Evaluation of the public charter schools program [electronic resource] : final report /

POLICY AND PROGRAM STUDIES SERVICE

Evaluation of the 

Public Charter Schools Program 

FINAL REPORT 

2004 

U.S. DEPARTMENT OF EDUCATION

DOC # 2004-08 

OFFICE OF THE DEPUTY SECRETARY

Evaluation of the 

Public Charter Schools Program: 

Final Report 

Prepared for: 

U.S. Department of Education 

Office of the Deputy Secretary 

Policy and Program Studies Service 

Kara Finnigan, University of Rochester 

Nancy Adelman, SRI International 

Lee Anderson, SRI International 

Lynyonne Cotton, St. Mary’s College of California 

Mary Beth Donnelly, SRI International 

Tiffany Price, SRI International 

SRI International 

Washington, D.C. 

2004 
This report was produced under U.S. Department of Education Contract No. ED98CO0074 with SRI 

International. Dena Patterson served as the contracting officer’s representative. The views expressed herein do 

not necessarily represent the positions or policies of the Department of Education. No official endorsement by 

the U.S. Department of Education of any product, commodity, service or enterprise mentioned in this 

publication is intended or should be inferred. 

U.S. Department of Education 

Rod Paige 

Secretary 

Office of the Deputy Secretary

Eugene Hickok 

Deputy Secretary

Policy and Program Studies Service

Alan L. Ginsburg 

Director 

Program and Analytic Studies Division

David Goodwin 

Director 

2004 

This report is in the public domain. Authorization to reproduce it in whole or in part is granted. While 

permission to reprint this publication is not necessary, the citation should be: U.S. Department of 

Education, Office of the Under Secretary, Evaluation of the Public Charter Schools Program: Final 

Report, Washington, D.C., 2004. 

To order copies of this report, write: 

ED Pubs 

Education Publications Center 

U.S. Department of Education 

P.O. Box 1398 

Jessup, MD 20794-1398 

Via fax, dial (301) 470-1244. 

Or via electronic mail, send your request to: edpubs@inet.ed.gov. 

You may also call toll-free: 1-877-433-7827 (1-877-4-ED-PUBS). If 877 service is not yet available in 

your area, call 1-800-872-5327 (1-800-USA-LEARN). Those who use a telecommunications device for the 

deaf (TDD) or a teletypewriter (TTY) should call 1-800-437-0833. 

To order online, point your Internet browser to: www.edpubs.org. 

This report is available on the Department's Web site at: http://www.ed.gov/rschstat/eval/choice/pcspfinal/index.html. 

On request, this publication is also available in alternative formats, such as Braille, large print, audiotape, 

or computer diskette. For more information, please contact the Department's Alternate Format Center at 

(202) 260-9895 or (202) 205-8113. 
ACKNOWLEDGMENTS 

The authors wish to thank several colleagues whose names do not appear in this volume: 

To Bonnee Groover, whose formatting skills, attention to detail and endless patience made 

completion of the report possible. 

To Andrea Lash, Bethany Martinez, Katherine Baisden and Vanessa Barrat who each contributed 

in a significant way to the analysis of comparative performance between charter schools and 

traditional public schools in five states. 

To Dena Patterson, the project officer for the study’s final year, who mastered a steep learning 

curve on work that had been going on for five years. 

To the members of the study’s Technical Work Group who reviewed the early findings from the 

third wave of data collection: Josephine Baker, James Goenner, Margaret Goertz, Ginger 

Hovenic, Bruno Manno and Eric Rofes. 

And finally, appreciation to the state coordinators, charter school authorizers, and charter school 

directors who took the time to complete telephone surveys in 2001-02. 

Contents 

List of Exhibits....................................................................................................................... vi 

Executive Summary............................................................................................................... ix 

Introduction...................................................................................................................... ix 

Context............................................................................................................................. ix 

Highlights......................................................................................................................... x

Evaluation Questions ....................................................................................................... x 

Key Findings.................................................................................................................... xi 

Chapter 1: Introduction and Overview .................................................................................. 1 

Overview of the Charter School Movement .................................................................... 2 

Organization of the Report............................................................................................... 8 

Chapter 2: Public Charter Schools Program Operations........................................................ 9 

The Growth of the Public Charter Schools Program ....................................................... 10 

State Criteria for Awarding Public Charter Schools Program Funds .............................. 11 

Public Charter Schools Program Subgrants: Awards to Schools..................................... 14 

Uses of Public Charter Schools Program Subgrants at the School-level......................... 16 

Capacity of State Education Agencies and Technical Assistance Provided by States .... 17 

Chapter 3: Characteristics of Charter Schools, Students, and Staff....................................... 19 

Characteristics of Charter Schools................................................................................... 19 

Characteristics of Charter School Students...................................................................... 23 

Charter School Teachers .................................................................................................. 26 

Parent Involvement .......................................................................................................... 28

Charter School Autonomy ............................................................................................... 30 

Flexibility ........................................................................................................................ 31 

Control ............................................................................................................................. 32 

The Role of Education Management Organizations and Community-Based 

Organizations ............................................................................................................. 32 

Chapter 4: Charter School Accountability and the Role of Authorizers ............................... 35 

Characteristics of Charter School Authorizers and Their Schools .................................. 35 

Why Authorizers Charter Schools ................................................................................... 36 

Authorizer Capacity ......................................................................................................... 38

Authorizer Services to Charter Schools........................................................................... 38 

The Charter School Accountability Process .................................................................... 39 

Accountability Phase 1: The Application Process ........................................................... 40 

Accountability Phase 2: The Monitoring Process............................................................ 41 

Accountability Phase 3: The Implementation of Sanctions............................................. 47 

Challenges to Charter School Accountability.................................................................. 50 

Chapter 5: Charter Schools and State Performance Standards .............................................. 53 

State-by-State Analysis of Charter Schools and Performance Standards........................ 54 

Chapter 6: Directions for Future Studies ............................................................................... 59 

References .............................................................................................................................. 61 
vi

Contents (continued) 

Appendices to the Chapters ................................................................................................... 65 

Appendix A: Chapter 1 .......................................................................................................... 67 

Appendix B: Chapter 2 .......................................................................................................... 73 

Appendix C: Chapter 3 .......................................................................................................... 81 

Appendix D: Chapter 4 .......................................................................................................... 89 

Appendix E: Chapter 5........................................................................................................... 99 

Exhibits 

Exhibit ES-1: Mean State Public Charter Schools Program Grant Amount, by Fiscal Year xi 

Exhibit ES-2: Characteristics of Students Attending Charter Schools and Traditional 

Public Schools, 1999-2000 ............................................................................ xiii 

Exhibit ES-3: Student Racial and Ethnic Composition in Charter Schools ......................... xv 

Exhibit ES-4: Percentage of Charter Schools, by Type of Authorizer ................................. xvi 

Exhibit ES-5: State Reporting Requirements for Charter Schools and Traditional 

Public Schools................................................................................................ xviii 

Exhibit 1-1: Primary Evaluation Questions ....................................................................... 1 

Exhibit 1-2: Data Collection Activities, by School Year................................................... 2 

Exhibit 1-3: Number of Operating Charter Schools, by School Year................................ 4 

Exhibit 1-4: Total Number of Charter Schools in Operation, by State.............................. 5 

Exhibit 1-5: Distribution of Charter Schools, by Type...................................................... 6 

Exhibit 1-6: Percentage of Charter Schools, by Type of Authorizer ................................. 7 

Exhibit 2-1: History of Public Charter Schools Program Appropriations, by Fiscal Year 11 

Exhibit 2-2: Mean State Public Charter Schools Program Grant Amount, by Fiscal Year 12 

Exhibit 2-3: State Reporting of Public Charter Schools Program Subgrant Criteria, 

by Type of Subgrant....................................................................................... 13 

Exhibit 2-4: Average Charter School Public Charter Schools Program Subgrant Amount, 

by State........................................................................................................... 15 

Exhibit 2-5: Start-up Funding Sources for Charter Schools .............................................. 16 

Exhibit 2-6: Charter School Reports of Uses of Public Charter Schools Program 

Start-up Subgrants.......................................................................................... 17 

Exhibit 3-1: Median Student Enrollment in Charter Schools ............................................ 20 

Exhibit 3-2: Median Student Enrollment in Charter Schools and Traditional 

Public Schools, by Grade-Level Configuration............................................. 21 

Exhibit 3-3: Grade-Level Configurations of Charter Schools and Traditional 

Public Schools................................................................................................ 22 

Exhibit 3-4: Primary Methods of Delivering Instruction in Charter Schools .................... 23 

Exhibit 3-5: Proportions of White and Minority Students in Charter Schools .................. 23 

Exhibit 3-6: Student Race and Ethnicity in Charter Schools and 

Traditional Public Schools............................................................................. 24 

Exhibit 3-7: Changes in Selected Characteristics of Students in Charter Schools ............ 25 

Exhibit 3-8: Comparisons of Selected Characteristics of Students in Charter Schools 

and Traditional Public Schools ...................................................................... 25 

Exhibit 3-9: Racial and Ethnic Distribution of Charter School Teachers.......................... 26 
vii

Contents (concluded) 

Exhibit 3-10: Racial and Ethnic Distribution of Teachers in Charter Schools and 

Traditional Public Schools............................................................................. 27 

Exhibit 3-11: Parental Involvement in Charter Schools and Traditional Public Schools.... 30 

Exhibit 3-12: Charter School Reports of Waivers ............................................................... 31 

Exhibit 3-13: Charter School Reports of Level of Authority............................................... 33 

Exhibit 3-14: Roles Played by Education Management Organizations and 

Community-Based Organizations, According to Charter School Directors.. 34 

Exhibit 4-1: Distribution of Charter School Authorizers, by Type.................................... 36 

Exhibit 4-2: Reasons Authorizers Sponsor Charter Schools.............................................. 37 

Exhibit 4-3: Assessment Strategies Used by Charter Schools........................................... 42 

Exhibit 4-4: Focus of Authorizer Monitoring .................................................................... 43 

Exhibit 4-5: Charter School Accountability to Multiple Entities....................................... 44 

Exhibit 4-6: Monitoring Responsibilities of State Charter School Office or Staff............ 46 

Exhibit 4-7: Reasons Given by Authorizers for Formal and Informal Sanctions.............. 49 

Exhibit 4-8: Challenges Reported by Authorizers and Difficulty Level............................ 50 

Exhibit 4-9: State Reporting Requirements for Charter Schools and 

Traditional Public Schools`............................................................................ 52 

Exhibit 5-1: Performance of Charter Schools and Traditional Public Schools.................. 55 

Executive Summary 

Introduction 

Charter schools are public schools that 

operate under a contract (or "charter"). The 

expectation is that these schools meet the 

terms of their charter or face closure by their 

authorizing bodies. As public schools, 

charter schools must also meet the 

accountability requirements of the federal 

Elementary and Secondary Education Act of 

1965 (ESEA), as amended by the No Child 

Left Behind Act of 2001 (NCLB).1

Since 1991 when Minnesota passed the first 

state charter school law, the charter school 

sector of public education has grown 

rapidly. By the 2002-03 school year, 39 

states and the District of Columbia had 

charter school laws in place, and more than 

2,700 charter schools were operating 

nationally, serving hundreds of thousands of 

students from every socioeconomic and 

demographic segment of the U.S. 

population. 

Federal support for charter schools began in 

1995 with the authorization of the Public 

Charter Schools Program (PCSP), 

administered by the U.S. Department of 

Education (ED).2

 The PCSP funds the state 

grant program discussed in this report, 

supports charter school research and 

demonstration programs and underwrites 

national charter school conferences. 

 

1

 The data presented in this study covers a period of 

time (1999-2002) prior to the enactment of NCLB. 

2

 The name of the Public Charter Schools Program 

(PCSP) changed to the Charter Schools Program 

(CSP) when the U.S. Department of Education issued 

nonregulatory guidance in August 2003. 

This report has a dual purpose: (1) to 

provide the public and education 

policymakers with findings from a 

descriptive examination of how the PCSP 

operates and (2) to continue documentation 

of the evolution of the charter school 

movement that began in 1995 under another 

federally funded study.3

 

Context 

The charter school sector includes a diverse 

array of schools categorized as newly 

created or converted from previous status as 

public or private schools. Although these 

schools are subject to the terms of an 

individual state’s charter school legislation, 

all charter school laws require that a 

designated bodythe charter school 

authorizerhold a school accountable for 

particular outcomes through the school’s 

individualized contract. Further, flexibility 

(freedom from many policies and 

regulations affecting traditional public 

schools) and autonomy (control over 

decisions) are central to this educational 

reform. This is the basic context in which 

the charter school movement has evolved 

and in which the PCSP operates. 

 

3

 RPP International conducted the first federally 

funded charter school study. In part, the study 

reported in this document extended the RPP study to 

provide a longitudinal portrait of charter schools. 
x

Highlights

Based on three years of data (collected in 

school years 1999-2000, 2000-01 and 

2001-02), the national evaluation of the 

PCSP found that: 

• PCSP money is the most prevalent 

source of start-up funding available to 

charter schools. Nearly two-thirds have 

received federal PCSP funds during their 

start-up phase. Charter schools primarily 

use PCSP funds to purchase technology 

and curricular and instructional 

materials, as well as to fund professional 

development activities. 

• Charter schools are more likely to serve 

minority and low-income students than 

traditional public schools but less likely 

to serve students in special education. 

• Charter schools, by design, have greater 

autonomy over their curriculums, 

budgets, educational philosophies, and 

teaching staff than do traditional public 

schools. Because some state charter 

school laws allow schools flexibility in 

hiring practices, charter schools as an 

overall group are less likely than 

traditional public schools to employ 

teachers meeting state certification 

standards. 

• In five case study states, charter schools 

are less likely to meet state performance 

standards than traditional public schools. 

It is impossible to know from this study 

whether that is because of the 

performance of the schools, the prior 

achievement of the students, or some 

other factor. The study design does not 

allow us to determine whether or not 

traditional public schools are more 

effective than charter schools. 

• Charter schools rarely face formal 

sanctions (revocation or nonrenewal). 

Furthermore, authorizing bodies impose 

sanctions on charter schools because of 

problems related to compliance with 

regulations and school finances rather 

than student performance. Authorizers 

have difficulty closing schools that are 

having problems. 

• During the time period examined by this 

study, little difference exists between the 

accountability requirements for charter 

schools and traditional public schools. 

Evaluation Questions 

The primary questions guiding this 

evaluation can be grouped into four 

overarching topic areas: 

The Public Charter Schools Program 

(1) How does the PCSP work and how do 

this federal grant program and its state 

grantees encourage the development of 

charter schools? 

(2) How do federally funded charter schools 

and school planners use their PCSP 

subgrants? 

Profile of the Charter School Sector 

(3) What are the characteristics of charter 

schools and the students and families 

who are involved with them? 

(4) What flexibility provisions are charter 

schools granted? 

Student Performance in Charter Schools 

(5) To what extent are charter schools 

meeting state standards for student 

performance and how do charter schools 

and traditional public schools compare 

in meeting these standards? 
xi

Charter School Authorizers 

(6) What are the characteristics and roles of 

authorizing bodies? 

(7) What types of accountability 

relationships do authorizers have with 

their schools? 

Several data sources inform answers to these 

questions: survey data from state charter 

school coordinators, charter school 

authorizers, and charter school directors; 

data from the federal Schools and Staffing 

Survey (SASS); and data from state 

departments of education. The evaluation 

team also conducted multiple site visits to 

12 charter schools in the following six 

states: Arizona, California, Massachusetts, 

Michigan, North Carolina, and Texas. 

Key Findings 

3⁄4 The PCSP is a targeted federal grant 

program that awards grants to states 

with charter school legislation. States, in 

turn, award subgrants to charter schools 

and charter school planning groups. At 

least 95 percent of the state grants 

currently reach charter schools, as 

required by the legislation. 

In FY 2001, 90 percent of the 37 states and 

the District of Columbia with charter school 

legislation received PCSP grants. The 

U.S. Department of Education (ED) 

competitively awards these grants to states 

on a three-year cycle, based on projected 

estimates of the level of chartering activity. 

Within the grant cycle, ED makes annual 

adjustments, as necessary. States withhold 

up to 5 percent of these PCSP grants for 

administration costs, and distribute the 

remaining 95 percent to schools in the form 

of PCSP subgrants. 

If a state with charter school legislation does 

not receive a PCSP award, individual charter 

schools within the state may apply directly 

to ED for a school grant. Charter schools in 

four states received grants through this 

provision in 2001-02. 

3⁄4 From FY 1995 through 2001, growth in 

the charter school sector kept pace with 

growth in federal appropriations for the 

PCSP program. During this period, the 

number of charter schools increased 

tenfold, as did the size of the average 

three-year state grant. 

Exhibit ES-1 

Mean State Public Charter Schools Program Grant Amount, by Fiscal Year 

$0

$500,000

$1,000,000

$1,500,000

$2,000,000

$2,500,000

$3,000,000

$3,500,000

$4,000,000

$4,500,000

$5,000,000

FY95 FY96 FY97 FY98 FY99 FY00 FY01

State PCSP Grant

Sources: SRI 1999-2000 and 2001-02 state coordinator surveys. 

Exhibit reads: In FY 2001, the mean state grant was $4.5 million. 
xii

State charter school coordinators and charter 

school directors confirmed the importance 

of the PCSP as a federal investment in 

charter school development. States may 

award two types of subgrants: (1) start-up 

subgrants to support planning and early 

implementation of charter schools and (2) 

dissemination subgrants to support charter 

schools in sharing their ideas and practices. 

Based on the 2001-02 survey of charter 

school directors, 61 percent reported that 

they had received a PCSP start-up subgrant 

and 19 percent had received a dissemination 

subgrant at some point in time.4

Federal appropriations for the PCSP grew 

steadily from $6 million in FY 1995 to 

$190 million in FY 2001 (increasing to 

$218.7 million in FY 2004). During the 

same period, the number of charter schools 

grew from approximately 250 to 2,700. 

PCSP awards to states have increased in 

size, from a mean state grant of $512,900 in 

FY 1995 to nearly $4.5 million in FY 2001 

(see Exhibit ES-1). This increase in state 

grant awards reflects growth in the PCSP 

annual appropriation coupled with a leveling 

off of the number of states with charter 

legislation. 

While the number of charter schools has 

continued to grow nationally, the growth is 

most substantial in a limited number of 

states. These states (for example, California 

and Florida) currently receive the largest 

PCSP grants. 

 

4

 These statistics derive from separate survey items 

and are not intended to be summed. These data may 

underestimate the percentage of schools with start-up 

subgrants because of school-level confusion about 

the funding source—the state versus ED. 

3⁄4 PCSP start-up and dissemination 

subgrants support professional 

development activities and technologyrelated purchases. In addition, schools 

used start-up subgrants to purchase 

curricular or instructional materials. 

Each state with a PCSP grant creates its own 

process and selection criteria for distributing 

the funds as subgrants to charter schools or 

planning groups.5 In general, start-up 

subgrants are more easily obtained than 

dissemination subgrants. The size of 

subgrants to charter schools or planning 

groups varies by state. The average school 

subgrant in FY 2001-02 ranged from 

$20,000 in one state to $263,000 in 

another—with most state averages tallying 

between $80,000 and $150,000. Most 

charter schools used PCSP start-up 

subgrants to purchase instructional materials 

(87 percent), fund professional development 

(79 percent), and purchase technology 

(78 percent). 

3⁄4 In comparison with traditional public 

schools, charter schools are smaller and 

employ fewer certified teachers than 

traditional public schools because of 

provisions in some state laws.6

 These 

schools are also more likely to serve 

more grade levels (e.g., K-12) than the 

typical public school. 

 

5

 The federal PCSP legislation places relatively few 

restrictions on the use of these funds. One prohibition 

is the use of PCSP funds to purchase a facility. 

6

 By law, some states afford charter schools more 

flexibility with respect to teacher certification 

provisions. 
xiii

Although the median enrollment in charter 

schools has been steadily rising (e.g., from 

137 students in 1998-99 to 190 students in 

2001-02), these schools remain considerably 

smaller than traditional public schools 

serving similar grade ranges. For example, 

according to data from the federal Schools 

and Staffing Survey (SASS), the median 

enrollment in charter high schools in 

1999-2000 (the most recent year of the 

Schools and Staffing Survey data) was 132 

compared with 675 in traditional public high 

schools. 

In addition, states provide flexibility to 

charter schools over many areas including 

hiring practices and the certification and 

licensure of their teachers. While charter 

schools must meet the accountability 

requirements of NCLB, they retain any 

flexibility provided to them in individual 

state chartering laws, especially in the area 

of teacher qualifications. One result of this 

flexibility may be that charter schools 

employ fewer traditionally certified 

teachers. According to the 1999-2000 SASS, 

79 percent of teachers in charter schools 

held certification, compared with 92 percent 

of teachers in traditional public schools. 

In contrast to the typical configuration of 

elementary, middle, and high schools, 

charter schools are more likely to contain 

either grades K-8 or grades K-12. More than 

one-third (35 percent) of charter schools are 

K-8 or K-12 schools, compared with 

8 percent of other public schools. Interviews 

with charter school staff and parents 

indicated that the K-8 and K-12 

configurations might be in response to the 

desire for students to avoid the difficult 

transitions between school-levels.7

 

3⁄4 Charter schools disproportionately 

attract students and families who are 

poor and who are from African 

American backgrounds. 

 

7

 Some research has found an association between 

grade level configuration and student academic and 

nonacademic performance (see Renchler, 2002, and 

Franklin et al., 1996). 

Exhibit ES-2 

Characteristics of Students Attending Charter Schools and 

Traditional Public Schools, 1999-2000 

Percentage of Students 

Student Characteristic 

Charter Schools 

(n=870) 

Traditional Public 

Schools (n=8,432) 

African American*** 27 17 

Hispanic*** 21 15 

White*** 46 63 

Free or reduced-price lunch*** 43 38 

Special education students with 

Individualized Education Programs (IEP)*** 9 12 

***p<.01 (Indicates significant difference between charter schools and traditional public schools in the 

percentage of students with various characteristics.) 

Source: 1999-2000 public charter school SASS survey and public school SASS survey. 

Exhibit reads: Of all students enrolled in charter schools in 1999-2000, 27 percent were African American, 

compared with 17 percent in traditional public schools. This difference is statistically significant. 
xiv

The profile of students who attend charter 

schools differs from traditional public 

schools, as illustrated in Exhibit ES-2. In 

1999-2000, charter schools served fewer 

white students and more minority students 

(including African American and Hispanic) 

than traditional public schools. Charter 

schools also served more students from lowincome families but fewer special education 

students with Individualized Education 

Plans (IEPs). 

Furthermore, the overall proportion of 

minority students attending charter schools 

has been increasing—in 2001-02, 

approximately two-thirds of students in 

charter schools were from minority 

backgrounds. As Exhibit ES-3 demonstrates, 

virtually all of the growth in minority 

enrollments is the result of increases in the 

percentage of African American students. 

Over the same period, the proportion of 

white students decreased and the proportion 

of Hispanic students remained fairly 

constant. 

3⁄4 Case studies of Texas, Colorado, 

Illinois, Massachusetts, and North 

Carolina show that more than half of the 

charter schools in these states were 

already meeting state performance 

standards in 2001-02, although charter 

schools were somewhat less likely than 

traditional public schools to meet 

standards.8

 These findings are not 

indicative of the impact of charter 

schools on student achievement. 

Furthermore, it is not possible to 

determine from this study whether or not 

traditional public schools are more 

effective than charter schools. 

 

8

 While the data analyzed predate the requirements of 

NCLB, these five states already had set school-level 

standards, perhaps in response to the Improving 

America’s Schools Act of 1994. 

Although the No Child Left Behind Act of 

2001 (NCLB) subjects charter schools to the 

same performance standards as traditional 

public schools, this study conducted case 

studies9 of five states during the period prior 

to NCLB and found that more than half of 

charter schools in each state were meeting 

state performance standards in 2001-02 

(with as many as 90 percent meeting 

performance standards in Colorado). 

However, because many charter schools 

tend to target students with educational 

disadvantages, some studies have shown that 

charter school students typically do not 

perform as well in school as students in 

other public schools. Charter schools in all 

five case study states were less likely than 

traditional public schools to meet 

performance standards even after controlling 

for several school characteristics. This 

finding, which does not imply a lack of 

charter school impact on student 

achievement, may be linked to the prior 

achievement of students or some other 

factor. The design of this study did not allow 

us to determine whether charter schools are 

more or less effective than traditional public 

schools. 

The purpose of this study’s student 

performance component was to determine 

whether charter schools met state 

performance standards and to determine 

how charter schools compared to traditional 

public schools in meeting these standards. 

The study originally intended to use studentlevel data, but in 2001-02, policy 

interpretations of the Family Educational 

Rights and Privacy Act (FERPA) precluded 

this. As a result, the study shifted its 

emphasis to school-level data, conducting an 

analysis in the states with adequate data. 

 

9

 Because these state analyses are not representative 

of the charter school universe, this evaluation refers 

to them as “case studies.” 
xv

The results of these analyses suggest that 

charter schools may have difficulty in 

meeting the high-stakes performance 

standards recently adopted by states under 

NCLB. Future studies should examine the 

extent to which charter schools serving high 

proportions of educationally disadvantaged 

students exhibit improved performance over 

time. 

3⁄4 Charter school authorizers monitor their 

schools for accountability purposes and 

provide direct services (often on a feefor-service basis). Authorizing bodies 

that charter many schools are likely to 

have an infrastructure for monitoring 

but are not likely to provide services. 

Authorizing bodies are a critical component of 

the charter school movement and include a 

variety of entities. In 2001-02, local school 

districts authorized 45 percent of charter 

schools, while state departments of education 

authorized 41 percent, and institutions of 

higher education authorized 12 percent. (See 

Exhibit ES-4.) (In addition, other entities, such 

as independent charter boards, authorized 

2 percent of charter schools.) It is interesting 

to note that although they authorize 45 percent 

of all charter schools, local education agencies 

represent 91 percent of the population of 

authorizers. State education agencies on the 

other hand, authorize 

41 percent of all charters but represent just 

3 percent of all authorizers.

There is a general expectation in the charter 

school sector that authorizers have a 

responsibility to regularly oversee charter 

school operations and progress toward 

meeting the goals in the charter. The reality is 

that only 36 percent of authorizers had a 

charter school office or staff in 2001-02, 

Exhibit ES-3 

Student Racial and Ethnic Composition in Charter Schools, 1998-99 to 2001-02 

0

10

20

30

40

50

60

1998-99

(n=975)

1999-2000

(n=870)

2000-01

(n=377)

2001-02

(n=444)

Percentage of Students

White

African American

Hispanic or Latino

Other Minority

Note: Racial and ethnic categories are based on current census categories and differ somewhat 

from RPP and SASS categories. Other Minority includes Asian, American Indian or Alaska 

Native and, in 2000-01 and 2001-02 only, Native Hawaiian or other Pacific Islander. 

Sources: 1998-99 data: Nelson et al. (2000); 1999-2000 data: Public Charter School SASS survey; 

2000-01 and 2001-02 data: SRI 2000-01 and 2001-02 charter school surveys. 

Exhibit reads: In 1998-99, 48 percent of students in charter schools were white, compared with 

46 percent in 1999-2000, 41 percent in 2000-01, and 37 percent in 2001-02. 
xvi

suggesting limited capacity to address charter 

school oversight. However, this finding 

varies by type of authorizer. For example, 

85 percent of states that are authorizers have 

an office or staff dedicated to charter school 

work. Because states are more likely to 

authorize a large number of schools, they 

may require an infrastructure to provide 

adequate oversight. 

Some authorizers, particularly local school 

districts, report that they provide a number of 

services to charter schools, the most common 

being administrative oversight, assistance in 

meeting state or federal regulations and 

special education services. Increasingly, 

authorizers report that schools must pay for 

these services. 

Exhibit ES-4 

Percentage of Charter Schools, 

by Type of Authorizer (2001-02) 

Percentage of Charter Schools

45%

41%

12% 2%

Local school districts

State boards of educ. or state educ. agencies

Universities or colleges

Other entities

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, local school districts 

authorized 45 percent of charter schools. 

3⁄4 Charter schools do not automatically 

have flexibility with respect to complying 

with state and federal regulations and 

often share authority over key decisions 

with their authorizers. Only 37 percent 

of charter school states automatically 

allow waivers of state regulations for 

charter schools. More commonly, 

charter schools must request specific 

waivers from the state. Few states (less 

than five) exempted charter schools from 

student assessment requirements in 

2001-02. 

In theory, charter schools enjoy flexibility or 

school-level control over key decisions not 

available to the typical school in exchange 

for accountability for specified outcomes. In 

reality, the autonomy of charter schools is 

limited by state policies, as well as by 

relationships with authorizers, education 

management organizations (EMOs) and 

community-based organizations (CBOs). 

Only 37 percent of states with charter 

schools granted them automatic waivers 

from state policies and regulations in 

2001-02, but 54 percent waived regulations 

on selected policies or allowed charter 

schools to request waivers on a case-by-case 

basis. Nine percent did not permit any 

waivers to charter schools. 

Furthermore, charter schools frequently 

share their school-level authority with one or 

more other entities. Schools were most 

likely to report sharing control with their 

authorizers. Some school directors reported 

sharing authority with EMOs or CBOs. 
xvii

3⁄4 Authorizers determine which schools to 

charter, monitor progress and 

performance and decide whether or not 

to renew the charter at the end of its 

term. However, more than half of all 

authorizers reported difficulty in closing 

a school that is having problems. In 

addition, the charter contract, with its 

tailored outcomes, may have diminished 

importance in the current high-stakes 

accountability environment. 

The charter school accountability process 

involves three phases: the application 

process, the monitoring process and the 

implementation of sanctions (if needed). 

During the application process, authorizing 

bodies screen applications, denying charters 

because of problems relating to, for 

example, proposed instructional strategies, 

governance procedures, accountability 

provisions, and business plans. 

The monitoring process occurs after 

authorizers have awarded charters to 

planning groups. Authorizers and states 

reserve legal authority to monitor charter 

schools, but other entities are also involved, 

resulting in a complex system of accounttability. Charter schools reported being 

monitored by their authorizers, governing 

boards, states and, in some cases, EMOs or 

CBOs. They reported that they are most 

accountable to their own governing boards. 

Authorizers have developed monitoring 

procedures and determined criteria for 

applying interventions or sanctions with 

little specific guidance from state charter 

school legislation. Authorizers reported 

monitoring nearly all of their schools on: 

compliance with federal or state regulations; 

student achievement results; enrollment 

numbers; financial record keeping and 

viability; and special education services. 

Finally, authorizing bodies have the 

authority to implement formal or informal 

sanctions against a school that fails to meet 

the terms of its charter. Results from the 

survey of authorizers show that few 

authorizers had implemented formal 

sanctions: only four percent of authorizers 

had not renewed a school’s charter and 

six percent had revoked a charter as of 

2001-02. (We are unable to compare these 

rates with the proportion of traditional 

public schools that have been sanctioned 

through closure or reconstitution.) Informal 

and less severe sanctions, such as written 

notification of concerns, were more 

common. Formal and informal sanctions 

were usually associated with problems 

relating to compliance with state and federal 

regulations and school finances. 

Authorizers report facing a wide range of 

challenges in sponsoring and providing 

support to charter schools, including 

inadequate financial or human resources. 

More important, more than half of 

authorizers report difficulty closing a school 

that is having problems—a key 

responsibility of authorizers in this 

educational reform. 

In the early years of the charter school 

movement’s development, charter schools—

at least theoretically—were more 

accountable for outcomes than other 

schools, by virtue of the terms of a charter 

contract. More recently, however, states 

have implemented reporting systems to track 

school inputs in addition to outcomes for all 

public schools. As Exhibit ES-5 indicates, 

little difference now exists between state 

reporting requirements for charter schools 

and those for traditional public schools. 
xviii

Exhibit ES-5 

State Reporting Requirements for Charter Schools and 

Traditional Public Schools 

Percentage of States

Reporting Requirement 

Required for 

Charter 

Schools 

Required for 

Traditional 

Public Schools 

Reporting student achievement results on required statewide 

assessments (n=35) 100 97 

Reporting on other student performance indicators, e.g., attendance 

rates (n=34) 97 97 

Reporting on enrollment numbers (n=32) 100 94 

Aligning of curriculum to state standards (n=31) 90 94 

Reporting on student demographics (n=31) 94 100 

Reporting on teacher qualifications (n=26) 100 96 

Reporting on teacher demographics (n=19) 89 100 

Reporting on school waiting list (n=11) 91 27 

Note: The number of respondents varies by accountability requirement because some states reported that these requirements were 

“not applicable” in their states. 

Note: The actual survey used the term “accountability requirements” to encompass both inputs and outputs. To avoid confusion 

with the current narrower definition of accountability in NCLB, we have used the term “reporting requirements” in this exhibit and 

accompanying text. 

Source: SRI 2001-02 state coordinator survey. 

Exhibit reads: All states (n=35) required that schools report student achievement results as part of their state accountability system. 

Of these, all required this for charter schools and 97 percent required this for traditional public schools.
Chapter 1 

Introduction and Overview 

The charter school movement is one of the 

fastest growing education reforms in the 

country and has strong appeal at all levels of 

the education system, from local communities 

to the U.S. Congress. The charter school 

movement also has bipartisan support 

nationally and in the states that have enacted 

charter school laws. The federal government 

encourages the development and 

implementation of charter schools through the 

Public Charter Schools Program, a major 

grant program administered by the 

U.S. Department of Education. 

Since Minnesota passed the first charter 

school law in 1991, the number of charter 

schools across the nation has increased 

steadily.10 As the movement has grown, 

however, it has struggled. Stories quickly 

emerged about the difficulties that charter 

schools faced in their first months and years 

(RPP International and University of 

Minnesota, 1997). In many cases, charter 

school founders underwrote the planning and 

early development of their schools out of their 

own pockets or by incurring debt. Finding, 

renting or buying, and renovating space were 

particular barriers. The Public Charter 

Schools Program (PCSP) was enacted in 1995 

to improve the financial circumstances of 

charter schools in these planning and early 

implementation stages; in later years the 

program extended its scope to successful 

charter schools that have been open for three 

or more years to assist them in disseminating 

their best practices to other public schools. 

 

10 As of spring 2004 when Washington passed its new 

law, 40 states have charter school legislation. 

This document is the final report of the first 

national evaluation of the PCSP. The 

evaluation began in October 1998 with two 

purposes: (1) to evaluate the PCSP and 

(2) to document the evolution of the charter 

school movement.11 The dual purposes of this 

evaluation are evident in the evaluation 

questions that guided the data collection and 

analysis activities. As illustrated in 

Exhibit 1-1, some questions focused 

specifically on the PCSP, while others 

addressed the general evolution of the charter 

school movement. 

 

11 The documentation of this movement began with 

RPP International’s National Study of Charter Schools, 

which tracked the development of charter schools 

nationwide from 1995 to 1999 (see Nelson et al., 

2000). 

Exhibit 1-1 

Primary Evaluation Questions 

• How does the PCSP work and how do this federal 

grant program and its state grantees encourage the 

development of charter schools? 

• How do federally funded charter schools or school 

planners use their PCSP subgrants? 

• What are the characteristics of charter schools and 

the students and families who are involved with 

them? 

• What flexibility provisions are charter schools 

granted? 

• To what extent are charter schools meeting state 

standards for student performance, and how do 

charter schools and traditional public schools 

compare in meeting these standards? 

• What are the characteristics and roles of 

authorizing bodies? 

• What types of accountability relationships do 

authorizers have with their schools? 
2

The evaluation involved data collection from 

multiple respondents at the national, state, and 

local levels to provide a comprehensive 

picture of both PCSP funding and the charter 

school movement generally. Exhibit 1-2 

provides details of the data collection 

activities by school year. While this report 

draws on findings from each year of data 

collection, it primarily focuses on the analysis 

of three surveys: (1) a survey of state charter 

school coordinators; (2) a survey of charter 

school authorizers; and, (3) a survey of 

charter school directors. The report also draws 

upon data collected by RPP International and 

the federal Schools and Staffing Survey 

(SASS), which included a charter school 

supplement in 1999-2000. For chapter 5, the 

study team acquired data on student 

performance from a number of states. The 

report focuses on findings from data gathered 

in 2001-02, the evaluation’s third year of data 

collection.12 Although this evaluation has 

reported previously on findings from earlier 

 

12 Refer to Appendix A-1 for detailed information on 

the data collection and analysis procedures. 

rounds of data collection, they are included in 

this report, as appropriate, to indicate data 

trends.13 

Overview of the Charter School 

Movement 

Public education in the United States is a 

multifaceted, multilayered system of 

institutions organizations, decision-makers, 

educators, parents and students. Every school, 

school district and state exists in—and is 

affected by—a complex web of laws, policies 

and conflicting demands from different 

stakeholders. Charter schools are a relatively 

new arrival on this scene and have matured 

within the same milieu of highly diverse 

school missions, accountability expectations, 

state laws, and federal policies that affect the 

rest of public schooling. 

 

13 For the Year 1 (1999-2000) report, see 

U.S. Department of Education (2000). For the Year 2 

(2000-01) report, see Anderson et al. (2002). The 

reference list provides URLs.

Exhibit 1-2 

Data Collection Activities, by School Year 

Data Collection Activity 

1998-

1999 

1999-

2000 

2000-

2001 

2001-

2002 

Telephone survey of state charter school coordinators X X 

Telephone survey of charter school authorizers X X X 

Telephone survey of charter schools X X 

Site visits to charter schools and authorizers X X 

Parent survey X‡ 

Focus groups with charter school planners, operators 

and authorizers X 

Charter school state student performance data X 

Federal file extraction X X 

Federal interviews X 

Data from the National Study of Charter Schools 

(RPP International) X 

Data from the Schools and Staffing Survey (SASS) X 

‡These data are not reported because of low response rates. 

Exhibit reads: Telephone surveys of state charter school coordinators were administered in school years 1999-2000 

and 2001-2002.
3

Charter schools have widespread political 

support because this educational reform has 

come to symbolize different things to diverse 

groups with contradictory agendas (Wells et 

al., 1999). Because charter school policies are 

a vehicle for change, rather than a particular 

approach or design, they can be considered 

“an opportunity, not a blueprint” (Nathan, 

1996). Although state charter laws vary, all 

share a common set of assumptions: (1) that 

accountability for outcomes will improve 

school performance and (2) that high levels of 

autonomy will allow schools to better meet 

student needs and, as a result, improve 

performance. Beyond these common 

assumptions, charter schools vary, by design, 

in a number of ways. For example, they may 

be either new schools or public or private 

conversions;14 they may be authorized by 

local districts or other agencies (e.g., 

universities); and they may target special 

populations of students or highlight particular 

curricular or instructional philosophies.15

Charter schools have five key features, 

according to Finn, Manno and Vanourek 

(2000, p.15): 

• They can be created by almost anyone. 

• They are exempt from most state and local 

regulations, essentially autonomous in 

their operations. 

• They are attended by youngsters whose 

parents choose them. 

• They are staffed by educators who are 

also there by choice. 

• They are liable to be closed for not 

producing satisfactory results. 

 

14 Note that private school conversions are not 

recognized under Federal definition and non-regulatory 

guidance. 

15 Several national studies provide information on these 

descriptive characteristics. See Nelson, et al., 2000; 

Anderson, et al., 2002; and Hill, et al., 2001. 

In general, those who wish to operate a 

charter school submit a proposal outlining the 

components of the school’s plan, including 

the instructional approach, the governance 

and financial arrangements, the specific 

educational outcomes, and the way in which 

the charter school will measure these 

outcomes (Geske, Davis et al., 1997). A 

legally designated authorizing body then 

determines whether it will “charter” the 

school and, as part of the bargain, hold it 

accountable to the terms of its charter. While 

charter schools may be similar to some 

specialty or magnet schools, they differ in the 

existence of performance agreements 

developed with their individual authorizers 

(Hill, Lake and Celio, 2002). 

Charter schools open for a variety of reasons. 

Some of these reasons, according to Nelson et 

al. (2000), included realizing an alternative 

vision of schooling (75 percent) and serving a 

special population of students (28 percent).16

Furthermore, charter school founders focus on 

more than test scores, including “how kids 

should be raised, the cultural content of the 

curriculum, the democratic or authoritative 

ways in which teachers relate to parents” 

(Finn et al., 2000, p.7). 

The phrase “charter school movement” is 

often used but is somewhat misleading 

because this reform comprises a variety of 

actors and institutions across the country. 

Since state charter school laws do not require 

a particular program or instructional 

approach, the missions and educational 

philosophies of charter schools vary, as do the 

types of students and communities they serve, 

the accountability requirements they face and 

the degree of flexibility they enjoy. For 

example, some charter schools deliver 

instruction through independent study, 

distance learning, and home schooling 

 

16 The survey allowed respondents to provide multiple 

responses. 
4

programs.17 Many of the differences are due to 

differences in state laws and regulations. One 

should keep in mind this high degree of 

variation when comparing charter schools 

with traditional public schools.18 

Within this context of high variability, the 

charter school movement, taken as a whole, 

has grown exponentially since the first two 

charter schools opened in Minnesota in 1992. 

During the 2002-03 school year, 2,695 

schools were in operation in 36 states and the 

District of Columbia. The largest increases 

occurred between school years 1998-99 and 

1999-2000 and between 2001-02 and 

2002-03—about 500 schools opened in each 

of these time periods (see Exhibit 1-3). 

 

17 While these schools are included in the study, they 

are not a focus of this report. 

18 “Traditional public schools” refers to all schools in 

the public school system that are not designated as 

charter schools. 

The number of charter schools in operation 

varies widely by state. As Exhibit 1-4 

illustrates, the numbers range from zero 

charter schools in both Indiana and New 

Hampshire to 349 schools in California. The 

average number of charter schools per state 

has increased from 47 in 1999-2000 to 58 in 

2001-02; this increase is largely due to growth 

in the number of schools in only a handful of 

states. Since fall 1999, just five states have 

had the total number of charter schools in 

their states increase by more than 30 

schools—California,19 Florida, Ohio, 

Pennsylvania, and Wisconsin. Many states 

experienced less dramatic increases during 

this time period (i.e., an increase of less than 

10 schools). 

 

19 California, in particular, increased by nearly 

50 percent, from 238 to 349 schools. 

Exhibit 1-3

Number of Operating Charter Schools, by School Year

0

500

1,000

1,500

2,000

2,500

3,000

1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03

Year

Number of Schools

Sources: 1992-93 through 1998-99 data: Nelson et al. (2000); 1999-2000 through 2001-02 data: SRI International. 

2002-03 data: Center for Education Reform (www.edreform.com) 

Exhibit reads: In 1992-93, the number of operating charter schools was two.
5

No relationship exists between the age of a 

state’s charter school law and its number of 

operational charter schools, although some of 

the states with the largest numbers of charter 

schools have had legislation in place for many 

years. (See Appendix A-2 for the year each 

state’s charter school law passed.) More 

important factors affect the number of schools 

in operation, including the state’s regulations 

regarding authorizing bodies, processes for 

charter approval and caps on the number of 

charter schools permitted in the state. For 

example, in New Hampshire the local level 

must approve a charter before it can move 

ahead to the state board of education for 

approval. This two-tiered process has 

hindered the development of charter schools 

in the state. (At the time of data collection, the 

New Hampshire state legislature was 

considering a bill that would bypass the role 

of the local level but this bill was later 

defeated.) 

In general, charter legislation allows two 

types of charter schools: those that are new 

and those that convert from an existing public 

or private school. A state’s charter school law 

Exhibit 1-4 

Total Number of Charter Schools in Operation, 

by State (as of January 1, 2002) 

State Number of Schools State Number of Schools 

California 349 Louisiana 20 

Arizona 287 New Mexico 20 

Michigan 188 Oregon 17 

Florida 185 Alaska 15 

Texas 180 Connecticut 14 

Wisconsin 108 Delaware 10 

North Carolina 93 Idaho 10 

Ohio 92 Nevada 10 

Colorado 87 Utah 9 

Pennsylvania 77 South Carolina 8 

Minnesota 68 Arkansas 6 

New Jersey 50 Hawaii 6a

Massachusetts 42 Virginia 6 

Georgia 38 Oklahoma 5a

District of Columbia 36 Rhode Island 3a

Kansas 28 Mississippi 1 

New York 24 Wyoming 1 

Illinois 23 Indiana 0 

Missouri 21 New Hampshire 0 

Total number of states: 

37 and the District of Columbiab

 Total number of schools: 2,137 a

The number of charter schools is based on SRI’s charter school sampling database for the 2001-02 

charter school survey (and includes schools in operation as of summer 2001) because these states did not 

complete the 2001-02 state survey. 

b

Iowa, Tennessee and Maryland passed charter school laws after data were collected for this report. 

Puerto Rico passed a charter school law in 1993 but was not included in data collection. 

Source: SRI 2001-02 state charter school coordinator survey. 

Exhibit reads: As of January 1, 2002, there were 349 charter schools in operation in California. 
6

determines the types of schools that may 

operate but the most common type is the 

newly created charter school. In 2001-02, 

newly created charter schools accounted for 

77 percent of charter schools, public 

conversions for 16 percent and private 

schools for 7 percent, as illustrated in 

Exhibit 1-5.20 Some states limit the types of 

schools that may operate. For example, New 

York has a cap of 100 on the total number of 

newly created schools but has no limit on the 

number of public school conversions.21 For 

more details regarding state caps please refer 

to this evaluation’s Year 1 report 

(U.S. Department of Education, 2000). 

 

20 In order to receive funds under the PCSP, a private 

school must close and reconstitute itself as a newly 

created charter school. 

21 New York does not allow private school conversions. 

State laws also stipulate which organizations 

can legally grant charters or authorize charter 

schools. Examples of authorizing bodies 

include local school districts, state education 

agencies, institutions of higher education, 

municipal governments, and special 

chartering boards. The accountability 

relationship between a charter school and its 

authorizing body is a distinguishing 

characteristic of this reform. This evaluation 

report examined three categories of 

authorizing bodies: local school districts, state 

departments of education, and universities.22 

 

22 Schools authorized by independent or special boards 

are a fourth category. However, this group was not 

included in these analyses because of the small number 

across the country and the low percentage of schools 

they sponsor. 

Exhibit 1-5 

Distribution of Charter Schools, by Type (2001-02) 

Public conversion

16%

Private conversion

7%

Newly created

77%

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, newly created schools accounted for 77 percent of charter schools. 
7

In 2001-02, most charter schools had local 

school districts or state departments of 

education as their authorizers (with 45 and 

41 percent of charter schools authorized by 

these groups, respectively) (see Exhibit 1-6).23

Because the role of authorizers has received 

relatively little attention in the charter school 

literature, this study surveyed a sample of 

authorizing bodies in each of the three rounds 

of data collection (see Chapter 4 for more 

details). 

 

23 Note that the proportion of schools authorized by 

different authorizer types is not the same as the 

proportion of the different types of authorizers. For 

example, a large proportion of schools were authorized 

at the state level but these state authorizing bodies 

represented a small proportion of the authorizer 

universe. Chapter 4 provides more details about the 

authorizer universe. 

Like all public schools, charter schools cannot 

discriminate on the basis of race, ethnicity, or 

other student characteristics and may not 

charge tuition. State charter school laws 

usually specify allowable admission criteria. 

In 2001-02, charter schools frequently used 

sibling preference, applications and residence 

requirements as admission criteria (see 

Appendix A-3 for details on the admission 

criteria used by charter schools). During the 

2001-02 school year, 66 percent of charter 

schools reported that they had more 

applicants for admission than they had the 

capacity to serve.24 

 

24 To be eligible for PCSP funds, charter schools that 

have more applicants for admission than they have the 

capacity to serve must select students using a lottery 

process. However, only 66 percent of the 

oversubscribed schools reported using a random 

selection process or lottery, and 38 percent reported 

using waiting lists or first-come, first-served 

procedures (multiple responses were allowed). 

Exhibit 1-6 

Percentage of Charter Schools, by Type of Authorizer (2001-02) 

Type of Authorizer 

Percentage of 

Charter Schools 

(n=477) 

Local school boards or districts, county boards or offices, intermediate 

school districts 45 

State boards of education, state education agencies, chief state school 

officers 41 

Universities, colleges, community colleges 12 

Other entities, including independent or special charter school boards 2 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, local school boards or districts, county boards or offices or intermediate school 

districts authorized 45 percent of charter schools. 
8

Organization of the Report 

This chapter provided an overview of the 

evaluation and charter schools. The remainder 

of this report addresses seven key evaluation 

questions, organized by chapter as follows: 

Chapter 2: Public Charter Schools 

Program Operations 

• How does the PCSP work and how do this 

federal grant program and its state 

grantees encourage the development of 

charter schools? 

• How do federally funded charter schools 

and school planners use their PCSP 

subgrants? 

Chapter 3: Characteristics of Charter 

Schools, Students, and Staff 

• What are the characteristics of charter 

schools and the students and families who 

are involved with them? 

• What flexibility provisions are charter 

schools granted? 

Chapter 4: Charter School Accountability 

and the Role of Authorizers 

• What are the characteristics and roles of 

authorizing bodies? 

• What types of accountability relationships 

do authorizers have with their schools? 

Chapter 5: Charter Schools and State 

Performance Standards 

• To what extent are charter schools 

meeting state performance standards for 

student performance, and how do charter 

schools and traditional public schools 

compare in meeting these standards? 

In addition to these five chapters, the 

Appendices contain details about the data 

sources and methods, as well as supporting 

data for each chapter. Finally, Chapter 6 

outlines possible directions for future studies 

that are suggested by this multi-method study. 
Chapter 2 

Public Charter Schools Program Operations 

Federal support for schools has a long history 

and takes many forms, from providing land 

for public schools and colleges in the 19th 

and early 20th centuries, to providing 

categorical program aid for many purposes 

today (vocational education, special 

education, etc.). The federal government also 

supports myriad school reform initiatives, 

including charter schools. By offering 

financial support to charter schools, the 

Congress and the U.S. Department of 

Education have given a national identity to an 

otherwise decentralized charter school 

movement. 

Federal interest in supporting the 

development of the charter school movement 

began in 1993 when the Public Charter 

Schools Program and Public Schools 

Redefinition Act were proposed. No action 

was taken, however, until the reauthorization 

of the Elementary and Secondary Education 

Act (ESEA) in 1994. The PCSP was enacted 

in 1995 as Title X, Part C, of ESEA. 

The PCSP is a competitive grant program 

currently administered by the Office of 

Innovation and Improvement in the 

U.S. Department of Education (ED). The aim 

of the original legislation was to support the 

planning, development, and initial 

implementation of charter schools during their 

first three years of existence. The 1998 

reauthorization, the Charter School Expansion 

Act (P.L. 105-278), broadened the scope of 

the PCSP by allowing states to award 

dissemination subgrants to older charter 

schools that had “demonstrated overall 

success”25 to enable them to share their 

promising practices with, and form bridges to, 

the larger public school system. Congress 

again reauthorized the PCSP as part of the No 

Child Left Behind Act of 2001 (P.L. 107-

110), resulting in two important changes: (1) 

local education agencies may not deduct 

funds from a school’s subgrant without prior 

agreement by the charter school; and, (2) 

charter school developers may receive 

planning subgrants prior to the award of a 

charter even if no authorizing body has been 

identified. 

This chapter addresses the following 

evaluation questions: 

• How does the PCSP work and how do this 

federal grant program and its state 

grantees encourage the development of 

charter schools? 

• How do federally funded charter schools 

and school planners use their PCSP 

subgrants? 

As prescribed by law, PCSP state grants are 

awarded to state education agencies for a 

period of up to three years;26 these funds, in 

25 According to the statute, schools have demonstrated 

success if they have made “substantial progress in 

improving student achievement,” have “high levels of 

parent satisfaction,” and have “the management and 

leadership necessary to overcome initial start-up 

problems and establish a thriving, financially viable 

charter school.” (For more details see the law: 

www.uscharterschools.org/pdf/fr/expansion_act.pdf.) 

26 PCSP funds are awarded to states for a period of up 

to three years. ED requires a full application only at the 

beginning of the three-year period and an annual 

progress report in subsequent years. Because federal 

regulations allow grantees to carry over funds from 

year to year, proposed figures may differ from actual 

amounts received. 
10

turn, are given to charter schools and charter 

school planners as school subgrants

according to criteria for allowable activities 

specified in federal law and other criteria 

added at the state level.27 

In fiscal year (FY) 2001, 37 states28 and the 

District of Columbia had charter school 

legislation. Of these, 90 percent (33 states 

plus the District of Columbia) received PCSP 

funds.29 Charter schools or charter school 

planners in four states (Arizona,30 Mississippi, 

New Hampshire, and Wyoming) applied to 

ED directly for funds. 

Charter schools have been part of the public 

education system for more than a decade, and 

the PCSP has been in existence for almost as 

long. During this period, the national charter 

school movement has evolved, and the PCSP 

has continued to grow—in terms of the 

program’s total funding, the mean state grant 

amount and the total number of state 

education agencies that have received PCSP 

grants. 

The PCSP is a key source of start-up support 

for charter schools across the country. 

Through the PCSP, states and the federal 

government have created an infrastructure for 

supporting the ongoing development of 

charter schools. Within this context, this 

chapter discusses findings related to growth in 

 

27 Allowable activities include planning and design of 

the education program; professional development of 

staff and acquisition of equipment, materials and 

supplies. PCSP funds may not be used to support a 

private school conversion or for construction costs. 

28 New Hampshire passed a charter school law in 1995 

but had no charter schools at the time of data 

collection. 

29 The number of states receiving PCSP funds is not 

equal to the number of states responding to the survey. 

See Appendix A-1 for more details. 

30 Arizona received a state PCSP grant from FY 1995 

to FY 1997 but was determined ineligible for a state 

PCSP grant in 1997. In subsequent years, charter 

schools in Arizona have applied directly to ED. 

PCSP federal funding parameters and state 

criteria for awarding PCSP funds; the average 

amounts and the uses of PCSP funds at the 

school level; and the role and capacity of state 

charter school offices. 

The Growth of the Public Charter 

Schools Program 

Finding: While growth in the number of 

states with charter legislation has tapered 

off, the amount of PCSP awards to states 

has increased and the number of charter 

schools has continued to grow. 

The PCSP has been a major source of funding 

for charter schools. Since the program’s initial 

$6 million appropriation in FY 1995, the total 

appropriation for the PCSP has substantially 

increased with an appropriation of 

$200 million for FY 2002. The most recent 

appropriation since data collection ended for 

this study was $218.7 million in FY 2004. 

Exhibit 2-1 displays the substantial increase 

in PCSP appropriations over time. 

Although the PCSP appropriations have 

grown exponentially, the number of states 

passing charter school laws has stabilized. In 

fact, since this evaluation’s first year of data 

collection in 1999-2000, only four states have 

passed charter legislation (Indiana in 2001, 

Iowa and Tennessee in 2002 and Maryland in 

2003). (Appendix B-1 illustrates these trends.) 

As a result of the program’s total 

appropriation growing faster than the number 

of states receiving PCSP grants, the mean 

state grant has increased every year since the 

beginning of the program from $512,900 in 

FY 1995 to nearly $4.5 million in FY 2001. 

Exhibit 2-2 illustrates the increase in PCSP 

grants to states by fiscal year. 
11

Some states have received particularly large 

increases in their PCSP grants. An example is 

Massachusettsthe PCSP grant increased 

from $2.2 million in FY 2000 to $4.1 million 

in FY 2001. However, not every state has had 

an increase and, in fact, reductions in PCSP 

grants are also possible. ED regulations 

permit grantees to carry over funds from one 

fiscal year to the next, but the amount carried 

over by each grantee “gives the Department 

discretion to consider those funds in 

determining whether to reduce the amount of 

new funds made available to the grant for the 

next budget period” (U.S. Department of 

Education, 1997). In states that have not 

realized the level of charter school growth 

projected in their original PCSP application, 

carryover funds may lead to reductions in 

future allocations. 

In addition to variation in the rate of charter 

school growth across states, other reasons 

relating to the funding eligibility of schools 

have led to state-level differences in PCSP 

awards. In some states, for example, 

operational charter schools have “aged out” 

and are no longer eligible for PCSP start-up 

funding, which is for schools to use in their 

first three years. Other states have reached 

caps on the number of charter schools allowed 

under state law. For these reasons, the gap 

between the maximum and minimum PCSP 

state grants has also increased. For example, 

in FY 2001, Floridawith high growth in its 

total number of charter schoolsreceived 

nearly $24 million, compared with $200,000 

awarded to Virginia where school districts 

have been slower to embrace their authority to 

approve charters. (Appendix B-2 illustrates 

the differences between the minimum and 

maximum state grants by fiscal year.) 

State Criteria for Awarding Public 

Charter Schools Program Funds 

Finding: Within federal funding 

parameters, states award subgrants to 

schools based on state-determined 

criteria. In general, dissemination 

subgrants and start-up subgrants have 

different criteria. 

Exhibit 2-1 

History of Public Charter Schools Program Appropriations, by Fiscal Year 

$0

$50

$100

$150

$200

$250

1995 1996 1997 1998 1999 2000 2001 2002 2003

Fiscal Year

Total Appropriations in Millions

Sources: Press releases and grant award information from ED’s Web site. 

Exhibit reads: In FY 1995, the total appropriation for the PCSP was $6 million. 
12

The U.S. Department of Education awards 

PCSP funds to state education agencies and, 

in turn, to charter schools as subgrants. All 

PCSP states award start-up subgrants but the 

states have different eligibility requirements 

(within federal guidelines). Two-thirds of the 

states award start-up subgrants only after the 

school has a charter in place (but before the 

school begins operating). The remaining 

states31 allow charter school planners to 

receive these subgrants, even before a school 

has a charter in place. 

Seventy-one percent of those states with 

PCSP grants that responded to the state 

coordinator survey32 reported that they 

allowed schools to receive PCSP funds at the 

 

31 These include Alaska, Arkansas, California, District 

of Columbia, Louisiana, Nevada, New York, Ohio, 

Oregon, South Carolina, and Wisconsin. 

32 State coordinator survey data from three states are 

not included in these analyses because they did not 

respond to the survey. 

dissemination stage.33 As prescribed by 

federal law, states may not spend more than 

10 percent of their total PCSP grant on 

dissemination activities. (The secretary of 

education may waive this requirement, 

however, under certain circumstances.) 

Asked what percentage of their FY 2001 

funds states reserved for dissemination 

purposes, states reported a range from 0 to 25 

percent of their 2001-02 allocations. The 

mean amount reserved for dissemination 

grants, however, was 9 percent and only three 

states reported reserving more than 10 percent 

for FY 2001. In addition, three states did not 

reserve any of their FY 2001 grant for 

dissemination activities but reported that they 

permitted schools to receive these funds. This 

 

33 The award of dissemination grants is a state option. 

Exhibit 2-2 

Mean State Public Charter Schools Program Grant Amount, by Fiscal Year 

$0

$500,000

$1,000,000

$1,500,000

$2,000,000

$2,500,000

$3,000,000

$3,500,000

$4,000,000

$4,500,000

$5,000,000

FY95 FY96 FY97 FY98 FY99 FY00 FY01

State PCSP Grant

Sources: SRI 1999-2000 and 2001-02 state coordinator surveys. 

Exhibit reads: In FY 2001, the mean state grant was $512,900. 
13

was the case for one of two reasons: (1) the 

state did not have charter schools in existence 

long enough or (2) the state did not have 

schools during that fiscal year that met the 

funding criteria (having “demonstrated 

overall success”). 

States design their own criteria for awarding 

PCSP funds and determining subgrant 

amounts within guidelines from the federal 

government. As illustrated in Exhibit 2-3, the 

criteria states use depends on the type of 

subgrant (i.e., start-up or dissemination). 

More than half of all states reported two 

criteria for funding decisions regarding both 

types of subgrants: a state ceiling on the total 

amount a school can receive and the quality 

of the proposal.34 However, despite these 

 

34 This evaluation did not probe the definition of 

proposal “quality”; therefore, additional studies would 

similarities, state coordinators reported 

differences in the process for awarding 

dissemination subgrants and the process for 

awarding start-up subgrants. For example, 

86 percent of states reported that the quality 

of a subgrant proposal was a criterion for 

dissemination subgrants, compared with 

61 percent for start-up subgrants. In addition, 

68 percent of states reported taking into 

account the recommendations of reviewers for 

dissemination subgrants, compared with 

39 percent for start-up subgrants. Finally, 

states were more likely to consider additional 

factors, such as level of need, in awarding 

dissemination subgrants. 

 

be required to better understand state criteria. 

Exhibit 2-3 

State Reporting of Public Charter Schools Program Subgrant Criteria, 

by Type of Subgrant (2001-02) 

0

10

20

30

40

50

60

70

80

90

100

Enrollment

Demographics

Negotiation of funding level

Quality of proposal

Reviewer recommendations

Level of need

Requested funding

State ceiling on award

Percentage of States

Start-up subgrants (n=31)

Dissemination subgrants (n=22)

 Source: SRI 2001-02 state coordinator survey. 

 Exhibit reads: In 2001-02, 45 percent of states reported that they took into account enrollment for start-up subgrants, 

 compared with 23 percent that did so for dissemination subgrants. 
14

Beyond the general criteria, states may give 

priority to charter schools that serve particular 

populations of students. The data collected in 

2001-02 may indicate a shift in states’ 

priorities from the data collected in 

1999-2000. For example, the percentage of 

states that reported giving priority to charter 

schools that served a “special population of 

students”35 decreased from 36 percent in 

1999-2000 to 26 percent in 2001-02. At the 

same time, however, the proportion of states 

targeting low-income communities36 increased 

from 14 percent in 1999-2000 to 23 percent in 

2001-02. 

Public Charter Schools Program 

Subgrants: Awards to Schools 

Finding: Subgrant award amounts vary 

by state. The average school subgrant in 

2001-02 ranged from $20,000 in 

Rhode Island to $263,000 in Oklahoma. 

An important—but elusive—part of this study 

has been an ongoing effort to understand the 

prevalence and range of subgrant awards 

within states. The study team intended to 

analyze data from several sources to 

determine (1) the proportion of charter 

schools nationally and by state that have 

received PCSP subgrants and (2) the average 

subgrant awards nationally and by state. 

Unfortunately, uneven data quality from the 

states that provided information37 led to 

difficulty in compiling a comprehensive 

 

35 States reporting this priority in 2001-02 include 

California, Florida, Michigan, New Mexico, Ohio, 

Oregon, Virginia, and Wisconsin. States did not clarify 

the type of population. 

36 States that reported targeting schools in low-income 

communities in 2001-02 include Florida, Michigan, 

New Mexico, Oregon, South Carolina, Virginia, and 

Wisconsin. 

37 Only 24 of the 35 states receiving PCSP grants 

provided school subgrant data. 

subgrant database.38 For example, some states 

submitted district-level award information or 

other aggregations, such as the cumulative 

amounts awarded to individual charter 

schools across multiple years. The mixed data 

quality is likely due to the fact that states are 

not required to provide subgrant information 

to ED as part of their reporting requirements. 

Two additional difficulties in determining the 

proportion of schools receiving funds per year 

are the changing numbers of operating charter 

schools in any given year and the fact that 

some states allow subgrants to be awarded to 

planners for schools that may never open. 

Overall, these data collection efforts did not 

result in clear-cut answers. 

As a result of these issues, the question 

regarding the proportion of charter schools 

that have received PCSP subgrants is based 

on the surveys of charter school directors. In 

2001-02, 61 percent of charter school 

directors reported receiving PCSP start-up 

subgrants at some point in time, and 

19 percent reported receiving dissemination 

subgrants.39 PCSP subgrants may be even 

more widespread than these data indicate 

since charter school directors in case study 

sites often had difficulty recognizing this 

funding source by name. 

 

38 The efforts to determine award amounts included: (1) 

acquisition of a database compiled by another 

researcher with details regarding PCSP funding from 

1995 to 1997 for some states; (2) extraction of 

information from PCSP program files, which contained 

annual reports but did not have systematic information 

on subgrants; (3) data collection from state charter 

school coordinators in 1999-2000; (4) a follow-up 

request in 2000-01 because not all states provided the 

information previously; and (5) a third and final round 

of data collection from states in 2001-02. 

39 Note that these statistics derive from separate survey 

items and are not intended to be summed. 
15

Average school subgrants vary widely across 

states. PCSP has recommended that states 

award schools a maximum of $450,000 over 

the course of three years, with the assumption 

that schools will receive up to $150,000 each 

year. However, most of the subgrants in a 

state, on average, are lower than this amount, 

as illustrated in Exhibit 2-4. Although one 

state’s average subgrant was only $20,000, 

the averages in a majority of states fell 

between $80,000 and $150,000 per school. 

(Appendix B-3 provides details about average 

state subgrants over time, while Appendix 

B-4 includes the minimum, maximum and 

average state subgrant in 2001-02.) 

An additional source of information on 

subgrant awards was ED’s conference for 

state charter school coordinators held in April 

2002. During this conference, ED estimated 

that the average subgrant was approximately 

$100,000an amount consistent with the 

study’s findings. State reports of their average 

subgrant amounts, which most likely included 

a mix of FY 2001 and FY 2002 funds from 

ED, are included in Appendix B-5. 

Exhibit 2-4 

Average Public Charter Schools Program Subgrant Amount, 

by State (FY 2001) 

$0 $50,000 $100,000 $150,000 $200,000 $250,000 $300,000

Texas 

Rhode Island 

Pennsylvania 

Oklahoma 

Ohio 

New York 

New Mexico 

Nevada 

Missouri 

Massachusetts 

Kansas 

Illinois 

Idaho 

Georgia 

District of Columbia 

Connecticut 

Colorado 

California 

Arkansas 

PCSP Subgrant

Source: SRI 2001-02 PCSP subgrant database. 

Exhibit reads: In FY 2001, the average PCSP subgrant amount given to charter schools in Arkansas was $216,667. 
16

PCSP funds are the most prevalent source of 

start-up funds available to charter schools 

across states (Exhibit 2-5). In other words, 

PCSP funds are widely accessibleall 

schools in states with charter legislation are 

able to apply for start-up funds from the 

PCSP program either through their state or by 

applying directly to the U.S. Department of 

Education. 

Besides PCSP funds, 84 percent of states 

reported private donors (e.g. Walton Family 

Foundation, Ford Foundation) as sources of 

start-up funds for charter schools. Moreover, 

nearly half (45 percent) of state coordinators 

reported the availability of state start-up 

funds. 

Uses of Public Charter Schools 

Program Subgrants at the School-level 

Finding: PCSP start-up and 

dissemination subgrants primarily fund 

professional development and 

technology. In addition, charter schools 

use start-up subgrants to purchase 

curricular and instructional materials. 

By design, start-up and dissemination 

subgrants serve different purposes. While the 

purpose of start-up subgrants is to allow the 

school to open and operate successfully 

during its first few years, the intent of 

dissemination subgrants is to facilitate the 

sharing of practices between charter schools 

and other public schools. According to federal 

law, to receive a dissemination grant, charter 

schools must be in existence for three 

consecutive years and must have 

demonstrated “overall success,” including 

academic achievement, high level of parental 

satisfaction, and strong management and 

leadership. As illustrated in Exhibit 2-6, 

charter schools used start-up funds to 

purchase curricular and instructional materials 

(88 percent), support professional 

PCSP 

Exhibit 2-5 

Start-up Funding Sources for Charter Schools, 2001-02 

0

20

40

60

80

100

Funding Source

Percentage of States (n=31)

Note: The number of states (n) is equal to 31 because three states (Hawaii, Oklahoma, and Rhode Island) did not complete 

the survey. Five states named “Other” sources that provided start-up funds to charter schools, including parents, the USDA 

Rural Development program and education management organizations. 

Source: SRI 2001-02 state coordinator survey. 

Exhibit reads: In 2001-02, 100 percent of states reported that the PCSP was a source of charter school start-up funds. 

Authorizer-16% Other sources-16%

Private 

donors

83% State 

45% 

PCSP 

100% 
17

development and training (80 percent), and 

purchase computer hardware and software 

(79 percent). 

As with start-up subgrants, charter schools 

report that dissemination subgrants support 

professional development and technology. In 

addition, schools target these funds toward 

slightly different activities, including travel 

and networking. The uses of dissemination 

funds at the school-level are included in 

Appendix B-6. 

Capacity of State Education Agencies 

and Technical Assistance Provided by 

States 

Finding: State charter school offices 

have responsibilities relating to the 

development, operation and monitoring 

of charter schools but most states have 

limited staff to perform these functions. 

According to federal charter law, states may 

retain up to 5 percent of PCSP funds for 

administrative expenses, and the data indicate 

that these funds play an important role in 

supporting state charter school activity. In 

2001-02, all states reported setting aside some 

portion of the 5 percent for administrative 

costs. According to state coordinators, these 

funds were primarily allocated toward salaries 

and benefits, travel, conferences, supplies, 

and training. 

Exhibit 2-6 

Charter School Reports of Uses of Public Charter Schools Program 

Start-up Subgrants, 2001-02 

0

10

20

30

40

50

60

70

80

90

100

Instructional

materials

Professional

development

Computer

materials

Consultants Recruiting &

public

relations

Renovations Staff salaries Renting or

leases

Uses of Funds

Percentage of Schools (n=292)
18

State charter school offices play an important 

role in the charter school movement, carrying 

out a variety of charter school responsibilities. 

Asked to name their four primary 

responsibilities, state coordinators reported 

the following: 

• Providing information and policy 

clarification to charter schools 

(74 percent); 

• Overseeing the application and charter 

approval process (71 percent); 

• Providing in-person technical assistance 

and professional development to charter 

schools (57 percent); and, 

• Monitoring student performance 

(43 percent). 

Regarding whether or not state charter school 

offices have the capacity to perform these 

tasks, nearly all state education agencies 

(97 percent) reported having an office or staff 

dedicated to charter schools in 2001-02, 

although capacity varied greatly. According 

to reports by state coordinators, charter school 

offices had, on average, three full-time 

equivalent professional and administrative 

staff members (FTEs) dedicated to charter 

work, although the most common (modal) 

response was to have one FTE.40 These data 

 

40 The data for FTE were as follows: mean=3, 

median=2, mode=1, maximum=12.5 (Ohio) and 

minimum=0.1 (Mississippi). State coordinators were 

asked: “Including yourself, how many full-time 

equivalent professional and administrative staff are 

dedicated to charter schools in your state?” Some states 

may have collaborative team approaches to working 

with charter schools; as a result, some states (e.g., 

Ohio) may report the involvement of a variety of staff 

members who do not solely focus on charter school 

activities. Typically, however, states have a small 

number of positions dedicated to charter school 

business. 

suggest that state charter school offices have 

very limited staff capacity to perform the 

many responsibilities related to charter school 

development, operation, and monitoring. For 

example, Arizona had 287 charter schools but 

only two FTEs dedicated to charter work in 

2001-02. Similarly, Wisconsin had 108 

charter schools but only 1.85 FTEs.

Since its creation eight years ago, the PCSP 

has become a centerpiece in the continuing 

growth of the charter school sector 

nationwide. During the PCSP’s lifetime, the 

number of charter schools nationwide has 

continued to grow rapidly, and additional 

states have passed charter laws. Moreover, the 

PCSP’s total appropriation has increased 

significantly, as has the average state grant 

and the total number of states that have 

received support through the program. While 

it is not clear whether this pattern of growth 

will continue, the emphasis on a strong role 

for charter schools in the No Child Left 

Behind Act of 2001 may mean that the PCSP 

and its grantees will play an even more 

prominent role in the educational reform 

agenda over the coming decade. 
Chapter 3 

Characteristics of Charter Schools, Students, and Staff 

One cannot understand the importance of 

charter schools as an educational reform 

without first examining the characteristics of 

these schools, as well as the types of students 

and teachers who choose them. Furthermore, 

it is important to consider how charter schools 

compare with traditional public schools on 

key characteristics. This study uses data from 

multiple sources, including the study’s charter 

school surveys administered in 2000-01 and 

2001-02; 1998-99 data from the National 

Study of Charter Schools conducted by 

RPP International; and data from the 1999-

2000 Schools and Staffing Survey (SASS),41

which included surveys of a nationally 

representative sample of schools and teachers, 

as well as a supplemental study of the 

universe of charter schools. 

In addition to providing descriptive details 

about charter schools, students and teachers, 

the second section of this chapter addresses 

the issues of flexibility, autonomy, and 

controlcentral aspects of charter school 

theory. 

This section of Chapter 3 focuses on the 

following evaluation question: 

• What are the characteristics of charter 

schools and the students and families 

who are involved with them? 

 

41 The SASS data allow comparisons between charter 

schools and traditional public schools. From this point 

forward, comparisons of charter schools and traditional 

public schools do not mention the source (SASS) and 

year (1999-2000) in the text for simplification 

purposes, but this information is included in every 

exhibit. 

Characteristics of Charter Schools 

Finding: The national profile of charter 

schools is changing, with increasing 

student enrollments and changing 

demographics. However, charter schools 

are distinct from traditional public 

schools because of lower total student 

enrollments, unique grade level 

configurations and a variety of 

instructional approaches. 

Student Enrollment. In the earliest years of 

the charter school movement, profiles of the 

charter school universe showed that the 

schools were generally very small—often 

well under 100 students (see, for example, the 

four reports produced by the National Study 

of Charter Schools covering the years 

1996-97, 1997-98, 1998-99 and 1999-2000). 

To a large extent, extreme smallness in a 

school’s first years has been an artifact of the 

common charter school implementation 

strategy of beginning with a limited number 

of grade levels and “growing” the school 

upward, downward or both. 

Charter schools are becoming larger. As the 

data presented in Exhibit 3-1 demonstrate, the 

median student enrollment for charter schools 

has steadily increased over the last few years, 

from 137 in 1998-99 to 190 in 2001-02.42

Some slight shifts have occurred in student 

enrollment by schools with different gradelevel configurations. K-8 charter schools have 

shown the greatest increase in median student 

enrollment since 1999-2000 (see 

Appendix C-1). Middle schools, on the other 

 

42 The longitudinal comparisons included in this report 

do not include tests of significance because the data are 

derived from different sources. 
20

100

110

120

130

140

150

160

170

180

190

200

1998-99 1999-2000 2000-01 2001-02

School Year

Number of Students

hand, have shown a pattern of decline in 

median enrollment, most likely related to the 

growth in the overall proportion of K-8 and 

K-12 charter schools, as the middle grades 

have become embedded in the larger gradelevel configurations. 

Exhibit 3-1 

Median Student Enrollment in Charter 

Schools, 1998-1999 to 2001-2002 

Sources: 1998-99 data: Nelson et al. (2000); 1999-2000 data: 

Public Charter School SASS survey; 2000-01 and 2001-02 

data: SRI 2000-01 and 2001-02 charter school surveys. 

Exhibit reads: In 1998-99, the median charter school enrollment 

was 137. 

Although the median enrollment in charter 

schools has increased over time, charter 

schools remain considerably smaller than 

traditional public schools (see Exhibit 3-2). In 

fact, traditional public schools were nearly 

three times as large as charter schools in 

1999-2000—458 students in traditional public 

schools compared with 169 students in charter 

schools. Across the different grade-level 

configurations, charter school median student 

enrollment ranged from 80 in primary schools 

(serving grades K-3) to 240 in schools with a 

K-12 configuration. Median enrollment in 

traditional public schools ranged from 180 in 

K-12 schools to 675 in high schools. Overall, 

the median student enrollment was highest in 

traditional public high schools. In contrast, 

charter high schools were among the smallest 

charter schools, perhaps because many were 

deliberately founded as alternatives to large, 

comprehensive high schools and frequently 

serve small numbers of students who have 

dropped out of traditional public schools. 

Grade-level Configuration. Charter school 

founders have the flexibility to choose their 

school’s grade level configuration based on a 

mission, philosophy of teaching, or child 

development approach. As a result of this 

flexibility, charter schools have implemented 

a variety of configurations, including nontraditional configurations, such as K-8 and 

K-12 schools. As Exhibit 3-3 indicates, 

charter schools were more likely to serve 

students in grades K-8 or K-12 than 

traditional public schools. Traditional public 

schools, on the other hand, were more likely 

to operate as standard elementary schools 

(i.e., serving grades K-6) than charter schools 

(49 compared with 22 percent, respectively). 
21

Exhibit 3-2 

Median Student Enrollment in Charter Schools and 

Traditional Public Schools, by Grade-Level Configuration (1999-2000) 

Note: Grade levels configurations follow conventions established by RPP and are defined as follows: Primary includes 

only grades K-3; Elementary begins with K and goes no higher than grade 6; Middle ranges from grade 5 to grade 9; 

Middle-high includes any of grades 6-8 and any of grades 9-12 and no grades K-5; High ranges from grade 9 to grade 12; 

K-8 includes any of grades K-3, any of grades 4-6 and any of grades 7-8; K-12 includes any of grades K-3, 4-6, any of 

grades 7-8 and any of grades 9-12. 

Sources: 1999-2000 public charter school SASS survey and public school SASS survey. 

Exhibit reads: In 1999-2000, the median student enrollment in the primary level configuration was 80 for charter schools, 

compared with 350 for traditional public schools. 

0 100 200 300 400 500 600 700 800

All schools

K-8

Elementary

High

Middle-high

Middle

K-12

Primary

Grade-Level Configuration

Median Enrollment

Traditional Public Schools

Charter Schools
22

Instructional Strategies. While charter 

schools have the opportunity to use 

alternative instructional strategies (e.g., 

distance learning), 91 percent of the charter 

schools surveyed in 2001-02 used classroombased instruction as their primary 

instructional delivery method as illustrated in 

Exhibit 3-4.43 A small proportion of charter 

schools delivered instruction in nontraditional ways in 2001-02, including 

independent study (9 percent), instruction via 

Internet or satellite (6 percent), and homebased instruction (sometimes referred to as 

“home schooling”) (2 percent). 

 

43 Exhibit 3-4 also indicates that a small proportion of 

charter schools combine multiple approaches (since the 

percentage of schools using the various instructional 

methods in one school year does not equal 

100 percent). 

Beyond delivering classroom-based 

instruction, charter schools were 

implementing a number of recent educational 

reforms. For example, charter schools were 

more likely than traditional public schools to 

use block scheduling (58 percent versus 

43 percent); interdisciplinary teaching 

(59 percent versus 48 percent); and 

“looping”44 (48 percent versus 26 percent). 

These differences were statistically significant 

(p<.01). This evidence suggests that charter 

schools were able to experiment with 

different instructional and organizational 

strategies. 

 

44 An approach in which teachers remain with students 

for two or more years. 

Exhibit 3-3 

Grade-Level Configurations of Charter Schools and 

Traditional Public Schools, 1999-2000 

0 10 20 30 40 50 60

Primary

Middle/high***

Middle***

K-12***

High***

Elementary***

K-8***

Grade Levels

Percentage of Schools

Charter Schools

Traditional Public Schools

***p<.01 (Indicates significant difference between charter schools and traditional public schools in the percentage of 

schools serving particular grade level configurations.) 

Note: Grade level configurations follow conventions established by RPP and are defined as follows: Primary includes 

only grades K-3; Elementary begins with K and goes no higher than grade 6; Middle ranges from grade 5 to grade 9; 

Middle-high includes any of grades 6-8 and any of grades 9-12 and no grades K-5; High ranges from grade 9 to grade 

12; K-8 includes any of grades K-3, any of grades 4-6 and any of grades 7-8; 

K-12 includes any of grades K-3, 4-6, any of grades 7-8 and any of grades 9-12. 

Sources: 1999-2000 public charter school SASS survey and public school SASS survey. 

Exhibit reads: In 1999-2000, 25 percent of charter schools served grades K-8, compared with 6 percent of traditional 

public schools. This difference is statistically significant. 
23

Characteristics of Charter School 

Students 

Finding: Compared with traditional 

public schools, charter schools enroll 

more African American students, fewer 

whites, and slightly higher proportions of 

students eligible for free and reducedprice lunches. Charter schools also 

attract high proportions of low 

performing students. 

The demographic characteristics of students 

in charter schools have been a topic of great 

interest to observers of this educational 

reform. Key questions center on the 

differences between the student populations 

of charter schools and traditional public 

schools, as well as the extent to which charter 

schools are targeting and enrolling specific 

student populations. 

Exhibit 3-4 

Primary Methods of Delivering Instruction in Charter Schools (2001-02) 

Method of Instruction Percentage of Schools (n=477) 

Classroom-based instruction 91 

Work- or community-based learning on a regular basis 6 

Instruction via Internet or satellite 6 

Distance learning 1 

Home-based instruction 2 

Independent study 9 

Note: Respondents could select more than one instructional delivery method. 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 91 percent of charter schools used classroom-based instruction. 

Exhibit 3-5 

Proportions of White and Minority Students in Charter Schools, 

1998-99 to 2001-02 

0

10

20

30

40

50

60

70

1998-99 1999-2000 2000-01 2001-02

School Year

Percentage of Students

White

Minority

Sources: 1998-99 data: Nelson et al. (2000); 1999-2000 data: Public Charter 

School SASS survey; 2000-01 and 2001-02 data: SRI 2000-01 and 2001-02 

charter school surveys. 

Exhibit reads: In 1998-99, 52 percent of students in charter schools were minority, 

compared with 48 percent of white students. 
24

Student Race and Ethnicity. The charter 

school student population during the 

2001-02 school year was 38 percent African 

American, 37 percent white, and 19 percent 

Hispanic (these data are included in 

Appendix C-2). Over time, there have been 

slight shifts in the charter school student 

population. The two most significant trends 

are the 11-percentage-point decrease in the 

proportion of white students and the 

14-percentage-point increase in the proportion 

of African American students between 

1998-99 and 2001-02. 

Exhibit 3-5 on the previous page illustrates 

the change in the proportion of all minority 

students compared with the proportion of 

white students. As this figure indicates, the 

charter school population of minority and 

white students was nearly evenly divided in 

1998-99 but by 2001-02 had shifted to 

approximately one-third white and two-thirds 

minority. This is an important change in the 

national profile of charter schools, one that 

future studies should examine more closely. 

A comparison of the racial and ethnic 

composition of student populations in charter 

schools and traditional public schools in the 

1999-2000 school year indicates that 

traditional public schools enrolled higher 

proportions of white students. In fact, 

traditional public schools enrolled 17 percent 

more white students than charter schools 

(Exhibit 3-6). Conversely, charter schools 

enrolled greater percentages of African 

American and Hispanic students than 

traditional public schools. 

Other Student Characteristics. Over time, a 

greater proportion of students enrolled in 

charter schools were eligible for free and 

reduced-price lunches, rising from 39 percent 

in 1998-99 to 53 percent in 2001-02, as 

illustrated in Exhibit 3-7. This increase may 

Exhibit 3-6 

Student Race and Ethnicity in Charter Schools and 

Traditional Public Schools, 1999-2000

0

10

20

30

40

50

60

70

White*** African American*** Hispanic or

Latino***

Other Minority***

Racial/Ethnic Category

Percentage of Students

Charter School Students

Traditional Public School Students

***p<.01 (Indicates significant difference between charter schools and traditional public schools in 

the percentage of students by race and ethnicity.) 

Sources: 1999-2000 public charter school SASS survey and public school SASS survey. 

Exhibit reads: In 1999-2000, 46 percent of charter school students were white, compared with 

63 percent of students in traditional public schools. This difference is statistically significant. 
25

be associated with the growth in minority 

populations discussed above or with a higher 

proportion of charter schools electing to 

participate in the National School Lunch 

Program. The proportion of limited English 

proficient (LEP) students enrolled in charter 

schools and the proportion with 

Individualized Education Programs (IEPs)45

have remained stable over time. 

 

45 Each public school child who receives special 

education and related services must have an IEP. 

Charter schools also had higher proportions of 

students eligible for free and reduced-price 

lunches and lower proportions of special 

education students with IEPs than traditional 

public schools. These differences were 

statistically significant. The difference in the 

proportion of LEP students in charter and 

traditional public schools was not statistically 

significant. Exhibit 3-8 compares charter 

schools and traditional public schools on 

these additional student characteristics. 

Exhibit 3-7 

Changes in Selected Characteristics of Students in 

Charter Schools, 1998-99 to 2001-02 

Percentage of Charter School Students 

Student Characteristic 1998-99 1999-2000 2000-01 2001-02 

Free or reduced-price lunch 39 43 46 53 

Individualized Education Program (IEP) 8 9 8 10 

Limited English proficient (LEP) 10 8 8 9 

Sources: 1998-99 data: Nelson et al. (2000); 1999-2000 data: Public Charter School SASS survey; 2000-01 and 

2001-02 data: SRI 2000-01 and 2001-02 charter school surveys. 

Exhibit reads: In 1998-99, 39 percent of charter school students were eligible for free or reduced-price lunches 

compared to 43 percent in 1999-2000, 46 percent in 2000-01 and 53 percent in 2001-02. 

Exhibit 3-8 

Comparisons of Selected Characteristics of Students in Charter Schools 

and Traditional Public Schools, 1999-2000 

 Percentage of Students 

Student Characteristic Charter schools Traditional public schools 

Free or reduced-price lunch*** 43 38 

Individualized Education Program (IEP)*** 9 12 

Limited English proficient (LEP) 8 7 

***p<.01 (Indicates significant difference between charter schools and traditional public schools in the percentage of 

students on various characteristics, e.g,. free or reduced-price lunch.) 

Sources: 1999-2000 public charter school SASS survey and public school SASS survey. 

Exhibit reads: In 1999-2000, 43 percent of charter school students were eligible for free or reduced-price lunches, 

compared with 38 percent of students in traditional public schools. This difference is statistically significant. 
26

Some charter schools actually seek out 

specific populations of students because of 

the school’s educational mission or program 

design. More than one quarter of charter 

schools targeted low-performing students, 

dropouts or potential dropouts, or students 

from low-income communities. However, 

many student populations were attracted to 

charter schools regardless of the school’s 

mission or design. While 28 percent of 

schools report targeting low-income and lowperforming students, 74 percent reported 

attracting these groups of students. Similarly, 

less than one-quarter of charter schools 

targeted gifted and talented or special 

education students, but more than half of the 

schools attracted these students. (See 

Appendix C-3 for more details on attracted 

and targeted student populations in charter 

schools.) 

Charter School Teachers 

Finding: Compared with traditional 

public school teachers, charter school 

teachers are more likely to be African 

American and less likely to have full 

state certification in the subject they 

teach. Teachers in charter schools 

participate in a variety of professional 

development activities in slightly higher 

proportions than traditional public 

school teachers. 

Teacher Race and Ethnicity. In 2001-02, 

whites accounted for 63 percent of charter 

school teaching staff. As Exhibit 3-9 

illustrates, there has been a slight decrease in 

the proportion of white teachers and an 

increase in the proportion of African 

American teachers over time. This trend is 

similar to the trend in the proportion of 

Exhibit 3-9 

Racial and Ethnic Distribution of Charter School Teachers, 1999-2000 to 2001-02 

0 10 20 30 40 50 60 70 80

White

African American

Hispanic or Latino

Other Minority

Racial/Ethnic Category

Percentage of Teachers

2001-02 (n=444 schools)

2000-01 (n=377 schools)

1999-2000 (n=870 schools)

Note: Charter school directors provided data on the race and ethnicity of teachers. Calculations were made 

by dividing the number of charter school teachers in each racial category by the total number of charter 

school teachers. “Other Minority” includes Asian, American Indian or Alaska Native and, in 2000-01 and 

2001-02, Native Hawaiian or other Pacific Islander.

Sources: 1999-2000 data: Public Charter School SASS survey; 2000-01 and 2001-02 data: SRI 2000-01 and 

2001-02 charter school surveys. 

Exhibit reads: In 1999-2000, 73 percent of charter school teachers were white, compared with 66 percent in 

2000-01 and 63 percent in 2001-02. 
27

African American charter school students, 

though not as pronounced. The proportions of 

teachers from most other ethnic minority 

groups have remained constant. 

As indicated in Exhibit 3-10, the majority of 

teachers in both charter schools and 

traditional public schools were white; 

however, charter schools had lower 

proportion of white teachers (73 percent) than 

traditional public schools (84 percent). 

Furthermore, charter schools had more 

African American teachers (16 percent) than 

did traditional public schools (9 percent). 

Future studies should examine these data 

more closely to determine why charter 

schools have greater proportions of African 

American teachers on staff. 

Certification. According to charter directors, 

nearly two-thirds of charter school teachers 

have full state certification for the subjects 

they teach (64 percent in 2001-02). 

Furthermore, 10 percent of charter school 

teachers had special education credentials and 

5 percent had bilingual credentials in 2001-02 

(in addition to full state certification). It is 

important to examine these data by state 

because frequently there is a relationship 

between differences in the proportion of 

teachers with certification and state policies 

governing teacher certification in charter 

schools. Flexibility over teacher hiring and 

certification practices is one of the important 

areas of autonomy that some charter schools 

enjoy. (For more details about state policies 

regarding teacher certification see Appendix 

C-4). Data from the 1999-2000 SASS survey 

revealed that charter school teachers were less 

likely to have full certification and more 

likely to have emergency credentials, 

including temporary, provisional, and 

probationary certifications, than their 

traditional public school peers.46 Overall, 

92 percent of traditional public school 

teachers had full certification in 

1999-2000, compared with 79 percent of 

teachers in charter schools. Teachers in 

charter schools were nearly three times more 

likely to have emergency credentials 

compared with teachers in traditional public 

schools (21 percent versus 8 percent). These 

differences between the proportion of teachers 

 

46 Note: These data are from the SASS survey 

administered to teachers, not schools. 

Exhibit 3-10 

Racial and Ethnic Distribution of Teachers in Charter Schools and 

Traditional Public Schools, 1999-2000 

Percentage of Teachers 

Racial and Ethnic Category Charter schools Traditional public schools 

White*** 73 84 

African American*** 16 9 

Hispanic or Latino*** 8 5 

Other Minority 3 2 

***p<.01 (Indicates significant difference between charter schools and traditional public schools 

in the percentage of teachers by race and ethnicity.) 

Sources: 1999-2000 public charter school SASS survey and public school SASS survey. 

Exhibit reads: In 1999-2000, 73 percent of charter school teachers were white, compared with 

84 percent of teachers in traditional public schools. This difference is statistically significant. 
28

with full certification and emergency 

credentials in traditional public schools and 

charter schools are statistically significant 

(p<.01). 

Future studies should examine charter schools 

and teacher certification in more depth, since 

the Elementary and Secondary Education Act 

of 1965 (ESEA), as amended by the No Child 

Left Behind Act of 2001 (NCLB) 

requirements for “highly qualified teachers” 

were not in place at the time of data 

collection. 

Professional Development Opportunities. 

The availability of professional development 

opportunities is an important indicator of 

teacher professionalism in schools. According 

to charter school directors, teachers had 

opportunities to develop their knowledge and 

skills beyond the traditionally ineffective oneshot workshops. Sixty two percent of charter 

school teachers received release time to work 

collaboratively with other instructional staff, 

and more than half were able to participate in 

peer observation and critiques. Approximately 

one-third of teachers were allowed release 

time for independent professional 

development activities. (See Appendix C-5 

for more data on charter school teachers’ 

professional development opportunities.) 

Based on data from the 1999-2000 SASS, 

which asked teachers about their participation 

in a variety of professional development 

activities over the last year, participation in 

professional development was high in both 

charter and traditional public schools. 

Teachers in both types of schools reported 

that they had regularly-scheduled 

collaboration with other teachers on 

instructional issues (74 percent in each case). 

Teachers in traditional public schools were 

more likely than teachers in charter schools to 

attend workshops, conferences, or training 

(95 percent of teachers in traditional public 

schools versus 90 percent of teachers in 

charter schools). In all other areas, including 

making observational visits to other schools, 

charter school teachers participated at higher 

rates than traditional public school teachers. 

In most cases, these differences in 

participation rates were statistically 

significant. (These data are included in 

Appendix C-6.) 

Parent Involvement 

Finding: Charter schools are more likely 

than traditional public schools to have 

high levels of parent involvement in the 

areas of budget decisions, governance, 

instructional issues, parent education 

workshops, and volunteering. 

A central tenet of the charter school 

movement is to provide “opportunities for 

educators and parents [emphasis added] to 

create the kinds of schools they believe make 

the most sense” (Nathan, 1996, p.1). 

Moreover, respondents in the charter schools 

visited for this evaluation typically referred to 

high levels of parent involvement and 

multiple avenues for parents to become 

involved. The survey data reinforced this 

finding. Overall, charter school directors 

reported that parents were involved with the 

schools in various capacities. More than half 

of charter school directors reported that 

parents played wide-ranging roles from 

performing clerical tasks for the school to 

serving on advisory boards. (See 

Appendix C-7.) The majority of school 

directors reported that parents were involved 

in a variety of activities including serving on 

school advisory committees (87 percent of 

schools reported parents playing this role), 

serving on school governing boards 

(82 percent of schools), and supervising lunch 

or field trips (81 percent of schools). 
29

When asked to identify specific areas in 

which parents were required to participate, 

most charter school directors reported that 

parents were voluntarily involved in school 

activities. 47 (Refer to Appendix C-8 for details 

on required and voluntary parent involvement 

activities.) This finding suggests that schools 

either did little to enforce parent contracts or 

that such written agreements with parents 

were not directly linked with specific 

activities but more general in nature. Data 

from the 2001-02 site visits support the notion 

that charter schools did little to enforce 

contracts, thus making any involvement 

essentially voluntary. 

According to SASS data, charter schools were 

more likely than traditional public schools to 

report high levels of parent involvement in 

several areas (see Exhibit 3-11). Schools 

reported activities in which at least 50 percent 

of parents participated, with the most frequent 

areas of involvement for both charter schools 

and traditional public schools being 

attendance at open houses and attendance at 

schoolwide parent-teacher conferences. 

 

47 A parent-volunteer requirement would normally be 

specified in a school’s charter and made known to 

parents at the time of application. Some researchers 

argue that these requirements serve as admission 

criterion that may exclude some families (Garcia and 

Garcia, 1996; Lopez, Wells, and Holme, 2002), while 

others argue that the requirements are an indication of 

the strong relationship that charter schools forge with 

parents (Nathan, 1996). 

These data indicate that charter schools and 

traditional public schools have similar levels 

of involvement in traditional parent 

involvement activities. 

There are indications, however, that charter 

schools may have higher degrees of parent 

involvement. For example, compared with 

traditional public schools, more than twice as 

many charter schools had high degrees of 

parent involvement in five areas: budget 

decisions, governance, instructional issues, 

parent education workshops, and 

volunteering. Nevertheless, the overall 

proportion of charter schools reporting parent 

involvement in these more intensive activities 

remains fairly low—less than 20 percent of 

charter schools reported high levels of 

involvement in nearly all of the areas (with 

the exception of volunteering).
30

Exhibit 3-11 

Parental Involvement in Charter Schools and Traditional Public Schools, 1999-2000 

(Percentage of schools in which parental involvement was at least 50 percent) 

***p<.01 (Indicates significant difference between charter schools and traditional public schools in the percentage of 

schools reporting high levels of parental involvement, i.e., more than 50 percent of parents involved.) 

Sources: 1999-2000 public charter school SASS survey and public school SASS survey. 

Exhibit reads: 74 percent of charter schools and 75 percent of traditional public schools reported that at least 

50 percent of parents attend school open houses. This difference is not statistically significant. 

Charter School Autonomy 

In theory, a charter specifies areas in which a 

school will have more autonomy than 

traditional public schools in exchange for 

particular accountability requirements. This 

section of Chapter 3 focuses on the extent to 

which charter schools enjoy additional 

flexibility from rules and regulations 

governing public schools and increased 

autonomy (or control) over school decisions.

Charter school theory posits that these schools 

will be exempted from bureaucratic 

requirements and will enjoy considerably 

more control over core operational elements 

such as the school year, daily schedules, 

curriculum and instruction, hiring of 

personnel and budgets than do traditional 

public schools. However, the actual flexibility 

and autonomy that charter schools have is 

dictated by the state legislation that allows 

their creation. As a result, there is 

considerable variation in the charter school 

flexibility story. 

0 10 20 30 40 50 60 70 80 90 100

Open houses

Parent conferences

Volunteering***

Parent education

workshops***

Instructional issues***

Governance***

Budget decisions***

Parent Involvement Activities

Percentage of Schools 

Traditional Public Schools (n=8,432)

Charter Schools (n=870)
31

Flexibility 

Finding: Only one-third of charter 

schools automatically receive waivers 

from state policies and regulations but 

many schools receive waivers on a caseby-case basis. Most state charter school 

policies do not allow exemptions from 

student assessment requirements. 

An assumption of charter school theory is that 

schools will receive waivers from the codes 

and regulations that apply to traditional public 

schools. Surveys with state directors indicate 

that the reality is differentonly 37 percent 

of states granted automatic waivers to charter 

schools in 2001-02 (automatic waivers do not 

include health, safety, and due process 

provisions). Nine percent48 did not permit any 

waivers of state laws, rules or regulations. 

The remaining 54 percent of states allow 

some (but not all) regulations to be waived 

automatically or on a case-by-case basis. 

 

48 Two states (Massachusetts and Utah) and 

Washington, D.C., were in this category. 

These findings indicate that a charter school’s 

degree of flexibility is closely associated with 

the state policy environment in which it is 

located.49 The majority of states reported that 

charter schools were exempt from 

requirements about the length of the school 

day or year (68 percent), staff hiring and 

firing policies (65 percent), and other teacher 

policies, such as teacher contract year and 

tenure requirements (61 percent). (See 

Appendix C-9 for more details about state 

exemptions.) 

Charter school directors reported using 

waivers in varying degrees. While more than 

half of charter schools received waivers from 

teacher certification requirements, staff hiring 

and firing policies, tenure and contract 

requirements, policies regarding teacher 

salaries and pay schedules and policies 

regarding the control of finances, many 

charter schools did not report these and other 

waivers (see Exhibit 3-12). Very few charter 

 

49 The school data support this finding. In 2001-02, 

only 49 percent of all charter school respondents 

reported that their charters enabled them to depart from 

or waive state laws or local policies that applied to 

other schools. 

Exhibit 3-12 

Charter School Reports of Waivers, 2001-02 

Type of Waiver Percentage of Schools (n=229) 

Teacher policies (contract year and tenure requirements) 61 

Teacher/staff hiring/firing policies 56 

Control of finances 56 

Teacher salary/pay schedule 56 

Teacher certification requirements 53 

Curriculum requirements 48 

Length of school day or year 43 

District student assessment policies 35 

Student admission policies 33 

Student attendance policies 21 

State student assessment policies 5 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 61 percent of charter schools reported receiving waivers on teacher policies. 
32

schools (5 percent) reported that they could 

waive state student assessment policies. 

Control 

Finding: Charter schools have control 

over many key decisions but frequently 

their school-level authority is shared with 

one or more other entities, including the 

school’s authorizing body or education 

management organization. 

Asked whether charter schools had full 

authority, shared authority, or no authority 

over a variety of school matters, more than 

half of all charter schools nationally reported 

that they had a full authority over key aspects 

of their own operations—with the exception 

of teacher certification requirements (see 

Exhibit 3-13). Schools were least likely to 

have authority over teacher certification 

policies. Authorizing bodies corroborated the 

degree of authority reported by schools. 

A little known fact illustrated in the exhibit 

that follows is that many charter schools 

report that they share authority over key 

decisions with another entity, e.g., the state or 

authorizer. The schools that shared authority 

with other groups were most likely to report 

sharing this control with their authorizers, 

except in the case of curriculum and teacher 

certification, for which authority was shared 

with the state. Schools associated with 

education management organizations (EMO) 

also reported sharing authority with these 

partners, as discussed in the following pages. 

The Role of Education Management 

Organizations and Community-Based 

Organizations 

Finding: For the nearly 20 percent of 

charter schools that partner with an 

education management organization or 

community-based organization, these 

entities have extensive roles in the 

development and ongoing operation of 

charter schools, from hiring staff to 

managing the school and ensuring 

compliance with state and federal 

regulations. 

Most states and authorizing bodies do not 

prohibit charter school relationships with 

EMOs,50 such as Edison Schools, and 

community-based organizations (CBO), such 

as Boys and Girls Clubs of America.51

However, the frequency of these relationships 

remains low: in 2001-02, 19 percent of 

charter schools had arrangements with EMOs 

and 16 percent had partnerships with CBOs. 

EMOs and CBOs, sometimes referred to as 

for-profits and nonprofits respectively, have 

emerged as prominent components of the 

charter school movement. In fact, within these 

relatively small percentages lies an important 

story. 

 

50 Education management organization, or EMO, are 

defined as, “A firm dedicated to operating schools or 

districts, usually on a for-profit basis” (Brown Carter, 

2003). 

51 Although state laws frequently require nonprofit 

organizations—usually the schools themselves—to 

hold the charter, charter schools in most states were 

free to establish relationships with EMOs and CBOs. 
33

Charter schools that had relationships with 

EMOs were more likely to have universities 

as their authorizers than other authorizing 

bodies. This finding was influenced by 

Michigan, which has large numbers of charter 

schools chartered by universities and also has 

a large proportion of charter schools managed 

by EMOs (such as National Heritage 

Academies).52 Schools chartered by state 

agencies were more likely to have 

arrangements with CBOs than schools 

chartered by other types of authorizers. 

According to charter school directors, EMOs 

play a variety of roles. In most of their 

schools, EMOs ensured compliance with state 

and federal regulations; provided technical 

assistance; administered personnel and 

benefits; administered the school’s budget; 

 

52 In Michigan, 74 percent of charter schools contract 

out services to EMOs (Miron and Nelson, 2002). Of 

the schools with university authorizers that reported 

EMO arrangements, all but one were in Michigan. 

and provided, leased, or located facilities (see 

Exhibit 3-14). On average, CBOs tended to 

play similar roles to EMOs (although slightly 

lower proportions of charter schools reported 

these roles for CBOs). However, a few 

differences stand out: charter schools with 

CBOs were much less likely to report 

involvement of CBOs in curriculum and 

instruction decisions or the hiring of staff. 

(Chapter 4 provides more detail on 

accountability roles played by EMOs and 

CBOs.) 

Exhibit 3-13 

Charter School Reports of Level of Authority, 2001-02 

Percentage of Schools (n=477) 

Type of Authority 

Full 

authority 

Shared 

authority 

No 

authority 

Daily schedule 84 16 <1 

Purchasing of supplies and equipment 79 21 <1 

Staff hiring, discipline and dismissal 72 26 1 

Other budgetary expenses, not including salaries and 

benefits 67 31 2 

Student disciplinary policies 64 36 <1 

Student assessment policies 63 33 4 

School calendar 59 37 5 

Curriculum 59 39 2 

Teacher certification requirements 45 45 9 

Sources: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 84 percent of charter schools reported having full authority over the school’s daily 

schedule, compared with 16 percent who reported sharing this authority and less than 1 percent who reported having 

no authority. 
34

Exhibit 3-14 

Roles Played by Education Management Organizations and Community-Based 

Organizations, According to Charter School Directors (2001-02) 

 Percentage of Schools 

Role 

Charter schools 

with EMOs 

(n=83) 

Charter schools 

with CBOs 

(n=74) 

Ensuring compliance with state and federal regulations 94 78 

Providing technical assistance and professional development 91 65 

Administering personnel and benefits 89 62 

Administering budget 87 70 

Providing, leasing, locating capital equipment or facilities 84 70 

Monitoring progress toward and compliance with terms of our 

charter 80 72 

Monitoring student performance 72 58 

Managing the overall operation or administration of this school 71 59 

Providing or brokering student services (e.g., special education or 

LEP) 68 49 

Representing school in negotiations with the charter school 

authorizer 65 69 

Directing the curriculum and instruction 64 38 

Providing seed or start-up funds 64 55 

Hiring staff 60 43 

Sources: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 94 percent of charter schools with EMOs reported that the EMO assisted their school in 

ensuring compliance with state and federal regulations, compared with 78 percent of charter schools with CBOs 

reporting that their CBO played this role. 
Chapter 4 

Charter School Accountability and the Role of Authorizers 

Authorizers play an important role in the 

charter movement, serving as “intermediaries” between charter schools and the state 

policymakers who created charter school 

legislation (Bulkley, 1999). In fact, these 

bodies are key partners in charter school 

accountability relationships, holding 

sponsored charter schools accountable for the 

goals listed in their charters, as well as in 

compliance with other state or federal laws 

and regulations. However, charter school 

legislation in the states has provided virtually 

no guidance on how authorizers should 

approach account-ability processes. 

Furthermore, both policymakers and 

researchers have largely overlooked 

authorizers (see, however, Palmer and Gau, 

2003, Hassel and Herdman, 2000). 

Understanding the role of authorizers is 

important to understanding this educational 

reform. 

Two question areas divide this chapter, with 

the first focusing on the roles of authorizing 

bodies and the second focusing on the 

accountability relationships with charter 

schools. 

This section discusses the types of authorizing 

bodies, the reasons they sponsor charter 

schools and the services they provide. 

However, charter schools are also accountable 

to other groups (Hill et al., 2001). Therefore, 

this chapter also discusses the multiple groups 

to whom charter schools are accountable, with 

an emphasis on authorizers because they have 

the unique authority to close or not renew a 

school when it fails to meet the goals stated in 

the school’s charter. As pioneers in 

performance-based accountability, charter 

school authorizers may inform broader 

attempts to hold public schools accountable 

(Hassel and Herdman, 2000; Vergari, 2001). 

Characteristics of Charter School 

Authorizers and Their Schools 

Finding: Local education agencies are 

the most common type of authorizer but 

grant the smallest number of charters, on 

average. In some states, state education 

agencies or universities may also 

authorize charter schools. These 

authorizing bodies tend to issue relatively 

larger numbers of charters. 

State laws determine which agencies may 

award charters or “authorize” charter schools. 

As discussed in Chapter 1, this report focuses 

on three types of authorizing bodies: local 

education agencies (including county offices 

of education), state education agencies and 

institutions of higher education.53 Exhibit 4-1 

illustrates the distribution of authorizers by 

type. In 2001-02, 91 percent of authorizers 

were local education agencies; relatively few 

were state education agencies or universities. 

From 2000-01 to 2001-02, the number of 

local authorizers increased by 175 and the 

universe overall increased by 186. 

 

53 In 2001-02 the evaluation did not sample the small 

number of “other” authorizers that includes municipal 

governments and independent chartering entities. 
36

Exhibit 4-1 

Distribution of Charter School 

Authorizers, by Type (2001-02) 

Universe of Authorizers, 

As of Summer 2001 (N=643)

91%

3% 5% 1%

Local (n=584) State (n=17)

University (n=35) Other (n=7)

Source: SRI 2001-02 authorizer sampling 

frame. 

Exhibit Reads: In 2001-02, 91 percent of the 

universe of authorizers were local school 

districts. 

Although the number of schools sponsored by 

individual authorizers was small (on average, 

authorizers had a total of three schools in 

operation), it is important to remember that 

relatively few authorizers chartered a large 

number of schools.54 Conversely, a large 

proportion of authorizers chartered a very 

small number of schools; three-quarters of 

authorizers had two or fewer schools in 

operation in 2001-02. 

 

54 In addition to having schools in operational and 

planning stages, some authorizers had schools that had 

been (1) chartered but never opened (8 percent), (2) 

closed as charters but continued to operate 

(1 percent), or (3) closed as charters and were no 

longer operating (10 percent). 

While local authorizers were the most 

common type of authorizer, they chartered the 

smallest number of schools, on average.55

States sponsor the largest number of schools, 

on average, with approximately five times as 

many state-authorized schools in operation 

compared with those of university authorizers 

and 15 times as many as those of local 

authorizers. 

Why Authorizers Charter Schools 

Finding: Authorizers’ reasons for 

awarding charters are changing over 

time, with a greater emphasis in 

2001-02 on creating competition and 

responding to political pressure. 

During focus groups in 1999-2000, charter 

school authorizers expressed concern about 

the extra work entailed in overseeing charter 

schools and uncertainty about their new roles. 

Authorizers become involved in sponsoring 

charter schools for a number of reasons 

despite these concerns. The top reasons that 

authorizers reported for sponsoring charter 

schools included: 

• Creating competition in the public school 

system; 

• Responding to public and political 

pressure; and, 

• Creating alternatives for students and 

parents. 

Authorizers’ primary reasons for sponsoring 

charter schools have shifted over a two-year 

period. As Exhibit 4-2 shows, authorizers 

were more likely to cite the creation of 

competition in the public school system as a 

 

55 These data were as follows: Local authorizers had 

two schools in operation, on average, and less than one 

school in planning. State authorizers had 30 schools in 

operation, on average, and five schools in planning. 

Universities had six schools in operation, on average, 

and less than one school in planning. These differences 

by type were statistically significant. 
37

reason for chartering activity in 2001-02 than 

they were the previous year. More 

respondents also noted that being responsive 

to public or political pressure was of 

importance. Conversely, in 2001-02, 

authorizers were less likely to report that the 

motivation for chartering included a desire to 

stimulate development of schools targeted to 

students with particular needs or to make a 

particular school or program available to 

more students. 

One possibility for this shift in rationale for 

sponsoring charter schools is that competition 

within the public school system is becoming 

more widely accepted. In addition, in some 

locations, transferring to charter schools may 

be an option for families whose children have 

been attending schools identified as failing 

under NCLB. It will be important for 

evaluators to continue documenting how 

NCLB requirements affect charter schools 

and authorizers. 

Not all authorizers have the same rationale for 

sponsoring charter schools. In fact, reasons 

for sponsoring schools differ by authorizer 

type. For example, state authorizers were 

much more likely than local authorizers and 

universities to view improving the public 

school system, creating competition, 

responding to public or political pressure, and 

fulfilling the mandates of the state charter law 

as “very important” in their decision to 

sponsor schools. (See Appendix D-1 for more 

detail about authorizers’ reasons for 

sponsoring charter schools by type of 

authorizer.) 

Exhibit 4-2 

Reasons Authorizers Sponsor Charter Schools, 2000-01 and 2001-02 

(Respondents Citing Reasons as “Very Important”) 

0 10 20 30 40 50 60 70 80 90 100

Stimulate development of schools

Increase program availability

Generate income

Target certain populations

Test educational practices

Improve public schools

Create alternatives for students and parents

Respond to public/political pressure

Create competition in the school system

Percentage of Schools

2001-02 (n=113)

2000-01 (n=100)

Sources: SRI 2000-01 and 2001-02 authorizer surveys. 

Exhibit reads: In 2000-01, 14 percent of authorizers reported that creating competition was a very important 

reason for sponsoring charter schools, compared with 71 percent of authorizers in 2001-02. 
38

Authorizer Capacity 

Finding: Only one-third of authorizing 

bodies have an office or staff dedicated to 

charter school activities, suggesting 

limited capacity to address charter school 

issues. However, charter authorizers may 

be developing their capacity through 

increased involvement in both formal 

and informal networks. 

A fundamental assumption of the philosophy 

of charter school accountability is that 

authorizers will track the progress of the 

charter schools that they sponsor to ensure 

that they are meeting the terms of their 

charters. To meet this responsibility, 

authorizers must have the capacity to take on 

this accountability role. One indicator of 

capacity is having a particular office or staff 

dedicated to charter school-related matters. 

During focus groups in 1999-2000, many 

authorizers stated that competing 

responsibilities made fulfilling their charter 

duties very difficult. This situation does not 

appear to have changed significantly over the 

years: Only 36 percent of authorizers had an 

office or staff dedicated to charter school 

work in 2001-02 (compared with 30 percent 

the year before). However, certain types of 

authorizers were more likely to have this 

arrangement. The majority of state authorizers 

(85 percent) had a charter school office or 

staff, compared with 64 percent of 

universities and 33 percent of local 

authorizers. These differences were 

statistically significant (p<.01). 

Another indicator of authorizer capacity is the 

number of full-time equivalent (FTE) staff 

dedicated to charter school work. Analysis of 

FTEs by types of authorizer reveals some 

interesting findings. Across all types of 

authorizers, an average of three FTE 

professional and administrative staff in the 

authorizing agency were dedicated to charter 

school work.56 Local and state authorizers had 

an average of three FTE staff, while 

university authorizers had seven FTE staff 

dedicated to charter work. The difference in 

the number of FTE staff by type of authorizer 

is statistically significant (p<.10). 

One way in which authorizers can build 

capacity is through involvement in both 

formal and informal networks with other 

charter school authorizers. According to 

Vergari (2001), “school authorizer offices 

stand to gain a great deal through the 

exchange of experiences, ideas and 

information with each other” (p. 138). In 

1999-2000 and 2000-01, many authorizers 

reported that they had too little interaction or 

discussion with other authorizers. By 2002, a 

general trend emerged in which more 

authorizers were becoming formally and 

informally involved with their peers. In 

2001-02, the percentage of authorizers not

involved in either type of network dropped to 

17 percent, from 39 percent in 2000-01. 

(Appendix D-2 provides more details.) 

Authorizer Services to Charter 

Schools 

Finding: Authorizing bodies deliver a 

wide range of services to their charter 

schools, including administrative 

oversight, assistance in meeting 

regulations and the provision of special 

education services. To an increasing 

extent, authorizing bodies provide these 

services at a cost to the schools. 

 

56 Please note that only the authorizers with an office or 

staff dedicated to charter schools answered the survey 

question on FTE. 
39

The accountability role of authorizers is the 

most important role they play; however, 

authorizers play an additional role (one that is 

seldom discussed) in providing assistance and 

services to their charter schools. In fact, 

85 percent of charter schools reported that 

their authorizers were a source of assistance. 

While the 2000-01 evaluation report for this 

study discusses this role of authorizers in 

detail, this section highlights additional 

findings and differences between the two 

years of data. 

Authorizers may either provide direct support 

or “broker” support by linking schools to 

other agencies. In 2001-02, authorizers were 

more likely to provide in-kind support to their 

schools than in the previous year. In addition, 

increasing proportions of authorizers were 

brokering support (that is, helping charter 

schools acquire services) in 

2001-02 compared with 2000-01. (See 

Appendix D-3 for more details.) In addition to 

providing assistance, many authorizers 

provide services ranging from general 

oversight to social services or supplies and 

equipment to their schools. Specifically, the 

most common types of assistance or services 

include: 

• Administrative oversight, monitoring and 

evaluation; 

• Assistance in meeting state and federal 

regulations; 

• Special education services; and 

• Special education testing and assessment. 

Services provided to schools varied greatly by 

type of authorizer, with local authorizers 

offering the widest range of services (see 

Appendix D-4). 

As discussed in the 2000-01 report, it is 

important to remember that schools do not 

necessarily receive these services free of 

charge. Instead, authorizers have a number of 

different financial arrangements with their 

charter schools, including retaining funds that 

pass through their agency to the school, 

charging a flat fee,57 and charging schools on 

a fee-for-service basis. In 2000-01, 

authorizers supplied their charter schools with 

the majority of the most widely provided 

services at no cost to the schools. Although in 

many cases authorizers still provide free 

services to their charter schools, the trend is 

toward retaining funds or charging a fee for 

service, depending on the service. 

The Charter School Accountability 

Process 

Beyond awarding charters, authorizing bodies 

are responsible for monitoring school 

performance in areas such as financial record 

keeping, special education services, 

compliance with regulations and student 

performance. If charter schools do not meet 

performance goals, authorizers may revoke or 

fail to renew a school’s charter. The purpose 

of this section of Chapter 4 is to examine 

these accountability relationships between 

charter schools and their authorizers. The 

section addresses the following evaluation 

question: 

• What types of accountability relationships 

do authorizers have with charter schools? 

The small amount of research focusing on 

accountability and the role of authorizers has 

found that variation in the amount of 

oversight that authorizers provide is linked to 

differences in type of authorizer (Hill et al., 

2001); to differences in philosophical views 

and capacity levels (Hassel and Herdman, 

2000); and to uncertainty about roles (Wells, 

1998). Bulkley (2001) argues that authorizers 

find themselves in an “accountability bind,” 

as they face challenges to the underlying 

 

57 According to NCLB, authorizers may not charge a 

flat fee for services unless this fee is mutually agreed 

upon by both the charter school and the authorizer. 

However, NCLB was not in effect at the time of this 

study. 
40

assumptions about charter school 

accountability ranging from difficulty in 

measuring performance to concerns that 

closing schools will reflect poorly on the 

charter school movement. 

The accountability process for charter schools 

involves three phases: the application process, 

the monitoring process and the 

implementation of sanctions. At all stages of 

the process, authorizers are the key players in 

charter school accountability. However, many 

other groups also play roles in charter school 

monitoringthe following section also 

discusses these groups and their overlapping 

roles. 

Accountability Phase 1: The 

Application Process 

Finding: Through the application 

process, authorizing bodies screen 

applications, denying charters because of 

problems relating to the proposed 

instructional strategies, governance 

procedures, accountability provisions 

and business plans. 

The charter school application process is the 

first stage in becoming a charter school, as 

founding groups seek an authorizer to 

“charter” their schools. Some authorizers 

view a rigorous application process as a way 

to influence school quality and improve the 

extent to which schools are held accountable 

(Bulkley, 2001). Authorizers may choose to 

take advantage of this opportunity to screen 

applicants (and deny charters) because closing 

a school that has already opened has proved 

to be difficult (Vergari, 2001; Bulkley, 2001). 

During this stage, authorizers focus on a 

number of different areas that they believe are 

critical to the successful operation and 

sustainability of a school. The 2000-01 report 

of this evaluation discussed the variety of 

application procedures for charter schools, 

such as formal deadlines and public hearings, 

and data from 2001-02 generally confirm 

these earlier findings. In addition, exactly the 

same proportion of authorizers (12 percent) 

provided additional assistance with funding or 

waivers to applicants targeting student groups 

from low-income communities in both years. 

For most authorizers, the application process 

is not a continuous process of reviewing 

applications. In 2001-02, for example, more 

than half of authorizers reported that they did 

not receive any charter applications in 2001.58

Even when they received applications, 

authorizers did not approve all applicants’ 

proposals. In fact, of the authorizers that 

reported receiving applications in 2001: 

• 21 percent did not charter any. 

• 19 percent chartered 1-25 percent of the 

applications. 

• 15 percent chartered 26-75 percent of the 

applications. 

• 45 percent chartered 76-100 percent of the 

applications. 

Local authorizers were likely to charter most 

(76-100 percent) of the applications they 

received, although they tended to receive 

fewer applications. States and universities, on 

the other hand, received larger numbers of 

applications but chartered lower proportions 

of them. 

Whether their criteria are explicit or not, 

authorizers rank certain application elements 

as most important, including accountability 

provisions, curricular and instructional 

strategies, mission and goals of the school, 

health and safety issues and the assessment 

system. More authorizers denied charter 

 

58 States (the type of authorizer that sponsors the most 

schools) were more likely than local and university 

authorizers to have received applications in 2001. 
41

applications because of problems or concerns 

with their applications in 2001-02 than in the 

previous year (33 percent versus 23 percent).59

Authorizers cited problems in the following 

areas as reasons for denying charter 

applications: 

• Curricular and instructional strategies; 

• Governance and management procedures; 

• Accountability provisions; and, 

• The business plan. 

Appendix D-5 provides more information 

about authorizers’ reasons for denying charter 

applications. 

Accountability Phase 2: The 

Monitoring Process 

Finding: Although authorizers and states 

reserve legal authority to monitor charter 

schools, multiple groups are involved in 

this process. The result is a complex 

system of accountability for charter 

schools. 

The monitoring process occurs after planning 

groups have been awarded charters by 

authorizers. This process is complex, and it is 

not clear to whom these schools are 

accountable despite the designation of the 

authorizer as the responsible agency in most 

charter laws. In fact, schools reported being 

monitored by various entities, including 

authorizers, their own governing boards, their 

states, and in some cases, EMOs or CBOs. 

 

59 Eighty-five percent of states denied charter 

applicants because of problems or concerns with their 

applications, compared with 52 percent of universities 

and 30 percent of local authorizers. 

This section discusses the different roles of 

these entities in monitoring charter schools 

and the overlapping nature of charter school 

monitoring. 

Clear and measurable goals are important 

components of accountability systems 

because authorizers have difficulty holding 

schools accountable when goals are vague or 

undefined (Wells, 1998). Both schools and 

authorizers reported that most charter schools 

had measurable goals in their charters, 

particularly goals focusing on academic 

achievement and student attendance. 

Authorizers reported that, on average, 

91 percent of their schools had measurable 

goals in student academic performance and 

79 percent had measurable goals in student 

attendance. 

Charter schools used a number of different 

assessment strategies, the most common 

being performance-based tests and normreferenced tests. Student interviews or 

surveys were the least common assessment 

strategy in charter schools. As Exhibit 4-3 

indicates, states, districts or authorizers tend 

to require norm- and criterion-referenced 

tests. On the other hand, schools tend to 

choose portfolios, student and parent surveys 

and the demonstration of students’ work as 

additional assessment tools. 
42

Monitoring by Authorizers. The monitoring 

role of authorizers flows directly from the 

philosophy of charter schools: Public agencies 

are responsible for holding these schools 

accountable to the terms of their charters. 

Paradoxically, this fundamental tenet of the 

charter school movement is left to authorizers 

themselves to put into practice, with little 

guidance from state charter laws. In this new 

system of accountability, authorizers must 

determine what is important to monitor and 

what procedures they will use to hold their 

schools accountable. 

Authorizers reported monitoring nearly all of 

their schools in the following areas: 

• Compliance with federal or state 

regulations; 

• Student achievement results on statewide 

assessments; 

• Enrollment numbers; 

• Financial record keeping and viability; 

and, 

• Special education services. 

Exhibit 4-4 indicates the areas that authorizers 

reported monitoring and provides the average 

percentages of schools they monitored.60 For 

example, authorizers monitored compliance 

with federal or state regulations for 91 percent 

of their schools, on average. 

For some authorizers (28 percent), the 

intensity of monitoring varied across schools 

because a portion of their charter schools 

received extra monitoring and oversight. 

States were more likely to vary the level of 

monitoring and oversight. In fact, 77 percent 

of state authorizers reported providing 

different levels of monitoring and oversight 

across schools, while only 25 percent of local 

and 38 percent of university authorizers 

reported such differences. These differences 

by type were statistically significant (p<.01). 

This variation is most likely linked to the 

greater number of schools chartered by state 

authorizers. 

 

60 The top five areas reportedly monitored by 

authorizers are the same for both authorizers 

(Exhibit 4-5) and schools (Appendix D-6). 

Exhibit 4-3 

Assessment Strategies Used by Charter Schools, 2001-02 

 Percentage of Schools 

Assessment Strategies 

Percentage of Schools 

(n=477) Using Strategy 

Use: 

school’s 

choice 

Use: required 

by state, 

district or 

authorizer 

Standardized norm-referenced tests 90 (n=430) 26 74 

Criterion-referenced test with proficiency levels 

or cut scores 82 (n=383) 35 65 

Performance-based tests 91 (n=434) 52 48 

Behavioral indicators, such as attendance, 

expulsion and college application rates 83 (n=392) 54 46 

Parent satisfaction surveys 86 (n=412) 74 26 

Student portfolios 76 (n=361) 81 19 

Student interviews or surveys 71 (n=340) 80 20 

Students’ demonstration of their work 86 (n=410) 82 18 

Note: Schools were first asked if they use a particular strategy. Only those who responded yes were asked to specify if 

the use was the school’s choice or required by the state, district, or authorizer. 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 90 percent of charter schools (n=430) reported using standardized norm-referenced tests. 

Twenty-six percent of those 430 schools chose to use this option, while 74 percent were required by the state, district, 

or authorizer. 
43

Authorizers used a number of different 

procedures or requirements to monitor their 

charter schools, including having schools 

report their standardized test scores, 

reviewing their compliance with regulations, 

and requiring fiscal audits and annual reports. 

(See Appendix D-7 for more detail about 

authorizer monitoring procedures.) 

How frequently authorizers monitored 

schools differs by specific topic or area. (See 

Appendix D-8 for data on the frequency of 

monitoring by authorizers.) Authorizers 

primarily monitored their charter schools 

“annually” or “more than once a 

year”depending on the focus of the 

monitoringrather than “at the end of the 

charter cycle.”61 They tended to monitor 

 

61 “Annually,” “more than once a year,” and “at the end 

financial record keeping and viability, 

enrollment numbers and special education 

services more than once a year. Monitoring 

these areas more frequently than other areas 

may be necessary because these are new 

schools with leaders who often lack 

experience managing a budget, and their 

viability depends on keeping enrollment 

numbers at a certain level. Furthermore, the 

provision of special education has always 

been an area of concern in charter schools, so 

it is not surprising that this was an area 

monitored more frequently. Authorizers tend 

to monitor student achievement results on 

statewide and other standardized tests 

annually (as they do for traditional public 

schools). 

 

of the charter cycle” were the survey categories. 

Exhibit 4-4 

Focus of Authorizer Monitoring, 2001-02 

Monitoring Focus 

Average Percentage of Schools 

Monitored by Authorizers 

(n=116)

Compliance with federal or state regulations 91 

Student achievement results on statewide assessment 91 

Enrollment numbers 91 

Financial record keeping and viability 89 

Special education services 85 

Alignment of curriculum to state standards 83 

Other student performance indicators, such as attendance rates 80 

School management, leadership or governance 78 

Student discipline and school safety 77 

Student achievement results on other standardized tests 75 

Student admission and selection procedures 74 

Student performance on performance-based tests 72 

Instructional practices 66 

Parent satisfaction 60 

Diversity of student body 58 

Parent or community involvement 58 

Staff performance and/or attendance 52 

Relationship with education management organizationa 41 

Staff satisfaction 39 

a

This item was answered by only those 78 authorizers that reported having relationships with EMOs. 

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, authorizers reported monitoring compliance with federal or state regulations for 

91 percent of their schools, on average. 
44

In addition to authorizers, charter schools 

reported being accountable to a number of 

other groups, as illustrated in Exhibit 4-5. The 

solid black arrows represent the core 

accountability requirements, usually specified 

in the state charter school legislation. Charter 

schools are accountable to their authorizers, 

who, in turn, are accountable to the state. (If 

the state were the authorizer, there would be a 

solid arrow between the state and the charter 

school.) The dotted arrows indicate 

accountability relationships that some charter 

schools have with governing bodies, states 

and EMOs or CBOs. Legally, these 

relationships may be of a different order than 

relationships with the authorizer or the state, 

but in some cases, charter schools consider 

them their most important reporting 

requirements. For example, it is quite 

common for charter directors to assert that 

they feel most accountable to their governing 

board (even though it could be argued that the 

school and the board are the same legal 

entity.) These complex accountability 

relationships can have varied effects, from 

strengthening the organizational viability of 

the school (positive) to creating confusing 

lines of accountability and detracting from the 

school’s core mission (negative) (Hill et al., 

2002). 

Exhibit 4-5 

Charter School Accountability to Multiple Entities 

Note: The solid black arrows represent the core accountability requirements, usually specified in the state 

charter school legislation. The dotted arrows indicate accountability relationships that some charter 

schools have with governing bodies, states and EMOs or CBOs. 

Exhibit reads: Charter schools have a legal accountability relationship with their authorizer but also 

report being accountable to their own governing bodies and education management 

organizations/community-based organizations. 

State Charter School 

Governing Board 

Education Management 

Organization/CommunityBased Organization Authorizer 

Charter 

School 
45

Monitoring by School Governing Bodies.

In both 2000-01 and 2001-02, charter schools 

reported that their own governing boards were 

more involved in monitoring their activities 

than their authorizing bodies, states, and other 

entities. These findings support past research 

(e.g., Hill et al., 2001). Monitoring by a 

school’s own governing body is an important 

internal accountability step, but it is only part 

of the overall monitoring process. 

The areas most commonly monitored by 

charter school governing bodies were staff 

performance or attendance (reported by 

77 percent of schools); student discipline and 

school safety (76 percent); parent satisfaction 

(74 percent); school management, leadership 

or governance (72 percent); parent or 

community involvement (71 percent); and 

instructional practices (71 percent). Again, in 

almost all monitoring areas, the school 

governing body was the most frequently 

identified monitoring entity. See 

Appendix D-9 for more detail on charter 

school reports of monitoring by school 

governing bodies. 

Monitoring by State Departments of 

Education. Like authorizers, state education 

agencies play a multifaceted monitoring role 

with respect to charter schools. This role 

begins with their implementation of state 

charter laws, which usually lay out in general 

terms the accountability expectations for state 

education agencies, authorizers and schools. 

Because charter schools are public schools, 

state accountability systems and oversight 

responsibilities apply to them. 

An additional twist on the state role is the fact 

that 43 percent of states with charter laws 

include state-level bodies or boards as charter 

school authorizers. This section presents the 

charter school monitoring role of states in 

general, with additional analyses separating 

the states that are authorizers from those that 

are not. This discussion, based primarily on 

data from the 2001-02 survey of state charter 

school coordinators, focuses on the 

monitoring role of the state. (For more 

information about the responsibilities of statelevel charter school offices, see Chapter 2.) 

Exhibit 4-6 indicates the primary monitoring 

responsibilities of states.62 As this exhibit 

shows, states that are authorizers have a role 

in monitoring their schools, particularly in 

overseeing the charter application process. 

What is interesting is that the states that are 

not authorizers also claimed a role in all of 

these areas, especially the monitoring of the 

charter school application process, student 

performance and progress toward the terms of 

the charter. 

Charter schools also reported that state-level 

bodies (including those that were authorizers) 

monitored their progress. At least one-third of 

charter schools reported state-level 

monitoring of the following areas: student 

achievement results on statewide assessments 

(63 percent); special education services 

(56 percent); compliance with federal or state 

regulations (55 percent); financial record 

keeping and viability (49 percent); other 

student performance indicators, such as 

attendance rates (41 percent); student 

performance on performance-based tests 

(41 percent); enrollment numbers 

(37 percent); and the alignment of curriculum 

to state standards (37 percent). In general, 

state-level monitoring is likely to increase 

under the requirements of NCLB. 

 

62 Respondents reported the top four responsibility 

areas of their charter school offices or staff. 
46

Like authorizers, states use a number of 

different strategies to monitor the progress of 

their charter schools, regardless of whether 

they are authorizers. Two of the most 

frequently used strategies by both types were 

annual reports and informal site visits to 

schools. Formal site visits were more 

common among states that were authorizers, 

and third-party evaluations were more 

common among non-authorizer states. On the 

other hand, both types frequently used fiscal 

audits. (See Appendix D-10 for more detail 

on state monitoring procedures for charter 

schools.) 

Monitoring by Education Management 

Organizations and Community-Based 

Organizations. As discussed in Chapter 3, 

19 percent of charter schools have an 

operational arrangement with Education 

Management Organizations (EMOs), and 

16 percent have an arrangement with 

Community-Based Organizations (CBOs). 

These charter schools reported an interesting 

and under-examined role of EMO and CBO 

relationships in monitoring their schools. 

More than two-thirds of charter schools with 

EMO and CBO relationships reported that 

these entities monitored their compliance with 

regulations and their progress toward the 

terms of their charters. In addition, more than 

half of charter schools with EMOs and CBOs 

reported that these entities monitored student 

performance. Appendix D-11 provides more 

details. 

Charter schools reported differences between 

the two groups, with EMOs reportedly 

involved in monitoring a larger proportion of 

charter schools for each category. For 

example, while more than 70 percent of 

charter schools reported that their EMO 

monitored student performance less than 

60 percent of charter schools with CBOs 

reported this. This monitoring role of EMOs 

and CBOs is an important area for future 

research. 

In summary, many more agencies have a role 

in monitoring charter schools than are 

typically involved in monitoring traditional 

public schools. Although the assumption is 

that authorizers are responsible for monitoring 

Exhibit 4-6 

Monitoring Responsibilities of State Charter School Office or Staff, 2001-02 

0 10 20 30 40 50 60 70 80 90 100

Overseeing charter application process

Monitoring student performance 

Monitoring progress toward terms of charter

Overseeing imposition of any sanctions

Monitoring fiscal or personnel management

Percentage of States

States that are not authorizers (n=20)

States that are authorizers (n=15)

Total (n=35)

Source: SRI 2001-02 state coordinator survey. 

Exhibit reads: In 2001-02, 20 percent of states that are not charter school authorizers reported that their charter school 

offices or staff monitor the fiscal and personnel management of charter schools, compared with 33 percent of states that 

are authorizers and 26 percent of all state respondents. 
47

their schools’ progress through requirements 

specified in state charter school laws, the 

charter schools themselves report multiple 

strands of reporting requirements. What 

makes these complicated monitoring 

relationships more confusing is that all of the 

monitoring groups do not have the authority 

to implement sanctions if the process 

uncovers problems or noncompliance with 

federal or state laws or regulations or with the 

school’s charter which is, legally speaking, a 

contract between the school and its 

authorizers. 

Accountability Phase 3: The 

Implementation of Sanctions 

Finding: Authorizers reported 

implementing informal sanctions more 

often than formal sanctions—few 

authorizers have revoked or not renewed 

a charter. 

Although many entities may monitor charter 

schools, authorizers are generally the only 

agencies with the authority to implement 

formal or informal sanctions as a result of the 

monitoring (some states may reserve this right 

through the state charter law).63 

Authorizers usually grant charters for three to 

five years, and when the term is completed, 

the school must apply for renewal. The formal 

sanctions used by authorizers are nonrenewal 

of the charter and revocation before the term 

of the charter is completed. In 2001-02, 

authorizers reported that from zero to 52 

schools were up for renewal. About half 

(48 percent) of authorizers had one or more 

schools up for renewal. 

 

63 States were asked whether formal sanctions had been 

imposed but these data were too general to estimate the 

prevalence of sanctions. Schools were also asked about 

sanctions but the sample consisted of charter schools in 

operation, providing an incomplete picture. 

Several studies have found that authorizers 

have difficulty closing schools due to political 

factors (Hassel and Herdman, 2000; Hill et 

al., 2001; Bulkley, 2001). Hassel and 

Herdman identify several disincentives for 

authorizers in terminating a charter, ranging 

from public relations to financial issues. 

Vergari (2001) argues that closing the school 

may damage the reputation of the authorizer. 

Based on survey data, small numbers of 

authorizers have revoked or not renewed 

charters.64 

On average, 96 percent of charter schools that 

had participated in the renewal process had 

their charters renewed by authorizers. Of the 

55 authorizers that had charter schools up for 

renewal, only four authorizers had not 

renewed one or more charters. Eight 

authorizers had charter schools that were still 

in the renewal process at the time of data 

collection, with the outcomes as yet unknown. 

(Comparing these rates with the proportion of 

traditional public schools that have been 

sanctioned through closure or reconstitution 

was beyond the scope of this study.) 

Revocation of the charter is another formal 

sanction seldom used by authorizers. 

Six percent of authorizers had implemented 

this sanction.65 States were more likely to 

have applied this sanction than the other types 

of authorizers: 38 percent of state authorizers 

compared with 5 percent of local authorizers 

and 14 percent of university authorizers. 

The total proportion of authorizers that either 

revoked or did not renew a charter in 2001-02 

was 12 percent (this is the same percentage of 

authorizers as in 2000-01). The authorizers 

that implemented formal sanctions reported 

 

64 According to the Center for Education Reform, 194 

charter schools had closed (of those that had been 

opened at one time) as of October 2002 (see 

www.edreform.com). 

65 Authorizers reported closing a total of 24 schools. 
48

that the sanctions were most frequently 

related to problems in the areas of compliance 

with federal or state regulations, financial 

record keeping and viability and special 

education services. 

Exhibit 4-7 provides the percentage of 

authorizers reporting problems that resulted in 

formal sanctions. Authorizers indicated that 

nonrenewal and revocation were linked to 

other areas besides academic performance 

problems, including problems relating to 

compliance with state and federal 

regulations66 and financial problems. 

Only 38 percent of authorizers reported 

implementing formal sanctions when a school 

was not meeting progress toward academic 

goals and 29 percent sanctioned their schools 

because of lack of progress toward specific 

goals in the charter. These findings support 

the argument by Hill et al. (2001) that lack of 

capacity leads authorizers to focus on 

bureaucratic measures, such as whether or not 

a school is fiscally solvent. 

Bulkley (2001) identified strategies 

authorizers use that fall short of closing a 

school. Several data collection efforts for the 

present study also identified informal 

sanctions: the authorizer survey, focus groups 

with authorizers, and site visits. Authorizers 

reported reluctance to implement formal 

sanctions without first trying to help the 

schools improve. Informal sanctions are more 

common than formal sanctions, with 

42 percent of authorizers implementing some 

type of informal sanction. Charter school 

directors reported that 12 percent of charter 

schools received written notification about 

problems, 9 percent were required to develop 

improvement plans, and 3 percent were 

placed 

 

66 This finding may seem counterintuitive given the 

theory of charter schools. However, as discussed in 

Chapter 3, many charter schools do not receive blanket 

waivers from regulations governing all public schools. 

on probationary status. Written notifications 

doubled over a two-year period, from 

6 percent in 2000-01 to 12 percent in 

2001-02. 

As Exhibit 4-7 illustrates, informal sanctions 

were imposed for the same reasons as formal 

sanctions: problems with financial record 

keeping and viability (64 percent) and with 

compliance with federal or state regulations 

(61 percent). Informal sanctions were more 

likely than formal sanctions to be linked to 

problems around school management, 

leadership or governance but were less likely 

to be associated with problems of academic 

progress, such as growth in student 

performance. When charter schools were not 

meeting the expectations of authorizers, 

58 percent of authorizers reported providing 

extra assistance to schools, whether or not 

they were placed under informal sanctions. 

The preceding discussion indicates that the 

application for a charter, monitoring and 

implementation of sanctions are all part of the 

charter school accountability process but that 

variation exists in the procedures used by 

authorizers and the degree of oversight 

exerted during all three stages. Complicating 

the accountability relationship between 

authorizers and schools is the fact that charter 

schools are accountable to multiple groups. 

However, the theory of charter school 

accountability may be even more affected by 

political and technical barriers faced by 

authorizers and by a shift toward increased 

accountability at the school-leveltwo 

important areas discussed in the following 

pages. 
49

Exhibit 4-7 

Reasons Given by Authorizers for Formal and Informal Sanctions67, 2001-02 

0 10 20 30 40 50 60 70 80 90

Compliance with regulations

Finances

Special education services

Health and safety issues

Staff performance or attendance

School management or governance

Progress toward academic goals 

Enrollment numbers

Achievement comparisons (charter v. other public schools)

Curricular or instructional strategies

Progress toward goals in charter

Growth in student performance

Student performance 

Parent or community involvement

Parent satisfaction

Progress toward nonacademic goals 

Relationship with EMO

School accreditation status

Staff satisfaction

Student diversity

Student admission or selection

Informal Sanctions (n=51)

Formal Sanctions (n=14)

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, 77 percent of the authorizers that reported implementing formal sanctions did so because of inadequacies 

in the area of compliance with federal or state regulations. Sixty-one percent of the authorizers that reported implementing informal

sanctions did so because of inadequacies in this area. 

 

67 “Formal” sanctions are identified in charter school laws and include (1) nonrenewal of the charter or (2) 

revocation before the term of the charter is completed. “Informal” sanctions, which are not mandated by law but 

have developed through the charter school implementation process, include providing written notification to schools 

about problems, requiring schools to develop improvement plans, and placing schools on probationary status. 
50

Challenges to Charter School 

Accountability 

Finding: Authorizers struggle in their 

new roles as the agents that hold charter 

schools accountable because of 

inadequate financial resources, difficulty 

closing schools, and limited staff 

capacity. Furthermore, the current 

accountability policy context calls into 

question the assumption that authorizers 

will hold charter schools accountable for 

individualized goals. 

While moving forward with the new system 

of holding schools accountable for the 

outcomes in their charters, many authorizers 

experience challenges that limit their ability 

to oversee schools. Past research has 

identified impediments relating to the lack of 

resources and capacity of authorizers (Hassel 

and Herdman, 2000; Vergari, 2001), the 

technical complexity of creating new 

accountability systems (Hassel and Herdman, 

2000; Vergari, 2001), the political 

environment (Hassel and Herdman, 2000; 

Vergari, 2001), and the delicate balance 

between giving schools flexibility and holding 

them accountable (Vergari, 2001). 

Exhibit 4-8 

Challenges Reported by Authorizers and Difficulty Level, 2001-02 

Challenges to Holding Charter Schools Accountable 

Percentage 

of 

Authorizers 

(n=114) 

Mean Rating of 

Difficulty to 

Overcome 

Challenge 

Inadequate financial resources 56 3.5 

Difficulty in closing a school that was having problems 55 3.3 

Lack of personnel dedicated to charter school work 48 3.5 

Lack of clarity and/or guidance about state charter school law 45 3.2 

Public confusion and/or lack of understanding of charter school 

concept 43 3.1 

Difficulty in measuring charter school progress 41 3.0 

Resistance from traditional public schools 38 3.1 

Lack of clarity and/or guidance about how other state policies 

relate to charter schools 37 3.4 

Competition with traditional public schools for 

funding/resources 37 3.6 

Difficulty determining your authorizing agency’s roles and 

responsibilities in charter schools 34 3.3 

Lack of clarity and/or guidance about how federal laws relate 

to charter schools 32 3.2 

Difficulty in creating a system to hold schools accountable 30 3.1 

Difficulty in setting targets for performance in charter schools 29 3.2 

Union opposition 23 3.1 

Community opposition 18 3.0 

Politicsa 12 3.6 

Note: Mean rating of difficulty is based on a 4-point scale, with “very difficult” equal to 4, “somewhat or 

moderately difficult” equal to 3, “slightly difficult” equal to 2 and “not at all difficult” equal to 1. a

 This category is based on the responses of 82 authorizers; the remainder reported that politics was “not 

applicable.” 

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, 56 percent of authorizers reported that “inadequate financial resources” was a challenge 

for them in fulfilling their duties. On a difficulty scale of 1 to 4, authorizers rated this challenge a 3.5. 
51

Exhibit 4-8 lists a wide range of challenges 

faced by authorizers, the most frequently 

reported being inadequate financial resources 

(56 percent) and difficulty closing a school 

that is having problems (55 percent). In 

addition, nearly half of the authorizers cited a 

lack of personnel as an impediment. As 

discussed previously, authorizers had three 

FTE staff, on average, dedicated to charter 

school work. In many cases, this number is 

insufficient for effective oversight of the 

schools they sponsor. Similar percentages of 

authorizers in 2001-02 and 2000-01 reported 

difficulty setting targets, creating 

accountability systems, and measuring 

progress. 

It is interesting to compare the prevalence of 

certain obstacles with authorizers’ perceived 

difficulty in overcoming them. As Exhibit 4-8 

illustrates, charter school authorizers cited 

inadequate financial resources as a common 

challenge and ranked it as one of the most 

difficult to overcome. Some barriers (e.g., 

politics and competition with traditional 

public schools) were not problems for as 

many authorizers but ranked high as difficult 

challenges to overcome. 

The data also indicate a statistically 

significant difference in the challenges faced 

by type of authorizers. For example, state 

authorizers reported that politics and 

resistance from traditional public schools 

were a bigger challenge than they were for 

local or university authorizers (see 

Appendix D-12 for more details). 

Beyond challenges faced by authorizers, the 

larger push for accountability at the state and 

federal levels also affects charter school 

accountability to some degree. Since the first 

charter schools began operating in the early 

1990s, state accountability systems have 

become more comprehensive and continue to 

be refined in response to requirements in 

federal legislation, most recently the 

Elementary and Secondary Education Act of 

1965 (ESEA), as amended by the No Child 

Left Behind Act of 2001 (NCLB). As of 

2001-02, more than three-fourths of state 

coordinators reported that their accountability 

systems included reporting on student 

demographics, alignment of curriculum to 

state standards, reporting of enrollment 

numbers, reporting on other student 

performance indicators, and reporting on 

student achievement results on statewide 

assessments.68 The different requirements 

included in statewide accountability systems, 

many of which are considered school 

“inputs,” are included in Exhibit 4-9. Few 

differences exist between those for traditional 

public schools and the requirements for 

charter schools. The largest differences were 

in the areas of teacher demographics 

(89 percent of states required charters to 

report, compared with 100 percent of states 

requiring traditional public schools to report) 

and waiting lists (91 percent of states required 

charters to report, compared with 27 percent 

of states requiring traditional public schools 

to report). 

These data suggest that state accountability 

systems that developed after the emergence of 

most state charter school laws have subsumed 

charter school accountability. As Anderson 

and Finnigan (2001) argue, the “traditional” 

accountability system, focusing on inputs, is 

embedded in the rules and regulations 

governing charter schools within the current 

public school system. In the current context, 

state accountability systems have challenged 

the principle that charter schools will be held 

accountable only for particular outcomes 

listed in their charters because charter schools 

are now held to the same requirements as 

other public schools in addition to measurable 

goals in the charter document. 

 

68 Note that then features of state accountability 

systems reported on here include both inputs and 

outputs (as results). 
52

Interestingly, the federal accountability 

provisions of NCLB specify converting 

persistently low-performing schools to charter 

schools as one option for restructuring them. 

Hence, charter schools are both subject to 

state and federal accountability requirements 

and possible outgrowths of these efforts. As 

of 2001-02, only 4 percent of authorizers and 

23 percent of states used this new spin on 

chartering as a reform strategy. This is an 

aspect of the development of the charter 

school movement that should continue to be 

tracked in the future. 

Charter school accountability in practice is 

only beginning to be understood. Beyond the 

technical aspects of who is monitoring charter 

schools and for what, charter school 

accountability in general appears limited by 

authorizer capacity issues, from personnel and 

resources to measuring progress. At present, 

few steps are being taken to improve this 

situation beyond an increased effort for 

authorizers to connect through the National 

Association of Charter School Authorizers 

(NACSA) which is supported by two ED 

“national leadership” grants. Assistance by 

states to the nearly one-third of authorizers 

that are having difficulty developing systems 

to hold schools accountable and setting 

performance targets for their schools would 

greatly improve the capacity of authorizers. 

Beyond these capacity issues, charter school 

accountability may be distorted by larger 

accountability contexts that did not exist 

when many of the charter laws were written. 

Future studies must examine how charter 

school accountability fits within state and 

federal accountability requirements. 

Exhibit 4-9 

State Reporting Requirements for Charter Schools and 

Traditional Public Schools, 2001-02 

Percentage of States

Reporting Requirements 

Required for 

Charter 

Schools 

Required for 

Traditional 

Public Schools 

Reporting student achievement results on required 

statewide assessments (n=35) 100 97 

Reporting on other student performance indicators, 

e.g., attendance rates (n=34) 97 97 

Reporting on enrollment numbers (n=32) 100 94 

Aligning of curriculum to state standards (n=31) 90 94 

Reporting on student demographics (n=31) 94 100 

Reporting on teacher qualifications (n=26) 100 96 

Reporting on teacher demographics (n=19) 89 100 

Reporting on school waiting list (n=11) 91 27 

Note: The number of respondents varies by accountability requirement because some states reported that these 

requirements were “not applicable” in their states. 

Note: The actual survey used the term “accountability requirements” to encompass both inputs and outputs. To 

avoid confusion with the current narrower definition of accountability in NCLB, we have used the term “reporting 

requirements” in this exhibit and accompanying text. 

Source: SRI 2001-02 state coordinator survey. 

Exhibit reads: All states (n=35) required that schools report student achievement results as part of their state 

accountability system. Of these, all required this for charter schools and 97 percent required this for traditional 

public schools.
Chapter 5 

Charter Schools and State Performance Standards 

The Elementary and Secondary Education Act 

of 1965 (ESEA), as amended by the No Child 

Left Behind Act of 2001 (NCLB) calls for 

states to hold all public schools, including 

charter schools, to the same standards of 

academic performance. Scant research 

currently exists on charter school 

performance, however, with the few existing 

studies relying on different methodologies 

and providing mixed results.69 Further, the 

unit of analysis varies from study to study. A 

handful of studies examine student level data 

in a single state over time (or “longitudinally”). For example, Solmon, Paark and 

Garcia (2001) studied the effects of charter 

school attendance on students in Arizona, and 

Gronberg and Jansen (2001) examined 

student performance in Texas charter schools. 

These studies suggest that at-risk students 

enrolled in charter schools for more than one 

or two years outperform students in 

traditional public schools. 

On the other hand, a number of studies 

examine performance within and across states 

using school-level data (See, for example, 

Greene, Forster, and Winters (2003) and 

Loveless (2002)). Beyond substantive reasons 

for using school-level data, these data are 

publicly available. Generally, these schoollevel studies (including the work reported on 

here) are exploratory and seek to examine the 

broader influence of charter schools on 

collective student performance. Unfortunately, these studies are inconsistent in the 

degree to which they acknowledge the 

 

69 Even within a single state and drawing on the same 

state databases, analyses have come to different 

conclusions. See, for example, three recent studies of 

student performance in California charter schools 

(Slovacek, Kunnan and Kim, 2002; Greene, Forster 

and Winters, 2003; Zimmer et al, 2003) in which the 

research groups used different analytic techniques on 

the same data to obtain differing results. 

problem of missing charter school data.70 Our 

own explorations suggest that missing data on 

charter schools in many state databases is a 

large problem, limiting the possible analyses 

to a small number of states. 

This chapter describes an analysis of the 

extent to which charter schools and traditional 

public schools met state performance 

standards in 2001-02 (the year preceding 

implementation of NCLB). 

Findings: More than half of the charter 

schools in Texas, Colorado, Illinois, 

Massachusetts, and North Carolina were 

meeting state performance standards. 

However, charter schools were less likely to 

meet performance standards compared with 

traditional public schools.

This evaluation does not examine the effect 

of charter schools on student learning. The 

study originally planned to use student-level 

data for this study component, but in 2001-02, 

current policy interpretations of the Family 

Educational Rights and Privacy Act (FERPA) 

precluded contractors from acquiring student 

data from states. Therefore, this analysis 

derives from school level cross-sectional data 

(i.e., data taken at one point in time). 

Analyses comparing standards-based 

performance levels of charter schools and 

traditional public schools have limitations but 

can inform and serve as the foundation for 

more sophisticated designs and analyses. 

 

70 In fact, reasons for missing data range from privacy 

issues in cases where grade level cohorts are small to 

delays in entering new charter schools into the state 

data system. 
54

Other researchers have conducted related but 

not identical analyses of student performance 

at the school-level. Loveless’s (2002) study, 

which compared the performance of charter 

schools and traditional public schools on state 

tests in 10 states (including Colorado, 

Massachusetts, and Texas), found that 

nationally charter schools scored significantly 

lower than other public schools. In contrast, 

Greene, Forster, and Winters (2003) 

compared charter schools serving general

populations with traditional public schools in 

the surrounding community in 11 states 

(including Colorado and Texas) and found 

that nationally, charter schools outperform

traditional public schools. An important 

difference between the study reported here 

and Greene’s study is that Greene et al. 

excluded charter schools targeting particular 

populations as well as some conversion 

charter schools, while this study included all 

charter schools. 

Thus, the finding in this chapter that charter 

schools were less likely than traditional public 

schools to meet a state’s performance 

standard joins an array of other studies with 

different analytic approaches and disparate 

findings about the success of charter schools. 

These data suggest that some charter schools 

may have difficulty meeting the high-stakes 

performance standards recently adopted by 

the states under NCLB. 

The remainder of this chapter provides 

additional detail about the specific analyses 

the evaluation undertook and discusses the 

results one state at a time. Because individual 

states have distinctive performance standards, 

it is nearly impossible to compare 

performance across states or to generalize 

these findings to other states. 

State-by-State Analysis of Charter 

Schools and Performance Standards 

The analysis of the rates at which charter 

schools met state performance standards 

included five states with adequate data to 

conduct the analyses in this evaluation and 

include Colorado, Illinois, Massachusetts, 

North Carolina, and Texas. These states act as 

case studies and are not representative of the 

charter school universe. 

Specifically, of the 36 states with operating 

charter schools in 2001-02, the five selected 

for case studies met three criteria. They had: 

(1) a school-level performance standard 

designated by the state that included all 

schools and, therefore, would permit 

meaningful comparisons of charter and 

traditional public schools; (2) adequate 

numbers of charter schools; and (3) adequate 

data for charter schools. (See Appendix E-1 

for additional details regarding the states 

excluded from this analysis.) State 

departments of education provided all the data 

used for these analyses, either directly or from 

public files available through their Web sites. 

The analysis sought to determine the extent to 

which charter schools and traditional public 

schools met state performance standards. For 

this evaluation, the “performance standard” 

for each of the five states was defined by the 

state as the school-level standard or 

benchmark that public schools in a state were 

expected to meet in 2001-02. State 

performance standards frequently included 

composite scores based on tests in multiple 

subject areas and at multiple grades. In some 

cases, other measures such as dropout rates 

were included in composite scores. The 

analysis is based on states’ designations of 

low performance (or other language, such as 

“unsatisfactory” progress). Details regarding 

each state’s accountability system and 

performance standard, as well as student 
55

demographics, are located in Appendices E-2 

through E-6. Exhibit 5-1 summarizes results 

of this step of the analysis. 

Overall, when not controlling for any 

background characteristics, more than onehalf of charter schools in each state included 

in this analysis met state performance 

standards in 2001-02the proportion of 

charter schools meeting state performance 

standards was even higher in Colorado 

(90 percent) and North Carolina (88 percent). 

However, charter schools met state 

performance standards at lower rates than 

traditional public schools. 

Another step in the analysis was a comparison 

of the performance of charter schools and 

traditional public schools, controlling for 

certain background variables, one at a time. 

The analysis employed control variables to 

statistically equate groups of charter schools 

and groups of traditional public schools. The 

relationship between meeting state 

performance standards and school type 

(charter or traditional public schools) was 

examined comparing similar populations 

along each of the following control variables: 

percent of the student body that is lowincome, percent of the student body that is 

minority (defined as students from any 

nonwhite racial or ethnic category), student 

mobility (student movement in and out of the 

school), and student enrollment (number of 

students).71 These data were included because 

of their availability in state-level databases 

and their possible associations with student 

performance, as established by many research 

studies over the years.72 

Finally, logistic regression was conducted in 

the two states with the largest numbers of 

charter schools and the most complete data 

(Texas and Colorado). This analysis 

examined the relationship between school 

 

71 Please note that the exact definition of these 

variables varies by state. In addition, not all variables 

were available in every state. For more details see 

Appendices E-2 through E-6. 

72 The following are examples of recent research 

linking student performance with student income and 

race (Denton and West, 2002; Donahue, Voelkl, 

Campbell and Mazzeo, 1999; Donahue, Finnegan, 

Lutkus, Allen and Campbell, 2001; National Center for 

Education Statistics, 2002); student mobility (Bryk, 

Thum, Easton and Luppescu, 1998); and school size 

(Cotton, 1996; The Rural School and Community 

Trust, 2002; Wasley, Fine, King, Powell, Holland, 

Gladden and Mosak, 2000). 

Exhibit 5-1 

Performance of Charter Schools and Traditional Public Schools, 2001-02 

Number and Percentage of Schools Meeting State Performance Standards 

Charter Schools Traditional Public Schools 

State

Number Meeting 

Standard 

Percentage 

Meeting Standard 

Number Meeting 

Standard 

Percentage 

Meeting Standard 

Texas*** 78 66 6,308 98 

Colorado*** 76 90 1,421 98 

Illinois*** 12 52 3,191 82 

Massachusetts*** 18 64 1,335 87 

North Carolina*** 75 88 2,022 100 

***p<.01 (Indicates significant association between charter schools and traditional public schools in the 

percentage meeting the state standard). 

Note: For all significant differences, the proportion of charter schools meeting the standard was lower than that of 

traditional public schools. No background variables were controlled for in these analyses. 

Exhibit reads: In Texas, 78 charter schools (66 percent) met the state performance standard in 2001-02 compared 

with 6,308 traditional public schools (98 percent). These numbers indicate that a significant association exists 

between performance and school type (i.e., charter schools or traditional public schools) in Texas (with a lower 

proportion of charter schools meeting the standard). 
56

type and performance while controlling for all 

of the available background variables 

simultaneously. 

The logistic regression results for both Texas 

and Colorado indicate that after controlling 

for all the background variables,73 type of 

school was associated with school 

performance. Specifically, being a charter 

school had a statistically significant positive 

association (p<.01) with not meeting the state 

performance standard. For more details on the 

logistic regression results in both Texas and 

Colorado, see Appendix E-2 and 

Appendix E-3. 

In addition to these logistic regression results, 

the following paragraphs provide highlights 

from each state regarding the variable by 

variable analyses (Appendices E-2 through 

E-6 provide additional details.) 

In each of the five states examined, charter 

schools were less likely to meet performance 

standards than traditional public schools. As 

discussed below, most of the differences 

occurred when charter schools and traditional 

public schools had above-average proportions 

of students eligible for free and reduced-price 

lunches and above average proportions of 

minority students. 

Texas: Traditional public schools in Texas 

met the state performance standard at higher 

rates than did charter schools. This finding 

did not change after controlling individually 

for the proportion of low-income students, the 

proportion of minority students, student 

mobility, and student enrollment.74 

 

73 The Texas analyses controlled for low-income, 

minority, student mobility, and enrollment, while the 

Colorado analyses controlled for low-income, minority, 

and enrollment (mobility data were not available in 

Colorado). 

74 A provision of the Texas charter school law deserves 

special mention. During the first six years of the 

charter school law in Texas (1995-2001), a large 

number of charter schools were established that 

Colorado: Compared to charter schools, 

traditional public schools in Colorado met the 

state performance standard at higher rates. 

This finding changed when steps were taken 

to control for background variables. Charter 

schools and traditional public schools met the 

state standard in Colorado at similar rates 

when they had lower than average proportions 

of low-income and minority students. The two 

types of schools also performed at similar 

rates when they had above average student 

enrollment numbers. 

Illinois: Charter schools in Illinois were less 

likely than traditional public schools to meet 

the state performance standard. When 

controlling for student mobility and 

enrollment, the differences in performance 

levels between charter and traditional public 

schools persisted. However, controlling for 

the proportion of minority students revealed 

that charter schools and traditional public 

schools met the performance standard at equal 

rates. In addition, charter schools and 

traditional public schools met the standard in 

equal rates when schools had lower than 

average proportions of low-income students. 

 

targeted at-risk students. (Note: Texas has 13 

categories of “students at-risk of dropping out” relating 

to a wide range of issues from low achievement to 

expulsion, pregnancy, and homelessness.) This 

provision was eliminated in 2001. However, as of 

2000-01 approximately one-third of charter schools in 

Texas continued to serve a large percentage of at-risk 

students. (For more information about charter schools 

serving at-risk populations in Texas, see 

http://www.tasb.org/advocacy/reports/2002/sept02/ 

charter_schools.html and http://www.tcer.org/tcer/ 

schools/yr6_executive_summary.pdf.) The other four 

states included in these analyses did not have this type 

of provision, but the Colorado, North Carolina, and 

Illinois charter school laws all include an emphasis on 

expanding the learning opportunities for “at-risk 

pupils.” 
57

Massachusetts: In Massachusetts, a higher 

proportion of traditional public schools than 

charter schools met the state performance 

standard. However, when both types of 

schools served lower than average proportions 

of low-income and minority students and had 

lower than average student enrollments, they 

met the state performance standard at similar 

rates. 

North Carolina: Charter schools in North 

Carolina were less likely to meet the state 

performance standard than were traditional 

public schools. After controlling for the 

proportion of minority students and 

enrollment, this finding did not change. 

Charter schools were less likely to meet state 

performance standards. 

In summary, charter schools met state 

performance standards at lower rates than 

traditional public schools in the five states. 

The few exceptions to this finding may raise 

more questions than they answer. For 

example, why do charter schools and 

traditional public schools in Colorado perform 

at similar rates when they have above average 

student enrollment numbers? 

These limited findings point to the importance 

of additional analyses of charter school 

performance because they do not illuminate 

the reasons that charter schools in the case 

study states were less likely to meet state 

performance standards. Additional studies 

could examine the extent to which charter 

schools that do not meet the state performance 

standards serve students with low prior 

achievement, for example, disabling 

conditions, or other characteristics that may 

hinder their achievement and the extent to 

which certain types of charter schools are 

more successful in meeting state performance 

standards. 

These findings do not indicate that charter 

schools were less effective than traditional 

public schools but suggest that many charter 

schools will have difficulty meeting the 

standards established by states under NCLB. 

With the passage of NCLB, all schools must 

meet school-level standards or face 

interventions in the current accountability 

policy context. Future studies should 

examine the extent to which charter schools 

exhibit growth in student performance in 

addition to whether they meet absolute 

standards. 

These data provide a single snapshot of 

school-level performance using state 

performance standards as a measure and 

controlling for some basic characteristics of 

students. The findings are important to (and 

should inform) the policy debate around the 

implementation of NCLB. However, many 

other variables affect student achievement. 

Therefore, future studies must examine 

charter school performance in more depth. 

Chapter 6 

Directions for Future Studies 

The key findings from this study illuminate 

many dimensions of the charter school 

movement but also indicate some directions 

in which to seek further knowledge about the 

ongoing development and impact of this 

education reform. Outlined below are a 

number of important questions for future 

study. 

• What is the impact of charter school 

attendance on the achievement of 

students? The current study compares 

how students at charter schools and 

traditional public schools perform on 

state standards and discusses the 

limitations of such analyses. In 

particular, the comparison could not 

fully take into account differences 

between the charter schools and 

traditional public schools in the 

students they serve. This analytic 

weakness is also a concern, to varying 

extents, in a growing body of research 

on the effects of charter schools in a 

number of states (see, for example, 

Horn and Miron 2000; Solmon et al. 

2001; Gronberg and Jansen 2001; 

Hanushek 2002; Loveless 2003; 

Zimmer et al. 2003; Bifulco and Ladd 

2003). Randomized controlled trials 

are one way to measure the impacts on 

student achievement, using a lottery of 

charter school applicants to create 

identical groups of charter school 

attendees and nonattendees and 

comparing their outcomes. ED is 

launching the first such study, 

focusing on the relationship between 

charter school impacts and the policy 

environments in which the schools 

operate. Additional randomized 

studies of charter schools or particular 

charter school strategies would 

provide stronger evidence about the 

effectiveness of this education reform 

approach. Depending on the actual 

evaluation questions posed and the 

availability of student-level data, other 

methods such as value-added or time 

series analyses could also be 

considered. 

• How is the demographic profile of 

charter school students evolving?

The demographic characteristics of 

charter school students have been a 

topic of great interest to observers and 

the current study provides some 

definition of the student population 

served by charter schools across the 

United States. How will this profile 

continue to evolve over time, as 

contrasted with that of traditional 

public schools and how will the 

profile vary by type of school 

(distance learning, classroom-based 

and so forth)? 

• Do features of charter schools 

attract African American teachers?

The significance of teacher race and 

the match between student race and 

teacher race are not well-developed 

areas of study. One area for further 

pursuit is to examine the factors 

associated with the relatively higher 

proportions of African American 

teachers in charter schools compared 

with traditional public schools. In 

addition, future studies should 

consider in their design the 

relationship between teacher race and 

student outcomes. 
60

• How does charter school 

accountability fit within state and 

federal accountability 

requirements? For better or worse, 

state accountability requirements and 

systems have challenged the principle 

that charter schools will be held 

accountable only for particular 

outcomes listed in their charters 

because charter schools are now held 

to the same requirements as other 

public schools as well as to the 

measurable goals in a charter. 

Accountability provisions in the 

implementation of the Elementary and 

Secondary Education Act of 1965 

(ESEA), as amended by the No Child 

Left Behind Act of 2001 (NCLB), 

seem likely to reinforce this 

development. It will be important to 

monitor the continuing impact of these 

state and federal requirements, on 

charter schools, in particular upon 

their autonomy and accountability 

processes. Furthermore, future studies 

should examine teacher qualifications 

in charter schools and traditional 

public schools given the requirements 

for “highly qualified teachers” in 

NCLB. 

• Are charter schools replacing 

poorly performing traditional 

public schools? NCLB’s 

accountability provisions specify 

converting persistently lowperforming schools to charter schools 

as one option for restructuring. As of 

2001-02, 23 percent of states and 

4 percent of authorizers reported 

creating charter schools for this 

purpose. This aspect of the charter 

school movement should be tracked 

and future studies should examine the 

use of charter schools as a solution to 

low performance. 

This report has provided information from 

multiple data sources on a range of topics: the 

role of the PCSP in encouraging the 

development of charter schools; the 

distinctive characteristics of students and 

teachers in charter schools; the limited 

autonomy of many charter schools; and the 

similarities between accountability 

requirements of charter schools and 

traditional public schools. In addition, the 

report contributes to the knowledge base 

about charter schools and student 

achievement, finding that many charter 

schools have difficulty meeting state 

performance standards. However, while the 

report answers some questions about the 

PCSP and about the charter school sector, it 

also raises new questions. At this juncture in 

the history of charter schools, additional 

research should be conducted to examine the 

impact of charter school attendance on 

achievement, the demographic shifts in 

charter school enrollments, the impact of 

NCLB on the flexibility provided to charter 

schools, and the use of “chartering” to 

improve school performance. While the 

knowledge base is growing about charter 

schools, more research is needed to illuminate 

the possibilities and limitations of this 

educational reform. 
References 

Anderson, L., and Finnigan, K. (2001). Charter school authorizers and charter school 

accountability. Paper presented at the annual meeting of the American Educational Research 

Association, Seattle, Wash. 

Anderson, L., Adelman, N., Finnigan, K., Cotton, L., Donnelly, M., and Price, T. (2002, 

November). A decade of public charter schools. Menlo Park, Calif.: SRI International. 

Available at http://www.sri.com/policy/cep/choice/yr2.pdf. 

Berman, P., Nelson, B., Ericson, J., Perry, R., and Silverman, D. (1998). A national study of 

charter schools: Second-year report. Washington, D.C.: U.S. Department of Education, 

Office of Educational Research and Improvement. Available at: 

http://www.ed.gov/pubs/charter98/index.html. 

Berman, P., Nelson, B., Perry, R., Silverman, D., Solomon, D., and Kamprath, N. (1999). The 

state of charter schools: Third-year report. Washington, D.C.: U.S. Department of 

Education, Office of Educational Research and Improvement. Available at: 

http://www.ed.gov/pubs/charter3rdyear/index.html. 

Bifulco, R. and Ladd, H. F. (2003, November). The impacts of charter schools on students 

achievement: Evidence from North Carolina. Paper presented at the Association of Public 

Policy Analysis and Management Annual Meeting, Washington, D.C. 

Bryk, A., Thum, Y., Easton, J., and Luppescu, S. (1998). Academic productivity of Chicago 

public elementary schools. Chicago: Consortium on Chicago School Research. 

Buddin, R. and Zimmer, R. (2003). A closer look at charter school student achievement.

Working paper. Santa Monica, Calif.: RAND Education. 

Bulkley, K. (1999). Charter school authorizers: A new governance mechanism? Educational 

Policy, 13, 674-697. 

Bulkley, K. (2001). Educational performance and charter school authorizers: The accountability 

bind. Educational Policy Analysis Archives, 9(37). Retrieved February 10, 2003, from 

http://epaa.asu.edu/epaa/v9n37.html. 

Cotton, K. (1996). School size, school climate and student performance. (School Improvement 

Research Series, Close-up #20). Portland, Ore.: Northwest Regional Educational Laboratory. 

Denton, K., and West, J. (2002). Children's reading and mathematics achievement in 

kindergarten and first grade. Washington, D.C.: U.S. Department of Education, Office of 

Educational Research and Improvement. Available at: 

http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2002125. 

Donahue, P. I., Finnegan, R. J., Lutkus, A. D., Allen, N. L., Campbell, J.R. (2001). The nation's 

report card: Fourth-grade reading 2000. Available at: 

http://nces.ed.gov/nationsreportcard/pubs/main2000/2001499.aspl. 

Donahue, P. I., Voelkl, K. E., Campbell, J. R., and Mazzeo, J. (1999). The NAEP 1998 reading 

report card for the nation and the states (NCES 1999-500). Washington, D.C.: U.S. 

Department of Education, Office of Education Research and Improvement. 

Finn, C., Manno, B., and Vanourek, G. (2000). Charter schools in action. Princeton, N.J.: 

Princeton University Press. 
62

Franklin, B., and Glascock, C. (1996, October). The relationship between grade configuration 

and student performance in rural schools. (ERIC document ED 403083). Paper presented at 

the Annual Conference of the National Rural Education Association, San Antonio, Texas. 

Garcia, George F., and Garcia, Mary (1996, November). Charter schools—Another top-down 

innovation. Educational Researcher, 25, 34-36. 

Geske, T. G., Davis, D., and Hingle, P. (1997). Charter schools: A viable public school choice 

option? Economics of Education Review, 16, 15-23. 

Greene, J., Forster, G., and Winters, M. (2003). Apples to apples: An evaluation of charter 

schools serving general student populations. (Education Working Paper 1). New York, N.Y.: 

Manhattan Institute for Policy Research. 

Gronberg, T., and Jansen, D. (2001). Navigating newly chartered waters: An analysis of Texas 

charter school performance. Austin, Texas: Texas Public Policy Foundation. 

Hanushek, E., Kain, J. F. and Rivkin, S. G. (2002). The impact of charter schools on academic 

achievement. Stanford, Calif.: Program of Research on Educational Outcomes, Stanford 

University. Retrieved April 29, 2004, from http: 

edpro.stanford.edu/eah/papers/charters.aea.jan03.pdf. 

Hassel, B., and Herdman, P. (2000, April). Charter school accountability: A guide to issues and 

options for charter authorizers. Charlotte, N.C.: Public Impact. 

Hill, P., Lake, R., and Celio, M. B. (2002). Charter schools and accountability in public 

education. Washington, D.C.: The Brookings Institution. 

Hill, P., Lake, R., Celio, M. B., Campbell, C., Herdman, P., and Bulkley, K. (2001). A study of 

charter school accountability. Washington, D.C.: U.S. Department of Education, Office of 

Educational Research and Improvement. 

Horn, J. and Miron, G. (2000). An evaluation of the Michigan charter school initiative: 

Performance, accountability, and impact. Kalamazoo, Mich.: The Evaluation Center, 

Western Michigan University. 

Lopez, A., Wells, A. S., and Holme, J. J. (2002). Creating charter school communities: Identity 

building, diversity and selectivity. In Wells, A. S. (ed.) Where charter school policy fails: 

The problems of accountability and equity (pp. 129-158). New York: Teachers College Press. 

Loveless, T. (2002). The 2002 Brown Center report on American education: How well are 

American students learning? Washington D.C.: The Brookings Institution. 

Loveless, T. (2003). The Brown Center Report on American Education. How well are American 

students learning? Washington, D.C.: The Brookings Institution. 

Miller, M., Rollefson, M., Garet, M., Berends, M., Adelman, N., Anderson, L., and Yamashiro, 

K. (2000, May). Summary and implications from a conference on student achievement and 

the evaluation of federal education programs: A working paper. Washington, D.C.: 

American Institutes for Research. 

Miron, G., and Nelson, C. (2002). What’s public about charter schools? Lessons learned about 

choice and accountability. Thousand Oaks, Calif.: Corwin Press, Inc. 

Nathan, J. (1996). Charter schools: Creating hope and opportunity for American education. San 

Francisco: Jossey-Bass Publishers. 

National Center for Education Statistics. (2002). Common Core of Data (CCD) quick facts. 
63

Retrieved May 1, 2002, from http://nces.ed.gov/ccd/quickfacts.html. 

Nelson, B., Berman, P., Ericson, J., Kamprath, N., Perry, R., Silverman, D., and Solomon, D. 

(2000). The state of charter schools 2000: Fourth-year report. Washington, D.C.: U.S. 

Department of Education, Office of Educational Research and Improvement. Retrieved on 

January 26, 2001, from http://ed.gov/PDFDocs/4yrrpt.pdf. 

Palmer, L. B., and Gau, R. (2003, June). Charter school authorizing: Are states making the 

grade? Washington, D.C.: Thomas B. Fordham Institute. Retrieved on August 15, 2003, 

from http://www.edexcellence.net/tbfinstitute/authorizers.html. 

Renchler, R. (2002). School organization: Grade span Trends and issues (ERIC document ED 

472994). Washington, D.C.: Office of Educational Research and Improvement. 

RPP International and University of Minnesota. (1997). A study of charter schools: First-year 

report. Washington, D.C.: U.S. Department of Education, Office of Educational Research 

and Improvement. Retrieved on October 10, 1998, from 

http://www.ed.gov/pubs/charter/index.html. 

The Rural School and Community Trust. (2002). Small works in Arkansas: How poverty and the 

size of schools and school districts affect schools performance in Arkansas. Washington, 

D.C.: Author. 

Slovacek, S., Kunnan, A., and Kim, H. (2002). California charter schools serving low-SES 

students: An analysis of the academic performance index. Paper presented at the 2002 

California Network of Educational Charters (CANEC) Conference, San Diego, Calif. 

Solmon, L., Paark, K., and Garcia, D. (2001). Does charter school attendance improve test 

scores? Phoenix, Ariz.: Goldwater Institute. 

U.S. Department of Education. (1997, July). Education Department General Administrative 

Regulations (EDGAR), Appendix A. Federal Register. Washington, D.C.: Retrieved on 

April 16, 2003, from http://www.ed.gov/offices/OCFO/grants/edgar/edgar_apxa.html. 

U.S. Department of Education. (2000). Evaluation of the public charter schools program: Year 

one evaluation report. Washington, D.C.: Office of the Under Secretary, Planning and 

Evaluation Service, Elementary and Secondary Program Division. Retrieved on March 17, 

2003, from http://www.ed.gov/offices/OUS/PES/chartschools/index.html. 

Vergari, S. (2001). Charter school authorizers: Public agents for holding charter schools 

accountable. Education and Urban Society, 33, 129-140. 

Wasley, P., Fine, M., King, S., Powell, L., Holland, N., Gladden, R., and Mosak, E. (2000). 

Small schools, great strides. New York, New York: Bank Street College of Education. 

Wells, A. S. (1998). Beyond the rhetoric of charter school reform: A study of ten California 

school districts. Los Angeles, Calif.: UCLA Charter School Study. 

Wells, A. S., Grutzik, C., Carnochan, S., Slayton, J. and Vasudeva, A. (1999, Spring). 

Underlying policy assumptions of charter school reform: The multiple meanings of a 

movement. Teachers College Record, 100, 513-35. 

Zimmer, R., Buddin, R., Chau, D., Daley, G., Gill, B., Guarino, C., Hamilton, L., Krop, C., 

McCaffrey, D., Sandler, M., and Brewer. D. (2003). Charter school operations and 

performance: Evidence from California. Santa Monica, Calif.: RAND Corporation. 

APPENDICES TO THE CHAPTERS 

APPENDIX A: 

CHAPTER 1 

Appendix A-1 

Data Collection and Analysis Procedures for 2001-02 

Data Collection for 2001-02 

Telephone survey of charter schools. A simple random sample of 585 charter schools in operation as of 

September 2001 was surveyed from February to May 2002. A total of 477 schools completed the survey, 

for a response rate of 82 percent.75 The typical respondent was the charter school’s director, principal, or 

an equivalent administrator at the school, and each survey lasted approximately 60 minutes. The survey 

instrument incorporated open- and closed-ended items on charter school characteristics, operations, 

accountability processes, flexibility, and support. SRI International designed the telephone survey, and 

LHK Partners, Inc., a survey firm in Newtown Square, Pa., administered this survey. Members of the SRI 

team trained LHK interviewers on the instrument and the evaluation’s purposes. 

Telephone survey of charter school authorizers. As of the summer of 2001, the universe of charter 

school authorizers that had awarded charters to schools included 643 agencies. A stratified random 

sample of 150 charter school authorizers was drawn from this universe. The sample was stratified into 

three categories: (1) local school board or district, or county board or office of education; (2) state board 

of education, state education agency, or chief state school officer; and (3) university, college, or 

community college.76 State and university authorizers were oversampled to provide robust comparisons 

across authorizer types. The authorizer survey was administered at the same time as the school survey 

(February to May 2002),77 and the interviews were also conducted by LHK Partners, Inc. A total of 118 

authorizers completed the telephone survey, for a response rate of 79 percent. Like the telephone survey 

of charter schools, the survey of charter school authorizers took about 60 minutes to complete. 

Survey questions were designed to document charter school authorizers’ characteristics and 

accountability relationships. The survey also included items on roles in supporting and assisting charter 

schools, along with the types of freedoms authorizers confer on their schools. Information on these topics 

was elicited through a combination of open- and closed-ended items. (2001-02 marked the third year SRI 

surveyed charter school authorizers, making authorizers the only data source included in every round of 

data collection.) 

Telephone survey of state charter school coordinators. The state charter school coordinators (or 

equivalent respondents) from 34 states and the District of Columbia were surveyed by phone in the winter 

and spring of 2002. Three states did not respond to the survey. Of the respondents, 30 states and the 

District of Columbia received PCSP funds and four did not. The survey instrument was a revised version 

of the 1999 instrument. As with the other telephone surveys, the state instrument consisted of open- and 

closed-ended items on charter school-related areas, as well as the state’s PCSP grant. SRI team members 

administered this survey. 

Site visits to charter schools and charter school authorizers. During the 2000-01 and 2001-02 school 

years, site visits were made to 12 charter schools and their authorizers in six states: Arizona, California, 

Massachusetts, Michigan, North Carolina, and Texas. Nine of these schools were visited a second time in 

2001-02. During these visits, interviews with school directors, teachers, parents, representatives from the 

schools’ authorizers, and, if applicable, representatives of EMOs and CBOs were conducted. The schools 

 

75 An analysis of the responses indicates that all states were represented in proportion to their share of the universe 

of charter schools. 

76 The small number of “other” authorizers were not sampled. 

77 Data collection was reopened in November 2002 to obtain a higher response rate from the original sample of 

state-level charter school authorizerseight of the 15 state authorizers were surveyed at that time. 
70

that were visited represented a wide range of educational approaches, from a school based on 

entrepreneurism to another linked to a local museum. The schools varied by grade levels served, total 

student enrollments, student demographics, and student socioeconomic status. A few of the schools had 

multiple campuses. 

Analysis of data from the 1999-2000 Schools and Staffing Survey (SASS). The Schools and Staffing 

Survey (SASS) is a major U.S. Department of Education data set designed to generate comprehensive 

information on a representative sample of schools and, importantly, the universe of charter schools in 

operation as of spring 1999. The SASS data set consists of several survey instruments at the school and 

teacher levels for both charter and traditional public schools. SASS was designed and piloted by the 

National Center for Education Statistics (NCES) and the American Institutes for Research’s Education 

Statistics Services Institute (ESSI). Data collection activities were conducted by the Census Bureau. 

SRI International obtained a license for the 1999-2000 SASS data set, which was released in summer 

2002. SASS data allow comparisons of charter schools and traditional public schools. 

Data Analysis for 2001-02 

This report includes analyses of both qualitative and quantitative data. Charter schools were sampled 

randomly; therefore, no weighting was necessary for this survey. The state charter school coordinator 

survey was administered to the universe of states with charter school laws, hence neither sampling nor 

weighting was required for this survey either. Some types of authorizers were oversampled; therefore, the 

authorizer data were weighted to correct for different sampling ratios across these types.78 These weighted 

percentages and means are used throughout this report, resulting in aggregate data that are generalizable 

to the universe of authorizers. However, for ease of reference, the unweighted number of respondents is 

provided in all tables—that is, a total sample size (or “n”) of 118 (unweighted) is listed in the tables rather 

than a total sample size of 636 (weighted). 

As in earlier reports from this evaluation, certain categories of authorizers were collapsed into larger 

categories for both technical and conceptual reasons. The resulting new types consist of “local 

authorizers,” which combines local education agencies and county boards or offices of education; “state 

authorizers”; and “university authorizers.” Independent or special charter boards were excluded from both 

analyses because of their low prevalence in each sample.79 

The quantitative analysis includes descriptive statistics and cross-tabulations. In addition, a variety of 

significance tests were conducted on both data sets, including analysis of variance, chi-squared tests, and 

Fisher’s exact tests80 for data involving categorical variables; analysis of variance or ANOVA for data 

involving continuous variables; t-tests of mean differences; and z-tests of differences in proportions. 

Although most of the 2001-02 report is based on quantitative data, qualitative data are used to explain 

some of the quantitative findings and provide examples. For the qualitative analysis, each site visit team 

wrote a site visit report after completing the visit, and a full-day debriefing meeting was held after the 

first two rounds of visits. In addition, open-ended survey questions were coded and analyzed by team 

members. 

 

78 The percentages and means for aggregate authorizer data were calculated after the following weights were 

applied: local authorizers (7.04), state authorizers (1.31), and university authorizers (1.59). The other survey data did 

not require sampling because the school survey was a simple random sample and the state survey was administered 

to the universe of states with charter school laws. 

79 Fewer than 2 percent of the schools in the 2001-02 sample were authorized by independent or special chartering 

boards or other authorizers, and those schools were dropped from the analyses of the school data by authorizer type. 

No independent or special chartering boards were included in the 2001-02 authorizer sample. 

80 Fisher’s exact test was used when, because of small cell sizes (n<5), chi-squared was not appropriate. 
71

Appendix A-2 

Age of State Charter School Law and Number of Charter Schools in Operation 

(as of Jan. 1, 2002) 

 Year Law Passed Total Number of Schools in Operation 

Minnesota 1991 68 

California 1992 349 

Colorado 1993 87 

Georgia 1993 38 

Massachusetts 1993 42 

Michigan 1993 188 

New Mexico 1993 20 

Wisconsin 1993 108 

Arizona 1994 287 

Hawaii 1994 6a 

Kansas 1994 28 

Alaska 1995 15 

Arkansas 1995 6 

Delaware 1995 10 

Louisiana 1995 20 

New Hampshire 1995 0 

Rhode Island 1995 3a

Texas 1995 180 

Wyoming 1995 1 

Connecticut 1996 14 

District of Columbia 1996 36 

Florida 1996 185 

Illinois 1996 23 

New Jersey 1996 50 

North Carolina 1996 93 

South Carolina 1996 8 

Mississippi 1997 1 

Nevada 1997 10 

Ohio 1997 92 

Pennsylvania 1997 77 

Idaho 1998 10 

Missouri 1998 21 

New York 1998 24 

Utah 1998 9 

Virginia 1998 6 

Oklahoma 1999 5a

Oregon 1999 17 

Indiana 2001 0 

Total number of states: 38b Total number of schools: 2,137 a

For states that did not complete the 2001-02 state survey, the number of charter schools is based on SRI’s charter school 

database for the 2001-02 school survey. This database included schools in operation as of summer 2001. b

Iowa, Tennessee, and Maryland passed charter school laws after data were collected for this report. Puerto Rico passed a charter 

school law in 1993, but Puerto Rico is not included in this table. 

Source: SRI 2001-02 state charter school coordinator survey. 

Exhibit reads: As of Jan. 1, 2002, Minnesota, which passed its charter school law in 1991, had 68 charter schools operating.
72

Appendix A-3 

Charter School Student Admission Criteria, 2001-02 

Admission Criterion Percentage of Schools (n=477) 

Sibling preference 64 

Application 63 

Residence 48 

Parent and/or student contracts 37 

Personal interviews 33 

Referrals 26 

Child of staff member 22 

Special student needs 20 

Academic records 20 

Recommendations 18 

Child of founder 11 

Standardized achievement tests 8 

Student aptitudes, skills, or talents 7 

Racial/Ethnic or ethnic background 7 

Admission tests 5 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 64 percent of charter schools used sibling preference as an admission 

criterion. 
APPENDIX B: 

CHAPTER 2 

Appendix B-1 

Total Number of States with Charter Laws and Number of States Receiving Public Charter 

Schools Program Funds, by Fiscal Year 

0

5

10

15

20

25

30

35

40

45

FY91 FY92 FY93 FY94 FY95 FY96 FY97 FY98 FY99 FY00 FY01 FY02

Number of States

Total number of states with charter

laws

Number of state education agencies

receiving PCSP funds

Note: This exhibit includes only funds allocated for grants to SEAs. The total PCSP appropriation also includes funds for 

national activities, such as the national charter schools conference, research projects, etc. 

Sources: SRI 1999-2000 and 2001-02 state coordinator surveys, U.S. Department of Education press releases, and 

http://www.uscharterschools.org. 

Exhibit reads: In FY 1991, one state had a charter law, and no states received PCSP funds. 
76

Appendix B-2 

Maximum and Minimum Public Charter Schools Program Grants, by Fiscal Year 

$0

$1,500,000

$3,000,000

$4,500,000

$6,000,000

$7,500,000

$9,000,000

$10,500,000

$12,000,000

$13,500,000

$15,000,000

$16,500,000

$18,000,000

$19,500,000

$21,000,000

$22,500,000

$24,000,000

FY95 FY96 FY97 FY98 FY99 FY00 FY01

PCSP State Grant

Minimum PCSP grant

Maximum PCSP grant

 Sources: SRI 1999-2000 and 2001-02 state coordinator surveys. 

Exhibit reads: In FY 1995, the minimum PCSP state grant was $250,000 and the maximum was $750,000. 
77

Appendix B-3 

Average Charter School Subgrant Amount, by State and Year 

 1998 1999 2000 2001 

Alaska $61,333 $45,232 $34,552 N/A 

Arkansas N/A81 $8,935 $45,114 $216,667 

California N/A $78,131 $102,696 $95,731 

Colorado $39,248 $91,176 $116,698 $140,968 

Connecticut $107,453 N/A $150,447 $162,596 

District of 

Columbia $120,090 $78,692 $90,753 $112,184 

Delaware $57,137 $52,236 $33,200 N/A 

Georgia $79,565 $34,865 $45,573 $148,807 

Idaho N/A $97,399 $136,842 $197,328 

Illinois $99,828 $85,407 $52,659 $65,304 

Kansas $25,000 $98,188 $16,586 $25,000 

Massachusetts N/A $148,523 $112,237 $104,255 

Minnesota $52,432 $52,553 $55,745 N/A 

Missouri N/A $79,110 $75,970 $80,000 

New Mexico N/A N/A $183,134 $165,484 

Nevada N/A N/A N/A $112,511 

New York N/A $137,828 $158,293 $126,834 

Ohio N/A $50,000 $65,410 $100,448 

Oklahoma N/A N/A N/A $263,267 

Pennsylvania $79,257 $63,531 $85,615 $77,496 

Rhode Island $68,182 $29,864 $116,171 $20,000 

South Carolina $59,593 $107,332 $147,955 N/A 

Texas $28,273 $33,770 $40,000 $56,192 

Wisconsin $33,532 $47,859 $71,094 N/A 

Note: In some cases, the subgrant amounts include monies carried over from a previous fiscal year. 

Source: SRI 2001-02 PCSP subgrant database. 

Exhibit reads: In 1998, Alaska had an average subgrant amount of $61,333, compared with $45,232 in 1999 and $34,552 in 2000. 

Data from 2001 was not available for Alaska. 

 

81 Data marked N/A was not available from the state at the time of the evaluation. 
78

Appendix B-4 

Range and Average of Subgrants Awarded to Charter Schools, by State (FY 2001) 

 Minimum Maximum Average 

Alaska N/A82 N/A N/A 

Arkansas $140,000 $300,000 $216,667 

California $15,000 $250,000 $95,731 

Colorado $22,000 $325,600 $140,968 

Connecticut $139,953 $185,239 $162,596 

District of Columbia $1,510 $365,183 $112,184 

Delaware N/A N/A N/A 

Georgia $123,000 $268,000 $148,807 

Idaho $131,065 $307,609 $197,328 

Illinois $8,000 $140,000 $65,304 

Kansas $25,000 $25,000 $25,000 

Massachusetts $42,085 $213,524 $104,255 

Minnesota N/A N/A N/A 

Missouri $80,000 $80,000 $80,000 

New Mexico $150,000 $304,862 $165,484 

Nevada $8,025 $150,000 $112,511 

New York $5,000 $150,000 $126,834 

Ohio $40,000 $250,000 $100,448 

Oklahoma $207,100 $372,500 $263,267 

Pennsylvania $825 $387,750 $77,496 

Rhode Island $20,000 $20,000 $20,000 

South Carolina N/A N/A N/A 

Texas $6,000 $85,000 $56,192 

Wisconsin N/A N/A N/A 

Note: In some cases, the subgrant amounts included monies carried over from a previous fiscal year if not used by a 

charter school. 

Source: SRI 2001-02 PCSP subgrant database. 

Exhibit reads: For FY 2001, the minimum subgrant in Arkansas was $140,000, and the maximum subgrant was 

$300,000. The average amount awarded to charter schools in Arkansas was $216,667. 

 

82 Data marked N/A was not available from the state at the time of the evaluation. 
79

Appendix B-5 

State Reports of Average Public Charter Schools Program Subgrants, by State 

Alaska $20,000 planning; $160,000 implementation. Subject to supplemental funds. 

Arkansas $10,000 preplanning; $300,000 over two years for implementation. 

California $450,000 over three years. 

Colorado Planning and implementation based on student enrollment. 

District of Columbia $350,000 

Florida 

$70,000 for two years. Subject to supplemental funds. Only awards 

implementation and dissemination subgrants. 

Georgia $200,000 for implementation. 

Hawaii Up to $450,000. 

Idaho Up to $450,000. 

Illinois Up to $50,000 for planning; $100,000-$150,000 for implementation. 

Kansas $270,000 

Louisiana Preplanning up to $20,000; $80,000 for two years of implementation. 

Massachusetts $450,000 over three years. 

Michigan $150,000 a year for three years (can be lower if enrollment drops). 

Missouri $80,000 to $100,000 for implementation. 

North Carolina $350,000 over three years. 

New Jersey $150,000 

Nevada $100,000 per year pre-charter; $150,000 implementation. 

New York $150,000 per year for planning and implementation. 

Ohio Two phases of implementation with $150,000 per phase. 

Oklahoma 

$300,000 to $450,000 depending on student population and anticipated 

growth. 

Oregon $50,000 planning; $150,000 implementation per year. 

Pennsylvania 

Amount unclear because federal funds mixed with state funds in allocating 

subgrants. Start-up awards based on new student enrollment numbers. 

Rhode Island 

State law limits federal amount to $150,000: $20,000 for planning and 

$130,000 for implementation. 

South Carolina $320,000 

Texas $150,000 

Utah 

$80,000 per year for three years; two schools have received double that 

amount. 

Wisconsin 

$10,000 for planning, can be supplemented to $40,000; $150,000 per year for 

implementation. 

Source: Reports by states at conference for state charter school coordinators convened by ED in Arlington, Va., April 2002. 

Exhibit Reads: Alaska reported that it awards its charter schools $20,000 in planning grants and $160,000 in implementation 

grants with figures subject to supplemental funds. 
80

Appendix B-6 

School Uses of Public Charter Schools Program Dissemination Subgrants, 2001-02 

0

10

20

30

40

50

60

70

80

90

Professional

Development

Technology Consultants Salaries Travel Networking

activities 

Percentage of schools (n=89)

 Sources: SRI 2001-02 charter school survey. 

 Exhibit reads: In 2001-02, 82 percent of schools reported spending PSCP dissemination funds on professional development. 
APPENDIX C: 

CHAPTER 3 

Appendix C-1 

Median Charter School Enrollment, by Grade-Level Configuration and School Year 

Median Enrollment 

Grade-Level Configuration 

1999-2000 

(n=870) 

2000-01 

(n=381) 

2001-02 

(n=477) 

All schools 169 171 190 

K-8 206 230 308 

Elementary 180 192 177 

High 132 100 150 

Middle-high 137 122 170 

Middle 140 155 154 

K-12 220 345 290 

Primary 75 131 100 

Note: Grade level configurations follow conventions established by RPP and are defined as follows: Primary 

includes only grades K-3; Elementary begins with K and goes no higher than grade 6; Middle ranges from grade 5 to 

grade 9; Middle-high includes any of grades 6-8 and any of grades 9-12 and no grades K-5; High ranges from grade 

9 to grade 12; K-8 includes any of grades K-3, any of grades 4-6, and any of grades 7-8; K-12 includes any of grades 

K-3, 4-6, any of grades 7-8, and any of grades 9-12. 

Sources: 1999-2000 data: Public Charter School SASS survey; 2000-01 and 2001-02 data: SRI 2000-01 and 2001-02 

charter school surveys. 1998-99 data were not available. 

Exhibit reads: In 1999-2000, the median enrollment of all charter schools was 169, compared with 171 in 2000-01 

and 190 in 2001-02. 

Appendix C-2 

Student Racial and Ethnic Composition in Charter Schools, 1998-99 to 2001-02 

 Percentage of Students 

Racial/Ethnic Category 1998-99 1999-2000 2000-01 2001-02 

White 48 46 41 37 

African American 24 27 28 38 

Hispanic or Latino 21 21 22 19 

American Indian or Alaska Native 3 1 2 3 

Asiana 3 3 3 1 

Native Hawaiian or other Pacific Islandera b b 1 1 a

Racial/ethnic categories were based on current census categories, and many differ somewhat from RPP and SASS 

categories. 

b

Not reported.

Sources: 1998-99 data: Nelson et al. (2000); 1999-2000 data: Public Charter School SASS survey; 2000-01 and 

2001-02 data: SRI 2000-01 and 2001-02 charter school surveys. 

Exhibit reads: In 1998-99, 48 percent of students in charter schools were white, compared with 46 percent in 

1999-2000, 41 percent in 2000-01, and 37 percent in 2001-02. 
84

Appendix C-3 

Targeted and Attracted Populations of Charter Schools, 2001-02 

Percentage of Schools 

(n=477) 

Student Population Targeted Attracted 

Students from low-income communities 28 74 

Low-performing students 28 74 

Dropouts or potential dropouts 26 49 

Students with specific academic interests (e.g., fine arts, math) 19 44 

Gifted and talented students 16 59 

Special education students 16 58 

Students of a particular cultural orientation 13 41 

Limited- or non-English-speaking students 12 26 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 28 percent of charter schools targeted students from low-income communities, and 74 percent 

of charter schools attracted students from low-income communities. 
85

Appendix C-4 

Teacher Certification Provisions for Charter Schools, by State 

(Excerpted from legislative analysis conducted by ECS, last updated April 2003) 

State Whether Teachers in a Charter School Must be Certified 

AK Yes 

AZ No 

AR Yes (unless waiver is granted in charter) 

CA Yes 

CO Yes (unless waiver is granted in charter) 

CT At least 50 percent must have standard certification; 50 percent may have alternative/temporary certification 

and be working towards standard certification. 

DE Yes (with exceptions) 

DC No 

FL Yes 

GA No 

HI Yes 

ID Yes 

IL 

Charter schools may employ non-certified teachers in possession of the following: bachelor’s degree, five 

year’s experience in area of degree; passing score on teacher’s tests; and evidence of professional growth. 

Mentoring must be provided to uncertified teachers. 

IN Teachers must be certified or in the process of becoming certified through Transition to Teaching. For those 

uncertified, teaching licenses must be obtained within three years of teaching at a charter school. 

IA Yes 

KS Yes 

LA Up to 25 percent of a charter school’s teachers may be uncertified if they meet other requirements. 

MA Yes; teachers hired after August 2000 must be certified or pass the teacher certification exam. 

MI Yes; except for faculty members from an IHE that teach in the IHE’s sponsored charter school(s). 

MN Yes 

MS Yes 

MO Up to 20 percent of full-time equivalent instructional staff at a charter school may be uncertified. 

NV Up to 30 percent of staff at a charter school may be uncertified. 

NH At least 50 percent of a charter school’s teaching staff must be certified or have three year teaching 

experience. 

NJ Yes 

NM Yes 

NY At least 30 percent (or five teachers) are permitted to have other credentials. Uncertified teachers must meet 

specified criteria. 

NC Up to 25 percent of teachers in elementary schools and 50 percent of teachers in secondary charter school 

may be uncertified. 

OH Yes 

OK Yes (unless waiver is granted in charter) 

OR At least 50 percent of a charter school’s teachers must be licensed by the Teacher Standards and Practices 

Commission (TSPC) and the non-TSPC-licensed staff must be registered by the TSPC 

PA Up to 25 percent of teachers may be uncertified 

RI Yes 

SC Up to 10 percent of teachers and 25 percent in start-ups may be uncertified 

TN Yes 

TX No 

UT Charter school’s teachers are required to hold certification or have completed competency under alternative 

programs 

VA Yes 

WI 

Yes; unless search for licensed teachers is unsuccessful, in which case a special charter school license is 

available for those with bachelor’s degree in their field who take six credits of training every year and are 

supervised by a regularly certified teacher. 

WY Yes 

Source: Education Commission of the States (April 2003) “State notes: Charter schools (Charter school teachers).” 

Retrieved from the World Wide Web at: http://www.ecs.org/clearinghouse/24/15/2415.htm. 

Exhibit reads: In Alaska, teachers in a charter school must be certified. 
86

Appendix C-5 

Charter School Teachers’ Participation in Professional Development, 2001-02 

Type of Professional Development Activity 

Percentage of 

Teachers who 

Participated 

Workshops sponsored by your school 92

Professional conferences or workshops 58 

Release time to work collaboratively with other instructional staff at your 

school 62 

Peer observation and critique 49 

Workshops sponsored by your charter school authorizer 41 

Courses offered by institutions of higher education 34 

Release time for independent professional development activities other than 

workshops and conferences 31 

Note: Data from charter school coordinators (n=477) was used to calculate the overall percentage of teachers involved in 

this activity. These data refer to the overall number of charter school teachers involved in this activity divided by the overall 

number of charter school teachers. 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2002-02, 92 percent of charter school teachers participated in workshops sponsored by their schools. 

Appendix C-6 

Teacher Participation in Professional Development Activities, 1999-2000 

Percentage of Teachers who 

Participated, 1999-2000

Professional Development Activity 

Charter Schools 

(n=2,847 

teachers) 

Traditional Public 

Schools (n=42,086 

teachers) 

Attended workshops, conferences, or training*** 90 95 

Had regularly-scheduled collaboration with other teachers 

on instructional issues 74 74 

Did individual or collaborative research on professionally 

interesting topic*** 52 47 

Participated in a formal mentor or peer observation 

program*** 48 42 

Made observational visits to other schools*** 41 34 

Took university course(s) for recertification or advanced 

certification*** 34 32 

Took university course(s) in main teaching assignment 

field*** 27 23 

Participated in network of teachers (organized 

externally)** 26 25 

Presented at workshops, conferences or training*** 25 22 

***p<.01, **p<.05 (Indicates significant difference between charter school teachers and traditional public school teachers in the 

proportion participating in each type of professional development activity.) 

Source: 1999-2000 Public Charter School Teacher and Public School Teacher SASS surveys. 

Exhibit reads: In 1999-2000, 90 percent of charter school teachers reported attending workshops, conferences, or training 

compared to 95 percent of traditional public school teachers. This difference is statistically significant. 
87

Appendix C-7 

Parental Involvement Activities in Charter Schools, 2001-02 

Activity 

Percentage of Schools 

(n=477)

Serve on school advisory committees 87 

Serve on school governing board 82 

Do work such as supervising lunch or a field trip 81 

Do school fundraising 77 

Do student or parent recruitment 70 

Attend workshops or classes offered by the school 67 

Serve as classroom aides or assistants to support instructional programs 61 

Supervise or direct extracurricular activities 57 

Do clerical tasks for the school 56 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 87 percent of charter schools reported that parents served on school advisory committees.

Appendix C-8 

Required and Voluntary Parent Involvement Activities in Charter Schools, 2001-02 

 Percentage of Schools 

Parent/Family Involvement Activities Required Voluntary Both 

Serve on school governing board (n=387) 14 75 11 

Serve on school advisory committees (n=416) 7 85 8 

Attend workshops or classes offered by the school (n=320) 5 89 6 

Serve as classroom aides or assistants to support 

instructional programs (n=292) 4 89 7 

Do work such as supervising lunch or a field trip (n=385) 3 90 6 

Do school fundraising (n=368) 3 92 5 

Supervise or direct extracurricular activities (n=273) 3 92 5 

Do clerical tasks for the school (n=268) 2 90 8 

Do student or parent recruitment (n=336) <1 96 3 

Note: The number of respondents varies by activity because some charter school directors reported that parents were not involved 

in these activities. 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 14 percent of charter schools required parents to serve on governing boards, 75 percent allowed 

parents to play this role voluntarily, and 11 percent said it was both required and voluntary. 
88

Appendix C-9 

State Policies Regarding Charter School Waivers and Exemptions 

from State or District Policies, 2001-02 

Area of State Waiver/Exemption 

Percentage of States that 

Allowed Waivers (n=34) 

Length of school day or year 68 

Teacher/staff hiring/firing policies 65 

Other teacher policies (e.g., teacher contract year, tenure 

requirements) 61 

Teacher salary/pay schedule 58 

Teacher certification requirements 56 

Control of finances/budget/ability to allocate funds 56 

District student assessment policies for school and classroom use 56 

Curriculum requirements 48 

Student admission policies 35 

Student attendance policies 32 

Continuing education unit requirements for teachers 29 

Incentives, rewards, or sanctions due to school performance 26 

State student assessment policies 12 

Sources: SRI 2001-02 state coordinator survey. 

Exhibit reads: In 2001-02, 68 percent of states allowed waivers or exemptions to charter schools regarding the length of the school 

day or year. 
APPENDIX D: 

CHAPTER 4 

Appendix D-1 

Reasons Authorizers Sponsor Charter Schools, by Type of Authorizer (2001-02) 

(Respondents Citing Reasons as “Very Important”) 

0 10 20 30 40 50 60 70 80 90 100

Improve public schools***

Improve educational program

quality***

Create competition in the

school system***

Create alternatives for

students and parents***

Create schools to

accommodate students

Stimulate development of

schools***

Increase program

availability***

Generate income**

Test curricular practices

Respond to public/political

pressure

Target certain populations

Keep students in public system

Fulfill state charter mandates

Percentage of Authorizers

Local (n=80) State (n=12) University (n=21)

***p<.01, **p<.05 (Indicates significant association between type of authorizer and reason for sponsoring charter schools.) 

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, 50 percent of university authorizers cited fulfilling state charter mandates as a “very important” 

reason to sponsor charter schools, compared with 92 percent of state authorizers and 45 percent of local authorizers. These data 

indicate a statistically significant association between type of authorizer and this reason for sponsoring schools. 
92

Appendix D-2 

Authorizer Involvement in Formal and Informal Networks, by Type (2001-02) 

0

10

20

30

40

50

60

70

80

90

100

No networks Only informal networks Only formal networks Both formal and informal

networks

Percentage of Authorizers

Local (n=82) State (n=13) University (n=22) Total (n=117)

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, 18 percent of local authorizers reported no involvement in networks, compared with 0 percent of state 

authorizers, 9 percent of university authorizers. In addition, 17 percent of all authorizers reported no involvement in networks. 
93

Appendix D-3 

Assistance Provided to Charter Schools by Authorizers, 2000-01 and 2001-02 

Direct Assistance

0

20

40

60

80

100

Technical

assistance

Monetary

support

In-kind support

Percentage of Authorizers

2000-01 (n=100) 2001-02 (n=118)

 

Brokered Assistance

0

20

40

60

80

100

Technical

assistance

Monetary

support

In-kind support

Percentage of Authorizers

2000-01 (n=100) 2001-02 (n=118)

Sources: SRI 2000-01 and 2001-02 authorizer surveys. 

Exhibit reads: In 2001-02, 78 percent of authorizers reported providing direct technical assistance to charter schools, compared with 86 

percent in 2001-02. Twenty-nine percent of authorizers in 2000-2001, and 55 percent in 2001-02 reported brokering technical assistance to 

charter schools. 

Appendix D-4 

Services Provided to Charter Schools, by Type of Authorizer (2001-02) 

Percentage of Authorizers 

Services 

Local 

(n=83)

State 

(n=13) 

University 

(n=22) 

Total 

(n=117) 

Administrative oversight, monitoring, and 

evaluation 81 85 68 80 

Assistance in meeting state or federal 

regulations** 76 85 45 74 

Special education services*** 79 46 10 74 

Special education testing and 

assessment*** 78 42 10 73 

Data management services*** 68 23 14 64 

Preparation of charter contract 60 38 48 59 

Budget preparation*** 61 8 18 57 

Human resources administration*** 60 8 14 56 

Reduced or free rent for facilities*** 57 8 9 53 

Bookkeeping*** 57 0 5 53 

Payroll services*** 56 15 0 52 

Food services*** 54 25 5 51 

Facility maintenance*** 54 8 9 50 

Purchasing*** 54 15 5 50 

Health services*** 52 23 5 49 

Transportation*** 52 8 5 48 

Supplies and equipment*** 48 17 14 46 

Social services*** 39 8 0 36 

***p<.01, **p<.05 (Indicates significant association between type of authorizer and services provided to charter schools.) 

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, 81 percent of local authorizers reported providing administrative oversight, monitoring, and evaluation services to 

the charter schools they authorized, compared with 85 percent of state authorizers, and 68 percent of university authorizers. These data 

indicate that no significant association exists between type of authorizer and the provision of administrative oversight, monitoring, and 

evaluation. Eighty percent of all authorizers provided these services.
94

Appendix D-5 

Authorizers’ Reasons for Denying Charter Applications, 2001-02 

Area of Inadequacy Percentage of Authorizers (n=47) 

Curricular/instructional strategies 71 

Governance procedures 57 

Accountability provisions 52 

Business plan 50 

School facilities 46 

Assessment system 44 

School leader backgrounds 44 

Special education services 40 

School mission/goals 38 

School location 37 

Parent/community involvement plan 33 

Student admission procedures 33 

Health/safety issues 32 

Targeted student population 31 

Personnel policies 29 

For-profit or nonprofit role 26 

Student discipline policies 15 

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, 71 percent of authorizers reported denying charter applicants because of 

inadequacies related to curricular or instructional strategies. 

Appendix D-6 

 Authorizer Monitoring, From the Charter School Perspective a (2001-02)

Percentage of Charter Schools 

Schools 

With Local 

Authorizer 

(n=214) 

Schools 

With State 

Authorizer 

(n=195) 

Schools With 

University 

Authorizer 

(n=57) 

Total 

(n=476) 

Finances 72 41 70 58 

Compliance with regulations 67 45 78 58 

Statewide assessment scores 71 43 67 58 

Special education services 70 37 55 53 

Enrollment numbers 58 38 59 50 

Other standardized test scores 48 35 70 45 

Performance-based test scores 42 30 47 45 

Curriculum alignment 50 36 65 45 

Other student performance indicators 56 35 39 45 

Student admission 47 32 57 42 

School governance 46 28 54 39 

Instructional practices 41 30 50 37 

Student discipline/safety 41 28 44 36 

a

 Only the items for which more than one-third of the total number of schools reported authorizer monitoring are included in this 

table. 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 72 percent of schools with local authorizers reported that their authorizer monitored their finances, 

compared with 41 percent with state authorizers, 70 percent with university authorizers. In addition, 58 percent of all schools

reported that their authorizer monitored the schools finances. 
95

Appendix D-7 

Authorizer Monitoring Procedures, 2001-02 

Average Percentage of Schools 

Monitored by Authorizers 

(n=118) 

Schools’ reporting of standardized test scores 94 

Review of compliance with state and federal regulations 93 

Fiscal audits (by authorizer or another agency) 93 

Annual reports from schools 91 

Review of progress toward goals listed in original charter 90 

Site visits to schools 85 

Information to schools about progress 82 

School-level self-evaluations 66 

Parent surveys 59 

Staff performance evaluations 53 

Staff surveys 48 

Student surveys 45 

School-level evaluations conducted by an independent entity 34 

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, authorizers reported monitoring 94 percent of their charter schools, on average, through the schools’ 

reporting of standardized test scores. 

Appendix D-8 

Frequency of Authorizer Monitoring, 2001-02 

 Percentage of Authorizers

End of the 

Charter Cycle83 Annually 

More Than 

Once a Year 

Instructional practices (n=72) 9 35 57 

Alignment of curriculum to state standards (n=94) 8 58 34 

Financial record keeping and viability (n=105) 3 25 72 

Compliance with federal or state regulations (n=106) 8 41 51 

Student achievement results on statewide assessment (n=105) 6 76 18 

Student achievement results on other standardized tests (n=80) 4 72 24 

Student performance on performance-based tests (n=81) 5 61 33 

Other student performance indicators, such as attendance rates (n=88) 5 47 49 

Student discipline and school safety (n=80) 3 42 55 

Diversity of student body (n=67) 7 55 38 

School management, leadership, or governance (n=90) 6 42 52 

Relationship with education management organization (n=27) 11 31 58 

Parent satisfaction (n=69) 15 57 28 

Parent or community involvement (n=67) 7 49 44 

Student admission and selection procedures (n=80) 13 52 35 

Enrollment numbers (n=105) 2 26 71 

Staff performance and/or attendance (n=53) 5 48 47 

Staff satisfaction (n=42) 7 55 38 

Special education services (n=94) 2 30 68 

Source: SRI 2001-02 authorizer survey 

Exhibit reads: In 2001-02, 9 percent of authorizers reported monitoring charter school instructional practices at the end of the charter 

cycle, compared with 35 percent of authorizers who monitored this annually and 57 percent who monitored this more than once a year. 

 

83 The charter school cycle is typically five years (U.S. Department of Education, 2000). 
96

Appendix D-9 

Charter School Reports on Areas Monitored by 

Charter School Governing Body,a

 2001-02 

0 10 20 30 40 50 60 70 80 90 100

Finances

Enrollment numbers

Other standardized test scores

Performance-based test scores

Other student performance indicators

School governance

Instructional practices

Student discipline/safety

Parent/community involvement

Parent satisfaction

Staff performance/attendance

Percentage of Schools (n=475)

a

 Areas included in table were reported by 60 percent or more of charter school respondents. 

Source: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 77 percent of charter schools reported being monitored on staff performance/attendance by their own 

governing bodies. 
97

Appendix D-10 

State Monitoring Procedures for Charter Schools, 2001-02 

0 10 20 30 40 50 60 70 80 90 100

Annual reports

Informal site visits

Fiscal audits

Formal site visits 

School self-evaluations

Third-party evaluations

Review progress toward charter goals 

Charter/noncharter school comparisons

Parent surveys

Student interviews/surveys

Percentage of States

States that are not authorizers (n=20)

States that are authorizers (n=15)

Total (n=35)

Source: SRI 2001-02 state coordinator survey. 

Exhibit reads: In 2001-02, 35 percent of states that are not authorizers reported monitoring charter schools through student 

interviews/surveys, compared with 47 percent of states that are authorizers and 40 percent of all states. 

Appendix D-11 

Monitoring Roles of EMOs and 

CBOs, as Reported by Schools (2001-02) 

0

20

40

60

80

100

Monitoring progress tow ard

charter terms

Monitoring student performance Ensuring compliance w ith

regulations

Percentage of Schools

Charter Schools w ith EMOs (n=83) Charter Schools w ith CBOs (n=74)

Sources: SRI 2001-02 charter school survey. 

Exhibit reads: In 2001-02, 80 percent of schools with EMOs reported that the EMOs monitored progress toward the 

charter terms, while 72 percent of schools with CBOs reported that CBOs monitored this area. 
98

Appendix D-12 

Authorizer Challenges, by Type of Authorizer (2001-02) 

Percentage of Authorizers 

Local 

(n=83) 

State 

(n=13) 

University 

(n=19) 

Lack of personnel 46 69 61 

Inadequate resources 56 69 44 

Lack of clarity about charter law 45 46 44 

Lack of clarity about other state laws 38 38 11 

Lack of clarity about federal laws 32 46 17 

Union opposition 23 42 11 

Community opposition 18 38 11 

Difficulty creating accountability system 29 31 39 

Difficulty setting performance targets 30 23 11 

Difficulty measuring progress 41 38 28 

Difficulty closing schools 54 77 67 

Politics*** 9 54 20 

Resistance from traditional public schools** 37 77 39 

Competition with traditional public schools for resources 37 46 21 

Public confusion about charters** 42 85 39 

Difficulty determining authorizer role 34 23 39 

***p<.01, **p<.05 (Indicates significant association between type of authorizer and particular challenge.) 

Source: SRI 2001-02 authorizer survey. 

Exhibit reads: In 2001-02, 46 percent of local authorizers reported lack of personnel as a barrier, compared with 69 percent of state 

authorizers and 61 percent of university authorizers. No significant association exists between type of authorizer and this barrier. 
APPENDIX E: 

CHAPTER 5 

Appendix E-1 

School Performance Design and Analysis Procedures 

The evaluation work on the performance of charter schools began with a student performance work 

group. In mid-1999, SRI, several other contractors, and Planning and Evaluation Service (PES) 

representatives (now Policy and Program Studies Service [PPSS]) formed a work group to collaborate on 

strategies for analyzing student performance through multiple evaluations sponsored by the U.S. 

Department of Education (ED). The work group convened a conference in October 1999 composed of 

experts, who ED charged with making a series of judgments and recommendations about the utility of 

state student performance systems for these evaluations.84 The PES student performance work group met 

for two years to determine the feasibility of implementing the recommendations of the experts at the 

conference and to discuss the design implications for the various studies. 

After receiving suggestions from the work group, the SRI study team compiled a state-by-state, schoollevel database for student performance analyses in Fall 2002. For the 36 states with charter schools as of 

Jan. 1, 2002, the study team collected the following information: 1) the number of charter schools in 

operation as of the 2001-02 school year; and, 2) a description of the state assessment system, including 

the state’s school-level performance standards.85 This preliminary analysis included an examination of the 

limitations of the assessment system. For example, if only Title I or only traditional public schools were 

included, comparisons necessary to answer the evaluation questions were not possible. Similarly, it was 

necessary to determine whether a large proportion of data was missing for charter schools or traditional 

public schools. Recognizing that appropriate comparisons between charter and traditional public schools 

require controls for student characteristics, the evaluation also identified data relating to school and 

student characteristics that were available in each state. 

As of early 2003, and based on assessment data for 2001-02, the 36 states with operating charter schools 

fell into five groups. Five states were in the first group of schools—the analysis of student performance in 

charter schools and traditional public schools was conducted for this group of schools (see Chapter 5). 

The remaining four groups were not included in the analysis. These groups include: 

• States with no accountability system or performance measure for public schools (nine states). 

State accountability systems are evolving rapidly, especially in response to the requirements of the 

Elementary and Secondary Education Act of 1965 (ESEA), as amended by the No Child Left 

Behind Act of 2001 (NCLB). Nevertheless, for the 2001-02 school year, several states did not have 

an accountability system that met the needs of this evaluation, either because the system did not 

apply to all schools or because it lacked a school-level performance standard, or both. Many states 

were in the process of developing their school-level targets (versus grade level and subject area 

targets, which were more common before NCLB), and others included only Title I schools in their 

accountability system. As a result, several states did not yet have data available that would allow 

comparisons of charter and traditional public schools against a school-level standard at the time of 

the analysis. In addition to these problems, a few states had only district-level performance targets. 

(The states with no school-level accountability system or performance standard were Georgia, 

Kansas, Michigan, Minnesota, Missouri, Ohio, Pennsylvania, District of Columbia, and 

Wisconsin.) 

 

84 For more information about the conference and the work group’s advanced preparation, see Miller et al. (2000). In 

brief, the experts determined that longitudinally linked, student-level scores from consecutive grade levels were 

required to answer most questions related to the effects of particular programs or educational interventions on 

student performance. However, relatively few states met this standard at that time. 

85 Each state’s “performance standard” was defined as the standard or benchmark that public schools were expected 

to meet in 2001-02. 
102

• States that exclude charter schools from their accountability systems (one state). This 

exclusion is related to the previous one, but it applies to the one state (New York) that excluded 

charter schools from the accountability system. 

• States with too few charter schools for meaningful analysis (16 states). Analyses were not 

conducted for states with fewer than 20 charter schools because student performance findings 

comparing charter schools with traditional public schools based on so few charter schools could 

not be interpreted with confidence. (The states with too few charter schools for analysis included 

Alaska, Arkansas, Connecticut, Delaware, Hawaii, Idaho, Louisiana, Mississippi, Nevada, 

Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Virginia, and Wyoming.) 

• States with large amounts of missing data for charter schools (five states). States in which 

student performance data were missing for more than 40 percent of charter schools were excluded 

to ensure that the findings were based on a solid majority (i.e., 60 percent or more) of those 

schools. (The states with data missing for more than 40 percent of charter schools included 

Arizona, California, Florida, New Jersey, and New Mexico.) 

The first step of the analysis was to determine the extent to which charter schools and traditional public 

schools met the state performance standard in each state. To make this analysis relevant to particular state 

contexts, the determination of “meeting the state standard” was based on each state’s accountability 

systems. Specifically, schools were designated as not meeting the state standard in this analysis if: 

• they were rated “unsatisfactory” (Colorado), 

• they were placed on a watch/warning list (Illinois), 

• they were rated as not making adequate yearly progress (Massachusetts), 

• they were rated “low performing” (North Carolina, Texas). 

The second step of this analysis was an examination of student performance for charter schools and 

traditional public schools, controlling for one background variable at a time. That is, each background 

variable was used to examine the relationship between meeting or not meeting state performance 

standards and school type (i.e., charter or traditional public schools). The following control variables were 

included in most states: student enrollment, percent minority,86 percent low-income,87 and mobility. For 

this analysis, each control variable was divided into halves (e.g., schools above and below the median of 

each variable).88 Significance tests were performed for these analyses, using the chi-squared test or, when 

fewer than five schools were in a cell, the Fisher’s exact test. These procedures were performed on data 

from all five states.89

 

86 This analysis defines students from any nonwhite racial or ethnic category as “minority”. 

87 This report uses the terms “low-income” and “high poverty” interchangeably. 

88 Medians were calculated on the basis of all schools in the state. 

89 To test the hypothesis that charter schools that have been in operation for more than a few years outperform newly 

opened schools (schools that may not have had enough time to fully implement and fine-tune their educational 

programs), the relationship between school performance (i.e., meeting the state performance standard) and the 

number of years the charter school had been in operation was examined. The analysis compared charter schools 

open one to two years with charter schools open for three or more years. In four states, these comparisons were not 

statistically significant, meaning the age of a charter school was not associated with its likelihood of meeting state 

performance standardsold and new schools met or did not meet the performance standard in equal proportions. In 

one state a significant difference existed, with older schools less likely to meet the performance standard than newer 

schools. 
103

The third step in the analysis was logistic regression. Logistic regression was conducted to examine the 

extent to which school type was associated with performance while controlling for all the available 

background variables simultaneously. Regressions were performed in the two states (Texas and Colorado) 

that had sufficient numbers of low-performing charter and traditional public schools and sufficient 

numbers of charter schools generally.90 Before conducting the regressions, tests of the collinearity of 

variables were conducted, finding that the variables were sufficiently distinct to include in the models. 

The outcome variable for this analysis was low performance or “not meeting the state performance 

standard.”91 

 

90 Logistic regression was not possible in the other states because they had either too few low-performing schools 

(North Carolina) or too few charter schools (Massachusetts, Illinois). 

91 Coded “1” for schools designated by the state as “not meeting the state performance standard,” “0” otherwise. 

This report uses the terms “not meeting the state performance standard” and “low performing” interchangeably. 

Appendix E-2 

Texas Student Performance Results 

The Texas accountability system included two systems: a standard system and an alternative system.92

The following pages only discuss the performance of charters within the standard system.93 The analysis 

discussed here draws on data from charter schools and traditional public schools that were assigned 

accountability ratings.94 Most traditional public schools (n=6,420) and approximately half of the charter 

schools in Texas (n=118) participated in the standard assessment system. Charter schools in the standard 

assessment system had higher mobility,95 higher proportions of nonwhite students, higher proportions of 

educationally disadvantaged students (“low-income”)96 and lower enrollment numbers than traditional 

public schools. The following table illustrates these differences. 

Descriptive Statistics (Texas), 2001-02 

Mean (Standard Deviation) 

Traditional public schools 

(n=6,420) 

Charter schools 

(n=118) 

All schools 

(n=6,538) 

Percent lowincome** 52 

(28) 

59 

(27) 

53 

(28) 

Percent 

minority*** 

55 

(32) 

76 

(31) 

55 

(32) 

Enrollment*** 

624 

(468) 

263 

(235) 

618 

(469) 

Mobility rate*** 

19 

(8) 

37 

(26) 

20 

(9) 

***p<.01, **p<.05 

Exhibit reads: In 2001-02 for Texas, the mean proportion of low-income students in traditional public schools was 52 percent 

with a standard deviation of 28 percent. This difference between the percent low-income in traditional public schools and charter 

schools is statistically significant. 

In 2001-02, 152 schools in the standard accountability system did not meet the Texas performance 

standard. Compared with traditional public schools, a higher proportion of charter schools in the standard 

system were low performing: 34 percent of charter schools, compared with 2 percent of traditional public 

schools. The following table illustrates this finding. 

 

92 The accountability system uses performance data on the Texas Assessment of Academic Skills (TAAS) and 

annual dropout rates to determine whether a school is rated as exemplary, recognized, acceptable, or low 

performing. The TAAS is administered in grades 3-8 and 10 in reading and writing, mathematics, science, and social 

studies, depending on the grade level. Schools are designated as low performing if students do not meet the 

performance standard in any tested subject area. 

93 The alternative system was not included in the analysis because it was discontinued to comply with NCLB. As of 

2002-03, all schools were included in one accountability system. 

94 A number of schools were excluded from the accountability system (and did not receive a school rating) by the 

state for the following reasons: (1) the school had poor data quality, (2) the campus served only primary grades, (3) 

the charter school was in its first year of operation, (4) the charter school had insufficient data to be evaluated. 

95 Student mobility in Texas was calculated by dividing the number of mobile students, defined as students in 

attendance at the school for less than 83 percent of the school year, by the total number of students. 

96 The percentage of educationally disadvantaged or “low-income” students in Texas was calculated as follows: the 

sum of the students coded as “eligible for free or reduced-price lunch[es] or eligible for other public assistance,” 

divided by the total number of students at the school. 
106

Performance of Traditional Public Schools and Charter Schools in Texas, 2001-02 

Number of Schools (Percentage) 

Met Standard Did Not Meet Standard Total 

Traditional public schools 

6,308 

(98) 

112 

(2) 

6,420 

(98)

Charter schools 

78 

(66) 

40 

(34) 

118 

(2)

Total 

6,386 

(98) 

152 

(2) 

6,538 

(100)

Significant association exists between school type and performance with a lower 

proportion of charter schools meeting the standard*** 

***p<.01 

Exhibit reads: In Texas, 6,308 traditional public schools (98 percent) met the state performance standard, and 

112 traditional public schools (2 percent) did not meet the standard in 2001-02. 

To better understand these differences in performance, the step two analyses were conducted to control 

for certain school characteristics, the proportion of minority students, the proportion of low-income 

students, the proportion of mobile students, and student enrollment. Charter schools are more likely to not 

meet the state performance standard even after comparing similar charter schools and traditional public 

schools based on the proportion of low-income students in the school. The gap between the proportion of 

low-performing charter schools and the proportion of low-performing traditional public schools is smaller 

in the group of schools with lower proportions of low-income students, but remains statistically 

significant. 

Analysis of School Type and Performance, Controlling for 

Proportion of Low-Income Students (Texas), 2001-02 

Number of Schools (Percentage) 

Low-Income: Below State Median Low-Income: Above State Median 

Met 

std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional public 

schools 

3,206 

(99) 

17 

(1) 

3,223

(99)

Traditional 

public schools 

3,102 

(97) 

95 

(3) 

3,197 

(98)

Charter schools 

32 

(76) 

10 

(24) 

42

(1) Charter schools 

46 

(61) 

30 

(39) 

76 

(2)

Total 

3,238 

(99) 

27 

(1) 

3,265

(100) Total 

3,148 

(96) 

125 

(4) 

3,273 

(100)

Significant association exists between school type 

and performance for schools with below average 

percentages of low-income students (with a lower 

proportion of charter schools meeting the 

standard)*** 

Significant association exists between school 

type and performance for schools with above 

average percentages of low-income students 

(with a lower proportion of charter schools 

meeting the standard)*** 

***p<.01 

Note: State median = 52 percent 

Exhibit reads: In Texas, 3,206 traditional public schools (99 percent) below the state median in the proportion of low-income 

students met the state performance standard, and 17 (1 percent) did not meet the standard in 2001-02. 
107

Examining the relationship between performance and school type while controlling for the proportion of 

minority students leads to a similar finding. Comparing charter and traditional public schools with similar 

populations of minority students does not change the finding: Different proportions of charter schools met 

the state standard compared to traditional public schools. Again, the gap between the proportion of charter 

and traditional public schools not meeting the standard was smaller for the schools with low numbers of 

minority students but remained statistically significant. 

Analysis of School Type and Performance, Controlling for 

Proportion of Minority Students (Texas), 2001-02 

Number of Schools (Percentage) 

Minority: 

 Below State Median 

Minority: 

Above State Median 

Met 

std. 

Did not meet 

std. Total 

Met 

std. 

Did not meet 

std. Total 

Traditional 

public schools 

3,349 

(99) 

20 

(<1) 

3,369

(99)

Traditional 

public schools 

2,959 

(97)

92 

(3) 

3,051 

(97)

Charter schools 

24 

(80) 

6 

(20) 

30

(1) Charter schools 

54 

(61)

34 

(39) 

88 

(3)

Total 

3,373 

(99) 

26 

(1) 

3,399

(100) Total 

3,013 

(96)

126 

(4) 

3,139 

(100)

Significant association exists between school type 

and performance for schools with below average 

percentages of minority students (with a lower 

proportion of charter schools meeting the 

standard)*** 

Significant association exists between school type 

and performance for schools with above average 

percentages of minority students (with a lower 

proportion of charter schools meeting the 

standard)*** 

***p<.01 

Note: State median = 55 percent 

Exhibit reads: In Texas, 3,349 traditional public schools (99 percent) with minority student populations below the state median met 

the state performance standard, and 20 (fewer than 1 percent) did not meet the standard in 2001-02. 

The study team also examined the effect of student mobility on the relationship between performance and 

school type. Controlling for mobility did not affect the relationship between school type and performance 

discussed above. 
108

Analysis of School Type and Performance, Controlling for 

Proportion of Mobile Students (Texas), 2001-02 

Number of Schools (Percentage in parenthesis) 

Mobility: 

Below State Median 

Mobility: 

Above State Median 

Met 

std. 

Did not meet 

std. Total 

Met 

std. 

Did not meet 

std. Total 

Traditional 

public schools 

3,622 

(99) 

33 

(1) 

3,655

(99)

Traditional 

public schools 

2,686 

(97)

79 

(3) 

2,765 

(97)

Charter schools 

40 

(89) 

5 

(11) 

45

(1) Charter schools 

38 

(52)

35 

(48) 

73 

(3)

Total 

3,662 

(99) 

38 

(1) 

3,700

(100) Total 

2,724 

(96)

114 

(4) 

2,838 

(100)

Significant association exists between school type 

and performance for schools with below average 

percentages of mobile students (with a lower 

proportion of charter schools meeting the 

standard)*** 

Significant association exists between school type 

and performance for schools with above average 

percentages of mobile students (with a lower 

proportion of charter schools meeting the 

standard)*** 

***p<.01 

Note: State median = 19 percent 

Exhibit reads: In Texas, 3,622 traditional public schools (99 percent) with student mobility rates below the state median met the 

state performance standard, and 33 (1 percent) did not meet the standard in 2001-02. 

Finally, the evaluation examined whether controlling for enrollment changed the findings above. Even 

after this step, the differences in the proportion of charter and traditional public schools meeting the Texas 

performance standard were statistically significant. 

Analysis of School Type and Performance Controlling for Enrollment (Texas), 2001-02 

Number of Schools (Percentage in parenthesis) 

Enrollment: 

 Below State Median 

Enrollment: 

Above State Median 

 Met std. 

Did not meet 

std. Total Met std. 

Did not meet 

std. Total 

Traditional 

public 

schools 

2,608 

(98) 

52 

(2) 

2,660

(96)

Traditional 

public 

schools 

3,700 

(98)

60 

(2) 

3,760 

(100)

Charter 

schools 

66 

(65) 

35 

(35) 

101

(4)

Charter 

schools 

12 

(71)

5 

(29) 

17 

(<1)

Total 

2,674 

(97) 

87 

(3) 

2,761

(100) Total 

3,712 

(98)

65 

(2) 

3,777 

(100)

Significant association exists between school type 

and performance for schools with below average 

student enrollment (with a lower proportion of 

charter schools meeting the standard)*** 

Significant association exists between school type 

and performance for schools with above average 

student enrollment (with a lower proportion of 

charter schools meeting the standard)*** 

***p<.01 

Note: State median = 469 students. 

Exhibit reads: In Texas, 2,608 traditional public schools (98 percent) with enrollments below the state median met the state 

performance standard, and 52 (2 percent) did not meet the standard in 2001-02. 

In summary, the finding that traditional public schools met the state performance standard at higher rates 

than charter schools did not change after controlling for the proportion of low-income and minority 

students, student mobility, or student enrollment. 
109

For Texas, the evaluation also included logistic regression. For Texas, the logistic regression model was: 

The log odds of meeting the state standard97

 = β0 + β1 (percent low-income) + β2 (percent 

minority) + β3 (enrollment: large school) + β4 (enrollment: small school) + β5 (mobility: high) + 

β6 (mobility: low) + β7 (charter98) + e (error term) 

As the formula illustrates, two variablesenrollment and mobilitywere converted from continuous to 

categorical because of the nonlinear relationship with the outcome in Texas.99 The findings from the 

logistic regression indicate that being a charter school was associated with not meeting the state 

performance standard, controlling for low-income students, minority students, student mobility, and 

student enrollment. 

Logistic Regression Results (Texas), 2001-02 

 DF Estimate Standard Error Chi-Square Pr > ChiSq

Intercept 1 -6.0056 0.3723 260.2098 <.0001 

Low-income 1 0.000331 0.00571 0.0034 0.9538 

Minority*** 1 0.0212 0.00552 14.7297 0.0001 

Enrollment: Large school 1 0.1723 0.2280 0.5710 0.4499 

Enrollment: Small school*** 1 0.8793 0.2455 12.8312 0.0003 

Mobility: High*** 1 0.9600 0.2117 20.5708 <.0001 

Mobility: Low** 1 -0.8579 0.3373 6.4674 0.0110 

Charter*** 1 2.4977 0.2850 76.8016 <.0001 

***p<.01, **p<.05 

Exhibit reads: The relationship between charter schools and the log odds of not meeting the state standard is significant 

(p<.01) after controlling for all of the other variables in the model. 

 

97 Coded “1” for not meeting the standard, “0” for meeting the standard. 

98 Coded “1” for charter and “0” for traditional public schools. 

99 These variables were converted to separate dummy variablesenrollment (small, medium, large) and mobility 

(low, medium, high). In each case, the middle category was the omitted category in the logistic regression. 

Appendix E-3 

Colorado Student Performance Results 

The Colorado assessment system involves student performance results on the state’s criterion-referenced 

test, the Colorado Student Assessment Program (CSAP).100 Performance data for 2001-02 are available 

from 1,451 traditional public schools and 84 charter schools in Colorado. Charter schools in Colorado 

have lower proportions of students eligible for free and reduced-price lunches101 (“low-income”), fewer 

nonwhite students, and lower average enrollments than traditional public schools, as illustrated in the 

following table. 

Descriptive Statistics (Colorado), 2001-02 

Mean (Standard Deviation in parenthesis) 

Traditional public schools 

(n=1,546) Charter schools (n=84)

All Schools 

(n=1,630) 

Percent lowincome*** 33 

(25) 

19 

(26) 

32 

(25) 

Percent 

minority*** 

32 

(26) 

23 

(25) 

32 

(26) 

Enrollment*** 

464 

(386) 

293 

(250) 

455 

(382) 

***p<.01 

Exhibit reads: In Colorado, the mean proportion of low-income students in traditional public schools is 33 percent with a 

standard deviation of 25 percent in 2001-02. This difference between the percent low-income in traditional public schools and 

charter schools is statistically significant. 

In 2001-02, 38 Colorado schools (2 percent) were not meeting the state’s performance standard. Ten 

percent of charter schools did not meet the standard, compared with 2 percent of traditional public 

schools. These differences were statistically significant, as indicated below. 

 

100 In 2001-02, students in Colorado in grades 3-5, 7-8, and 10 participated in the Colorado Student Assessment 

Program (CSAP), a test that measures student achievement against the Colorado Model Content Standards. 

Depending on grade level, CSAP includes tests of reading and writing, mathematics, and science. Colorado schools 

are rated based on their overall CSAP scores from grades 3-10 and the ACT for all 11th-graders and designated as 

Excellent Academic Performance, High Academic Performance, Average Academic Performance, Low Academic 

Performance, or Unsatisfactory. Schools rated as Unsatisfactory are considered by the state not to have met the 

Colorado performance standard. If a school serves multiple levels (e.g., K-8 serves both elementary and middle), it 

receives more than one accountability rating. Note: In less than five cases, a charter school had an unsatisfactory 

rating for one level and a satisfactory rating for another. These schools were dropped from the analysis. 

101 In Colorado, “low-income students” are defined as those students eligible for free or reduced-price lunches. 

Student mobility data were not available in Colorado. 
112

Performance of Traditional Public Schools and Charter Schools (Colorado), 2001-02 

Number of Schools (Percentage) 

 Met standard Did Not Meet Standard Total 

Traditional public schools

1,421 

(98) 

25 

(2) 

1,446 

(95) 

Charter schools 

76 

(90) 

8 

(10) 

84 

(5) 

Total 

1,497 

(98) 

33 

(2) 

1,530 

(100) 

Significant association exists between school type and performance with a 

lower proportion of charter schools meeting the standard*** 

***p<.01 

Exhibit reads: In Colorado, 1,421 traditional public schools (98 percent) met the state 

performance standard, and 30 (2 percent) did not meet the standard in 2001-02. 

To better understand these differences, the step two analyses were conducted in Colorado, as well. As the 

table indicates, controlling for the proportion of low-income students produced mixed findings. In schools 

with higher than average (above the state median) proportions of low-income students, traditional public 

schools were more likely than charter schools to meet the state performance standard. However, among 

schools with lower than average proportions of low-income students, traditional public schools and 

charter schools performed at similar levels. 

Analysis of School Type and Performance, Controlling for 

Proportion of Low-Income Students (Colorado), 2001-02 

Number of Schools (Percentage) 

Low-income: 

Below State Median 

Low-income: 

Above State Median 

Met 

std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional public 

schools 

697 

(99) 

6 

(1) 

703

(92)

Traditional 

public 

schools 

724 

(97)

19 

(3) 

743 

(97)

Charter schools 

59 

(97) 

2 

(3) 

61

(8)

Charter 

schools 

17 

(74)

6 

(26) 

23 

(3)

Total 

756 

(99) 

8 

(1) 

764

(100) Total 

741 

(97)

25 

(3) 

766 

(100)

No significant association exists between school 

type and performance for schools with below 

average percentages of low-income students (i.e., 

charter schools and traditional public schools meet 

the standard in equal proportions) 

Significant association exists between school 

type and performance for schools with above 

average percentages of low-income students 

(with a lower proportion of charter schools 

meeting the standard)*** 

***p<.01 

Note: State median = 28 percent. 

Exhibit reads: In Colorado, 697 traditional public schools (99 percent) below the state median for the proportion of lowincome students met the state performance standard, and six (1 percent) did not meet the standard in 2001-02.
113

The data indicate a similar finding when examining charter schools and traditional public schools 

controlling for the percentage of minority students. Traditional public schools and charter schools met 

state performance standards in similar rates when comparing schools with lower than average proportions 

of minority students. However, when comparing schools with higher than average proportions of minority 

students, charter schools were less likely to meet the Colorado performance standard compared with 

traditional public schools. 

Analysis of School Type and Performance, Controlling for 

Proportion of Minority Students (Colorado), 2001-02 

Number of Schools (Percentage) 

Minority: 

Below State Median 

Minority: 

Above State Median 

Met 

std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total

Traditional public 

schools 

725 

(100) 

3 

(<1) 

728

(93)

Traditional public 

schools 

696 

(97) 

22 

(3) 

718 

(96)

Charter schools 

54 

(98) 

1 

(2) 

55

(7) Charter schools 

22 

(76) 

7 

(24) 

29 

(4)

Total 

779 

(99) 

4 

(1) 

783

(100) Total 

718 

(96) 

29 

(4) 

747 

(100)

No significant association exists between school 

type and performance for schools with below 

average percentages of minority students (i.e., 

charter schools and traditional public schools 

meet the standard in equal proportions) 

Significant association exists between school type 

and performance for schools with above average 

percentages of minority students (with a lower 

proportion of charter schools meeting the 

standard)*** 

***p<.01 

Note: State median = 23 percent. 

Exhibit reads: In Colorado, 725 traditional public schools (99 percent) below the state median for the proportion of minority 

students met the state performance standard, and three (fewer than 1 percent) did not meet the standard in 2001-02. 

In addition to finding that charter schools and traditional public schools meet the state standard in 

Colorado in similar rates when they have lower than average proportions of low-income and minority 

students, the evaluation team found that the two types of schools perform at similar rates when they have 

above average student enrollment numbers, as illustrated below. However, charter schools met the 

standard in lower rates when they had smaller than average enrollments compared with similar traditional 

public schools. 
114

Analysis of School Type and Performance, Controlling for Enrollment (Colorado), 2001-02 

Number of Schools (Percentage) 

Enrollment: 

Below State Median 

Enrollment: 

Above State Median 

 Met std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total

Traditional 

public schools 

646 

(97) 

17 

(3) 

663

(92)

Traditional 

public schools 

775 

(99)

8 

(1) 

783 

(97)

Charter 

schools 

52 

(87) 

8 

(13) 

60

(8) Charter schools 

24 

(100) 0 

24 

(3)

Total 

698 

(97) 

25 

(3) 

723

(100) Total 

799 

(99)

8 

(1) 

807 

(100)

Significant association exists between school 

type and performance for schools with below 

average student enrollment (with a lower 

proportion of charter schools meeting the 

standard)*** 

No significant association exists between 

school type and performance for schools with 

above average student enrollment (i.e., charter 

schools and traditional public schools meet the 

standard in equal proportions) 

***p<.01 

Note: State median = 389 students. 

Exhibit reads: In Colorado, 646 traditional public schools (97 percent) below the state median for student enrollment 

met the state performance standard, and 17 (3 percent) did not meet the standard. 

Student performance results in Colorado also included step three of these analyses: logistic regression. 

For Colorado, the logistic regression model was as follows:

The log odds of meeting the state standard102

 = β0 + β1 (low-income: high proportions) + β2 

(low-income: low proportions) + β3 (percent minority) + β4 (enrollment: large school) + β5 

(enrollment: small school) + β6 (charter103) + e (error term) 

Minority is a continuous variable, and enrollment and low-income were converted from continuous to 

categorical after preliminary analyses showed that these variables had a nonlinear relationship with the 

dependent variable.104 The findings from the logistic regression indicate that being a charter school was 

associated with not meeting the standard, controlling for low-income students, minority students, and 

student enrollment. 

 

102 Coded “1” for not meeting the standard, “0” for meeting the standard. 103 Coded “1” for charter and “0” for traditional public schools. 

104 These variables were converted to separate dummy variablesenrollment (small, medium, large) and lowincome (low proportions, medium proportions, and high proportions). In each case, the middle category was the 

omitted category in the logistic regression. 
115

Logistic Regression Results (Colorado), 2001-02 

 DF Estimate Standard Error Chi-Square Pr > ChiSq

Intercept 1 -9.0955 1.0323 77.6330 <0.0001 

Low-income: High Proportions 1 0.4832 0.7793 0.3845 0.5352 

Low-income: Low Proportions*** 1 2.0552 0.8399 5.9881 0.0144 

Minority*** 1 0.0625 0.0101 38.2964 <0.0001 

Enrollment: Large school 1 0.3653 0.6496 0.3162 0.5739 

Enrollment: Small school*** 1 2.1779 0.57777 14.2107 0.0002 

Charter*** 1 1.5875 0.5263 9.0985 0.0026 

***p<.01 

Exhibit reads: The relationship between charter schools and the log odds of not meeting the state standard is significant 

(p<.01) after controlling for all of the other variables in the model. 

Appendix E-4 

Illinois Student Performance Results 

The assessment system in Illinois places schools on a warning or watch list based on student achievement 

results.105 The analysis of performance in Illinois is based on data from 23 charter schools and 3,892 

traditional public schools. Charter schools in Illinois have lower average enrollments and higher 

proportions of minority and low-income students106 than the state’s traditional public schools. In addition, 

charter schools have higher than average mobility rates.107

Descriptive Statistics (Illinois), 2001-02 

Mean (Standard Deviation in parenthesis) 

Traditional public schools 

(n=3,892) Charter schools (n=23)

All schools 

(n=3,915) 

Percent lowincome*** 36 

(31) 

62 

(33) 

36 

(30) 

Percent minority*** 

33 

(36) 

83 

(28) 

33 

(36) 

Enrollment 

519 

(444) 

376 

(536) 

518 

(445) 

Mobility rate 

17 

(13) 

20 

(16) 

17 

(13) 

***p<.01 

Exhibit reads: In Illinois, the mean proportion of low-income students in traditional public schools is 36 percent with a standard 

deviation of 31 percent in 2001-02. The difference in the percent low-income in traditional public schools and charter schools is 

statistically significant. 

The 2001-02 performance ratings indicate that 712 charter and traditional public schools did not meet the 

Illinois performance standard. Of these 712 schools, 11 were charter schools, representing 48 percent of 

the charter school population, compared with 18 percent of traditional public schools, as shown below. 

 

105 The assessment system in Illinois used two tests in 2001-02: the Illinois Standards Achievement Test (ISAT) for 

grades 3-5 and 7-8, and the Prairie State Achievement Examination (PSAE) for grade 11. The ISAT includes tests in 

reading, writing, math, science, and social studies, depending on the grade level. Schools are placed on the 

Academic Early Warning List based on composite scores. 

106 In Illinois, “low-income students” are defined as pupils from families receiving public assistance, living in 

institutions for neglected or delinquent children, being supported in foster homes with public funds, or eligible to 

receive free or reduced-price lunches. 

107 In Illinois, “mobility rate” is defined as the number of times students enroll in or leave a school during the year. 
118

Performance of Traditional Public Schools and Charter Schools (Illinois), 2001-02 

 Number of Schools (Percentage) 

 Met standard Did not meet standard Total 

Traditional public schools 

3,191 

(82) 

701 

(18) 

3,892 

(99)

Charter 

12 

(52) 

11 

(48) 

23 

(1)

Total 

3,203 

(82) 

712 

(18) 

3,915 

(100)

Significant association exists between school type and performance with a lower 

proportion of charter schools meeting the standard*** 

***p<.01 

Exhibit reads: In Illinois, 3,191 traditional public schools (82 percent) met the state performance standard, 

and 701 (18 percent) did not meet the standard in 2001-02. 

When controlling for student mobility and enrollment, the differences in performance levels between 

charter and traditional public schools persisted (see tables that follow). However, non-significant results 

were found when comparing similar groups of charter and traditional public schools based on the 

proportion of low-income and minority students, as discussed below. 

Analysis of School Type and Performance, Controlling for Mobility (Illinois), 2001-02 

Number of Schools (Percentage) 

Mobility: 

 Below State Median 

Mobility: 

Above State Median 

Met 

std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional 

public schools 

1,866 

(96) 

77 

(4) 

1,943

(100)

Traditional 

public schools 

1,325 

(68) 

624 

(32) 

1,949 

(99)

Charter schools 

6 

(67) 

3 

(33) 

9

(<1) Charter schools 

5 

(42) 

7 

(58) 

12 

(1)

Total 

1,872 

(96) 

80 

(4) 

1,952

(100) Total 

1,330 

(68) 

631 

(32) 

1,961 

(100)

Significant association exists between school type 

and performance for schools with below average 

mobility rates (with a lower proportion of charter 

schools meeting the standard)*** 

Significant association exists between school type 

and performance for schools with above average 

mobility rates (with a lower proportion of charter 

schools meeting the standard)** 

***p<.01, **p<.05 

Note: State median = 14 percent. 

Exhibit reads: In Illinois, 1,866 traditional public schools (96 percent) below the state median for student mobility met the state 

performance standard, and 77 (4 percent) did not meet the standard in 2001-02. 
119

Analysis of School Type and Performance, Controlling for Enrollment (Illinois), 2001-02 

Number of Schools (Percentage) 

Enrollment: 

Below State Median 

Enrollment: 

Above State Median 

Met 

std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional public 

schools 

1,749 

(90) 

186 

(10) 

1,935 

(99) 

Traditional public 

schools 

1,442 

(74) 

515 

(26) 

1,957 

(100)

Charter schools 

12 

(60) 

8 

(40) 

20 

(1) Charter schools 

0 

(0) 

3 

(100) 

3 

(<1)

Total 

1,761 

(90) 

194 

(10) 

1,955 

(100) Total 

1,442 

(74) 

518 

(26) 

1,960 

(100)

Significant association exists between school type 

and performance for schools with below average 

student enrollment (with a lower proportion of 

charter schools meeting the standard)*** 

Significant association exists between school type 

and performance for schools with above average 

student enrollment (with a lower proportion of 

charter schools meeting the standard)** 

***p<.01, **p<.05 

Note: State median = 403 students. 

Exhibit reads: In Illinois, 1,749 traditional public schools (90 percent) below the state median for student enrollment met the state 

performance standard, and 186 (10 percent) did not meet the standard in 2001-02. 

As the table below indicates, among schools with lower than average proportions of low-income students, 

performance differences between charter schools and traditional public schools were not statistically 

significant.108 Charter schools with higher than average proportions of low-income students met the 

standard less frequently when compared with similar traditional public schools. 

Analysis of School Type and Performance, Controlling for 

Proportion of Low-Income Students (Illinois), 2001-02 

Number of Schools (Percentage) 

Low-Income: 

Below State Median 

Low-Income: 

Above State Median 

Met 

std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional 

public schools 

1,907 

(98)

40 

(2) 

1,947 

(100) 

Traditional public 

schools 

1,284 

(66) 

661 

(34) 

1,945 

(99)

Charter schools 

5 

(83)

1 

(17) 

6 

(<1) Charter schools 

7 

(41) 

10 

(59) 

17 

(1)

Total 

1,912 

(98)

41 

(2) 

1,953 

(100) Total 

1,291 

(66) 

671 

(34) 

1,962 

(100)

No significant association exists between school type 

and performance for schools with below average lowincome students (i.e., charter schools and traditional 

public schools meet the standard in equal proportions) 

Significant association exists between school type 

and performance for schools with above average 

low-income students (with a lower proportion of 

charter schools meeting the standard)** 

**p<.05 

Note: State median = 28 percent. 

Exhibit reads: In Illinois, 1,907 traditional public schools (98 percent) below the state median for the proportion of low-income students 

met the state performance standard, and 40 (2 percent) did not meet the standard. 

 

108 Please note that only six charter schools had lower than average proportions of low-income students. Therefore, 

these findings must be interpreted with caution. 
120

The Illinois data reveal an interesting finding: charter schools and traditional public schools met the 

standard in similar rates when schools with similar proportions of minority students were compared. This 

finding held for both schools with above average proportions of minority students and those with below 

average proportions. As illustrated below, controlling for minority indicates that charter schools and 

traditional public schools met the performance standard in equal rates. 

Analysis of School Type and Performance, Controlling for 

Proportion of Minority Students (Illinois), 2001-02 

Number of Schools (Percentage) 

Minority: 

Below State Median 

Minority: 

Above State Median 

Met 

std. 

Did not 

meet std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional 

public schools 

1,905 

(98) 

48 

(2) 

1,953 

(100) 

Traditional 

public schools 

1,286 

(66) 

653 

(34) 

1,939 

(99)

Charter schools 

1 

(100) 

0 

(0) 

1 

(<1) Charter schools 

11 

(50) 

11 

(50) 

22 

(1)

Total 

1,906 

(98) 

48 

(2) 

1,954 

(100) Total 

1,297 

(66) 

664 

(34) 

1,961 

(100)

No significant association exists between school type 

and performance for schools with below average 

minority students (i.e., charter schools and traditional 

public schools meet the standard in equal 

proportions) 

No significant association exists between school 

type and performance for schools with above 

average minority students (i.e., charter schools 

and traditional public schools meet the standard 

in equal proportions) 

Note: State median = 16 percent. 

Exhibit reads: In Illinois, 1,905 traditional public schools (98 percent) below the state median for the proportion of minority students 

met the state performance standard, and 48 (2 percent) did not meet the standard in 2001-02. 
Appendix E-5 

Massachusetts Student Performance Results 

The Massachusetts school performance system rates schools according to whether they make adequate 

yearly progress (AYP).109 The analysis of performance in Massachusetts is based on charter schools and 

traditional public schools with available performance data.110 A total of 1,538 traditional public schools 

and 28 charter schools are included in the analysis. Charter schools in Massachusetts have lower 

enrollments and higher proportions of minority and low-income students111 than traditional public schools 

in the state, as indicated in the following table.112 

Descriptive Statistics (Massachusetts), 2001-02 

Mean (Standard Deviation in parenthesis) 

Traditional public schools 

(n=1,538) 

Charter schools 

(n=28) 

All Schools 

(n=1,566) 

Percent lowincome*** 29 

(28) 

47 

(25) 

29 

(28) 

Percent minority*** 

24 

(28) 

48 

(40) 

24 

(28) 

Enrollment*** 

511 

(360) 

342 

(290) 

508 

(360) 

***p<.01 

Exhibit reads: In Massachusetts, the mean proportion of low-income students in traditional public schools is 29 percent with a 

standard deviation of 28 percent in 2001-02. The difference between the percent low-income in traditional public schools and 

charter schools is statistically significant.

The 2001-02 AYP performance ratings indicate that 213 charter and traditional public schools did not 

meet the Massachusetts performance standard. Thirty six percent of charter schools were represented 

among these 213 schools, compared with 13 percent of traditional public schools. 

 

109 AYP is determined every two years on the basis of performance and improvement ratings on the Massachusetts 

Comprehensive Assessment System (MCAS) tests in English language arts and mathematics. The most recent 

ratings issued (Cycle II) are based on tests administered in 2001 and 2002 at grades 4, 7, 8, and 10. AYP is 

determined for school-level (elementary, middle, high) and subject area. Schools that are designated as not meeting 

AYP, or as low performing, by the state are subject to sanctions and receive assistance. For this analysis, if a school 

was designated as low performing in either subject area for any grade level tested, it was considered low performing, 

consistent with state procedures. 

110 The analysis is based on data for 66 percent of the charter schools in Massachusetts (28 of the 42 charter 

schools). According to the Massachusetts Department of Education, these schools are missing performance data 

either because they had too few students or because they were not open for the full performance cycle (two years) on 

which the performance ratings are based. Approximately 300 traditional public schools were also excluded from this 

analysis because they did not receive a performance rating. 

111 In Massachusetts, “low-income” is defined as a student who is eligible for free or reduced-price lunches, receives 

Transitional Aid to Families benefits, or is eligible for food stamps. 

112 This analysis does not include mobility, as in other states, because Massachusetts does not maintain such data. 
122

Performance of Traditional Public Schools and Charter Schools (Massachusetts), 2001-02 

Number of Schools (Percentage) 

 Met standard Did not meet standard Total 

Traditional public 

schools 

1,335 

(87) 

203 

(13) 

1,538 

(98) 

Charter schools 

18 

(64) 

10 

(36) 

28 

(2) 

Total 

1,353 

(86) 

213 

(14) 

1,566 

(100) 

Significant association exists between school type and performance with a 

lower proportion of charter schools meeting the standard*** 

***p<.01 

Exhibit reads: In Massachusetts, 1,335 traditional public schools (87 percent) met the state 

performance standard, and 203 (13 percent) did not meet the standard in 2001-02. 

Charter schools and traditional public schools in Massachusetts serving lower than average proportions of 

low-income and minority students and schools with lower student enrollments met the state performance 

standard at similar rates. These findings are discussed below. 

As the following table indicates, charter schools and traditional public schools that had lower than 

average proportions of low-income students met the state performance standard in equal rates. However, 

charter schools that had higher proportions of low-income students were less likely to meet the 

performance standard than their traditional public school peers. 

Analysis of School Type and Performance, Controlling for 

Proportion of Low-Income Students (Massachusetts), 2001-02 

Number of Schools (Percentage) 

Low-income: 

Below State Median 

Low-income: 

Above State Median 

Met 

std. 

Did not meet 

std. Total 

Met 

std. 

Did not meet 

std. Total

Traditional 

public 

schools 

749 

(100) 

3 

(<1) 

752

(99)

Traditional 

public 

schools 

537

(73)

200 

(27) 

737 

(98)

Charter 

schools 

4 

(100) 

0 

(0) 

4

(1)

Charter 

schools 

8

(44)

10 

(56) 

18 

(2)

Total 

753 

(100) 

3 

(<1) 

756

(100) Total 

545 

(72)

210 

(28) 

755 

(100)

No significant association exists between school 

type and performance for schools with below 

average percentages of low-income students (i.e., 

charter schools and traditional public schools meet 

the standard in equal proportions) 

Significant association exists between school type 

and performance for schools with above average 

percentages of low-income students (with a lower 

proportion of charter schools meeting the 

standard)*** 

***p<.01 

Note: State median = 17 percent. 

Exhibit reads: In Massachusetts, 749 traditional public schools (100 percent) below the state median for the proportion of lowincome students met the state performance standard, and three (fewer than 1 percent) did not meet the standard in 2001-02. 
123

The data reveal a similar finding when controlling for the proportion of minority students at the school. 

Charter schools with below average proportions of minority students met the Massachusetts performance 

standard at a rate similar to that of traditional public schools (see below). However, charter schools that 

serve higher than average proportions of minority students were less likely to meet the state standard 

(compared with similar traditional public schools). 

Analysis of School Type and Performance, Controlling for 

Proportion of Minority Students (Massachusetts), 2001-02 

Number of Schools (Percentage) 

Minority: 

Below State Median 

Minority: 

Above State Median 

Met 

std. 

Did not meet 

std. Total 

Met 

std. 

Did not meet 

std. Total 

Traditional 

public schools 

722 

(98) 

17 

(2) 

739

(99)

Traditional 

public schools 

610

(77)

186 

(23) 

796 

(98)

Charter schools 

11 

(100) 

0 

(0) 

11

(1)

Charter 

schools 

7

(41)

10 

(59) 

17 

(2)

Total 

733 

(98) 

17 

(2) 

750

(100) Total 

617 

(76)

196 

(24) 

813 

(100)

No significant association exists between school type 

and performance for schools with below average 

percentages of minority students (i.e., charter 

schools and traditional public schools meet the 

standard in equal proportions) 

Significant association exists between school type 

and performance for schools with above average 

percentages of minority students (with a lower 

proportion of charter schools meeting the 

standard)*** 

***p<.01 

Note: State median = 10 percent. 

Exhibit reads: In Massachusetts, 722 traditional public schools (98 percent) below the state median for the proportion of minority 

students met the state performance standard, and 17 (2 percent) did not meet the standard in 2001-02. 

Finally, charters that had low student enrollments performed as well as traditional public schools. The 

following table illustrates this finding. However, among schools that served larger numbers of students, 

traditional public schools met the performance standards at higher rates than charter schools. 
124

Analysis of School Type and Performance, Controlling for Enrollment (Massachusetts), 2001-02 

Number of Schools (Percentage) 

Enrollment: 

Below State Median 

Enrollment: 

Above State Median 

Met 

std. 

Did not meet 

std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional 

public schools 

555 

(85) 

101 

(15) 

656 

(97) 

Traditional public 

schools 

777 

(88) 

102 

(12) 

879 

(99)

Charter schools 

15 

(79) 

4 

(21) 

19

(3) Charter schools 

3 

(33) 

6 

(67) 

9 

(1)

Total 

570 

(84) 

105 

(16) 

675 

(100) Total 

780 

(88) 

108 

(12) 

888 

(100)

No significant association exists between school type 

and performance for schools with below average 

student enrollment (i.e., charter schools and 

traditional public schools meet the standard in equal 

proportions) 

Significant association exists between school 

type and performance for schools with above 

average student enrollment (with a lower 

proportion of charter schools meeting the 

standard)*** 

***p<.01 

Note; State median = 439 students. 

Exhibit reads: In Massachusetts, 555 traditional public schools (85 percent) below the state median for student enrollment met the state 

performance standard, and 101 (15 percent) did not meet the standard in 2001-02. 
Appendix E-6 

North Carolina Student Performance Results 

The accountability system in North Carolina is called the “ABCs of Public Education.”113 This system 

includes a standard and an alternative system—the standard system is the focus of this analysis.114 The 

student performance analysis is based on 2,029 traditional public schools and 85 charter schools. Charter 

schools in North Carolina were more likely to be small and to serve larger proportions of minority 

students than traditional public schools.115 

Descriptive Statistics (North Carolina), 2001-02 

Mean (Standard Deviation in parenthesis) 

Traditional public schools 

(n=1,538) Charter schools (n=85) 

All Schools 

(n=2,114) 

Percent low-income 

48 

(23) Not reported 48 (23) 

Percent minority** 

41 

(27) 

47 

(37) 

41 

(28) 

Enrollment*** 

489 

(391) 

135 

(108) 

474 

(390) 

***p<.01, **p<.05 

Exhibit reads: In North Carolina, the mean proportion of low-income students in traditional public schools is 48 percent with a

standard deviation of 23 percent. 

In 2001-02, 17 schools did not meet the North Carolina performance standard (see below). These data 

indicate that 12 percent of charter schools did not meet the state performance standard, compared with 

less than 1 percent of traditional public schools. 

 

113 Under North Carolina’s accountability system students are tested in grades 3-12 in reading, writing, mathematics, 

science, and social studies, depending on the grade level. The school-level performance standard is based on an 

assessment of student test scores and demographic variables to determine whether schools are meeting expected 

growth, high growth, and performance based on composite scores. As part of the accountability system, each school 

is assigned one or more of the following categories of recognition: exemplary growth, expected growth, high 

growth, school of excellence, school of distinction, school of progress, priority school, 25 most improved K-8 

schools, 10 most improved high schools, low-performing, no recognition. Schools that are designated as low 

performing do not meet state standards and, as a result, are not eligible for rewards. 

114 The alternative system is not included because it did not have any charter schools. 

115 Mobility rates and proportion of low-income students are not included in this analysis. North Carolina does not 

maintain data on student mobility for any public school. In addition, it does not compile data on the proportion of 

students eligible for free or reduced-price lunches in charter schools. 
126

Performance of Traditional Public Schools and Charter Schools (North Carolina), 2001-02 

Number of Schools (Percentage) 

 Met standard Did not meet standard Total 

Traditional public school 

2,022 

(100) 

7 

(<1) 

2,029 

(96)

Charter 

75 

(88) 

10 

(12) 

85 

(4)

Total 

2,097 

(99) 

17 

(1) 

2,114 

(100)

Significant association exists between school type and performance with a lower 

proportion of charter schools meeting the standard*** 

***p<.01 

Exhibit reads: In North Carolina, 2,022 traditional public schools (100 percent) met the state performance 

standard, and 7 (fewer than 1 percent) did not meet the standard in 2001-02. 

Additional analyses were conducted to examine the association between performance and school type 

after controlling for the proportion of minority students and enrollment of schools. Charter schools were 

less likely to meet state performance standards when comparing similar groups of charter and traditional 

public schools based on these demographic characteristics (see tables that follow). In other words, 

controlling for demographics did not alter the finding that charter schools were less likely to meet state 

performance standards. 

Analysis of School Type and Performance, Controlling for 

Proportion of Minority Students (North Carolina), 2001-02 

Number of Schools (Percentage) 

Minority: 

Below State Median 

Minority: 

Above State Median 

Met 

std. 

Did not meet 

std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional 

public schools 

988 

(100) 

0 

(0) 

988

(96)

Traditional 

public schools 

988 

(100) 

5 

(<1) 

993 

(96)

Charter schools 

43 

(98) 

1 

(2) 

44

(4) Charter schools 

32 

(78) 

9 

(22) 

41 

(4)

Total 

1,031 

(100) 

1 

(<1) 

1,032

(100) Total 

1,020 

(99) 

14 

(1) 

1,034 

(100)

Significant association exists between school type 

and performance for schools with below average 

percentages of minority students (with a lower 

proportion of charter schools meeting the 

standard)**

Significant association exists between school type 

and performance for schools with above average 

percentages of minority students (with a lower 

proportion of charter schools meeting the 

standard)***

***p<.01, **p<.05 

Note: State median = 37 percent. 

Exhibit reads: In North Carolina, 998 traditional public schools (100 percent) below the state median for the proportion of minority 

students met the state performance standard, and zero did not meet the standard in 2001-02. 
127

Analysis of School Type and Performance, Controlling for Enrollment (North Carolina), 2001-02 

Number of Schools (Percentage) 

Enrollment: 

Below State Median 

Enrollment: 

Above State Median 

Met 

std. 

Did not meet 

std. Total 

Met 

std. 

Did not 

meet std. Total 

Traditional 

public schools 

951 

(99) 

1 

(<1) 

952

(92)

Traditional 

public schools 

1,025 

(100) 

4 

(<1) 

1,029 

(99)

Charter schools 

69 

(88) 

9 

(12) 

78

(8) Charter schools 

6 

(86) 

1 

(14) 

7 

(1)

Total 

1,020 

(99) 

10 

(1) 

1,030

(100) Total 

1,031 

(100) 

5 

(<1) 

1,036 

(100)

Significant association exists between school type 

and performance for schools with below average 

student enrollment (with a lower proportion of 

charter schools meeting the standard)***

Significant association exists between school type 

and performance for schools with above average 

student enrollment (with a lower proportion of 

charter schools meeting the standard)**

***p<.01, **p<.05 

Note: State median = 325 percent. 

Exhibit reads: In North Carolina, 951 traditional public schools (99 percent) below the state median for student enrollment met

the state performance standard, and one (fewer than 1 percent) did not meet the standard in 2001-02.