,while only 15.9% of students at Hope High School metthat goal. At the Met 29.3% of students achieved thestandard or above in math in 2004, while only 9.2% ofstudents at Hope High were at or above the standard.

Oak Valley SchoolThe school is so small that the state does not publish testresults; however, the fact that 22 of 30 students come tothe school from outside their districts suggests that parentsare very pleased with the work their children are doing atOak Valley.

Tacoma School of the ArtsStudents come to TSOTA from all over the city ofTacoma, so it is difficult to compare their performancewith that of students in another high school. The nearestschool, Foss High School, has a different demographic,with 47.1% of students eligible for free or reduced-pricemeals in contrast to 7.7% of students at TSOTA.Students at TSOTA are doing well academically. Theannual dropout rate is .7% while the dropout rate at Fossis 5.6%. In 2003-2004, 91.1% of TSOTA students metthe standard in reading, 63.1% met the standard in math,94.7% met the standard for writing, and 57.3% met the

standard for science. In comparison, 57.8% of Fossstudents met the reading standard, 32.7% met the mathstandard, 61.6% met the writing standard, and 17.2% metthe science standard.

The STAR SchoolThis school is so small that the state does not publish test results.

Todd Beamer High SchoolTodd Beamer High School opened in the fall of 2003 with1,350 students in three academies. The WASL results arenot reported by academy, but by whole school. Resultsachieved by students at Todd Beamer are roughly thesame as those achieved by students at Federal Way HighSchool: 66.9% of Beamer students met the standard forreading, 40.6% for math, 67.8% for writing, and 26.4%met the science standard. By comparison, 67.55% ofstudents at Federal Way High School met the readingstandard, 38% met the math standard, 72.1% met thewriting standard, and 21% met the science standard.

H.S. Truman High SchoolTruman is an alternative high school so this comparisonlooks at the school’s progress over the past few years sinceit converted to an internship model, as well as againstFederal Way High School. In 2002-2003, 29% of Trumanstudents met the standard for reading, but by 2003-2004this number jumped to 72.3%. In 2002-2003, 9.1% ofTruman students met the math standard, but 14.9% ofthem accomplished this in 2003-2004. By 2003-2004,63.8% of Truman students met the standard for writing,whereas the year before only 33.3% had been able to meetthe standard. In science there was a slight jump, from 9.1to 10.9%. Scores for Truman students still lag behindthose of students at Federal Way High School in all areasexcept reading, where 67.5% of Federal Way students metthe standard in comparison to 72.3% of students atTruman High School. The progress Truman students havemade in recent years, however, is outstanding.

World of OpportunityStudents at WOO do not take state tests; however, inspring of 2005, 50 had passed their GED and more than25 were in college. Without the program and supportWOO gives, it is very doubtful any of these studentswould have passed the GED or been accepted at college.


Appendix 4 • Grid of Strategies

The checked items indicate which cost-saving strategies each school uses. Leaders from each of the schools highlighted inthis report are available to discuss the strategies they have employed. Contact information for each school is listed inAppendix 2.

Blaney/ Laurel-Avalon Beamer Schroder Bronx Camino FALA High Tech IDDS Concord MATCH MET Oak Valley TSOTA STAR Truman WOO

STAFFING

Juggling act X X XUsing rules strategically X X XDouble duty X X X X X X XSharing administrative duties X X X X XSharing staff X X X XImporting teachers XAccepting no substitutes XUsing mentors and tutors X X X X X X X X X XAdjunct faculty XUsing volunteers X X X X X X X X XMultiple credentials X X X XHiring young, first-time teachers X X X X XHiring retirees and career changers X XProject- and intern-based curriculum X X X XIncentives X

EDUCATIONAL PROGRAM

Focused mission X X X XAttracting students XCourse cycles X XOrdering supplies X

SERVICES

Transportation X X X X X X X X X XFood service X X X X X X X X X X X X X X X XAthletics X X X X X X X X X X X X X X XHealth care X X X X X X X X X X

SOURCES OF FUNDING

Renting space X XBusiness located in school XTrading spaces X X XBuilding shared space X XSelling memberships XRenting computers XGrants X X X X X X X X XDonations X X X X X X XCommunity support X X X X X X XSharing space X X X X X X XRecruiting students X X X

Blaney/ Laurel-

115continued >

Avalon Beamer Schroder Bronx Camino FALA High Tech IDDS Concord MATCH MET Oak Valley TSOTA STAR Truman WOO

SOURCES OF FUNDING (CONT.)Federal funding:

Title 1 XImpact aid XWork study XEmergency repairs XQZAB X

Carl D. Perkins Vocational Grants XNew Market Tax Credit XState programs X X

FACILITIES DESIGN AND CONSTRUCTION

Smaller school sites X X X X X X X X X XDonated site X X X XDesign and construction X X X X X X X X X X X X XFree design X XConstruction bidding XRenovation X X X X X X X X X XAdditions X X X XLeasing X X XUsing students and volunteers X X X X X XFlexibility/adaptability/multiple uses X X X XSharing space with partners X X X XReducing square footage/student X X X XSharing space for athletics and extra-curricular activities X X X XPartnerships in Construction X X X X X X X X X X XGreen construction X X X X

FURNISHINGS, FIXTURES, EQUIPMENT

Donated X X X X XBuild it yourself X X X X XShared internally X X X X X X X XFlexibility/mobility X X X X X XIT built rebuilt by students X XIT serviced by students X

MAINTENANCE

Preventive maintenance XReducing trash and graffiti X X X X X X X X X X X X X X XStudents, parents, teachers maintaining building X X X X X X X X XStudents, parents, teachers maintaining grounds X X X X XUsing computers effectively for maintenance and operations X

OPERATIONS

Janitorial staff X X X X XSecurity staff X X X X X X X X X X X X X X

UtilitiesNatural lighting X X X X X X X X X XDual fuel X X XAutomatic light sensors X X XSolar power X X XKeeping cool X XAutomatic water shut-off X X X XWind power XComputerized control system X X XRetrofitting X


117

The Dollars & Sense team selected schools on the basis ofthe following criteria:

SIZE: Schools that had 400 or fewer students or no morethan the following number of students per grade:

High School: 100 Middle School: 75Elementary: 50

REPRESENTATIVENESS: Schools located in a. Urban, rural, and suburban communities.b. Different regions of the country.

TYPE: Schools that were organized bya. Different grade configurations.b. Different forms of governance including:

traditional, charter, alternative, conversion.

EQUITY: Schools that were accessible to students of allbackgrounds.

FACILITY PROJECTS: Schools that had completedconstruction of a new facility, addition, or renovation ofan existing building, within the previous 10 years.

Appendix 5 • Criteria for Selecting and Evaluating Schools

PROGRAM and the way in which the facility supportsthe program:

The team selected schools that met the following criteriaas fully as possible, and whose facilities supported thesecriteria. For the purpose of evaluating schools, the teamdeveloped a detailed rubric of ways in which the facilityshould support the program, and indicators that suggestthe degree to which it is succeeding.

Smaller size is potentially helpful in increasing connectionsamong students, educators, and the community,particularly in marginalized and impoverished communitiesand neighborhoods, but it does not guarantee success.Leadership, program, and climate are essential ingredientsin a good small school, but best practice can vary to reflectlocal history, culture, and conditions.

The Dollars & Sense team was committed to includingsmall schools that in their programs honor and cultivate:

• autonomy, • good leadership,• a defined mission, • making choices, to ensure coherence and support

the distinctive nature of the school,• developing good working relationships within the

school and between the school and community, • equity, • respect, • a supportive culture, and • a commitment to academic achievement.

This list of characteristics was informed by Linda Darling-Hammond’s work in Redesigning Schools, What Matters and WhatWorks and the Gates Attributes of High Achieving Schools.

Case Study Selection Criteria:


118

CriteriaSustained Relationships(Student/Teacher/Staff/Administration)

ImportanceIncrease student successIncrease sense of belongingIncrease mutual trustImprove academic performanceIncrease engagement in schoolIncrease teacher retentionImprove teacher effectivenessIncrease school/community sustainabilityIncrease empowermentIncrease opportunities for responsibilityIncrease ‘informed’ cooperationIncrease social capacityIncrease civic capacity

IndicatorsStudents/teachers in multi-year associationsStudents socialize with adultsStudent behavior toward visitorsBullying vs. support and cooperationStudent academic successStudents meet with adults outside of classTeachers know students’ namesPrincipal knows students’ namesGuidance Counselor/Advisor knows namesParticipation in after-school activitiesTeachers/students/staff in facility after schoolTeacher moraleTeacher retention rateTeam teachingParents and community volunteersParent participation in school eventsRespectful frankness, opportunity for

honest discussionGraduation rateLoss rate from grade to grade (high school)Process of change/implementation in placeOngoing student/community projects

Authenticity andAccomplishment

Increase student successIncreased pride in selfIncrease pride in schoolIncrease level of performanceIncrease options for studentsIncrease level of academic workLead to sustainability – individual and

communityIncrease community options for

sustainability

Student work evaluated by performance and portfolio

Student work is visible throughout the schoolsPerformance assessment is an essential

component of the programStudent work with and in the communityStudents are engaged in classes and

independent workStudents are involved with long-term

community projectsIEP or equivalent for all students/program is

learner-basedStudents look cheerful/contentPost-secondary options goodLocal entrepreneurs hire students

Site Visit GuidelinesForm for School Program Evaluation:


119

Form for School Program Evaluation (continued):

CriteriaEquity; Democracy

ImportanceIncrease empowermentIncrease responsibilityReduce fear, increase trustIncrease sustainabilityIncrease safety for allRight thing to doIncrease student success

IndicatorsSchool is open and available to students of

all backgroundsStudent body is economically, racially diverse,

if possible, given local demographicsStudents, staff, teachers, parents, community

have opportunity and confidence toexpress views

Students, staff, teachers, parents, community feel their opinions matter and areconsidered

Students, teachers, parents, community voice opinions

People ask open-ended questionsProgram provides for a range of opportunities

for successModes of teaching are adjusted to individual

student backgrounds, talents, interests andthe nature of past performance

Students mix at lunch, optional activitiesTeachers come from diverse backgrounds, agesStudents joke across cultural differencesStudents have friends from diverse

backgrounds

Collaboration/Partnership Increase teacher effectivenessIncrease student achievementIncrease sustainability of schoolIncrease sustainability of communityAdd to quality of lifeEmpower peopleIncrease responsibility

Students have opportunities to work with other students

Students have opportunities to work in the community

Community people and organizations are involved in the daily life of the school

People work collaboratively to realize goalsStudents listen to each otherStaff and teachers value working

collaborativelyTeachers have opportunities to work/plan

togetherTeachers valued by communityTeacher retention rate/morale


120

CriteriaCommunity

ImportanceImprove quality of lifeIncrease trustSupport academic programIncrease sustainabilityIncrease safetyIncrease empowermentIncrease responsibility

IndicatorsThe members of the school and community

interact, have a sense of responsibility toand for each other

Members of the school community express a mutual sense of identity with and for eachother

People operate from a base of community, not competition or power

Community spaces augment and support the school facility and program

Diverse and active parent groupsParents and other community people

use the schoolWork of adults in the community is on display

CriteriaFlexible/Internally

ImportanceSupport varying teaching stylesSupport varying learning stylesIncrease community useEmpower all usersIncrease responsible and individualized use

IndicatorsWalls, furnishings, some décor can be in

multiple configurationsTeachers and students can and do change

walls, furnishings, to support multipleteaching and learning strategies

Community can use (with/without students)

Adaptable/New Uses Increase sustainability of schoolIncrease community involvementIncrease community support

Facility can easily be used by community groups

Facility can easily be used for many purposesFacility is used on a regular basis for many

purposesPortions of the building, or the entire facility

could be adapted to an alternate use

Form for School Program Evaluation (continued):

Form for School Facility Evaluation:


121

CriteriaSafe/Secure

ImportanceIncrease trust among occupantsSupport success in all areasIncrease pride in schoolMake parents feel at easeHelp parents support school/kidsHelp community feel welcome

IndicatorsOccupants feel safe and secure in and

around facilityThere are few discipline problems and little to

no vandalism or crime on school groundsSpaces promote safety and a sense of securityStudents are trusted to leave classrooms and

buildingPeople in the community visit the school

Comfortable Support academic programMake people feel empoweredMake people feel welcomeIncrease use of buildingIncrease responsible use of resources

Thermostats can be controlled in individual rooms

Day-lighting prevalentNoise level is not distractingBuilding is accessibleScale is appropriate to age level of studentsFurniture and furnishings

attractive/comfortable/used

Welcoming Put people at easeIncrease parent participationSupport academic programPut people at advantage not

disadvantage—at their bestDecrease defensivenessIncrease cooperation

Occupants and visitors feel welcome – are greeted when entering

Signage is clear, invitingSpace is available for community and

parent usePeople less stressedParents and community people use the spaceVisitors don’t get easily lost

Offers Sense ofPlace/Ownership

Increase student successIncrease parent supportIncrease community supportIncrease financial supportIncrease short- and long-term commitment

to school/communityIncrease long/short term sustainability of

school/community

Design and décor reflect the location, history and culture of the area

Design and décor reflect location, history and culture of people attending the school

Scale and form of facility are appropriate to the area

Community helps care for facilityStudents help care for facilityStudents can learn about their heritage from

the design and décor in the schoolThere are traditions specific to the building that have

evolved through use and over time that are respected

Students help minimize wear and tear, trash and use of utilities

There are ways for current users of the building to influence what it looks like and how it functions

Form for School Facility Evaluation (continued):


122

CriteriaPersonalized

ImportanceIncrease student confidenceIncrease student enjoymentIncrease student empowermentIncrease student responsibilityIncrease teacher confidenceIncrease teacher moraleIncrease teacher empowermentIncrease teacher effectivenessIncrease teacher retentionSupport academic program

IndicatorsStudents have individual space which they

have adapted to personal preferencesStudents have places to socialize with peersStudents have places to socialize with adultsTeachers and staff have spaces to meet and

collaborateThe facility offers useful spaces that support

teacher collaboration in planning and development of programs

There are spaces appropriate for a wide range of academic activities

Students have access to technology that permits individual research and study

There are appropriate spaces for support services

Aesthetically Pleasing Increase pride in schoolIncrease enjoyment of spaceIncrease use of spaceIncrease quality of life in school

Space is interesting, enjoyable, and even funSpace is aesthetically pleasingSpace is well-cared for, clean and without

graffiti or damage

Encourages CommunityInteraction

Increase support for studentsIncrease empowermentIncrease responsibilityIncrease community optionsAdd to community amenitiesIncrease sustainabilitySupport academic programAdd to quality of life for all

Building provides spaces for parents and community people to use while volunteering

Building provides spaces for parents and community to meet with students/teachers

There is space available for students, teachers, parents and community members to meet on a regular basis for discussion of issues relating to the school

The original building is a product of collaboration with the community

The building reflects community goals and answers community needs

The community participates in ongoing policy and decisions about the facility

Community people mentor students in the school and outside the school



123

CriteriaOutdoor Spaces areIntegral to Facility

ImportanceSupport academic programIncrease optionsIncrease empowermentIncrease responsibilityMaximize ecological integrityAdd to quality fo life for all

IndicatorsLandscaping is well cared forLandscaping is appropriateGrounds are free from litterOutdoor spaces provide places for people to

relax, study and meetOutdoor spaces provide places for recreation

and athletics

Cost Effective Increase sustainabilityMaximize ecological integrityAdd to quality of life/options

Maintenance is regular, pro-active, thoroughCosts for maintenance, operations minimal

Innovative Increase sustainabilitySupport academic programIncrease prideIncrease confidence in schoolIncrease sustainability

Creative, thoughtful strategies reduce costsCreative strategies maximize use of building


1

Construction Costs of Smaller and Larger Schools

Craig B. Howley Ohio University

September 5, 2004

Prepared for the KnowledgeWorks Foundation and the Dollars and Sense II team

2

Construction Costs of Smaller and Larger Schools: Final Report

Problem Statement Across a range of contemporary studies, including those that examine tested achievement, researchers have concluded that benefits to students appear to accumulate more densely in smaller schools than in larger schools. With respect to tested achievement, smaller schools have been shown especially beneficial for the most impoverished quartile of students; smaller size is also associated with greater equity of achievement measured as weakness of the relationship between socioeconomic status and achievement (Howley, Craig & Howley, in press). High school size above about 1,000 students has also been shown to be detrimental to the tested achievement of students regardless of socioeconomic status or poverty (Lee & Smith, 1997). Smaller high schools have also been shown to be consistently associated with lower dropout rates and higher rates of participation in co-curricular activities (Fetler, 1989; Morgan & Alwin, 1980) . Benefits for smaller schools are also claimed, on the basis of a somewhat weaker base of evidence, for teacher collegiality, school safety, and students social and emotional development and well-being (Cotton, 2001). Reasonably sized schools would enroll fewer students than those shown to be associated with harmful academic consequences, and they would vary in sizes less than those upper limits according to community need and preference—all else equal. Part of that “all else” concerns the cost of building smaller schools. Resistance to building smaller schools

Informing the resistance to building smaller rather than larger schools—including the unfriendly mega-schools larger than 1,000 or 1,500 students—is the durable presumption that smaller schools cannot be built cost effectively. In both rural and urban areas, in fact, the construction regulations promulgated by states and districts tend to impose size restrictions that impede the construction of new smaller schools because the prevailing wisdom accepts the proposition that economies of scale reduce construction costs per student or per square foot. Although several large urban school districts (New York, Chicago, and Los Angeles among them) have attempted to create smaller schools within their huge systems, sustaining and extending these efforts has proven difficult, in part because these districts themselves are so large (Meier, 1995) . In rural areas, by contrast, the continued existence of smaller schools encounters the threat of closure whenever school facilities are planned.

Presumptions, of course, need to be questioned. What is the evidence informing the durable presumption? Is the evidentiary base deep and wide? Does the available evidence support the presumption strongly, weakly, or not at all? Are smaller schools not cost effective as compared to larger schools? Are cost differentials, if they exist, large? Are smaller schools more costly in all ways and under all conditions? This study focuses on just two of the relevant questions, and is part of the larger research effort designed to inform the development of Dollars and Sense II. Research questions

The specific research questions addressed by this quantitative study are as follows: (1) Are planned costs of construction are equal for smaller as compared to larger schools?

and

3

(2) How do construction costs vary in light of school and context variables including planned enrollment, square footage, district revenues, locale, socioeconomic conditions, and ethnicity?

Relevant Literature We conducted several searches of relevant databases in order to assess the available research. The result of these searches could be considered shocking. Very few instances of bona fide research studies of school construction costs exist in the professional education literature. The ERIC database indexed (between 1966 and 2003) four assessments conducted by project team member Paul Abramson, and just 3 other publications. All of these appeared in practitioner magazines rather than in research journals proper. The median length of these articles was a meager four pages.

In his latest annual school construction report Abramson (Abramson, 2004) observes that smaller high schools (smallest quartile, with a median planned enrollment of 375) appear (1) to be less costly per square foot and more spacious than larger high schools and (2) no more costly per student than the largest quartile of schools (with a median planned enrollment of 2,100 students). The smallest quartile of elementary and middle schools, by contrast, were no more expensive than the largest quartile of these schools, but they were more expensive per student because the smaller school allocated about 50% more square footage per student than did the largest schools. The lack of contextual variables, however, is a distinct limit on the interpretation of these results, and Abramson notes that intriguing questions remain. His observation should be read as understatement given the nearly total lack of research on this strategically important educational topic.

Use of a variety of ERIC descriptors and search strategies yielded consistently similar

results: the relevant literature is thin and superficial. Searching on “school size” and “construction costs” we discovered a single peer-reviewed research study (Azari-Rad, Philips, & Prus, 2002). This study was based on a national data set of accepted bid prices for schools built between 1991 and 1999.1 The data set used by the researchers contained information similar to that supplied for the current study (that is, very few variables and many cases). Analyses examined all schools together and high schools separately. Size of school in this study was measured by total square feet rather than by planned enrollment, and on this basis the researchers concluded that doubling the square feet of a project increases costs by 91%. The researchers also conclude that a two-school, as compared to a one-school option, would increase construction costs by an average of 4.7%. Azari-Rad and colleagues, however, report that this modest advantage is offset when “very large schools” are planned, because these projects generate demand for local construction talent that can increase costs from 8% to 12%. Their best advice for saving on construction costs has nothing to do with project size (again, size is defined as total square feet); they advise that school construction be planned for economic downturns, when costs will be lower due to weak demand for construction. A doubled unemployment rate is associated, they suggest, with a 21% decrease in school construction costs. The researchers estimate, also, that paying higher “prevailing wage rates” (union wages) does not result in cost savings, perhaps because construction is more effectively managed by firms paying the higher wages—better management offsetting higher wages. This is a useful and interesting study, which reaches some surprising insights using very few variables, but it hardly constitutes an

1 The data were collected by a subsidiary of McGraw Hill (F.W. Dodge)

4

adequate literature. Azari-Rad and colleagues, not surprisingly, refer to no prior literature in their study.

The absence of research about school construction costs, however, does not indicate a lack of interest. The construction industry magazines actively report on school construction, and business magazines and newspapers also report on the trends and issues that emerge in school construction.

School construction has general economic importance quite apart from its importance to the professional education community, to parents of school-aged children, and local taxpayers. School construction is prominent in the overall construction industry. For instance, a reporter for Building Design and Construction reported in 2002,

Builders completed more than $62 billion worth of U.S. educational facilities during 2001, an increase of 13 percent over the total for 2000. This growth was identical to the growth rate recorded between 1999 and 2000, so there’s no disputing the “star” status of the educational sector [of the industry]. (Delano, 2002)

School construction has an important effect on the generation of decently-paid local or regional jobs, and—obviously—on the profitability of construction companies. In the domain, the niceties of education figure as politically and economically less important issues. The dominant presumption that larger is cheaper prevails in general, quite apart from the issue of the proportional costs and benefits. The construction industry would arguably be the last entity to object to the construction of mega-schools enrolling of 2,000 to 5,000 students and destined for pedagogical disaster. Larger projects entail larger management and design fees, often calculated as a percentage of the cost of the amounts budgeted for materials and labor. School construction, moreover, figures in political agendas in educationally irrelevant ways, as well. A 1998 article in Crain’s Cleveland Business reported on the “prevailing wage” controversy in Ohio. In 1997, according to the article (Ford, 1998), the state legislature ended “the decades-old practice of requiring the payment of prevailing wages on school projects...at the urging of business lobbyists who argued the change would save school districts money.” The Ohio School Facilities Commission, which funds school construction in Ohio, affirmed that it did not have the data to support the claimed savings. The single research article in this literature(Azari-Rad et al., 2002), published three years later, finds that such savings are unlikely (based on national, not Ohio, data). Data Sets This study analyzes two national data sets to answer the research questions, a basic data set of recently planned construction projects at three educational levels (elementary, middle, and high school) and an augmented data set of high schools only. The augmented data set is derived from the basic construction data set; the augmentation consists of added variables that enable the team to address the second research question (“the extent to which construction costs for larger and smaller school in light of contextual variables including locale, socioeconomic status, and ethnicity”).

Basic construction data set. The project team acquired from Paul Abramson, author of School Planning and Management’s “Annual Construction Report,” a dataset of 3,471 cases of

5

school construction projects (new buildings according to survey respondents) begun in the years 1989 to 2003 for planned completion in the years 1996 to 2009. This data set included all cases gathered via survey research from 8 years of the report that pertained to self-reported new construction. It contained just 8 unique variables: state, zip code, educational level (elementary, middle, and high school), total cost, planned enrollment, planned square feet, start year, and end year. Useful derived variables can also be calculated from the unique variables, including cost per square foot, cost per student, and square feet per students, and smaller versus larger schools (e.g., cases split at the median on planned enrollment). Regions can also be fashioned from groups of states, but such analysis was not a concern in this study (Abramson, 2004). The project team was concerned, however, not only to carry out analyses with the 3,471 cases in the basic data set, but also to carry out more extensive analyses than would be possible with 8 variables and the relevant derived variables in the basic data set. For this reason, the project team developed and executed a plan to augment cases in the data set with additional variables.

Augmented data set. Because the cases in the basic data set were schools, the team decided to attempt a of match these cases with schools in the Common Core of Data (CCD), the annual census of U.S. schools and districts maintained by the National Center for Education Statistics (NCES). The CCD includes information about the locale of a school, its gradespan, its minority enrollment, and its subsidized meal rate. Additionally, each school can be linked to information about the district that operates the school. The basic construction data set (n=3,471), however, contained very little information on which to attempt the intended match to CCD cases. State and zip code were the only relevant variables on which to attempt the match. Grade-level (elementary, middle, and secondary), a variable in the construction data set was clue, but does not match any CCD variable; the CCD posts highest and lowest grade in a school. The construction data set also included planned enrollment as a variable, which would obviously vary to an unknown extent from actual enrollment. The accuracy of any match based on such information, the team reasoned, would need to be assessed. Test match process. For these reasons, a procedure was developed to assess the possibility of obtaining matches in which the team could reasonably vest confidence. Inaccurate matches would embed error into any dataset created in this way, and a high rate of mistaken matches could well obscure any patterns that might otherwise have emerged. Poor matches, in short, would pose a serious threat to internal validity. We selected 40 cases at random from the dataset and attempted a match on the basis of zip code, educational level (elementary, middle, and high school), and planned enrollment. We used the CCD “school locator” system, an online resource of NCES (http://nces.ed.gov/ccd/ schoolsearch/). We searched for all schools in a given zip code and selected the closest apparently matched case within the zip code to assign a school in our basic data set. When all cases had been matched in this tentative way, we placed phone calls (using the “matched” school’s phone number as listed in the CCD). Four schools could not be reached on repeated attempts because phone numbers were incorrect.

6

In 28 of the 36 remaining cases (78%), the success of the matching effort seemed probable. We asked a series of questions to arrive at this judgment, with the key question asking if the school had experienced substantial new construction in recent years (additions, renovations, or if the school were newly constructed). We also asked about the nature of the substantial new construction. Although the construction data set had identified all cases as new construction, in fact our respondents reported that 11 of the 28 were not new construction, but instead involved a substantial variety of projects including new roofs, new heating and air conditioning installations, portable buildings, and additions. That is, among these schools, only about 60% of the matched schools (17 of 28) could be confirmed as the new construction of an entire school.

Revised matching strategy. The information from the test match process was not totally discouraging. The good news seemed to be that matches could be made with a measurable degree of reliability (ranging from low to high; we regarded our obtained rate of 78% as promising, but not adequate). The bad news was the unexpected range of projects, when we had expected only new construction. Taken together, the good and the bad news seemed to point to the need to devise an improved match strategy that would yield a data set adequate to our intentions to perform a more close-grained analysis than was possible with the few variables in the basic construction data set. Such an improvement, we reasoned, would come about if we could restrict the number of CCD schools presented for matching. Two steps to accomplish this restriction recommended themselves to us.

First, high schools are less numerous than elementary schools in any locality; they are larger in general and they comprise fewer grade levels (typically 4 grades). This is the likely reason that the basic data set classifies just 803 cases of its cases as high schools, but 1,868 of its cases as elementary schools.

Second, within the high school group, zip codes that were unique among high schools in

the basic data set would also reduce the range of CCD schools available for a prospective match. Our experience with the basic data set and the matching process suggested that when multiple schools were represented in the basic data set with a single zip code, the probability of making a mismatch was heightened, but when this was not the case, the probability of a mismatch decreased. Finally, in view of the literature on school size and the positions on school size take in Dollars and Sense (Lawrence, B. et al., 2003), we chose to focus our attention on high schools enrolling 1,000 or fewer students as representative of what Lawrence and colleagues (2002) referred to as “reasonably sized” high schools. The resultant data set would comprise a purposive sample of newly constructed high schools planned nationally between 1989 and 2003 to house 1,000 or fewer students.

We applied these three selection rules and extracted 296 cases of the 803 in the basic data

set. We again attempted matches, but also imposed an additional decision rule intended to exclude cases in which circumstances were likely to inflate chances for a mismatch, as follows. We identified the district to which the case in the basic data set belonged (based on the zip code variable) and, if the district operated more than two high schools, excluded the school from further effort. Of the 296 cases, just 12 cases were excluded for this reason. In another 12 cases,

7

no likely match was evident. Additionally, 3 cases were also excluded because of evidently erroneous information in the basic data set (e.g., zip code mismatch with state).

Thus, 269 cases were available for further effort (i.e., confirmation calls to all apparently

matched schools). Attempts were made to reach all 269 schools, including multiple calls over a span of three weeks. At the end of May, 2004, calling was discontinued as further effort yielded no additional contacts. We had contacted 244 schools.

In making contact, we posed two questions to respondents: (1) Is this the only high

school in the district? and (2) Was this high school newly built within the past 10 years? Of the 244 schools contacted, 79.1% (193) were the only high school in their district and 86.5% (211) were newly built schools. Table 1 provides the relevant crosstabulation. Table 1. Crosstabulation of answers to “two or fewer high schools in district” and “newly built school”

Only high school in district?

Newly built high school?

no yes row total

yes 15 (6.1%) 178 (73.0%) 193no 18 (7.4%) 33 (13.5%) 51

column total 33 (13.5%) 211 (86.5%) 244

The crosstabulations in Table 1 also provide confirmation of the success of the purposive sampling rules in identifying newly built schools. For districts in which the contacted high school was not the only high school, respondents at 35% of matched schools we contacted denied that they had been newly built (18 of 51)—marginally better than with the test match cases (60%, see preceding discussion). For high schools unique to their operating districts, however, respondents at only 8% of schools contacted (15 of 193) made this denial. With this information in hand, the team opted to perform analyses only on schools that were confirmed new construction.

Much can be specified about these schools, in contrast to those in the basic data set. Adding school-level and district-level information to the cases from information in the CCD provided upwards of 100 additional variables. Analyses employ a more limited number of variables, of course, including the following: locale (rural and nonrural), educational attainment of the adult population, proportion of population living in poverty, total revenue per student, proportion of school enrollment from minority background, and district enrollment. Choice of variables for analysis is fully explained in the discussion of analytic method.

Preparing the basic (unaugmented) construction data set for analysis. Experience with constructing the augmented high school data set and with initial analyses on that data set, surfaced two concerns with the basic (unaugmented) construction data set. Our concerns for the limited analyses possible were, first, strategic, and second, tactical or technical.

8

First, we made the strategic decision to segment the data set into 3 panels based on educational level: elementary, middle, and secondary. This is a strategic decision based on differences among educational levels. Although we had no way to assess the accuracy of the assignment of level in the data set, there was, we agreed, ample reason to suspect that cost patterns and relevant relationship could differ by educational level, since curricula, pedagogy, characteristics of students, and average size of buildings differ considerably by educational level. Indeed, Abramson’s previous reports had indicated the need for this decision, which positioned the project to perform three sets of analyses with the basic data set.

Second, because we were most interested in new construction, we realized that extreme

values in a key derived variable—cost per square foot of the indicated construction project (ostensibly a new school)—were more likely than the middle of the distribution to be associated with two troublesome conditions: (1) recording error (i.e., values simply entered incorrectly) and (2) very non-representative construction projects. The particular concern was with suspiciously low costs-per-square foot (under $50 or $60 per square foot), especially in light of our knowledge that a portion of these projects did not involve the construction of entirely new schools. A considered a number of alternatives for dealing with this threat were considered, including removal of univariate and multivariate outliers. Such outliers, however, all fell at the upper end of the distribution. Thus, our concern with low values in the data set (e.g., project costs per square foot under $50 or $60) were not addressed, and adopted a different, and commonly used, approach, trimming the data set by removing the top and bottom 5% of cases on cost per square foot in each segment (elementary, middle, and high school). In every instance, the trimmed data sets produced more normally distributed values for cost per square foot (reduced from a maximum skewness of 9.66, among middle schools, in the untrimmed data set to a maximum of 1.47, among elementary schools, in the trimmed data set).

Because we had obtained, at the high school level, a confirmed set of newly constructed

schools, we were able to compare values of key variables in the adjusted basic data set and in the augmented high school data set (n = 211). Measures of central tendency, variation, and range were similar in the two data sets. The augmented data set, of course, pertains only to schools planned to enroll 1,000 or fewer students. Of the 723 high schools in the trimmed data set, by comparison, 316 enroll more than 1000 students. Table 2 reports the comparisons, which report values that differ in an anticipated direction, but which are unexpectedly close in view of the differing sizes represented by the two data sets. Table 2. Central tendency and variation in two high school data sets Purposive high school data set

(n=211) High schools in trimmed basic data set

(n=723) $/ft2 $/student ft2/student $/ft2 $/student ft2/student mean $105.88 $21,759.84 210.03 $110.01 $20,919.70 197.01median $97.67 $19,296.25 200.34 $103.45 $18,561.48 173.33SD $44.37 $11,301.71 85.35 $33.03 $10,918.31 102.76

Differences in values of the focal variables in the two data sets. Important distinctions in the large trimmed basic construction data set (n = 3,124 after trimming) and the much smaller augmented high school data (n = 211). These distinctions concern, on one hand, of size (the

9

smaller augmented data set is restricted to high schools planned to enroll 1000 or fewer students) and, on the other hand, of the apparent nature of the construction costs (the trimmed basic construction data set represents a variety of construction costs—principally but by no means exclusively new construction—associated with elementary, middle, and secondary schools of all sizes. The discussion of results will remind readers of these distinctions in subsequent sections of this report. Methods of Analysis This section details the statistical methods used to address the research questions. Analysis includes descriptive statistics, tests of differences in means, as well correlational and regression analysis (ordinary least squares). Schools are the unit of analysis in all instances. Analyses with the augmented high school data set. The basic construction data set (n = 3,471 prior to trimming) identified any school as a high school if its highest grade were grade 12. Examination of the augmented data set (n = 211), with CCD information on gradespan available as a result of the matching process, identified three types of high schools based on gradespan configuration. The most common type (80% of cases, or n = 168) were the expected 9-12 high schools, currently the typical pattern in the U.S. About 20% of cases (n = 41) spanned grades 6, 7, or 8 to 12 (78% were 7-12 high schools). Two schools were of neither type: one was a 1-12 school and one was a 10-12 school. These two “other” schools were dropped from analysis and the 9-12 and 6,7,8-12 schools (henceforth referred to as “6-12” schools) were analyzed separately.

The augmented data set of newly constructed schools included 211 cases. Importing CCD data and developing an array of derived variables resulted in a data set with 119 relevant variables—as compared to the 17 unique and derived variables ultimately available in the trimmed basic construction dataset (n = 3,124) used for the analyses described above.

The researcher performed the same sorts of analyses for each panel (9-12 high schools

and 6-12 high schools) that were conducted with the trimmed basic construction data set. The analyses with the augmented high school data set, however, were more fine-grained because of the availability of additional variables. Comparison of means were conducted for (1) size (smaller vs. larger); (2) district wealth (lower vs. higher); (3) locale (rural vs. nonrural); and (4) locale by wealth (rural, nonrural x low-wealth, high-wealth). Stepwise regression analyses were conducted on three dependent variables (namely, total cost, cost per student, and cost per square foot) with, in each case, eight powerful predictor variables (namely, district size, district revenue per student, percentage of adult population attaining high school graduation, percentage of families above the poverty line in the district, planned school enrollment, project end date, rurality [dummy variable], and percentage of African-American families in the district [log transform]). The variables represented the most recent data that could be obtained at the time of the import procedure—varying from 1997 to 2001 depending on the variable. Variable measurements, then, were roughly coeval with the reported construction time-span for these schools (1994-2003). It is useful to note that cost per student is simply the product of cost per square foot and square feet per student (cost/student = cost/ft2 * ft2/student). Logically, for each school, cost per student is determined by these two variables. Empirically, however, their relationship to one another across a large number of schools might be anticipated to exhibit some anomalies. That is, the relationship of cost per student and cost per square foot might be mediated systematically by indirect influences on building footprint.

10

Analyses with the trimmed basic construction data set. The trimmed basic construction data set, as noted previously, has comparatively few variables available for analysis (data set prepared for analysis as described previously). The researcher conducted two series of analyses: (1) for all schools by level (elementary, middle, and high school) and (2) for reasonably sized schools by level. Mean differences reported apply to differences by schools divided at the median of enrollment in each group (i.e., the set used for each series of analysis—all schools by level or reasonably-sized schools by level). The upper limits for “reasonably sized schools” are those adopted in Dollars and Sense (Lawrence, B. et al., 2003): 500, 750, and 1,000 students (in planned enrollment in this study) for elementary, middle, and high schools, respectively. In each instance, schools planned to enroll fewer than the median number of students are “smaller” schools and those planned to enroll at least the median are “larger” schools. Table 3 reports the applicable medians. Table 3. Median size for all schools by level and comparison group

level comparison group

N of

schools

median enrollment

elementary all schools 1,681 600 reasonably-sized schools 609 400 middle all schools 720 750 reasonably-sized schools 371 575 high school all schools 723 912 reasonably-sized schools 407 567

Table 3 discloses two trends worth noting at this juncture. First, observe that in this data set, among elementary and middle schools, the medians for the two school groups (all schools and reasonably sized schools) are closer than are the medians for the two high school groups (67% and 77% for elementary and middle schools, as compared to 62% for high schools). Second, and consistent with this observation, observe that 36% of all elementary schools, 52% of all middle schools, and 56% of all high schools can be classified as smaller than the respective upper limits of “reasonable size.” These trends suggest that, overall, planning for high school construction in 1989-2002 could be considered more reasonable with respect to size than planning for middle schools, and much more reasonable than planning for elementary schools. The analyses for the trimmed basic construction data set include, for each school group (all schools and reasonably sized schools), the following: (1) descriptive statistics (total cost, planned enrollment, total square feet planned, cost per student, cost per square foot, and square feet per student); (2) tests of mean differences in total cost, cost per student, cost per square foot, and square feet per student; (3) correlations among these variables; and (4) regressions of total cost per student on planned enrollment, total square feet, and projected end date. Even with just three predictor variables, the proportion of explained variance is comparatively high, ranging from about 45% to about 60% across the numerous analyses.

11

Results The specific research quantitative questions addressed by this study are (1) whether or not planned costs of construction are equal for smaller as compared to larger schools and (2) the extent to which cost-related construction variables can be predicted by contextual variables including locale, socioeconomic status, and ethnicity. Results are reported first for the augmented high school data set, then the trimmed basic construction data set (pages o-p). Tables are provided for selected results because of the large number of analyses completed; some analyses are described narratively, without the support of tables. Results for the augmented high school data set Results for the augmented data set of reasonably-sized high schools are reported in two sections. The first describes results for 9-12 high schools and the second describes results for 6-, 7- , 8-12 high schools.

Results for 9-12 high schools. Table 4 presents descriptive statistics (mean, standard deviation, and skewness) for 16 selected variables of interest.

Table 4. Descriptive statistics for selected variables (9-12 schools, augmented data set) variable mean SD skewness

planned enrollment 590 203 - 0.12 actual enrollment 01-02 589 364 + 0.19

planned cost $12,795,140 $8,290,367 + 1.79 planned ft2 122,722 57,138 + 1.02

planned cost per ft2 $102.43 $40.74 + 1.82 planned cost per student $21,508.58 $10,792.78 + 1.41

planned ft2 per student 215.49 85.70 + 0.94 school subsidized meal rate 25.93% 17.88% + 1.17

school percent minority 14.98% 21.08% + 2.11 school percent African American 6.28% 14.46% + 3.71 district adults w/at least HS grad 35.02% 7.37% - 0.22 instructional expend per student $4,065.31 $668.38 + 0.85

district revenue per student $7586.95 $1,601.68 + 1.48 district membership 2,791 4,710 + 6.65

district families above poverty line 89.59% 8.36% - 1.57 rurality (1= rural) .49 .50 + 0.24

note. n =168 schools from 30 states (AL, 4; CA, 1; CO, 5; CT, 2; DE, 1; GA 9; IA, 2; ID, 5; IN, 1; KS, 2; KY, 5; MA, 3; ME 1, MI, 24; MS, 7; NC, 1; NE, 3; NH, 2; NY, 1; OH, 27; OK, 3; OR, 2; PA, 4; SC, 1; TN, 11; TX, 33; UT, 2; VA, 1; WA, 2; WV, 3)

Several pertinent observations about the values in Table 4 can be made. First, the mean planned and actual enrollments are identical, though the actual enrollments are predictably more variable than the planned enrollments. Second, costs per square foot in this data set are consistent with those reported by Abramson for 2003, with $108 cited as the national median cost per square foot for new schools planned to enroll fewer than 850 students (Abramson, 2004). Third, and relatedly, the restrictive selection rules (unique zip code, high school unique in district) did not produce sharply different means on cost. The fact that the values are close is some indication of generalizability, in fact, as does the representation of 30 states among the 168 cases. Fourth, rural schools (CCD locale codes 7 and 8) comprise about half the sample

12

(compared to the national prevalence of about 30%). Our selection rules are doubtless responsible for this difference (high schools located in a zip code unique to the database, enrolling fewer than 1,000 students in districts maintaining a single high school)—rural schools enrolling fewer students on average than suburban and urban schools. Table 5 reports the CCD locales for all schools in the augmented data set. One locale (large town) is missing from the data set; this locale is the least prevalent nationally and its absence is predictable. Table 5. Locale among schools in augmented data set

locale

N

large city 1 mid-size city 3 fringe of large city 23 fringe of mid-size city 16 small-town 42 rural (nonmetro) 42 rural (metro) 41 Total

168

The discussion first considers planned construction costs for smaller versus larger 9-12 high schools (divided at the median of planned enrollment). Table 6 reports the results, comparing smaller and larger schools across a number of variables, focusing on 4 cost-related variables (total cost, cost per square foot, cost per students, and square feet per student) but also including comparisons on 6 contextual variables. Interestingly, 3 of the 4 cost-related variables exhibit significant differences. By comparison, none of the 6 contextual variables exhibits significant differences. Smaller and larger schools exist in contexts that do not differ on average; means and variability are very similar.

13

Table 6. Cost differences, smaller versus larger planned enrollments in augmented data set

smaller schools (n=87) larger schools (n=81) variable mean SD mean SD

planned enrollment**** 426 118 767 100

actual enrollment (01) 487 359 699 338planned cost**** $9,530,127 $5,570,334 $16,302,007 $9,277,583

cost per ft2** $95.73 $35.89 $109.61 $44.48cost per student $22,171.65 $11,332.37 $20,796.40 $10,203.66

ft2 per student**** 238.94 98.62 190.32 60.30state rev per student 56% 15% 54% 14%

rurality .49 .50 .49 .50high school grads 34% 8% 36% 7%

district mdn fam inc $46,465.69 $12,158.86 $48,662.81 $12,370.11percent fam nonpov 89% 8% 91% 8%dist revenue per stud $7,490.05 $1,612.59 $7,688.62 $1,593.79

note 1. ****p<.0005; * p < .05 note 2. Schools divided on median of planned enrollment, i.e., 600 students. note 3. Schools from 27 states represent smaller schools (36 of 87 cases from MI and TX). note 4. Schools from 18 states represent larger schools (31 of 81 cases from MI and OH). With respect to costs, smaller schools are less expensive than larger schools per square foot, but they are no less expensive per student than larger schools. Further, the smaller schools in this data set allocate 26% more space to each student as compared the larger schools. Table 6 distinguishes smaller from larger schools on the basis of planned enrollment. Planned enrollment, however, is arguably not a fair representation of the size of an actual school. Planned enrollments may not, in fact, materialize. Indeed, the extent to which planned and actual enrollments are related is easily assessed in the augmented high school data set by correlating planned enrollment and actual (2001) enrollment; the correlation is r = .42; the two measures of size share just 16% of variance. Confirmation of this meager relationship suggests the appropriateness of comparing schools according to actual, rather than planned, size. The dependent cost variables, however, remain the same (cots, square feet, and cost per planned enrollment and per square foot). It might be objected that the enrollment-proportional cost factor should be cost per actual enrollment, but at the point of cost estimation and contract award, cost per student was cost per planned enrollment; actual enrollment was necessarily unknown. However, dividing schools into smaller versus larger on the basis of actual enrollment seems the more appropriate decision, since smaller and larger size is a matter of “actual” reality, not planned “reality.” Table 7 reports comparable statistics for the 168 grade 9-12 high schools on the basis of these decisions.

14

Table 7. Cost differences, smaller versus larger actual enrollments in augmented data set smaller schools (n=84) larger schools (n=84)

variable mean SD mean SD

planned enrollment**** 512 201 668 173actual enrollment (01)**** 339 117 839 356

planned cost**** $9,668,289 $6,740,692 $15,921,992 $8,547,602cost per ft2 $99.80 $45.94 $105.05 $34.85

cost per student** $18,895.83 $10,353.47 $24,121.33 $10,647.07ft2 per student* 200.26 94.24 230.73 73.67

state rev per student**** 59% 12% 51% 16%rurality**** .64 .48 .35 .48

high school grads 36% 7% 34% 8%district mdn fam inc*** $44,440.54 $8,888.24 $50,609.50 $14,312.64

percent fam nonpov 89% 9% 90% 8%dist revenue per stud $7,592.71 $1,746.13 $7,581.18 $1,453.61

note 1. *** *p<.0005; *** p < .001 ** p < .01; * p < .05; note 2. Schools divided on median of actual 2001 enrollment, i.e., 552 students. note 3. Schools from 19 states represent smaller schools (31 of 84 cases from MI and OH). note 4. Schools from 27 states represent larger schools (37 of 84 cases from MI, OH, and TX).

This alternative analysis gives results somewhat different from those reported in Table 6. The pattern of results, however, is consistent. First, once again three of four cost-related measures show significant differences. The results for smaller schools in Table 7 are the converse of those in the results given in Table 6: here, cost per square foot is not significantly different in larger and smaller schools, whereas cost per student is lower in smaller schools. Also, planned square footage per student is less in smaller schools than in larger schools in Table 7, and this is the opposite of the result in Table 6. In Table 7 (but not in Table 6) three of the six contextual variables are significant—state revenue is proportionately higher in districts that operate the smaller schools, the smaller schools are located more frequently in rural places, and the median family incomes are lower in the districts in which the smaller schools are located.

Where does the pattern consistency lie among all these individual differences? Together Table 6 and Table 7 rather consistently suggest that, if anything, smaller as compared to larger “reasonably sized” 9-12 high schools (those enrolling no more than 1,000 students) are less costly to build (per square foot according to Table 6, and per student in Table 7). Neither analyses suggests that larger reasonably sized high schools are less costly to build. The results for smaller schools are the same whether or not one distinguishes between larger and smaller schools as planned or as subsequently (in reality) operated. The ensuing analyses report comparisons based on planned enrollment only, on the view that construction decision making is founded on plans, not on an impossible foreknowledge of subsequent reality. The analyses in Tables 6 and 7 nonetheless attest to the robustness of the findings that address the main research question. Smaller reasonably-sized high schools are no more, and probably less, costly to build than larger reasonably sized high schools.

15

The augmented data set permits a comparison of construction costs for smaller and larger high schools in more and less well-resourced locations. Two such comparisons seem salient: (1) smaller and larger schools in low- and high-wealth districts and (2) smaller and larger schools in rural and non-rural locales. Table 8 reports average cost-related data for smaller as compared to larger schools in these wealth and locale categories. Table 8. Average costs for smaller and larger schools by wealth and by locale. Panel 1. costs for size by wealth

size wealth $/ ft2 $/Student ft2/studentsmaller lower $89.36 $20,183.18 234.7 higher $103.72 $24,595.68 246.0larger lower $96.66 $18,171.53 185.0 higher $119.98 $22,896.30 194.6total lower $92.53 $19,310.65 213.2 higher $112.54 $23,674.33 218.1

Panel 2. costs for size by locale size locale $/ ft2 $/Student ft2/studentsmaller rural $102.47 $21,262.38 219.6 non-rural $101.46 $24,753.90 243.9larger rural $104.82 $19,485.94 187.4 non-rural $112.94 $22,033.08 194.8total non-rural $107.87 $23,235.97 216.5 rural $103.53 $20,459.12 205.0

note 1. Annual total revenue per student median = $7,395.21; lower-wealth < median; higher wealth ≥ median. note 2. Size categories determined at median of planned size (600 students). note 3. Rural and non-rural are represented by an equal number of cases. The findings reported in Table 8 are consistent with what might be predicted on the basis of extant information about locale and wealth differences. That is, one would expect average costs to be lower in rural and low-wealth districts. Arguably, however, expenditures on construction—and not, strictly speaking, the costs of purchasing the same thing—will be lower in districts with lower annual revenues and in rural communities. These predictions are borne out in Table 8. Rural schools and low-wealth schools are built for less (per square foot and per student) than non-rural and high-wealth schools. This pattern characterizes smaller and larger schools within wealth and locale categories, as well, with one exception. Cost per square foot in smaller rural schools is about the same as in smaller non-rural schools. One might also expect that higher-wealth and non-rural schools would provide more space per student, and this supposition is also confirmed by the analysis represented in Table 8, where the distinction is consistent for smaller and larger schools within wealth and locale categories. One caveat pertains to Table 8: differences of statistical significance are not reported. Of greatest interest is the character of the overall pattern, which is quite consistent, not which differences are statistically significant. There is ample reason to suspect that this pattern of differences would prove statistically significant in a large, representative data set.

16

The second research question was “How do construction costs vary in light of school and context variables including planned enrollment, square footage, district revenues, locale, socioeconomic conditions, and ethnicity?” The study approached this question via a series of regression analyses that employed a uniform set of nine predictor variables. The predictor variables were (1) total planned square feet, (2) planned enrollment, (3) average district enrollment for 1997-2001, (4) average district revenue per student for 1997-2001, (5) proportion of adult population 25 and older who were high school graduates in 2000, (6) percent of district families with children living above the poverty line in 2000, (7) percentage of students in the school who were African American (natural logarithm transform), (8) planned completion date, and (9) rurality (rural=1). Total cost is used as a predictor variable in the regression for square feet per student, where the values of the variable are divided by 1,000,000 to render the regression coefficient interpretable (transformation of a variable by a constant alters neither the underlying distribution nor the findings). The analyses regressed four dependent variables on this set of predictors: (1) total cost in dollars, (2) cost in dollars per square foot, (3) cost in dollars per student, and (4) total feet square per student. Variables in the predictor set that were used to compute any of the dependent variables were excluded from use as predictor variables in that analysis, as follows: (a) Total feet square per student was excluded as predictor variable in the regression that used cost per square foot as the dependent variable; (b) total planned enrollment was excluded as a predictor from the regression for cost per student; and (c) both square feet and planned enrollment were excluded as predictors from the regression for square feet per student. In all cases, the analyses were conducted as stepwise regressions. Because there is so little empirical research on either scale-related school construction or on school construction costs at all, hypothesizing any particular relationships is difficult so far as cost per square foot, cost per student, of square feet per student is concerned. Hypothesizing influences on total cost, however, is perhaps less problematic. Among the predictor variables, those numbered 1-6 would likely exhibit positive relationship to total cost: square feet and enrollment are principal cost factors in any construction project. District enrollment is a measure of district size and is related in most educational and sociological research to urbanicity and to measures of affluence—each of which influences costs upward, indirectly and perhaps directly. Revenue per student, high school graduates (a measure of educational attainment), and percent of families above the poverty line are all related to local purchasing power, which can be predicted to increase contracted cost (affluent communities simply buy more). Planned completion date embeds a measure of inflation, such that later dates will be associated with higher costs. Rurality would be predicted to be associated with lower total costs because wages are lower in rural locales than elsewhere. Which variables will exert effective influence in an OLS regression model, however, cannot be predicted. Table 9 reports the results for the regression analyses. Panel 1 reports results for total cost, panel 2 for cost per square foot, panel 3 cost per student, and panel 4 square feet per student. Information about only the final step of the regressions is reported in Table 9. For all regressions, however, every applicable predictor was specified for possible entry (except as noted previously in the case of variables used to construct the dependent variables).

17

Table 9. Regression for variables predicting cost-related measures of new high schools (n=168) Panel 1: total cost in dollars

variable B s.e. B β t p square feet 115.122 5.08 .800 22.662 .000 end date 758,036.90 139217.05 .191 5.445 .000 revenue per student 704.89 170.29 .146 4.139 .000 rural -1,498,198.82 575,478.73 -.091 -2.603 .010 families above poverty line 82,402.92 35,708.28 .081 2.308 .022

note. First step adjusted R2 = .68; ∆R2 from first to fifth step = .08; fifth step adjusted R2 = .76 Panel 2: cost in dollars per square foot

variable B s.e. B β t p end date 5.380 1.237 .273 4.348 .000 revenue per student .006 .001 .271 4.366 .000 planned enrollment .027 .012 .138 2.197 .029 families above poverty line .674 .314 .133 2.143 .033

note. First step adjusted R2 = .12; ∆R2 from first to fourth step = .13; fourth step adjusted R2 = .25 Panel 3: cost in dollars per student

variable B s.e. B β t p square feet .102 .010 .541 9.938 .000 revenue per student 1.703 .344 .269 4.948 .000

note. First step adjusted R2 = .33; ∆R2 from first to second step = .07; second step adjusted R2 = .40 Panel 4: square feet per student

variables B s.e. B β million dollars of planned cost 4.304 .646 .412 6.666 .000 end date -14.112 2.565 -.340 -5.502 .000

note. First step adjusted R2 = .12; ∆R2 from first to second step = .11; second step adjusted R2 = .23 Table 9 reveals a number of tendencies. First, most of the variance (76%) in the total cost of reasonably-sized 9-12 high schools (represented in the augmented data set) is accounted for by five variables. By far the most influential variable is square feet (as would be expected).

18

For all other cost-related dependent variables, the predictor variables account for less than half the variance (40% for cost per student; 25% for cost per square foot, and 23% for square feet per student). For the total cost equation, which includes five of the nine predictor variables, square feet alone accounts for 68% of the variance in cost (β = .80). The influence of the five variables in Table 9 can be interpreted directly from the unstandardized regression coefficients (the “B” values) as follows:

(1) Every thousand square feet increases mean total cost by about $115,000 ($115 per foot times 1000), all else equal.

(2) For the schools in this data set, a one year change in construction completion date increases costs by about $758,000 (β = .19).

(3) Every additional one-hundred dollars in revenue per student is associated with a mean cost increase of about $7,000 dollars (β = .15).

(4) Every percent of increase in families with incomes above the poverty line predicts an $82,000 increase in total cost (β = .08) .

(5) The only negative total cost influence is whether or not the school is located in a rural place. That condition, all else equal—i.e., with the influence of square feet, end date, and the other predictors statistically removed—deflates mean total cost by an estimated $1.5 million (β = -.09).

Cost per student is predicted by square feet and by revenue per student. Building footprint exerts the stronger influence (β = .54), but with that influence controlled, district funding affluence separately influences cost per student (β = .27). Every hundred additional square feet increases mean cost per student by about $1.00, and every dollar of additional revenue per student increases mean cost per student by $1.70. With respect to the influence of revenue per student, one might hypothesize that, whatever the building footprint, decision makers in more affluent communities purchase more costly amenities (quality of flooring, faΗade treatment, HVAC options and so forth). Cost per square foot is predicted by four variables, mostly strongly (β = .27) by end date (inflation factor) and revenue per student (β = .27). Every additional year for completion date is associated with a $5.38 increase in mean cost per square foot, and every $1000 of additional revenue per student increases cost per square foot by about $6.00. Weaker influences are planned enrollment (β = .14) and percentage of district families with incomes above the poverty line (β = .13). Every increment of 100 students of planned enrollment increases mean cost by $2.70 per square foot and every 10% change in the proportion of families above the poverty line increases cost per square foot by an estimated $6.70. Square feet per student is predicted by two variables of approximately equal influence, million dollars of planned cost (β = .41) and end date (β = -.34). Each million dollars of planned total cost is associated with an additional 4.3 square feet per student. In this equation, end date is uniquely associated with a reduction in square feet per student of approximately 14 square feet per student. All else equal, this finding could indicate that one way planners have countered rising total costs is to reduce the footprint of schools, and this reduction over time would logically translate into a reduced space allocation per student.

19

Results for 6-, 7-, and 8-12 high schools. The sample size for these schools is one-fourth that for the 9-12 schools. This fact limits the sorts of analyses that can be undertaken. For this reason, the study limits inquiry to the first research question: “Are planned costs of construction are equal for smaller as compared to larger schools?”

Table 10 provides descriptive statistics for the same set of selected variables used to describe the 9-12 schools in Table 4, with the addition of two measures of cohort enrollment. These variables are the relevant school size measures for this data set, which comprises schools of 5-7 grades. Cohort size is merely total enrollment (planned or actual) divided by the number of grade levels. This method of computing school size removes the confounding influence on size of gradespan. Table 10. Descriptive statistics for selected variables (6-12 schools, augmented data set)

variable mean SD skewness planned enrollment 496 219 + 0.47

actual enrollment 01-02 398 221 + 0.77 planned cohort enrollment 84.5 42.4 + 0.88

actual cohort enrollment 67.8 40.6 + 0.96 planned cost $11,689,487 $9,754,693 + 2.13

planned ft2 92,186 49,250 + 0.54 planned cost per ft2 $118.75 $55.79 + 2.36

planned cost per student $22,890.21 $13,460.22 + 1.31 planned ft2 per student 190.68 82.95 + 1.60

school subsidized meal rate 25.75% 13.62% - 1.11 school percent minority 9.76% 13.30% + 2.68

school percent African American 3.27% 6.95% + 2.81 district adults w/at least HS grad 38.27% 8.15% + 0.21 instructional expend per student $4,280.48 $1,228.71 + 2.21

district revenue per student $7903.28 $2,059.03 + 2.27 district membership 1,925 2,738 + 2.91

district families above poverty line 89.31% 6.66% - 0.21 rurality (1= rural) .78 .42 - 1.41

note: n = 41 schools from 18 states (AL, 1; AR, 2; FL, 1; GA, 1; ID, 3; IN, 1; MA, 3; MI, 4; MS, 1; NM, 1; NY, 2; OH, 9; TN, 2; TX, 5; VA, 1; WA, 1, WV, 1, WY, 2)

20

Inspection of data in Tables 10 and 4 reveals variables of likely difference. The mean differences are sufficiently numerous to warrant the separate analyses reported here. Six variables common to both data sets in fact exhibit significant differences (five variables with p < .01). For 6-12 and 9-12 schools, respectively, the variables exhibiting significant mean differences are: (1) planned enrollment (496 v. 590), (2) actual enrollment (398 v. 589), (3) planned square feet (92,000 v. 123,000), (4) planned cost per square foot ($119 v. $102), (5) percentage of adults with high school diploma (38.3% v. 35%) and (6) locale (78% rural v. 49% rural). The schools seem to comprise an arguably different population, much as do elementary, middle, and secondary schools.

Table 11 reports statistics for three cost-related variables in smaller as compared to larger

schools—based on the division of the sample at the median of planned cohort enrollment (74 students per grade). Only total planned cost shows a statistically significant difference—larger schools, of course, are more costly. Comparisons per unit cost (per square foot and per student) are not more costly. The small sample size (about 20 schools in each size category) makes conclusions somewhat problematic, and a larger sample (e.g., n = 150) could confirm additional significant differences in either cost per square foot or cost per student, if not both. The observed differences for 6-12 schools favor smaller schools. The small sample size simply means that observers cannot accept these differences as real and not as the artifacts of mere chance. Table 11. Costs for smaller as compared to larger 6-12 high schools (n = 41)

smaller schools larger schools variable mean SD mean SD p power

planned cost** $7,378,524 $5,043,184 $16,216,000 $11,464,885 .003 .497

cost per ft2 $110.44 $41.04 $127.48 $68.00 .335 .707cost per student $22,265.62 $12,727,66 $23,546.04 $14,492.18 .765 .846

note 1. schools divided by size at the median of cohort enrollment (74 students per grade) note 2. ** p < .01 note 3. n = 21 smaller schools in 11 states (OH, 5; TX, 3; AR, ID, MI, WY, 2; FL, GA, MA MS, TN, 1) note 4. n = 20 larger schools in 13 states (OH, 4; MA, MI, NY, TX, 2; AL, ID, IN, NM, TN, VA, WA, WV, 1)

In all the comparisons for this report that address the first research question, the particular concern with size centers around the historical presumption that smaller schools are proportionately more expensive to build (per square foot or per student). Small sample sizes make it more difficult for analyses to reject the null hypothesis (of no significant difference) and thereby affirm a difference. In this case, with a very small sample, it is fair to ask if the analysis has overlooked a result that, though not evident, is nonetheless present. That is, the study can and should assess the probability that it has failed to identify a difference that really exists (type II error). Again, this is a particular concern with a very small, nongeneralizable sample such as this one for 6-12 high schools. The probability of avoiding a type II error is reported in Table 11 (column labeled “power”). The last two columns should be interpreted as follows: The observed differences favoring smaller may well be the result of chance (p = .335 and .746), but it is nonetheless unlikely that the analysis has overlooked a difference that really exists (power = .707 and .846). The analysis, in other words, is even more likely to have avoided a Type II than a Type I error.

21

Research question two is more difficult to approach in this small sample of 6-12 schools, which presents limited opportunities for exploring predictor variables. The study did conduct regression analyses similar to those for 9-12 schools, and results are summarized narratively, without tables.

Total cost is well predicted (adjusted R2 = .85) by four independent variables: (1) total

square feet, the strongest influence (β = .81, explaining 75% of variance); (2) revenue per student; and (3) rurality. The latter two variables account for the additional 10% in explained variance. Rurality is a negative influence on total cost, as in the regressions reported in Table 9.

Cost per square foot is associated with a single significant predictor, revenue per student

(β=.64, adjusted R2 = .36). District affluence rather significantly influences total cost upward among these schools, in contrast to 9-12 schools; for these schools, as well, this single predictor accounts for nearly 50% more in explained variance over the four predictors in the comparable equation for 9-12 schools.

Cost per student in these 6-12 schools is jointly predicted by two variables, square feet

and revenue per student, both positively and at approximately equal strength of influence (β = .54 and .44, respectively). Together these two variables explain nearly half the observed variance (adjusted R2 = .47).

Square feet per student is predicted weakly and solely by total cost (β=.36, adjusted R2 =

.10). More costly buildings in the small sample tend to provide more proportionately more space than less costly buildings.

With a clearer grasp of the differences between 6-12 and 9-12 high schools in this data

set, one additional analysis relevant to research question two (How do construction costs vary in light of school and context variables including planned enrollment, square footage, district revenues, locale, socioeconomic conditions, and ethnicity?) an additional regression analysis is possible. This analysis uses high school type (9-12 or not 9-12) as a predictor variable in a combined high school data set (n = 209, i.e., excluding the 1-12 school and the 10-12 school). This analysis used cost per square foot and cost per student as dependent measures. Cost per square foot was not significantly influenced by high school type, but cost per student was. Table 12 reports the results. Table 12. Regression of cost per student for 9-12 and 6-12 high schools.

variable B s.e. B β t p square feet .105 .011 .554 9.950 .000revenue per student 1.708 .345 .269 4.952 .0009-12 high school -3719.051 1524.72 -.137 -2.439 .016rural -2591.732 1193.45 -.120 -2.172 .031

note 1. n = 206 due to missing data; adjusted R2 = .416 note 2. dummy variable values: rural = 1; 9-12 high school =1

22

The comparable analysis for 9-12 high schools alone (n = 168) is given in panel 3 of Table 9, and for 6-12 schools only (n = 41) in the preceding paragraphs. With 6-12 schools represented in the data set and with a dummy variable for high school type (9-12 school = 1), four variables predict cost per student in the combined data set. In contrast to the regression reported in panel 3 of Table 9, both rural and 9-12 school enter the equation as significant predictors, explaining slightly more variance (42% vs. 40%). Both variables tend to reduce cost per student, but their unique influence in this regard is comparatively weak (i.e., together they explain an additional 2% of the observed variance in cost per student, once the strong influence of square feet and revenue per student is taken into account).

Summary of findings for research question 1 in the augmented data set. Are planned

costs of construction are equal for smaller as compared to larger high schools? The clear preponderance of evidence is that high schools ranging from rather small to the

upper limit of reasonable size (1,000 students in grades 9-12) are equally costly to build by proportionate measures (cost per square foot and cost per student). The smallest high school in this data set was planned to enroll 138 students, and the largest 962 students.

Once the decision is made not to build very large schools, costs of construction for larger

and smaller high schools cannot be said to vary appreciably by planned enrollment. In other words, for reasonably-sized schools, economies of scale for construction costs were not in evidence in this data set.

For analyses based on planned enrollment of 9-12 schools, cost per square foot, but not

cost per student, was significantly less (by about $15.00 per square foot) in smaller schools. For analyses based on actual subsequent enrollment of 9-12 schools, cost per student, but not cost per square foot, was significantly less in smaller schools (by about $1,300 per student). Larger schools were not shown to be significantly less expensive than smaller schools on either measure. For 6-12 schools, no significant differences in cost per square foot or cost per student were found in any analyses.

Summary of findings for research question 2 in the augmented data set. Research

question 2 asks, “How do construction costs vary in light of school and context variables including planned enrollment, square footage, district revenues, locale, socioeconomic conditions, and ethnicity?” In light of the findings from question 1, research question 2 asks, “What does predict cost?”

Total cost is well predicted for 9-12 high schools (in declining order of strength of

influence) by footprint, inflation (construction end date), local revenue wealth, rural location, and proportion of families not living in poverty. Rural location tends to reduce cost, whereas the other predictors are associated with increased costs. For 6-12 high schools, most of the variance in total costs (85%) is accounted for by footprint, local revenue wealth, and rural location. Here too, rural location tends to reduce costs.

Cost per square foot is less well predicted, but for 9-12 schools, inflation (construction

end date), local revenue wealth, planned enrollment, and proportion of families not living in poverty account for 25% of the observed variance. For 6-12 schools, 36% of the variance is accounted for by local revenue wealth alone.

23

Cost per student is more fully predicted than cost per square foot. For 9-12 schools, 40%

of the observed variance is accounted for by footprint and locale revenue wealth. For 6-12 schools nearly half of the variance (47%) is accounted for by the same two variables. The study also found that cost per student could be slightly better explained (42% instead of 40% of variance accounted for) in a combined 6-12 and 9-12 data set (n = 209) by two additional variables, high school gradespan configuration (i.e., 6-12 or 9-12 school) and rural location. The 9-12 configuration and rural location were shown to exert weak (but statistically significant) negative influences on cost per student.

Results for the Trimmed Basic Construction Data Set (N=3,124)

The trimmed basic construction data set provides the opportunity to compare the cost-related variables considered in the augmented high school data set (total cost, cost per square foot, cost per student, and square feet per student) for two types of size comparison and three levels of school. The three levels are (1) elementary, (2) middle, and (3) secondary. The two size types are (1) smaller and larger schools among all schools at the given level (n=3,124) and (2) smaller and larger reasonably-sized schools (upper limits of 500, 750, and 1,000 for elementary, middle, and secondary schools, respectively; n=1,387).

The data set does not specify grade span configuration (Abramson, 2004), so it is not

possible in these comparisons to use students per grade to scale the school size measure; instead, as in the 9-12 analyses for the augmented data set, the median of total enrollment is used to divide smaller and larger schools. Many studies employ this measure, but, as noted above, when comparing school size among schools with varied gradespans, total enrollment is a troublesome measure of size (Howley, Craig, 2000).

The analysis plan provides 6 tests of the null hypothesis embedded in the first research

question, “Are planned costs of construction are equal for smaller as compared to larger schools?” Because contextual variables are absent from the trimmed full data set, however, approaching the second research question, which centers on the influence of contextual variables, is difficult to say the least. Nonetheless, some attention is given to regressions of cost and cost per student on some of the other variables in the data set, which for the purposes of the second research question constitute a sort of limited context. That is, a given total cost for a project, for instance, does exist in the “context” of that project, which context can be described as entailing a certain anticipated enrollment and a certain specification of square feet. Some circumscribed conclusions have been developed on this basis.

Results for research question 1 (trimmed full data set). In the analyses that follow, dividing the sample at the median produces groups of differing numbers of cases rather than the equal number of cases that might be expected. The reason for this is that planned enrollments cluster around appealing values—multiples of 10, 50, and 100. At the elementary level, for instance, 186 schools have planned enrollments at the median (600 students), whereas only 2 schools fall at the median for middle schools (750) students), and only 1 school at the median for high schools (912 students). Clearly the elementary data set presents a problem for the planned analysis. The approach adopted by the study was to assign the median value to the “larger” half, because the median falls above the study’s criterion for “reasonable” size at the elementary level (i.e., 500 students). Results were calculated for the alternative coding, however, and were not found to differ.

24

Before addressing the research question, this report establishes the significant cost and size differences that prevail by educational level in this data set. Table 13 reports descriptive statistics for the data set. Medians are included for comparison with those reported by (Abramson, 2004). Table 13. Descriptive statistics by educational level, trimmed basic construction data set panel 1: all schools

level variable mean mdn SD total square feet 73,755.02 70,000.00 27,948.34elementarya planned enrollment 609.55 600.00 231.88 total cost $8,347,206.30 $7,400,000.00 $4,360,566.90 total square feet 111,861.07 109,500.00 42,356.21middleb planned enrollment 774.51 750.00 314.33 total cost $12,285,831.63 $11,000,000.00 $6,200,057.12 total square feet 177,377.13 160,000.00 96,460.76highc planned enrollment 1,063.26 912.00 660.28 total cost $19,740,985.98 $17,180,000.00 $12,557,578.48

note 1. a n = 1,681; b n = 720; c n = 723 note 2. Mean differences by level are significant at p <.0001 for each variable. panel 2: reasonably sized schools

level variable mean mdn SD total square feet 62,065.97 60,000.00 27,060.640elementarya planned enrollment 381.04 400.00 112.72 total cost $7,072,361.90 $6,260,000.00 $3,918,955.15 total square feet 89,138.88 88,485.00 34,280.32middleb planned enrollment 537.09 575.00 157.78 total cost $9,841,491.80 $9,000,000.00 $5,185,683.74 total square feet 124,877.67 120,000.00 61,343.09highc planned enrollment 583.89 567.00 249.17 total cost $13,334,546.80 $11,763,200.00 $8,141,073.12

note 1. reasonable size defined as planned enrollment less than 500, 750, and 1,000 for elementary, middle, and

high schools, respectively. note 2. a n = 609; b n = 371; c n = 407 note 3. Mean differences by level are significant at p <.0001 for each variable.

25

The two panels in Table 13 show total square feet, planned enrollment, and total cost for all school and for reasonably-sized schools, but without reference to the smaller and larger schools in each group (those contrasts appear in Tables 14 and 15). Reasonably-size schools have a smaller footprint, enroll fewer students, and have lower price tags than all schools—exactly what one would expect.

Several observations are worth articulating with respect to the data in Table 13. First, the

major cost-related factors also clearly and robustly differ by educational level whether for all schools or for reasonably sized schools. This fact warrants the separate analyses by level that were planned, consistent with (Abramson, 2004).

Second, the reduction in footprint for smaller schools is not proportional to the reduction

in enrollment. The median planned enrollment for all elementary schools, for instance, is 600, and for reasonably-size schools it is 400—a reduction of 33 percent. Foot print for elementary schools, however, differs by just 14 percent (from 70,000 to 60,000 square feet). Such disparities vary by level, but characterize middle and high schools as well. Reasonably-sized schools allocate more space per student on average than do all schools.

Second, the total cost differential (again using elementary schools as the example) is also

not proportional to the difference in median size for the two groups. The 33% reduction in enrollment is reflected in a 16% difference in total cost ($7.4 million vs. $6.26 million). The proportionately larger footprint for reasonably-size schools would likely be associated with a higher cost per student.

Third, differences in planned enrollments across levels (with elementary schools the

reference level) differ by group. Among all schools, middle schools enroll 25% more students than elementary schools, and high schools enroll 52% more. Among reasonably-size schools, however, middle schools enroll 44% more students than elementary schools, and high schools enroll 42% more. This pattern is reflective of the fact that total planned enrollment is not at all distributed randomly, but tends to clump, and, moreover, “clumps” rather differently by level and group. Among all schools, for instance, elementary and middle schools exhibit overall negative skewness (the clumping is at larger enrollments). For reasonably sized schools, however, the skewness is positive (clumping at smaller enrollments). The degree of skewness, however, is comparatively modest (usually less than 1 in either direction).

Table 14 and Table 15 succinctly present a great deal of information about mean

differences between smaller and larger schools, at each level, and for each group. Table 14 reports means and standard deviations for two groups of schools—all 3,124 schools (the “A” rows in the Table) and 1,387 reasonably sized schools (the “RS” rows in the Table)—and within each group for three levels (i.e., elementary, middle, and high schools). Table 15 uses the Table 14 means to compute mean differences, where they are subjected to tests of significance and assessed for effect size (effect sizes are reported only for mean differences that prove statistically significant—non significant differences are considered to have an effect size of 0).

26

Table 14. Descriptive statistics by level and size category (all schools and reasonably-sized schools)

note 1. A = all schools (n = 3,124); RS = reasonably-sized schools (n = 1,387) note 2. for all schools, n’s for smaller and larger schools, respectively: elementary (747, 934), middle (371, 349), high (362, 361) note 3. for reasonably-sized schools, n’s for smaller and larger schools, respectively: elementary (320, 289), middle (187, 184), high (204, 203)

Elementary Middle High

smaller larger smaller larger smaller larger

variable Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD $/ft2 A

$114.63 $37.74 $114.33 $39.77 $109.42 $34.75 $109.21 $32.41 $103.53 $31.55 $116.519 $33.24

RS $114.94 $37.75 $115.51 $38.83 $105.38 $33.22 $113.53 $35.87 $99.54 $30.64 $109.79 $31.34

$/stud A $19,464.50 $13,392.54 $12,151.55 $5,215.79 $18,801.16 $8,938.13 $14,733.65 $6,028.73 $24,024.89 $12,523.75 $17,805.92 $7,898.28

RS $24,733.75 $16,454.43 $16,013.30 $9,230.29 $19,501.96 $10,033.32 $18,088.93 $7,628.70 $26,028.54 $13,678.40 $21,494.37 $10,050.56

ft2/stud A 172.51 111.86 107.33 27.96 174.03 71.35 134.68 32.96 236.92 119.66 156.99 60.10

RS 221.43 144.18 138.52 58.96 187.85 87.60 159.99 45.82 269.00 140.15 194.60 66.77

27

Table 15. Mean differences (smaller – larger) by level and school group (all & reasonably-sized)

Elementary Middle High group variable diff.

(S-L) p e.s. diff.

(S-L) p e.s. diff.

(S-L) p e.s.

ALL $/ft2 $1.50 .432 ---- $0.21 .934 ---- -$12.99 .000 -.39 $/stud $6,247.21 .000 +.60 $4,067.51 .000 +.51 $6,218.97 .000 +.57 ft2/stud 54.74 .000 +.65 39.35 .000 +.66 79.93 .000 +.78RS $/ft2 -$0.58 .853 ---- -8.15 .024 -.23 -$10.24 .001 -.33 $/stud $8,720.45 .000 +.94 $1,413.03 .127 ---- $4,534.17 .000 +.45 ft2/stud 82.91 .000 +1.41 27.86 .000 +.61 74.40 .000 +1.11 note 1. same Ns for groups and subgroups by level as reported in Table 14 note 2. diff.= mean difference (smaller minus larger); p = significance; e.s. = effect size note 3. Effect size (e.s.) computed for statistically significant differences and is computed as ratio of difference in

means to SD for entire group of schools (i.e., all schools, both smaller and larger, in ALL or RS, as appropriate); see notes 4 and 5 for the pooled standard deviations

note 4. Pooled standard deviations, for all schools and reasonably-sized schools (used to calculate effect sizes).

level & group $/ft2 $/stud. ft2/stud. E ALL $38.87 $10,390.61 83.91 RS $38.83 $9,230.29 58.96

M ALL $33.61 $7,927.25 59.44 RS $35.87 $7,628.70 45.82 H ALL $33.03 $10,918.31 102.76 RS $31.34 $10,050.56 66.77

The mean differences and effect sizes given in Table 15 provide the evidence to answer research question 1. The two unit cost variables provide contradictory testimony, and the information of square foot per student helps to clarify matters somewhat. The conclusion for one level, however, is not ambiguous. Smaller reasonably-sized schools at the middle level are less costly to build than larger ones, to judge from analyses of these data. Cost per square foot is less, and cost per student is not statistically different as compared to larger schools. Generalizations from the information reported in Table 15 follow.

First, cost per square foot is either not different for larger and smaller schools, or smaller schools are less costly. Larger schools are never less costly per square foot. This pattern persists across both groups of schools—all schools and reasonably-sized schools. Smaller high schools in the all-schools group, and smaller middle and smaller high schools in the reasonably-sized schools group, are less costly per square foot than their within-level counterparts. At other levels in both groups, no significant difference in cost per square foot is evident. For the three (of six) cases in which smaller schools are less costly, effect sizes are moderate, varying from about two-tenths of a standard deviation to about four-tenths of a standard deviation.

Second, cost per student is always higher in smaller as compared to larger schools,

whatever the group, with one exception in 6 comparisons—smaller middle schools in the reasonably-sized group are no more costly per student than larger reasonably-sized schools. The

28

effect sizes for these differences are strong, ranging from half a standard deviation (middle schools in the all-schools group) to nearly a full standard deviation (elementary schools in the reasonably-sized schools group).

Third, square feet per student is consistently higher among smaller schools (i.e., for all

levels in both schools groups). This allocation of space logically explains the tendency of smaller schools to exhibit higher contracted costs per student. The effect sizes that describe the magnitude of the difference are even stronger than those for cost per student, ranging from about three-fifths of a standard deviation (middle school, reasonably-sized schools group) to nearly a standard deviation and a half (elementary schools, reasonably-sized schools group).

Results for the trimmed full data set (research question 2). The trimmed construction

data set contains none of the contextual variables that would have permitted the study to investigate research question 2. Indeed, that fact motivated the study to develop the augmented data set on reasonable-sized high schools, where such results were reported (regressing cost on a standard set of 9 contextual variables).

One can nonetheless regress total cost on total planned enrollment, total square feet, and

planned completion date. In analyses summarized in Table 16, the study conducted such regressions for all schools and reasonably sized schools.

Table 16. Regressions of total cost on three predictor variables panel 1: all schools (n = 3,124)

elementary (n=1,681)

middle (n=720)

high (n=723)

variable β p β p β p planned enrollment NS -.07 .042 .10 .000 total square feet .71 .000 .81 .000 .78 .000 end date .12 .000 .17 .000 .14 .000 Adjusted R2 .52 .63 .79 panel 2: reasonably-sized schools (n = 1,387)

elementary (n=609)

middle (n=371)

high (n=407)

variable β p β p β p planned enrollment NS NS NS total square feet .81 .000 .77 .000 .83 .000 end date .10 .000 .15 .000 .15 .000 Adjusted R2 .65 .64 .76

Table 16 shows that total costs are strongly influenced by these three variables, but that

total square feet exerts the preponderant influence on total cost. For all schools (panel 1), and

29

with total square feet in the equation, planned enrollment (i.e., school size) exerts a small but contradictory additional influence on total cost. At the elementary level, planned enrollment does not exert an separate influence. For middle schools, the small additional influence is negative (the larger the planned enrollment, the lower the total cost—by a small amount); for high schools, however, the small influence is positive, and the larger the planned enrollment, even with total square feet in the equation, the higher the cost. The later the planned completion, in all cases, the higher the cost (representing an inflation factor).

For reasonably sized schools, planned enrollment does not exert an additional influence

on total cost, with total planned square feet in the equation.

The answer to question 1 however—ambiguity about costs of smaller as compared to larger schools—highlights the importance of cost per student. Interested parties might well ask “What determines cost per student given the findings on question 1?” The existing variables, as it turns out, can address this question rather well, though without much surprise. At the school level, cost per student is determined by cost per square foot and square feet per student: this is simply a mathematical relationship (cost / students = cost / ft2 x ft2/ students). It may be worth representing this mathematical relationship empirically in order to determine the empirical strength of the two main cost factors in this data set, and the next analysis does so. As possible influences in the empirical data set, likely independent variables that might influence cost per student, in addition to square feet per student and cost per square foot include whether a school is smaller or larger, and dummy variables for school level. Table 17 reports the results of regressing cost per student on such variables. Collinearity analysis for this equation shows that variance inflation factors are all under 2.0. The threat from multicollinearity, therefore, is small. This simply means that the independent variables are sufficiently unrelated to one another to permit an acceptable multiple regression analysis. This is still something of a surprise given that cost per square foot and square feet per student are logically related. Their relationship turns out to be negative and quite modest (r = -.22). Recall, as well, that regression analysis, as in this analysis, is typically additive (and not multiplicative as in the mathematical relationship of cost per student, cost per square foot, and square feet per student). Table 17. Regression of cost per student on selected predictors

variable B se B β t p $/ft2 139.645 1.351 .497 103.387 .000 ft2/student 103.158 .631 .880 163.563 .000 high school -341.659 146.478 -.014 -2.332 .020 elementary -205.239 122.419 -.010 -1.677 .094 larger -239.478 105.739 -.012 -2.265 .024

note 1. adjusted R2 = .93 note 2. dummy variables (high school, elementary, larger) coded 1 for the presence of the condition As logically expected, the proportion of explained variance is nearly 100 percent (i.e., 93%). This efficiency is the artifact of the fact that the dependent variable is the product of the most

30

influential independent variables. What is perhaps interesting in Table 16, however, is the empirically assessed magnitude of the influence of the predictor variables. Square feet per student is the strongest influence on cost per student (β = .88; part correlation = .78), while cost per square foot exerts an influence about half as strong (β = .50; part correlation = .49). The dummy variables explain negligible amounts of additional variance in cost per student and therefore, despite their statistical significance, are not practically significant.

An ancillary analysis: smaller reasonably sized schools compared to megaschools. This study has focused on construction costs for reasonably-sized schools, with some attention to comparing the cost of such schools with the costs for all schools in the trimmed construction database. This focus of concern reflects the accumulating research that demonstrates the academic benefits of smaller size, especially for students in impoverished communities. In such communities the smallest schools should be maintained according to some school-size researchers (Howley, Craig & Howley, in press). The cost differentials disclosed in the forgoing analyses have been comparatively modest, and cannot be said to be either definitive or exorbitant. Impoverished communities, however, are usually in the position of having to “nickel and dime” themselves when school construction is at issue, so that even a modest differential pose a considerable challenge: the capacity of impoverished communities to secure financing is much more restricted than it is among affluent communities.

When such communities cannot provide supplemental funding, or when they can provide

none at all, the imperative—from state and local officials—is to minimize construction costs. As the previous analyses, suggest, the most efficient way to minimize total cost is to minimize the square feet allocated per student. Hence in such communities, and especially such communities in urban areas, the tendency has been to construct much larger schools—schools with a larger footprint, enrolling many students (too many students from the standpoint of academically reasonable size).

One comparison was not initially planned, but now seems advisable: the comparison

between appropriately smaller schools (beneficial to student achievement in impoverished communities) and the megaschools to which impoverished students are too often consigned. For this comparison, the study determined the median planned enrollment, at each level (elementary, middle, and high), of schools not in the “reasonably-sized” group, and identified all schools with planned enrollments above the median as “megaschools.” These medians were 700 students for elementary schools, 1,000 for middle schools and 1,600 for high schools. Recall that the smaller of the reasonably-sized schools had maximum planned enrollments of 400, 575, and 567 (for elementary, middle, and high schools, respectively).

Table 18 provides the same information as given in Tables 14 and 15, except that the

analysis makes only one set of comparisons instead of two; Table compares (1) the smaller reasonably-sized schools (i.e., “appropriately sized schools”) with (2) schools larger (at each level) than the previously described upper limits of reasonable size (i.e., “megaschools”). In other words, these Tables compare the costs for smaller schools thought to be suitable for impoverished communities with costs for schools deemed to be unsuitable anywhere, and well beyond the upper limits of “reasonable size” adopted here and in (Lawrence, B. et al., 2003), but which unsuitable sizes often prevail in impoverished urban settings and in some impoverished rural settings (particularly in the Southeast, where large, consolidated rural school districts prevail).

31

Table 18 exhibits startling contrasts between appropriately small and megaschools, including the cost variables. The statistics are consistent with those reported in Tables 14 and 15, not surprisingly, but are much more extreme. The effect sizes for planned enrollment merely reflect the decision to contrast smaller, reasonably-size schools with schools that many observers would characterize as “megaschools.” The mean enrollments for these very large schools are about 900 at the elementary level, 1,300 at the middle-school level, and about 2,150 at the high school level.

Schools this large are comparatively rare. For instance, in the 2001-2002 school year, the

nation maintained 9,881 regular-education public K-6 schools, but just 466 of these schools enrolled more than 900 students—most of them (n=332) in large cities and the exurban and suburban fringe of large cities. Similarly, there were 11,385 9-12 high schools, but just 921 enrolled more than 2,150 students—787 of them in large cities, mid-size cities, and the fringe of large cities. Interestingly, megaschools are far more common in rural areas than in small-towns, by a factor of about 20. All data for these analyses are taken from the 2001-2002 Common Core of Data school file (Common Core of Data (2001-2002), 2003). These comparative data, read next to the multi-year trimmed construction data set, constitute a cause for alarm: 520 elementary schools (more than the number of K-6 megaschools in existence in 2001-2002) were planned during the several years represented in the construction data base; but the numbers of such middle and high schools planned is also disturbing news. Table 18. Appropriately small schools compared to megaschools, by level Panel 1: Elementary schools

Panel 2: Middle schools

appropriately small (AS)

n=320 megaschool (MS)

n= 520 comparison (AS-MS) variable Mean SD Mean SD difference p e.s.

planned N 295.21 88.36 871.96 161.391 -576.75 .000 -2.49total cost $6,636,601 $3,624,091 $10,162,797 $5,068,126 -$3,526,196 .000 -.81

$/ft2 $114.94 $37.75 $112.712 $41.69406 $2.22 .426 -----$/stud $24,733.75 $16,454.42 $11,675.10 $5,126.37 $13,058.65 .000 1.26

ft2/stud 221.43 144.18 104.46 24.2889 116.97 .000 1.39

appropriately small (AS)

n=187 megaschool (MS)

n=116 comparison (AS-MS) variable Mean SD Mean SD difference p e.s.

planned N 408.45 111.80 1,292.33 207.393 -883.87 .000 -2.81total cost $7,675,342 $4,252,611 $17,407,219 $6,154,691 -$9,731,876 .000 -1.57

$/ft2 $105.38 $33.22 $109.56 $29.86 -$4.18 .270 -----$/stud $19,501.96 $10,033.32 $13,658.84 $4,973.12 $5,843.12 .000 .74

ft2/stud 187.846 87.60 124.75 28.38 63.10 .000 1.06

32

Panel 3: High schools

appropriately small (AS) n=204

megaschool (MS) n=134 comparison (AS-MS)

variable Mean SD Mean SD difference p e.s. planned N 373.12 127.817 2,156.88 358.47 -1,783.76 .000 -2.70total cost $9,632,047 $6,043,470 $32,209,290 $13,920,734 -$22,577,242 .000 -1.80

$/ft2 $99.5446 $30.64 $122.63 $38.17 -$23.08 .000 -.70$/stud $26,028.54 $13,678.40 $15,028.89 $6,278.70 $10,999.65 .000 1.01

ft2/stud 269.00 140.15 128.348 49.30 140.65 .000 1.37 note 1. e.s. = effect sizes; effect sizes computed using standard deviation for all schools at each level in the entire

trimmed sample (see Table 13 and Table 15 note for these values). note 2. Mean difference is the mean for appropriately small schools minus the mean for megaschools (AS-MS). note 3. Upper limits for the appropriately small schools are the medians for the reasonably-sized schools in Tables

14 and 15 (i.e., 400, 575, and 567 for elementary, middle, and secondary schools, respectively). Lower limits for the megaschools are the medians for schools not reasonably-sized schools (i.e., 700, 1,000, and 1,600, respectively for the three levels.

In any case, the effect sizes for planned enrollment are a measure of absolute difference in the scale of appropriately sized schools and megaschools, with the overall standard deviation for all schools in the trimmed data set at each level serving as the metric of difference. The schools indeed occupy opposite ends of the size spectrum in each case, with their means on planned enrollment separated by two-an-a-half to three standard deviations of planned enrollment. Simple ratios show the same result, of course: the appropriately sized elementary and middle schools were planned to enroll on average 1/3 as many students as megaschools, and appropriately sized high schools were planned to enroll just 1/6 as many as secondary megaschools. On the other hand, the megaschools exhibit total costs that show more moderate effect sizes, ranging from .8 to 1.8. At the elementary level, megaschools cost an average of 1.5 times the “appropriately small” schools, middle-level megaschools an average 2.25 times more, and high-school-level megaschools 3.33 times as much. These statistics embed the differences in cost per student and space allocation per student between appropriately small and megaschools, as well: each exhibiting very strong effect sizes of about 1.0. Megaschools allocate 25 to 50% less space per student that appropriately small schools, and, hence, costs per student are much less in megaschools, almost half as much at the elementary school level. Cost per square foot also tends in the same direction as in Table 15, and opposite in direction to cost per student. Again in Table 18 as in the Table 15 statistics for “all schools,” appropriately small schools as compared to megaschools are either no less costly per square foot, or, at the high school level, they are cheaper. In Table 18, the effect size for cost per square foot is twice that in Table 15. Megaschools carry a hefty cost of $122.63 per square foot as compared to $99.54 for appropriately small high schools.

33

Summary of findings for research question 1 with the trimmed full construction data set. Question 1 is addressed in three iterations with the trimmed full construction data set, as follows: (1) for smaller as compared to larger elementary, middle, and high schools in the full data set (n=3,124); (2) for smaller and larger reasonably-sized school drawn from the trimmed full construction data set (n=1,387); and (3) for appropriately small schools as compared to megaschools.

Across all three analyses, cost per square foot is not higher in smaller than in larger

schools; in fact, it is significantly and consistently lower for smaller high schools, whether smaller and larger is defined over the full data set, the reasonably-sized schools, or for appropriately small as compared to megaschools. In fact, megaschools at the high school level are the most expensive category per square foot of all categories considered.

At the same time, it is clearly and consistently the case that smaller schools allocate more

space (square footage) per student than larger schools, whatever the group—all schools, reasonably-sized schools, or the extreme group (appropriately small versus megaschools). Megaschools at the elementary and high-school levels allocate less than half the square feet per student of appropriately small schools. This difference consistently and clearly renders cost per student lower in larger schools as compared to smaller schools, much lower in the case of megaschools.

Summary of findings for research question 2 with the trimmed full construction

data set. Although no investigation of the typical contextual variables was possible with the trimmed full data set, the study did examine the comparative influence of cost per student and cost per square foot on total project cost in two analyses. Among all schools and among reasonably-sized schools, the regression of total cost on planned enrollment, total feet squared, and construction end date revealed total square feet as the strongest influence. With total square feet in the equation, planned enrollment did not explain significant additional variance in total cost. The explanatory power of these simple equations is comparatively strong, accounting, for instance at the high school level of between 75 and 80 percent of the variance in total cost (again, with footprint exerting by far the strongest influence and accounting for at least 60 percent of the variance alone). The consistent, moderate influence of construction end date predictably mirrors the effects of inflation (costs are not reported in constant dollars).

The other relevant analysis conducted by the study also compared the extent of influence

on total cost of cost per square foot and cost per student. In fact, the three variables are mathematically related (the dependent variable being the product of the independent variables), but in regression analysis (in an additive model with adequate technical characteristics), it can be seen that the influence of cost per square feet is the stronger determinant of total cost. Additionally, being a high school and being a larger school (among all schools at a given level) significantly reduce total, though not at a practically significant level of influence.

Discussion In this section, the report extends the consideration of findings, organized by research question. Following this consideration, the report considers the limitations of the study. Research question 1. This study defined megaschools as those enrolling more than 700 students at the elementary level, more than 1,000 at the middle-school level, and more than 1,600

34

at the high school level, and these “megaschools” exhibited mean enrollments of about 900, 1,300, and 2,150 students at the respective levels. Schools this large occupy approximately the upper 10 percent of the school-size distribution, and they are most prevalent in major metropolitan areas according to data from the National Center for Education Statistics (Common Core of Data (2001-2002), 2003).

The analyses show that megaschools can clearly be constructed much more cheaply per student than appropriately small schools (the smaller half of the study’s “reasonably-sized” schools). Megaschools may, however, cost more per square foot, especially at the high school level. The cost differential per student is nonetheless substantial. Schools of three to six times the size of appropriately small schools realize savings of up to 50 percent of the contracted cost per student as compared to the appropriately small schools. Differences in cost per student between smaller and larger schools are less marked in other comparisons made by the study.

Moreover, once the decision is taken not to build schools larger than reasonable (not

larger than 500, 850, or 1,000 students), the cost advantages of schools of larger schools is cut by one-half to two-thirds. In fact, in that part of the study that used the augmented data set of reasonably-sized high schools, cost per student was not statistically different from that for larger schools (Table 6). In the case of 6-12 high schools in the augmented data set, the study found no significant difference in either cost per square foot or cost per student.

These findings, however, are somewhat inconsistent with those for reasonably-sized

schools in the trimmed construction data set. In those analyses by level, costs per square foot were less in smaller than in larger schools or not significantly different from those in larger schools. Costs per student, however, were more in smaller elementary and high schools, but not in smaller and larger middle schools.

What might account for the contradictory findings? The most likely cause is that the

augmented data set is a purposive sample, whereas the full data set represents most of the construction projects known to the industry during the span of data collection. If the two samples are sufficiently different, analysis of relationships in each could exhibit divergent results. This may be the case.

The preponderance of evidence in this study, however, is that smaller schools are not

necessarily more costly than larger schools, but that “appropriately small schools” are indeed more costly than “megaschools.” Schools that are far too large to educate students—especially when located in impoverished communities—can be built cheaply. Under certain conditions, moreover, it seems that smaller and larger schools can actually be built for unit costs (per student and per square foot) that are not appreciably different. This conclusion is warranted by the analyses with the augmented data set—that is, in the case of the set of 9-12 and 6-12 high defined by the decision rules used to create the augmented data set. In any case, the sound decision not to build schools that are too large, sharply reduces the cost disparities evident when comparing appropriately small schools with megaschools.

Research question 2. The contextual features associated with construction costs (apart

from size in square feet and inflation) notably include local revenue wealth, rural location, and proportion of district families not living in poverty. Rural location reduces cost, whereas the two measures of local affluence increase costs. Hypothetically, affluent communities provision the

35

schools built in their midst better than those built in impoverished communities. The data to test this hypothesis, however, are not available. These conclusions are the result of analysis with the purposive sample of the augmented data set and may not pertain to all schools. The direction of influence, however, is certainly consistent with expectations.

The full trimmed data set included now proper contextual variables. Nonetheless, a large

proportion of total cost was shown to be predicted by total square feet. This variable was more influential on total cost, in fact, than total planned enrollment. This finding is important because the way that larger schools realize cost efficiencies (to the extent that they do realize them, an assertion that depends on policy context) is by reducing the amount of space allocated per student. Megaschools, the study found, allocate about half as much space per student as do appropriately small schools.

Not surprisingly, then, the study concludes that total costs and proportionate costs are

driven upward by affluence, and downward by poverty. This means—again, not surprisingly—that impoverished communities are quite unlikely to get the appropriately small schools that would benefit their children academically. This bad political economy for schooling could be broken were policy makers and planners to relinquish as planning options the possibility of building unreasonably large schools (schools most commonly imposed on impoverished communities because of their lower cost per student).

Limitations. As noted in the previous section, findings from the augmented data set

should not be generalized to the nation. The sample is a purposive one, designed to permit the study to address important research questions in light of opportunities sharply limited in the full data set. Findings were developed at a cost—the cost of generalizability.

Generalizability, however, is also an issue with the full data set because the data set

represents many projects nationwide, but the projects themselves can hardly be said to be representatively sponsored. Instead, the construction projects in the data set are those that localities—through a variety of measures—were enabled to consider. The determinates of this good fortune, however, are not likely to be random acts of grace and kindness. Instead, the choice of which community may construct a school is the result of political and economic contests not well represented by any information in the data set.

In both data sets, as well, varying types of error are evident. Experience with developing

the augmented data set, as noted previously, suggests some threats to internal validity. These problems concerned (1) the inability to distinguish which planned projects represented in the full data base were subsequently abandoned and never built, (2) the exact grades included in the planned schools, and (3) the inclusion of a comparatively large proportion of construction projects that were not actually newly constructed buildings. These problems pose comparatively mild threats to the overall pattern of results, but are likely to make the parameter estimates less accurate (i.e., values of the means, regression weights, and similar statistics). The augmented data set avoids these difficulties, but is still subject to an unknown degree of error associated with incorrect matches with schools in the Common Core of Data. To the best of the researchers’ knowledge, matching errors are limited to a few cases at most.

These four threats are next described in some detail. The first three represent threats to

internal validity and the latter a threat to external validity.

36

First, the basic construction data set may contain 12 to 15 percent of schools that were never actually built so far as one can judge from the experience of developing the augmented data set, which entailed contacting extant schools whose records were matched with information in the construction data set. Including such schools in the study confuses planning with reality—it mixes fiction with fact and treats the mixture as all fact. That is, the inclusion of schools that were not actually built, for some reason (e.g., economic or political impracticality) introduces additional error into the data set.

Second, it seems likely that 20 percent or more of the high schools in the trimmed data

set are not 9-12 high schools; some will be K-12 schools, 10-12 schools, and other configurations. Inability to distinguish among them (through lack of variables related to grades in the schools) is tantamount to an inability to acknowledge between-level cost differences. Failure to make these distinctions can contribute to conditions that produce anomalous results.

Third, a fair proportion of the projects reported as new construction were not actually

new construction, so far as one can judge from the contacts made during the development of the augmented data set. This means the data set embeds a certain proportion of cases that should be excluded. The augmented data set was able to make this exclusion, and the study dealt with the threat in the full data set by the processing of trimming from it the top and bottom 5 percent of cases on cost per square foot.

Fourth, in creating the augmented data set, the study opted for a purposive sample (e.g.,

only high school in district). The results of such a sample cannot be generalized to the population of all high schools with any measurable certainty. Indeed, the sample is more rural and less urban than the population of high schools nationally. In other words, the pattern observed in this sample may represent conditions only among schools with very similar characteristics.

Recommendations School construction and school size are major issues for rural, suburban, and urban communities, for differing and often contested reasons. In rural areas, school construction usually entails a varied schedule of consolidations. Sometimes the consolidations are contingent on the combination of two or more small districts; sometimes, combining of schools; and sometimes the collection of all schools in a district on a single rural campus.

The overall effect of these rural consolidations is to remove schools from functional communities, to make them larger, and to locate them more remotely. In larger rural districts, these schools can achieve the status of megaschools as defined in this report. In smaller rural districts, the single-campus model results in a campus-equivalent of the megaschool.

In suburban areas, megaschools schools are often desired for a nexus of reasons cost

efficiencies per student mean that funds can be reallocated to better materials, labs and common spaces, and other architectural amenities. Over-stated curricular claims also make it seem that a school of 2,000 can educate students better, and, finally, the largest schools field the most visible and (on some terms) most competitive sports teams, often prominent statewide. Affluent suburban communities are themselves competitive places, in which civic organizations and families pride themselves on having and being “the best.” Large, well appointed schools with successful sports teams are an emblem of success in many, though by no means all, such places.

37

The megaschool can and is contested in suburban communities, however. Some citizens and some educators are conversant with the literature on the scaling of schools and understand that upper limits of size are educationally relevant. The author of this report has been frequently contacted by citizens from suburban areas concerned to see the construction, for instance, of two 1,200-student high schools instead of a single massive school enrolling 2,400 students.

In urban areas most citizens and educators seem to acknowledge that megaschools do

damage to the students and communities they are supposed to serve. Perhaps the reason for this apparent consensus is that the racial and economic character of large cities dramatically altered in the second half of the twentieth century, resulting in the famed urban concentrations of impoverished “minorities” who became our cities’ majorities. These majorities inherited the patterns of school size constructed by urban reformers in the first half of the last century, and in a time when the populations enrolled often (but not always) included a substantial middle class. This pattern of schooling has been shown repeatedly to serve urban students, families, communities poorly indeed. In recent years, structural reforms to urban schooling have been sponsored that center around making schools that are more humanly scaled. Among these reforms has been the creation of much smaller high schools. They often succeed academically and fail politically.

In view of the continuing contest over school size, a contest that is partly economic and

partly political, this report offers a set of brief recommendations differentially addressed to citizens, educational planners, and policy makers. The recommendations are only partly based on the study reported here. They constitute a personal best-judgment set of prerogatives and are offered as a practical conclusion to this report. Recommendations to citizens and taxpayers. These recommendations to citizens assume a measure of good will toward students and, also, that interests of citizens and taxpayers are coincident. Good will in this instance means good will toward students’ academic learning and quality of life in schools.

1. Challenge plans to build megaschools wherever they are proposed. No sound educational reason exists for having schools anywhere larger than about (more or less) 500, 750, and 1,000 students.

2. Advocate for appropriately smaller schools. School size can be pegged to the affluence of a community. The smallest schools are most appropriate in the most impoverished communities. Planned enrollments in reasonably sized schools should range from 100 or even less at the elementary level in impoverished communities to about 1,000 at the high school level in the most affluent communities.

3. Become informed about school size issues. Consult the practical summaries first (Howley, C., 2001) and (Lawrence, B. et al., 2003). Follow the leads further in the reference lists in these two works.

38

4. Organize politically. Being informed individually is not sufficient. A vocal group of well-informed citizens stands a fair chance of prevailing as they work their networks in concert with each other.

Recommendations for planners.

Who are school planners? They usually are state officials, local professional leaders (especially superintendents), school boards, and sometimes advisory boards of citizens or other citizen representatives. The latter group usually exerts little effective influence, and the preceding advice to citizens applies to them in their capacity as planners as well as more generally. This set of recommendations is directed at “the usual suspects.”

1. Plan schools that are sized appropriately to advance student achievement.

The publication from the Educational Research Service titled Research on Smaller Schools: What Education Leaders Need to Know to Make Better Decisions (Howley, 2001) gives an accessible summary of research on the achievement advantages of smaller schools. Read it. It’s short, comprehensive, and practical.

2. Plan construction that makes, rather than breaks, connections to communities.

Communities need schools as much as schools need communities. Community members (especially parents) need schools that welcome them. Megaschools and rural consolidations remove schools from communities, and they are unwelcoming to all, but the are odious in impoverished communities.

3. Recognizing that square feet per student is the key cost-related feature of new construction, develop innovative ways to engage the issue.

The Hermit Crab Solution (Lawrence, 2004), though developed for rural communities, provides examples of “thinking outside the box.”

4. Work with communities rather than against them.

Some community members will want megaschools, it’s true. But surveys of the general public generally confirm a preference for schools that are not too large. Work strenuously to build smaller schools for impoverished communities.

5. Challenge arbitrary impositions of minimum size from funding agencies.

There is no research to warrant statewide minimum sizes; all such standards are arbitrary. Ohio, for instance, requires that no school be built for fewer than 350 students. A more educationally responsible standard is the one adopted in this report for “reasonably sized” schools.

Recommendations for policymakers. 1. Create a policy climate that supports the construction of smaller schools.

Much work needs to be done to help communities secure reasonably sized schools. Some of this work concerns revision to the policy frameworks based on mid-20th

39

century conceptions that larger schools were preferable to smaller ones. Today, unreasonably large elementary schools are being constructed at an alarming rate; and the problem of building middle and high schools that are too large remains. These developments are, in fact, largely the result of policy failures.

2. The range of school-construction policy changes needed is broad. Begin the steady work of needed policy changes now.

The changes include they way school construction is funded (a dramatic equity issue), the specifications for school buildings and building sites, provisions for securing and honoring community engagement in the construction planning, and indirect influences such as rigid expectations for curriculum, technology, and teacher certification.

3. Do not impose high minimum sizes for schools.

If smaller size must be rationed, the research-based option dictates that smaller schools be allocated to impoverished communities. If size minima are unavailable, they should be low—as low as 50, for instance (3,000 regular schools of this size existed in 2000-2001).

4. Create policies to facilitate the building of new smaller schools in big cities and their

suburbs.

Megaschools are plagues in metropolitan areas. Schools-within-schools are poor alternatives (they often fail, and the failures tend to re-establish tracking systems). The challenge is to create smaller schools in real time and real space.

5. Develop alternatives to the myriad forms of rural school consolidation.

Many appropriately small rural schools still exist, but their existence is continually threatened as new schools are built. Every time an appropriately small rural school is closed in an impoverished community, the odds are that the achievement of their (former) students suffers. Alternatives should be planned, developed, and deployed.

The professional experience of the author suggests that, among the preceding recommendations, citizens are generally receptive and convinced, educational planners occasionally receptive but generally unassertive, and policy makers comparatively skeptical and unconvinced. This observation is, of course, almost completely hypothetical. But it is based on the experience of frequent contact from citizens, occasional contact from planners, and almost no contact from interested state-level policy makers. One must agree that policy makers confront very broad and complex difficulties, whereas the difficulties for citizens are immediate and compelling (“they are trying to force a school that is far too large on us”). States that confront a massive building program have taken the path of least resistance, and perhaps understandably: many poor communities need schools, and schools to house students can be built cheaply if made as large as possible.

40

Problem solved, problem created. The solved problem is fiscal; the created problem is educational—lower levels of achievement and greater inequity of achievement. Policy makers must grapple with both finance and achievement. Some critics charge that the accountability movement harbors a deeply inscribed hidden agenda—the undoing of public schooling. A systematic policy that, on one hand, exhorts educators to raise achievement levels and, on the other hand, systematically subverts those efforts (i.e., through sustaining unreasonably large schools and closing appropriately small schools) could be understood as evidence supporting the charge. The public should be better served.

Educational planners can take a mediate role between concerned (and sometimes affronted) citizens and policy makers remotely wrestling with enduring dilemmas. The recommendations given point them in that direction. Their role is to question and challenge unproductive facilities policies, not simply to plan buildings in line with extant policies.

41

References

Abramson, P. (2004, February). Ninth annual school construction report. School Planning and Management, 2-16.

Azari-Rad, H., Philips, P., & Prus, M. (2002). Making hay when it rains: The effect prevailing wage regulations, scale economies, seasonal, cyclical and local business patterns have on school construction costs. Journal of Education Finance, 27, 997-1012.

Breaking ranks: Changing an American Institution. (4th ed.)(1999). Reston, VA: Author. Common Core of Data (2001-2002). (2003). from http://nces.ed.gov/ccd/pubschuniv.asp Cotton, K. (2001). New Small learning communities: Findings from recent literature.

Unpublished manuscript, Portland. Delano, D. (2002). Education market still earning honor roll mention. Building Design and

Construction, 43(4), 19. Fetler, M. (1989). School dropout rates, academic performance, size, and poverty: Correlates of

Educational reform. Educational Evaluation and Policy Analysis, 11(2), 109-116. Ford, T. (1998, November 16). Schools ponder prevailing wage for projects. Crain's Cleveland

Business, p. p. 18. Howley, C. (2000). Research on smaller schools: What education leaders need to know to make

better decisions: Educational Research Arlington, Virginia Service. Howley, C. (2001). Research on smaller schools: What education leaders need to know to make

better decisions. Unpublished manuscript, Arlington, VA. Howley, C., & Howley, A. (in press). School size and the influence of socioeconomic status on

student achievement: Confronting the threat of size bias in national data sets. Educational Policy Analysis Archives.

Lawrence, B., Bingler, S., Diamond, B., Hill, B., Hoffman, J., Howley, C., et al. (2003). Dollars and sense: The cost effectiveness of small schools.

Lawrence, B. K. (2004). The hermit crab solution: Creative alternatives for improving rural school facilities and keeping them close. Charleston, WV: AEL.

Lee, V., & Smith, J. (1997). High school size: Which works best, and for whom? Educational Policy Analysis Archives, 19(3), 205-227.

Meier, D. (1995). The power of their ideas: Lessons for America from a small school in Harlem. Boston: Beacon Press.

Morgan, D., & Alwin, D. (1980). When less is more: School size and social participation. Social Psychology Quarterly, 43, 241-252.

Documents

Dollars & Sense II - Architects of Achievement